[{"id":"oa:W4413961449","type":"article-journal","title":"Diagnosis of nontuberculous mycobacterial infections using genomics and artificial intelligence-machine learning approaches: scope, progress and challenges","abstract":"The nontuberculous mycobacterial (NTM) infections cause morbidity and mortality in individuals who are immunocompromised and those with lung conditions. The timely diagnosis of NTM infections is thus the need of the hour for appropriate management of the disease. In this context, genomics has played a pivotal role in diagnosis of NTM by targeting various conserved regions which are useful for species identification and diagnosis. Also, the exploring of whole genome of nontuberculous mycobacteria has made species identification easier and has revolutionized the diagnostic landscape of NTM. The refinement of Whole Genome Sequencing (WGS) and the advent of targeted Next Generation Sequencing (tNGS) and metagenomic NGS (mNGS) has helped in bringing down the cost without compromising the quality in NTM diagnostics. The advent of artificial intelligence (AI) technologies has made NTM diagnosis even easier by analyzing complex genomic data and providing faster results. Thus, this comprehensive review discusses the strides made in genomics and AI based approaches in the diagnosis of NTM infections and the way forward for harnessing this potential to the maximum for the benefit of mankind.","author":[{"family":"Murthy","given":"Madhan"},{"family":"Gupta","given":"Vivek"},{"family":"Maurya","given":"Anand"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fmicb.2025.1665685","URL":"https://doi.org/10.3389/fmicb.2025.1665685","source":"openalex"},{"id":"oa:W4409607246","type":"article-journal","title":"Artificial Intelligence in Midwifery: A Scoping Review of Current Applications, Future Prospects, and Midwives’ Perspectives","abstract":"Background/Objectives: Artificial intelligence (AI) is considered one of the core technological advancements of Industry 4.0, expected to transform various sectors, including healthcare. Midwifery can greatly benefit from AI; however, its current use, its future potential, and midwives’ attitudes remain underexplored. This study aimed to investigate the implementation of, prospects of, and attitudes of midwives toward AI. Methods: A scoping review was carried out, following the PRISMA guidelines. The search was conducted in Pubmed, Scopus, and Web of Science, from database inception to 2 February 2025. Results: Eight studies met the inclusion criteria. Although AI is not yet widely implemented in midwifery, it has notable potential. Several potential benefits were recorded, such as the enhancement of clinical education through personalized learning tools, such as AI-driven virtual patients and customized assessments, as well as a reduction in clinical errors via predictive models and real-time monitoring technologies. The adoption of AI is therefore expected to improve quality of care, particularly in perinatal and neonatal settings. However, it was found that the integration remains limited due to two key obstacles: ethical concerns (e.g., data privacy) and a notable level of anxiety or hesitation among midwives, associated with low levels of digital health literacy. Conclusions: It is important to form a relevant framework regarding the use of AI in midwifery, addressing ethical concerns and skepticism. Additionally, targeted educational interventions are needed to enhance midwives’ AI literacy and alleviate concerns. In general, it is essential to overcome these barriers to accelerate AI adoption in midwifery and unlock its full potential in perinatal care.","author":[{"family":"Giaxi","given":"Paraskevi"},{"family":"Vivilaki","given":"Victoria"},{"family":"Sarella","given":"Angeliki"},{"family":"Gourounti","given":"Kleanthi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/healthcare13080942","URL":"https://doi.org/10.3390/healthcare13080942","source":"openalex"},{"id":"oa:W4411200075","type":"article-journal","title":"Transforming surgical planning and procedures through the synergistic use of additive manufacturing, advanced materials and artificial intelligence: challenges and opportunities","abstract":"Additive manufacturing (AM) is a powerful approach in healthcare to augment the functionalities of patient-specific medical products and surgical tools. One such area of the healthcare industry is surgical planning and procedures, where the benefits of AM can revolutionize the industry. AM technologies, commonly known as three-dimensional (3D) printing, can change the conventional surgical methodology from the \"open-detect-operate-close\" mode to the \"detect-open-operate-close\" mode. However, the use of 3D printing in surgical planning has been hampered by the limited availability of literature reports thoroughly examining the advantages and drawbacks of this technology in clinical settings. Hence, this review explores the widespread use of additive manufacturing, multi-materials, metamaterials, 4D printing, and artificial intelligence in surgical planning for complex surgical procedures of the spine and in orthopedics, dentistry, cardiology, gynecology, and neurology. This review focuses on meticulously adjusting the lattice structure of metamaterials during 3D printing to achieve specific mechanical properties. It further delves into 4D printing to achieve dynamic capabilities in 3D printed models for better integration with the host tissue. Furthermore, it highlights the key aspects of combining AM with artificial intelligence/machine learning (AI/ML) models in healthcare to automate 3D model production and thereby reduce human intervention. This comprehensive review offers bioengineers, clinical scientists, and clinicians a platform to explore AM and its potential for addressing pre- and post-surgical operation challenges, providing valuable insights for biomedical engineering and healthcare advancements.","author":[{"family":"Tripathi","given":"Shivi"},{"family":"Ansari","given":"Aftab"},{"family":"Singh","given":"Manisha"},{"family":"Dash","given":"Madhusmita"},{"family":"Kumar","given":"Prasoon"},{"family":"Singh","given":"Harpreet"},{"family":"Panda","given":"Biranchi"},{"family":"Nukavarapu","given":"Syam"},{"family":"Camciunal","given":"Gulden"},{"family":"Li","given":"Bingbing"},{"family":"Jain","given":"Prashant"},{"family":"Jayaganthan","given":"Rengaswamy"},{"family":"Mehboob","given":"Hassan"},{"family":"Junaedi","given":"Harri"},{"family":"Nanda","given":"Himansu"},{"family":"Chen","given":"Guoping"},{"family":"Kundu","given":"Subhas"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1039/d5mh00501a","URL":"https://doi.org/10.1039/d5mh00501a","source":"openalex"},{"id":"oa:W4412582607","type":"article-journal","title":"Generative artificial intelligence in medicine: a mixed-methods survey of UK general practitioners","abstract":"Objective To explore the opinions of general practitioners (GPs) in the UK about the use of generative artificial intelligence (AI) tools in primary care. Methods and analysis At the beginning of 2024, using a convenience sample, we administered an online mixed-methods survey to registered GPs currently working in the UK. Results A total of 1006 GPs responded, with 53% being male and 54% over 46 years old. One-fifth of GPs reported having used AI for clinical practice, with male doctors and those in bigger cities being more likely to have used it. 80% of respondents expressed a need for more training in understanding these tools. GPs at least somewhat agreed AI would improve documentation (59%) and patient information gathering (56%). 55% felt AI could increase inequities and 54% saw potential for patient harm, but 47% believed it could enhance healthcare efficiency. GPs who used these tools were significantly more optimistic about the scope for generative AI in improving clinical tasks. One-third of GPs left comments that were classified into four major themes: (1) lack of familiarity and understanding with AI, (2) role of AI in clinical practice, (3) concerns about AI and (4) AI and the future of healthcare. Conclusions This study highlights UK GPs’ developing perspectives on generative AI in clinical practice, emphasising the need for more training. Many GPs reported a lack of knowledge and experience with this technology, although a portion already used non-medical grade technology for clinical tasks, with the risks that this entails.","author":[{"family":"Kharko","given":"Anna"},{"family":"Locher","given":"Cosima"},{"family":"Torous","given":"John"},{"family":"Rosch","given":"Sophie"},{"family":"Hägglund","given":"Maria"},{"family":"Gaab","given":"Jens"},{"family":"Mcmillan","given":"Brian"},{"family":"Sundemo","given":"David"},{"family":"Mandl","given":"Kenneth"},{"family":"Blease","given":"Charlotte"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/bmjdhai-2025-000051","URL":"https://doi.org/10.1136/bmjdhai-2025-000051","source":"openalex"},{"id":"oa:W4408688362","type":"article-journal","title":"Artificial Intelligence in the Service of Medicine: Current Solutions and Future Perspectives, Opportunities, and Challenges.","abstract":"Objective: This article aims to identify the opportunities and risks of Artificial Intelligence tools (AIT) applied to clinical practice, while also reflecting on their impact on the doctor-patient relationship. Materials and Methods: The authors conducted a systematic literature review following the PRISMA guidelines, selecting the period from 2019 to October 2024. Academic databases PubMed and Scopus were drawn upon by using the keywords and searchstrings \"artificial intelligence\", \"healthcare\", \"informed consent\", and \"doctor-patient relationship\" in titles, abstracts, and keywords. Results and Discussion: AIT has proven useful in significantly reducing the time spent on bureaucratic tasks and minimizing errors compared to traditional medicine. However, their effectiveness is highly influenced by the quantity and quality of data used for training. Additionally, there is an issue with the transparency of the decision-making process because AIT and even their programmers are unable to explain their diagnostic and therapeutic recommendations. Therefore, human supervision of AI work is essential. Conclusions: The potential risks of AI for patient safety and personal data security necessitate that governments urge those involved in the production of AI tools to adhere to specific ethical standards developed with the participation of all stakeholders, including patients.","author":[{"family":"Marinelli","given":"Susanna"},{"family":"Paola","given":"Lina"},{"family":"Stark","given":"Michael"},{"family":"Vergallo","given":"Gianluca"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7417/ct.2025.5192","URL":"https://doi.org/10.7417/ct.2025.5192","source":"openalex"},{"id":"oa:W4410708995","type":"article-journal","title":"Critical thinking in the age of generative AI: implications for health sciences education","abstract":"Generative artificial intelligence (genAI) systems are progressively transforming health science education and research by assisting clinicians in diagnosis and structuring specific intervention regimens (MIR et al., 2023). Moreover, these technologies serve educators in yielding simple concept-based educational modules tailored as per student's requirements and large language models (LLMs) and analogous models display the potential for automated streamlined literature reviews, prompt generation of interpretations and conclusions, and effortless drafting of manuscripts within seconds (MIR et al., 2023;Al Kuwaiti et al., 2023;Gupta et al., 2024) thereby, offering a potentially high level of convenience and efficiency. Furthermore, many researchers contend that genAI has the potential to completely automate research processes, including drafting proposals, analyzing data, and composing concluding reports (Almansour and Alfhaid, 2024;Preiksaitis and Rose, 2023). Hence, these merits indicate a prospective future for genAI, particularly in domains that demand processing of large scale data and iterative analyses.However, despite these evident positive outcomes, concerns persist regarding the quality of AIgenerated outputs, which may include inaccuracies in text reporting that contribute to misinformation, logical inconsistencies, outdated or unverified claims, and hallucinated references, all of which undermine the academic credibility of writing (Athaluri et al., 2023;Farrelly and Baker, 2023;Sittig and Singh, 2024). In addition, a lack of transparency in datasets exacerbates ethical challenges and biases (Norori et al., 2021). In an era characterized by the expedited advancement and refinement of genAI, it is necessary to critically evaluate whether these systems encourage critical thinking and human intelligence or subtly undermine them. This raises an important question: Are we unintentionally relinquishing the cognitive abilities that have propelled scientific and clinical advancements, as healthcare professionals progressively integrate genAI into clinical and academic domains? The dilemma lies in balancing the efficiency of genAI with the preservation of essential human cognitive skills, such as critical thinking, ethical reasoning, and independent problem-solving. At its core, this dilemma focuses on how health professionals, medical trainees, and early career researchers apply outcomes produced by genAI as overreliance risk supporting passive dependence on algorithm produced outcomes in a context where time and cognitive capacity are consistently constrained. The expertise involving scientific accuracy, ethical judgment, and diagnostic reasoning are cultivated through proactive contribution by integrating knowledge in innovative and contextually relevant approaches, critically evaluating evidence, and grappling with uncertainty that genAI cannot substitute (Passerini et al., 2025;Shoja et al., 2023).The significance of the risk of cognitive complacency is emphasized globally, as industry and academia compete to implement genAI tools, and have rapidly accelerated publications related to AI, reflecting both enthusiasm and apprehension. However, medical professionals and researchers may demonstrate overdependence on genAI tools due to faster processing, thus, reducing opportunities for independent problem-solving and critical thinking (Shoja et al., 2023;Zhai et al., 2024). Additionally, in medical research where precision plays an important factor, underlying biases in data training of genAI can propagate false information (Norori et al., 2021). The inefficacy of plagiarism detection software to recognize text generated by genAI, undermines conventional ethical integrity measures as the output may be erroneously identified as genuine scholarly writing (Farrelly and Baker, 2023), though some exceptions exist (Elkhatat et al., 2023;Weber-Wulff et al., 2023). This attitude leads to a workforce adept at utilizing genAI but defic","author":[{"family":"Naqvi","given":"Waqar"},{"family":"Ganjoo","given":"Rohini"},{"family":"Rowe","given":"Michael"},{"family":"Pashine","given":"Aishwarya"},{"family":"Mishra","given":"Gaurav"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/frai.2025.1571527","URL":"https://doi.org/10.3389/frai.2025.1571527","source":"openalex"},{"id":"oa:W4412191263","type":"article-journal","title":"Artificial Intelligence in Risk Stratification and Outcome Prediction for Transcatheter Aortic Valve Replacement: A Systematic Review and Meta-Analysis","abstract":"Background/Objectives: Transcatheter aortic valve replacement (TAVR) has been introduced as an optimal treatment for patients with severe aortic stenosis, offering a minimally invasive alternative to surgical aortic valve replacement. Predicting these outcomes following TAVR is crucial. Artificial intelligence (AI) has emerged as a promising tool for improving post-TAVR outcome prediction. In this systematic review and meta-analysis, we aim to summarize the current evidence on utilizing AI in predicting post-TAVR outcomes. Methods: A comprehensive search was conducted to evaluate the studies focused on TAVR that applied AI methods for risk stratification. We assessed various ML algorithms, including random forests, neural networks, extreme gradient boosting, and support vector machines. Model performance metrics—recall, area under the curve (AUC), and accuracy—were collected with 95% confidence intervals (CIs). A random-effects meta-analysis was conducted to pool effect estimates. Results: We included 43 studies evaluating 366,269 patients (mean age 80 ± 8.25; 52.9% men) following TAVR. Meta-analyses for AI model performances demonstrated the following results: all-cause mortality (AUC = 0.78 (0.74–0.82), accuracy = 0.81 (0.69–0.89), and recall = 0.90 (0.70–0.97); permanent pacemaker implantation or new left bundle branch block (AUC = 0.75 (0.68–0.82), accuracy = 0.73 (0.59–0.84), and recall = 0.87 (0.50–0.98)); valve-related dysfunction (AUC = 0.73 (0.62–0.84), accuracy = 0.79 (0.57–0.91), and recall = 0.54 (0.26–0.80)); and major adverse cardiovascular events (AUC = 0.79 (0.67–0.92)). Subgroup analyses based on the model development approaches indicated that models incorporating baseline clinical data, imaging, and biomarker information enhanced predictive performance. Conclusions: AI-based risk prediction for TAVR complications has demonstrated promising performance. However, it is necessary to evaluate the efficiency of the aforementioned models in external validation datasets.","author":[{"family":"Shojaei","given":"Shayan"},{"family":"Mousavi","given":"Asma"},{"family":"Kazemian","given":"Sina"},{"family":"Armani","given":"Shiva"},{"family":"Maleki","given":"Saba"},{"family":"Fallahtafti","given":"Parisa"},{"family":"Arashlow","given":"Farzin"},{"family":"Daryabari","given":"Yasaman"},{"family":"Naderian","given":"Mohammadreza"},{"family":"Alkhouli","given":"Mohamad"},{"family":"Rana","given":"Jamal"},{"family":"Mehrani","given":"Mehdi"},{"family":"Jenab","given":"Yaser"},{"family":"Hosseini","given":"Kaveh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jpm15070302","URL":"https://doi.org/10.3390/jpm15070302","source":"openalex"},{"id":"oa:W4410421263","type":"article-journal","title":"Harnessing artificial intelligence for transforming dementia care: Innovations in early detection and treatment","abstract":"Dementia, particularly Alzheimer's Disease, continues to be a significant global health concern, driven by increasing prevalence as the population ages. Early detection and accurate diagnosis are essential for improving patient outcomes and mitigating the associated healthcare burden. Artificial intelligence (AI) has emerged as a powerful tool in dementia care, providing innovative approaches to the early detection, diagnosis, and management of these neurodegenerative conditions. This review examines the role of AI in revolutionizing dementia care by focusing on its application in neuroimaging, biomarker identification, predictive modeling, and therapeutic interventions.","author":[{"family":"Habbal","given":"Saadeddine"},{"family":"Mian","given":"Maamoon"},{"family":"Imam","given":"Musa"},{"family":"Tahiri","given":"Jihane"},{"family":"Amor","given":"Adam"},{"family":"Reddy","given":"PH"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.bosn.2025.05.001","URL":"https://doi.org/10.1016/j.bosn.2025.05.001","source":"openalex"},{"id":"oa:W4415401926","type":"article-journal","title":"Artificial Intelligence-Based Epileptic Seizure Prediction Strategies: A Review","abstract":"Epilepsy, a chronic neurological disorder marked by recurrent and unpredictable seizures, poses significant risks of injury and compromises patient quality of life. The accurate forecasting of seizures is paramount for enabling timely interventions and improving safety. Since the 1970s, research has increasingly focused on analyzing bioelectrical signals for this purpose. In recent years, artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), has emerged as a powerful tool for seizure prediction. This review, conducted by PRISMA guidelines, analyzes studies from 2020 to August 2025. It explores the evolution from traditional ML classifiers toward advanced DL architecture, including convolutional and recurrent neural networks and transformer-based frameworks, applied to bioelectrical signals. While these approaches show promising performance, significant challenges persist in patient generalization, standardized evaluation, and clinical validation. This review synthesizes current advancements, provides a critical analysis of methodological limitations, and outlines future directions for developing robust, clinically relevant seizure prediction systems to enhance patient autonomy and outcomes.","author":[{"family":"Perez-Sanchez","given":"Andrea"},{"family":"Valtierrarodriguez","given":"Martin"},{"family":"De-Santiago-Perez","given":"JJ"},{"family":"Perez-Ramirez","given":"Carlos"},{"family":"García-Pérez","given":"Arturo"},{"family":"Amézquita-Sánchez","given":"Juan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ai6100274","URL":"https://doi.org/10.3390/ai6100274","source":"openalex"},{"id":"oa:W4414793290","type":"article-journal","title":"Artificial Intelligence in Glioma Diagnosis: A Narrative Review of Radiomics and Deep Learning for Tumor Classification and Molecular Profiling Across Positron Emission Tomography and Magnetic Resonance Imaging","abstract":"Background: This narrative review summarizes recent progress in artificial intelligence (AI), especially radiomics and deep learning, for non-invasive diagnosis and molecular profiling of gliomas. Methodology: A thorough literature search was conducted on PubMed, Scopus, and Embase for studies published from January 2020 to July 2025, focusing on clinical and technical research. In key areas, these studies examine AI models’ predictive capabilities with multi-parametric Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET). Results: The domains identified in the literature include the advancement of radiomic models for tumor grading and biomarker prediction, such as Isocitrate Dehydrogenase (IDH) mutation, O6-methylguanine-dna methyltransferase (MGMT) promoter methylation, and 1p/19q codeletion. The growing use of convolutional neural networks (CNNs) and generative adversarial networks (GANs) in tumor segmentation, classification, and prognosis was also a significant topic discussed in the literature. Deep learning (DL) methods are evaluated against traditional radiomics regarding feature extraction, scalability, and robustness to imaging protocol differences across institutions. Conclusions: This review analyzes emerging efforts to combine clinical, imaging, and histology data within hybrid or transformer-based AI systems to enhance diagnostic accuracy. Significant findings include the application of DL to predict cyclin-dependent kinase inhibitor 2A/B (CDKN2A/B) deletion and chemokine CCL2 expression. These highlight the expanding capabilities of imaging-based genomic inference and the importance of clinical data in multimodal fusion. Challenges such as data harmonization, model interpretability, and external validation still need to be addressed.","author":[{"family":"Christodoulou","given":"Rafail"},{"family":"Pitsillos","given":"Rafael"},{"family":"Papageorgiou","given":"Platon"},{"family":"Petrou","given":"Vasileia"},{"family":"Vamvouras","given":"Georgios"},{"family":"Rivera","given":"Ludwing"},{"family":"Papageorgiou","given":"Sokratis"},{"family":"Solomou","given":"Elena"},{"family":"Georgiou","given":"Michalis"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/eng6100262","URL":"https://doi.org/10.3390/eng6100262","source":"openalex"},{"id":"oa:W4415233076","type":"article-journal","title":"Implementing artificial intelligence and machine learning algorithms for optimized crop management: a systematic review on data-driven approach to enhancing resource use and agricultural sustainability","abstract":"Artificial intelligence (AI) and machine learning (ML) are transforming agriculture by enabling data-driven decisions that elevate productivity and sustainability. This review synthesises 95 studies published between 2013 and 2023 that evaluate applications across crop monitoring, yield prediction, and resource optimisation. Reported model accuracies for neural networks, decision trees, and deep learning reached up to 93 percent; deep learning was most accurate but least interpretable. Reported benefits include a 25 percent increase in yield, a 28 percent reduction in costs, 40 percent efficiency gains, 22 percent water savings, 28 percent fertilizer savings, and 35 percent lower nitrogen runoff. Adoption barriers persist, including poor data quality, expensive infrastructure, limited digital literacy, and ethical concerns around data ownership and bias. Integrated, enterprise-scale platforms favor large farms, while mobile AI applications yield 15-30 percent gains for smallholders. Converging technologies blockchain, IoT, and robotics enable integration, and automation can lower labor and input requirements by 35 percent. The review points to the importance of inclusion policies, transparent systems, and global governance. Overall, AI/ML are drivers of socio-technical transition consistent with Sustainability Transitions Theory, necessitating multidisciplinary strategies for sustainable, climate-resilient food systems.","author":[{"family":"Ugwu","given":"Okechukwu"},{"family":"Ogenyi","given":"Fabian"},{"family":"Alum","given":"Esther"},{"family":"Eze","given":"Val"},{"family":"Basajja","given":"Mariam"},{"family":"Ugwu","given":"Jovita"},{"family":"Ugwu","given":"Chinyere"},{"family":"Ejemot-Nwadiaro","given":"Regina"},{"family":"Okon","given":"Michael"},{"family":"Egba","given":"Simeon"},{"family":"Ejim","given":"Uti"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/23311932.2025.2569982","URL":"https://doi.org/10.1080/23311932.2025.2569982","source":"openalex"},{"id":"oa:W4416975840","type":"article-journal","title":"Artificial intelligence in oncology drug development and management: a precision medicine perspective","abstract":"The management of oncology drugs is inherently complex, facing challenges such as high development costs, prolonged timelines, and substantial inter-patient heterogeneity. Recent advances in artificial intelligence (AI) have introduced transformative capabilities across the entire cancer drug lifecycle-from target discovery and compound screening to clinical trial optimization, individualized therapy, toxicity management, supply chain logistics, and regulatory oversight. AI enables precise target identification, accelerates virtual drug screening and molecular design, and enhances clinical trial efficiency through intelligent patient stratification and adaptive protocols. Moreover, AI facilitates personalized treatment decision-making, early prediction of drug resistance, and real-time toxicity surveillance, while improving pharmacovigilance and post-market drug evaluation using real-world data. Here, \"post-market drug evaluation\" refers to real-world effectiveness and safety assessment using spontaneous reports (e.g., FAERS/VigiBase) and EHR/claims-based outcomes, rather than cost-effectiveness analyses. Examples include EHR-NLP to surface immune-related adverse events, AI-assisted trial recruitment and adaptive designs, and individualized dosing frameworks (e.g., CURATE.AI). Despite its enormous promise, AI-driven oncology drug management faces notable challenges in data integration, model interpretability, clinical translation, fairness, and regulatory governance. This review comprehensively summarizes the current applications of AI in oncology pharmacology, highlights key opportunities and barriers, and explores future directions at the intersection of AI, precision medicine, and cancer therapeutics. Future priorities include prospective multi-site evaluations, fairness auditing, and continuous post-market algorithmovigilance.","author":[{"family":"Fang","given":"Caixia"},{"family":"Zhou","given":"Pengfa"},{"family":"Zhang","given":"Xuerong"},{"family":"He","given":"Yongsheng"},{"family":"Yang","given":"Qingwei"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fonc.2025.1609827","URL":"https://doi.org/10.3389/fonc.2025.1609827","source":"openalex"},{"id":"oa:W4405986075","type":"article-journal","title":"Digitalization and Artificial Intelligence as Motivators for Healthcare Professionals","abstract":"Background: Digitalization and artificial intelligence technologies are navigating and strengthening human labour practices and organizational performance in healthcare. Research has shown that digitization and AI can help healthcare professionals through real-time insights and recommendations derived from extensive datasets. These modern technologies are advancing beyond being mere instruments in the health sector, now acting as partners to aid healthcare professionals in making better predictions and decisions by offering timely insights and suggestions derived from extensive datasets, as well as pinpoint potential health issues with greater accuracy and speed. It is estimated also that significantly can be supported the management of workplace well-being. Objective: This article delves into how AI and digitalization help healthcare professionals by boosting efficiency, meeting their professional and personal needs, and showcasing how they can enhance employee mental health and well-being. Results: It is crucial to recognize that issues arise from the intrinsic complexity and opaque nature of AI, the risk of job loss, and the disruption of the conventional interaction between physicians and patients. Nonetheless, AI in the healthcare facilities should not be seen as a danger to human employees. Instead, AI strive to support healthcare employees, enabling them to allocate more time to complex and crucial tasks. By automating tasks that are repetitive and mundane, these new technologies can lessen the burden on healthcare professionals, allowing them to dedicate more time to caring for patients and engaging in valuable interactions. Conclusions: The integration of AI and digitalization technologies into healthcare presents both opportunities and challenges for employee motivation and job performance. Although it can improve effectiveness and lower stress levels, it is important to carefully address worries about employment stability and maintaining personal connections in healthcare. Organizations need to prioritize creating a workspace that encourages and assists employees in adjusting to new technological changes","author":[{"family":"Karaferis","given":"Dimitris"},{"family":"Dimitra","given":"Balaska"},{"family":"Yannis","given":"Pollalis"}],"issued":{"date-parts":[[2025]]},"DOI":"10.33425/2690-8077.1170","URL":"https://doi.org/10.33425/2690-8077.1170","source":"openalex"},{"id":"oa:W4409001773","type":"article-journal","title":"Clinical Application of Artificial Intelligence in Digital Breast Tomosynthesis","abstract":"Digital breast tomosynthesis (DBT) provides improved cancer detection and lower recall rates when compared with full-field digital mammography (DM) and has been widely adopted for breast cancer screening. However, adopting DBT presents new challenges such as an increased number of acquired images resulting in longer interpretation times. Artificial intelligence (AI) offers numerous opportunities to enhance the advantages of DBT and mitigate its shortcomings. Research in the DBT AI domain has grown significantly and AI algorithms play a key role in the screening and diagnostic phases of breast cancer detection and characterization. The application of AI may streamline the workflow and reduce the time required for radiologists to interpret images. In addition, AI can minimize radiation exposure and enhance lesion visibility in synthetic two-dimensional DM images. This review provides an overview of AI technology in DBT, its clinical applications, and future considerations.","author":[{"family":"Chang","given":"Jung"},{"family":"Lee","given":"Weonsuk"},{"family":"Bahl","given":"Manisha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3348/jksr.2025.0011","URL":"https://doi.org/10.3348/jksr.2025.0011","source":"openalex"},{"id":"oa:W4407925758","type":"article-journal","title":"Comparing Artificial Intelligence and manual methods in systematic review processes: protocol for a systematic review","abstract":"Objectives This systematic review aims to evaluate the effectiveness of automated methods using artificial intelligence (AI) in conducting systematic reviews, with a focus on both performance and resource utilization compared to human reviewers. Study Design and Setting This systematic review and meta-analysis protocol follows the Cochrane Methodology protocol and review guidance. We searched five bibliographic databases to identify potential studies published in English from 2005. Two independent reviewers will screen the titles and abstracts, followed by a full-text review of the included articles. Any discrepancies will be resolved through discussion and, if necessary, referral to a third reviewer. The risk of bias (RoB) in included studies will be assessed at the outcome level using the revised Cochrane risk-of-bias tool for randomized trials and the RoB In Non-randomized Studies - of Interventions for non-randomized studies. Where appropriate, we plan to conduct meta-analysis using random-effects models to obtain pooled estimates. We will explore the sources of heterogeneity and conduct sensitivity analyses based on prespecified characteristics. Where meta-analysis is not feasible, a narrative synthesis will be performed. Results We will present the results of this review, focusing on performance and resource utilization metrics. Conclusion This systematic review will evaluate the effectiveness of automated methods, especially AI tools in systematic reviews, aiming to synthesize current evidence on their performance, resource utilization, and impact on review quality. The findings will inform evidence-based recommendations for systematic review authors and developers on implementing automation tools to optimize review efficiency while maintaining methodological rigor. In addition, we will identify key research gaps to guide future development of AI-assisted systematic review methods. Plain Language Summary A systematic review is a thorough and organized summary of all relevant studies on a specific topic. These reviews are important for gathering evidence to guide health care decisions, but they often take a lot of time and effort. Recently, tools using artificial intelligence (AI) have been developed to speed up this process. We will conduct a systematic review to see how well these AI tools perform compared to human reviewers. We will examine studies from 2005 that have used AI to conduct systematic reviews. We will assess how well AI tools find the right information, how much time and work they save, and how easy and reliable they are for users. This study aims to help researchers choose the best AI tools to make systematic reviews faster and more efficient without losing quality.","author":[{"family":"Pang","given":"Xuenan"},{"family":"Saifurrahman","given":"Km"},{"family":"Berhane","given":"Sarah"},{"family":"Yao","given":"Xiaomei"},{"family":"Kothari","given":"Kavita"},{"family":"Taneri","given":"Petek"},{"family":"Thomas","given":"James"},{"family":"Devane","given":"Declan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jclinepi.2025.111738","URL":"https://doi.org/10.1016/j.jclinepi.2025.111738","source":"openalex"},{"id":"oa:W4414283228","type":"article-journal","title":"Cross-border higher education cooperation under the dual context of artificial intelligence and geopolitics: opportunities, challenges, and pathways","abstract":"This paper examines the profound impact of artificial intelligence (AI) and geopolitics on cross-border higher education cooperation. AI has the potential to enhance educational accessibility and collaboration efficiency by enabling personalized learning, virtual classrooms, open resource platforms, and open-source research collaborations, ultimately helping bridge global educational gaps. However, significant challenges arise, such as techno-nationalism (e.g., semiconductor export controls), data sovereignty conflicts (e.g., GDPR restrictions), divergent algorithmic values, and the expanding digital divide. To address these challenges, this study proposes several solutions: the creation of an inclusive technological ecosystem (including open-source platforms, shared computing power, and cross-cultural models); the development of mutual recognition mechanisms (such as data stratification and standard harmonization); the strengthening of South-South cooperation through digital public goods; and the reconstruction of ethical consensus, emphasizing cultural diversity and human-in-the-loop principles. Notably, China has actively contributed to these efforts through technological empowerment (e.g., National Smart Education Platform, Luban Workshops), regulatory input (e.g., Global Governance Initiative), and infrastructure support. Looking ahead, the paper argues for the establishment of an “Intelligent Education Community,” guided by the principles of “extensive consultation, joint contribution, shared benefits, and wise governance,” to ensure that AI advances global educational equity and promotes human progress.","author":[{"family":"Zhu","given":"Yaoshun"},{"family":"Zhu","given":"Yaoshun"},{"family":"Zhu","given":"Zhitao"},{"family":"Xu","given":"Wenyao"},{"family":"Zhu","given":"Yaoshun"},{"family":"Zhu","given":"Yaoshun"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/feduc.2025.1656518","URL":"https://doi.org/10.3389/feduc.2025.1656518","source":"openalex"},{"id":"oa:W4400667193","type":"article-journal","title":"Procedural Content Generation via Generative Artificial Intelligence","abstract":"The attempt to utilize machine learning in procedural content generation (PCG) has been made in the past. In this survey paper, we investigate how generative artificial intelligence (AI), which saw a significant increase in interest in the mid-2010s, is being used for PCG. We review applications of generative AI for the creation of various types of content, including terrains, items, and even storylines. While generative AI is effective for PCG, building high-performance models requires not only handling customized content and ensuring quality and diversity, but also securing sufficient training data. For PCG research to advance further, addressing these challenges is essential. Thus, we also give special consideration to research that explores innovative generation techniques, model architectures, and approaches suited for limited-data scenarios.","author":[{"family":"Mao","given":"Xinyu"},{"family":"Yu","given":"Wanli"},{"family":"Okawara","given":"Yuya"},{"family":"Zhan","given":"Xueying"},{"family":"Yamada","given":"Kazunori"},{"family":"Zielewski","given":"Michael"}],"issued":{"date-parts":[[2026]]},"DOI":"10.4036/iis.2026.r.01","URL":"https://doi.org/10.4036/iis.2026.r.01","source":"openalex"},{"id":"oa:W4415685397","type":"article-journal","title":"Explainable artificial intelligence for gait analysis: advances, pitfalls, and challenges - a systematic review","abstract":"Machine learning (ML) has emerged as a powerful tool to analyze gait data, yet the \"black-box\" nature of many ML models hinders their clinical application. Explainable artificial intelligence (XAI) promises to enhance the interpretability and transparency of ML models, making them more suitable for clinical decision-making. This systematic review, registered on PROSPERO (CRD42024622752), assessed the application of XAI in gait analysis by examining its methods, performance, and potential for clinical utility. A comprehensive search across four electronic databases yielded 3676 unique records, of which 31 studies met inclusion criteria. These studies were categorized into model-agnostic (n = 16), model-specific (n = 12), and hybrid (n = 3) interpretability approaches. Most applied local interpretation methods such as SHAP and LIME, while others used Grad-CAM, attention mechanisms, and Layer-wise Relevance Propagation. Clinical populations studied included Parkinson's disease, stroke, sarcopenia, cerebral palsy, and musculoskeletal disorders. Reported outcomes highlighted biomechanically relevant features such as stride length and joint angles as key discriminators of pathological gait. Overall, the findings demonstrate that XAI can bridge the gap between predictive performance and interpretability, but significant challenges remain in standardization, validation, and balancing accuracy with transparency. Future research should refine XAI frameworks and assess their real-world clinical applicability across diverse gait disorders.","author":[{"family":"Xiang","given":"Liangliang"},{"family":"Gao","given":"Zixiang"},{"family":"Yu","given":"Peimin"},{"family":"Fernandez","given":"Justin"},{"family":"Gu","given":"Yaodong"},{"family":"Wang","given":"Ruoli"},{"family":"Gutierrez-Farewik","given":"Elena"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fbioe.2025.1671344","URL":"https://doi.org/10.3389/fbioe.2025.1671344","source":"openalex"},{"id":"oa:W7118193480","type":"article-journal","title":"Patients’ views on the use of artificial intelligence in healthcare: Artificial Intelligence Survey Aachen (AISA)—a prospective survey","abstract":"OBJECTIVES: The use of AI is gaining relevance in healthcare. There is limited information regarding the views of patients on AI in healthcare. The aim of our study was to assess the views of patients on the use of AI in healthcare with an on-site questionnaire. MATERIALS AND METHODS: Patients in our tertiary hospital with a diagnostic imaging appointment were invited to complete a paper-based questionnaire between December 2022 and October 2023. We asked about socio-demographic data, experience, knowledge, and their opinion on the use of AI in healthcare, focusing on the fields (1) diagnostics, (2) therapy, and (3) triage. RESULTS: Out of a total of 198 patients (mean age 49.41 ± 17.6 years, 99 female), 91.5% stated that they expected benefits from the implementation of AI in healthcare, although 73.4% rated their knowledge of AI as moderate to none. The majority of patients were in favour of using AI in diagnostics (87.2%) and therapy (73.1%), while only 28.2% approved its use in patient triage. 84.0% wanted to be informed about the use of AI in at least one of the mentioned areas. Participants with higher education, higher self-assessed knowledge of AI and personal experience with AI showed greater approval for AI in healthcare. CONCLUSION: Our interviewed patients have a rather open attitude towards AI in healthcare, with differentiated views depending on the topic; patients are in favour of the use of AI, especially in diagnostics and to a lesser extent also for therapy support, but they reject its use for triage. CRITICAL RELEVANCE STATEMENT: Overall, the results emphasise the need for widespread efforts to address patient concerns about AI in healthcare, including enhancing understanding and acceptance while protecting marginalised groups. This will help clinical radiology to adopt AI more effectively. KEY POINTS: There is limited information on patients' views of AI in healthcare, often focused on specific groups, limiting generalizability. Patients are open to AI in healthcare, supporting its use in diagnostics and therapy, but rejecting its use for triage. Overall, patients want to be informed about AI usage and participants with higher education and AI experience showed more approval.","author":[{"family":"Baldus","given":"Sophie"},{"family":"Wiesmann","given":"Martin"},{"family":"Habel","given":"U"},{"family":"Gerhards","given":"Anna"},{"family":"Hasan","given":"Dimah"},{"family":"Weyland","given":"Charlotte"},{"family":"Truhn","given":"Daniel"},{"family":"Hasl","given":"Marian"},{"family":"Clemens","given":"Benjamin"},{"family":"Nikoubashman","given":"Omid"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1186/s13244-025-02159-3","URL":"https://doi.org/10.1186/s13244-025-02159-3","source":"openalex"},{"id":"oa:W4409649298","type":"article-journal","title":"Trust, Trustworthiness, and the Future of Medical AI: Outcomes of an Interdisciplinary Expert Workshop","abstract":"Trustworthiness has become a key concept for the ethical development and application of artificial intelligence (AI) in medicine. Various guidelines have formulated key principles, such as fairness, robustness, and explainability, as essential components to achieve trustworthy AI. However, conceptualizations of trustworthy AI often emphasize technical requirements and computational solutions, frequently overlooking broader aspects of fairness and potential biases. These include not only algorithmic bias but also human, institutional, social, and societal factors, which are critical to foster AI systems that are both ethically sound and socially responsible. This viewpoint article presents an interdisciplinary approach to analyzing trust in AI and trustworthy AI within the medical context, focusing on (1) social sciences and humanities conceptualizations and legal perspectives on trust and (2) their implications for trustworthy AI in health care. It focuses on real-world challenges in medicine that are often underrepresented in theoretical discussions to propose a more practice-oriented understanding. Insights were gathered from an interdisciplinary workshop with experts from various disciplines involved in the development and application of medical AI, particularly in oncological imaging and genomics, complemented by theoretical approaches related to trust in AI. Results emphasize that, beyond common issues of bias and fairness, knowledge and human involvement are essential for trustworthy AI. Stakeholder engagement throughout the AI life cycle emerged as crucial, supporting a human- and multicentered framework for trustworthy AI implementation. Findings emphasize that trust in medical AI depends on providing meaningful, user-oriented information and balancing knowledge with acceptable uncertainty. Experts highlighted the importance of confidence in the tool's functionality, specifically that it performs as expected. Trustworthiness was shown to be not a feature but rather a relational process, involving humans, their expertise, and the broader social or institutional contexts in which AI tools operate. Trust is dynamic, shaped by interactions among individuals, technologies, and institutions, and ultimately centers on people rather than tools alone. Tools are evaluated based on reliability and credibility, yet trust fundamentally relies on human connections. The article underscores the development of AI tools that are not only technically sound but also ethically robust and broadly accepted by end users, contributing to more effective and equitable AI-mediated health care. Findings highlight that building AI trustworthiness in health care requires a human-centered, multistakeholder approach with diverse and inclusive engagement. To promote equity, we recommend that AI development teams involve all relevant stakeholders at every stage of the AI lifecycle-from conception, technical development, clinical validation, and real-world deployment.","author":[{"family":"Goisauf","given":"Melanie"},{"family":"Abadía","given":"Mónica"},{"family":"Akyüz","given":"Kaya"},{"family":"Bobowicz","given":"Maciej"},{"family":"Buyx","given":"Alena"},{"family":"Colussi","given":"Ilaria"},{"family":"Fritzsche","given":"Marie"},{"family":"Lekadir","given":"Karim"},{"family":"Marttinen","given":"Pekka"},{"family":"Mayrhofer","given":"Michaela"},{"family":"Mészáros","given":"János"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/71236","URL":"https://doi.org/10.2196/71236","source":"openalex"},{"id":"oa:W7119481192","type":"article-journal","title":"Artificial intelligence and multimodal imaging in orthopaedics: from technological advances to clinical translation","abstract":"The integration of multimodal medical imaging with artificial intelligence (AI) is potentially catalysing a paradigm shift in orthopaedic diagnosis and treatment, moving beyond experience-based practices toward intelligent, data-driven precision medicine. This narrative review synthesizes recent key evidence across imaging modalities and AI frameworks, and highlights the translational gap that persists between algorithmic development and real-world clinical implementation. By combining complementary information from X-ray, CT, MRI, PET, ultrasound, and biochemical data, multimodal AI overcomes the inherent limitations of single-modality approaches, enabling more comprehensive structural, functional, and metabolic assessments. Recent advances demonstrate broad applications, including accurate fracture detection and classification, differentiation of benign and malignant bone tumours, quantitative assessment of osteoarthritis, risk prediction for osteoporosis, and intelligent preoperative planning and intraoperative navigation. Moreover, multimodal AI facilitates efficacy prediction and personalised treatment decision-making, positioning future systems as AI-assisted decision-support tools that support surgeons in surgical strategy, implant design, and long-term follow-up. Nevertheless, significant challenges remain, particularly in data heterogeneity, model generalisation, interpretability, and clinical integration. Progress in constructing standardised multimodal databases, developing self-supervised and multi-task learning strategies, and strengthening ethical-regulatory frameworks will be essential for clinical translation. Ultimately, multimodal AI holds immense potential to transition from laboratory validation to routine practice, delivering safer, more efficient, and precise diagnostic and therapeutic solutions for orthopaedic patients.","author":[{"family":"Luo","given":"Guangan"},{"family":"Tan","given":"Shuanglong"},{"family":"Luo","given":"Lincong"},{"family":"Hu","given":"Konghe"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/fmed.2025.1728248","URL":"https://doi.org/10.3389/fmed.2025.1728248","source":"openalex"},{"id":"oa:W4406610264","type":"article-journal","title":"Equitable artificial intelligence for glaucoma screening with fair identity normalization","abstract":"Glaucoma is the leading cause of irreversible blindness globally. Research indicates a disproportionate impact of glaucoma on racial and ethnic minorities. Existing deep learning models for glaucoma detection might not achieve equitable performance across diverse identity groups. We developed fair identify normalization (FIN) module to equalize the feature importance across different identity groups to improve model performance equity. The optical coherence tomography (OCT) measurements were used to categorize patients into glaucoma and non-glaucoma. The equity-scaled area under the receiver operating characteristic curve (ES-AUC) was adopted to quantify model performance equity. With FIN for racial groups, the overall AUC and ES-AUC increased from 0.82 to 0.85 and 0.77 to 0.81, respectively, with the AUC for Blacks increasing from 0.77 to 0.82. With FIN for ethnic groups, the overall AUC and ES-AUC increased from 0.82 to 0.84 and 0.77 to 0.80, respectively, with the AUC for Hispanics increasing from 0.75 to 0.79.","author":[{"family":"Shi","given":"Min"},{"family":"Luo","given":"Yan"},{"family":"Tian","given":"Yu"},{"family":"Shen","given":"Lucy"},{"family":"Zebardast","given":"Nazlee"},{"family":"Eslami","given":"Mohammad"},{"family":"Kazeminasab","given":"Saber"},{"family":"Boland","given":"Michael"},{"family":"Friedman","given":"David"},{"family":"Pasquale","given":"Louis"},{"family":"Wang","given":"Mengyu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01432-5","URL":"https://doi.org/10.1038/s41746-025-01432-5","source":"openalex"},{"id":"oa:W4413833887","type":"article-journal","title":"Artificial Intelligence in Higher Education: Predictive Analysis of Attitudes and Dependency Among Ecuadorian University Students","abstract":"This study examines the relationship between attitudes toward artificial intelligence (AI) and AI dependency among Ecuadorian university students. A cross-sectional design was used, applying two validated instruments: the Artificial Intelligence Dependence Scale (DAI) and the General Attitudes Toward Artificial Intelligence Scale (GAAIS), with a sample of 540 students. Structural equation modeling (SEM) assessed how both positive and negative attitudes predict dependency levels. Results indicate a moderate level of AI dependency and an ambivalent attitudinal profile. Both attitudinal dimensions significantly predicted dependency, suggesting dual-use behaviors shaped by perceived utility and ethical concerns. Urban students reported higher dependency and greater sensitivity to AI-related risks, highlighting digital inequalities. Although the SEM model showed adequate comparative fit (CFI = 0.976; TLI = 0.973), residual indicators (RMSEA = 0.075) suggest further refinement is needed. This study contributes to underexplored Latin American contexts and emphasizes the need for equity-driven digital literacy strategies in higher education. Findings support pedagogical frameworks promoting critical thinking, ethical reasoning, and responsible AI use. The study aligns with Sustainable Development Goals 4 (Quality Education) and 10 (Reduced Inequalities), reinforcing the importance of inclusive, learner-centered approaches to AI integration.","author":[{"family":"Arce","given":"Carla"},{"family":"Gavilanes","given":"Jaime"},{"family":"Arce","given":"Edgar"},{"family":"Haro","given":"Edgar"},{"family":"Jurado","given":"Diego"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/su17177741","URL":"https://doi.org/10.3390/su17177741","source":"openalex"},{"id":"oa:W4407263148","type":"article-journal","title":"AI versus human-generated multiple-choice questions for medical education: a cohort study in a high-stakes examination","abstract":"BACKGROUND: The creation of high-quality multiple-choice questions (MCQs) is essential for medical education assessments but is resource-intensive and time-consuming when done by human experts. Large language models (LLMs) like ChatGPT-4o offer a promising alternative, but their efficacy remains unclear, particularly in high-stakes exams. OBJECTIVE: This study aimed to evaluate the quality and psychometric properties of ChatGPT-4o-generated MCQs compared to human-created MCQs in a high-stakes medical licensing exam. METHODS: A prospective cohort study was conducted among medical doctors preparing for the Primary Examination on Emergency Medicine (PEEM) organised by the Hong Kong College of Emergency Medicine in August 2024. Participants attempted two sets of 100 MCQs-one AI-generated and one human-generated. Expert reviewers assessed MCQs for factual correctness, relevance, difficulty, alignment with Bloom's taxonomy (remember, understand, apply and analyse), and item writing flaws. Psychometric analyses were performed, including difficulty and discrimination indices and KR-20 reliability. Candidate performance and time efficiency were also evaluated. RESULTS: Among 24 participants, AI-generated MCQs were easier (mean difficulty index = 0.78 ± 0.22 vs. 0.69 ± 0.23, p < 0.01) but showed similar discrimination indices to human MCQs (mean = 0.22 ± 0.23 vs. 0.26 ± 0.26). Agreement was moderate (ICC = 0.62, p = 0.01, 95% CI: 0.12-0.84). Expert reviews identified more factual inaccuracies (6% vs. 4%), irrelevance (6% vs. 0%), and inappropriate difficulty levels (14% vs. 1%) in AI MCQs. AI questions primarily tested lower-order cognitive skills, while human MCQs better assessed higher-order skills (χ² = 14.27, p = 0.003). AI significantly reduced time spent on question generation (24.5 vs. 96 person-hours). CONCLUSION: ChatGPT-4o demonstrates the potential for efficiently generating MCQs but lacks the depth needed for complex assessments. Human review remains essential to ensure quality. Combining AI efficiency with expert oversight could optimise question creation for high-stakes exams, offering a scalable model for medical education that balances time efficiency and content quality.","author":[{"family":"Law","given":"Alex"},{"family":"So","given":"Jerome"},{"family":"Lui","given":"Chun"},{"family":"Choi","given":"Yu"},{"family":"Cheung","given":"Koon"},{"family":"Hung","given":"Kevin"},{"family":"Graham","given":"Colin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12909-025-06796-6","URL":"https://doi.org/10.1186/s12909-025-06796-6","source":"openalex"},{"id":"oa:W4409516537","type":"article-journal","title":"An integrated model to evaluate the transparency in predicting employee churn using explainable artificial intelligence","abstract":"Recent studies focus on machine learning (ML) algorithms for predicting employee churn (ECn) to save probable economic loss, technology leakage, and customer and knowledge transference. However, can human resource professionals rely on algorithms for prediction? Can they decide when the process of prediction is not known? Due to the lack of interpretability, ML models' exclusive nature and growing intricacy make it challenging for field experts to comprehend these multifaceted black boxes. To address the concern of interpretability, trust and transparency of black-box predictions, this study explores the application of explainable artificial intelligence (XAI) in identifying the factors that escalate the ECn, analysing the negative impact on productivity, employee morale and financial stability. We propose a predictive model that compares the best two top-performing algorithms based on the performance metrics. Thereafter, we suggest applying an explainable artificial intelligence based on Shapley values, i.e., the Shapley Additive exPlanations approach (SHAP), to identify and compare the feature importance of top-performing algorithms logistic regression and random forest analysis on our dataset. The interpretability of the predictive outcome unboxes the predictions, enhancing trust and facilitating retention strategies.","author":[{"family":"Chaudhary","given":"Meenu"},{"family":"Gaur","given":"Loveleen"},{"family":"Chakrabarti","given":"Amlan"},{"family":"Singh","given":"Gurmeet"},{"family":"Jones","given":"Paul"},{"family":"Kraus","given":"Sascha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jik.2025.100700","URL":"https://doi.org/10.1016/j.jik.2025.100700","source":"openalex"},{"id":"oa:W4409487051","type":"article-journal","title":"Artificial intelligence-assisted multimodal imaging for the clinical applications of breast cancer: a bibliometric analysis","abstract":"BACKGROUND: Breast cancer (BC) remains a leading cause of cancer-related mortality among women globally, with increasing incidence rates posing significant public health challenges. Recent advancements in artificial intelligence (AI) have revolutionized medical imaging, particularly in enhancing diagnostic accuracy and prognostic capabilities for BC. While multimodal imaging combined with AI has shown remarkable potential, a comprehensive analysis is needed to synthesize current research and identify emerging trends and hotspots in AI-assisted multimodal imaging for BC. METHODS: This study analyzed literature on AI-assisted multimodal imaging in BC from January 2010 to November 2024 in Web of Science Core Collection (WoSCC). Bibliometric and visualization tools, including VOSviewer, CiteSpace, and the Bibliometrix R package, were employed to assess countries, institutions, authors, journals, and keywords. RESULTS: A total of 80 publications were included, revealing a steady increase in annual publications and citations, with a notable surge post-2021. China led in productivity and citations, while Germany exhibited the highest citation average. The United States demonstrated the strongest international collaboration. The most productive institution and author are Radboud University Nijmegen and Xi, Xiaoming. Publications were predominantly published in Computerized Medical Imaging and Graphics, with Qian, XJ's 2021 study on BC risk prediction under deep learning frameworks being the most influential. Keyword analysis highlighted themes such as \"breast cancer\", \"classification\", and \"deep learning\". CONCLUSIONS: AI-assisted multimodal imaging has significantly advanced BC diagnosis and management, with promising future developments. This study offers researchers a comprehensive overview of current frameworks and emerging research directions. Future efforts are expected to focus on improving diagnostic precision and refining therapeutic strategies through optimized imaging techniques and AI algorithms, emphasizing international collaboration to drive innovation and clinical translation.","author":[{"family":"Hou","given":"Chenke"},{"family":"Huang","given":"Ting"},{"family":"Hu","given":"Keke"},{"family":"Ye","given":"Zhifeng"},{"family":"Guo","given":"Junhua"},{"family":"Zhou","given":"Heran"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s12672-025-02329-1","URL":"https://doi.org/10.1007/s12672-025-02329-1","source":"openalex"},{"id":"oa:W4407450687","type":"article-journal","title":"Impact of artificial intelligence on future clinical pharmacy research and scholarship","abstract":"Abstract Almost every facet of modern biomedical research involves artificial intelligence (AI). This ACCP commentary forecasts the role of AI in clinical pharmacy research and scholarship. The potential benefits/opportunities together with the limitations/challenges of AI are reviewed for stages of the scientific method including (1) developing the research question(s), study design, and execution; (2) data analysis; and (3) reporting and dissemination of clinical pharmacy research. Benefits and opportunities of AI in clinical pharmacy research include streamlining hypothesis generation and facilitating study design, overcoming limitations of traditional statistical analysis techniques, facilitating manuscript development and dissemination, and expediting peer review. Limitations and challenges of AI include the introduction of biases in subject recruitment; generation of false information, also known as “AI hallucinations”; concern of “black box” analyses that are difficult to validate; potential legal liabilities; lack of accountability; and the need for investigators to ensure the accuracy and integrity of AI‐generated content. In summary, rapid progress of AI capabilities has great potential to revolutionize and accelerate clinical pharmacy research and scholarship; however, it is also imperative to recognize and mitigate the challenges and limitations introduced by AI.","author":[{"family":"Chan","given":"Alexandre"},{"family":"Baker","given":"William"},{"family":"Abazia","given":"Daniel"},{"family":"Bauman","given":"Jerry"},{"family":"Devane","given":"CL"},{"family":"Goodlet","given":"Kellie"},{"family":"Hall","given":"Natalie"},{"family":"Hicks","given":"JK"},{"family":"Jones","given":"Ellen"},{"family":"Lu","given":"Chi‐hua"},{"family":"Moore","given":"Donald"},{"family":"Nelson","given":"Nicholas"},{"family":"Putney","given":"Kaylee"},{"family":"Sosa","given":"Aracely"},{"family":"Trujillo","given":"Toby"},{"family":"Zhou","given":"Crystal"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/jac5.70003","URL":"https://doi.org/10.1002/jac5.70003","source":"openalex"},{"id":"oa:W4406994545","type":"article-journal","title":"Generative Artificial Intelligence Transparency in scientific writing: the GAIT 2024 guidance","abstract":"Background: Generative Artificial Intelligence (GAI) tools are increasingly used in research. At present, there is no standardised approach to reporting GAI use. We aimed to produce guidance to support authors in the use of GAI in scientific writing. Methods: A steering group of academic surgeons with experience in GAI developed draft statements for best practice in reporting GAI use. These statements were refined through iterative discussions using a nominal group technique. A broad network of surgeons and surgical researchers were invited to participate in an online consultation exercise to validate these statements by ranking using a Likert scale. A pre-planned threshold of ≥70% of participants scoring a statement ≥7 would lead to acceptance. Participants were additionally surveyed on the use, opportunities, and risks. Thematic analysis was completed using ChatGPT. Results: The steering group developed five draft statements, which were validated in the online consultation exercise by 124 participants from 46 countries. Four draft statements were accepted based on this exercise and consolidated into the final Generative AI Transparency (GAIT) guidance: (1) GAI use should be reported in a GAIT statement; (2) GAI use should be mapped using the Contributor Roles Taxonomy; (3) specific prompts used should be reported; (4) authors should retain final responsibility for their work. Example statements to be included in manuscripts include: (1) ChatGPT-4o was used in November 2024 to check and edit statistical code (formal analysis) and edit small sections of the manuscript text for clarity (writing: review & editing). Prompts used are reported in the supplement. The authors should retain final responsibility for their work; (2) No Generative Artificial Intelligence was used to produce, draft, or edit this guidance paper. Conclusion: The GAIT 2024 guidance will support transparent, structured reporting of the use of generative AI in scientific writing, supporting the integrity of research outputs.","author":[{"family":"Linder","given":"Cortland"},{"family":"Nepogodiev","given":"Dmitri"}],"issued":{"date-parts":[[2025]]},"DOI":"10.62463/surgery.134","URL":"https://doi.org/10.62463/surgery.134","source":"openalex"},{"id":"oa:W4408631719","type":"article-journal","title":"Comparing the Effectiveness of Artificial Intelligence Models in Predicting Ovarian Cancer Survival: A Systematic Review","abstract":"BACKGROUND: This systematic review investigates the use of machine learning (ML) algorithms in predicting survival outcomes for ovarian cancer (OC) patients. Key prognostic endpoints, including overall survival (OS), recurrence-free survival (RFS), progression-free survival (PFS), and treatment response prediction (TRP), are examined to evaluate the effectiveness of these algorithms and identify significant features that influence predictive accuracy. RECENT FINDINGS: A thorough search of four major databases-PubMed, Scopus, Web of Science, and Cochrane-resulted in 2400 articles published within the last decade, with 32 studies meeting the inclusion criteria. Notably, most publications emerged after 2021. Commonly used algorithms for survival prediction included random forest, support vector machines, logistic regression, XGBoost, and various deep learning models. Evaluation metrics such as area under the curve (AUC) (18 studies), concordance index (C-index) (11 studies), and accuracy (11 studies) were frequently employed. Age at diagnosis, tumor stage, CA-125 levels, and treatment-related factors were consistently highlighted as significant predictors, emphasizing their relevance in OC prognosis. CONCLUSION: ML models demonstrate considerable potential for predicting OC survival outcomes; however, challenges persist regarding model accuracy and interpretability. Incorporating diverse data types-such as clinical, imaging, and molecular datasets-holds promise for enhancing predictive capabilities. Future advancements will depend on integrating heterogeneous data sources with multimodal ML approaches, which are crucial for improving prognostic precision in OC.","author":[{"family":"Asadi","given":"Farkhondeh"},{"family":"Rahimi","given":"Milad"},{"family":"Ramezanghorbani","given":"Nahid"},{"family":"Almasi","given":"Sohrab"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/cnr2.70138","URL":"https://doi.org/10.1002/cnr2.70138","source":"openalex"},{"id":"oa:W4415171327","type":"article-journal","title":"Artificial Intelligence in Gynaecology Oncology","abstract":"Plain Language Summary Artificial intelligence (AI) is an emerging powerful technology that differs from traditional computer programs in its ability to learn from its results and enhance performance, mimicking human intelligence, hence the name. AI is already an important part of most computer‐based tasks in our daily lives. Everyday examples include internet search engines and products that provide face recognition or predict the outbreak of diseases. Research interests in AI appear to be limited to, hence constricted by, available pre‐existing information and datasets rather than addressing patients' priorities and clinical needs. The National Institute for Health and Care Excellence in England noted that current medical technologies using AI lack robust research backing and NHS patient involvement. While some AI‐based products are currently in clinical use—for example, in identifying abnormal cells in cervical smears—AI remains largely in the research phase in gynaecology oncology. Researchers have reported good results of its performance in fields such as prediction of lymph node involvement in cervical, endometrial and ovarian cancers—which are important for treatment planning, distinguishing benign from malignant pelvic masses—and cervical cancer screening in low‐ and high‐income countries. AI products have learning that is supervised and some AI modalities use technology that learns from itself. There are ethical concerns surrounding the use of AI in health care. Many of these concerns relate to the quality of data used in training AI systems, i.e., data should be inclusive so that results can be applicable in the future irrespective of race, ethnicity, socioeconomic background or place of residence. It is also not clear who should take responsibility for clinical recommendations made by AI systems: is it the doctor using it, the hospital employing the doctor or the creators of the AI product? Concerns have also been raised regarding how the roll‐out of AI might affect jobs for doctors, nurses and administration staff. AI is expected to contribute to health care in many positive ways. This can be achieved with good scrutiny and appropriate legislations to protect patients' health and privacy in addition to identifying important research and implementation areas through a collaborative partnership among investors, investigators, clinicians and patients.","author":[{"family":"Sawan","given":"Saladin"},{"family":"Eftekhari","given":"N"},{"family":"Lintonreid","given":"Kristofer"},{"family":"Wood","given":"N"},{"family":"Numan","given":"Tricia"},{"family":"Aboagye","given":"Eric"},{"family":"Angione","given":"Claudio"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/1471-0528.70005","URL":"https://doi.org/10.1111/1471-0528.70005","source":"openalex"},{"id":"oa:W4414687094","type":"article-journal","title":"Artificial intelligence in ventricular arrhythmias and sudden cardiac death: A guide for clinicians","abstract":"Sudden cardiac death (SCD) from ventricular arrhythmias (VAs) remains a leading cause of mortality worldwide. Traditional risk stratification, primarily based on left ventricular ejection fraction (LVEF) and other coarse metrics, often fails to identify a large subset of patients at risk and frequently leads to unnecessary device implantations. Advances in artificial intelligence (AI) offer new strategies to improve both long-term SCD risk prediction and near-term VAs forecasting. In this review, we discuss how AI algorithms applied to the 12-lead electrocardiogram (ECG) can identify subtle risk markers in conditions such as hypertrophic cardiomyopathy (HCM), arrhythmogenic right ventricular cardiomyopathy (ARVC), and coronary artery disease (CAD), often outperforming conventional risk models. We also explore the integration of AI with cardiac imaging, such as scar quantification on cardiac magnetic resonance (CMR) and fibrosis mapping, to enhance the identification of the arrhythmogenic substrate. Furthermore, we investigate the application of data from implantable cardioverter-defibrillators (ICDs) and wearable devices to predict ventricular tachycardia (VT) or ventricular fibrillation (VF) events before they occur, thereby advancing care toward real-time prevention. Amid these innovations, we address the medicolegal and ethical implications of AI-driven automated alerts in arrhythmia care, highlighting when clinicians can trust AI predictions. Future directions include multimodal AI fusion to personalize SCD risk assessment, as well as AI-guided VT ablation planning through imaging-based digital heart models. This review provides a comprehensive overview for general medical readers, focusing on peer-reviewed advances globally in the emerging intersection of AI, VAs, and SCD prevention.","author":[{"family":"Antoun","given":"Ibrahim"},{"family":"Li","given":"Xin"},{"family":"Abdelrazik","given":"Ahmed"},{"family":"Eldesouky","given":"Mahmoud"},{"family":"Thu","given":"Kaung"},{"family":"Ibrahim","given":"Mokhtar"},{"family":"Dhutia","given":"Harshil"},{"family":"Somani","given":"Riyaz"},{"family":"Ng","given":"GA"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.ipej.2025.09.005","URL":"https://doi.org/10.1016/j.ipej.2025.09.005","source":"openalex"},{"id":"oa:W4410448132","type":"article-journal","title":"Artificial intelligence-guided distal radius fracture detection on plain radiographs in comparison with human raters","abstract":"BACKGROUND: The aim of this study was to compare the performance of artificial intelligence (AI) in detecting distal radius fractures (DRFs) on plain radiographs with the performance of human raters. METHODS: We retrospectively analysed all wrist radiographs taken in our hospital since the introduction of AI-guided fracture detection from 11 September 2023 to 10 September 2024. The ground truth was defined by the radiological report of a board-certified radiologist based solely on conventional radiographs. The following parameters were calculated: True Positives (TP), True Negatives (TN), False Positives (FP), and False Negatives (FN), accuracy (%), Cohen's Kappa coefficient, F1 score, sensitivity (%), specificity (%), Youden Index (J Statistic). RESULTS: In total 1145 plain radiographs of the wrist were taken between 11 September 2023 and 10 September 2024. The mean age of the included patients was 46.6 years (± 27.3), ranging from 2 to 99 years and 59.0% were female. According to the ground truth, of the 556 anteroposterior (AP) radiographs, 225 cases (40.5%) had a DRF, and of the 589 lateral view radiographs, 240 cases (40.7%) had a DRF. The AI system showed the following results on AP radiographs: accuracy (%): 95.90; Cohen's Kappa: 0.913; F1 score: 0.947; sensitivity (%): 92.02; specificity (%): 98.45; Youden Index: 90.47. The orthopedic surgeon achieved a sensitivity of 91.5%, specificity of 97.8%, an overall accuracy of 95.1%, F1 score of 0.943, and Cohen's kappa of 0.901. These results were comparable to those of the AI model. CONCLUSION: AI-guided detection of DRF demonstrated diagnostic performance nearly identical to that of an experienced orthopedic surgeon across all key metrics. The marginal differences observed in sensitivity and specificity suggest that AI can reliably support clinical fracture assessment based solely on conventional radiographs.","author":[{"family":"Ramadanov","given":"Nikolai"},{"family":"John","given":"Patric"},{"family":"Hable","given":"Robert"},{"family":"Schreyer","given":"A"},{"family":"Shabo","given":"Simon"},{"family":"Prill","given":"Robert"},{"family":"Salzmann","given":"Mikhail"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s13018-025-05888-9","URL":"https://doi.org/10.1186/s13018-025-05888-9","source":"openalex"},{"id":"oa:W4407277279","type":"article-journal","title":"Harnessing artificial intelligence for predicting breast cancer recurrence: a systematic review of clinical and imaging data","abstract":"Breast cancer is a leading cause of mortality among women, with recurrence prediction remaining a significant challenge. In this context, artificial intelligence application and its resources can serve as a powerful tool in analyzing large amounts of data and predicting cancer recurrence, potentially enabling personalized medical treatment and improving the patient's quality of life. Thus, the systematic review examines the role of AI in predicting breast cancer recurrence using clinical data, imaging data, and combined datasets. Support Vector Machine (SVM) and Neural Networks, especially when applied to combined data, demonstrate strong potential in improving prediction accuracy. SVMs are effective with high-dimensional clinical data, while Neural Networks in genetic and molecular analysis. Despite these advancements, limitations such as dataset diversity, sample size, and evaluation standardization persist, emphasizing the need for further research. AI integration in recurrence prediction offers promising prospects for personalized care but requires rigorous validation for safe clinical application.","author":[{"family":"Silveira","given":"Jaqueline"},{"family":"Silva","given":"Alexandre"},{"family":"Lima","given":"Mariana"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s12672-025-01908-6","URL":"https://doi.org/10.1007/s12672-025-01908-6","source":"openalex"},{"id":"oa:W4408588988","type":"article-journal","title":"Leveraging Artificial Intelligence and Radiomics for Improved Nasopharyngeal Carcinoma Prognostication","abstract":"INTRODUCTION: Nasopharyngeal carcinoma (NPC) typically presents as advanced disease due to the lack of significant symptoms in the early stages. Accurate prognostication is therefore challenging as current methods based on anatomical staging often lack the granularity to differentiate between patients with differing prognoses. This study investigates the potential of radiomics to improve the prediction of locoregional recurrence (LRR) and overall survival in patients with NPC. METHODS: Radiomic features were extracted from radiotherapy planning CT scans for 294 NPC patients divided into training (n = 147) and validation (n = 147) sets. A feature selection step utilising feature clustering and mutual information classifier to select six key radiomic features was employed to reduce redundancy and improve interpretability. Models were trained using clinical data, radiomic features, and these in combination to predict 2-year LRR, with performance assessed on the left-out independent validation set. RESULTS: Combining radiomic features with clinical data resulted in the best performance for predicting 2-year LRR (Area Under the Curve, AUC 0.76) compared to prediction using clinical or radiomic features alone (mean AUC 0.56 and 0.57, respectively). Risk stratification based on the combined model was significant for LRR-free survival and overall survival (p < 0.01). Key radiomic features included tumour size, intensity distribution, overall textural patterns, and distribution of fine and coarse textured regions. DISCUSSION: Radiomics holds promise for improving NPC risk stratification, potentially allowing for personalised treatment strategies. The most important radiomics feature, maximum 2D diameter, suggests a need to reconsider tumour size as a prognostic criterion despite its current exclusion from TNM staging. Larger prospective studies are needed to validate these findings.","author":[{"family":"Shannon","given":"Nicholas"},{"family":"Lyer","given":"Narayanan"},{"family":"Chua","given":"Melvin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/cam4.70706","URL":"https://doi.org/10.1002/cam4.70706","source":"openalex"},{"id":"oa:W4414232095","type":"article-journal","title":"Exploring the use and impact of artificial intelligence in higher education in Africa","abstract":"Artificial intelligence is transforming higher education, but its adoption in developing countries has been overlooked. This study analysed this gap through a comprehensive systematic literature review, analysing AI's use and impact in higher education. A total of 1,521 articles were identified through databases such as Scopus, Web of Science, EBSCOhost, and Google Scholar, with 63 selected for review. The findings indicate that AI is used in various higher education applications, including tutoring, administrative tasks, instruction, curriculum development, and facilitating new skills acquisition. Many benefits were derived from its use, such as multitasking, ease of workloads, and customised learning. However, challenges exist, including abuse by learners, resource limitations, skill gaps, data security concerns, and ethical issues. This study provides a unique conceptual analysis of AI's impact on higher education institutions, focusing on developing nations. Surprisingly, few explicit studies exist on AI adoption in higher education in these countries. The findings will inform policymakers, educators, and stakeholders about AI's potential to reshape higher education in developing countries, guiding strategic efforts to harness benefits while mitigating risks.","author":[{"family":"Pasipamire","given":"Notice"},{"family":"Chigwada","given":"Josiline"},{"family":"Maturure","given":"Rosemary"}],"issued":{"date-parts":[[2025]]},"DOI":"10.33902/jpsp.202532046","URL":"https://doi.org/10.33902/jpsp.202532046","source":"openalex"},{"id":"oa:W4412190343","type":"article-journal","title":"Unlocking the Potential of Artificial Intelligence in Pharma Research and Development: Insights from Investor and Researcher Perspectives","abstract":"The integration of artificial intelligence into drug discovery processes represents a major innovation in pharmaceutical research and development. This study investigates the role of AI investments in enhancing research efficiency, addressing implementation challenges, and shaping stakeholder perspectives. Via a structured explanatory research design, the study applies a quantitative methodology based on survey data collected from researchers, investors, and pharmaceutical executives across the USA and United Kingdom. The questionnaire examined respondents’ experiences with artificial intelligence tools, investment patterns, and perceived research outcomes. Statistical methods such as logistic regression and chi-square tests were employed to analyze correlations between investment strategies and research efficiency. Findings indicate that while artificial intelligence improves productivity – in predictive modeling and data analysis – barriers such as high infrastructure costs, inadequate training, and regulatory uncertainty persist. Notably, 70% of participants plan to increase AI investments within the next five years, and 80% regard artificial intelligence as essential or very important to the future of drug discovery. However, successful implementation appears to correlate with firm size and access to technical resources, suggesting disparities in AI readiness across the industry. Recommendations include expanding artificial intelligence training programs, strengthening infrastructure, and fostering closer collaboration between investors and researchers. Ethical considerations, including data privacy and regulatory compliance, are also emphasized. The pilot study provides foundational insights for a full-scale investigation and offers practical guidance for optimizing artificial intelligence integration in pharmaceutical research and development.","author":[{"family":"Kritikos","given":"Jacob"},{"family":"Sarantopoulos","given":"Andreas"},{"family":"Roumeliotis","given":"Anastasios"},{"family":"Vasiliades","given":"Julia"},{"family":"Matsinas","given":"Ioannis"}],"issued":{"date-parts":[[2025]]},"DOI":"10.61093/hem.2025.2-01","URL":"https://doi.org/10.61093/hem.2025.2-01","source":"openalex"},{"id":"oa:W4414043654","type":"article-journal","title":"Artificial intelligence in allergy and immunology: Recent developments, implementation challenges, and the road toward clinical impact","abstract":"Artificial intelligence (AI) is increasingly recognized for its capacity to transform medicine. While publications applying AI in allergy and immunology have increased in number, clinical implementation substantially lags behind other specialties. By mid-2024, over 1,000 US Food and Drug Administration-approved AI-enabled medical devices existed, but none specifically addressed allergy and immunology. This gap partly reflects the field's limited reliance on imaging, which facilitated early AI breakthroughs in radiology and pathology. This narrative review examines recent AI developments, including large language models and AI agents, evaluating their applicability to allergy and immunology practice. We analyze current and potential applications, emphasizing those demonstrating clinical value while identifying implementation barriers amplified by allergic diseases' unique complexities, including data privacy concerns, bias, reliability constraints, and evolving regulatory frameworks. To bridge the persistent research-to-implementation gap, we propose a 6-point road map: (1) prioritize impactful applications, (2) define clinically relevant benchmarks, (3) enforce rigorous governance, (4) transition to operationalization, (5) promote clinical adoption through trustworthy AI, and (6) establish life cycle management. This road map builds on established implementation frameworks while incorporating critical field-specific considerations unique to allergy and immunology. Through this approach, we provide a perspective for advancing AI in allergy and immunology from academic promise to tangible clinical benefit.","author":[{"family":"Breugel","given":"Merlijn"},{"family":"Greenhawt","given":"Matt"},{"family":"Eguíluzgracia","given":"Ibon"},{"family":"Torres","given":"Marı́a"},{"family":"Anagnostou","given":"Aikaterini"},{"family":"Koppelman","given":"Gerard"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jaci.2025.08.022","URL":"https://doi.org/10.1016/j.jaci.2025.08.022","source":"openalex"},{"id":"oa:W4416721771","type":"article-journal","title":"The Role of Artificial Intelligence in Pharmacy Practice and Patient Care: Innovations and Implications","abstract":"Artificial Intelligence (AI) is reshaping pharmacy practice by enhancing decision-making, personalizing therapy, and improving medication safety. AI applications now span drug discovery, clinical decision support, and adherence monitoring. This narrative review explores key innovations, practical applications, and the implications of AI integration in pharmacy practice, with a focus on emerging tools, pharmacist roles, and ethical considerations. The review was conducted using literature from PubMed/MEDLINE, Scopus, Web of Science, and Google Scholar. Thematic synthesis included AI-based drug interaction checkers, Clinical Decision Support Systems (CDSS), telepharmacy, pharmacogenomics, and predictive analytics. AI enhances clinical decision-making, reduces medication errors, and supports precision medicine. AI tools support pharmacists and healthcare professionals in optimizing care. However, data privacy, algorithmic bias, and workflow integration continue to pose challenges. AI holds transformative potential in pharmacy, though its integration requires overcoming ethical and workflow-related challenges. Ethical and regulatory vigilance, coupled with pharmacist training and interdisciplinary collaboration, is essential to realize the full potential of AI.","author":[{"family":"Alam","given":"Aftab"},{"family":"Shah","given":"Syed"},{"family":"Rabbani","given":"Syed"},{"family":"Eltanani","given":"Mohamed"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/biomedinformatics5040065","URL":"https://doi.org/10.3390/biomedinformatics5040065","source":"openalex"},{"id":"oa:W4416509484","type":"article-journal","title":"Recent Advances in the Application of Artificial Intelligence in Microalgal Cultivation","abstract":"Microalgae are unicellular, industrially important organisms that are used extensively in a range of industrial, environmental, and biorefinery applications. They can produce lipids, carbohydrates, and possibly additional vital bioactive substances. The increasing popularity of artificial intelligence (AI) in microalgae research can be attributed to its algorithms’ ability to manage the complexity of unexpected biosystems. In the case of microalgae-based biorefineries, AI technology can also help uncover system dynamics and uncertainties, provide helpful predictive analytics, and expedite the optimisation process. AI is used in microalgal cultivation to optimise carbon capture, biomass production, and conditions for growth. Additionally, it is employed for genome editing, automated monitoring, and lipid accumulation enhancement. However, its uses are broad and constantly growing. Furthermore, critical environmental parameters in microalgae culture, including temperature, light intensity, pH, dissolved oxygen, and nutrient levels, may be continually monitored and controlled by internet of things (IoT)-based devices. This review comprehensively summarises the latest applications of AI technology in the field of microalgae cultivation and the role of IoT-based automatic control.","author":[{"family":"Rayamajhi","given":"Vijay"},{"family":"Hussain","given":"Mudasir"},{"family":"Shin","given":"Hyun‐woung"},{"family":"Jung","given":"Sang"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/pr13123764","URL":"https://doi.org/10.3390/pr13123764","source":"openalex"},{"id":"oa:W7160167964","type":"article-journal","title":"Artificial Intelligence and Big Data in Accounting: The Case of Commercial Banks in Jordan","abstract":"The integration of Artificial Intelligence (AI) and Big Data analytics is transforming the accounting and financial practices of commercial banks, particularly in emerging markets like Jordan. As the banking sector shifts from a product-centered to a customer-focused approach, the ability to analyze large volumes of data becomes essential for gaining insights into customer behavior, preferences, and financial patterns. This research explores how AI and Big Data technologies are being adopted by Jordanian commercial banks to enhance accounting processes, improve decision-making, reduce customer churn, and personalize services. By leveraging predictive modeling, segmentation, and behavioral analytics, banks can refine credit risk assessments, optimize marketing strategies, and improve customer satisfaction. Despite the promising benefits, challenges related to data quality, technological infrastructure, and ethical concerns remain. This study provides a contextual analysis of how AI and Big Data are reshaping the accounting landscape in Jordan’s banking sector and offers recommendations for their effective implementation.","author":[{"family":"Bader","given":"Ayman"},{"family":"Qtaish","given":"Attalah"},{"family":"Odeh","given":"Khalel"},{"family":"Sad","given":"Hanadi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.15849/zjjb.v1i03.49","URL":"https://doi.org/10.15849/zjjb.v1i03.49","source":"openalex"},{"id":"oa:W4409331282","type":"article-journal","title":"Bridging the Past and Future of Clinical Data Management: The Transformative Impact of Artificial Intelligence","abstract":"Abstract: Effective clinical data management is fundamental to clinical research and regulatory submissions. Modern clinical trials have increasingly adopted web-based electronic data capture (EDC) systems, which enhance data collection efficiency but introduces challenges in data integration and quality. This scoping review explores the transformative role of artificial intelligence and machine learning in evolving CDM into clinical data science. In the review, we followed the PRISMA-ScR guidelines and analyzed the literature from 2008 to 2025 using Scopus, Web of Science, and PubMed databases. A total of 26 papers were included and categorized into those related to clinical data management, natural language processing, and general artificial intelligence/machine learning adoption in clinical data management. The integration shows promise in enhancing data analysis, automating data cleaning, and predicting critical outcomes. The key emerging trends include risk-based quality monitoring, blockchain technology, remote monitoring, and patient-centric approaches involving wearables and mobile applications. The results clearly indicate a substantial increase in data volume in Phase III trials, underscoring the need for advanced technologies natural language processing offers significant potential in interpreting unstructured text data, thereby improving the clinical data management processes. The review concludes on different artificial intelligence/machine learning techniques like natural language processing, predictive analytics, and automation technologies, and their applications in improving data quality and streamlining clinical data workflows. Keywords: clinical data management, artificial intelligence, machine learning, natural language processing, clinical data science, electronic data capture","author":[{"family":"Musik","given":"Szymon"},{"family":"Sasin-Kurowska","given":"Joanna"},{"family":"Pańczyk","given":"Mariusz"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2147/oajct.s509921","URL":"https://doi.org/10.2147/oajct.s509921","source":"openalex"},{"id":"oa:W4411606457","type":"article-journal","title":"Responsible scaling of artificial intelligence in healthcare: standardization meets customization","abstract":"Abstract Organizations across the globe are progressively investing in artificial intelligence (AI) innovations to meet today’s healthcare challenges. Meanwhile, public policy increasingly emphasizes the need for these innovations to be ‘scaled’. As scholars emphasize, scaling innovations is never just ‘more of the same’, but requires adapting innovations to local contexts. In this perspective paper, we aim to explore and draw attention to the tensions and possible alignments between standardization and customization that should lead to a responsible scaling of AI in healthcare. We approach responsible scaling building on the notion of socio-technical configurations. Configurations are unique assemblies of technological and non-technological components, including human factors, integrated in different ways to meet particular local requirements. We explore how conceptualizing AI tools and the broader socio-technical ecosystems in which they are integrated as configurations can offer a framework for envisioning possible pathways for responsibly scaling AI. We contend that standardization and customization can be employed synergistically within AI configurations. Standardization can be an important driver of innovation at the level of configurational components of healthcare AI, as well as the interoperability between these components. Thereby, standardization can expand the configurational options that local AI implementations can draw from and lay a foundation for local customization of healthcare AI ecosystems at the architectural level. Accordingly, we propose key considerations for innovators and policymakers to boost the configurability of healthcare AI, and discuss the need for, and challenges of shaping of healthcare AI configurations at the local scale.","author":[{"family":"Lukkien","given":"DRM"},{"family":"Nap","given":"Henk"},{"family":"Peine","given":"Alexander"},{"family":"Minkman","given":"Mirella"},{"family":"Moors","given":"Ellen"},{"family":"Boon","given":"Wouter"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s10676-025-09842-5","URL":"https://doi.org/10.1007/s10676-025-09842-5","source":"openalex"},{"id":"oa:W4416784217","type":"article-journal","title":"Artificial intelligence in cardiovascular diagnostics: a systematic review and descriptive analysis of clinical applications and diagnostic performance","abstract":"INTRODUCTION: Artificial intelligence (AI) is rapidly transforming cardiovascular diagnostics by enhancing early disease detection, risk stratification, and clinical decision-making. Recent studies have shown the effectiveness of AI in analyzing electrocardiograms (ECGs) and cardiac imaging, predicting adverse cardiovascular outcomes, and enabling personalized care. AI models have also demonstrated potential in community-based and population-specific applications, signaling a shift toward data-driven, precision cardiovascular medicine. STUDY OBJECTIVE: To systematically evaluate the clinical applications and diagnostic performance of AI in the detection and risk assessment of cardiovascular diseases. METHODS: A systematic search was conducted in PubMed, Google Scholar, and ScienceDirect for English-language articles published between January 2020 and June 2025. Eligible studies included randomized controlled trials and observational designs with free full-text availability. QUADAS-2 tool for diagnostic accuracy studies and the PROBAST for prognostic or risk-prediction models, were used for quality assessment. Of 30 eligible articles, 14 high-quality studies were included. RESULTS: Across the included studies, AI-based diagnostic tools demonstrated consistently high performance for cardiovascular disease detection. Reported area under the curve (AUC) values ranged from 0.804 to 0.991, with most ≥ 0.88, indicating robust discriminative accuracy across diverse modalities including ECG analysis, cardiac imaging, and predictive risk modeling. Although formal pooling was not conducted due to methodological heterogeneity, the descriptive synthesis highlighted strong and consistent performance in applications such as heart failure, coronary artery disease, and arrhythmia detection. Variability in study design and reporting limited direct comparison, but overall trends support the potential of AI systems to enhance diagnostic precision across cardiovascular contexts. CONCLUSION: AI-driven diagnostic tools demonstrate consistently high accuracy across cardiovascular applications, supporting their potential to complement clinical decision-making. However, variability in study design and limited external validation highlight the need for standardized evaluation and transparent reporting before widespread clinical integration.","author":[{"family":"Niazai","given":"Ahsanullah"},{"family":"Jamil","given":"Hajra"},{"family":"Hameed","given":"Maryam"},{"family":"Sheikh","given":"Sarah"},{"family":"Nisar","given":"Mah"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12872-025-05327-x","URL":"https://doi.org/10.1186/s12872-025-05327-x","source":"openalex"},{"id":"oa:W4410990229","type":"article-journal","title":"Fibromyalgia: one year in review 2025","abstract":"Fibromyalgia (FM) is a chronic syndrome characterised by widespread pain, high prevalence, and a significant impact on quality of life. Despite extensive research, its pathogenesis and treatment remain only partially understood, driving continued investigation throughout 2024. Dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis and sympathetic nervous system has been linked to chronic stress responses and neuroinflammation, with neuroimaging and preclinical studies confirming altered pain and stress processing. Low-grade inflammation and metabolic disturbances, including cytokine imbalance and increased adipose tissue infiltration, further exacerbate symptoms. Alterations in the gut microbiota contribute to immune and emotional dysregulation. MRI studies continue to reveal brain changes that differentiate FM from other chronic pain disorders. Multi-omics approaches, including transcriptomic and metabolomic analyses, show promise as diagnostic biomarkers. Mitochondrial dysfunction also emerges as a key factor, since impaired energy metabolism seems to correlate with symptom severity. From a clinical perspective, recent studies have explored under-recognised aspects of FM, such as sexual and cognitive dysfunction, the role of gender, environmental exposures, and the disease's impact on relationships and family life. The differential diagnosis of FM and long COVID has ignited discussion about potential shared mechanisms. Conversely, residual pain in inflammatory diseases remains insufficiently addressed. Therapeutically, non-pharmacological strategies, particularly physical activity and psychosocial interventions, remain fundamental. Emerging areas such as non-invasive neuromodulation, psychedelic therapies, and the integration of technologies like virtual reality and artificial intelligence are opening new frontiers in treatment, patient care, and research. These advances underscore the multifactorial nature of FM and the need for personalised, interdisciplinary approaches.","author":[{"family":"Iannuccelli","given":"Cristina"},{"family":"Favretti","given":"Martina"},{"family":"Dolcini","given":"Giulio"},{"family":"Carlo","given":"Marco"},{"family":"Pellegrino","given":"Greta"},{"family":"Bazzichi","given":"Laura"},{"family":"Atzeni","given":"Fabiola"},{"family":"Lucini","given":"Daniela"},{"family":"Varassi","given":"Giustino"},{"family":"Leoni","given":"Matteo"},{"family":"Fornasari","given":"Diego"},{"family":"Conti","given":"Fabrizio"},{"family":"Salaffi","given":"Fausto"},{"family":"Sarziputtini","given":"Piercarlo"},{"family":"Franco","given":"Manuela"}],"issued":{"date-parts":[[2025]]},"DOI":"10.55563/clinexprheumatol/buhd2z","URL":"https://doi.org/10.55563/clinexprheumatol/buhd2z","source":"openalex"},{"id":"oa:W7131698947","type":"article-journal","title":"Artificial Intelligence agents for biological research: a survey","abstract":"The rapid growth of biological data and experimental complexity has motivated increasing interest in artificial intelligence (AI) systems that extend beyond static prediction toward autonomous reasoning and action. While recent computational models achieve strong predictive performance, they largely operate as passive tools within human-driven research workflows. In contrast, AI agents integrate reasoning, planning, tool invocation, and feedback-driven refinement, enabling more adaptive and interactive forms of biological analysis. This survey provides a systematic synthesis of recent progress in biological AI agents by reviewing over 100 representative studies across clinical analytics, molecular and drug design, multi-omics analysis, and knowledge discovery. We introduce a unified 5D taxonomy that organizes existing work along task domains, system architectures, interaction modes, evaluation strategies, and resource integration. Building on this framework, we analyze common design patterns, highlight emerging capabilities enabled by agentic paradigms, and identify key open challenges, including reliability, privacy, scalability, and standardized evaluation. Collectively, this survey clarifies the conceptual and methodological landscape of biological AI agents and outlines directions toward more robust, transparent, and collaborative agent-based systems for biological research. To serve as a living resource for the community, we curated a GitHub repository that includes resources and benchmark summaries, available at https://github.com/MineSelf2016/biological_agents_survey.","author":[{"family":"Qi","given":"Cong"},{"family":"Wang","given":"Wenbo"},{"family":"Jiang","given":"Siqi"},{"family":"Liu","given":"Q"},{"family":"Song","given":"Xun"},{"family":"Fang","given":"Hanzhang"},{"family":"Wei","given":"Zhi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1093/bib/bbag075","URL":"https://doi.org/10.1093/bib/bbag075","source":"openalex"},{"id":"oa:W4414103623","type":"article-journal","title":"Artificial intelligence in medical imaging empowers precision neoadjuvant immunochemotherapy in esophageal squamous cell carcinoma","abstract":"Neoadjuvant immunochemotherapy (nICT) has demonstrated significant potential in improving pathological response rates and survival outcomes for patients with locally advanced esophageal squamous cell carcinoma (ESCC). However, substantial interindividual variability in therapeutic outcomes highlights the urgent need for more precise predictive tools to guide clinical decision-making. Traditional biomarkers remain limited in both predictive performance and clinical feasibility. In recent years, the application of artificial intelligence (AI) in medical imaging has expanded rapidly. By incorporating voxel-level feature maps, the combination of radiomics and deep learning enables the extraction of rich textural, morphological, and microstructural features, while autonomously learning high-level abstract representations from clinical CT images, thereby revealing biological heterogeneity that is often imperceptible to conventional assessments. Leveraging these high-dimensional representations, AI models can provide more accurate predictions of nICT response. Future advancements in foundation models, multimodal integration, and dynamic temporal modeling are expected to further enhance the generalizability and clinical applicability of AI. AI-powered medical imaging is poised to support all stages of perioperative management in ESCC, playing a pivotal role in high-risk patient identification, dynamic monitoring of therapeutic response, and individualized treatment adjustment, thereby comprehensively advancing precision nICT.","author":[{"family":"Fu","given":"Jia"},{"family":"Huang","given":"Xiao‐ying"},{"family":"Fang","given":"Mengjie"},{"family":"Feng","given":"Xinliang"},{"family":"Zhang","given":"Xu"},{"family":"Xie","given":"Xuebin"},{"family":"Zheng","given":"Zhuozhao"},{"family":"Dong","given":"Di"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/jitc-2025-012468","URL":"https://doi.org/10.1136/jitc-2025-012468","source":"openalex"},{"id":"oa:W4409204055","type":"manuscript","title":"Artificial Intelligence in Orthopedic Medical Education: A Comprehensive Review of Emerging Technologies and Their Applications","abstract":"Integrating artificial intelligence (AI) and mixed reality (MR) into orthopedic education has transformed learning. This review examines AI-powered platforms like Microsoft HoloLens, Apple Vision Pro, and HTC Vive Pro, which enhance anatomical visualization, surgical simulation, and clinical decision-making. These technologies improve spatial understanding of musculoskeletal structures, refine procedural skills with haptic feedback, and personalize learning through AI-driven adaptive algorithms. Generative AI tools like ChatGPT further support knowledge retention and provide evidence-based insights on orthopedic topics. AI-enabled platforms and generative AI tools help address challenges in standardizing orthopedic education. However, we still face many barriers that relate to standardizing data, algorithm evaluation, ethics, and the curriculum. AI is used in preoperative planning and predictive analytics in the postoperative period that bridges theory and practice. AI and MR are key to supporting innovation and scalability in orthopedic education. However, technological innovation relies on collaborative partnerships to develop equitable, evidence-informed practices that can be implemented in orthopedic education. For sustained impact, innovation must be aligned with pedagogical theories and principles. We believe that orthopedic medical educator's future critical role will be to enhance the next generation of competent clinicians.","author":[{"family":"Sporn","given":"Kyle"},{"family":"Kumar","given":"Rahul"},{"family":"Paladugu","given":"Phani"},{"family":"Ong","given":"Joshua"},{"family":"Sekhar","given":"Tejas"},{"family":"Vaja","given":"Swapna"},{"family":"Hage","given":"Tamer"},{"family":"Waisberg","given":"Ethan"},{"family":"Gowda","given":"Chirag"},{"family":"Jagadeesan","given":"Ram"},{"family":"Zaman","given":"Nasif"},{"family":"Tavakkoli","given":"Alireza"}],"issued":{"date-parts":[[2025]]},"DOI":"10.20944/preprints202504.0306.v1","URL":"https://doi.org/10.20944/preprints202504.0306.v1","source":"preprints"},{"id":"oa:W4407922620","type":"article-journal","title":"Health Care Professionals’ Concerns About Medical AI and Psychological Barriers and Strategies for Successful Implementation: Scoping Review","abstract":"BACKGROUND: The rapid progress in the development of artificial intelligence (AI) is having a substantial impact on health care (HC) delivery and the physician-patient interaction. OBJECTIVE: This scoping review aims to offer a thorough analysis of the current status of integrating AI into medical practice as well as the apprehensions expressed by HC professionals (HCPs) over its application. METHODS: This scoping review used the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines to examine articles that investigated the apprehensions of HCPs about medical AI. Following the application of inclusion and exclusion criteria, 32 of an initial 217 studies (14.7%) were selected for the final analysis. We aimed to develop an attitude range that accurately captured the unfavorable emotions of HCPs toward medical AI. We achieved this by selecting attitudes and ranking them on a scale that represented the degree of aversion, ranging from mild skepticism to intense fear. The ultimate depiction of the scale was as follows: skepticism, reluctance, anxiety, resistance, and fear. RESULTS: In total, 3 themes were identified through the process of thematic analysis. National surveys performed among HCPs aimed to comprehensively analyze their current emotions, worries, and attitudes regarding the integration of AI in the medical industry. Research on technostress primarily focused on the psychological dimensions of adopting AI, examining the emotional reactions, fears, and difficulties experienced by HCPs when they encountered AI-powered technology. The high-level perspective category included studies that took a broad and comprehensive approach to evaluating overarching themes, trends, and implications related to the integration of AI technology in HC. We discovered 15 sources of attitudes, which we classified into 2 distinct groups: intrinsic and extrinsic. The intrinsic group focused on HCPs' inherent professional identity, encompassing their tasks and capacities. Conversely, the extrinsic group pertained to their patients and the influence of AI on patient care. Next, we examined the shared themes and made suggestions to potentially tackle the problems discovered. Ultimately, we analyzed the results in relation to the attitude scale, assessing the degree to which each attitude was portrayed. CONCLUSIONS: The solution to addressing resistance toward medical AI appears to be centered on comprehensive education, the implementation of suitable legislation, and the delineation of roles. Addressing these issues may foster acceptance and optimize AI integration, enhancing HC delivery while maintaining ethical standards. Due to the current prominence and extensive research on regulation, we suggest that further research could be dedicated to education.","author":[{"family":"Árvai","given":"Nóra"},{"family":"Katonai","given":"Gellért"},{"family":"Meskó","given":"Bertalan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/66986","URL":"https://doi.org/10.2196/66986","source":"openalex"},{"id":"oa:W4411156068","type":"article-journal","title":"Artificial Intelligence in Laryngeal Cancer Detection: A Systematic Review and Meta-Analysis","abstract":"(1) Background: The early detection of laryngeal cancer is crucial for achieving superior patient outcomes and preserving laryngeal function. Artificial intelligence (AI) methodologies can expedite the triage of suspicious laryngeal lesions, thereby diminishing the critical timeframe required for clinical intervention. (2) Methods: We included all studies published up to February 2025. We conducted a systematic search across five major databases: MEDLINE, EMCARE, EMBASE, PubMed, and the Cochrane Library. We included 15 studies, with a total of 17,559 patients. A risk of bias assessment was performed using the QUADAS-2 tool. We conducted data synthesis using the Meta Disc 1.4 program. (3) Results: A meta-analysis revealed that AI demonstrated high sensitivity (78%) and specificity (86%), with a Pooled Diagnostic Odds Ratio of 53.77 (95% CI: 27.38 to 105.62) in detecting laryngeal cancer. The subset analysis revealed that CNN-based AI models are superior to non-CNN-based models in image analysis and lesion detection. (4) Conclusions: AI can be used in real-world settings due to its diagnostic accuracy, high sensitivity, and specificity.","author":[{"family":"Alabdalhussein","given":"Ali"},{"family":"Al-Khafaji","given":"Mustafa"},{"family":"Al-Busairi","given":"Rusul"},{"family":"Al-Dabbagh","given":"Shahad"},{"family":"Khan","given":"Waleed"},{"family":"Anwar","given":"Fahim"},{"family":"Raheem","given":"Taghreed"},{"family":"Elkrim","given":"Mohammed"},{"family":"Sahota","given":"Raguwinder"},{"family":"Mair","given":"Manish"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/curroncol32060338","URL":"https://doi.org/10.3390/curroncol32060338","source":"openalex"},{"id":"oa:W4411469847","type":"article-journal","title":"Leveraging In Silico and Artificial Intelligence Models to Advance Drug Disposition and Response Predictions Across the Lifespan","abstract":"Incorporating inter-individual differences in drug disposition and responses is essential for ensuring the safe and effective use of drugs in real-world patients. Despite ongoing efforts, lower participation of children, older individuals, pregnant and breastfeeding women, postmenopausal women, and people with disease states and disabilities in drug clinical trials is frequent, and it requires multifaceted strategies and tools to evaluate drug exposure and responses in broad populations. The availability of modeling and simulation tools, such as physiologically based pharmacokinetic (PBPK) and quantitative systems pharmacology/toxicology (QSP/QST) modeling, enables the application of virtual populations that reflect the differences in drug disposition and responses for disease states and different stages of the lifespan. These models integrate clinical trial and real-world data (RWD) to predict drug exposure, efficacy, and safety. Additionally, machine learning (ML) and artificial intelligence (AI) offer powerful tools for analyzing large datasets and identifying key physiological determinants of drug response across the lifespan. This review discusses the application of in silico and AI models to advance the prediction of drug exposure and responses across the lifespan, including examples of virtual populations in PBPK and QSP/QST models. A case study on QST modeling for drug-induced liver injury (DILI) in postmenopausal women is presented, along with opportunities and challenges in applying AI for modeling physiological determinants of drug dosing in individuals ranging in age from 12 to > 80 years old in drug development.","author":[{"family":"Yang","given":"Kyunghee"},{"family":"González","given":"Daniel"},{"family":"Woodhead","given":"Jeffrey"},{"family":"Bhargava","given":"Pallavi"},{"family":"Ramanathan","given":"Murali"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/cts.70272","URL":"https://doi.org/10.1111/cts.70272","source":"openalex"},{"id":"oa:W4410974516","type":"article-journal","title":"Evaluating the Application of Artificial Intelligence and Ambient Listening to Generate Medical Notes in Vitreoretinal Clinic Encounters","abstract":"Purpose: Analyze the application of large language models (LLM) to listen to and generate medical documentation in vitreoretinal clinic encounters. Subjects: Two publicly available large language models, Google Gemini 1.0 Pro and Chat GPT 3.5. Methods: Patient-physician dialogues simulating vitreoretinal clinic scenarios were scripted to simulate real-world encounters and recorded for standardization. Two artificial intelligence engines were given the audio files to transcribe the dialogue and produce medical documentation of the encounters. Similarity of the dialogue and LLM transcription was assessed using an online comparability tool. A panel of practicing retina specialists evaluated each generated medical note. Main Outcome Measures: The number of discrepancies and overall similarity of LLM text compared to scripted patient-physician dialogues, and scoring on the physician documentation quality instrument-9 (PDQI-9) of each medical note by five retina specialists. Results: On average, the documentation produced by AI engines scored 81.5% of total possible points in documentation quality. Similarity between pre-formed dialogue scripts and transcribed encounters was higher for ChatGPT (96.5%) compared to Gemini (90.6%, p<0.01). The mean total PDQI-9 score among all encounters from ChatGPT 3.5 (196.2/225, 87.2%) was significantly greater than Gemini 1.0 Pro (170.4/225, 75.7%, p=0.002). Conclusion: The authors report the aptitude of two popular LLMs (ChatGPT 3.5 and Google Gemini 1.0 Pro) in generating medical notes based on audio recordings of scripted vitreoretinal clinical encounters using a validated medical documentation tool. Artificial intelligence can produce quality vitreoretinal clinic encounter medical notes after listening to patient-physician dialogues despite case complexity and missing encounter variables. The performance of these engines was satisfactory but sometimes included fabricated information. We demonstrate the potential utility of LLMs in reducing the documentation burden on physicians and potentially streamlining patient care.","author":[{"family":"Patel","given":"Neeket"},{"family":"Lacher","given":"Corey"},{"family":"Huang","given":"Alan"},{"family":"Kolomeyer","given":"Anton"},{"family":"Bavinger","given":"JC"},{"family":"Carroll","given":"Robert"},{"family":"Kim","given":"Benjamin"},{"family":"Tsui","given":"Jonathan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2147/opth.s513633","URL":"https://doi.org/10.2147/opth.s513633","source":"openalex"},{"id":"oa:W7117585964","type":"article-journal","title":"Physician Perspectives on the Impact of Artificial Intelligence on the Therapeutic Relationship in Mental Health Care: Qualitative Study","abstract":"Background: The therapeutic relationship is a professional partnership between clinicians and patients that supports open communication and clinical decision-making. This relationship is critical to the delivery of effective mental health care. The integration of artificial intelligence (AI) into mental health care has the potential to support accessibility and personalized care; however, little is known about how AI might affect the dynamics of the therapeutic relationship. Objective: This study aimed to ascertain how physicians anticipate AI tools will impact the therapeutic relationship in mental health care. Methods: We conducted 42 in-depth interviews with psychiatrists and family medicine practitioners to investigate physician perceptions regarding the impact of AI on mental health care. Results: Physicians identified several disruptions from AI use, noting that these tools could impact the dyad of the patient-physician relationship in ways that are both positive and negative. The main themes that emerged included potential disruptions to the therapeutic relationship, shifts in shared decision-making dynamics, and the importance of transparent AI use. Participants suggested that AI tools could create efficiencies that allow for relationship building as well as help avoid issues with miscommunication during psychotherapeutic interactions. However, they also expressed concerns that AI tools might not adequately capture aspects of the therapeutic relationship, such as empathy, that are vital to mental health care. Physicians also raised issues related to the impact that AI tools will have on maintaining relationships with patients. Conclusions: As AI applications become increasingly integrated into mental health care, it is crucial to assess how this integration may support or disrupt the therapeutic relationship. Physician acceptance of emerging AI tools may be highly dependent on how well the human elements of mental health care are preserved.","author":[{"family":"Weir","given":"Isabel"},{"family":"Stroud","given":"Austin"},{"family":"Stout","given":"Jeremiah"},{"family":"Barry","given":"Barbara"},{"family":"Athreya","given":"Arjun"},{"family":"Bobo","given":"William"},{"family":"Sharp","given":"Richard"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/81970","URL":"https://doi.org/10.2196/81970","source":"openalex"},{"id":"oa:W4411327768","type":"article-journal","title":"Artificial intelligence: a promising tool for the clinical cardiologist","abstract":"INTRODUCTION: Artificial intelligence (AI) has emerged as a revolutionary technology that is changing clinical practice, including management of patients with cardiovascular diseases. AREAS COVERED: From a clinical practice perspective, this manuscript reviews the impact of AI on the management of cardiovascular diseases, and current challenges and opportunities. For this purpose, a systematic search was conducted on PubMed (MEDLINE), using the MeSH terms [Artificial intelligence] + [Cardiology] + [Cardiovascular] up to February 2025. Original data from clinical trials, observational studies and reviews of interest were reviewed. EXPERT OPINION: Cardiovascular diseases remain the first cause of morbidity, disability, and death worldwide, mainly owing to late diagnosis, insufficient control of cardiovascular risk factors, and poor use of guideline-recommended therapies. Moreover, the high prevalence of cardiac disease increases stress on the health system, which is already overloaded, challenging its capacity to provide quality patient care. AI-based algorithms may assist clinicians by promoting personalized medicine, improving efficiency, and better anticipating outcomes. Although some AI-based technical solutions are currently implemented, most will be ready for use in the coming years. Nonetheless, many challenges, barriers, and ethical concerns remain, and the effective implementation of AI in routine practice will take some time. In this context, it seems necessary to increase medical knowledge of how AI works, its impact on cardiovascular diseases, and its potential translation to clinical practice.","author":[{"family":"Escobar","given":"Carlos"},{"family":"Fácila","given":"Lorenzo"},{"family":"Vidal-Pérez","given":"Rafael"},{"family":"Lapeña","given":"Alberto"},{"family":"Vivas","given":"David"},{"family":"Martín","given":"Ana"},{"family":"Manzanofernández","given":"Sergio"},{"family":"Caballero","given":"Eva"},{"family":"Barrios","given":"Vivencio"},{"family":"Freixapamias","given":"Román"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/14779072.2025.2520830","URL":"https://doi.org/10.1080/14779072.2025.2520830","source":"openalex"},{"id":"oa:W4410859807","type":"article-journal","title":"The Human Voice as a Digital Health Solution Leveraging Artificial Intelligence","abstract":"The human voice is an important medium of communication and expression of feelings or thoughts. Disruption in the regulatory systems of the human voice can be analyzed and used as a diagnostic tool, labeling voice as a potential \"biomarker\". Conversational artificial intelligence is at the core of voice-powered technologies, enabling intelligent interactions between machines. Due to its richness and availability, voice can be leveraged for predictive analytics and enhanced healthcare insights. Utilizing this idea, we reviewed artificial intelligence (AI) models that have executed vocal analysis and their outcomes. Recordings undergo extraction of useful vocal features to be analyzed by neural networks and machine learning models. Studies reveal machine learning models to be superior to spectral analysis in dynamically combining the huge amount of data of vocal features. Clinical applications of a vocal biomarker exist in neurological diseases such as Parkinson's, Alzheimer's, psychological disorders, DM, CHF, CAD, aspiration, GERD, and pulmonary diseases, including COVID-19. The primary ethical challenge when incorporating voice as a diagnostic tool is that of privacy and security. To eliminate this, encryption methods exist to convert patient-identifiable vocal data into a more secure, private nature. Advancements in AI have expanded the capabilities and future potential of voice as a digital health solution.","author":[{"family":"Muddaloor","given":"Pratyusha"},{"family":"Baraskar","given":"Bhavana"},{"family":"Shah","given":"Hriday"},{"family":"Gopalakrishnan","given":"Keerthy"},{"family":"Sood","given":"Divyanshi"},{"family":"Pasupuleti","given":"Prem"},{"family":"Singh","given":"Akshay"},{"family":"Mitra","given":"Dipankar"},{"family":"Hoskote","given":"Sumedh"},{"family":"Iyer","given":"Vivek"},{"family":"Helgeson","given":"Scott"},{"family":"Arunachalam","given":"Shivaram"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/s25113424","URL":"https://doi.org/10.3390/s25113424","source":"openalex"},{"id":"oa:W4411624374","type":"article-journal","title":"Early warning score and feasible complementary approach using artificial intelligence-based bio-signal monitoring system: a review","abstract":"Early warning score (EWS) have become an essential component of patient safety strategies in healthcare environments worldwide. These systems aim to identify patients at risk of clinical deterioration by evaluating vital signs and other physiological parameters, enabling timely intervention by rapid response teams. Despite proven benefits and widespread adoption, conventional EWS have limitations that may affect their ability to effectively detect and respond to patient deterioration. There is growing interest in integrating continuous multimodal monitoring technologies and advanced analytics, particularly artificial intelligence (AI) and machine learning (ML)-based approaches, to address these limitations and enhance EWS performance. This review provides a comprehensive overview of the current state and potential future directions of AI-based bio-signal monitoring in early warning system. It examines emerging trends and techniques in AI and ML for bio-signal analysis, exploring the possibilities and potential applications of various bio-signals such as electroencephalography, electrocardiography, electromyography in early warning system. However, significant challenges exist in developing and implementing AI-based bio-signal monitoring systems in early warning system, including data acquisition strategies, data quality and standardization, interpretability and explainability, validation and regulatory approval, integration into clinical workflows, and ethical and legal considerations. Addressing these challenges requires a multidisciplinary approach involving close collaboration between healthcare professionals, data scientists, engineers, and other stakeholders. Future research should focus on developing advanced data fusion techniques, personalized adaptive models, real-time and continuous monitoring, explainable and reliable AI, and regulatory and ethical frameworks. By addressing these challenges and opportunities, the integration of AI and bio-signals into early warning systems can enhance patient monitoring and clinical decision support, ultimately improving healthcare quality and safety. In conclusion, integrating AI and bio-signals into the early warning system represents a promising approach to improve patient care outcomes and support clinical decision-making. As research in this field continues to evolve, it is crucial to develop safe, effective, and ethically responsible solutions that can be seamlessly integrated into clinical practice, harnessing the power of innovative technology to enhance patient care and improve individual and population health and well-being.","author":[{"family":"Park","given":"Dogeun"},{"family":"So","given":"Kwangsub"},{"family":"Prabhakar","given":"Sunil"},{"family":"Kim","given":"Chulho"},{"family":"Lee","given":"Jae"},{"family":"Sohn","given":"Jong‐hee"},{"family":"Kim","given":"Jong"},{"family":"Lee","given":"Sang‐hwa"},{"family":"Won","given":"Dong"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s13534-025-00486-4","URL":"https://doi.org/10.1007/s13534-025-00486-4","source":"openalex"},{"id":"oa:W4416308911","type":"article-journal","title":"The role of artificial intelligence in enhancing breast cancer screening and diagnosis: A review of current advances","abstract":"Breast cancer (BCA) remains the most prevalent cancer globally and the leading cause of cancer-related mortality among women, with rising incidence rates driven by genetic, lifestyle, and environmental factors. Early detection through precise screening is essential to improve prognosis and survival; yet, challenges persist, especially in resource-limited areas. Recent advances in Artificial Intelligence (AI), particularly machine learning and deep learning algorithms, have illustrated significant potential to enhance breast cancer screening, diagnosis, and treatment personalization. This review highlights the multifaceted role of AI in BCA management, encompassing its applications in image-based screening modalities, genomic and immunologic profiling, and drug discovery. AI-driven approaches offer diagnostic accuracy, cost-effectiveness, time-saving, and individualized treatment regimens. Despite promising developments, further research is crucial to overcome current challenges and regulatory hurdles in clinical settings. This article highlights the positive aspects of AI technologies in advancing BCA care and the importance of continued interdisciplinary research to optimize their implementations in breast cancer workflows.","author":[{"family":"Firuzpour","given":"Faezeh"},{"family":"Heydari","given":"Mohammad"},{"family":"Aram","given":"Cena"},{"family":"Alishvandi","given":"Ali"}],"issued":{"date-parts":[[2025]]},"DOI":"10.34172/bi.30984","URL":"https://doi.org/10.34172/bi.30984","source":"openalex"},{"id":"oa:W4407591319","type":"article-journal","title":"Antibiotics and Artificial Intelligence: Clinical Considerations on a Rapidly Evolving Landscape","abstract":"The growing interest in leveraging artificial intelligence (AI) tools for healthcare decision-making extends to improving antibiotic prescribing. Large language models (LLMs), a type of AI trained on extensive datasets from diverse sources, can process and generate contextually relevant text. While their potential to enhance patient outcomes is significant, implementing LLM-based support for antibiotic prescribing is complex. Here, we specifically expand the discussion on this crucial topic by introducing three interconnected perspectives: (1) the distinctive commonalities, but also the crucial conceptual differences, between the use of LLMs as assistants in scientific writing and in supporting antibiotic prescribing in real-world practice; (2) the possibility and nuances of the expertise paradox; and (3) the peculiarities of the risk of error when considering LLMs to support complex tasks such as antibiotic prescribing.","author":[{"family":"Giacobbe","given":"Daniele"},{"family":"Guastavino","given":"Sabrina"},{"family":"Marelli","given":"Cristina"},{"family":"Murgia","given":"Ylenia"},{"family":"Mora","given":"Sara"},{"family":"Signori","given":"Alessio"},{"family":"Rosso","given":"Nicola"},{"family":"Giacomini","given":"Mauro"},{"family":"Campi","given":"Cristina"},{"family":"Piana","given":"Michele"},{"family":"Bassetti","given":"Matteo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s40121-025-01114-5","URL":"https://doi.org/10.1007/s40121-025-01114-5","source":"openalex"},{"id":"oa:W4416087215","type":"article-journal","title":"The Use of Artificial Intelligence (AI) to Support Dietetic Practice Across Primary Care: A Scoping Review of the Literature","abstract":"Background/objectives: The nutrition care process (NCP) is an evidence-based practice framework used in Medical Nutrition Therapy for the prevention, treatment, and management of non-communicable chronic health conditions. This review aimed to explore available artificial intelligence (AI)-integrated technologies across the NCP in dietetic primary care, their uses, and their impacts on the NCP and patient outcomes. Method: Six databases were searched: MEDLINE, Embase, PsycINFO, Scopus, IEEE, and ACM digital library. Eligible studies were published between January 2007 and August 2024 and included human adult studies, AI-integrated technologies in the dietetic primary care setting, and patient-related outcomes. Extracted details focused on participant characteristics, dietitian involvement, and the type of AI system and its application in the NCP. Results: Ninety-seven studies were included. Three different AI systems (image or audio recognition, chatbots, and recommendation systems) were found. These were implemented in web-based or smartphone applications, wearable sensor systems, smart utensils, and software. Most AI-integrated technologies could be incorporated into one or more NCP stages. Seventy-nine studies reported user- or patient-related outcomes, with mixed findings, but all highlighted efficiencies of using AI. Higher patient engagement was observed with Chatbots. Seventeen studies raised concerns encompassing ethics and patient safety. Conclusions: AI systems show promise as a clinical support tool across most stages of the NCP. Whilst they have varying degrees of accuracy, AI demonstrates potential in improving efficiency, supporting personalised nutrition, and enhancing chronic disease management outcomes. Integrating AI education into dietetic training and professional development will be essential to ensure safe and effective use in practice.","author":[{"family":"Ngo","given":"Kaitlyn"},{"family":"Mekhail","given":"Simone"},{"family":"Chan","given":"Virginia"},{"family":"Li","given":"Xinyi"},{"family":"Yin","given":"Annabelle"},{"family":"Choi","given":"Ha"},{"family":"Allmanfarinelli","given":"Margaret"},{"family":"Chen","given":"Juliana"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/nu17223515","URL":"https://doi.org/10.3390/nu17223515","source":"openalex"},{"id":"oa:W4412390161","type":"article-journal","title":"The Helicobacter pylori AI-clinician harnesses artificial intelligence to personalise H. pylori treatment recommendations","abstract":"Helicobacter pylori (H. pylori) is the most common carcinogenic pathogen globally and the leading cause of gastric cancer. Here, we develop a reinforcement learning-based AI Clinician system to personalise treatment selection and evaluate its ability to improve eradication success compared to clinician-prescribed therapies. The model is trained and internally validated on 38,049 patients from the retrospective European Registry on Helicobacter pylori Management (Hp-EuReg), using independent state deep Q-learning (isDQN) to recommend optimal therapies based on patient characteristics such as age, sex, antibiotic allergies, country, and pre-treatment indication. In internal validation using real-world Hp-EuReg data, AI-recommended therapies achieve a 94.1% success rate (95% CI: 93.2-95.0%) versus 88.1% (95% CI: 87.7-88.4%) for clinician-prescribed therapies not aligned with AI suggestions-an improvement of 6.0%. Results are replicated in an external validation cohort (n = 7186), confirming generalisability. The AI system identifies optimal treatment strategies in key subgroups: 65% (n = 24,923) are recommended bismuth-based therapies, and 15% (n = 5898) non-bismuth quadruple therapies. Random forest modelling identifies region and concurrent medications as patient-specific drivers of AI recommendations. With nearly half the global population likely to contract H. pylori, this approach lays the foundation for future prospective clinical validation and shows the potential of AI to support clinical decision-making, enhance outcomes, and reduce gastric cancer burden.","author":[{"family":"Higgins","given":"KM"},{"family":"Nyssen","given":"Olga"},{"family":"Southern","given":"Joshua"},{"family":"Laponogov","given":"Ivan"},{"family":"Miralles-Marco","given":"Ana"},{"family":"Cabeza-Segura","given":"Manuel"},{"family":"Jiménez-Martí","given":"Elena"},{"family":"Castillo","given":"Josefa"},{"family":"Leja","given":"Mārcis"},{"family":"Poļaka","given":"Inese"},{"family":"Carneiro","given":"Fátima"},{"family":"Figueiredo","given":"Céu"},{"family":"Ferreira","given":"Rui"},{"family":"Barros","given":"Rita"},{"family":"Moreira","given":"Letícia"},{"family":"Cuatrecasas","given":"Míriam"},{"family":"Fernándezesparrach","given":"Glòria"},{"family":"Matysiak-Budnik","given":"Tamara"},{"family":"Martin","given":"JP"},{"family":"Jonaitis","given":"Laimas"},{"family":"Kupčinskas","given":"Juozas"},{"family":"Jonaitis","given":"Paulius"},{"family":"Dinisribeiro","given":"Mário"},{"family":"Coimbra","given":"Miguel"},{"family":"Pereira","given":"Ana"},{"family":"Fontes","given":"Filipa"},{"family":"Spaander","given":"Manon"},{"family":"Honing","given":"Judith"},{"family":"Sedola","given":"Stefano"},{"family":"Pescino","given":"Junior"},{"family":"Maravic","given":"Zorana"},{"family":"Martins","given":"Ana"},{"family":"Veselkov","given":"Dennis"},{"family":"Gisbert","given":"Javier"},{"family":"Fleitas","given":"Tania"},{"family":"Veselkov","given":"Kirill"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41467-025-61329-5","URL":"https://doi.org/10.1038/s41467-025-61329-5","source":"openalex"},{"id":"oa:W4410454801","type":"article-journal","title":"Artificial intelligence unlocks the healthcare data lake","abstract":"Artificial intelligence (AI) is poised to revolutionize surgical care by leveraging the vast and complex “data lake” of healthcare information. This perspective piece outlines how AI may harness structured and unstructured data to improve patient outcomes. Advances in deep learning and foundational models have enabled the development of predictive analytics, automated clinical documentation, personalized patient chatbots, remote monitoring, and enhanced medical imaging. Examples include the ACS NSQIP risk calculator, Sepsis ImmunoScore, startups in ambient transcription, and cutting-edge AI applications in intraoperative imaging and real-time diagnostics. However, the adoption of AI in healthcare requires overcoming challenges, including data privacy, bias, integration into clinical workflows, interoperability, cost, ethical concerns, and regulatory hurdles. As AI technologies evolve, collaboration between surgeons and scientists will be critical to ensure ethical, patient-centered designs. This manuscript calls for surgeons to lead AI applications role in surgery, bridging technology with meaningful use cases to positively align with clinical practice.","author":[{"family":"Talwar","given":"Ankoor"},{"family":"Talwar","given":"Abhinav"},{"family":"Talwar","given":"Abhinav"},{"family":"Talwar","given":"Abhinav"},{"family":"Broach","given":"Robyn"},{"family":"Ungar","given":"Lyle"},{"family":"Hashimoto","given":"Daniel"},{"family":"Fischer","given":"John"}],"issued":{"date-parts":[[2025]]},"DOI":"10.20517/ais.2024.109","URL":"https://doi.org/10.20517/ais.2024.109","source":"openalex"},{"id":"oa:W4413857435","type":"article-journal","title":"Artificial Intelligence in Clinical Decision-Making: A Scoping Review of Rule-Based Systems and Their Applications in Medicine","abstract":"Artificial intelligence (AI) has become increasingly integrated into clinical workflows, with rule-based clinical decision support systems (CDSS) emerging as one of its most mature and widely adopted applications. These systems rely on rule engines, that is, software components that apply predefined conditional logic (if/then rules) to patient data, to deliver alerts, diagnostic suggestions, or treatment recommendations. By embedding expert knowledge into structured rule sets and utilizing inference engines to process them, rule-based CDSS provides transparent, interpretable, and adaptable decision support. Although their use has expanded significantly over the past decade, evolving from simple decision aids to advanced tools incorporating AI and real-time analytics, a comprehensive synthesis of their applications, effectiveness, and technological evolution remains lacking. This scoping review aims to examine the current landscape of rule engine implementations in medicine, focusing on their clinical functionalities, evaluated outcomes, technological characteristics, and geographic adoption patterns across different medical domains. Following established scoping review methodology, we conducted a systematic search of PubMed and Scopus databases (2007-2023). Of 437 initially identified records, 28 studies met our inclusion criteria after rigorous screening. Data were extracted on study characteristics, clinical applications, rule engine technologies, and implementation outcomes, with particular attention to temporal trends and geographic distribution. The analysis revealed several key findings. The United States accounted for 46.42% of studies, demonstrating significant geographic concentration. Technologically, implementations evolved from early SQL-based systems to contemporary approaches integrating machine learning and natural language processing. Clinically, the rule engine showed particular effectiveness in chronic disease management (approximately 30% focused on diabetes care) and demonstrated measurable improvements, such as 30% reductions in adverse drug events. However, challenges persisted in system interoperability and clinician adoption across multiple studies. Our analysis of 28 studies demonstrates that rule engines have demonstrated substantial potential to enhance clinical decision-making and healthcare efficiency, though their adoption remains uneven geographically and is technically constrained in some settings. Based on our findings, we recommend: (1) developing standardized implementation frameworks to address interoperability challenges, (2) expanding research and deployment in underrepresented regions, and (3) investing in hybrid systems that combine rule-based logic with machine learning capabilities. These insights provide valuable guidance for healthcare organizations seeking to implement or optimize rule engine technologies in clinical practice.","author":[{"family":"Alnattah","given":"Ashraf"},{"family":"Jajroudi","given":"Mahdie"},{"family":"Fadafen","given":"Seyyed"},{"family":"Manzari","given":"Mahdi"},{"family":"Eslami","given":"Saeid"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.91333","URL":"https://doi.org/10.7759/cureus.91333","source":"openalex"},{"id":"oa:W4415273825","type":"article-journal","title":"Optimized Ensemble Machine-Learning-Driven Transparent Thyroid Cancer Prediction Using Explainable Artificial Intelligence","abstract":"Thyroid cancer disease diagnosis is a critical medical challenge, requiring accurate and reliable predictions to support clinical decision-making. Patients may suffer from an incomplete diagnosis while using traditional machine learning models. In this study, we present an optimized ensemble machine-learning framework for predicting thyroid cancer disease. The methodology integrates multiple classifiers, including support vector machine, random forest, Naïve Bayes, K-nearest neighbors, and decision tree. The final classification decision is determined with the help of soft voting by a total predictive support vector machine, which selects the classifier with the highest confidence score among the ensemble models. The ensemble strategy enhances predictive accuracy and robustness by combining the strengths of individual classifiers. The model was trained and evaluated, achieving an impressive accuracy of 0.9633 and an area under the receiver operating characteristic curve of 0.9914. The proposed method of this study is very accurate, but there is still a black box problem. To overcome this issue and to ensure interpretability, Explainable Artificial Intelligence techniques, including Shapley Additive Explanations and Local Interpretable Model-agnostic Explanations, are implemented, providing insights into feature contributions towards the performance of the proposed method and model decisions. The dataset contains a total of 30 features with 3,772 different cases consisting of two classes. Received: 16 June 2025 | Revised: 14 August 2025 | Accepted: 24 August 2025 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The data that support the findings of this study are openly available in Kaggle at https://www.kaggle.com/datasets/bidemiayinde/thyroid-sickness-determination. The data that support the findings of this study are openly available in Kaggle at https://www.kaggle.com/datasets/bhargavchirumamilla/thyroid-cancer-risk-dataset. The data that support the findings of this study are openly available in Kaggle at https://www.kaggle.com/datasets/sikandaraidev/thyroid-dataset. Author Contribution Statement Syed Younus Ali: Conceptualization, Software, Data curation, Writing – original draft, Writing – review & editing. Bilal Shoaib Khan: Conceptualization, Methodology, Resources, Writing – original draft, Writing – review & editing, Visualization, Supervision. Abdul Hanan Khan: Software, Validation. Muhammad Adnan Khan: Formal analysis, Investigation. Asghar Ali Shah: Validation, Investigation. Sagheer Abbas: Formal analysis, Project administration. Khan Muhammad Adnan: Methodology, Resources.","author":[{"family":"Ali","given":"Syed"},{"family":"Shoaib","given":"Bilal"},{"family":"Khan","given":"Abdul"},{"family":"Khan","given":"Muhammad"},{"family":"Shah","given":"Asghar"},{"family":"Abbas","given":"Sagheer"},{"family":"Adnan","given":"Khan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.47852/bonviewjcce52026503","URL":"https://doi.org/10.47852/bonviewjcce52026503","source":"openalex"},{"id":"oa:W4408999757","type":"article-journal","title":"Comparing Artificial Intelligence–Generated and Clinician-Created Personalized Self-Management Guidance for Patients With Knee Osteoarthritis: Blinded Observational Study","abstract":"BACKGROUND: Knee osteoarthritis is a prevalent, chronic musculoskeletal disorder that impairs mobility and quality of life. Personalized patient education aims to improve self-management and adherence; yet, its delivery is often limited by time constraints, clinician workload, and the heterogeneity of patient needs. Recent advances in large language models offer potential solutions. GPT-4 (OpenAI), distinguished by its long-context reasoning and adoption in clinical artificial intelligence research, emerged as a leading candidate for personalized health communication. However, its application in generating condition-specific educational guidance remains underexplored, and concerns about misinformation, personalization limits, and ethical oversight remain. OBJECTIVE: We evaluated GPT-4's ability to generate individualized self-management guidance for patients with knee osteoarthritis in comparison with clinician-created content. METHODS: This 2-phase, double-blind, observational study used data from 50 patients previously enrolled in a registered randomized trial. In phase 1, 2 orthopedic clinicians each generated personalized education materials for 25 patient profiles using anonymized clinical data, including history, symptoms, and lifestyle. In phase 2, the same datasets were processed by GPT-4 using standardized prompts. All content was anonymized and evaluated by 2 independent, blinded clinical experts using validated scoring systems. Evaluation criteria included efficiency, readability (Flesch-Kincaid, Gunning Fog, Coleman-Liau, and Simple Measure of Gobbledygook), accuracy, personalization, and comprehensiveness and safety. Disagreements between reviewers were resolved through consensus or third-party adjudication. RESULTS: GPT-4 outperformed clinicians in content generation speed (530.03 vs 37.29 words per min, P<.001). Readability was better on the Flesch-Kincaid (mean 11.56, SD 1.08 vs mean 12.67 SD 0.95), Gunning Fog (mean 12.47, SD 1.36 vs mean 14.56, SD 0.93), and Simple Measure of Gobbledygook (mean 13.33, SD 1.00 vs mean 13.81 SD 0.69) indices (all P<.001), though GPT-4 scored slightly higher on the Coleman-Liau Index (mean 15.90, SD 1.03 vs mean 15.15, SD 0.91). GPT-4 also outperformed clinicians in accuracy (mean 5.31, SD 1.73 vs mean 4.76, SD 1.10; P=.05, personalization (mean 54.32, SD 6.21 vs mean 33.20, SD 5.40; P<.001), comprehensiveness (mean 51.74, SD 6.47 vs mean 35.26, SD 6.66; P<.001), and safety (median 61, IQR 58-66 vs median 50, IQR 47-55.25; P<.001). CONCLUSIONS: GPT-4 could generate personalized self-management guidance for knee osteoarthritis with greater efficiency, accuracy, personalization, comprehensiveness, and safety than clinician-generated content, as assessed using standardized, guideline-aligned evaluation frameworks. These findings underscore the potential of large language models to support scalable, high-quality patient education in chronic disease management. The observed lexical complexity suggests the need to refine outputs for populations with limited health literacy. As an exploratory, single-center study, these results warrant confirmation in larger, multicenter cohorts with diverse demographic profiles. Future implementation should be guided by ethical and operational safeguards, including data privacy, transparency, and the delineation of clinical responsibility. Hybrid models integrating artificial intelligence-generated content with clinician oversight may offer a pragmatic path forward.","author":[{"family":"Du","given":"Kai"},{"family":"Li","given":"Ao"},{"family":"Zuo","given":"Qi"},{"family":"Zhang","given":"Chen"},{"family":"Guo","given":"Ren"},{"family":"Chen","given":"Ping"},{"family":"Du","given":"Wei"},{"family":"Li","given":"Shu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/67830","URL":"https://doi.org/10.2196/67830","source":"openalex"},{"id":"oa:W4412161112","type":"article-journal","title":"Perceptions of mental health professionals towards artificial intelligence in mental healthcare: a cross-sectional study","abstract":"Background: Artificial intelligence (AI) holds significant potential for enhancing mental health care, but uptake is limited, potentially impacted by demographic factors of healthcare professionals. Further, while AI use in Saudi Arabia is progressive, there is minimal exploration of its role and impact within mental health services. Objective: This study presents a unique exploration of psychiatric professional's perceptions of AI in mental health care in Jeddah, Saudi Arabia. Methods: A cross-sectional online survey was conducted with a sample of mental health professionals from two governmental mental health hospitals in Jeddah, Saudi Arabia. The study tool was made up of two sections, the first consisting of sociodemographic questions and the second was the Shinners Artificial Intelligence Perception (SHAIP) questionnaire assessing the perceptions towards AI in mental healthcare, with data analyzed using IBM SPSS Statistical software. Results: A total of 251 mental health professionals, mostly females (56.6%), aged 31-40 (50%), married (45%), and nurses (55.4%). Only 24.3% used AI in practice, though 85.7% were aware of AI. Participants positively rated AI's impact (mean item range: 3.48-3.75) and felt unprepared for role-specific AI (mean 2.78). Nurses and those aware of AI had higher AI impact perceptions (p<0.0001) Specialty and AI awareness affected AI preparedness (p=0.001, p=0.029). Discussion: The study provides insights into mental health professionals ' views on AI in mental healthcare, emphasizing the need for targeted education to improve AI literacy and preparedness among Saudi healthcare professionals. It highlights the importance of ethical AI implementation to enhance patient care and advance psychiatric practice in the region.","author":[{"family":"Sharif","given":"Loujain"},{"family":"Almabadi","given":"Reem"},{"family":"Alahmari","given":"Alhanouf"},{"family":"Alqurashi","given":"Fai"},{"family":"Alsahafi","given":"Fidaa"},{"family":"Qusti","given":"Shahad"},{"family":"Akash","given":"Walaa"},{"family":"Mahsoon","given":"Alaa"},{"family":"Poudel","given":"Dev"},{"family":"Sharif","given":"Khalid"},{"family":"Wright","given":"Rebecca"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fpsyt.2025.1601456","URL":"https://doi.org/10.3389/fpsyt.2025.1601456","source":"openalex"},{"id":"oa:W4416293174","type":"article-journal","title":"Artificial Intelligence Education in Radiology Training: A Systematic Review of Effectiveness, Barriers, and Future Directions","abstract":"RATIONALE AND OBJECTIVES: The purpose of this systematic review study was to characterize the current landscape of various artificial intelligence (AI) education in radiology, summarizing existing curricula, outcomes, challenges, and future directions for effective integration into residency training. MATERIALS AND METHODS: A comprehensive search of PubMed, Web of Science, Embase, and Google Scholar identified relevant published studies up to June 19, 2025. RESULTS: Of the 2646 studies screened, 14 studies evaluated the performance of AI-based training programs for radiology trainees; among these, 92.9% (13/14) reported improvements in trainees' performance, including better diagnostic precision and interpretation (57.2%, 8/14), greater trainee confidence (57.2%, 8/14), hands-on experience with AI platforms (85.7%, 12/14), increased AI knowledge (85.7%, 12/14), engagement with AI-based case learning (35.7%, 5/14), understanding of AI ethics and bias (7.1%, 1/14), and acceptance of AI-assisted learning (78.6%, 11/14), whereas one study (7.1%, 1/14) found no significant benefit. Performance evaluation metrics varied across studies, with 35.7% (5/14) reporting a higher median of sensitivity, specificity, and accuracy (72%, 80%, and 81.3%) after AI training compared to before AI training (62.2%, 78.9%, and 76.5%, respectively), and 28.6% (4/14) showing improved AI knowledge scores. Hands-on simulations and didactic lectures were the most common AI training formats (78.6% and 71.4%). Risks and concerns included over-reliance on AI, limited exposure to complex or rare cases, and a lack of feedback. Recommendations highlighted the need for AI-faculty teaching, broader content coverage, and standardized multi-center AI-training programs to facilitate wider adoption. CONCLUSION: 92.9% of studies showed that AI-based training can enhance radiology trainees' knowledge, interpretive skills, or diagnostic performance, especially for junior trainees; however, its safe adoption requires standardized curricula with diverse cases, mentorship, workflow integration, and robust evaluation, with larger studies needed to confirm generalizability.","author":[{"family":"Keshavarz","given":"Pedram"},{"family":"Mohammadigoldar","given":"Zahra"},{"family":"Bedayat","given":"Arash"},{"family":"Razi","given":"Farzad"},{"family":"Satei","given":"Alexander"},{"family":"Arsene","given":"Camelia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.acra.2025.10.049","URL":"https://doi.org/10.1016/j.acra.2025.10.049","source":"openalex"},{"id":"oa:W4414667728","type":"article-journal","title":"Application of Explainable Artificial Intelligence Based on Visual Explanation in Digestive Endoscopy","abstract":"At present, artificial intelligence (AI) has shown significant potential in digestive endoscopy image analysis, serving as a powerful auxiliary tool for the accurate diagnosis and treatment of gastrointestinal diseases. However, mainstream models represented by deep learning are often characterized as complex \"black boxes,\" with decision-making processes that are difficult for humans to interpret. The lack of interpretability undermines physicians' trust in model results and hinders the broader use of models in clinical practice. To address this core challenge, Explainable AI (XAI) has emerged to enhance the transparency of decision-making, thereby establishing a foundation of trust for human-machine collaboration. The review systematically reviews 34 articles (7 articles in esophagogastroduodenoscopy, 13 articles in colonoscopy, 9 articles in endoscopic ultrasonography, and 5 articles in wireless capsule endoscopy), focusing on the research progress and applications of XAI in the field of digestive endoscopic image analysis, with particular emphasis on the visual explanation-based methods. We first clarify the definition and mainstream classification of XAI, then introduce the principles and characteristics of key XAI methods based on visual explanation. Subsequently, we review the applications of these methods in digestive endoscopy image analysis. Lastly, we explore the obstacles presently faced in this domain and the future directions. This study provides a theoretical basis for constructing a trustworthy and transparent AI-assisted digestive endoscopy diagnosis and treatment system and promotes the implementation and application of XAI in clinical practice.","author":[{"family":"Cai","given":"Xiaohan"},{"family":"Zhang","given":"Zexin"},{"family":"Zhao","given":"Siqi"},{"family":"Liu","given":"Wentian"},{"family":"Fan","given":"Xiaofei"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bioengineering12101058","URL":"https://doi.org/10.3390/bioengineering12101058","source":"openalex"},{"id":"oa:W4415133485","type":"article-journal","title":"Exploring the Intersection of Nursing Leadership and Artificial Intelligence: Scoping Review","abstract":"Background: As artificial intelligence (AI) technology permeates health care settings, nurse leaders must position themselves to shape its development, implementation, and impact, guiding meaningful change that benefits nurses and care delivery. Nurse leaders possess the capacity to influence decisions, shape practice, and ensure the delivery of ethical, safe, and high-quality care. While AI technology is reshaping many aspects of health care delivery, there is limited knowledge on how nurse leaders perceive and experience this shift. Objective: This scoping review aimed to explore the intersection of nursing leadership and AI technology in health care by mapping current evidence, identifying key concepts, and highlighting knowledge gaps within the literature. Methods: This scoping review was guided by the Joanna Briggs Institute methodology and reported on using the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) checklist. A systematic search of 4 electronic databases (CINAHL [EBSCO Information Services], Ovid MEDLINE [Wolters Kluwer], PsycINFO [American Psychological Association], and Scopus [Elsevier]) was conducted for English-language, peer-reviewed literature published between 2014 and 2025. Gray literature was also reviewed. Articles were included if they met the inclusion criteria by exploring the population of nurse leaders and the concept of AI technology within the context of health care settings and were published in English from May 2014 forward. A total of 26 articles were included in the analysis. Qualitative content analysis and numerical summary supported the inductive identification and synthesis of data categories. Results: Of the 26 articles included, 8 were empirical (qualitative, quantitative, or mixed methods), and 18 were conceptual or theoretical articles. Although 1 article was Canadian, there were no empirical studies conducted by Canadian researchers. The qualitative content analysis of the primary search findings revealed 6 overarching data categories: (1) leading digital transformation and technology integration, (2) AI technology and the nursing role: reshaping practice, (3) ethical considerations of AI technology for nurse leaders, (4) AI technology as a facilitator of innovative leadership, (5) education and training on AI technology in nursing practice, and (6) influence of AI technology on the work environment. Conclusions: This review confirms that nurse leaders play an essential role in shaping the future of health care in the context of AI technology. Although this review highlights a growing recognition of nursing leadership as a crucial driver of AI technology integration in health care, there is a lack of research to guide practice, policy, and leadership development through education, despite emerging interest and a recent increase in empirical work. The findings accentuate the need for increased investment in nurse-led research and leadership development to ensure that AI systems are designed, implemented, and evaluated in a manner that upholds ethical care, equity, and professional nursing values. As health care systems increasingly adopt AI technology, nurse leaders must be equipped with the knowledge, tools, and support required to lead transformative change and act as AI technology directors.","author":[{"family":"Burford","given":"Jessica"},{"family":"Booth","given":"Richard"},{"family":"Mcintyre","given":"Amanda"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/80085","URL":"https://doi.org/10.2196/80085","source":"openalex"},{"id":"oa:W4409803332","type":"article-journal","title":"AIFM-ed Curriculum Framework for Postgraduate Family Medicine Education on Artificial Intelligence: Mixed Methods Study","abstract":"BACKGROUND: As health care moves to a more digital environment, there is a growing need to train future family doctors on the clinical uses of artificial intelligence (AI). However, family medicine training in AI has often been inconsistent or lacking. OBJECTIVE: The aim of the study is to develop a curriculum framework for family medicine postgraduate education on AI called \"Artificial Intelligence Training in Postgraduate Family Medicine Education\" (AIFM-ed). METHODS: First, we conducted a comprehensive scoping review on existing AI education frameworks guided by the methodological framework developed by Arksey and O'Malley and Joanna Briggs Institute methodological framework for scoping reviews. We adhered to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) checklist for reporting the results. Next, 2 national expert panels were conducted. Panelists included family medicine educators and residents knowledgeable in AI from family medicine residency programs across Canada. Participants were purposively sampled, and panels were held via Zoom, recorded, and transcribed. Data were analyzed using content analysis. We followed the Standards for Reporting Qualitative Research for panels. RESULTS: An integration of the scoping review results and 2 panel discussions of 14 participants led to the development of the AIFM-ed curriculum framework for AI training in postgraduate family medicine education with five key elements: (1) need and purpose of the curriculum, (2) learning objectives, (3) curriculum content, (4) organization of curriculum content, and (5) implementation aspects of the curriculum. CONCLUSIONS: Using the results of this study, we developed the AIFM-ed curriculum framework for AI training in postgraduate family medicine education. This framework serves as a structured guide for integrating AI competencies into medical education, ensuring that future family physicians are equipped with the necessary skills to use AI effectively in their clinical practice. Future research should focus on the validation and implementation of the AIFM-ed framework within family medicine education. Institutions also are encouraged to consider adapting the AIFM-ed framework within their own programs, tailoring it to meet the specific needs of their trainees and health care environments.","author":[{"family":"Tolentino","given":"Raymond"},{"family":"Hersson-Edery","given":"Fanny"},{"family":"Yaffe","given":"Mark"},{"family":"Rahimi","given":"Samira"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/66828","URL":"https://doi.org/10.2196/66828","source":"openalex"},{"id":"oa:W4415777022","type":"article-journal","title":"Redefining oral healthcare through artificial intelligence: a review of current applications and a roadmap for the future of dentistry","abstract":"The integration of artificial intelligence (AI) into dental practice represents a significant technological advancement in oral healthcare. This narrative review examines current research across AI domains including machine learning, deep learning, computer vision, natural language processing, and generative modeling to assess how AI is influencing clinical workflows, diagnostic procedures, treatment planning, and surgical interventions in dentistry. The central research question addresses how artificial intelligence is currently being applied in dental practice and what opportunities and challenges exist for its future implementation. Evidence from peer-reviewed studies demonstrates variable efficacy of AI in supporting diagnostic decisions for caries, periodontal disease, and oral cancer; assisting orthodontic planning and prosthodontic CAD/CAM workflows; and enabling robotic surgery and predictive modeling. While some AI models show promising accuracy metrics, questions remain about their generalizability and real-world performance. The review explores integration of AI through electronic health records, decision-support systems, and emerging digital twin technologies. Interdisciplinary collaborations with biomedical engineering, computational biology, and materials science are examined. Critical challenges including algorithmic bias, data privacy, informed consent, liability frameworks, and adoption barriers such as technological infrastructure, clinician acceptance, and regulatory uncertainties are discussed. Future directions emphasize federated learning for collaborative model development; explainable AI to improve transparency; integration with multi-omics data; and adaptive systems capable of continuous learning. This review provides a balanced assessment of AI's potential in dentistry while acknowledging significant implementation challenges that must be addressed to ensure equitable and effective patient care.","author":[{"family":"Dua","given":"Bharti"},{"family":"Gupta","given":"Rajiv"},{"family":"Bhargava","given":"Akshay"},{"family":"Bhardwaj","given":"Anupam"},{"family":"Jain","given":"Meena"},{"family":"Tripathi","given":"Siddhi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s44398-025-00013-6","URL":"https://doi.org/10.1186/s44398-025-00013-6","source":"openalex"},{"id":"oa:W4414658925","type":"article-journal","title":"Enhanced intrusion detection in cybersecurity through dimensionality reduction and explainable artificial intelligence","abstract":"Cybersecurity is one of the applications of controls, procedures, and technologies for protecting data, networks, programs, and systems from potential cyber threats. Malicious threats have become complex, and the leading task is to recognize obfuscated and mysterious malware, as the malware inventors utilize dissimilar evasion models for data covering to avert recognition by intrusion detection systems (IDSs). Artificial intelligence (AI) usage in cybersecurity is gradually becoming familiar, but the main task is the absence of interpretability and transparency of AI methods. Explainable AI (XAI) can tackle this problem by improving the understandability of AI techniques, permitting cyber-security experts to comprehend the decisions created by these methods and to recognize biases or errors. Recently, Machine learning (ML) and deep learning (DL) models have delivered automatic analytical intrusion detection procedures, providing numerous advantages. This study proposes an Enhanced Intrusion Detection in Cybersecurity through Dimensionality Reduction and Explainable Artificial Intelligence with Attention Mechanism in Deep Learning (EIDCDR-XAIADL) model. The main intention of the proposed EIDCDR-XAIADL model is to deliver a robust cybersecurity system that combines XAI to address the attacks. Initially, the proposed EIDCDR-XAIADL technique performs data normalization by using mean normalization to ensure uniform scaling of network traffic data. The multiverse optimization (MVO) technique selects the most appropriate and discriminative features. For the cybersecurity attack classification process, the hybrid of convolutional neural network (CNN), bi-directional gated recurrent unit (BiGRU), and attention mechanism (CNN-BiGRU-AM) technique is implemented. Moreover, the antlion optimization (ALO) technique adjusts the hyperparameter values of the CNN-BiGRU-AM method optimally and results in more excellent classification performance. Finally, Shapley Additive Explanations (SHAP) is utilized as an XAI technique to enhance threat detection and decision-making by providing trustworthy insights into AI-driven security systems. The experimental evaluation of the EIDCDR-XAIADL approach is examined under dual datasets. The experimental validation of the EIDCDR-XAIADL approach demonstrated a superior accuracy value of 99.19% and 99.12% under NSLKDD and CICIDS 2017 datasets.","author":[{"family":"Alamro","given":"Hayam"},{"family":"Alahmari","given":"Sultan"},{"family":"Nemri","given":"Nadhem"},{"family":"Aljebreen","given":"Mohammed"},{"family":"Alhashmi","given":"Asma"},{"family":"Alamro","given":"Sulaiman"},{"family":"Alqazzaz","given":"Ali"},{"family":"Duhayyim","given":"Mesfer"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-06761-9","URL":"https://doi.org/10.1038/s41598-025-06761-9","source":"openalex"},{"id":"oa:W4416331297","type":"article-journal","title":"Artificial intelligence algorithms in orthopaedics: A narrative review of methods and clinical applications","abstract":"This narrative review evaluates the role of artificial intelligence (AI) algorithms in orthopaedic surgery and distinguishes itself by explaining how the main algorithmic approaches function and illustrating each with orthopaedic examples. Machine learning methods, including regression, classification and reinforcement learning, have been applied to fracture detection, prediction of revision risk and modelling of outcomes after arthroplasty and sports injury. Deep learning and convolutional neural networks have improved fracture classification, implant surveillance and segmentation of cartilage and meniscal tissue on magnetic resonance imaging. Neural networks such as FracNet and YOLO-based systems demonstrate growing capability in trauma imaging. Natural language processing has automated the extraction of operative and registry data, while large language models are emerging for diagnostic support and education. Generative artificial intelligence (GAI) have produced synthetic musculoskeletal images to expand data sets. Computer vision and image processing underpin robotic-assisted surgery and preoperative planning, and federated learning enables multicentre collaboration while protecting privacy. Each algorithm offers strengths in accuracy, efficiency or scalability, but also carries bias, transparency, computational cost and lack of external validation. This review explores how these algorithms are shaping orthopaedics, highlighting their benefits, limitations and challenges. Rigorous validation, transparent reporting and governance are essential for safe clinical use. Level of Evidence: N/A.","author":[{"family":"Rosen","given":"Jamie"},{"family":"Russell","given":"Jemima"},{"family":"Kartik","given":"Prerna"},{"family":"Vellabaldacchino","given":"Martinique"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/jeo2.70549","URL":"https://doi.org/10.1002/jeo2.70549","source":"openalex"},{"id":"oa:W4408488823","type":"article-journal","title":"Regulatory and legal challenges of Artificial Intelligence in the U.S. Healthcare System: Liability, Compliance, and Patient Safety","abstract":"This study explores Regulatory and Legal Challenges of Artificial Intelligence in the U.S. Healthcare System: Liability, Compliance, and Patient Safety. It also examines the challenges associated with AI integration, including ethical concerns, data privacy risks, and regulatory compliance, as providing insights into legal frameworks governing AI in healthcare. A qualitative research approach was employed, involving a comprehensive review of existing literature, and regulatory policies. Peer-reviewed journals, government publications, and industry reports were analyzed to assess the effectiveness, challenges, and future implications of AI-driven healthcare solutions. The study reveals that AI significantly enhances healthcare delivery by improving diagnostic accuracy, enabling personalized treatments, and optimizing hospital workflows. Machine learning models and natural language processing facilitate early disease detection, while robotic process automation streamlines administrative processes. However, challenges such as algorithmic bias, data security concerns, and the need for stringent regulatory oversight persist. Regulatory frameworks such as HIPAA, GDPR, and FDA guidelines provide necessary compliance structures but require continuous updates to keep pace with AI advancements. The paper therefore concludes that AI is revolutionizing healthcare, offering significant benefits in efficiency and patient outcomes. However, successful implementation necessitates a balanced approach that integrates ethical considerations, data protection measures, and regulatory frameworks. Future research should focus on enhancing AI transparency, addressing biases, and ensuring that AI-driven healthcare solutions remain patient-centered and legally compliant.","author":[{"family":"Osifowokan","given":"Adewale"},{"family":"Agbadamasi","given":"Tessy"},{"family":"Adukpo","given":"Tobias"},{"family":"Mensah","given":"Nicholas"}],"issued":{"date-parts":[[2025]]},"DOI":"10.30574/wjarr.2025.25.3.0807","URL":"https://doi.org/10.30574/wjarr.2025.25.3.0807","source":"openalex"},{"id":"oa:W4408862063","type":"article-journal","title":"Individual dynamic capabilities and artificial intelligence in health operations: Exploration of innovation diffusion","abstract":"This research investigates the integration of individual dynamic capabilities (IDC), artificial intelligence (AI), and the Technology Acceptance Model (TAM) within health operations to evaluate their role in fostering innovation diffusion in healthcare. A convergent, multifaceted research approach encompassing quantitative and qualitative methodologies was employed, commencing with a systematic review of the extant literature. This was then complemented by the execution of focus group sessions involving 21 participants. The main objective of this sequential exploratory design was to synthesize existing research present an empirical validation of real-world case studies, and assess AI deployment challenges that influence operational efficiency and service quality in healthcare organizations. The findings underscore the importance of IDC in advancing healthcare practices by driving cross-functional adaptation, facilitating AI implementation, and ensuring smooth operational transformation in line with healthcare standards and best practices. The findings offer valuable insights for operational and executive-level decision-makers aiming to optimize health operations by integrating IDC and AI technologies, enhancing patient care, service quality, and innovative health solutions. • Integration of AI and IDC enhances healthcare operational efficiency and regulatory compliance. •Individual dynamic capabilities foster adaptability, technological adoption, and continuous learning in healthcare operations. •AI-driven predictive analytics streamline decision-making and improve patient care outcomes. •IDC and AI synergistically optimize data interoperability and regulatory standards in healthcare systems. •Leadership commitment and cross-functional collaboration drive successful AI and IDC implementation.","author":[{"family":"Pesqueira","given":"António"},{"family":"Sousa","given":"Maria"},{"family":"Pereira","given":"Rúben"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.ibmed.2025.100239","URL":"https://doi.org/10.1016/j.ibmed.2025.100239","source":"openalex"},{"id":"oa:W4412024302","type":"article-journal","title":"The Use of Generative Artificial Intelligence (AI) in Academic Research: A Review of the Consensus App","abstract":"Consensus App is an academic search engine designed to change how researchers access and synthesize information. It helps researchers quickly browse the growing body of academic literature by offering insights at both the topic and paper levels. We evaluate the Consensus App's potential to transform academic research, its ethical implications, and the reasons behind its underrepresentation in academic literature. We seek to provide a balanced perspective on the app's current and future influence in academic research. This paper is based on a rapid review of the literature to see how the Consensus App is used and reported in the literature. Our review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We focused on identifying applications, benefits, and ethical concerns related to the Consensus App. The search was conducted on December 23, 2024, across 210 academic databases. The databases from which articles were retrieved include Web of Science (N=6), MEDLINE (N=2), Academic Search Ultimate (N=1), and Fuente Académica Plus (N=1). In addition to the database searches, five additional editorials were identified through targeted manual searches of high-impact journals. In total, 10 papers were included in the final review. ChatGPT-4.5 was used to assist in synthesizing key themes across the articles, focusing on application, benefits, and ethical concerns related to the Consensus App and the broader use of artificial intelligence (AI) in scholarly work. The reviewed articles revealed that the use of the Consensus App is surprisingly low, which may suggest underreporting by its users. Researchers may also not be aware of it. These studies showed how the app has been limitedly used in the literature. Despite its advantages, we identified ethical concerns in the reviewed studies. Despite its potential, the Consensus App remains underutilized and significantly underreported in academic literature. Therefore, it is important for academic institutions, journal editors, and researchers to collaboratively develop standardized reporting guidelines when AI is involved in the process of manuscript development. The eventual goal is to lead to a more transparent reporting of AI usage in research.","author":[{"family":"Apata","given":"Olukayode"},{"family":"Kwok","given":"Oi‐man"},{"family":"Lee","given":"Yuan‐hsuan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.87297","URL":"https://doi.org/10.7759/cureus.87297","source":"openalex"},{"id":"oa:W4408589460","type":"article-journal","title":"Artificial Intelligence in Nuclear Cardiac Imaging: Novel Advances, Emerging Techniques, and Recent Clinical Trials","abstract":"Cardiovascular disease (CVD) is a leading cause of death, accounting for over 30% of annual global fatalities. Ischemic heart disease, in turn, is a frontrunner of worldwide CVD mortality. With the burden of coronary disease rapidly growing, understanding the nuances of cardiac imaging and risk prognostication becomes paramount. Myocardial perfusion imaging (MPI) is a frequently utilized and well established testing modality due to its significant clinical impact in disease diagnosis and risk assessment. Recently, nuclear cardiology has witnessed major advancements, driven by innovations in novel imaging technologies and improved understanding of cardiovascular pathophysiology. Applications of artificial intelligence (AI) to MPI have enhanced diagnostic accuracy, risk stratification, and therapeutic decision-making in patients with coronary artery disease (CAD). AI techniques such as machine learning (ML) and deep learning (DL) neural networks offer new interpretations of immense data fields, acquired through cardiovascular imaging modalities such as nuclear medicine (NM). Recently, AI algorithms have been employed to enhance image reconstruction, reduce noise, and assist in the interpretation of complex datasets. The rise of AI in nuclear medicine (AI-NM) has proven itself groundbreaking in the efficiency of image acquisition, post-processing time, diagnostic ability, consistency, and even in risk-stratification and outcome prognostication. To that end, this narrative review will explore these latest advances in AI in nuclear medicine and its rapid transformation of the cardiac diagnostics landscape. This paper will examine the evolution of AI-NM, review novel AI techniques and applications in nuclear cardiac imaging, summarize recent AI-NM clinical trials, and explore the technical and clinical challenges in its implementation of artificial intelligence.","author":[{"family":"Golub","given":"Ilana"},{"family":"Thummala","given":"Abhinav"},{"family":"Morad","given":"Tyler"},{"family":"Dhaliwal","given":"Jasmeet"},{"family":"Elisarraras","given":"Francisco"},{"family":"Karlsberg","given":"Ronald"},{"family":"Cho","given":"Geoffrey"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jcm14062095","URL":"https://doi.org/10.3390/jcm14062095","source":"openalex"},{"id":"oa:W4417071200","type":"article-journal","title":"Artificial Intelligence in Qualitative Research: Insights From Experts via Reflexive Thematic Analysis","abstract":"The rapid advancement of artificial intelligence (AI) is increasingly shaping research methodologies across disciplines. However, its integration in qualitative research remains controversial due to epistemological, ethical, and human-centered concerns. This study explores the perspectives of 14 expert qualitative researchers from socio-anthropological and healthcare fields working in Italian academic and hospital settings, with a focus on the opportunities, challenges, and future directions of AI use in qualitative inquiry. Through semi-structured interviews and reflexive thematic analysis, four main themes were developed. First, participants expressed ambivalent attitudes-balancing curiosity with technophobia and emphasizing the need for human oversight and contextual interpretation. Second, an anthropological and philosophical dimension was constructed, underscoring the importance of reflexivity, creativity, and researcher identity as essential counterbalances to AI's mechanistic tendencies. Third, researchers acknowledged AI's practical benefits in tasks such as transcription and data management, and they remained skeptical of its ability to perform complex interpretative work. Finally, ethical and sustainability concerns were raised, including algorithmic bias, data privacy, and the environmental impact of AI technologies. The findings reveal persistent epistemological tensions but also highlight emerging opportunities for AI to enhance research efficiency and accessibility, provided that human interpretative agency remains central. Participants stressed the importance of developing robust ethical frameworks, fostering critical reflexivity, and adopting innovative conceptual approaches to responsibly integrate AI into qualitative research and education. This study offers valuable insights for scholars and practitioners navigating the evolving landscape of AI in qualitative inquiry, advocating a balanced approach that leverages AI's potential while safeguarding the human core of qualitative research.","author":[{"family":"Dellafiore","given":"Federica"},{"family":"Saba","given":"Andreina"},{"family":"Collaro","given":"Concetta"},{"family":"Artioli","given":"Giovanna"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1177/10497323251389800","URL":"https://doi.org/10.1177/10497323251389800","source":"openalex"},{"id":"oa:W4414179015","type":"article-journal","title":"Improving Sepsis Prediction in the ICU with Explainable Artificial Intelligence: The Promise of Bayesian Networks","abstract":"Background/Objectives: Sepsis remains one of the leading causes of mortality worldwide, characterized by a complex and heterogeneous clinical presentation. Despite advances in patient monitoring and biomarkers, early detection of sepsis in the intensive care unit (ICU) is often hampered by incomplete data and diagnostic uncertainty. In recent years, machine learning models have been proposed as predictive tools, but many function as opaque “black boxes”, meaning that humans are unable to understand algorithmic reasoning, poorly suited to the uncertainty-laden clinical environment of critical care. Even when post-hoc interpretability methods are available for these algorithms, their explanations often remain difficult for non-expert clinicians to understand. Methods: In this clinical perspective, we explore the specific advantages of probabilistic graphical models, particularly Bayesian Networks (BNs) and their dynamic counterparts (DBNs), for sepsis prediction. Results: Recent applications of AI models in sepsis prediction have demonstrated encouraging results, such as DBNs achieving an AUROC of 0.94 in early detection, or causal probabilistic models in hospital admissions (AUROC 0.95). These models explicitly represent clinical reasoning under uncertainty, handle missing data natively, and offer interpretable, transparent decision paths. Drawing on recent studies, including real-time sepsis alert systems and treatment-effect modeling, we highlight concrete clinical applications and their current limitations. Conclusions: We argue that BNs present a great opportunity to bridge the gap between artificial intelligence and bedside care through human-in-the-loop collaboration, transparent inference, and integration into clinical information systems. As critical care continues to move toward data-driven decision-making, Bayesian models may offer not only technical performance but also the epistemic humility needed to support clinicians facing uncertain, high-stakes decisions.","author":[{"family":"Agard","given":"Geoffray"},{"family":"Roman","given":"Christophe"},{"family":"Guervilly","given":"Christophe"},{"family":"Ouladsine","given":"Mustapha"},{"family":"Boyer","given":"Laurent"},{"family":"Hraiech","given":"Sami"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jcm14186463","URL":"https://doi.org/10.3390/jcm14186463","source":"openalex"},{"id":"oa:W4416783267","type":"article-journal","title":"Development and influencing factors of artificial intelligence literacy and computational thinking in Chinese university students","abstract":"This study investigates the developmental status and influencing factors of artificial intelligence (AI) literacy and computational thinking (CT) literacy among undergraduates in China's \"four new\" majors. Guided by the Technology Acceptance Model, Social Cognitive Theory, and Constructivist Learning Theory, the research employs a questionnaire survey to assess student's AI and CT literacy, as well as the impact of subject background and AI tool usage. Statistical analyses (t-test, ANOVA, Pearson correlation) revealed statistically significant positive associations between dimensions of AI literacy and CT; however, effect sizes were uniformly small (|r| < .10), indicating that these associations-while detectable in a large sample-have limited practical magnitude and should be interpreted with caution. Intelligent Thinking exhibited the comparatively strongest association with critical thinking, though the magnitude warrants cautious interpretation. Disciplinary differences are evident: new engineering students excel at algorithmic thinking, while new liberal arts students show strengths in human-machine collaboration. Moreover, group mean differences were observed across usage-frequency categories; however, we did not fit non-linear models, and further research is needed to verify any non-linear patterns. These findings are consistent with the co-development of AIL and CT and may inform the design of discipline-specific, differentiated educational strategies in higher education.","author":[{"family":"Hu","given":"Zhihua"},{"family":"He","given":"Huili"},{"family":"Zhang","given":"Chunqu"},{"family":"Guan","given":"Yurong"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-26888-z","URL":"https://doi.org/10.1038/s41598-025-26888-z","source":"openalex"},{"id":"oa:W4416466328","type":"article-journal","title":"Neuro-symbolic AI for auditable cognitive information extraction from medical reports","abstract":"Large language models (LLMs) such as GPT-4 can interpret free text, but unreliable answers, opaque reasoning, and privacy risks limit their use in healthcare. In contrast, rule-based artificial intelligence (AI) provides transparent and reproducible results but struggles with free text. We aimed to combine the strengths of both approaches to test whether such a hybrid system can autonomously and reliably extract clinical data from diagnostic imaging reports. We developed a neuro-symbolic AI that connects GPT-4 with a rule-based expert system through a semantic integration platform. GPT-4 extracted candidate facts from free-text reports, while the expert system verified them against medical rules, producing traceable, deterministic labels. We evaluated the system on 206 consecutive prostate cancer PET/CT scan reports, requiring extraction of 26 clinical parameters per report, generating 5356 data points, and answering three study questions: study inclusion, recurrent cancer identification, and prostate-specific antigen (PSA) level retrieval. Outputs were compared against physician-derived references, and discrepancies were reviewed by a blinded adjudicator. Here we show that neuro-symbolic AI outperforms GPT-4 alone and matches physicians in structuring and analysing reports. GPT-4 alone achieves F1 scores of 0.63 for study inclusion and 0.95 for recurrence detection, with 96.6% correct PSA values. Physicians reach F1 scores of 1.00 and 0.99, with 98.1% PSA accuracy. The neuro-symbolic AI scores twice 1.00 with 100% PSA accuracy and delivers always an auditable chain of reasoning. It intercepts two intentionally introduced reports with residual identifiers, preventing unintended transfer of sensitive data. Unlike standalone LLMs, neuro-symbolic AI can safely automate data extraction for clinical research and may provide a path toward trustworthy AI in healthcare practice. Medical doctors often write reports as free text, which is hard to reuse for research or care. A large language model is software that reads and writes text by imitating large networks of brain cells. This type of artificial intelligence can extract and organize important information from medical reports. But its reasoning is opaque, answers can be wrong, and it raises privacy concerns. Rule-based artificial intelligence is transparent, responds correctly, and is privacy-protecting but struggles with free text. We combined both artificial intelligence types, so each offsets the other’s weaknesses. We tested the system on 206 prostate cancer imaging reports, where it extracted information correctly, showed how it reached its answers, and protected sensitive data. Pairing large language models with rule-based systems could make artificial intelligence safer, more trustworthy, and more useful in healthcare. Prenosil, Weitzel et al. unify semantically the large language model GPT-4 with a rule-based expert system to turn free-text radiology reports into structured, privacy-safe data. In a proof-of-concept on 206 prostate cancer PET/CT reports, the resulting neuro-symbolic artificial intelligence matches physicians and supports clinical trials at scale.","author":[{"family":"Prenosil","given":"George"},{"family":"Weitzel","given":"Thilo"},{"family":"Bello","given":"Shaibu"},{"family":"Mingels","given":"Clemens"},{"family":"Manzini","given":"Giulia"},{"family":"Meier","given":"Lorenz"},{"family":"Shi","given":"Kuangyu"},{"family":"Rominger","given":"Axel"},{"family":"Afsharoromieh","given":"Ali"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s43856-025-01194-x","URL":"https://doi.org/10.1038/s43856-025-01194-x","source":"openalex"},{"id":"oa:W4409170484","type":"article-journal","title":"Leveraging Artificial Intelligence for Personalized Rehabilitation Programs for Head and Neck Surgery Patients","abstract":"Background: Artificial intelligence (AI) and large language models (LLMs) are increasingly used in healthcare, with applications in clinical decision-making and workflow optimization. In head and neck surgery, postoperative rehabilitation is a complex, multidisciplinary process requiring personalized care. This study evaluates the feasibility of using LLMs to generate tailored rehabilitation programs for patients undergoing major head and neck surgical procedures. Methods: Ten hypothetical head and neck surgical clinical scenarios were developed, representing oncologic resections with complex reconstructions. Four LLMs, ChatGPT-4o, DeepSeek V3, Gemini 2, and Copilot, were prompted with identical queries to generate rehabilitation plans. Three senior clinicians independently assessed their quality, accuracy, and clinical relevance using a five-point Likert scale. Readability and quality metrics, including the DISCERN score, Flesch Reading Ease, Flesch–Kincaid Grade Level, and Coleman–Liau Index, were applied. Results: ChatGPT-4o achieved the highest clinical relevance (Likert mean of 4.90 ± 0.32), followed by DeepSeek V3 (4.00 ± 0.82) and Gemini 2 (3.90 ± 0.74), while Copilot underperformed (2.70 ± 0.82). Gemini 2 produced the most readable content. A statistical analysis confirmed significant differences across the models (p < 0.001). Conclusions: LLMs can generate rehabilitation programs with varying quality and readability. ChatGPT-4o produced the most clinically relevant plans, while Gemini 2 generated more readable content. AI-generated rehabilitation plans may complement existing protocols, but further clinical validation is necessary to assess their impact on patient outcomes.","author":[{"family":"Marcaccini","given":"Gianluca"},{"family":"Seth","given":"Ishith"},{"family":"Novo","given":"Jennifer"},{"family":"Mcclure","given":"Vicki"},{"family":"Sacks","given":"Brett"},{"family":"Lim","given":"Kaiyang"},{"family":"Ng","given":"Sally"},{"family":"Cuomo","given":"Roberto"},{"family":"Rozen","given":"Warren"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/technologies13040142","URL":"https://doi.org/10.3390/technologies13040142","source":"openalex"},{"id":"oa:W4408302723","type":"article-journal","title":"Assessing Physician Confidence in Artificial Intelligence: Insights from Iran","abstract":"This study aims to explore physician confidence in artificial intelligence within the context of Iran's healthcare system. We aim to identify the factors influencing acceptance levels among physicians, particularly focusing on how job status and specialization impact attitudes toward AI. Specifically, we will investigate whether Type of specialization practitioners view AI differently relative to each other and whether academic physicians have distinct perspectives compared to their non-academic counterparts. By addressing these questions, we hope to contribute valuable insights that can facilitate the integration of AI technologies into medical practice and ultimately enhance healthcare delivery in Iran.This study aims to explore physician confidence in artificial intelligence within the context of Iran's healthcare system. We aim to identify the factors influencing acceptance levels among physicians, particularly focusing on how job status and specialization impact attitudes toward AI. Specifically, we will investigate whether Type of specialization practitioners view AI differently relative to each other and whether academic physicians have distinct perspectives compared to their non-academic counterparts. By addressing these questions, we hope to contribute valuable insights that can facilitate the integration of AI technologies into medical practice and ultimately enhance healthcare delivery in Iran.","author":[{"family":"Firuzpour","given":"Faezeh"},{"family":"Abdolalipour","given":"Elaheh"},{"family":"Rezaeiroushan","given":"Narjes"},{"family":"Barancheshmeh","given":"Maryam"},{"family":"Moradi","given":"Niloufar"},{"family":"Targhi","given":"Hadiseh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.61186/ist.202502.02.01","URL":"https://doi.org/10.61186/ist.202502.02.01","source":"openalex"},{"id":"oa:W4407171053","type":"article-journal","title":"Medical students and ChatGPT: analyzing attitudes, practices, and academic perceptions","abstract":"BACKGROUND: ChatGPT, a chatbot launched by OpenAI in November 2022, has generated both excitement and concern within the healthcare education, research, and practice communities. This study aimed to explore the knowledge, perceptions, attitudes, and practices of undergraduate medical students regarding the use of ChatGPT and similar chatbots in their academic work. METHODS: An anonymous, structured questionnaire was developed using Google Forms and administered to medical students as part of a cross-sectional study. The survey targeted undergraduate medical students from four governorates in Egypt. The questionnaire link was distributed through social media platforms, including Facebook and WhatsApp. The survey comprised four sections: socio-demographic characteristics, perceptions, attitudes, and practices. RESULTS: The survey achieved a response rate of 96%, with 614 out of 640 approached students participating. Prior to the study, most respondents (78.5%) had personal experience using it. Overall, respondents demonstrated positive perceptions, attitudes, and practices toward ChatGPT, with mean scores exceeding 3 for all three variables: 3.99 ± 0.60 for perceptions, 3.01 ± 0.46 for attitudes, and 3.55 ± 0.55 for practices. In general, the students exhibited a high degree of trust in the model, with approximately half trusting the accuracy and reliability of the information provided by ChatGPT. However, more than two-thirds expressed apprehension about its potential misuse in medical education, and around 60% were concerned about the accuracy of information ChatGPT might generate on complex medical topics. CONCLUSIONS: Medical students show strong interest and trust in using ChatGPT and similar chatbots for academic purposes but have concerns about the reliability of the information and potential misuse in medical education. The use of AI tools should follow ethical guidelines set by academic institutions, with regular updates to keep pace with technological progress. Future research should focus on the impact of AI on education and personal development, especially among young people.","author":[{"family":"Abdelhafiz","given":"Ahmed"},{"family":"Farghly","given":"Maysa"},{"family":"Sultan","given":"Eman"},{"family":"Abouelmagd","given":"Moaz"},{"family":"Ashmawy","given":"Youssef"},{"family":"Elsebaie","given":"Eman"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12909-025-06731-9","URL":"https://doi.org/10.1186/s12909-025-06731-9","source":"openalex"},{"id":"oa:W4413137421","type":"article-journal","title":"Generative Artificial Intelligence Tools in Medical Research (GAMER): Protocol for a Scoping Review and Development of Reporting Guidelines","abstract":"BACKGROUND: The integration of artificial intelligence (AI) has revolutionized medical research, offering innovative solutions for data collection, patient engagement, and information dissemination. Powerful generative AI (GenAI) tools and other similar chatbots have emerged, facilitating user interactions with virtual conversational agents. However, the increasing use of GenAI tools in medical research presents challenges, including ethical concerns, data privacy issues, and the potential for generating false content. These issues necessitate standardization of reporting to ensure transparency and scientific rigor. OBJECTIVE: The development of the Generative Artificial Intelligence Tools in Medical Research (GAMER) reporting guidelines aims to establish comprehensive, standardized guidelines for reporting the use of GenAI tools in medical research. METHODS: The GAMER guidelines are being developed following the methodology recommended by the Enhancing the Quality and Transparency of Health Research (EQUATOR) Network, involving a scoping review and expert Delphi consensus. The scoping review searched PubMed, Web of Science, Embase, CINAHL, PsycINFO, and Google Scholar (for the first 200 results) using keywords like \"generative AI\" and \"medical research\" to identify reporting elements in GenAI-related studies. The Delphi process involves 30-50 experts with ≥3 years of experience in AI applications or medical research, selected based on publication records and expertise across disciplines (eg, clinicians and data scientists) and regions (eg, Asia and Europe). A 7-point-scale survey will establish consensus on checklist items. The testing phase invites authors to apply the GAMER checklist to GenAI-related manuscripts and provide feedback via a questionnaire, while experts assess reliability (κ statistic) and usability (time taken, 7-point Likert scale). The study has been approved by the Ethics Committee of the Institute of Health Data Science at Lanzhou University (HDS-202406-01). RESULTS: The GAMER project was launched in July 2023 by the Evidence-Based Medicine Center of Lanzhou University and the WHO Collaborating Centre for Guideline Implementation and Knowledge Translation, and it concluded in July 2024. The scoping review was completed in November 2023. The Delphi process was conducted from October 2023 to April 2024. The testing phase began in March 2025 and is ongoing. The expected outcome of the GAMER project is a reporting checklist accompanied by relevant terminology, examples, and explanations to guide stakeholders in better reporting the use of GenAI tools. CONCLUSIONS: GAMER aims to guide researchers, reviewers, and editors in the transparent and scientific application of GenAI tools in medical research. By providing a standardized reporting checklist, GAMER seeks to enhance the clarity, completeness, and integrity of research involving GenAI tools, thereby promoting collaboration, comparability, and cumulative knowledge generation in AI-driven health care technologies. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/64640.","author":[{"family":"Luo","given":"Xufei"},{"family":"Tham","given":"Yih"},{"family":"Daher","given":"Mohammad"},{"family":"Bian","given":"Zhaoxiang"},{"family":"Chen","given":"Yaolong"},{"family":"Estill","given":"Janne"},{"family":"Group","given":"Gamer"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/64640","URL":"https://doi.org/10.2196/64640","source":"openalex"},{"id":"oa:W4406304903","type":"article-journal","title":"Artificial intelligence-enhanced diagnosis of degenerative joint disease using temporomandibular joint panoramic radiography and joint noise data","abstract":"This study aimed to develop an artificial intelligence (AI) model for the screening of degenerative joint disease (DJD) using temporomandibular joint (TMJ) panoramic radiography and joint noise data. A total of 2631 TMJ panoramic images were collected, resulting in a final dataset of 3908 images (2127 normal (N) and 1781 DJD (D)) after excluding indeterminate cases and errors. AI models using GoogleNet were evaluated with six different combinations of image data, clinician-detected crepitus, and patient-reported joint noise. The model that integrated all joint noise data with imaging demonstrated the highest performance, achieving an F1-score of 0.72. Another model, which incorporated both imaging and crepitus, also achieved the same F1-score but had lower D recall (0.55 vs. 0.67) and N precision (0.71 vs. 0.74). The AI models outperformed orofacial pain specialists when provided with imaging alone or in combination with all joint noise data. These findings suggest that AI-enhanced DJD diagnosis using TMJ panoramic radiography and joint noise data offers a promising approach for early detection and improved patient care. The results underscore AI's capability to integrate diverse diagnostic factors, providing a comprehensive and accurate assessment that surpasses traditional methods.","author":[{"family":"Choi","given":"Eunhye"},{"family":"Shin","given":"Seokwon"},{"family":"Lee","given":"Kijin"},{"family":"An","given":"Tien"},{"family":"Lee","given":"Richard"},{"family":"Kim","given":"Sunmin"},{"family":"Kim","given":"Sunmin"},{"family":"Son","given":"Youngdoo"},{"family":"Kim","given":"ST"},{"family":"Kim","given":"ST"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-024-83750-4","URL":"https://doi.org/10.1038/s41598-024-83750-4","source":"openalex"},{"id":"oa:W4416122045","type":"article-journal","title":"Efficiency of Artificial Intelligence in Three-Dimensional Reconstruction of Medical Imaging","abstract":"Three-dimensional (3D) reconstruction is necessary for visualizing complex anatomy and supporting clinical decision-making in radiology. However, traditional techniques often struggle with limitations in scalability, speed, and reproducibility. The recent emergence of artificial intelligence (AI) has enabled a new generation of reconstruction tools that offer greater automation and new clinical capabilities. Motivated by the demand for efficient imaging, researchers have applied deep learning to overcome longstanding barriers, with impacts spanning diagnosis, surgical planning, and disease monitoring. This review included peer-reviewed studies published within the last 10 years, focusing exclusively on adult human imaging published in the English language, as anatomical development, imaging protocols, and clinical decision pathways differ significantly in pediatric populations. While this improved applicability to adult radiology, it limited insight into emerging pediatric and preclinical research. Searches focused on AI-driven 3D reconstruction across different radiologic modalities. Articles were selected based on the following predefined inclusion criteria: adult human participants, AI-based 3D reconstruction, clinical validation, and relevance to radiologic practice. Studies including pediatric or animal subjects, preclinical-only experimentation, non-English-language text, or without applied clinical evaluation were excluded. Our results showed that AI has improved the accuracy, speed, and clinical utility of 3D reconstruction in multiple specialties. Deep learning models such as U-Net, V-Net, DenseVNet, and generative adversarial networks have achieved high segmentation accuracy, often reporting Dice scores >0.90. These models have also been used for tumor detection, surgical planning, and reducing radiation exposure. However, challenges such as high computational requirements, lack of standardized datasets, limited real-world validation, and ethical concerns remain. In conclusion, 3D reconstruction is transforming radiology with more accurate patient-specific images for improved clinical decision-making. While it is rapidly being integrated into practice, these technologies still have limitations that need to be addressed. Nonetheless, with improved and ongoing innovation, AI has the potential to become a catalyst for precise imaging and patient care.","author":[{"family":"Nikhita"},{"family":"Bannur","given":"Dhyaan"},{"family":"Cerdas","given":"Maria"},{"family":"Saeed","given":"Arisha"},{"family":"Imam","given":"Bashir"},{"family":"Thandi","given":"Ravtej"},{"family":"Anusha","given":"HC"},{"family":"Reddy","given":"PH"},{"family":"Ali","given":"Ramsha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.96580","URL":"https://doi.org/10.7759/cureus.96580","source":"openalex"},{"id":"oa:W7124980116","type":"article-journal","title":"Artificial intelligence in the prevention and early detection of postpartum depression: a systematic review and meta-analysis","abstract":"Objective: Postpartum depression is a frequent complication after childbirth, affecting maternal health, infant development, and family well-being. This study evaluated the role of artificial intelligence (AI) in preventing and detecting postpartum depression early. Methods: A systematic search was conducted in Scopus, PubMed, Web of Science, and CINAHL for studies (2020-2025) applying AI to identify postpartum depression. PRISMA guidelines guided selection and appraisal. Two random-effects meta-analyses estimated pooled sensitivity and accuracy based on total sample size and reported metrics. Results: Of 1,857 records, 16 studies met inclusion criteria. Machine learning models (Random Forest, XGBoost, neural networks) showed greater accuracy than traditional methods. Integration of AI with medical records and social media data enabled earlier, personalized detection. Reported challenges included algorithmic bias, data privacy, and implementation barriers. Pooled sensitivity was 69% (95% CI: 55-81%; n=277,496) and accuracy 79% (95% CI: 73-85%; n=306,156). Conclusions: AI shows promise for enhancing postpartum depression detection and prevention but requires addressing ethical, technical, and educational challenges to achieve equitable clinical integration. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420251004175, identifier CRD420251004175.","author":[{"family":"Ruger-Navarrete","given":"Azahara"},{"family":"Gómez-Ferrera","given":"María"},{"family":"Mérida-Yáñez","given":"Beatriz"},{"family":"Vázquez-Lara","given":"Juana"},{"family":"Gómez-Salgado","given":"Juan"},{"family":"García-Oliva","given":"Sofía"},{"family":"Vázquez-Lara","given":"María"},{"family":"Rodríguez-Díaz","given":"Luciano"},{"family":"Antúnez-Calvente","given":"Irene"},{"family":"Fernández-Carrasco","given":"Francisco"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fpsyt.2025.1734102","URL":"https://doi.org/10.3389/fpsyt.2025.1734102","source":"pubmed"},{"id":"oa:W7128732859","type":"article-journal","title":"AIONS Consensus Conference on Definitions of Artificial Intelligence Surgery, Surgomics and Robotics","abstract":"This Consensus Statement was jointly developed by the Editorial Board Members of Artificial Intelligence Surgery and the Artificial Intelligence Organization for the Next Generation of Surgeons (AIONS). The initiative began in February 2025 and proceeded through iterative drafting of definitions, online meetings, expert subgroup revisions, an online validation survey, an in-person Consensus Conference, and a final online meeting to confirm revisions and review the conference manuscript. Votes greater than or equal to 80 percent were considered validating. Definitions were sought for: (1) Surgery, (2) Endoluminal Surgery, (3) Percutaneous Surgery, (4) Robot, (5) Surgical Robot, (6) Robot-Assisted Surgery, (7) Telemanipulator Surgery, (8) Remote Surgery, (9) Collaborative Robotic (Cobotic) Surgery, (10) Robotic Surgery, (11) Artificial Intelligence Surgery, (12) Surgomics, (13) Surgical Multiomics, (14) Non-Invasive Surgery, (15) Digital Surgery, (16) Computer-Assisted Surgery, and (17) Cybersurgery. All candidate definitions achieved at least 80 percent approval in an online vote prior to the Consensus Conference. The in-person meeting occurred on 26 September 2025 at the Orto Botanico, University of Padova, Italy, where 11 definitions were ratified. The definition of Surgery was deemed premature and invalidated. Surgomics and Surgical Multiomics were determined to be distinct entities and were therefore revoted online after the meeting. Collaborative Robotics was clarified as requiring co-local presence of the surgeon and robot. Definitions for Percutaneous Surgery and Robot were amended and validated during a follow-up online vote on 11 November. Ultimately, all 17 definitions were validated. This Consensus provides terminology, rationale, and strategic direction for the surgical field as artificial intelligence, robotics, and data science reshape surgical practice. Future Consensus Conferences are planned to update definitions as the field evolves.","author":[{"family":"Gumbs","given":"Andrew"},{"family":"Michele","given":"Diana"},{"family":"Rawicz-Prusyński","given":"Karol"},{"family":"Spolverato","given":"Gaya"},{"family":"Frigerio","given":"Isabella"},{"family":"Hilal","given":"Mohammad"},{"family":"Bannone","given":"Elisa"},{"family":"Croner","given":"Roland"},{"family":"Mas","given":"Francesca"},{"family":"Simone","given":"Belinda"},{"family":"Friebe","given":"Michael"},{"family":"Giovinazzo","given":"Francesco"},{"family":"Grasso","given":"SV"},{"family":"Ishizawa","given":"Takeaki"},{"family":"Karcz","given":"Konrad"},{"family":"Khalpey","given":"Zain"},{"family":"Milone","given":"Luca"},{"family":"Messaoudi","given":"Nouredin"},{"family":"Ozmen","given":"MM"},{"family":"Passias","given":"Peter"},{"family":"Rashidian","given":"Niki"},{"family":"Ross","given":"Sharona"},{"family":"Schnelldorfer","given":"Thomas"},{"family":"Szold","given":"Amir"},{"family":"Nawrat","given":"Zbigniew"},{"family":"Dagher","given":"Ibrahim"}],"issued":{"date-parts":[[2026]]},"DOI":"10.20517/ais.2025.113","URL":"https://doi.org/10.20517/ais.2025.113","source":"openalex"},{"id":"oa:W4417114461","type":"article-journal","title":"Unveiling the Algorithm: The Role of Explainable Artificial Intelligence in Modern Surgery","abstract":"Artificial Intelligence (AI) is rapidly transforming surgical care by enabling more accurate diagnosis and risk prediction, personalized decision-making, real-time intraoperative support, and postoperative management. Ongoing trends such as multi-task learning, real-time integration, and clinician-centered design suggest AI is maturing into a safe, pragmatic asset in surgical care. Yet, significant challenges, such as the complexity and opacity of many AI models (particularly deep learning), transparency, bias, data sharing, and equitable deployment, must be surpassed to achieve clinical trust, ethical use, and regulatory approval of AI algorithms in healthcare. Explainable Artificial Intelligence (XAI) is an emerging field that plays an important role in bridging the gap between algorithmic power and clinical use as surgery becomes increasingly data-driven. The authors reviewed current applications of XAI in the context of surgery-preoperative risk assessment, surgical planning, intraoperative guidance, and postoperative monitoring-and highlighted the absence of these mechanisms in Generative AI (e.g., ChatGPT). XAI will allow surgeons to interpret, validate, and trust AI tools. XAI applied in surgery is not a luxury: it must be a prerequisite for responsible innovation. Model bias, overfitting, and user interface design are key challenges that need to be overcome and will be explored in this review to achieve the integration of XAI into the surgical field. Unveiling the algorithm is the first step toward a safe, accountable, transparent, and human-centered surgical AI.","author":[{"family":"Lopes","given":"Sara"},{"family":"Mascarenhas","given":"Miguel"},{"family":"Fonseca","given":"João"},{"family":"Fernandes","given":"Gabriela"},{"family":"Leitemoreira","given":"Adelino"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/healthcare13243208","URL":"https://doi.org/10.3390/healthcare13243208","source":"openalex"},{"id":"oa:W4412990763","type":"article-journal","title":"A Narrative Review of Theranostics in Neuro-Oncology: Advancing Brain Tumor Diagnosis and Treatment Through Nuclear Medicine and Artificial Intelligence","abstract":"This narrative review explores the integration of theranostics and artificial intelligence (AI) in neuro-oncology, addressing the urgent need for improved diagnostic and treatment strategies for brain tumors, including gliomas, meningiomas, and pediatric central nervous system neoplasms. A comprehensive literature search was conducted through PubMed, Scopus, and Embase for articles published between January 2020 and May 2025, focusing on recent clinical and preclinical advancements in personalized neuro-oncology. The review synthesizes evidence on novel theranostic agents-such as Lu-177-based radiopharmaceuticals, CXCR4-targeted PET tracers, and multifunctional nanoparticles-and highlights the role of AI in enhancing tumor detection, segmentation, and treatment planning through advanced imaging analysis, radiogenomics, and predictive modeling. Key findings include the emergence of nanotheranostics for targeted drug delivery and real-time monitoring, the application of AI-driven algorithms for improved image interpretation and therapy guidance, and the identification of current limitations such as data standardization, regulatory challenges, and limited multicenter validation. The review concludes that the convergence of AI and theranostic technologies holds significant promise for advancing precision medicine in neuro-oncology, but emphasizes the need for collaborative, multidisciplinary research to overcome existing barriers and enable widespread clinical adoption.","author":[{"family":"Christodoulou","given":"Rafail"},{"family":"Papageorgiou","given":"Platon"},{"family":"Pitsillos","given":"Rafael"},{"family":"Woodward","given":"Amanda"},{"family":"Papageorgiou","given":"Sokratis"},{"family":"Solomou","given":"Elena"},{"family":"Georgiou","given":"Michalis"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ijms26157396","URL":"https://doi.org/10.3390/ijms26157396","source":"openalex"},{"id":"oa:W4407660154","type":"article-journal","title":"Explainable Artificial Intelligence Models for Predicting Depression Based on Polysomnographic Phenotypes","abstract":"Depression is a common mental health disorder and a leading contributor to mortality and morbidity. Despite several advancements, the current screening methods have limitations in enabling the robust and automated detection of depression, thereby hindering early diagnosis and timely intervention. This study aimed to develop explainable artificial intelligence (AI) models to predict depression using polysomnographic phenotype data, ensuring high predictive performance while providing clear insights into the importance of features influencing the risk of depression. Advanced machine learning algorithms such as random forest, extreme gradient boosting, categorical boosting, and light gradient boosting machines were employed to train and validate the predictive AI models. Phenotype data from subjective health questionnaires, clinical assessments, and demographic factors were analyzed. The explainable AI models identified the important features, and their performance was evaluated using cross-validation. The study population, comprising 114 control participants and 39 individuals with depression, was stratified based on validated depression-scoring methods. The proposed explainable AI models achieved an F1-score of 85%, verifying their high reliability in predicting depression. Key features influencing the risk of depression, such as anxiety disorders, sleep efficiency, and demographic factors, offer actionable insights for clinical practice, highlighting the transparency of these models. This study proposed and developed explainable AI models based on polysomnographic phenotype data for the automated detection of depression and verified that these models help improve mental health diagnostics, enabling timely interventions.","author":[{"family":"Enkhbayar","given":"Doljinsuren"},{"family":"Ko","given":"Jonathan"},{"family":"Oh","given":"Sejong"},{"family":"Ferdushi","given":"Rumana"},{"family":"Kim","given":"Jae"},{"family":"Key","given":"Jaehong"},{"family":"Urtnasan","given":"Erdenebayar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bioengineering12020186","URL":"https://doi.org/10.3390/bioengineering12020186","source":"openalex"},{"id":"oa:W7118566795","type":"article-journal","title":"Breeding Smarter: Artificial Intelligence and Machine Learning Tools in Modern Breeding—A Review","abstract":"Climate challenges, along with a projected global population increase of 2 billion by 2080, are intensifying pressures on agricultural systems, leading to biodiversity loss, land use constrains, soil fertility declining, and changes in water cycles, while crop yields struggle to meet the rising food demand. These challenges, coupled with evolving legislation and rapid technology advancements, require innovative sustainable agricultural solutions. By reshaping farmers’ daily operations, real-time data acquisition and predictive models can support informed decision-making. In this context, smart farming (SM) applied to plant breeding can improve efficiency by reducing inputs and increasing outputs through the adoption of digital and data-driven technologies. Examples include the investment on common ontologies and metadata standards for phenotypes and environments, standardization of HTP protocols, integration of prediction outputs into breeding databases, and selection workflows, as well in building multi-partner field networks that collect diverse envirotypes. This review outlines how AI and machine learning (ML) can be integrated in modern plant breeding methodologies, including genomic selection (GS) and genetic algorithms (GAs), to accelerate the development of climate-resilient and sustainably performing crop varieties. While many reviews address smart farming or smart breeding independently, herein, these domains are bridged to provide an understandable strategic landscape by enhancing breeding efficiency.","author":[{"family":"Garcia-Oliveira","given":"Ana"},{"family":"Dwivedi","given":"Sangam"},{"family":"Chander","given":"Subhash"},{"family":"Nelimor","given":"Charles"},{"family":"Moneim","given":"Diaa"},{"family":"Ortíz","given":"Rodomiro"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/agronomy16010137","URL":"https://doi.org/10.3390/agronomy16010137","source":"openalex"},{"id":"oa:W4408245445","type":"article-journal","title":"Artificial intelligence for early detection of lung cancer in GPs’ clinical notes: a retrospective observational cohort study","abstract":"BACKGROUND: The journey of >80% of patients diagnosed with lung cancer starts in general practice. About 75% of patients are diagnosed when it is at an advanced stage (3 or 4), leading to >80% mortality within 1 year at present. The long-term data in GP records might contain hidden information that could be used for earlier case finding of patients with cancer. AIM: To develop new prediction tools that improve the risk assessment for lung cancer. DESIGN AND SETTING: Text analysis of electronic patient data using natural language processing and machine learning in the general practice files of four networks in the Netherlands. METHOD: Files of 525 526 patients were analysed, of whom 2386 were diagnosed with lung cancer. Diagnoses were validated by using the Dutch cancer registry, and both structured and free-text data were used to predict the diagnosis of lung cancer 5 months before diagnosis (4 months before referral). RESULTS: The algorithm could facilitate earlier detection of lung cancer using routine general practice data. Discrimination, calibration, sensitivity, and specificity were established under various cut-off points of the prediction 5 months before diagnosis. Internal validation of the best model demonstrated an area under the curve of 0.88 (95% confidence interval [CI] = 0.86 to 0.89), which shrunk to 0.79 (95% CI = 0.78 to 0.80) during external validation. The desired sensitivity determines the number of patients to be referred to detect one patient with lung cancer. CONCLUSION: Artificial intelligence-based support enables earlier detection of lung cancer in general practice using readily available text in the patient files of GPs, but needs additional prospective clinical evaluation.","author":[{"family":"Schut","given":"Martijn"},{"family":"Luik","given":"Torec"},{"family":"Vagliano","given":"Iacopo"},{"family":"Rios","given":"Miguel"},{"family":"Helsper","given":"Charles"},{"family":"Asselt","given":"Kristel"},{"family":"Wit","given":"Niek"},{"family":"Abuhanna","given":"Ameen"},{"family":"Weert","given":"Henk"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3399/bjgp.2023.0489","URL":"https://doi.org/10.3399/bjgp.2023.0489","source":"openalex"},{"id":"oa:W4412047269","type":"article-journal","title":"A model based on artificial intelligence for the prediction, prevention and patient-centred approach for non-communicable diseases related to metabolic syndrome","abstract":"Metabolic syndrome (MetS) is related to non-communicable diseases (NCDs) such as type 2 diabetes (T2D), metabolic-associated steatotic liver disease (MASLD), atherogenic dyslipidaemia (ATD), and chronic kidney disease (CKD). The absence of reliable tools for early diagnosis and risk stratification leads to delayed detection, preventable hospitalizations, and increased healthcare costs. This study evaluates the impact of Transformer-based artificial intelligence (AI) model in predicting and managing MetS-related NCDs compared to classical machine learning models. Electronical medical data registered in the MIMIC-IV v2.2database from 183 958 patients with at least two recorded medical visits were analysed. A two-stage AI approach was implemented: (1) pretraining on 60% of the dataset to capture disease progression patterns, and (2) fine-tuning on the remaining 40% for disease-specific predictions. Transformer-based models was compared with traditional machine learning approaches (Random Forest and Linear Support Vector Classifier [SVC]), evaluating predictive performance through AUC and F1-score. The Transformer-based model significantly outperformed classical models, achieving higher AUC values across all diseases. It also identified a substantial number of undiagnosed cases compared to documented diagnoses fold increase for CKD 2.58, T2D 0.78, dyslipidaemia 1.89, hypertension 3.33, MASLD 5.78, and obesity 4.07. Diagnosis delays ranged from 90 to 500 days, with 35% of missed intervention opportunities occurring within the first five appointments. These delays correlated with an 84% increase in hospitalizations and a 69% rise in medical procedures. This study demonstrates that Transformer-based AI models offer superior predictive accuracy over traditional methods by capturing complex temporal disease patterns. Their integration into clinical workflows and public health strategies could enable scalable, proactive MetS management, reducing undiagnosed cases, optimizing resource allocation, and improving population health outcomes.","author":[{"family":"Clarós","given":"Alejandro"},{"family":"Ciudin","given":"Andreea"},{"family":"Muria","given":"Jordi"},{"family":"Llull","given":"Lluis"},{"family":"Mola","given":"Jose"},{"family":"Pons","given":"Martí"},{"family":"Castán","given":"Javier"},{"family":"Cruz","given":"Juan"},{"family":"Simó","given":"Rafael"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/eurpub/ckaf098","URL":"https://doi.org/10.1093/eurpub/ckaf098","source":"openalex"},{"id":"oa:W4413035028","type":"article-journal","title":"The impact of an artificial intelligence enhancement program on healthcare providers’ knowledge, attitudes, and workplace flourishing","abstract":"Background The integration of AI into healthcare influences healthcare providers’ knowledge, attitudes, and workplace flourishing, grounded in key theoretical frameworks. Social cognitive theory suggests AI-enhanced programs may shape knowledge acquisition and decision-making. The Theory of Planned Behavior helps explain how perceptions of AI affect professional attitudes. Meanwhile, workplace flourishing aligns with positive organizational psychology, emphasizing autonomy and engagement factors potentially impacted by AI adoption. We aimed to examine the impact of artificial intelligence enhancement programs on the knowledge, attitudes, and workplace flourishing of healthcare providers. Methods The present study was a quasi-experimental study conducted on healthcare providers at Zagazig University Hospital. The data was gathered using a self-administered three-domain tool, including an artificial intelligence knowledge domain, general attitudes toward artificial intelligence domain, and a workplace flourishing domain. Results Regarding the artificial intelligence technologies knowledge, attitude, and flourishing at work scales, post-intervention scores of all domains showed a statistically significant increase compared to pre-intervention, with a percent increase in knowledge score, attitude, and flourishing at work score were 123.14, 74.28, and 10.63%, respectively. Post-intervention attitude score was significantly positively correlated with knowledge score ( p = 0.001). In addition, age and years of experience were negatively correlated with changes in knowledge and attitude. Conclusion Artificial intelligence training is essential for enhancing healthcare providers’ knowledge and alleviating their concerns regarding its integration into healthcare. Clinical trial registration Identifier PACTR202403647083094; https://pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=27347","author":[{"family":"Nofal","given":"Hanaa"},{"family":"Mohamed","given":"Amal"},{"family":"Almadani","given":"Noura"},{"family":"Mahfouz","given":"Rasha"},{"family":"Bahri","given":"Hibah"},{"family":"Ali","given":"Hossam"},{"family":"Elrafey","given":"Dina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fpubh.2025.1639333","URL":"https://doi.org/10.3389/fpubh.2025.1639333","source":"openalex"},{"id":"oa:W4413323193","type":"article-journal","title":"A high-resolution, nanopore-based artificial intelligence assay for DNA replication stress in human cancer cells","abstract":"DNA replication stress is a hallmark of cancer that is exploited by chemotherapies. Current assays for replication stress have low throughput and poor resolution whilst being unable to map the movement of replication forks genome-wide. We present a new method that uses nanopore sequencing and artificial intelligence to map forks and measure their rates of movement and stalling in melanoma and colon cancer cells treated with chemotherapies. Our method can differentiate between fork slowing and fork stalling in cells treated with hydroxyurea, as well as inhibitors of ATR, WEE1, and PARP1. These different therapies yield different characteristic signatures of replication stress. We assess the role of the intra-S-phase checkpoint on fork slowing and stalling and show that replication stress dynamically changes over S-phase. Finally, we demonstrate that this method is applicable and consistent across two different flow cell chemistries (R9.4.1 and R10.4.1) from Oxford Nanopore Technologies. This method requires sequencing on only one nanopore flow cell per sample, and the cost-effectiveness enables functional screens to determine how human cancers respond to replication-targeted therapies.","author":[{"family":"Jones","given":"Mathew"},{"family":"Kumar","given":"Subash"},{"family":"Pfuderer","given":"Pauline"},{"family":"Bonfimmelo","given":"Alexis"},{"family":"Pagan","given":"Julia"},{"family":"Clarke","given":"Paul"},{"family":"Totañes","given":"Francis"},{"family":"Merrick","given":"Catherine"},{"family":"Mcclelland","given":"Sarah"},{"family":"Boemo","given":"Michael"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41467-025-63168-w","URL":"https://doi.org/10.1038/s41467-025-63168-w","source":"openalex"},{"id":"oa:W4417359252","type":"article-journal","title":"Artificial Intelligence and the Future of Cardiac Implantable Electronic Devices: Diagnostics, Monitoring, and Therapy","abstract":"Cardiac implantable electronic devices (CIEDs) such as pacemakers, implantable cardioverter-defibrillators (ICDs), and cardiac resynchronisation therapy (CRT) devices are generating unprecedented volumes of data in both inpatient and remote settings. Artificial intelligence (AI) techniques are increasingly being applied to enhance the management of these devices and the patients who rely on them. Recent advances demonstrate that machine learning (ML) and deep learning (DL) can improve diagnostic capabilities (for example, by detecting arrhythmias and predicting clinical events), streamline remote monitoring workflows, and optimise device-based therapies. Key applications include AI-driven algorithms that accurately detect true arrhythmias while filtering out false alerts from pacemakers and implantable monitors, neural network models that predict ventricular arrhythmias weeks before ICD shocks, and personalised models that forecast which heart failure patients will respond to CRT. Moreover, novel approaches such as natural language processing (NLP) and reinforcement learning are being explored to integrate diverse data sources and to enable devices to self-adjust their programming. This narrative review summarises the major applications of AI in the CIED domain-diagnostics, remote monitoring, and therapy optimisation-with an emphasis on the recent literature over the past five years. The review highlights important studies and randomised trials in each area, discusses the variety of AI techniques employed, and outlines future directions and challenges (including data standardisation, validation in clinical trials, and regulatory considerations) for translating these innovations into routine clinical care.","author":[{"family":"Antoun","given":"Ibrahim"},{"family":"Alkhayer","given":"Alkassem"},{"family":"Abdelrazik","given":"Ahmed"},{"family":"Eldesouky","given":"Mahmoud"},{"family":"Thu","given":"Kaung"},{"family":"Dhutia","given":"Harshil"},{"family":"Somani","given":"Riyaz"},{"family":"Ng","given":"GA"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jcm14248824","URL":"https://doi.org/10.3390/jcm14248824","source":"openalex"},{"id":"oa:W4410898059","type":"article-journal","title":"When One Size Does not Fit All—Artificial Intelligence in Australian Rural Health","abstract":"AIMS: Artificial intelligence (AI) is having an increasing impact on many aspects of our day-to-day lives. This change is also true in healthcare, with various tools being developed to hasten burdensome administrative tasks and increase overall healthcare efficiency, particularly in metropolitan centres. CONTEXT: AI has remained comparatively clear of rural, regional and remote Australian hospitals, where it has the potential to provide significant benefits. Like previous health technology implementations, rural workforce requirements for AI maintenance and support may hinder AI deployment in these areas. While AI has been implemented successfully in metropolitan areas, these models may have limited translatability to rural health settings with significantly different administrative and healthcare systems. APPROACH: AI may assist with key issues in rural centres such as resource allocation and timely patient transfer for higher level care. While the potential benefits of AI in rural centres are clear, one must consider key factors in rural centres that may limit the success of AI in these hospitals. Smaller rural populations may limit the ability to train location-specific models, and connectivity issues may impede their effective use. CONCLUSION: Specific efforts are required to realise potential benefits of medical AI for rural Australia; addressing connectivity and workforce issues in rural areas is vital to allow for AI and large language models to help benefit rural centres.","author":[{"family":"Hains","given":"Lewis"},{"family":"Kovoor","given":"J"},{"family":"Stretton","given":"Brandon"},{"family":"Gupta","given":"Aashray"},{"family":"Zaka","given":"Ammar"},{"family":"Carmichael","given":"Gavin"},{"family":"Kefalianos","given":"John"},{"family":"Sin","given":"Win"},{"family":"Leslie","given":"Alasdair"},{"family":"Booth","given":"Andrew"},{"family":"Satheakeerthy","given":"Shrirajh"},{"family":"Beath","given":"Alexander"},{"family":"Arafat","given":"Yasser"},{"family":"Jacob","given":"Mathew"},{"family":"Bruening","given":"Martin"},{"family":"Chan","given":"Wengonn"},{"family":"Bacchi","given":"Stephen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/ajr.70037","URL":"https://doi.org/10.1111/ajr.70037","source":"openalex"},{"id":"oa:W4413304061","type":"article-journal","title":"University english teaching evaluation using artificial intelligence and data mining technology","abstract":"This work intends to drive reform and innovation in English teaching evaluation and support personalized English instruction. It utilizes deep learning (DL) and artificial intelligence (AI)-driven data mining technology to explore a reliable and efficient method for university English teaching evaluation. By employing DL, this work explores innovative English teaching models and introduces a Bayesian framework to enable personalized teaching strategies. In the data mining process, the Transformer architecture is applied to English teaching evaluations. This capitalizes on its powerful feature extraction and sequence modeling capabilities to gain a comprehensive understanding and precise evaluation of students' English proficiency. Additionally, an AI-based method for English teaching evaluation is proposed. Data from the English teaching and evaluation system for Computer Science students in the 2018 class at Tianjin University of Science and Technology are collected, analyzed, and processed. Group profiles of students are created to predict exam outcomes. The findings show that over 70% of students engage in active English learning only occasionally, with a higher proportion among females. More than 80% of males recognize the importance of listening and speaking skills, a sentiment shared by over 90% of female students. In terms of factors influencing students' passing exams, scores in various question types play a central role, significantly impacting final grades. These scores reflect students' mastery of English knowledge and application abilities. This work applies the Transformer architecture from natural language processing to the education domain, achieving interdisciplinary integration and innovation. This cross-disciplinary approach not only enriches teaching assessment methods but also provides new solutions for broader educational challenges. The proposed method enhances the objectivity and accuracy of teaching evaluation, minimizing the influence of human bias assessment results.","author":[{"family":"Huang","given":"Qiuyang"},{"family":"Li","given":"Wenling"},{"family":"Muhamad","given":"Mohd"},{"family":"Nawi","given":"Nur"},{"family":"Liu","given":"Xutao"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-16498-0","URL":"https://doi.org/10.1038/s41598-025-16498-0","source":"openalex"},{"id":"oa:W4412423297","type":"article-journal","title":"Technology Landscape Review of In-Sensor Photonic Intelligence: From Optical Sensors to Smart Devices","abstract":"Optical sensors have undergone significant evolution, transitioning from discrete optical microsystems toward sophisticated photonic integrated circuits (PICs) that leverage artificial intelligence (AI) for enhanced functionality. This review systematically explores the integration of optical sensing technologies with AI, charting the advancement from conventional optical microsystems to AI-driven smart devices. First, we examine classical optical sensing methodologies, including refractive index sensing, surface-enhanced infrared absorption (SEIRA), surface-enhanced Raman spectroscopy (SERS), surface plasmon-enhanced chiral spectroscopy, and surface-enhanced fluorescence (SEF) spectroscopy, highlighting their principles, capabilities, and limitations. Subsequently, we analyze the architecture of PIC-based sensing platforms, emphasizing their miniaturization, scalability, and real-time detection performance. This review then introduces the emerging paradigm of in-sensor computing, where AI algorithms are integrated directly within photonic devices, enabling real-time data processing, decision making, and enhanced system autonomy. Finally, we offer a comprehensive outlook on current technological challenges and future research directions, addressing integration complexity, material compatibility, and data processing bottlenecks. This review provides timely insights into the transformative potential of AI-enhanced PIC sensors, setting the stage for future innovations in autonomous, intelligent sensing applications.","author":[{"family":"Zhou","given":"Hong"},{"family":"Li","given":"Dongxiao"},{"family":"Lee","given":"Chengkuo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/aisens1010005","URL":"https://doi.org/10.3390/aisens1010005","source":"openalex"},{"id":"oa:W4408310331","type":"article-journal","title":"Artificial Intelligence Assisted Creativity: Conceptualization, Instrument Development and Validation","abstract":"ABSTRACT Considering the pivotal role of creativity across various eras and the rapid integration of Artificial Intelligence (AI) in both creative processes and education, this study introduces and provides validity evidence for the AI‐assisted Creativity Questionnaire (AICQ). This new 16‐item instrument aims to quantify human creative potential in AI‐assisted endeavors. Initially, a diverse cohort of 322 university students in Taiwan completed the AICQ from November to December 2023. Through exploratory factor analysis (EFA) of responses from this sample, three distinct factors emerged: (1) AI‐assisted functional creativity (AIFC), (2) AI‐assisted visual artistic creativity (AIVAC), and (3) AI‐assisted ideational creativity (AIIC). One item was removed due to cross‐loading. Subsequently, in September 2024, 330 university students in Taiwan engaged with both the AICQ and the 28‐item Creative Behavior Inventory (CBI). Based on responses from this sample, construct validity evidence for the AICQ was examined using confirmatory factor analysis (CFA), which affirmed that a three‐factor model provided a good fit to the data. Gender differences in AICQ scores were found in these subsequent data. Overall, data from participants provided evidence for the reliability as well as convergent, concurrent, and discriminant validity of the scores obtained from the AICQ.","author":[{"family":"Mok","given":"Pui"},{"family":"Chuang","given":"Hsueh‐hua"},{"family":"Cheng","given":"Ming‐min"},{"family":"Smith","given":"Thomas"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/jocb.70004","URL":"https://doi.org/10.1002/jocb.70004","source":"openalex"},{"id":"doi:10.1186/s12909-025-06640-x","type":"article-journal","title":"Design strategies for artificial intelligence based future learning centers in medical universities","abstract":"BACKGROUND: This study explores the acceptance of artificial intelligence(AI) tools in medical students and its influencing factors, thus providing theoretical basis and practical guidance for the construction of future learning centers in medical universities. METHODS: This study comprehensively applied the unified theory of acceptance and use of technology(UTAUT), expectancy confirmation theory (ECT), and innovation diffusion theory (IDT) to analyze the data through structural equation modeling. RESULTS: Effort expectancy (EE), facilitating condition (FC), social influence (SI), and satisfaction (SA) significantly influence medical students' continuance intention (CI) to use artificial intelligence tools. Relative advantage (RA) has a significant impact on medical students' satisfaction (SA) with artificial intelligence tools. Personal innovativeness (PI) plays a significant positive moderating role in the relationships between facilitating condition (FC) and continuance intention (CI), as well as between satisfaction (SA) and continuance intention (CI). CONCLUSIONS: The construction of AI-based future learning centers in medical universities should attach importance to providing personalized learning paths, ensuring technical support and training, creating a collaborative and innovative environment, and showcasing the comparative advantage of tools.","author":[{"family":"Xiaowen","given":"Yang"},{"family":"Jingjing","given":"Ding"},{"family":"Biao","given":"Wang"},{"family":"Shenzhong","given":"Zhang"},{"family":"Yana","given":"Wu"},{"family":"Ding","given":"Jingjing"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12909-025-06640-x","URL":"https://doi.org/10.1186/s12909-025-06640-x","source":"openalex"},{"id":"doi:10.1093/bjrai/ubae017","type":"article-journal","title":"Multimodal artificial intelligence models for radiology","abstract":"Abstract Artificial intelligence (AI) models in medicine often fall short in real-world deployment due to inability to incorporate multiple data modalities in their decision-making process as clinicians do. Clinicians integrate evidence and signals from multiple data sources like radiology images, patient clinical status as recorded in electronic health records, consultations from fellow providers, and even subtle clues using the appearance of a patient, when making decisions about diagnosis or treatment. To bridge this gap, significant research effort has focused on building fusion models capable of harnessing multi-modal data for advanced decision making. We present a broad overview of the landscape of research in multimodal AI for radiology covering a wide variety of approaches from traditional fusion modelling to modern vision-language models. We provide analysis of comparative merits and drawbacks of each approach to assist future research and highlight ethical consideration in developing multimodal AI. In practice, the quality and quantity of available training data, availability of computational resources, and clinical application dictates which fusion method may be most suitable.","author":[{"family":"Tariq","given":"Amara"},{"family":"Banerjee","given":"Imon"},{"family":"Trivedi","given":"Hari"},{"family":"Gichoya","given":"Judy"},{"family":"Gichoya","given":"Judy"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/bjrai/ubae017","URL":"https://doi.org/10.1093/bjrai/ubae017","source":"openalex"},{"id":"doi:10.2196/88511","type":"article-journal","title":"AI Integration in Spanish Undergraduate Medical Education: National Cross-Sectional Study.","abstract":"Background: Artificial intelligence (AI) is reshaping clinical practice and redefining the competencies future physicians will need. International bodies, such as the Association of American Medical Colleges, have called for structured AI training in medical curricula. Despite growing international consensus, no systematic nationwide evaluation had been conducted in Spain prior to this study. Objective: This study aimed to characterize the presence, type, and curricular features of AI-related training across all Spanish universities offering an official medical degree and to assess differences by institutional ownership and geographic region. Methods: This cross-sectional study was conducted from July to September 2025. Universities were the unit of analysis. A census of all institutions offering an officially recognized medical degree was obtained from the Register of Universities, Centers and Degrees; all 52 eligible institutions were included. Publicly available curricula and course guides for the 2025-2026 academic year were reviewed by 2 independent researchers and validated by an external evaluator. Courses were classified as (1) a specific AI course (AI as primary topic, accounting for >50% of syllabus), (2) an AI-similar course (a digital health or biomedical informatics course referencing AI as secondary content), or (3) not AI-related training. Course-level variables included ownership (public or private), region, status (compulsory or elective), European Credit Transfer and Accumulation System (ECTS) credits, academic year, and department. All analyses were descriptive. Potential sources of bias were addressed through predefined classification criteria, duplicate independent extraction, and external dataset verification. Results: Of 52 universities, 36 (69.2%) were public and 16 (30.8%) were private. A total of 10 (19.2%) institutions offered at least one specific AI course; 6 (11.5%) included an AI-similar course. Overall, 16 (30.8%) universities had incorporated AI in some form; 36 (69.2%) institutions had not incorporated AI. Rates were similar for public (7/36, 19.4%) and private institutions (3/16, 18.8%). Identified courses ranged from 3 to 6 ECTS credits, representing an average of 1.17% of the 360-credit degree; most were elective. Only the University of Jaén offered a compulsory course with AI content. Marked regional disparities were observed: Andalusia led with 5 of 9 (55.6%) universities offering a specific AI course, while 10 autonomous communities had no universities with any AI-related training. Conclusions: This study delivers the first census-based, reproducible, national assessment of AI integration in Spanish undergraduate medical education. Unlike prior work focused on individual programs or nonstandardized definitions, we applied a consistent taxonomic framework reusable for longitudinal monitoring and international benchmarking. Findings reveal a heterogeneous, predominantly elective, and low-weight curricular landscape with striking interregional inequities. These results inform curriculum reform, accreditation standards, and faculty development priorities and support the establishment of minimum national competency standards and systematic monitoring to ensure equitable AI literacy among future physicians in Spain.","author":[{"family":"Janeiro","given":"Ana"},{"family":"Pereira","given":"Karina"},{"family":"Mayol","given":"Julio"},{"family":"Crespo","given":"Javier"},{"family":"Carballo","given":"Fernando"},{"family":"Cabello","given":"Juan"},{"family":"Ramos-Casals","given":"Manuel"},{"family":"Corbacho","given":"Bibiana"},{"family":"Turnés","given":"Juan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.2196/88511","URL":"https://doi.org/10.2196/88511","source":"europepmc"},{"id":"doi:10.21203/rs.3.rs-8051581/v1","type":"article-journal","title":"Deploying Medical AI in Low-Resource Settings: A Scoping Review of Challenges and Strategies","abstract":"Abstract Artificial intelligence (AI) is transforming global healthcare by improving diagnostic accuracy, efficiency, and clinical decision-making. However, its implementation in low-resource settings (LRS) remains constrained by weak digital infrastructure, fragmented data systems, and limited governance capacity. This human-centered scoping review synthesizes recent evidence to identify the main challenges and practical strategies for deploying medical AI in low- and middle-income countries (LMICs). A total of thirty peer-reviewed Q1/Q2 studies published between 2020 and 2025 were analyzed thematically across four domains: digital infrastructure and connectivity, data quality and local capacity, ethics and governance, and policy and sustainability. The findings reveal that successful AI deployment in LMICs depends less on algorithmic sophistication and more on stable systems, trustworthy data, and empowered professionals. Key barriers include unstable electricity, limited internet access, outdated hardware, insufficient AI literacy, and weak governance. Recommended strategies emphasize investing in resilient digital infrastructure, building interoperable data repositories based on HL7/FHIR standards, expanding continuous training programs, establishing fairness audits, and embedding AI governance into national health strategies with sustainable financing. Ultimately, sustainable and equitable medical AI requires embedding human values—transparency, privacy, and equity—into every phase of design and deployment. Meaningful innovation in global health depends on amplifying human judgment with compassionate, trustworthy, and context-aware AI systems, rather than replacing it. The review confirms that properly implemented AI systems can strengthen health service delivery, enhance diagnostic accuracy, and reduce inequalities in access to care.","author":[{"family":"Al-Ganad","given":"Abdulelah"},{"family":"Al-Shahethi","given":"Ahmed"},{"family":"Al-Dhaifi","given":"Othman"},{"family":"Hajeb","given":"Essam"},{"family":"Hajeb","given":"Huwaida"},{"family":"Almotarreb","given":"Ahmed"}],"issued":{"date-parts":[[2025]]},"DOI":"10.21203/rs.3.rs-8051581/v1","URL":"https://doi.org/10.21203/rs.3.rs-8051581/v1","source":"preprints"},{"id":"oa:W4411329945","type":"article-journal","title":"Artificial Intelligence Outperforms Physicians in General Medical Knowledge, Except in the Paediatrics Domain: A Cross-Sectional Study","abstract":"Generative artificial intelligence (genAI) shows promising results in clinical practice. This study compared a GPT-4-turbo virtual assistant with physicians from Italy, France, Spain, and Portugal on medical knowledge derived from national exams while analysing knowledge retention over time and domain-specific performance. Via a digital platform, 17,144 physicians provided 221,574 answers to 600 exam questions between December 2022 and February 2024. Physicians were stratified by years since graduation and specialty, and the assistant answered the same questions in each native language. Differences in proportions of correct answers were tested with binomial logistic regression (odds ratios, 95% CI) or Fisher’s exact test (α = 0.05). The assistant outperformed physicians in all countries (72–96% vs. 46–62%; logistic regression, p < 0.001). Physicians also trailed the assistant across most knowledge domains (p < 0.001), except paediatrics (45% vs. 52%; Fisher, p = 0.60). Accuracy declined with seniority, falling 4–10% between the youngest and oldest cohorts (logistic regression, p < 0.001). Overall, genAI exceeds practising doctors on broad medical knowledge and may help counter knowledge attrition, though paediatrics remains a domain requiring targeted refinement.","author":[{"family":"Miranda","given":"JMG"},{"family":"Pereira-Silva","given":"Raquel"},{"family":"Guichard","given":"João"},{"family":"Meneses","given":"Jorge"},{"family":"Carreira","given":"Andreia"},{"family":"Seixas","given":"Daniela"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bioengineering12060653","URL":"https://doi.org/10.3390/bioengineering12060653","source":"openalex"},{"id":"oa:W4409293417","type":"article-journal","title":"Digitalisation and artificial intelligence development. A cross-country analysis","abstract":"Purpose This study aims to examine how cultural dimensions affect digitalisation, particularly artificial intelligence (AI) development and its implementation strategies in 38 countries. Design/methodology/approach We use a mixed methodology to obtain a deeper insight into a complex and, at the same time, understudied phenomenon, e.g. the global impact of cultural values on digitalisation and AI development. Combining quantitative and qualitative analyses, we shed new light on the empirical manifestations of cultural dimensions and expert perceptions. Quantitative analysis uses regression models based on data sourced from the GLOBE and Digital Competitiveness Ranking studies. Qualitative insights are drawn from epistemic interviews with experts representing four countries, namely China, Italy, Russia and the USA, which belong to different cultural clusters. The choice to have countries representing different clusters follows the logic of the GLOBE Project and aims at gaining insight into different cultural groups and scaling up the results to other countries in the same cluster. Findings Key findings highlight risk aversion as a pivotal factor, encompassing uncertainty avoidance, future orientation, institutional collectivism, power distance and performance orientation practices. Expert interviews elucidate how these cultural values influence national strategies, regulatory frameworks, stakeholder engagement, competitive dynamics and future skill requirements in AI implementation and development. Research limitations/implications Further research is recommended to delve deeper into the intricate relationship between country culture, technological adoption and economic progress, enriching global understanding of digitalisation dynamics and AI implementation and development. The research data can be further developed and expanded to include other GLOBE clusters not covered in this study, allowing for a deeper analysis. Inviting more experts from each region would enhance the breadth of perspectives. Moreover, similar studies could be conducted within the context of a single country to provide a more detailed examination. Practical implications Implications for policymakers underscore the need to integrate cultural considerations into digitalisation strategies, ensuring alignment with societal values and optimising economic outcomes. Businesses are encouraged to adopt culturally sensitive approaches to AI implementation to build trust and foster engagement within diverse cultural contexts. Originality/value The originality of this study lies in its exploration of the intersection between digitalisation, AI implementation and cultural dimensions – an area that has so far received limited attention in the available literature. While comparative studies often focus on the technical or policy aspects of digital advancements, this research highlights how cultural factors can facilitate or hinder these developments. Moreover, the study employs a mixed-methods approach, combining quantitative analysis with qualitative insights. This approach allows deeper understanding and cross-validation of findings, enhancing the study’s credibility and offering a more nuanced perspective on the role of culture in shaping digital transformation.","author":[{"family":"Cannavale","given":"Chiara"},{"family":"Claudio","given":"Lorenza"},{"family":"Королева","given":"Диана"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1108/ejim-07-2024-0828","URL":"https://doi.org/10.1108/ejim-07-2024-0828","source":"openalex"},{"id":"oa:W4409331637","type":"article-journal","title":"The Role of Artificial Intelligence in the Diagnosis and Management of Rheumatoid Arthritis","abstract":"Background and Objectives: Artificial intelligence has emerged as a transformative tool in healthcare, offering capabilities such as early diagnosis, personalised treatment, and real-time patient monitoring. In the context of rheumatoid arthritis, a chronic autoimmune disease that demands timely intervention, artificial intelligence shows promise in overcoming diagnostic delays and optimising disease management. This study examines the role of artificial intelligence in the diagnosis and management of rheumatoid arthritis, focusing on perceived benefits, challenges, and acceptance levels among healthcare professionals and patients. Materials and Methods: A cross-sectional study was conducted using a detailed questionnaire distributed to 205 participants, including rheumatologists, general practitioners, and rheumatoid arthritis patients from Romania. The study used descriptive statistics, chi-square tests, and logistic regression to analyse AI acceptance in rheumatology. Data visualisation and multiple imputations addressed missing values, ensuring accuracy. Statistical significance was set at p < 0.05 for hypothesis testing. Results: Respondents with prior experience in artificial intelligence perceived it as more useful for early diagnosis and personalised management of RA (p < 0.001). Familiarity with artificial intelligence concepts positively correlated with acceptance in routine rheumatology practice (ρ = 1.066, p < 0.001). The main barriers identified were high costs (36%), lack of medical staff training (37%), and concerns regarding diagnostic accuracy (21%). Although less frequently mentioned, data privacy concerns remained relevant for a subset of respondents. The study revealed that artificial intelligence could improve diagnostic accuracy and rheumatoid arthritis monitoring, being perceived as a valuable tool by professionals familiar with digital technologies. However, 42% of participants cited the lack of data standardisation across medical systems as a major barrier, underscoring the need for effective interoperability solutions. Conclusions: Artificial intelligence has the potential to revolutionise rheumatoid arthritis management through faster and more accurate diagnoses, personalised treatments, and optimised monitoring. Nevertheless, challenges such as costs, staff training, and data privacy need to be addressed to ensure efficient integration into clinical practice. Educational programmes and interdisciplinary collaboration are essential to increase artificial intelligence adoption in rheumatology.","author":[{"family":"Vlad","given":"Amalia"},{"family":"Popazu","given":"Corina"},{"family":"Lescai","given":"Alina"},{"family":"Voinescu","given":"Doina"},{"family":"Baltă","given":"Alexia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/medicina61040689","URL":"https://doi.org/10.3390/medicina61040689","source":"openalex"},{"id":"oa:W4410831215","type":"article-journal","title":"Artificial intelligence in bronchoscopy: a systematic review","abstract":"BACKGROUND: Artificial intelligence (AI) systems have been implemented to improve the diagnostic yield and operators' skills within endoscopy. Similar AI systems are now emerging in bronchoscopy. Our objective was to identify and describe AI systems in bronchoscopy. METHODS: A systematic review was performed using MEDLINE, Embase and Scopus databases, focusing on two terms: bronchoscopy and AI. All studies had to evaluate their AI against human ratings. The methodological quality of each study was assessed using the Medical Education Research Study Quality Instrument (MERSQI). RESULTS: 1196 studies were identified, with 20 passing the eligibility criteria. The studies could be divided into three categories: nine studies in airway anatomy and navigation, seven studies in computer-aided detection and classification of nodules in endobronchial ultrasound, and four studies in rapid on-site evaluation. 16 were assessment studies, with 12 showing equal performance and four showing superior performance of AI compared with human ratings. Four studies within airway anatomy implemented their AI, all favouring AI guidance to no AI guidance. The methodological quality of the studies was moderate (mean MERSQI 12.9 points, out of a maximum 18 points). INTERPRETATION: 20 studies developed AI systems, with only four examining the implementation of their AI. The four studies were all within airway navigation and favoured AI to no AI in a simulated setting. Future implementation studies are warranted to test for the clinical effect of AI systems within bronchoscopy.","author":[{"family":"Cold","given":"Kristoffer"},{"family":"Vamadevan","given":"Anishan"},{"family":"Laursen","given":"Christian"},{"family":"Bjerrum","given":"Flemming"},{"family":"Singh","given":"Suveer"},{"family":"Konge","given":"Lars"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1183/16000617.0274-2024","URL":"https://doi.org/10.1183/16000617.0274-2024","source":"openalex"},{"id":"oa:W4414041124","type":"article-journal","title":"Generative artificial intelligence for automated data extraction from unstructured medical text","abstract":"Objectives: Unstructured data, such as procedure notes, contain valuable medical information that is frequently underutilized due to the labor-intensive nature of data extraction. This study aims to develop a generative artificial intelligence (GenAI) pipeline using an open-source Large Language Model (LLM) with built-in guardrails and a retry mechanism to extract data from unstructured right heart catheterization (RHC) notes while minimizing errors, including hallucinations. Materials and Methods: A total of 220 RHC notes were randomly selected for pipeline development and 200 for validation from the Pulmonary Vascular Disease Registry. The pipeline comprised three main components: the Engineered Preload Framework (EPF), which integrated schemas and instructions; the LLM module, enhanced by reasoning capabilities; and the validation and retry mechanism, which ensured data accuracy through iterative self-correction. A clinical expert manually extracted data from the validation cohort to establish the ground truth. Pipeline performance was evaluated using precision, recall, and F1 score. Additionally, the dataset was stratified into quartiles to assess the pipeline's ability to handle varying levels of data availability. Results: The pipeline achieved 99.0% precision, 85.0% recall, and a 91.5% F1 score, with an overall accuracy of 90% when evaluated at the note level. The most common error was missed values (5.2%), while hallucinations were the least frequent (<0.01%). Discussion and Conclusion: This study demonstrates the feasibility of a robust GenAI pipeline for automating structured data extraction from unstructured RHC procedure notes. The approach highlights the potential of LLMs in medical data mining, improving research efficiency and clinical applications.","author":[{"family":"Dao","given":"Phuong"},{"family":"Quesada","given":"Luisa"},{"family":"Hassan","given":"Syed"},{"family":"Campo","given":"MI"},{"family":"Johnson","given":"Shelsey"},{"family":"Ghose","given":"S"},{"family":"Estépar","given":"Raúl"},{"family":"Waxman","given":"Aaron"},{"family":"Washko","given":"George"},{"family":"Rahaghi","given":"Farbod"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/jamiaopen/ooaf097","URL":"https://doi.org/10.1093/jamiaopen/ooaf097","source":"openalex"},{"id":"oa:W4414529263","type":"article-journal","title":"Artificial Intelligence-Driven Multi-Omics Approaches in Glioblastoma","abstract":"Glioblastoma (GBM) is the most common and aggressive primary brain tumor in adults. It is characterized by a high degree of heterogeneity, meaning that although these tumors may appear morphologically similar, they often exhibit distinct clinical outcomes. By associating specific molecular fingerprints with different clinical behaviors, high-throughput omics technologies (e.g., genomics, transcriptomics, and epigenomics) have significantly advanced our understanding of GBM, particularly of its extensive heterogeneity, by proposing a molecular classification for the implementation of precision medicine. However, due to the vast volume and complexity of data, the integrative analysis of omics data demands substantial computational power for processing, analyzing and interpreting GBM-related data. Artificial intelligence (AI), which mainly includes machine learning (ML) and deep learning (DL) computational approaches, now presents a unique opportunity to infer valuable biological insights from omics data and enhance the clinical management of GBM. In this review, we explored the potential of integrating multi-omics, imaging radiomics and clinical data with AI to uncover different aspects of GBM (molecular profiling, prognosis, and treatment) and improve its clinical management.","author":[{"family":"Morello","given":"Giovanna"},{"family":"Cognata","given":"Valentina"},{"family":"Guarnaccia","given":"Maria"},{"family":"Gentile","given":"Giulia"},{"family":"Cavallaro","given":"Sebastiano"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ijms26199362","URL":"https://doi.org/10.3390/ijms26199362","source":"openalex"},{"id":"oa:W7119491114","type":"article-journal","title":"Artificial intelligence, extended reality, and emerging AI–XR integrations in medical education","abstract":"Introduction: Artificial intelligence (AI) and extended reality (XR)-including virtual, augmented, and mixed reality-are increasingly adopted in health-professions education. However, the educational impact of AI, XR, and especially their combined use within integrated AI-XR ecosystems remains incompletely characterized. Objective: To synthesize empirical evidence on educational outcomes and implementation considerations for AI-, XR-, and combined AI-XR-based interventions in medical and health-professions education. Methods: Following PRISMA and PICO guidance, we searched three databases (Scopus, PubMed, IEEE Xplore) and screened records using predefined eligibility criteria targeting empirical evaluations in health-professions education. After deduplication (336 records removed) and two-stage screening, 13 studies published between 2019 and 2024 were included. Data were extracted on learner population, clinical domain, AI/XR modality, comparators, outcomes, and implementation factors, and narratively synthesized due to heterogeneity in designs and measures. Results: The 13 included studies involved undergraduate and postgraduate learners in areas such as procedural training, clinical decision-making, and communication skills. Only a minority explicitly integrated AI with XR within the same intervention; most evaluated AI-based or XR-based approaches in isolation. Across this mixed body of work, studies more often than not reported gains in at least one outcome-knowledge or skills performance, task accuracy, procedural time, or learner engagement-relative to conventional instruction, alongside generally high acceptability. Recurrent constraints included costs, technical reliability, usability, faculty readiness, digital literacy, and data privacy and ethics concerns. Conclusions: Current evidence on AI, XR, and emerging AI-XR integrations suggests promising but preliminary benefits for learning and performance. The small number of fully integrated AI-XR interventions and the methodological limitations of many primary studies substantially limit the certainty and generalizability of these findings. Future research should use more rigorous and standardized designs, explicitly compare AI-only, XR-only, and AI-XR hybrid approaches, and be coupled with faculty development, robust technical support, and alignment with competency-based assessment.","author":[{"family":"Tene","given":"Talía"},{"family":"Lopez","given":"Diego"},{"family":"Veloz","given":"Marlene"},{"family":"Oviedo","given":"Byron"},{"family":"Tene-Fernandez","given":"Richard"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/fdgth.2025.1740557","URL":"https://doi.org/10.3389/fdgth.2025.1740557","source":"openalex"},{"id":"oa:W4411326274","type":"article-journal","title":"Enhancing nucleic acid delivery by the integration of artificial intelligence into lipid nanoparticle formulation","abstract":"The advent of messenger RNA (mRNA) therapeutics has revolutionized medicine, with its potential underscored by rapid advancements during the COVID-19 pandemic. Despite its promise, nucleic acid delivery remains a formidable challenge due to enzymatic degradation, cellular uptake barriers, and endosomal trapping. Therapeutic lipid nanoparticles (LNPs), pioneered in the 1970s, have emerged as the gold standard for delivering mRNA and other nucleic acids, offering unparalleled advantages in stability, biocompatibility, and cellular targeting. This review explores the evolution and design of LNPs, focusing on their role in hematologic therapies and platelet transfection, where unique challenges arise due to platelets' anucleate nature. The paper systematically evaluates the composition of LNPs, highlighting the role of ionizable, cationic, and neutral lipids in optimizing delivery efficiency, stability, and immune response modulation. Strategies to overcome platelet transfection barriers, including tailored lipid compositions and particle engineering, are discussed alongside advances in artificial intelligence (AI) for predictive nanoparticle design. Furthermore, it examines various nucleic acid cargoes, including mRNA, siRNA, and miRNA, and their therapeutic potential in addressing platelet-related disorders and advancing personalized medicine. Finally, the review delves into emerging technologies and the integration of AI to overcome existing barriers in nucleic acid delivery. By fostering interdisciplinary collaboration, this work aims to catalyze discoveries in LNP-based therapeutics and transformative advancements in hematologic treatments.","author":[{"family":"Amoako","given":"Kagya"},{"family":"Mokhammad","given":"Amir"},{"family":"Malik","given":"Afrida"},{"family":"Yesudasan","given":"Sumith"},{"family":"Wheba","given":"Anas"},{"family":"Olagunju","given":"Oluwanifemi"},{"family":"Gu","given":"Sean"},{"family":"Yarovinsky","given":"Timur"},{"family":"Faustino","given":"EVS"},{"family":"Nguyen","given":"Juliane"},{"family":"Hwa","given":"John"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fmedt.2025.1591119","URL":"https://doi.org/10.3389/fmedt.2025.1591119","source":"openalex"},{"id":"oa:W4415695229","type":"article-journal","title":"Toward governance of artificial intelligence in pediatric healthcare","abstract":"AI is transforming healthcare, yet pediatric adoption remains limited and governance is underdeveloped. We review existing frameworks and identify pediatric-specific gaps: insufficient stakeholder engagement, developmentally appropriate consent/assent, limited bias mitigation, and unclear accountability. An analysis of FDA-cleared pediatric SaMDs shows radiology dominance while other specialties lag. We call for a pediatric-centric governance approach emphasizing transparency, inclusive participation, equitable data practices, and rigorous post-deployment monitoring to ensure safe, responsible integration.","author":[{"family":"Richter","given":"Felix"},{"family":"Holmes","given":"Emma"},{"family":"Richter","given":"Florian"},{"family":"Guttmann","given":"Katherine"},{"family":"Duong","given":"Son"},{"family":"Gangadharan","given":"Sandeep"},{"family":"Schadt","given":"Eric"},{"family":"Salmasian","given":"Hojjat"},{"family":"Gelb","given":"Bruce"},{"family":"Glicksberg","given":"Benjamin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-02000-7","URL":"https://doi.org/10.1038/s41746-025-02000-7","source":"openalex"},{"id":"oa:W4410004931","type":"article-journal","title":"Artificial Intelligence in Maritime Cybersecurity: A Systematic Review of AI-Driven Threat Detection and Risk Mitigation Strategies","abstract":"The maritime industry is undergoing a digital transformation, integrating automation, artificial intelligence (AI), and the Internet of Things (IoT) to enhance operational efficiency and safety. However, this technological evolution has also increased cybersecurity vulnerabilities, exposing vessels, ports, and maritime communication networks to sophisticated cyber threats. This systematic review, conducted following the PRISMA guidelines, examines the current landscape of AI-driven cybersecurity solutions in maritime environments. By analyzing peer-reviewed studies and industry reports, this review identifies key AI methodologies, including machine-learning-based intrusion detection systems, anomaly detection mechanisms, predictive threat modeling, and AI-enhanced zero-trust architectures. This study assesses the effectiveness of these techniques in mitigating cyber risks, explores their implementation challenges, and highlights existing research gaps. The findings indicate that AI-powered solutions significantly enhance real-time threat detection and response capabilities in maritime networks, yet issues such as data scarcity, regulatory constraints, and adversarial attacks on AI models remain unresolved. Future research directions should focus on integrating AI with blockchain, federated learning, and quantum cryptographic techniques to strengthen maritime cybersecurity frameworks.","author":[{"family":"Miller","given":"Tymoteusz"},{"family":"Durlik","given":"Irmina"},{"family":"Kostecka","given":"Ewelina"},{"family":"Sokołowska","given":"S"},{"family":"Kozlovska","given":"Polina"},{"family":"Zwolak","given":"Rafał"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/electronics14091844","URL":"https://doi.org/10.3390/electronics14091844","source":"openalex"},{"id":"oa:W4409524471","type":"article-journal","title":"Prospects for the Use of Artificial Intelligence in Personalized Medicine, Pharmaceutical Design and Education","abstract":"The article reviews modern approaches to the use of artificial intelligence in personalized medicine, pharmaceutical design and education, in particular in the pharmaceutical industry, in the processes of search, development, new drugs, personalization of pharmacotherapy and in the education of specialists in these processes. An analysis of the main areas of application of artificial intelligence, its advantages and challenges, as well as the impact on pharmaceutical design in the creation of new drugs is carried out. Special attention is paid to the role of artificial intelligence in personalized medicine, prediction of clinical and pharmacological properties, optimization of clinical trials. Ethical and regulatory aspects of integrating artificial intelligence into medical and pharmaceutical education are considered. Prospects for further development and improvement of the implementation of artificial intelligence in medicine, pharmaceutical design and education are identified. The potential for artificial intelligence to accelerate drug discovery, reduce costs, and enhance treatment precision through real-time analysis of vast datasets is particularly highlighted. Additionally, educational curricula incorporating artificial intelligence-based simulations and virtual reality tools for training future pharmaceutical professionals are explored.","author":[{"family":"Dovzhuk","given":"Viktoria"},{"family":"Konovalova","given":"L"},{"family":"Dovzhuk","given":"Natela"},{"family":"Konovalov","given":"Serhii"},{"family":"Konoshevych","given":"Liudmyla"},{"family":"Motorna","given":"Natalia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.53933/sspmpm.v5i2.182","URL":"https://doi.org/10.53933/sspmpm.v5i2.182","source":"openalex"},{"id":"oa:W4412823383","type":"article-journal","title":"Harnessing artificial intelligence of things for cardiac sensing: current advances and network-based perspectives","abstract":"Background: With the rapid advancements in science and technology, artificial intelligence (AI) has become increasingly integral to various medical applications, including medical devices and assistive healthcare tools. Extensive research highlights the significant potential of AI in the development of Internet of Things (IoT)-enabled medical devices, particularly in the field of cardiac sensing. Methods: This study explores and synthesizes current advancements and future directions of AI-driven IoT applications in cardiac sensing, highlighting their significance. Utilizing a bibliometric approach, we visualize key focus areas, emerging trends, and the evolutionary trajectory of this interdisciplinary field. Results: As of December 2024, relevant literature at the intersection of IoT, cardiac sensors, and AI was systematically retrieved from the SCIE and ESCI indices. Using CiteSpace, we conducted a comprehensive visualization analysis of countries/regions, academic publications, organizations, authors, citations, and key terminologies. A total of 2,128 papers were included in the analysis. Conclusion: From our perspective, current advancements in AI-powered IoT cardiac sensors primarily focus on optimizing AI algorithms, such as deep learning techniques, and enhancing the functionality of smart wearable devices for precision medicine. Looking ahead, we anticipate that this field will increasingly prioritize data privacy protection, particularly in the era of large language models, to address emerging challenges and ensure sustainable growth. In summary, we need to continue harnessing the power of AI-powered IoT for cardiac sensing as part of public health strategies to enable early detection of heart diseases.","author":[{"family":"Ren","given":"Hao"},{"family":"Jing","given":"Fengshi"},{"family":"Ma","given":"Yuan"},{"family":"Wang","given":"Ruining"},{"family":"He","given":"Chaocheng"},{"family":"Wang","given":"Yufan"},{"family":"Zhou","given":"Jian"},{"family":"Sun","given":"Yu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fpubh.2025.1569887","URL":"https://doi.org/10.3389/fpubh.2025.1569887","source":"openalex"},{"id":"oa:W4411244936","type":"article-journal","title":"Artificial Intelligence in Chronic Disease Management for Aging Populations: A Systematic Review of Machine Learning and NLP Applications","abstract":"As China's elderly population grows rapidly and the aging society arrives, the number of elderly patients with chronic diseases (mainly including chronic cardiovascular and cerebrovascular diseases, respiratory diseases, etc) continues to increase, significantly impacting individuals, families, and society. Geriatric Chronic Disease Management in China faces multiple challenges, including unequal distribution of medical resources, lack of professional management teams, insufficient health education, improper medication management, inadequate psychological support, insufficient medical insurance coverage, and insufficient family support. The rise of artificial intelligence (AI) technology (eg, machine learning, deep learning, NLP, computer vision) offers possibilities for improving Geriatric Chronic Disease Management, including optimizing the distribution of medical resources, supplementing professional management teams, popularizing health education, optimizing medication management, enhancing psychological support, improving medical insurance efficiency and accuracy, and strengthening family support. However, the application of AI in Geriatric Chronic Disease Management still faces challenges such as the data scarcity, model generalization, clinician adoption, alignment of AI decision-making with clinical guidelines, Integration with existing healthcare systems, privacy and security, user acceptance, ethics and law. To overcome these challenges, interdisciplinary collaboration is needed to promote the rational and effective application of AI technology, aiming to achieve healthy aging. This paper systematically reviews the current status, challenges, and future directions of AI application in Geriatric Chronic Disease Management.","author":[{"family":"Feng","given":"Gang"},{"family":"Weng","given":"Falin"},{"family":"Lu","given":"Wei"},{"family":"Xu","given":"Libin"},{"family":"Wenxiang","given":"Zhu"},{"family":"Tan","given":"Man"},{"family":"Weng","given":"Pengjuan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2147/ijgm.s516247","URL":"https://doi.org/10.2147/ijgm.s516247","source":"openalex"},{"id":"oa:W4410951986","type":"article-journal","title":"Artificial Intelligence and Public Sector Auditing: Challenges and Opportunities for Supreme Audit Institutions","abstract":"The application of artificial intelligence (AI) is growing exponentially in public entities, contributing to the improvement of the design and provision of services, as well as to the internal management and efficiency of public institutions. However, the potential of artificial intelligence systems for the public sector also entails a set of risks related, among other areas, to privacy, confidentiality, security, transparency or bias and discrimination. The Supreme Audit Institutions (SAIs), when auditing public services and policies, must adapt their human and technological resources to this new scenario. This paper analyses the implications of AI penetration in the public sector, as well as the challenges that these technological developments pose to SAIs to improve effectiveness and efficiency in their auditing tasks. This paper presents a conceptual and exploratory analysis, informed by documentary evidence and case illustrations. Given the dynamic evolution of AI research, the findings should be interpreted as a contribution to ongoing debates, rather than definitive conclusions. It also reviews the status of the audits of systems based on algorithms carried out by some SAIs.","author":[{"family":"Genaro-Moya","given":"Dolores"},{"family":"Hernández","given":"Antonio"},{"family":"Godz","given":"Mariia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/world6020078","URL":"https://doi.org/10.3390/world6020078","source":"openalex"},{"id":"oa:W4410969909","type":"article-journal","title":"Improving Explainability and Integrability of Medical AI to Promote Health Care Professional Acceptance and Use: Mixed Systematic Review","abstract":"BACKGROUND: The integration of artificial intelligence (AI) in health care has significant potential, yet its acceptance by health care professionals (HCPs) is essential for successful implementation. Understanding HCPs' perspectives on the explainability and integrability of medical AI is crucial, as these factors influence their willingness to adopt and effectively use such technologies. OBJECTIVE: This study aims to improve the acceptance and use of medical AI. From a user perspective, it explores HCPs' understanding of the explainability and integrability of medical AI. METHODS: We performed a mixed systematic review by conducting a comprehensive search in the PubMed, Web of Science, Scopus, IEEE Xplore, and ACM Digital Library and arXiv databases for studies published between 2014 and 2024. Studies concerning an explanation or the integrability of medical AI were included. Study quality was assessed using the Joanna Briggs Institute critical appraisal checklist and Mixed Methods Appraisal Tool, with only medium- or high-quality studies included. Qualitative data were analyzed via thematic analysis, while quantitative findings were synthesized narratively. RESULTS: Out of 11,888 records initially retrieved, 22 (0.19%) studies met the inclusion criteria. All selected studies were published from 2020 onward, reflecting the recency and relevance of the topic. The majority (18/22, 82%) originated from high-income countries, and most (17/22, 77%) adopted qualitative methodologies, with the remainder (5/22, 23%) using quantitative or mixed method approaches. From the included studies, a conceptual framework was developed that delineates HCPs' perceptions of explainability and integrability. Regarding explainability, HCPs predominantly emphasized postprocessing explanations, particularly aspects of local explainability such as feature relevance and case-specific outputs. Visual tools that enhance the explainability of AI decisions (eg, heat maps and feature attribution) were frequently mentioned as important enablers of trust and acceptance. For integrability, key concerns included workflow adaptation, system compatibility with electronic health records, and overall ease of use. These aspects were consistently identified as primary conditions for real-world adoption. CONCLUSIONS: To foster wider adoption of AI in clinical settings, future system designs must center on the needs of HCPs. Enhancing post hoc explainability and ensuring seamless integration into existing workflows are critical to building trust and promoting sustained use. The proposed conceptual framework can serve as a practical guide for developers, researchers, and policy makers in aligning AI solutions with frontline user expectations. TRIAL REGISTRATION: PROSPERO CRD420250652253; https://www.crd.york.ac.uk/PROSPERO/view/CRD420250652253.","author":[{"family":"Liu","given":"Yushu"},{"family":"Liu","given":"Chenxi"},{"family":"Zheng","given":"Jianing"},{"family":"Xu","given":"Chang"},{"family":"Wang","given":"Dan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/73374","URL":"https://doi.org/10.2196/73374","source":"openalex"},{"id":"oa:W7141119996","type":"article-journal","title":"Human-in-the-Loop Artificial Intelligence: A Systematic Review of Concepts, Methods, and Applications","abstract":"The integration of human judgment into artificial intelligence (AI) systems has emerged as a key research direction, particularly for high-stakes applications where full automation remains insufficient. Human-in-the-Loop (HITL) AI represents a field that combines machine learning capabilities with human oversight, feedback, and decision-making at various stages of the AI pipeline. This survey provides a systematic review of HITL approaches, covering theoretical foundations, technical methods, ethical considerations, and domain-specific applications. We propose a unified taxonomy that categorizes HITL systems based on loop placement, interaction granularity, and temporal characteristics. This review synthesizes findings from healthcare, autonomous systems, cybersecurity, and other high-risk domains where human oversight is essential. We also examine the challenges of scalability, cognitive load, and trust calibration that affect the practical deployment of HITL systems. The final section outlines open research directions and introduces a framework for designing effective human-AI collaborative systems.","author":[{"family":"Lazaros","given":"Konstantinos"},{"family":"Vrahatis","given":"Aristidis"},{"family":"Kotsiantis","given":"Sotiris"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/e28040377","URL":"https://doi.org/10.3390/e28040377","source":"openalex"},{"id":"oa:W4410958358","type":"article-journal","title":"Exploring the determinants and effects of artificial intelligence (AI) hallucination exposure on generative AI adoption in healthcare","abstract":"Artificial intelligence (AI) hallucinations—erroneous outputs that generate misleading or nonsensical content—pose significant risks in contexts where consumers seek health information, as inaccuracies in this domain could lead to harmful outcomes. This study aims to explore determinants of AI hallucination exposure (HEX), examine the potential HEX's direct and mediating effects on adoption intentions, and integrate HEX into the Theory of Planned Behavior (TPB) to advance generative AI adoption models. An eight-factor measurement model, grounded in the TPB and incorporating constructs such as perceived usefulness, attitude toward AI, perceived risk, subjective norms, perceived behavioral control, user trust, AI hallucination exposure, and behavioral intention, was developed and tested using structural equation modeling (SEM). The study concludes that perceived behavioral control is a significant determinant of AI hallucination exposure (HEX), while subjective norms exert a direct influence on behavioral intention (BI) to adopt generative AI chatbots. AI hallucination exposure does not statistically significantly mediate the relationship between key antecedents and the use of generative AI chatbots. These findings advance the Theory of Planned Behavior (TPB) by integrating AI-specific risks like HEX while underscoring the need to refine theoretical models to account for contexts where technological reliability—rather than user perceptions alone—drives adoption decisions.","author":[{"family":"Jin","given":"Longyun"},{"family":"Shen","given":"Zijun"},{"family":"Alhur","given":"Anas"},{"family":"Naeem","given":"Salman"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1177/02666669251340954","URL":"https://doi.org/10.1177/02666669251340954","source":"openalex"},{"id":"oa:W4409275342","type":"article-journal","title":"Geopolymer foam concrete: a review of pore characteristics, compressive strength and artificial intelligence in GFC strength simulations","abstract":"Geopolymer foam concrete (GFC) has emerged as a sustainable alternative in construction, utilizing waste-derived binders to enhance material properties while addressing environmental concerns. This review examines the impact of precursor materials, foam agents, nanomaterials, fibers, and curing temperatures on GFC’s pore structure and compressive strength. GFC typically achieves compressive strengths of 1–10 MPa, with porosity significantly influencing performance. For instance, GFC with densities of 280–865 kg/m 3 exhibits compressive strengths of 1.10–8.13 MPa and thermal conductivities of 0.08–0.20 W/(m K), making it suitable for insulation applications. The integration of alternative materials, such as industrial by-products, enhances sustainability without compromising mechanical properties. Precise control over curing parameters, particularly heating rates, is critical to optimizing porosity and strength. The addition of nanomaterials improves mechanical performance and introduces self-sensing capabilities, expanding GFC’s potential for advanced industrial applications. Fibers further enhance toughness and crack resistance, broadening its usability. Machine learning algorithms offer promising tools for optimizing GFC formulations and structural designs, enabling the development of high-performance, sustainable materials tailored to specific engineering needs. By systematically evaluating the effects of alternative materials, nanotechnology, fibers, and curing conditions, this review highlights the potential of GFC to revolutionize sustainable construction. Finally, GFC represents a transformative advancement in eco-friendly construction materials. Its integration of waste-derived binders, nanomaterials, and advanced optimization techniques positions it as a key solution for sustainable infrastructure, balancing performance, durability, and environmental responsibility.","author":[{"family":"Abdellatief","given":"Mohamed"},{"family":"Hassanien","given":"Alaa"},{"family":"Mortagi","given":"Mohamed"},{"family":"Hamouda","given":"Hassan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s44416-025-00003-x","URL":"https://doi.org/10.1007/s44416-025-00003-x","source":"openalex"},{"id":"oa:W4406296262","type":"article-journal","title":"Artificial Intelligence for Cervical Spine Fracture Detection: A Systematic Review of Diagnostic Performance and Clinical Potential","abstract":"Study DesignSystematic review.ObjectiveArtificial intelligence (AI) and deep learning (DL) models have recently emerged as tools to improve fracture detection, mainly through imaging modalities such as computed tomography (CT) and radiographs. This systematic review evaluates the diagnostic performance of AI and DL models in detecting cervical spine fractures and assesses their potential role in clinical practice.MethodsA systematic search of PubMed/Medline, Embase, Scopus, and Web of Science was conducted for studies published between January 2000 and July 2024. Studies that evaluated AI models for cervical spine fracture detection were included. Diagnostic performance metrics were extracted and included sensitivity, specificity, accuracy, and area under the curve. The PROBAST tool assessed bias, and PRISMA criteria were used for study selection and reporting.ResultsEleven studies published between 2021 and 2024 were included in the review. AI models demonstrated variable performance, with sensitivity ranging from 54.9% to 100% and specificity from 72% to 98.6%. Models applied to CT imaging generally outperformed those applied to radiographs, with convolutional neural networks (CNN) and advanced architectures such as MobileNetV2 and Vision Transformer (ViT) achieving the highest accuracy. However, most studies lacked external validation, raising concerns about the generalizability of their findings.ConclusionsAI and DL models show significant potential in improving fracture detection, particularly in CT imaging. While these models offer high diagnostic accuracy, further validation and refinement are necessary before they can be widely integrated into clinical practice. AI should complement, rather than replace, human expertise in diagnostic workflows.","author":[{"family":"Liawrungrueang","given":"Wongthawat"},{"family":"Cholamjiak","given":"Watcharaporn"},{"family":"Promsri","given":"Arunee"},{"family":"Jitpakdee","given":"Khanathip"},{"family":"Sunpaweravong","given":"Sompoom"},{"family":"Kotheeranurak","given":"Vit"},{"family":"Sarasombath","given":"Peem"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1177/21925682251314379","URL":"https://doi.org/10.1177/21925682251314379","source":"openalex"},{"id":"oa:W4416028051","type":"article-journal","title":"Concept-based Explainable Artificial Intelligence: A Survey","abstract":"The field of explainable artificial intelligence emerged in response to the growing need for more transparent and reliable models. However, using raw features to provide explanations has been discussed in several works lately, advocating for more user-understandable explanations. To address this issue, a wide range of papers proposing Concept-based eXplainable Artificial Intelligence (C-XAI) methods have been published in recent years. Nevertheless, a unified categorization and precise field definition are still missing. This paper fills the gap by offering a thorough review of C-XAI approaches. We identify and define different concepts and explanation types. We propose a taxonomy comprising nine categories and guidelines for selecting a suitable category based on the application context. Additionally, we discuss common evaluation strategies including metrics, human evaluations, and datasets employed, aiming to assist the development of future methods. Overall, we believe this survey will assist researchers, practitioners, and domain experts in enhancing their understanding and contributing to the progress of this innovative field.","author":[{"family":"Poeta","given":"Eleonora"},{"family":"Ciravegna","given":"Gabriele"},{"family":"Pastor","given":"Eliana"},{"family":"Cerquitelli","given":"Tania"},{"family":"Baralis","given":"Elena"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1145/3774643","URL":"https://doi.org/10.1145/3774643","source":"openalex"},{"id":"oa:W4406038522","type":"article-journal","title":"An Innovative Artificial Intelligence Based Decision Making System for Public Health Crisis Virtual Reality Rehabilitation","abstract":"The COVID-19 disease caused by the SARS-CoV-2 virus was declared by the World Health Organization (WHO) as a spreadable viral disease. During the COVID pandemic, there was difficulty in notifying the Decision-Making System (DMS) about the rapid and precise triage of patients admitted to the emergency wards. As a method to achieve the aim and develop digital healthcare revolutions in data and analytics, digital healthcare information was established. Artificial Intelligence (AI) is a robust automation tool for sustainability in the context of the COVID-19 health crisis on big datasets. Besides, the gap between AI investment and commercial real-time application, which are the initial digital technology development curves, has been identified. It was discovered that AI’s new applications are grounded in Digital Transformation Mapping (DTM) for the DMS of Health Crises. The fast inventions in AI and Machine Learning (ML) have implications for amazingly preventive and clinical healthcare, and for the association, ML was developed as a predictable attention. Billions of smartphones, massive online datasets, linked wireless wearable devices, comparatively cost-effective computing resources and improved ML and Nural Language Processing (NLP) are leveraged by these rapid responses, with the trained dataset of 65% and evaluated in the other 35%, the renowned ML models for structured data like Support Vector Machine (SVM), Multinomial Naive Bayes (MNB), Logistic Regressive Tree (LRT), Decision Tree (DT), Stochastic Gradient Booster (SGB), and Random Forest (RF) are used for simulating new unidentified data. AI-DTM challenges DMS of Health Crises (COVID-19) and the drawbacks of critically contributing risk factors to healthcare diseases. Meanwhile, a comprehensive collection of healthcare datasets over what is spreadable would be required to save human lives, train AI, and limit cost-effective health risks.","author":[{"family":"Ghanimi","given":"Hayder"},{"family":"Ayasrah","given":"Firas"},{"family":"Jadala","given":"Vijaya"},{"family":"Manjunath","given":"TC"},{"family":"Balasaranya","given":"K"},{"family":"Srinivasarao","given":"B"}],"issued":{"date-parts":[[2025]]},"DOI":"10.53759/7669/jmc202505044","URL":"https://doi.org/10.53759/7669/jmc202505044","source":"openalex"},{"id":"oa:W4407791699","type":"article-journal","title":"Artificial Intelligence and Breast Cancer Management: From Data to the Clinic","abstract":"Breast cancer (BC) remains a significant threat to women's health worldwide. The oncology field had an exponential growth in the abundance of medical images, clinical information, and genomic data. With its continuous advancement and refinement, artificial intelligence (AI) has demonstrated exceptional capabilities in processing intricate multidimensional BC-related data. AI has proven advantageous in various facets of BC management, encompassing efficient screening and diagnosis, precise prognosis assessment, and personalized treatment planning. However, the implementation of AI into precision medicine and clinical practice presents ongoing challenges that necessitate enhanced regulation, transparency, fairness, and integration of multiple clinical pathways. In this review, we provide a comprehensive overview of the current research related to AI in BC, highlighting its extensive applications throughout the whole BC cycle management and its potential for innovative impact. Furthermore, this article emphasizes the significance of constructing patient-oriented AI algorithms. Additionally, we explore the opportunities and potential research directions within this burgeoning field.","author":[{"family":"Feng","given":"Kaixiang"},{"family":"Yi","given":"Zongbi"},{"family":"Xu","given":"Binghe"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/cai2.159","URL":"https://doi.org/10.1002/cai2.159","source":"openalex"},{"id":"oa:W4412660449","type":"article-journal","title":"The Role of Artificial Intelligence in Strategic Planning and Competitive Advantage","abstract":"This study investigates how Artificial Intelligence (AI) affects strategic planning and helps industries improve their competitive positions. With digital changes becoming more overwhelming, organisations need to act fast and AI gives them the tools and guidelines they need for foresight, agility and decision intelligence. By studying more than 80 recent academic articles on the topic, the paper explores how AI merges with approaches like Resource-Based View (RBV), SWOT and Porter’s Five Forces and boosts them by using predictive analytics, natural language processing (NLP) and machine learning (ML). Looking at Amazon, Google, IBM, Tesla and SMEs reveals how AI helps improve efficiency, speed up innovation, develop well-known brands and improve interaction with the ecosystem. The findings also show that unmanaged ethical problems, governance gaps and differences in goals can hold back AI. The results point out that AI should be considered both a useful technology and a strategic partner, creating continuous value when it supports corporate aims, ethics and leadership. In addition, the paper advocates for additional research that brings together different experts to solve the challenges in AI policy, planning and its social results. Because of these insights, organisations, scholars and policymakers gain the support they need in the dynamic AI-based business environment.","author":[{"family":"Jafari","given":"Mahdi"},{"family":"Shahbazi","given":"Avesta"},{"family":"Kawsar","given":"Md"},{"family":"Davoudi","given":"Seyed"},{"family":"Janani","given":"S"}],"issued":{"date-parts":[[2025]]},"DOI":"10.59857/mvzl8684","URL":"https://doi.org/10.59857/mvzl8684","source":"openalex"},{"id":"oa:W4406001297","type":"article-journal","title":"Predicting pediatric patient rehabilitation outcomes after spinal deformity surgery with artificial intelligence","abstract":"Adolescent idiopathic scoliosis (AIS) is the most common type of scoliosis, affecting 1–4% of adolescents. The Scoliosis Research Society-22R (SRS-22R), a health-related quality-of-life instrument for AIS, has allowed orthopedists to measure subjective patient outcomes before and after corrective surgery beyond objective radiographic measurements. However, research has revealed that there is no significant correlation between the correction rate in major radiographic parameters and improvements in patient-reported outcomes (PROs), making it difficult to incorporate PROs into personalized surgical planning. The objective of this study is to develop an artificial intelligence (AI)-enabled surgical planning and counseling support system for post-operative patient rehabilitation outcomes prediction in order to facilitate personalized AIS patient care. A unique multi-site cohort of 455 pediatric patients undergoing spinal fusion surgery at two Shriners Children’s hospitals from 2010 is investigated in our analysis. In total, 171 pre-operative clinical features are used to train six machine-learning models for post-operative outcomes prediction. We further employ explainability analysis to quantify the contribution of pre-operative radiographic and questionnaire parameters in predicting patient surgical outcomes. Moreover, we enable responsible AI by calibrating model confidence for human intervention and mitigating health disparities for algorithm fairness. The best prediction model achieves an area under receiver operating curve (AUROC) performance of 0.86, 0.85, and 0.83 for individual SRS-22R question response prediction over three-time horizons from pre-operation to 6-month, 1-year, and 2-year post-operation, respectively. Additionally, we demonstrate the efficacy of our proposed prediction method to predict other patient rehabilitation outcomes based on minimal clinically important differences (MCID) and correction rates across all three-time horizons. Based on the relationship analysis, we suggest additional attention to sagittal parameters (e.g., lordosis, sagittal vertical axis) and patient self-image beyond major Cobb angles to improve surgical decision-making for AIS patients. In the age of personalized medicine, the proposed responsible AI-enabled clinical decision-support system may facilitate pre-operative counseling and shared decision-making within real-world clinical settings. The goal of this study is to develop a planning and counseling support system for predicting how well patients recover after surgeries. This should allow for more personalized care for scoliosis (spinal curvature) patients. We collected data from 455 pediatric patients who underwent spinal fusion surgery at different locations and used this data to train computer learning methods to predict outcomes after surgery. We show that our proposed computer method can predict the outcome of patient rehabilitation well for short-term (6-month and 1-year) and long-term (2-year) results. We applied additional tests to our method to calculate how well it works and measure fairness to provide a straight-forward, trustworthy, and fair method for real-world clinical use. Shi et al. develop an AI-enabled surgical planning and counseling support system for post-operative rehabilitation outcome predictions to facilitate personalized adolescent idiopathic scoliosis (AIS) patient care. They suggest additional attention to sagittal parameters and patient self-image beyond Cobb angles to improve surgical decision-making.","author":[{"family":"Shi","given":"Wenqi"},{"family":"Giuste","given":"Felipe"},{"family":"Zhu","given":"Yuanda"},{"family":"Tamo","given":"Ben"},{"family":"Nnamdi","given":"Micky"},{"family":"Hornback","given":"Andrew"},{"family":"Carpenter","given":"Ashley"},{"family":"Hilton","given":"Coleman"},{"family":"Iwinski","given":"Henry"},{"family":"Wattenbarger","given":"JM"},{"family":"Wang","given":"May"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s43856-024-00726-1","URL":"https://doi.org/10.1038/s43856-024-00726-1","source":"openalex"},{"id":"oa:W4407799493","type":"article-journal","title":"Research progress on artificial intelligence technology-assisted diagnosis of thyroid diseases","abstract":"With the rapid development of the \"Internet + Medical\" model, artificial intelligence technology has been widely used in the analysis of medical images. Among them, the technology of using deep learning algorithms to identify features of ultrasound and pathological images and realize intelligent diagnosis of diseases has entered the clinical verification stage. This study is based on the application research of artificial intelligence technology in medical diagnosis and reviews the early screening and diagnosis of thyroid diseases. The cure rate of thyroid disease is high in the early stage, but once it deteriorates into thyroid cancer, the risk of death and treatment costs of the patient increase. At present, the early diagnosis of the disease still depends on the examination equipment and the clinical experience of doctors, and there is a certain misdiagnosis rate. Based on the above background, it is particularly important to explore a technology that can achieve objective screening of thyroid lesions in the early stages. This paper provides a comprehensive review of recent research on the early diagnosis of thyroid diseases using artificial intelligence technology. It integrates the findings of multiple studies and that traditional machine learning algorithms are widely used as research objects. The convolutional neural network model has a high recognition accuracy for thyroid nodules and thyroid pathological cell lesions. U-Net network model can significantly improve the recognition accuracy of thyroid nodule ultrasound images when used as a segmentation algorithm. This article focuses on reviewing the intelligent recognition technology of thyroid ultrasound images and pathological sections, hoping to provide researchers with research ideas and help clinicians achieve intelligent early screening of thyroid cancer.","author":[{"family":"Yang","given":"Lina"},{"family":"Wang","given":"Xinyuan"},{"family":"Zhang","given":"Shixia"},{"family":"Cao","given":"Kun"},{"family":"Yang","given":"Jianjun"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fonc.2025.1536039","URL":"https://doi.org/10.3389/fonc.2025.1536039","source":"openalex"},{"id":"oa:W4410007777","type":"article-journal","title":"Medical Image Segmentation: A Comprehensive Review of Deep Learning-Based Methods","abstract":"Medical image segmentation is a critical application of computer vision in the analysis of medical images. Its primary objective is to isolate regions of interest in medical images from the background, thereby assisting clinicians in accurately identifying lesions, their sizes, locations, and their relationships with surrounding tissues. However, compared to natural images, medical images present unique challenges, such as low resolution, poor contrast, inconsistency, and scattered target regions. Furthermore, the accuracy and stability of segmentation results are subject to more stringent requirements. In recent years, with the widespread application of Convolutional Neural Networks (CNNs) in computer vision, deep learning-based methods for medical image segmentation have become a focal point of research. This paper categorizes, reviews, and summarizes the current representative methods and research status in the field of medical image segmentation. A comparative analysis of relevant experiments is presented, along with an introduction to commonly used public datasets, performance evaluation metrics, and loss functions in medical image segmentation. Finally, potential future research directions and development trends in this field are predicted and analyzed.","author":[{"family":"Gao","given":"Yuxiao"},{"family":"Jiang","given":"Yang"},{"family":"Peng","given":"Yanhong"},{"family":"Yuan","given":"Fujiang"},{"family":"Zhang","given":"Xinyue"},{"family":"Wang","given":"Jianfeng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/tomography11050052","URL":"https://doi.org/10.3390/tomography11050052","source":"openalex"},{"id":"oa:W4409308174","type":"article-journal","title":"Opportunities and challenges with artificial intelligence in allergy and immunology: a bibliometric study","abstract":"Introduction: The fields of allergy and immunology are increasingly recognizing the transformative potential of artificial intelligence (AI). Its adoption is reshaping research directions, clinical practices, and healthcare systems. However, a systematic overview identifying current statuses, emerging trends, and future research hotspots is lacking. Methods: This study applied bibliometric analysis methods to systematically evaluate the global research landscape of AI applications in allergy and immunology. Data from 3,883 articles published by 21,552 authors across 1,247 journals were collected and analyzed to identify leading contributors, prevalent research themes, and collaboration patterns. Results: Analysis revealed that the USA and China are currently leading in research output and scientific impact in this domain. AI methodologies, especially machine learning (ML) and deep learning (DL), are predominantly applied in drug discovery and development, disease classification and prediction, immune response modeling, clinical decision support, diagnostics, healthcare system digitalization, and medical education. Emerging trends indicate significant movement toward personalized medical systems integration. Discussion: The findings demonstrate the dynamic evolution of AI in allergy and immunology, highlighting the broadening scope from basic diagnostics to comprehensive personalized healthcare systems. Despite advancements, critical challenges persist, including technological limitations, ethical concerns, and regulatory frameworks that could potentially hinder further implementation and integration. Conclusion: AI holds considerable promise for advancing allergy and immunology globally by enhancing healthcare precision, efficiency, and accessibility. Addressing existing technological, ethical, and regulatory challenges will be crucial to fully realizing its potential, ultimately improving global health outcomes and patient well-being.","author":[{"family":"Xiao","given":"Ningkun"},{"family":"Huang","given":"Xin‐lin"},{"family":"Wu","given":"Yujun"},{"family":"Li","given":"Baoheng"},{"family":"Zang","given":"Wanli"},{"family":"Shinwari","given":"Khyber"},{"family":"Tuzankina","given":"Irina"},{"family":"Черешнев","given":"ВА"},{"family":"Liu","given":"Guojun"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fmed.2025.1523902","URL":"https://doi.org/10.3389/fmed.2025.1523902","source":"openalex"},{"id":"oa:W4406064793","type":"article-journal","title":"What is the influence of psychosocial factors on artificial intelligence appropriation in college students?","abstract":"BACKGROUND: In recent years, the adoption of artificial intelligence (AI) has become increasingly relevant in various sectors, including higher education. This study investigates the psychosocial factors influencing AI adoption among Peruvian university students and uses an extended UTAUT2 model to examine various constructs that may impact AI acceptance and use. METHOD: This study employed a quantitative approach with a survey-based design. A total of 482 students from public and private universities in Peru participated in the research. The study utilized partial least squares structural equation modeling (PLS-SEM) to analyze the data and test the hypothesized relationships between the constructs. RESULTS: The findings revealed that three out of the six hypothesized factors significantly influenced AI adoption among Peruvian university students. Performance expectancy (β = 0.274), social influence (β = 0.355), and AI learning self-efficacy (β = 0.431) were found to have significant positive effects on AI adoption. In contrast to expectations, ethical awareness, perceived playfulness, AI readiness and AI anxiety did not have significant impacts on AI appropriation in this context. CONCLUSION: This study highlights the importance of practical benefits, the social context, and self-confidence in the adoption of AI within Peruvian higher education. These findings contribute to the understanding of AI adoption in diverse educational settings and provide a framework for developing effective AI implementation strategies in higher education institutions. The results can guide universities and policymakers in creating targeted approaches to enhance AI adoption and integration in academic environments, focusing on demonstrating the practical value of AI, leveraging social networks, and building students' confidence in their ability to learn and use AI technologies.","author":[{"family":"Enríquez","given":"Benicio"},{"family":"Valle","given":"María"},{"family":"Ballesteros","given":"Marco"},{"family":"Castillo","given":"Julie"},{"family":"Várgas","given":"Carmen"},{"family":"Torres","given":"Isaac"},{"family":"León","given":"Pedro"},{"family":"Tirado","given":"Karina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s40359-024-02328-x","URL":"https://doi.org/10.1186/s40359-024-02328-x","source":"openalex"},{"id":"oa:W4411391856","type":"article-journal","title":"Application progress of artificial intelligence in managing thyroid disease","abstract":"Artificial intelligence (AI) has been used to study thyroid diseases since the 1990s. Previously, it mainly concentrated on the diagnosis of thyroid function and distinguishing benign from malignant thyroid nodules. With the rapid development of machine and deep learning, AI has been widely used in multiple areas of thyroid disease management, including image analysis, pathological diagnosis, personalized treatment, patient monitoring, and follow-up. This review systematically examines the evolution of AI applications in thyroid disease management since the 1990s, with a focus on diagnostic innovations, therapeutic personalization, and emerging challenges in clinical implementation. AI not only reduces the subjectivity associated with ultrasound examinations but also enhances the differentiation rate of benign and malignant thyroid nodules, thereby reducing the frequency of unnecessary fine-needle aspirations. AI synthesizes multimodal data, such as ultrasound, electronic health records, and wearable sensors, for continuous health monitoring. This integration facilitates the early detection of subclinical recurrence risk, particularly in patients who have undergone thyroidectomy. Despite the broad prospects of AI applications, challenges related to data privacy, model interpretability, and clinical applicability remain. This review critically evaluates studies across the ultrasound, CT/MRI, and histopathology domains, while addressing barriers to clinical translation, such as data heterogeneity and ethical concerns.","author":[{"family":"Lu","given":"Qing"},{"family":"Wu","given":"Yu"},{"family":"Chang","given":"Jing"},{"family":"Zhang","given":"Li"},{"family":"Lv","given":"Qing"},{"family":"Sun","given":"Hui"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fendo.2025.1578455","URL":"https://doi.org/10.3389/fendo.2025.1578455","source":"openalex"},{"id":"oa:W4408313032","type":"article-journal","title":"Artificial Intelligence in Temporomandibular Joint Disorders: An Umbrella Review","abstract":"OBJECTIVES: Given the complexity of temporomandibular joint disorders (TMDs) and their overlapping symptoms with other conditions, an accurate diagnosis necessitates a thorough examination, which can be time-consuming and resource-intensive. Consequently, innovative diagnostic tools are required to increase TMD diagnosis efficiency and precision. Therefore, the purpose of this umbrella review was to examine the existing evidence about the usefulness of artificial intelligence (AI) in TMD diagnosis. MATERIAL AND METHODS: A comprehensive search of the literature was performed from inception to November 30, 2024, in PubMed-MEDLINE, Embase, and Scopus databases. This review evaluated systematic reviews (SRs) and meta-analyses (MAs) that reported TMD patients/datasets, any AI model as intervention, no treatment, placebo as comparator and accuracy, sensitivity, specificity, or predictive value of AI models as outcome. The extracted data were complemented with narrative synthesis. RESULTS: Out of 1497 search results, this umbrella review included five studies. One of the five articles was an SR while the other four were SRMAs. Three studies focused on patients with temporomandibular joint (TMJ) problems as a group, whereas two were specific to temporomandibular joint osteoarthritis (TMJOA). The included studies reported the use of imaging datasets as samples, including cone-beam computed tomography (CBCT), magnetic resonance imaging (MRI), and panoramic radiography. The studies reported an accuracy level ranging from 0.59 to 1. Four studies reported sensitivity levels ranging from 0.76 to 0.80. Four studies reported specificity values ranging from 0.63 to 0.95 for TMJ conditions. However, only one study provided the area under the curve (AUC) in the diagnosis of TMDs. CONCLUSIONS: AI has the ability to provide faster, more accurate, sensitive, and objective diagnosis of TMJ condition. However, the performance is determined on the AI models and datasets used. Therefore, before implementing AI models in clinical practice, it is essential for researchers to extensively refine and evaluate the AI application.","author":[{"family":"Mehta","given":"Vini"},{"family":"Tripathy","given":"Snehasish"},{"family":"Noor","given":"Toufiq"},{"family":"Mathur","given":"Ankita"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/cre2.70115","URL":"https://doi.org/10.1002/cre2.70115","source":"openalex"},{"id":"oa:W4411725460","type":"article-journal","title":"Leadership in radiology in the era of technological advancements and artificial intelligence","abstract":"Radiology has evolved from the pioneering days of X-ray imaging to a field rich in advanced technologies on the cusp of a transformative future driven by artificial intelligence (AI). As imaging workloads grow in volume and complexity, and economic as well as environmental pressures intensify, visionary leadership is needed to navigate the unprecedented challenges and opportunities ahead. Leveraging its strengths in automation, accuracy and objectivity, AI will profoundly impact all aspects of radiology practice-from workflow management, to imaging, diagnostics, reporting and data-driven analytics-freeing radiologists to focus on value-driven tasks that improve patient care. However, successful AI integration requires strong leadership and robust governance structures to oversee algorithm evaluation, deployment, and ongoing maintenance, steering the transition from static to continuous learning systems. The vision of a \"diagnostic cockpit\" that integrates multidimensional data for quantitative precision diagnoses depends on visionary leadership that fosters innovation and interdisciplinary collaboration. Through administrative automation, precision medicine, and predictive analytics, AI can enhance operational efficiency, reduce administrative burden, and optimize resource allocation, leading to substantial cost reductions. Leaders need to understand not only the technical aspects but also the complex human, administrative, and organizational challenges of AI's implementation. Establishing sound governance and organizational frameworks will be essential to ensure ethical compliance and appropriate oversight of AI algorithms. As radiology advances toward this AI-driven future, leaders must cultivate an environment where technology enhances rather than replaces human skills, upholding an unwavering commitment to human-centered care. Their vision will define radiology's pioneering role in AI-enabled healthcare transformation. KEY POINTS: Question Artificial intelligence (AI) will transform radiology, improving workflow efficiency, reducing administrative burden, and optimizing resource allocation to meet imaging workloads' increasing complexity and volume. Findings Strong leadership and governance ensure ethical deployment of AI, steering the transition from static to continuous learning systems while fostering interdisciplinary innovation and collaboration. Clinical relevance Visionary leaders must harness AI to enhance, rather than replace, the role of professionals in radiology, advancing human-centered care while pioneering healthcare transformation.","author":[{"family":"Wichtmann","given":"Barbara"},{"family":"Paech","given":"Daniel"},{"family":"Pianykh","given":"Oleg"},{"family":"Huang","given":"Susie"},{"family":"Seltzer","given":"Steven"},{"family":"Brink","given":"James"},{"family":"Fennessy","given":"Fiona"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00330-025-11745-4","URL":"https://doi.org/10.1007/s00330-025-11745-4","source":"openalex"},{"id":"oa:W4406405008","type":"article-journal","title":"Artificial Intelligence‐Enhanced, Closed‐Loop Wearable Systems Toward Next‐Generation Diabetes Management","abstract":"Recent advancements in wearable healthcare have led to commercially accessible continuous glucose monitoring systems (CGMs) for diabetes management. However, CGMs only monitor glucose levels and lack therapeutic functions, prompting the development of closed‐loop systems that use monitored glucose levels to guide insulin dosing. While promising, these devices also pose risks, such as insulin overdosing, which can cause hypoglycemia. This review summarizes recent advances in integrating artificial intelligence methods with conventional CGMs. The developments in wearable CGMs and progress in insulin delivery technologies are explored, and existing algorithms for glucose prediction in closed‐loop systems are reviewed. Additionally, emerging trends in optimizing these algorithms to enhance the safety and security of closed‐loop insulin delivery systems are highlighted.","author":[{"family":"Huang","given":"Wei"},{"family":"Pang","given":"Ivo"},{"family":"Bai","given":"Jing"},{"family":"Cui","given":"Bin‐bin"},{"family":"Qi","given":"Xiaojuan"},{"family":"Zhang","given":"Shiming"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/aisy.202400822","URL":"https://doi.org/10.1002/aisy.202400822","source":"openalex"},{"id":"oa:W4411691260","type":"article-journal","title":"Challenges in the Rapid and Responsible Integration of Generative Artificial Intelligence (AI) Into a New Medical School Curriculum","abstract":"This study describes a systematic approach to integrate generative artificial intelligence (AI) into our new medical education curriculum at the Quillen College of Medicine while maintaining academic integrity. As medical schools navigate the widespread adoption of large language models, we implemented a five-step process to address the policy recommendations of the Association of American Medical Colleges (AAMC) on AI integration. First, surveys assessed AI usage patterns among students, revealing increasing adoption (from 24% to 77%) between May 2024 and February 2025. Second, clear professionalism guidelines were established, prohibiting AI use in generating learning objectives, writing Subjective, Objective, Assessment, and Plan (SOAP) notes, or completing assignments while permitting its use in research applications or otherwise in specific course settings when given permission to do so by faculty. Third, an institutional grant ensured equitable access to AI platforms for all students entering in Fall 2024. Fourth, AI was integrated into controlled educational settings, particularly within problem-based learning (PBL) for first-year students and team-based learning (TBL) for second-year students, with structured evaluation criteria. Finally, specialized training on ethical AI use was provided to students transitioning to clinical clerkships. Survey data indicated that students found AI exposure beneficial (91% agreement) and helpful for researching learning objectives (94% agreement), though confidence in AI's accuracy was lower (85% agreement). Students prioritized summarizing learning materials and testing understanding as important AI applications while valuing the ability to function as clinicians both with and without AI. Our approach demonstrates a balanced integration strategy that encourages responsible AI adoption while maintaining educational integrity in medical training.","author":[{"family":"Rueff","given":"Gabrielle"},{"family":"Monaco","given":"Paul"},{"family":"Rusinol","given":"Antonio"},{"family":"Hernandez","given":"Mark"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.86796","URL":"https://doi.org/10.7759/cureus.86796","source":"openalex"},{"id":"oa:W4406025662","type":"article-journal","title":"In the era of responsible artificial intelligence and digitalization: business group digitalization, operations and subsidiary performance","abstract":"Abstract With the rapid development of digital technologies, responsible AI has become a critical focus for ensuring ethical and socially conscious advancements in business and operations management. The integration of responsible AI practices in business groups’ digital transformations is essential to mitigate potential risks and maximize the positive impact on operational efficiency, supply chain performance, and subsidiary performance. This study aims to examine the consequences and mechanisms through which responsible group digitalization influences business group’s operation management, as manifested in subsidiary performance within the context of the digital economy. Analyzing data from 202 affiliated subsidiaries, we examine the role of HRM collaboration and technological turbulence in facilitating group digitalization. This study enriches the operations management literature and expands the application of ethical and responsible AI practices in digitalization by investigating the relationship between business group digitalization and business operations. Furthermore, this study provides practical implications pertaining to how ethical and responsible practices can guide group digital transformations, business operations and enhance the performance of subsidiaries.","author":[{"family":"Sun","given":"Wei"},{"family":"Ren","given":"Shuang"},{"family":"Tang","given":"Guiyao"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s10479-024-06453-z","URL":"https://doi.org/10.1007/s10479-024-06453-z","source":"openalex"},{"id":"oa:W4406001440","type":"article-journal","title":"A real world evaluation of an innovative artificial intelligence tool for population-level breast cancer screening","abstract":"In resource-constrained countries like India, mammography-based breast screening is challenging to implement. This state-wide study, funded by the Government of Punjab, evaluated the use of Thermalytix, a low-cost, radiation-free AI tool, for breast cancer screening. Community health workers, trained to raise awareness, mobilized women aged 30 and above for screening. Thermalytix triaged women into five risk categories based on thermal images, with high-risk women recalled for diagnostic imaging. Over 18 months, 15,069 women were screened across 183 locations in Punjab. The median age was 41 years, and 69.9% were asymptomatic. Of 460 women testing positive (recall rate 3.1%), 268 underwent follow-up imaging, and 27 were confirmed with breast cancer, yielding a detection rate of 0.18%. The positive predictive value of biopsy performed was 81.81%, and the median diagnostic interval was 21 days, with therapy initiation within 30 days. The study demonstrates the potential of Thermalytix for effective population-level breast cancer screening in low-resource settings.","author":[{"family":"Adapa","given":"Karthik"},{"family":"Gupta","given":"Ashu"},{"family":"Singh","given":"Sandeep"},{"family":"Kaur","given":"Hitinder"},{"family":"Trikha","given":"Abhinav"},{"family":"Sharma","given":"Ajoy"},{"family":"Rahul","given":"Kumar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-024-01368-2","URL":"https://doi.org/10.1038/s41746-024-01368-2","source":"openalex"},{"id":"oa:W4413611966","type":"article-journal","title":"Artificial Intelligence in Planning for Spine Surgery","abstract":"PURPOSE OF REVIEW: There has been an expanding role of artificial intelligence (AI) and machine learning (ML) in spine surgery, particularly in operative planning, intraoperative navigation, and postoperative management. With a focus on patient-specific surgical strategies, AI technologies offer new possibilities for improving surgical accuracy, reducing risks, and enhancing patient outcomes in spine care. RECENT FINDINGS: AI models have shown strong accuracy in preoperative planning, with neural networks outperforming traditional algorithms in patient selection and outcome prediction. Advances in 3D modeling, supported by machine learning, enable efficient, patient-specific anatomical reconstructions, reducing manual segmentation time from hours to seconds. In intraoperative navigation, AI-driven virtual and augmented reality systems enhance screw placement precision and reduce radiation exposure by up to 90%, improving workflow and safety. Additionally, real-time AI-based decision support has decreased operative time and postoperative risks, while postoperative AI applications now support mortality risk stratification and discharge planning, yielding significant predictive accuracy for adverse events and extended stays. AI technologies are transforming spine surgery by increasing surgical precision, optimizing clinical workflows, and personalizing patient care. While challenges remain regarding data diversity and ethical considerations, ongoing innovations indicate that AI will continue to refine spine surgery through personalized and efficient care solutions.","author":[{"family":"Ali","given":"Iyad"},{"family":"Bakaes","given":"Yianni"},{"family":"Macleod","given":"James"},{"family":"Lee","given":"T"},{"family":"Cho","given":"Sia"},{"family":"Hsu","given":"Wellington"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s12178-025-09992-5","URL":"https://doi.org/10.1007/s12178-025-09992-5","source":"openalex"},{"id":"oa:W4413125615","type":"article-journal","title":"Emerging Applications of Artificial Intelligence in Edge Computing: A Comprehensive Review","abstract":"Edge computing, coupled with Artificial Intelligence (AI), represents a paradigm shift in data processing, enabling real-time analytics and decision-making at the source. By distributing computation closer to the data origin, this integration addresses the critical challenges of latency, bandwidth, and privacy, which traditional cloud-centric systems face. AI algorithms deployed on Edge devices enable localized intelligence, facilitating transformative advancements in domains such as healthcare, smart cities, industrial automation, and autonomous systems. This review comprehensively examines the synergistic relationship between AI and Edge computing, highlighting key applications, challenges, and future research opportunities. Moreover, we emphasize the critical need for lightweight AI models, energy-efficient systems, and robust security measures to fully harness the potential of Edge AI in an increasingly connected world.","author":[{"family":"Dintakurthy","given":"Yeswanth"},{"family":"Innmuri","given":"Rama"},{"family":"Vanteru","given":"Ashutosh"},{"family":"Thotakuri","given":"Arun"}],"issued":{"date-parts":[[2025]]},"DOI":"10.71426/jmt.v1.i2.pp175-185","URL":"https://doi.org/10.71426/jmt.v1.i2.pp175-185","source":"openalex"},{"id":"oa:W4410732863","type":"article-journal","title":"Novel Blended Learning on Artificial Intelligence for Medical Students: Qualitative Interview Study","abstract":"Abstract Background Artificial intelligence (AI) systems are becoming increasingly relevant in everyday clinical practice, with Food and Drug Administration–approved AI solutions now available in many specialties. This development has far-reaching implications for doctors and the future medical profession, highlighting the need for both practicing physicians and medical students to acquire the knowledge, skills, and attitudes necessary to effectively use and evaluate these technologies. Currently, however, there is limited experience with AI-focused curricular training and continuing education. Objective This paper first introduces a novel blended learning curriculum including one module on AI for medical students in Germany. Second, this paper presents findings from a qualitative postcourse evaluation of students’ knowledge and attitudes toward AI and their overall perception of the course. Methods Clinical-year medical students can attend a 5-day elective course called “Medicine in the Digital Age,” which includes one dedicated AI module alongside 4 others on digital doctor-patient communication; digital health applications and smart devices; telemedicine; and virtual/augmented reality and robotics. After course completion, participants were interviewed in semistructured small group interviews. The interview guide was developed deductively from existing evidence and research questions compiled by our group. A subset of interview questions focused on students’ knowledge, skills, and attitudes regarding medical AI, and their overall course assessment. Responses were analyzed using Mayring’s qualitative content analysis. This paper reports on the subset of students’ statements about their perception and attitudes toward AI and the elective’s general evaluation. Results We conducted a total of 18 group interviews, in which all 35 (100%) participants (female=11, male=24) from 3 consecutive course runs participated. This produced a total of 214 statements on AI, which were assigned to the 3 main categories “Areas of Application,” “Future Work,” and “Critical Reflection.” The findings indicate that students have a nuanced and differentiated understanding of AI. Additionally, 610 statements concerned the elective’s overall assessment, demonstrating great learning benefits and high levels of acceptance of the teaching concept. All 35 students would recommend the elective to peers. Conclusions The evaluation demonstrated that the AI module effectively generates competences regarding AI technology, fosters a critical perspective, and prepares medical students to engage with the technology in a differentiated manner. The curriculum is feasible, beneficial, and highly accepted among students, suggesting it could serve as a teaching model for other medical institutions. Given the growing number and impact of medical AI applications, there is a pressing need for more AI-focused curricula and further research on their educational impact.","author":[{"family":"Oftring","given":"Zoe"},{"family":"Deutsch","given":"Kim"},{"family":"Tolks","given":"Daniel"},{"family":"Jungmann","given":"Florian"},{"family":"Kühn","given":"Sebastian"},{"family":"Oftring","given":"Zoe"},{"family":"Kuhn","given":"Sebastian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/65220","URL":"https://doi.org/10.2196/65220","source":"openalex"},{"id":"oa:W4406704087","type":"article-journal","title":"A Systematic Review: The Role of Artificial Intelligence in Lung Cancer Screening in Detecting Lung Nodules on Chest X-Rays","abstract":"Background: Lung cancer remains one of the leading causes of cancer-related deaths worldwide. Artificial intelligence (AI) holds significant potential roles in enhancing the detection of lung nodules through chest X-ray (CXR), enabling earlier diagnosis and improved outcomes. Methods: Papers were identified through a comprehensive search of the Web of Science (WOS), Scopus, and Ovid Medline databases for publications dated between 2020 and 2024. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, 34 studies that met the inclusion criteria were selected for quality assessment and data extraction. Results: AI demonstrated sensitivity rates of 56.4–95.7% and specificities of 71.9–97.5%, with the area under the receiver operating characteristic (AUROC) values between 0.89 and 0.99, compared to radiologists’ mean area under the curve (AUC) of 0.81. AI performed better with larger nodules (>2 cm) and solid nodules, showing higher AUC values for calcified (0.71) compared to non-calcified nodules (0.55). Performance was lower in hilar areas (30%) and lower lung fields (43.8%). A combined AI-radiologist approach improved overall detection rates, particularly benefiting less experienced readers; however, AI showed limitations in detecting ground-glass opacities (GGOs). Conclusions: AI shows promise as a supplementary tool for radiologists in lung nodule detection. However, the variability in AI results across studies highlights the need for standardized assessment methods and diverse datasets for model training. Future studies should focus on developing more precise and applicable algorithms while evaluating the effectiveness and cost-efficiency of AI in lung cancer screening interventions.","author":[{"family":"Ramli","given":"Puteri"},{"family":"Aizuddin","given":"Azimatun"},{"family":"Ahmad","given":"Norfazilah"},{"family":"Hamid","given":"Zuhanis"},{"family":"Ismail","given":"Khairil"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/diagnostics15030246","URL":"https://doi.org/10.3390/diagnostics15030246","source":"openalex"},{"id":"oa:W4413895322","type":"article-journal","title":"Phenotypic Selectivity of Artificial Intelligence–Enhanced Electrocardiography in Cardiovascular Diagnosis and Risk Prediction","abstract":"BACKGROUND: Artificial intelligence (AI)-enhanced ECG (AI-ECG) models are often designed to detect specific anatomical and functional cardiac abnormalities. Understanding the selectivity of their phenotypic associations is essential to inform their clinical use. Here, we sought to assess whether AI-ECG models function as condition-specific classifiers or broader cardiovascular risk markers. METHODS: We included 4 distinct study populations drawn from both electronic health records and prospective cohort studies. We deployed 6 image-based AI-ECG models: 5 validated models for the detection of left ventricular systolic dysfunction, aortic stenosis, mitral regurgitation, left ventricular hypertrophy, and a composite model for structural heart disease; and 1 negative control AI-ECG model for biological sex. Additionally, we developed 6 experimental models designed to identify noncardiovascular conditions. Diagnosis codes from electronic health records and cohorts were transformed into interpretable phenotypes using a phenome-wide association study framework. We assessed associations of AI-ECG probabilities with cross-sectional phenotypes using logistic regression and with new-onset cardiovascular diseases using Cox regression. Pearson correlation coefficients were calculated to compare phenotypic signatures. RESULTS: <10⁻⁶), whereas the sex model did not show a similar pattern. All AI-ECG models were significantly associated with their respective target phenotype but also showed similar or stronger associations with a broad range of other cardiovascular phenotypes. Phenotypic associations were similar across AI-ECG models trained for different conditions, which was not observed in models for noncardiovascular conditions. Correlation of phenotype association patterns between models was high (0.67-0.96). This pattern was consistent across all models and external data sets and in both cross-sectional and prospective analyses. CONCLUSIONS: Despite being developed to detect specific cardiovascular conditions, AI-ECG models detect the presence and predict the future development of a broad range of cardiovascular diseases with similar propensity. This challenges their role as binary diagnostic tools and instead supports their use as broader cardiovascular biomarkers.","author":[{"family":"Croon","given":"Philip"},{"family":"Dhingra","given":"Lovedeep"},{"family":"Biswas","given":"Dhruva"},{"family":"Oikonomou","given":"Evangelos"},{"family":"Khera","given":"Rohan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1161/circulationaha.125.076279","URL":"https://doi.org/10.1161/circulationaha.125.076279","source":"openalex"},{"id":"oa:W4406158982","type":"article-journal","title":"Artificial Intelligence in Metal–Organic Frameworks from 2013 to 2024: A Bibliometric Analysis","abstract":"Abstract The purpose of this work is to analyze the development direction and prospects in the field of artificial intelligence (AI) in metal–organic frameworks (MOFs) and to provide reference information for related research and industry personnel. The scientific papers on AI in MOFs published in Web of Science database from 2013 to mid-2024 were collected. Bibliometric methods and knowledge mapping visualization software were used to analyze the papers. Both quantitative statistics and qualitative comparative analysis of global scientific papers were done in terms of annual paper trends, papers by major countries, authors, institutions, journals and research topics, respectively. The results showed that the number of published papers has increased in recent years. The top three productive countries are China, the USA and Germany, respectively. The top three productive institutions are Guangzhou University, Northwestern University and Chinese Academy of Sciences, respectively. Reference co-citation analysis classifies references into four clusters, and keyword co-occurrence analysis divides keywords into six clusters. Bibliometric and network analyses were utilized to examine the distribution of research outcomes, enabling scholars to discern the prevailing trends and focal points within the domain of AI-MOFs.","author":[{"family":"Cao","given":"Jian"},{"family":"Zhou","given":"Ling"},{"family":"Gan","given":"Fan"},{"family":"You","given":"Zhipeng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s11837-024-07065-5","URL":"https://doi.org/10.1007/s11837-024-07065-5","source":"openalex"},{"id":"oa:W4411690548","type":"article-journal","title":"Performance of 7 Artificial Intelligence Chatbots on Board-style Endodontic Questions","abstract":"INTRODUCTION: The aim of this study was to assess the overall performance of artificial intelligence chatbots in answering board-style endodontic questions. METHODS: One hundred multiple choice endodontic questions, following the style of American Board of Endodontics Written Exam, were generated by two board-certified endodontists. The questions were submitted to the following chatbots, three times in a row: Gemini Advanced, Gemini, Microsoft Copilot, GPT-3.5, GPT-4o, GPT-4.0, and Claude 3.5 Sonnet. The chatbot was asked to choose the correct response and to explain the justification. The response to the question was considered \"correct\" only if the chatbot picked the right choice in ALL 3 attempts. The quality of reasoning as to why the chatbot selected the answer choice was scored using a three-ordinal scale (0, 1, 2). Two calibrated reviewers scored all 2100 responses independently. Categorical data were analyzed using Chi-square test; ordinal data were analyzed using Kruskal-Wallis and Mann-Whitney tests. RESULTS: The accuracy scores ranged from 48% (Microsoft Copilot) to 71% (Gemini Advanced, GPT-3.5, and Claude 3.5 Sonnet) (P < .05). Gemini Advanced, Gemini, and Microsoft Copilot showed similar performance regardless of the question source (textbook or literature) (P > .05). GPT-3.5, GPT-4o, GPT-4.0 and Claude 3.5 Sonnet performed significantly better with textbook-based questions (P < .05). Reasoning scores showed different distribution among chatbots (P < .05). Gemini Advanced had the highest rate of score 2 (81%) and the lowest rate of score 0 (18.5%). CONCLUSIONS: Comprehensive assessment of seven AI chatbots' performance on board-style endodontic questions revealed their capacities and limitations as educational resources in the field of endodontics.","author":[{"family":"Jalali","given":"Poorya"},{"family":"Mohammadrahimi","given":"Hossein"},{"family":"Wang","given":"Fengming"},{"family":"Sohrabniya","given":"Fatemeh"},{"family":"Ourang","given":"Seyed"},{"family":"Tian","given":"Yuke"},{"family":"Martinho","given":"Frederico"},{"family":"Nosrat","given":"Ali"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.joen.2025.06.014","URL":"https://doi.org/10.1016/j.joen.2025.06.014","source":"openalex"},{"id":"oa:W4412763832","type":"article-journal","title":"Artificial intelligence in pharmacovigilance: a narrative review and practical experience with an expert-defined Bayesian network tool","abstract":"BACKGROUND: Pharmacovigilance is vital for monitoring adverse drug reactions (ADRs) and ensuring drug safety. Traditional methods are slow and inconsistent, but artificial intelligence (AI), through automation and advanced analytics, improves efficiency and accuracy in managing increasing data complexity. AIM: To explore AI's practical applications in pharmacovigilance, focusing on efficiency, process acceleration, and task automation. It also examines the use of an expert-defined Bayesian network for causality assessment in a Pharmacovigilance Centre, demonstrating its impact on decision-making. METHOD: A comprehensive literature narrative review was conducted in MEDLINE (via PubMed), Scopus, and Web of Science using a set of targeted keywords, including but not limited to \"pharmacovigilance\", \"artificial intelligence\", \"adverse drug reactions\" and \"drug safety\". Relevant studies were analysed without restrictions on publication year or language. The search was carried out in January 2025. RESULTS: AI has greatly improved pharmacovigilance by streamlining signal detection, surveillance, and ADR reporting automation. Techniques like data mining and automated signal detection have expedited safety signal identification, while duplicate detection has enhanced data precision in safety evaluations. AI has also refined real-world evidence analysis, deepening drug safety and efficacy insights. Predictive models now anticipate ADRs and drug-drug interactions, enabling proactive patient care. At a regional pharmacovigilance center, the implementation of an expert-defined Bayesian network has optimized causality assessment, reducing processing times from days to hours, minimizing subjectivity, and improving the reliability of drug safety evaluations. CONCLUSION: AI holds significant promise for enhancing pharmacovigilance practices, yet its practical application remains primarily confined to academic research, with integration hindered by data quality issues, regulatory barriers, and the need for more transparent algorithms.","author":[{"family":"Algarvio","given":"Rogério"},{"family":"Conceição","given":"Jaime"},{"family":"Rodrigues","given":"Pedro"},{"family":"Ribeirovaz","given":"Inês"},{"family":"Ferreiradasilva","given":"Renato"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s11096-025-01975-3","URL":"https://doi.org/10.1007/s11096-025-01975-3","source":"openalex"},{"id":"oa:W4412403910","type":"article-journal","title":"A bibliometric analysis of clinical studies on artificial intelligence in emergency medicine","abstract":"BACKGROUND: Interest in artificial intelligence (AI) and machine learning (ML) has grown rapidly in recent years due to the success of modern algorithms across various domains. Emergency departments are fast-paced and resource-constrained environments where timely decision-making is critical. These characteristics make them ideal settings for the integration of AI technologies, which have shown potential to enhance diagnostic accuracy and optimize patient outcomes. This study aims to identify and characterize the scientific literature on AI and ML applications in emergency departments over the past decade. METHODS: A comprehensive search was conducted in the Web of Science database on June 20, 2024. Articles published between 2015 and 2024 were considered. The search was performed using the keywords \"artificial intelligence\" or \"machine learning\" in all fields, limited to the \"emergency medicine\" category. The analysis of the articles included descriptive data on primary publication characteristics, such as the number of authors, citations, country of origin of the coauthors, and journal names. Bibliometric indicators were analyzed to identify publication trends and research themes, cluster analyses of keywords, and thematic maps. RESULTS: A total of 321 articles were analyzed. The average number of citations per article was 10.04, and the annual growth rate was 37.87%. Most publications originated from the United States. Resuscitation, American Journal of Emergency Medicine, Injury-International Journal of the Care of the Injured, and Resuscitation Plus published 107 articles. In 2024, the trending topic of the articles was \"health,\" while \"care\" was the most popular in the last 10 years. The top 5 niche themes were \"medical,\" \"digital transformation,\" \"education,\" \"database,\" and \"emergency care systems.\" CONCLUSION: This bibliometric analysis highlights the growing role of AI in emergency medicine. The findings provide insight into current research directions and may help inform future investigations in this evolving field.","author":[{"family":"Limon","given":"Önder"},{"family":"Bayram","given":"Başak"},{"family":"Çetin","given":"Murat"},{"family":"Limon","given":"Gülsüm"},{"family":"Dirican","given":"Nigar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1097/md.0000000000043282","URL":"https://doi.org/10.1097/md.0000000000043282","source":"openalex"},{"id":"oa:W4408523111","type":"article-journal","title":"Medical Laboratories in Healthcare Delivery: A Systematic Review of Their Roles and Impact","abstract":"Medical laboratories (MLs) are vital in global healthcare delivery, enhancing diagnostic accuracy and supporting clinical decision-making. This systematic review examines the multifaceted contributions of ML, emphasizing their importance in pandemic preparedness, disease surveillance, and the integration of innovative technologies such as artificial intelligence (AI). Medical laboratories are equally crucial to clinical practices, offering essential diagnostic services to identify diseases like infections, metabolic disorders, and malignancies. They monitor treatment effectiveness by analyzing patient samples, enabling healthcare providers to optimize therapies. Additionally, they support personalized medicine by tailoring treatments based on genetic and molecular data and ensure test accuracy through strict quality control measures, thereby enhancing patient care. The methodology for this systematic review follows the PRISMA-ScR guidelines to systematically map evidence and identify key concepts, theories, sources, and knowledge gaps related to the roles and impact of MLs in public health delivery. This review involved systematic searching and filtering of literature from various databases, focusing on studies from 2010 to 2024, primarily in Africa, Asia, and Europe. The selected studies were analyzed to assess their outcomes, strengths, and limitations regarding MLS roles, impacts, and integration within healthcare systems. The goal was to provide comprehensive insights and recommendations based on the gathered data. The article highlights the challenges that laboratories face, especially in low- and middle-income countries (LMICs), where resource constraints hinder effective healthcare delivery. It discusses the potential of AI to improve diagnostic processes and patient outcomes while addressing ethical and infrastructural challenges. This review underscores the necessity for collaborative efforts among stakeholders to enhance laboratory services, ensuring that they are accessible, efficient, and capable of meeting the evolving demands of healthcare systems. Overall, the findings advocate for strengthened laboratory infrastructures and the adoption of advanced technologies to improve health outcomes globally.","author":[{"family":"Adekoya","given":"Adebola"},{"family":"Okezue","given":"Mercy"},{"family":"Menon","given":"Kavitha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/laboratories2010008","URL":"https://doi.org/10.3390/laboratories2010008","source":"openalex"},{"id":"oa:W4410952859","type":"article-journal","title":"The Emergence of Artificial Intelligence-Guided Karyotyping: A Review and Reflection","abstract":"Artificial intelligence (AI) has entered the medical subspecialty of cytogenetics with the recent introduction of AI-guided karyotyping into the clinical laboratory. Karyotyping is an essential component of the cytogenetic analysis process; however, it is both labor-intensive and time-consuming. The introduction of AI algorithms into karyotyping software streamlines this process to provide accurate and abundant auto-karyotyped images for laboratory professionals to review and, also, alters the paradigm for chromosome analysis. Herein, we provide an overview of the AI-guided karyotyping products currently available for clinical use, discuss their utilization in the cytogenetics laboratory, and highlight changes AI-guided karyotyping has brought for early users. Finally, we reflect on our own laboratory observations and experience to discuss issues and practices that may need to adapt to best utilize this promising new technology.","author":[{"family":"Rosenblum","given":"Lynne"},{"family":"Holmes","given":"Julia"},{"family":"Taghiyev","given":"Agshin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/genes16060685","URL":"https://doi.org/10.3390/genes16060685","source":"openalex"},{"id":"oa:W4412112859","type":"article-journal","title":"Artificial intelligence in prostate cancer","abstract":"ABSTRACT: Prostate cancer (PCa) ranks as the second most prevalent malignancy among men worldwide. Early diagnosis, personalized treatment, and prognosis prediction of PCa play a crucial role in improving patients' survival rates. The advancement of artificial intelligence (AI), particularly the utilization of deep learning (DL) algorithms, has brought about substantial progress in assisting the diagnosis, treatment, and prognosis prediction of PCa. The introduction of the foundation model has revolutionized the application of AI in medical treatment and facilitated its integration into clinical practice. This review emphasizes the clinical application of AI in PCa by discussing recent advancements from both pathological and imaging perspectives. Furthermore, it explores the current challenges faced by AI in clinical applications while also considering future developments, aiming to provide a valuable point of reference for the integration of AI and clinical applications.","author":[{"family":"Li","given":"Wei"},{"family":"Hu","given":"Rui"},{"family":"Zhang","given":"Quan"},{"family":"Yu","given":"Zhangsheng"},{"family":"Deng","given":"Longxin"},{"family":"Zhu","given":"Xinhao"},{"family":"Xia","given":"Yujia"},{"family":"Song","given":"Zijian"},{"family":"Cimadamore","given":"Alessia"},{"family":"Chen","given":"Fei"},{"family":"López-Beltrán","given":"Antonio"},{"family":"Montironi","given":"Rodolfo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1097/cm9.0000000000003689","URL":"https://doi.org/10.1097/cm9.0000000000003689","source":"openalex"},{"id":"oa:W4416225738","type":"article-journal","title":"The human factor in explainable artificial intelligence: clinician variability in trust, reliance, and performance","abstract":"Explainable Artificial Intelligence (XAI) is proposed as essential for high-risk applications like healthcare, where it aims to enhance user trust. However, studies often rely on automated metrics rather than user evaluation. We adapt a prototype-based XAI model for image-based gestational age (GA) estimation and evaluate its impact on trust, reliance, and performance, including a novel measure of appropriate reliance. Ten sonographers completed a 3-stage reader study assessing the XAI model's impact on GA estimates. Model predictions reduced clinician mean absolute error (MAE) from 23.5 to 15.7 days, and explanations had a further non-significant reduction to 14.3 days. However, the impact of explanations varied across participants, with some performing worse with explanations than without. Additionally, although explanations increased participant confidence, they had no significant effect on trust or reliance on the model. These counterintuitive results highlight potential pitfalls in deploying XAI, emphasising the need for human studies to capture clinician variability.","author":[{"family":"Nicolson","given":"Angus"},{"family":"Bradburn","given":"Elizabeth"},{"family":"Gal","given":"Yarin"},{"family":"Papageorghiou","given":"Aris"},{"family":"Noble","given":"JA"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-02023-0","URL":"https://doi.org/10.1038/s41746-025-02023-0","source":"openalex"},{"id":"oa:W4409763938","type":"article-journal","title":"Artificial intelligence in the service of sustainable shipping","abstract":"Abstract Shipping constitutes an international endeavor that undoubtedly encapsulates one of the most crucial industries of our time. Owing to shipping, societies can enjoy a variety of goods, effectively transported from one place to another throughout the world, heavily contributing to the global economy and competitive advantages, so much so that Smith and Ricardo would be most proud; yet, the surge of interest in cost minimization and the systemic and traditional focus on accounting costs (that societies are less willing to absorb) have allotted an industry, which is imperative due to its globalized nature, but on the other hand, the said nature has caused negative externalities, including extensive environmental pollution and hazards for human and ecosystemic health. As the contemporary paradigm is one of self-regulated industries, which acknowledge that profitability goes hand in hand with sustainability, in recent years, shipping strives to align itself with sustainability initiatives. This paper provides, through a structured literature review and the use of qualitative data analysis software, the current sustainability practices that influence the shipping industry, to provide a topology as to the hurdles and opportunities that sustainability is yet to face.","author":[{"family":"Prousaloglou","given":"Periklis"},{"family":"Kyriakopoulou-Roussou","given":"Maria"},{"family":"Stavroulakis","given":"Peter"},{"family":"Tsioumas","given":"Vangelis"},{"family":"Papadimitriou","given":"Stratos"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s40722-025-00390-0","URL":"https://doi.org/10.1007/s40722-025-00390-0","source":"openalex"},{"id":"oa:W4406591822","type":"article-journal","title":"Data-Driven Civil Engineering: Applications of Artificial Intelligence, Machine Learning, and Deep Learning","abstract":"Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) are a great advantage that is coming to civil engineering in ways that detail accuracy can be enhanced, many tasks automated, and predictive modeling improved. Across some of the significant subdomains, these technologies allow for eminent progress in structural health monitoring, geotechnical engineering, hydraulic systems, construction management. Currently, AI-powered models such as Artificial Neural Networks (ANNs), fuzzy logic, and evolution-based algorithms allow engineers to predict failure, optimize design, and better resource management of infrastructures. Yet, despite the potential, the adoption of AI, ML, and DL into civil engineering faces a host of challenges including data availability, computational complexity, model interpretability, integration with traditional systems, etc. High-quality, real-time data collection remains expensive and the resource-intensive nature of DL models limits their application to a large scale. In addition, the \"black-box\" nature of these models raises ethical and regulatory issues especially in decisions related to safety. Against this backdrop, this paper reviews current and potential applications of AI, ML, and DL in civil engineering within the framework of benefits and limitations of AI, ML, and DL, focusing on comparisons. Besides that, the paper outlines future directions regarding cloud computing, explainable AI, and regulatory frameworks. With all these changes within the scope of the discipline, AI-driven technologies will be major in safe, efficient, and sustainable infrastructure systems, provided that success is specifically dependent on addressing these key challenges.","author":[{"family":"Jain","given":"Rituraj"},{"family":"Singh","given":"Sitesh"},{"family":"Palaniappan","given":"Damodharan"},{"family":"Parmar","given":"Kumar"},{"family":"Premavathi","given":"T"}],"issued":{"date-parts":[[2025]]},"DOI":"10.31127/tuje.1581564","URL":"https://doi.org/10.31127/tuje.1581564","source":"openalex"},{"id":"oa:W4406849936","type":"article-journal","title":"Attitudes toward artificial intelligence and robots in healthcare in the general population: a qualitative study","abstract":"Background: The growth of the use of artificial intelligence (AI) and robotic solutions in healthcare is accompanied by high expectations for improved efficiency and quality of services. However, the use of such technologies can be a source of anxiety for patients whose expectations and experiences with such technology differ from medical staff's. This study assessed attitudes toward AI and robots in delivering health services and performing various tasks in medicine and related fields in Polish society. Methods: 50 semistructured in-depth interviews were conducted with participants of diversified socio-demographic profiles. The interviewees were initially recruited for the interviews in a convenience sample; then, the process was continued using the snowballing technique. The interviews were transcribed and analyzed using the MAXQDA Analytics Pro 2022 program (release 22.7.0). An interpretative approach to qualitative content analysis was applied to the responses to the research questions. Results: The analysis of interviews yielded three main themes: positive and negative perceptions of the use of AI and robots in healthcare and ontological concerns about AI, which went beyond objections about the usefulness of the technology. Positive attitudes toward AI and robots were associated with overall higher trust in technology, the need to adequately respond to demographic challenges, and the conviction that AI and robots can lower the workload of medical personnel. Negative attitudes originated from convictions regarding unreliability and the lack of proper technological and political control over AI; an equally important topic was the inability of artificial entities to feel and express emotions. The third theme was that the potential interaction with machines equipped with human-like traits was a source of insecurity. Conclusions: The study showed that patients' attitudes toward AI and robots in healthcare vary according to their trust in technology, their recognition of urgent problems in healthcare (staff workload, time of diagnosis), and their beliefs regarding the reliability and functioning of new technologies. Emotional concerns about contact with artificial entities looking or performing like humans are also important to respondents' attitudes.","author":[{"family":"Smoła","given":"Paulina"},{"family":"Młoźniak","given":"Iwona"},{"family":"Wojcieszko","given":"Monika"},{"family":"Zwierczyk","given":"Urszula"},{"family":"Kobryn","given":"Mateusz"},{"family":"Rzepecka","given":"Elżbieta"},{"family":"Duplaga","given":"Mariusz"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fdgth.2025.1458685","URL":"https://doi.org/10.3389/fdgth.2025.1458685","source":"openalex"},{"id":"oa:W4407931364","type":"article-journal","title":"Artificial intelligence-enhanced electrocardiography for the identification of a sex-related cardiovascular risk continuum: a retrospective cohort study","abstract":"BACKGROUND: Females are typically underserved in cardiovascular medicine. The use of sex as a dichotomous variable for risk stratification fails to capture the heterogeneity of risk within each sex. We aimed to develop an artificial intelligence-enhanced electrocardiography (AI-ECG) model to investigate sex-specific cardiovascular risk. METHODS: In this retrospective cohort study, we trained a convolutional neural network to classify sex using the 12-lead electrocardiogram (ECG). The Beth Israel Deaconess Medical Center (BIDMC) secondary care dataset, comprising data from individuals who had clinically indicated ECGs performed in a hospital setting in Boston, MA, USA collected between May, 2000, and March, 2023, was the derivation cohort (1 163 401 ECGs). 50% of this dataset was used for model training, 10% for validation, and 40% for testing. External validation was performed using the UK Biobank cohort, comprising data from volunteers aged 40-69 years at the time of enrolment in 2006-10 (42 386 ECGs). We examined the difference between AI-ECG-predicted sex (continuous) and biological sex (dichotomous), termed sex discordance score. FINDINGS: AI-ECG accurately identified sex (area under the receiver operating characteristic 0·943 [95% CI 0·942-0·943] for BIDMC and 0·971 [0·969-0·972] for the UK Biobank). In BIDMC outpatients with normal ECGs, an increased sex discordance score was associated with covariate-adjusted increased risk of cardiovascular death in females (hazard ratio [HR] 1·78 [95% CI 1·18-2·70], p=0·006) but not males (1·00 [0·63-1·58], p=0·996). In the UK Biobank cohort, the same pattern was seen (HR 1·33 [95% CI 1·06-1·68] for females, p=0·015; 0·98 [0·80-1·20] for males, p=0·854). Females with a higher sex discordance score were more likely to have future heart failure or myocardial infarction in the BIDMC cohort and had more male cardiac (increased left ventricular mass and chamber volumes) and non-cardiac phenotypes (increased muscle mass and reduced body fat percentage) in both cohorts. INTERPRETATION: Sex discordance score is a novel AI-ECG biomarker capable of identifying females with disproportionately elevated cardiovascular risk. AI-ECG has the potential to identify female patients who could benefit from enhanced risk factor modification or surveillance. FUNDING: British Heart Foundation.","author":[{"family":"Sau","given":"Arunashis"},{"family":"Sieliwończyk","given":"Ewa"},{"family":"Patlatzoglou","given":"Konstantinos"},{"family":"Pastika","given":"Libor"},{"family":"Mcgurk","given":"Kathryn"},{"family":"Ribeiro","given":"Antônio"},{"family":"Ribeiro","given":"Antônio"},{"family":"Ho","given":"Jennifer"},{"family":"Peters","given":"Nicholas"},{"family":"Ware","given":"James"},{"family":"Tayal","given":"Upasana"},{"family":"Kramer","given":"Daniel"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.landig.2024.12.003","URL":"https://doi.org/10.1016/j.landig.2024.12.003","source":"openalex"},{"id":"oa:W4407028552","type":"article-journal","title":"Convergence of evolving artificial intelligence and machine learning techniques in precision oncology","abstract":"The confluence of new technologies with artificial intelligence (AI) and machine learning (ML) analytical techniques is rapidly advancing the field of precision oncology, promising to improve diagnostic approaches and therapeutic strategies for patients with cancer. By analyzing multi-dimensional, multiomic, spatial pathology, and radiomic data, these technologies enable a deeper understanding of the intricate molecular pathways, aiding in the identification of critical nodes within the tumor's biology to optimize treatment selection. The applications of AI/ML in precision oncology are extensive and include the generation of synthetic data, e.g., digital twins, in order to provide the necessary information to design or expedite the conduct of clinical trials. Currently, many operational and technical challenges exist related to data technology, engineering, and storage; algorithm development and structures; quality and quantity of the data and the analytical pipeline; data sharing and generalizability; and the incorporation of these technologies into the current clinical workflow and reimbursement models.","author":[{"family":"Fountzilas","given":"Elena"},{"family":"Pearce","given":"Tillman"},{"family":"Baysal","given":"Mehmet"},{"family":"Chakraborty","given":"Abhijit"},{"family":"Tsimberidou","given":"Apostolia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01471-y","URL":"https://doi.org/10.1038/s41746-025-01471-y","source":"openalex"},{"id":"oa:W4409291977","type":"article-journal","title":"Application of Artificial Intelligence in Medical Imaging: Current Status and Future Directions","abstract":"ABSTRACT A revolution in medical diagnosis and treatment is being driven by the use of artificial intelligence (AI) in medical imaging. The diagnostic efficacy and accuracy of medical imaging are greatly enhanced by AI technologies, especially deep learning, that performs image recognition, feature extraction, and pattern analysis. Furthermore, AI has demonstrated significant promise in assessing the effects of treatments and forecasting the course of diseases. It also provides doctors with more advanced tools for managing the conditions of their patients. AI is poised to play a more significant role in medical imaging, especially in real‐time image processing and multimodal fusion. By integrating multiple forms of image data, multimodal fusion technology provides more comprehensive disease information, whereas real‐time image analysis can assist surgeons in making more precise decisions. By tailoring treatment regimens to each patient's unique needs, AI enhances both the effectiveness of treatment and the patient experience. Overall, AI in medical imaging promises a bright future, significantly enhancing diagnostic precision and therapeutic efficacy, and ultimately delivering higher‐quality medical care to patients.","author":[{"family":"Yang","given":"Yixin"},{"family":"Ye","given":"Lan"},{"family":"Feng","given":"Zhanhui"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/ird3.70008","URL":"https://doi.org/10.1002/ird3.70008","source":"openalex"},{"id":"oa:W4410386467","type":"article-journal","title":"Perspectives on the Current and Future State of Artificial Intelligence in Medical Genetics","abstract":"Artificial intelligence (AI) is rapidly transforming numerous aspects of daily life, including clinical practice and biomedical research. In light of this rapid transformation, and in the context of medical genetics, we assembled a group of leaders in the field to respond to the question about how AI is affecting, and especially how AI will affect, medical genetics. The authors who contributed to this collection of essays intentionally represent different areas of expertise, career stages, and geographies, and include diverse types of clinicians, computer scientists, and researchers. The individual pieces cover a wide range of areas related to medical genetics; we expect that these pieces may provide helpful windows into the ways in which AI is being actively studied, used, and considered in medical genetics.","author":[{"family":"Solomon","given":"Benjamin"},{"family":"Cheatham","given":"Morgan"},{"family":"Guimarães","given":"Thales"},{"family":"Duong","given":"Dat"},{"family":"Haendel","given":"Melissa"},{"family":"Hsieh","given":"Tzung‐chien"},{"family":"Javanmardi","given":"Behnam"},{"family":"Johnson","given":"Britt"},{"family":"Krawitz","given":"Peter"},{"family":"Kruszka","given":"Paul"},{"family":"Laurent","given":"Tim"},{"family":"Lee","given":"Ni‐chung"},{"family":"Mcwalter","given":"Kirsty"},{"family":"Michaelides","given":"Michel"},{"family":"Mohnike","given":"Klaus"},{"family":"Pontikos","given":"Nikolas"},{"family":"Sacoto","given":"María"},{"family":"Shwetar","given":"Yousif"},{"family":"Ustach","given":"Vincent"},{"family":"Waikel","given":"Rebekah"},{"family":"Woof","given":"William"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/ajmg.a.64118","URL":"https://doi.org/10.1002/ajmg.a.64118","source":"openalex"},{"id":"oa:W4412584217","type":"article-journal","title":"Artificial Intelligence in healthcare: Transformative applications, ethical challenges, and future directions in medical diagnostics and personalized medicine","abstract":"The harmonization of Artificial Intelligence (AI) in the healthcare sector has revolutionized medical diagnostics, treatment planning, and patient management. Over the past decade, AI-powered technologies have demonstrated significant potential in improving accuracy, efficiency, and accessibility in healthcare services. Machine learning algorithms and deep learning models have been employed for disease prediction, early diagnosis, and personalized medicine, enhancing patient outcomes. AI-driven robotic surgeries, virtual health assistants, and predictive analytics have optimized medical workflows, reducing human errors and optimizing resource utilization. Despite these advancements, challenges such as data privacy, ethical concerns, and the need for regulatory frameworks remain significant barriers to widespread adoption. This paper explores the evolution of AI in healthcare, focusing on its applications, benefits, and limitations. Through a comprehensive analysis of past developments and current trends, this study highlights the transformative role of AI in reshaping the medical landscape. As technology continues to grow, AI is poised to play an even more critical role in future healthcare innovations, ultimately improving the quality of patient care and medical decision-making.","author":[{"family":"Hossain","given":"Md"},{"family":"Rahman","given":"Md"},{"family":"Hossan","given":"Kazi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.30574/ijsra.2025.15.1.0954","URL":"https://doi.org/10.30574/ijsra.2025.15.1.0954","source":"openalex"},{"id":"oa:W4406035098","type":"article-journal","title":"Federated Learning Lifecycle Management for Distributed Medical Artificial Intelligence Applications: A Case Study on Post-Transcatheter Aortic Valve Replacement Complication Prediction Solution","abstract":"The evolution of artificial intelligence (AI) has unveiled considerable prospects for delivering efficacious solutions in the medical domain. Nevertheless, existing legal frameworks and concerns regarding data privacy associated with medical information impose substantial constraints on implementing AI solutions in this domain. Federated learning is a paradigm that enables the training of machine learning models in a decentralized manner without transferring data to a central repository, allowing model development while preserving data privacy across medical and other industries. This study provided a comprehensive framework for applying federated learning to AI solutions in the medical domain. It advocates a sustainable learning ecosystem by overseeing federated learning servers and clients and evaluating performance by managing the federated learning lifecycle. To enhance its practical relevance, this framework includes a detailed process for continuous lifecycle management, involving model deployment, aggregation, testing, evaluation, versioning, and real-time monitoring through the FedOps platform, supporting a sustainable solution. In this study, the feasibility of the proposed methodology was verified using a post-transcatheter aortic valve replacement (TAVR) complication–prediction framework. The performance of the solution after transitioning to a federated learning approach was compared with that of an existing centralized solution. The findings indicated no statistically significant difference in performance between the two methodologies. This implies that federated learning can augment data usability and facilitate the integration of AI technologies into the medical domain, where the preservation of data privacy is critically important.","author":[{"family":"Jung","given":"Min"},{"family":"Song","given":"Inseo"},{"family":"Lee","given":"Kang"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/app15010378","URL":"https://doi.org/10.3390/app15010378","source":"openalex"},{"id":"oa:W4406927485","type":"article-journal","title":"Explainable AI for Healthcare: Training Healthcare Workers to Use Artificial Intelligence Techniques to Reduce Medical Negligence in Ghana’s Public Health Act, 2012 (Act 851)","abstract":"This analysis examines whether Ghana’s Public Health Act, 2012 (Act 851) imposes adequate legal responsibilities on healthcare facilities concerning personnel training on artificial intelligence (AI) systems and implementation of medical negligence reduction measures. Through an evaluative review of Act 851 provisions on staff qualifications, technology deployment, quality care, safety planning, and risk management benchmarks relative to precedents in Ghana and other countries, critical gaps in binding regulations to incentivize organizational capacity building for mitigating errors, hazards and liabilities from substandard practices were identified. Key recommendations include amending Act 851 to mandate credentialing assurance frameworks, clinical audits, risk assessment models and transparency requirements around reporting quality indicators. Strengthening policy directives will compel internal monitoring, governance, and accountability among healthcare facilities as multilayered negligence prevention strategies. Scientific contributions highlight deficiencies in Ghana’s health legislation regarding contemporary challenges like AI adoption risks and propose legal reforms to modernize regulations to support safer, responsible healthcare delivery nationwide.","author":[{"family":"Mensah","given":"George"},{"family":"Mıjwıl","given":"Maad"},{"family":"Abotaleb","given":"Mostafa"},{"family":"Ali","given":"Guma"},{"family":"Dutta","given":"Pushan"},{"family":"Mzili","given":"Toufik"},{"family":"Eid","given":"Marwa"}],"issued":{"date-parts":[[2025]]},"DOI":"10.70470/edraak/2025/001","URL":"https://doi.org/10.70470/edraak/2025/001","source":"openalex"},{"id":"oa:W4410019803","type":"article-journal","title":"Evaluation of an Ambient Artificial Intelligence Documentation Platform for Clinicians","abstract":"Importance: The increase of electronic health record (EHR) work negatively impacts clinician well-being. One potential solution is incorporating an ambient artificial intelligence (AI) documentation platform. Objective: To understand clinician experience before and after implementing ambient AI. Design, Setting, and Participants: This quality improvement study was a pilot evaluation with before and after survey and EHR metrics conducted at a large health care organization in Northern and Central California. Clinicians were purposively sampled to be representative of region and specialty. Ambient AI was implemented in April 2024 with EHR data from 3 months before and after implementation. Data were analyzed from May to September 2024. Exposure: Ambient AI access. Main Outcomes and Measures: Metrics of time were examined in notes per appointment, off-hour EHR activities (5:30 pm to 7:00 am on weekdays and nonscheduled weekends and holidays), documentation note length, progress note length, NASA Task Load Index (NASA-TLX) score, mini-Z burnout question, and overall experience. It was hypothesized that time in notes per appointment would decrease and clinical well-being would improve. Logistic regression and linear mixed-effect models were used. Results: Among 100 clinicians (53 male [53.0%]; mean [SD] age, 48.9 [11.0] years), 58 clinicians (58.0%) were in primary care and 92 clinicians had EHR metrics. Among 57 clinicians who completed both preimplementation and postimplementation surveys, there was a decrease in burnout from 24 clinicians (42.1%) to 20 clinicians (35.1%), although this was not a significant difference (P = .12). Mean (SD) NASA-TLX scores all decreased after using ambient AI: mental demand of note writing (12.2 [4.0] to 6.3 [3.7]), hurried or rushed pace (13.2 [4.0] to 6.4 [4.2]), and effort to accomplish note writing (12.5 [4.1] to 7.4 [4.3]) (all P < .001). Mean (SD) time in notes per appointment significantly decreased from 6.2 (4.0) to 5.3 (3.5) minutes (P < .001), with a bigger decrease for female vs male clinicians (8.1 [3.9] to 6.7 [3.6] minutes vs 4.7 [3.5] to 4.2 [3.1] minutes; P = .001). More primary care clinicians (33 of 38 clinicians [85.8%]) reported that ambient AI improved overall satisfaction at work compared with clinicians in medical (4 of 11 clinicians [36.4%]) and surgical (4 of 8 clinicians [50.0%]) subspecialties (P < .001). After adjusting for participant characteristics, model results suggested that mean scores for NASA-TLX decreased for mental demand (-6.12 [95% CI, -7.52 to -4.72]), hurried or rushed pace (-6.96 [95% CI, -8.42 to -5.50]), and effort to accomplish note writing (-5.57 [95% CI, -6.93 to -4.21]), while mean time in note taking decreased by less than 1 minute per appointment (0.91 minutes [95% CI, -1.20 to -0.62 minutes]) (all P < .001). Conclusions and Relevance: This study found that ambient AI was associated with improved overall experience and time in notes for clinicians but with varying outcomes by sex and specialty. Future research should investigate outcomes after widescale expansion of this rapidly evolving technology.","author":[{"family":"Stults","given":"Cheryl"},{"family":"Deng","given":"Sien"},{"family":"Martinez","given":"Meghan"},{"family":"Wilcox","given":"Joseph"},{"family":"Szwerinski","given":"Nina"},{"family":"Chen","given":"Kevin"},{"family":"Driscoll","given":"Stephanie"},{"family":"Washburn","given":"Joanna"},{"family":"Jones","given":"Veena"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1001/jamanetworkopen.2025.8614","URL":"https://doi.org/10.1001/jamanetworkopen.2025.8614","source":"openalex"},{"id":"oa:W4406797838","type":"article-journal","title":"Artificial Intelligence-Empowered Radiology—Current Status and Critical Review","abstract":"Humanity stands at a pivotal moment of technological revolution, with artificial intelligence (AI) reshaping fields traditionally reliant on human cognitive abilities. This transition, driven by advancements in artificial neural networks, has transformed data processing and evaluation, creating opportunities for addressing complex and time-consuming tasks with AI solutions. Convolutional networks (CNNs) and the adoption of GPU technology have already revolutionized image recognition by enhancing computational efficiency and accuracy. In radiology, AI applications are particularly valuable for tasks involving pattern detection and classification; for example, AI tools have enhanced diagnostic accuracy and efficiency in detecting abnormalities across imaging modalities through automated feature extraction. Our analysis reveals that neuroimaging and chest imaging, as well as CT and MRI modalities, are the primary focus areas for AI products, reflecting their high clinical demand and complexity. AI tools are also used to target high-prevalence diseases, such as lung cancer, stroke, and breast cancer, underscoring AI's alignment with impactful diagnostic needs. The regulatory landscape is a critical factor in AI product development, with the majority of products certified under the Medical Device Directive (MDD) and Medical Device Regulation (MDR) in Class IIa or Class I categories, indicating compliance with moderate-risk standards. A rapid increase in AI product development from 2017 to 2020, peaking in 2020 and followed by recent stabilization and saturation, was identified. In this work, the authors review the advancements in AI-based imaging applications, underscoring AI's transformative potential for enhanced diagnostic support and focusing on the critical role of CNNs, regulatory challenges, and potential threats to human labor in the field of diagnostic imaging.","author":[{"family":"Obuchowicz","given":"Rafał"},{"family":"Lasek","given":"Julia"},{"family":"Wodziński","given":"Marek"},{"family":"Piórkowski","given":"Adam"},{"family":"Strzelecki","given":"Michał"},{"family":"Nurzyǹska","given":"Karolina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/diagnostics15030282","URL":"https://doi.org/10.3390/diagnostics15030282","source":"openalex"},{"id":"oa:W7129033494","type":"article-journal","title":"Psychometric Properties of the Chinese Version of the Medical Artificial Intelligence Readiness Scale for Medical Students (MAIRS-MS)","abstract":"This study aimed to evaluate the psychometric properties of the Chinese version of the Medical Artificial Intelligence Readiness Scale for Medical Students (MAIRS-MS) and assess its applicability among undergraduate students in medicine and health-related disciplines in China. Using Brislin’s translation model, we translated and culturally adapted the original (English) scale to produce the Chinese version. Between May and July 2024, we used convenience sampling to recruit undergraduate students from a medical university in Changsha City, Hunan Province, China (population size = 520). We collected a total of 480 valid responses (participation rate = 92.3%) through the Wenjuanxing platform. We analyzed data using R and AMOS 29.0. Content validity was supported by two rounds of Delphi expert consultation, with item-level content validity indices (I-CVI) ranging from 0.80 to 1.00 and a scale-level index (S-CVI) of 0.95. The Chinese version of the MAIRS-MS showed good internal consistency, with a total Cronbach’s alpha coefficient of 0.90 and subscale coefficients all exceeding 0.78. The split-half reliability was 0.94, and the test-retest reliability was 0.95. Exploratory factor analysis supported the instrument’s original four-factor structure—Cognition, Ability, Vision, and Ethics—with a KMO value of 0.93, cumulative variance explained of 67.2%, and all item loadings greater than 0.40. Confirmatory factor analysis indicated a good model fit (χ2/df = 1.11, RMSEA = 0.02, CFI = 0.93, TLI= 0.99). In sum, the Chinese version of the MAIRS-MS demonstrated satisfactory content validity, internal consistency, and structural validity, supporting its use as a reliable tool for assessing AI readiness among Chinese undergraduate health professional students.","author":[{"family":"Luo","given":"Chuhong"},{"family":"Xie","given":"Siqi"},{"family":"Yuan","given":"Rong"},{"family":"Li","given":"Pingshuang"},{"family":"Yang","given":"Can"},{"family":"Cao","given":"Jixia"},{"family":"He","given":"YD"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1080/10401334.2026.2627455","URL":"https://doi.org/10.1080/10401334.2026.2627455","source":"europepmc"},{"id":"oa:W4412033531","type":"article-journal","title":"Impact of artificial intelligence on academic performance in medical education: A systematic review","abstract":"Artificial intelligence (AI) plays a significant role in improving the quality of medical education. This study aimed to investigate the application of AI in medical education. A systematic review was conducted of all educational intervention studies in medical courses from January 1986 to 2023. Of the 16755 studies initially identified by our search, 7387 remained after removing duplicates. After that, 6205 studies were excluded for the title and abstract screening. Then, 15 full-text articles were included in our final review. The following keywords were used: artificial intelligence, machine intelligence, medical education, medical teaching, precision medical teaching, and precision medical education. A total of 16745 articles were identified from ISI, PubMed, Scopus, and Educational Resources and Information Center (ERIC) databases. Fourteen studies met the eligibility criteria. The quality of the included articles was appraised by the best evidence medical education (BEME) review of the education portfolio. PICO consists of P = medical student, I: use of AI, C = do not use AI, and I = enhance health professions students’ knowledge, attitudes, and skills. The included studies were synthesized and categorized according to the Kirkpatrick model. Educational intervention outcomes were categorized into three parts [1] : improvement in learners’ knowledge (N = 6 studies) [2] ; enhancement of students’ attitudes (N = 3 studies); and [3] acquisition of learners’ skills (N = 10 studies). The reviewed studies examined the impact of AI and virtual reality on enhancing health profession students’ knowledge, attitudes, and skills. However, more research is still needed on integrating AI into diverse curriculum models and sustaining AI’s role in real-world education systems.","author":[{"family":"Kalantarion","given":"Masomeh"},{"family":"Heidari","given":"Mehrsa"},{"family":"Khajeali","given":"Nasrin"},{"family":"Khorrami","given":"Zahra"},{"family":"Amini","given":"Mitra"}],"issued":{"date-parts":[[2025]]},"DOI":"10.4103/jehp.jehp_2071_23","URL":"https://doi.org/10.4103/jehp.jehp_2071_23","source":"openalex"},{"id":"doi:10.1007/s13534-025-00523-2","type":"article-journal","title":"Institutionalizing convergence education for medical artificial intelligence.","abstract":"As artificial intelligence (AI) becomes increasingly central to modern healthcare, medical education must move beyond passive knowledge transfer and adopt a system-wide approach to convergence training. This narrative review shares a 5-year case study from Seoul National University College of Medicine (SNU Medicine), which developed a comprehensive, multi-level model for integrating AI into medical education. Instead of relying on pilot programs or piecemeal curriculum updates, SNU Medicine established a governance-driven, modular framework that includes institutional infrastructure, interdisciplinary teaching strategies, cross-campus credit integration, and alignment with national digital health policies. Based on this long-term case, we propose four key design principles-modularity, transdisciplinary alignment, infrastructure-curriculum coupling, and policy embeddedness-as a framework for creating scalable and sustainable convergence education in medical AI. While rooted in Korea's unique policy environment, this model provides transferable insights for medical institutions worldwide, particularly those operating within public or policy-constrained environments.","author":[{"family":"Park","given":"Tae"},{"family":"Seo","given":"Jong"},{"family":"Yoon","given":"Hyung"},{"family":"Lee","given":"Kyu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s13534-025-00523-2","URL":"https://doi.org/10.1007/s13534-025-00523-2","source":"europepmc"},{"id":"doi:10.5847/wjem.j.1920-8642.2025.095","type":"article-journal","title":"Performance of a novel medical artificial intelligence large language model on supporting decision-making for emergency patients with suspected sepsis.","abstract":"BACKGROUND: Large language models (LLMs) are being explored for disease prediction and diagnosis; however, their effi cacy for early sepsis identifi cation in emergency departments (EDs) remains unexplored.This study aims to evaluate MedGo, a novel medical LLM, as a decision-support tool for clinicians managing patients with suspected sepsis.METHODS: This retrospective study included anonymized medical records of 203 patients (mean age 79.9±10.2years) with confi rmed sepsis from a tertiary hospital ED between January 2023 and January 2024.MedGo performance across nine sepsis-related assessment tasks was compared with that of two junior (<3 years of experience) and two senior (>10 years of experience) ED physicians.Assessments were scored on a 5-point Likert scale for accuracy, comprehensiveness, readability, and case-analysis skills.RESULTS: MedGo demonstrated diagnostic performance comparable to that of senior physicians across most metrics, achieving a median Likert score of 4 in accuracy, comprehensiveness, and readability.MedGo signifi cantly outperformed junior physicians (P<0.001 for accuracy and case-analysis skills).MedGo assistance significantly enhanced both junior (P<0.001) and senior (P<0.05)physicians' diagnostic accuracy.Notably, MedGo-assisted junior physicians achieved accuracy levels comparable to those of unassisted senior physicians.MedGo maintained consistent performance across varying sepsis severities.CONCLUSION: MedGo shows significant diagnostic efficacy for sepsis and effectively supports clinicians in the ED, particularly enhancing junior physicians' performance.Our study highlights the potential of MedGo as a valuable decision-support tool for sepsis management, paving the way for specialized sepsis AI models.","author":[{"family":"Jiang","given":"Sen"},{"family":"Liu","given":"Xiandong"},{"family":"Liu","given":"Tong"},{"family":"Gu","given":"Yi"},{"family":"An","given":"Bo"},{"family":"Wang","given":"Chunxue"},{"family":"Zhao","given":"Dongyang"},{"family":"Zhang","given":"Haitao"},{"family":"Tang","given":"Lunxian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5847/wjem.j.1920-8642.2025.095","URL":"https://doi.org/10.5847/wjem.j.1920-8642.2025.095","source":"europepmc"},{"id":"doi:10.5281/zenodo.22021373","type":"article-journal","title":"MYCIN: A RULE-BASED EXPERT SYSTEM FOR ANTIMICROBIAL THERAPY SELECTION IN MEDICAL DIAGNOSIS — A REVIEW AND CRITICAL ANALYSIS","abstract":"ABSTRACT Expert systems represent one of the earliest and most influential branches of applied artificial intelligence, aiming to replicate the decision-making capability of a human specialist within a narrow, well-defined domain. Among the pioneering efforts in this field, MYCIN, developed at Stanford University in the mid-1970s, stands out as a landmark rule-based system designed to assist physicians in diagnosing bacterial infections, particularly bacteremia and meningitis, and in recommending appropriate antimicrobial therapy. This paper presents a comprehensive review of MYCIN's architecture, knowledge representation scheme, inferencing strategy, and its use of certainty factors to manage diagnostic uncertainty. The system's backward-chaining inference engine, its separation of the knowledge base from the inference mechanism, and its natural language explanation facility are examined in detail. The paper further evaluates MYCIN's diagnostic performance as reported in early validation studies, discusses the ethical and legal barriers that prevented its clinical deployment, and draws comparisons with contemporaneous and later expert systems such as DENDRAL, INTERNIST-I, and CADUCEUS. Finally, the paper reflects on MYCIN's lasting influence on modern clinical decision-support systems and artificial intelligence-based diagnostic tools, arguing that its architectural principles continue to inform the design of contemporary knowledge-based and hybrid AI systems in healthcare. Keywords: Expert System; MYCIN; Artificial Intelligence; Medical Diagnosis; Rule-Based Reasoning; Certainty Factor; Knowledge-Based System; Clinical Decision Support","author":[{"family":"Radhalakshmi","given":"GVP"},{"family":"Vsrimuki"},{"family":"Sprethika"},{"family":"Drssuganyadevi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22021373","URL":"https://doi.org/10.5281/zenodo.22021373","source":"datacite"},{"id":"doi:10.5281/zenodo.22021374","type":"article-journal","title":"MYCIN: A RULE-BASED EXPERT SYSTEM FOR ANTIMICROBIAL THERAPY SELECTION IN MEDICAL DIAGNOSIS — A REVIEW AND CRITICAL ANALYSIS","abstract":"ABSTRACT Expert systems represent one of the earliest and most influential branches of applied artificial intelligence, aiming to replicate the decision-making capability of a human specialist within a narrow, well-defined domain. Among the pioneering efforts in this field, MYCIN, developed at Stanford University in the mid-1970s, stands out as a landmark rule-based system designed to assist physicians in diagnosing bacterial infections, particularly bacteremia and meningitis, and in recommending appropriate antimicrobial therapy. This paper presents a comprehensive review of MYCIN's architecture, knowledge representation scheme, inferencing strategy, and its use of certainty factors to manage diagnostic uncertainty. The system's backward-chaining inference engine, its separation of the knowledge base from the inference mechanism, and its natural language explanation facility are examined in detail. The paper further evaluates MYCIN's diagnostic performance as reported in early validation studies, discusses the ethical and legal barriers that prevented its clinical deployment, and draws comparisons with contemporaneous and later expert systems such as DENDRAL, INTERNIST-I, and CADUCEUS. Finally, the paper reflects on MYCIN's lasting influence on modern clinical decision-support systems and artificial intelligence-based diagnostic tools, arguing that its architectural principles continue to inform the design of contemporary knowledge-based and hybrid AI systems in healthcare. Keywords: Expert System; MYCIN; Artificial Intelligence; Medical Diagnosis; Rule-Based Reasoning; Certainty Factor; Knowledge-Based System; Clinical Decision Support","author":[{"family":"Radhalakshmi","given":"GVP"},{"family":"Vsrimuki"},{"family":"Sprethika"},{"family":"Drssuganyadevi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22021374","URL":"https://doi.org/10.5281/zenodo.22021374","source":"datacite"},{"id":"doi:10.17605/osf.io/q7wkn","type":"article-journal","title":"The Gap Between Artificial Intelligence Use and Disclosure in Academic Publishing: A Scoping Review Protocol","abstract":"This scoping review will systematically identify, map, and synthesize empirical evidence on the gap between artificial intelligence use and disclosure in academic publishing. The review will include studies reporting empirical data on the prevalence of artificial intelligence tool use in academic research or manuscript preparation, studies reporting the prevalence of artificial intelligence disclosure statements in published scholarly literature, and studies examining factors associated with disclosure or non-disclosure behaviour in academic contexts. The review will follow Joanna Briggs Institute methodology for scoping reviews and will be reported according to the PRISMA extension for Scoping Reviews. Eligible evidence will include surveys, content analyses, bibliometric or database studies, qualitative studies, experimental studies, and relevant grey literature where primary empirical data and methodology are reported. The planned synthesis will quantify the disclosure gap, defined as the difference between survey-reported artificial intelligence use prevalence and content-analysis disclosure prevalence, and will synthesize individual, contextual, and structural predictors of non-disclosure. Expected outputs include a structured quantification of the artificial intelligence disclosure gap across available evidence and a predictor synthesis framework identifying factors that may explain non-disclosure across disciplines, publication contexts, and policy environments.","author":[{"family":"Poirier","given":"Philippe"},{"family":"Parent-Harvey","given":"Stephanie"},{"family":"Bozzo","given":"Isabella"},{"family":"Harvey","given":"Edward"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/q7wkn","URL":"https://doi.org/10.17605/osf.io/q7wkn","source":"datacite"},{"id":"doi:10.48550/arxiv.2608.18260","type":"manuscript","title":"Redakto - The Incognito Tab for LLMs","abstract":"Large Language Models (LLMs) are being increasingly used in everyday applications. A major challenge in the context of LLMs or Artificial Intelligence (AI) in general is to ensure privacy when using them, meaning that personally identifiable information (PII) is removed from any text that enters an LLM. These challenges have become more urgent with novel EU legislation. Uncertainty around LLM usage with respect to privacy concerns in EU countries can be a major blocker for the speed of innovation and transfer from research to applications. Here we present \\textbf{Redakto}, a tool that can be used for anonymizing text prior to feeding it to an LLM or other downstream text processing. We provide state-of-the-art functionalities for both redaction of PII but also when used for pseudonymization. These functionalities are exposed such that they can easily be used by end-users, through the Redakto web application, and by developers and researchers, via REST APIs and model context protocol (MCP) hooks. The implementation is fully open source, requires modest compute resources, and can be readily deployed on local hardware. In contrast to prior work and in order to better assess the quality of the anonymized texts, we conduct extensive empirical evaluations on textual data from legal and medical domain with respect to both privacy and utility of the redacted texts. Our empirical results demonstrate that the texts anonymized with different redaction strategies achieve utility scores on par with the original texts, suggesting that anonymization with Redakto can be used for LLM tasks without substantial negative impact for the tasks we explored.","author":[{"family":"Saha","given":"Saurav"},{"family":"Röhr","given":"Tom"},{"family":"Bießmann","given":"Felix"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2608.18260","URL":"https://doi.org/10.48550/arxiv.2608.18260","source":"datacite"},{"id":"doi:10.48550/arxiv.2608.18166","type":"manuscript","title":"TractoGraphVLM: A Unified Vision-Language Framework for White Matter Tractography","abstract":"Vision language models have transformed 2D medical imaging, yet extending them to 3D white matter tractography remains challenging due to the complex topology of fiber bundles. We introduce TractoGraphVLM, a unified framework for four tasks, bundle classification, text-to-tract retrieval, anatomical captioning, and visual question answering, built on a shared GPS architecture, training procedure, and read-out design. Fiber bundles are represented as streamline graphs whose nodes encode 3D position and tangent orientation. A General, Powerful, Scalable (GPS) graph transformer produces bundle embeddings aligned with a frozen BiomedBERT text encoder via contrastive learning, while a BioGPT decoder with visual prefix tokens generates captions and answers. A single shared encoder and decoder is trained jointly across all four tasks and evaluated from one checkpoint. Trained on HCP Young Adult subjects, TractoGraphVLM achieves 91.8% bundle classification accuracy, 84.7% retrieval R@1, BLEU-4=20.1, ROUGE-L=66.8, and 66.4% VQA accuracy on a held-out test set. The same checkpoints transfer zero-shot to HCP Aging subjects, with a modest drop on discriminative tasks and a larger drop on generative tasks, showing robustness to age and acquisition shift. Language supervision yields richer representations than label-only training, recovering structure like hemisphere and fiber family, carried by captions but never given as a label. Swapping only the visual encoder, graphs preserving fiber orientation outperform volumetric baselines, with GPS giving the best balance. Generative metrics measure consistency with a structured knowledge base rather than independent clinical text; even so, TractoGraphVLM shows that classifying, retrieving, describing, and answering questions about a white matter bundle can be served by one jointly trained model that learns transferable neuroanatomy from language alone.","author":[{"family":"Kumar","given":"Gurucharan"},{"family":"Mendola","given":"Janine"},{"family":"Shmuel","given":"Amir"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2608.18166","URL":"https://doi.org/10.48550/arxiv.2608.18166","source":"datacite"},{"id":"doi:10.5281/zenodo.20672289","type":"article-journal","title":"Healthcare Resource Allocation Predictor: A Machine Learning Approach for Optimizing Healthcare Resource Distribution","abstract":"This paper presents a Healthcare Resource Allocation Predictor using Machine Learning to improve the efficiency of healthcare resource management. Hospitals often face challenges such as increasing patient admissions, limited ICU beds, staff shortages, and inefficient inventory management. The proposed system utilizes machine learning techniques to analyze healthcare data and predict future resource requirements, including patient admissions, ICU occupancy, staffing needs, and medical inventory demand. The study discusses the application of predictive analytics, artificial intelligence, and healthcare data processing to support proactive decision-making in hospitals. Various machine learning approaches, including Linear Regression, Random Forest, XGBoost, and LSTM Neural Networks, are explored for healthcare forecasting and resource optimization. The proposed framework aims to reduce resource shortages, improve patient care quality, enhance emergency preparedness, and optimize operational costs. This work contributes to the development of intelligent, data-driven healthcare management systems and highlights the potential of machine learning technologies in modern healthcare infrastructure.","author":[{"family":"Chouhan","given":"Anshika"},{"family":"Khan","given":"Khushboo"},{"family":"Jharbade","given":"Kiran"},{"family":"Patel","given":"Radha"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20672289","URL":"https://doi.org/10.5281/zenodo.20672289","source":"datacite"},{"id":"doi:10.5281/zenodo.20672290","type":"article-journal","title":"Healthcare Resource Allocation Predictor: A Machine Learning Approach for Optimizing Healthcare Resource Distribution","abstract":"This paper presents a Healthcare Resource Allocation Predictor using Machine Learning to improve the efficiency of healthcare resource management. Hospitals often face challenges such as increasing patient admissions, limited ICU beds, staff shortages, and inefficient inventory management. The proposed system utilizes machine learning techniques to analyze healthcare data and predict future resource requirements, including patient admissions, ICU occupancy, staffing needs, and medical inventory demand. The study discusses the application of predictive analytics, artificial intelligence, and healthcare data processing to support proactive decision-making in hospitals. Various machine learning approaches, including Linear Regression, Random Forest, XGBoost, and LSTM Neural Networks, are explored for healthcare forecasting and resource optimization. The proposed framework aims to reduce resource shortages, improve patient care quality, enhance emergency preparedness, and optimize operational costs. This work contributes to the development of intelligent, data-driven healthcare management systems and highlights the potential of machine learning technologies in modern healthcare infrastructure.","author":[{"family":"Chouhan","given":"Anshika"},{"family":"Khan","given":"Khushboo"},{"family":"Jharbade","given":"Kiran"},{"family":"Patel","given":"Radha"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20672290","URL":"https://doi.org/10.5281/zenodo.20672290","source":"datacite"},{"id":"doi:10.26187/deakin.33286674","type":"article-journal","title":"Use of generative artificial intelligence (AI) in psychiatry and mental health care: A systematic review","abstract":"Objectives: Tools based on generative artificial intelligence (AI) such as ChatGPT have the potential to transform modern society, including the field of medicine. Due to the prominent role of language in psychiatry, e.g., for diagnostic assessment and psychotherapy, these tools may be particularly useful within this medical field. Therefore, the aim of this study was to systematically review the literature on generative AI applications in psychiatry and mental health. Methods: We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The search was conducted across three databases, and the resulting articles were screened independently by two researchers. The content, themes, and findings of the articles were qualitatively assessed. Results: The search and screening process resulted in the inclusion of 40 studies. The median year of publication was 2023. The themes covered in the articles were mainly mental health and well-being in general - with less emphasis on specific mental disorders (substance use disorder being the most prevalent). The majority of studies were conducted as prompt experiments, with the remaining studies comprising surveys, pilot studies, and case reports. Most studies focused on models that generate language, ChatGPT in particular. Conclusions: Generative AI in psychiatry and mental health is a nascent but quickly expanding field. The literature mainly focuses on applications of ChatGPT, and finds that generative AI performs well, but notes that it is limited by significant safety and ethical concerns. Future research should strive to enhance transparency of methods, use experimental designs, ensure clinical relevance, and involve users/patients in the design phase.","author":[{"family":"Kolding","given":"S"},{"family":"Lundin","given":"Robert"},{"family":"Hansen","given":"L"},{"family":"Ostergaard","given":"Sd"}],"issued":{"date-parts":[[2026]]},"DOI":"10.26187/deakin.33286674","URL":"https://doi.org/10.26187/deakin.33286674","source":"datacite"},{"id":"doi:10.48550/arxiv.2608.18036","type":"manuscript","title":"Harnessing Magnitude-Only and Complex Measurements for Improved Dynamic MRI Reconstruction with Learned Priors","abstract":"MRI reconstruction methods for undersampled k-space data naturally utilize complex-valued measurements. Parallel developments in sparse phase retrieval have shown that magnitude-only measurements may provide complementary information for signal recovery. However, their use in MRI reconstruction remains largely unexplored, due to lack of practical settings where informative magnitude measurements can be obtained without additional scan time. In this work, we investigate the use of auxiliary k-space magnitude information for accelerated steady-state dynamic MRI reconstruction, and demonstrate strong consistency of k-space magnitudes across time-frames. Building on this observation, we propose $\\mathbb{C}+\\text{Mag}$, a magnitude-informed physics-driven deep learning reconstruction method. The proposed method employs an ADMM-based unrolling framework with a novel magnitude-aware data-fidelity formulation, where quadratically smoothed optimization and momentum-based updates are introduced to address the non-differentiability and non-convexity of the magnitude constraints. Experiments on retrospectively undersampled cine MRI and phase-contrast flow MRI datasets, as well as prospectively undersampled real-time cine MRI acquisitions, demonstrate improved artifact suppression, sharper anatomical recovery, and better preservation of phase information compared to conventional PD-DL methods, which is further supported through blinded expert reader evaluations.","author":[{"family":"Saberi","given":"Mahdi"},{"family":"Alçalar","given":"Yaşar"},{"family":"Gülle","given":"Merve"},{"family":"Shenoy","given":"Chetan"},{"family":"Akçakaya","given":"Mehmet"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2608.18036","URL":"https://doi.org/10.48550/arxiv.2608.18036","source":"datacite"},{"id":"doi:10.5281/zenodo.22004205","type":"article-journal","title":"IA COMO HERRAMIENTA DE AUTODIAGNÓSTICO Y AUTOMEDICACIÓN EN PERÚ 2025","abstract":"Introducción: La irrupción de la inteligencia artificial (IA) generativa ha transformado la búsqueda de información de salud. En contextos de alta informalidad como la Provincia Constitucional del Callao, Perú, esta herramienta podría estar sustituyendo la consulta médica y exacerbando la automedicación. Objetivo: Analizar el comportamiento de usuarios que emplearon IA para el autodiagnóstico y la experiencia de farmacéuticos frente a solicitudes derivadas de estos diagnósticos. Métodos: Estudio cuantitativo, descriptivo-correlacional y de corte transversal. Se evaluaron 402 usuarios y 235 farmacéuticos mediante dos instrumentos validados (EUIA-AD y EFDIA). Se realizaron análisis factoriales exploratorios, regresión logística binaria y Análisis de Perfiles Latentes (LPA). Resultados: El 68.4% de los usuarios carecía de seguro de salud y el 62.7% se automedicó guiado por la IA, destacando un consumo de antibióticos sin receta del 28.1%. Los modelos logísticos no fueron significativos. El LPA identificó dos tipologías de usuarios (\"Adoptantes Creyentes\" y \"Escépticos Ocasionales\"), ambas con tasas de automedicación superiores al 61% (p = .830). En los farmacéuticos, el 87.2% percibió un vacío regulatorio y el 43% dispensó bajo presión, sin importar su nivel de proactividad (p = .638). Conclusiones: La automedicación en el Callao es una constante estructural que no es predicha por la confianza en la IA. Las herramientas algorítmicas se acoplan a un ecosistema sanitario fragmentado, trasladando el riesgo al mostrador. Se sugiere la necesidad que la Dirección General de Insumos, Drogas y Medicamentos (DIGEMID) actualice la normativa de dispensación para proteger al farmacéutico comunitario frente a los diagnósticos generados por IA. Palabras clave: Inteligencia artificial; automedicación; farmacia comunitaria; regulación sanitaria; estudio cuantitativo. Abstract Introduction: The emergence of generative artificial intelligence (AI) has transformed health information-seeking behaviors. In highly informal contexts such as the Constitutional Province of Callao, Peru, this tool could be replacing medical consultations and exacerbating self-medication. Objective: To analyze the behavior of users who employed AI for self-diagnosis and the experience of pharmacists facing requests derived from these diagnoses. Methods: Quantitative, descriptive-correlational, and cross-sectional study. 402 users and 235 pharmacists were evaluated using two validated instruments (EUIA-AD and EFDIA). Exploratory factor analyses, binary logistic regression, and Latent Profile Analysis (LPA) were performed. Results: 68.4% of users lacked health insurance, and 62.7% self-medicated guided by AI, with 28.1% consuming antibiotics without a prescription. Logistic models were not significant. LPA identified two user typologies (\"Believing Adopters\" and \"Occasional Skeptics\"), both showing self-medication rates above 61% (p = .830). Among pharmacists, 87.2% perceived a regulatory void and 43% dispensed under pressure, regardless of their proactivity level (p = .638). Conclusions: Self-medication in Callao is a structural constant not predicted by trust in AI. Algorithmic tools attach to a fragmented healthcare ecosystem, shifting risk to the pharmacy counter. It is suggested that Peru's Directorate General of Medicines, Supplies and Drugs (DIGEMID) should update dispensing regulations to protect community pharmacists against AI-generated diagnoses. Keywords: Artificial intelligence; self-medication; community pharmacy; health regulation; quantitative methods. La correspondencia concerniente a este artículo debe dirigirse a Edgar S., Inciso Mendo, Docente Medicina Humana. Universidad Norbert Wiener. er7219677@gmail.com. Wilmer A., Alania Yauri. Docente Investigador. Tecnólogo Médico en Laboratorio Clínico y Anatomía Patológica Universidad Tecnológica del Perú.investigador.peru.2020@gmail.com. Edilberto, Pérez Torres. Docente Departamento de Pediatría. Universidad Nac","author":[{"family":"Inciso Mendo","given":"Edgar"},{"family":"Alania Yauri","given":"Wilmer"},{"family":"Valero Quispe","given":"Javier"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22004205","URL":"https://doi.org/10.5281/zenodo.22004205","source":"datacite"},{"id":"doi:10.5281/zenodo.22004206","type":"article-journal","title":"IA COMO HERRAMIENTA DE AUTODIAGNÓSTICO Y AUTOMEDICACIÓN EN PERÚ 2025","abstract":"Introducción: La irrupción de la inteligencia artificial (IA) generativa ha transformado la búsqueda de información de salud. En contextos de alta informalidad como la Provincia Constitucional del Callao, Perú, esta herramienta podría estar sustituyendo la consulta médica y exacerbando la automedicación. Objetivo: Analizar el comportamiento de usuarios que emplearon IA para el autodiagnóstico y la experiencia de farmacéuticos frente a solicitudes derivadas de estos diagnósticos. Métodos: Estudio cuantitativo, descriptivo-correlacional y de corte transversal. Se evaluaron 402 usuarios y 235 farmacéuticos mediante dos instrumentos validados (EUIA-AD y EFDIA). Se realizaron análisis factoriales exploratorios, regresión logística binaria y Análisis de Perfiles Latentes (LPA). Resultados: El 68.4% de los usuarios carecía de seguro de salud y el 62.7% se automedicó guiado por la IA, destacando un consumo de antibióticos sin receta del 28.1%. Los modelos logísticos no fueron significativos. El LPA identificó dos tipologías de usuarios (\"Adoptantes Creyentes\" y \"Escépticos Ocasionales\"), ambas con tasas de automedicación superiores al 61% (p = .830). En los farmacéuticos, el 87.2% percibió un vacío regulatorio y el 43% dispensó bajo presión, sin importar su nivel de proactividad (p = .638). Conclusiones: La automedicación en el Callao es una constante estructural que no es predicha por la confianza en la IA. Las herramientas algorítmicas se acoplan a un ecosistema sanitario fragmentado, trasladando el riesgo al mostrador. Se sugiere la necesidad que la Dirección General de Insumos, Drogas y Medicamentos (DIGEMID) actualice la normativa de dispensación para proteger al farmacéutico comunitario frente a los diagnósticos generados por IA. Palabras clave: Inteligencia artificial; automedicación; farmacia comunitaria; regulación sanitaria; estudio cuantitativo. Abstract Introduction: The emergence of generative artificial intelligence (AI) has transformed health information-seeking behaviors. In highly informal contexts such as the Constitutional Province of Callao, Peru, this tool could be replacing medical consultations and exacerbating self-medication. Objective: To analyze the behavior of users who employed AI for self-diagnosis and the experience of pharmacists facing requests derived from these diagnoses. Methods: Quantitative, descriptive-correlational, and cross-sectional study. 402 users and 235 pharmacists were evaluated using two validated instruments (EUIA-AD and EFDIA). Exploratory factor analyses, binary logistic regression, and Latent Profile Analysis (LPA) were performed. Results: 68.4% of users lacked health insurance, and 62.7% self-medicated guided by AI, with 28.1% consuming antibiotics without a prescription. Logistic models were not significant. LPA identified two user typologies (\"Believing Adopters\" and \"Occasional Skeptics\"), both showing self-medication rates above 61% (p = .830). Among pharmacists, 87.2% perceived a regulatory void and 43% dispensed under pressure, regardless of their proactivity level (p = .638). Conclusions: Self-medication in Callao is a structural constant not predicted by trust in AI. Algorithmic tools attach to a fragmented healthcare ecosystem, shifting risk to the pharmacy counter. It is suggested that Peru's Directorate General of Medicines, Supplies and Drugs (DIGEMID) should update dispensing regulations to protect community pharmacists against AI-generated diagnoses. Keywords: Artificial intelligence; self-medication; community pharmacy; health regulation; quantitative methods. La correspondencia concerniente a este artículo debe dirigirse a Edgar S., Inciso Mendo, Docente Medicina Humana. Universidad Norbert Wiener. er7219677@gmail.com. Wilmer A., Alania Yauri. Docente Investigador. Tecnólogo Médico en Laboratorio Clínico y Anatomía Patológica Universidad Tecnológica del Perú.investigador.peru.2020@gmail.com. Edilberto, Pérez Torres. Docente Departamento de Pediatría. Universidad Nac","author":[{"family":"Inciso Mendo","given":"Edgar"},{"family":"Alania Yauri","given":"Wilmer"},{"family":"Valero Quispe","given":"Javier"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22004206","URL":"https://doi.org/10.5281/zenodo.22004206","source":"datacite"},{"id":"doi:10.17632/wndbd5r26y.5","type":"article-journal","title":"A Primary Chest X-ray Dataset of Normal Bangladesh","abstract":"📌 Steps to Reproduce This dataset contains a collection of primary chest X-ray images acquired from Epic Chittagong, Bangladesh. The dataset is designed for the study and development of deep learning and machine learning models for pneumonia detection and classification. This dataset contains 3,355 primary chest X-ray images collected from Epic Chittagong, Bangladesh, categorized into two classes: (1) Normal (2) Pneumonia 📊 Dataset Composition --------------------------------- Training Data : =&gt; Normal: 321 images =&gt; Pneumonia: 321 images =&gt; Total Training Samples: 642 Testing Data : ------------------- Normal: 1,363 images =&gt; Pneumonia: 1,350 images =&gt; Total Testing Samples: 2,713 👉 Grand Total: 3,355 X-ray images 📂 Folder Structure : ------------------------- /Chest_Xray_EpicChittagong_Dataset/ ├── train/ │ ├── Normal/ │ └── Pneumonia/ ├── test/ │ ├── Normal/ │ └── Pneumonia/ 📷 Image Details : ------------------------ Format: JPEG / PNG Modality: Chest X-ray (CXR) Color: Grayscale Source: Epic Chittagong, Bangladesh 2025 Status: Primary dataset (raw and unprocessed) 🧪 Applications : --------------------- =&gt; Pneumonia vs. Normal chest X-ray classification =&gt; Deep learning model training (CNN, transfer learning) =&gt; Benchmarking medical imaging algorithms =&gt; Computer-aided diagnosis (CAD) =&gt; Radiology research and teaching 📬 Contact : ------------------ For questions or collaboration Email: hiramdirfanulkabir@gmail.com 🎓 Department of Computer Science and Engineering 🏛️ Institutions : -------------------- Epic Chittagong, Bangladesh National Institute of Textile Engineering and Research University of Dhaka 📚 Categories : ---------------------- Computer Science, Radiology, Health Sciences, Artificial Intelligence, Computer Vision, Medical Imaging, Pneumonia, Chest X-ray, Deep Learning, Machine Learning","author":[{"family":"Hira","given":"Md"},{"family":"Bithee","given":"Mst"},{"family":"Ahmed","given":"Shafee"},{"family":"Akter","given":"Laboni"},{"family":"Anonna","given":"Mst"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/wndbd5r26y.5","URL":"https://doi.org/10.17632/wndbd5r26y.5","source":"datacite"},{"id":"doi:10.17632/wndbd5r26y","type":"article-journal","title":"A Primary Chest X-ray Dataset of Normal Bangladesh","abstract":"📌 Steps to Reproduce This dataset contains a collection of primary chest X-ray images acquired from Epic Chittagong, Bangladesh. The dataset is designed for the study and development of deep learning and machine learning models for pneumonia detection and classification. This dataset contains 3,355 primary chest X-ray images collected from Epic Chittagong, Bangladesh, categorized into two classes: (1) Normal (2) Pneumonia 📊 Dataset Composition --------------------------------- Training Data : =&gt; Normal: 321 images =&gt; Pneumonia: 321 images =&gt; Total Training Samples: 642 Testing Data : ------------------- Normal: 1,363 images =&gt; Pneumonia: 1,350 images =&gt; Total Testing Samples: 2,713 👉 Grand Total: 3,355 X-ray images 📂 Folder Structure : ------------------------- /Chest_Xray_EpicChittagong_Dataset/ ├── train/ │ ├── Normal/ │ └── Pneumonia/ ├── test/ │ ├── Normal/ │ └── Pneumonia/ 📷 Image Details : ------------------------ Format: JPEG / PNG Modality: Chest X-ray (CXR) Color: Grayscale Source: Epic Chittagong, Bangladesh 2025 Status: Primary dataset (raw and unprocessed) 🧪 Applications : --------------------- =&gt; Pneumonia vs. Normal chest X-ray classification =&gt; Deep learning model training (CNN, transfer learning) =&gt; Benchmarking medical imaging algorithms =&gt; Computer-aided diagnosis (CAD) =&gt; Radiology research and teaching 📬 Contact : ------------------ For questions or collaboration Email: hiramdirfanulkabir@gmail.com 🎓 Department of Computer Science and Engineering 🏛️ Institutions : -------------------- Epic Chittagong, Bangladesh National Institute of Textile Engineering and Research University of Dhaka 📚 Categories : ---------------------- Computer Science, Radiology, Health Sciences, Artificial Intelligence, Computer Vision, Medical Imaging, Pneumonia, Chest X-ray, Deep Learning, Machine Learning","author":[{"family":"Hira","given":"Md"},{"family":"Bithee","given":"Mst"},{"family":"Ahmed","given":"Shafee"},{"family":"Akter","given":"Laboni"},{"family":"Anonna","given":"Mst"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/wndbd5r26y","URL":"https://doi.org/10.17632/wndbd5r26y","source":"datacite"},{"id":"oa:W4406186727","type":"article-journal","title":"Artificial Intelligence for Patient Safety and Surgical Education in Neurosurgery","abstract":"Neurosurgery has evolved alongside technological innovations; however, these advances have also introduced greater complexity into clinical practice. Neurosurgery remains a demanding and high-risk field that requires a broad range of skills. Artificial intelligence (AI) has immense potential in neurosurgery given its ability to rapidly analyze large volumes of clinical data generated in modern clinical environments. An expanding body of literature has demonstrated that AI enhances various aspects of neurosurgery, including diagnostics, prognostication, decision-making, data management, education, and clinical studies. AI applications are expected to reduce medical errors and costs, broaden healthcare accessibility, and ultimately boost patient safety and surgical education. Nevertheless, AI application in neurosurgery remains practically limited because of several challenges, such as the diversity and volume of clinical training data collection, concerns regarding data quality, algorithmic bias, transparency (explainability and interpretability), ethical issues, and regulatory implications. To comprehensively discuss the potential benefits, future directions, and limitations of AI in neurosurgery, this review examined recent studies on AI technology and its applications in this field, focusing on intraoperative decision support and surgical education.","author":[{"family":"Sugiyama","given":"Taku"},{"family":"Sugimori","given":"Hiroyuki"},{"family":"Tang","given":"Minghui"},{"family":"Fujimura","given":"Miki"}],"issued":{"date-parts":[[2025]]},"DOI":"10.31662/jmaj.2024-0141","URL":"https://doi.org/10.31662/jmaj.2024-0141","source":"openalex"},{"id":"oa:W4411426688","type":"article-journal","title":"An In-depth overview of artificial intelligence (AI) tool utilization across diverse phases of organ transplantation","abstract":"Artificial Intelligence (AI) offers a revolutionary approach to improve decision-making in medicine through the use of advanced computational tools. Its ability to analyze large and complex datasets enables a thorough evaluation of multiple factors, leading to a deeper understanding of medical procedures. Numerous studies have demonstrated that AI has made significant advancements in areas such as organ allocation, donor-recipient matching, and immunosuppression protocols in organ transplantation. The transplantation process consists of three key stages: pre-transplant evaluation, the surgical procedure, and post-transplant management. AI can enhance all three stages by analyzing and integrating data from histopathological reports, lab results, radiological features, and patient demographics to aid in matching donors and recipients. Additionally, AI supports robotic-assisted surgery and optimizes post-transplant regimens while evaluating complications. Various researches have utilized machine learning (ML) to predict medication bioavailability immediately after transplantation and assess the risk of post-transplant complications based on factors like genetic phenotypes, age, gender, and body mass index. This review aims to gather information on AI applications across various stages of organ transplantation and elaborate the strategies and tools relevant to these processes.","author":[{"family":"Arjmandmazidi","given":"Shiva"},{"family":"Heidari","given":"Hamid"},{"family":"Ghasemnejad","given":"Tohid"},{"family":"Mori","given":"Zeinab"},{"family":"Molavi","given":"Leila"},{"family":"Meraji","given":"Amir"},{"family":"Kaghazchi","given":"Shadi"},{"family":"Aghdam","given":"Elnaz"},{"family":"Montazersaheb","given":"Soheila"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12967-025-06488-1","URL":"https://doi.org/10.1186/s12967-025-06488-1","source":"openalex"},{"id":"oa:W4406919111","type":"article-journal","title":"Artificial intelligence methods applied to longitudinal data from electronic health records for prediction of cancer: a scoping review","abstract":"BACKGROUND: Early detection and diagnosis of cancer are vital to improving outcomes for patients. Artificial intelligence (AI) models have shown promise in the early detection and diagnosis of cancer, but there is limited evidence on methods that fully exploit the longitudinal data stored within electronic health records (EHRs). This review aims to summarise methods currently utilised for prediction of cancer from longitudinal data and provides recommendations on how such models should be developed. METHODS: The review was conducted following PRISMA-ScR guidance. Six databases (MEDLINE, EMBASE, Web of Science, IEEE Xplore, PubMed and SCOPUS) were searched for relevant records published before 2/2/2024. Search terms related to the concepts \"artificial intelligence\", \"prediction\", \"health records\", \"longitudinal\", and \"cancer\". Data were extracted relating to several areas of the articles: (1) publication details, (2) study characteristics, (3) input data, (4) model characteristics, (4) reproducibility, and (5) quality assessment using the PROBAST tool. Models were evaluated against a framework for terminology relating to reporting of cancer detection and risk prediction models. RESULTS: Of 653 records screened, 33 were included in the review; 10 predicted risk of cancer, 18 performed either cancer detection or early detection, 4 predicted recurrence, and 1 predicted metastasis. The most common cancers predicted in the studies were colorectal (n = 9) and pancreatic cancer (n = 9). 16 studies used feature engineering to represent temporal data, with the most common features representing trends. 18 used deep learning models which take a direct sequential input, most commonly recurrent neural networks, but also including convolutional neural networks and transformers. Prediction windows and lead times varied greatly between studies, even for models predicting the same cancer. High risk of bias was found in 90% of the studies. This risk was often introduced due to inappropriate study design (n = 26) and sample size (n = 26). CONCLUSION: This review highlights the breadth of approaches to cancer prediction from longitudinal data. We identify areas where reporting of methods could be improved, particularly regarding where in a patients' trajectory the model is applied. The review shows opportunities for further work, including comparison of these approaches and their applications in other cancers.","author":[{"family":"Moglia","given":"Victoria"},{"family":"Johnson","given":"Owen"},{"family":"Cook","given":"GE"},{"family":"Kamps","given":"Marc"},{"family":"Smith","given":"Lesley"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12874-025-02473-w","URL":"https://doi.org/10.1186/s12874-025-02473-w","source":"openalex"},{"id":"oa:W4407729171","type":"article-journal","title":"Advancing Medical Research Through Artificial Intelligence: Progressive and Transformative Strategies: A Literature Review","abstract":"Background and Aims: Artificial intelligence (AI) has become integral to medical research, impacting various aspects such as data analysis, writing assistance, and publishing. This paper explores the multifaceted influence of AI on the process of writing medical research papers, encompassing data analysis, ethical considerations, writing assistance, and publishing efficiency. Methods: The review was conducted following the PRISMA guidelines; a comprehensive search was performed in Scopus, PubMed, EMBASE, and MEDLINE databases for research publications on artificial intelligence in medical research published up to October 2023. Results: AI facilitates the writing process by generating drafts, offering grammar and style suggestions, and enhancing manuscript quality through advanced models like ChatGPT. Ethical concerns regarding content ownership and potential biases in AI-generated content underscore the need for collaborative efforts among researchers, publishers, and AI creators to establish ethical standards. Moreover, AI significantly influences data analysis in healthcare, optimizing outcomes and patient care, particularly in fields such as obstetrics and gynecology and pharmaceutical research. The application of AI in publishing, ranging from peer review to manuscript quality control and journal matching, underscores its potential to streamline and enhance the entire research and publication process. Overall, while AI presents substantial benefits, ongoing research, and ethical guidelines are essential for its responsible integration into the evolving landscape of medical research and publishing. Conclusion: The integration of AI in medical research has revolutionized efficiency and innovation, impacting data analysis, writing assistance, publishing, and others. While AI tools offer significant benefits, ethical considerations such as biases and content ownership must be addressed. Ongoing research and collaborative efforts are crucial to ensure responsible and transparent AI implementation in the dynamic landscape of medical research and publishing.","author":[{"family":"Alqudimat","given":"Ahmad"},{"family":"Fares","given":"Zainab"},{"family":"Elaarag","given":"Mai"},{"family":"Osman","given":"Maha"},{"family":"Alzoubi","given":"Raed"},{"family":"Aboumarzouk","given":"Omar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/hsr2.70200","URL":"https://doi.org/10.1002/hsr2.70200","source":"openalex"},{"id":"oa:W4407683030","type":"article-journal","title":"Digital Twin and Artificial Intelligence in Machining: A Bibliometric Analysis","abstract":"The past decade has witnessed an exodus toward smart and lean manufacturing methods. The trend includes integrating intelligent methods into sustainable manufacturing systems purposely to improve the machining efficiency, reduce waste and also optimize productivity. Manufacturing systems have seen transformations from conventional methods, leaning towards smart manufacturing in line with the industrial revolution 4.0. Since the manufacturing process encompasses a wide range of human development capacity, it is essential to analyze its developmental trends, thereby preparing us for future uncertainties. In this work, we have used a Bibliometric analysis technique to study the developmental trends relating to machining, digital twins and artificial intelligence techniques. The review comprises the current activities in relation to the development to this area. The article comprises a Bibliometric analysis of 464 articles that were acquired from the Web of Science database, with a search period until November 2024. The method of obtaining the data includes retrieval from the database, qualitative analysis and interpreting the data via visual representation. The raw data obtained were redrawn using the origin software, and their visual interpretations were represented using the VOSviewer software (VOSviewer_1.6.19). The results obtained indicate that the number of publications related to the searched keywords has remarkably increased since the year 2018, achieving a record maximum of over 80 articles in 2024. This is indicative of its increasing popularity. The analysis of the articles was conducted based on the author countries, journal types, journal names, institutions, article types, major and micro research areas. The findings from the analysis are meant to provide a bibliometric explanation of the developmental trends in machining systems towards achieving the IR 4.0 goals. Additionally, the results would be helpful to researchers and industrialists that intend to achieve optimum and sustainable machining using digital twin technologies.","author":[{"family":"Suleiman","given":"Dambatta"},{"family":"Li","given":"QX"},{"family":"Li","given":"Benkai"},{"family":"Zhang","given":"Yanbin"},{"family":"Zhang","given":"Bo"},{"family":"Liu","given":"Danyang"},{"family":"Zhang","given":"Wenqiang"},{"family":"Zhou","given":"Zhigang"},{"family":"Feng","given":"Yuewen"},{"family":"Bie","given":"Qingfeng"},{"family":"Yin","given":"Xianxin"},{"family":"Wang","given":"Lesan"},{"family":"Li","given":"Changhe"}],"issued":{"date-parts":[[2025]]},"DOI":"10.70322/ism.2025.10005","URL":"https://doi.org/10.70322/ism.2025.10005","source":"openalex"},{"id":"oa:W4410992552","type":"article-journal","title":"Artificial intelligence and academic integrity in nursing education: A mixed methods study on usage, perceptions, and institutional implications","abstract":"BACKGROUND: The rise of artificial intelligence (AI) use in higher education has generated substantial debate among academics and students, given the potential for students to engage in academic misconduct through the misuse of AI. Academics argue that AI poses a serious threat to the foundational development of nurses through the questionable integrity of AI-generated academic work and by undermining the development of critical thinking skills essential for professional practice. However, there is limited research on nursing students' integration of AI technologies in their studies. METHOD: This study utilised a convergent parallel mixed methods approach to develop a multiphase approach with convergent parallel techniques for the qualitative and quantitative phases. The quantitative method utilised a Qualtrics-powered online survey to engage 188 nursing students, exploring various domains related to AI use. In the qualitative phase, in-depth interviews with 13 purposively sampled students provided deeper insights. The qualitative data were analysed using an inductive thematic analysis approach, while the quantitative data were analysed using SPSS. RESULT: In the survey, 24 % of respondents reported using AI, ranging from moderate to extensive usage. In logistics regression analysis, hearing about AI (OR = 3.9; CI 1.07-10.2; p < 0.05), the belief that AI was useful in the studies (OR = 5.5; CI 1.7-17.3; p < 0.01), and the perception that learning to use AI is easy (OR = 3.4; CI 1.1-11.1; p < 0.05) predicted AI use. Qualitative findings revealed that all students used AI for various academic purposes. The 'fascinating', 'intelligent' and 'efficient' nature of AI in handling 'time-consuming' academic tasks motivated its use. However, concerns about breaching academic integrity and the value of achieving success through personal effort served as deterrents. CONCLUSION: The findings suggest that while AI's efficiency drives students to adopt it, they remain cautious about its ethical implications, leading to uncertainty in its application within academic practices. This highlights the critical need for institutional support and explicit guidelines on responsible AI integration in educational settings.","author":[{"family":"Zgambo","given":"Maggie"},{"family":"Costello","given":"Martina"},{"family":"Buhlmann","given":"Melanie"},{"family":"Maldon","given":"Justine"},{"family":"Anyango","given":"Edah"},{"family":"Adama","given":"Esther"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.nedt.2025.106796","URL":"https://doi.org/10.1016/j.nedt.2025.106796","source":"openalex"},{"id":"oa:W4412488540","type":"article-journal","title":"Artificial intelligence-enhanced electrocardiography to predict regurgitant valvular heart diseases: an international study","abstract":"BACKGROUND AND AIMS: Valvular heart disease (VHD) is a significant source of morbidity and mortality, though early intervention can improve outcomes. This study aims to develop artificial intelligence-enhanced electrocardiography (AI-ECG) models to diagnose and predict future moderate or severe regurgitant VHDs (rVHDs), including mitral regurgitation (MR), tricuspid regurgitation (TR), and aortic regurgitation (AR). METHODS: The AI-ECG models were developed in a data set of 988 618 ECG and transthoracic echocardiogram pairs from 400 882 patients from Zhongshan Hospital, Shanghai, China. The AI-ECG models used a residual convolutional neural network with a discrete-time survival loss function. External evaluation was performed in outpatients from a secondary care data set from Beth Israel Deaconess Medical Center, Boston, USA, consisting of 34 214 patients with linked echocardiography. RESULTS: In the internal test set, the AI-ECG models accurately predicted future significant MR [C-index 0.774, 95% confidence interval (CI) 0.753-0.792], AR (0.691, 95% CI 0.657-0.720), and TR (0.793, 95% CI 0.777-0.808). In age- and sex-adjusted Cox models, the highest risk quartile had a hazard ratio (HR) of 7.6 (95% CI 5.8-9.9, P < .0001) for risk of future significant MR, compared with the lowest risk quartile. For future AR and TR, the equivalent HRs were 3.8 (95% CI 2.7-5.5) and 9.9 (95% CI 7.5-13.0), respectively. These findings were confirmed in the transnational external test set. Imaging association analyses demonstrated AI-ECG predictions were associated with subclinical chamber remodelling. CONCLUSIONS: This study developed AI-ECG models to diagnose and predict rVHDs and validated the models in a transnational and ethnically distinct cohort. The AI-ECG models could be utilized to guide surveillance echocardiography in patients at risk of future rVHDs, to facilitate early detection and intervention.","author":[{"family":"Liang","given":"Yixiu"},{"family":"Sau","given":"Arunashis"},{"family":"Zeidaabadi","given":"Boroumand"},{"family":"Barker","given":"Joseph"},{"family":"Patlatzoglou","given":"Konstantinos"},{"family":"Pastika","given":"Libor"},{"family":"Sieliwończyk","given":"Ewa"},{"family":"Whinnett","given":"Zachary"},{"family":"Peters","given":"Nicholas"},{"family":"Yu","given":"Ziqing"},{"family":"Liu","given":"Xi"},{"family":"Wang","given":"Shuo"},{"family":"Lu","given":"Hongyang"},{"family":"Kramer","given":"Daniel"},{"family":"Waks","given":"Jonathan"},{"family":"Su","given":"Yangang"},{"family":"Ge","given":"Junbo"},{"family":"Ng","given":"Fu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/eurheartj/ehaf448","URL":"https://doi.org/10.1093/eurheartj/ehaf448","source":"openalex"},{"id":"oa:W4411903243","type":"article-journal","title":"Perceived worries in the adoption of artificial intelligence among nurses in neonatal intensive care units","abstract":"INTRODUCTION: Artificial Intelligence (AI) comprises computational algorithms designed to analyze data, learn patterns, and execute tasks traditionally requiring human cognition. These models can support public health initiatives, expedite clinical care, and improve diagnosis accuracy. Thus, artificial intelligence in healthcare sectors has the potential to enhance nursing care by assisting nurses with tasks like documentation, workflow improvement, and decision-making, while reducing workforce stress. This study, guided by the Technology Acceptance Model (TAM), assesses perceived worries regarding AI adoption among nurses in neonatal intensive care units (NICUs). METHODS: A cross-sectional quantitative design was employed using convenience sampling. Data were collected using the Worries of Applying AI in Healthcare Questionnaire (WAAI-HCQ) from 227 NICU nurses across nine hospitals in the West Bank (January 2-March 3, 2025). SPSS version 26 was used for analysis. RESULTS: Participants demonstrated intermediate levels of AI awareness (M = 2.7, SD = 0.5) and limited prior AI experience (M = 2.3, SD = 0.5). Total AI-related worries were moderate (M = 3.2, SD = 0.9), with healthcare provider-related concerns being highest. Multiple linear regression (R² = 0.846) identified education level (B = 0.074, p = 0.026), AI awareness (B = 2.006, p < 0.001), and AI experience (B = -0.959, p < 0.001) as significant predictors, explaining 84.6% of the variance in AI-related worries. CONCLUSIONS: NICU nurses in Palestine exhibit moderate AI awareness and concerns, highlighting the need for targeted education and training to address knowledge gaps and facilitate AI integration. This study contributes new knowledge specifically for conflict-affected, resource-constrained NICU settings, where AI implementation faces unique challenges. CLINICAL TRIAL NUMBER: Not applicable.","author":[{"family":"Ayed","given":"Ahmad"},{"family":"Batran","given":"Ahmad"},{"family":"Aqtam","given":"Ibrahim"},{"family":"Malak","given":"Malakeh"},{"family":"Ejheisheh","given":"Moath"},{"family":"Farajallah","given":"Mosaab"},{"family":"Farraj","given":"Lamees"},{"family":"Alkhatib","given":"Sanaa"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12912-025-03318-z","URL":"https://doi.org/10.1186/s12912-025-03318-z","source":"openalex"},{"id":"oa:W7125491435","type":"article-journal","title":"Artificial Intelligence Drives Advances in Multi-Omics Analysis and Precision Medicine for Sepsis","abstract":"Sepsis is a life-threatening syndrome characterized by marked clinical heterogeneity and complex host-pathogen interactions. Although traditional mechanistic studies have identified key molecular pathways, they remain insufficient to capture the highly dynamic, multifactorial, and systems-level nature of this condition. The advent of high-throughput omics technologies-particularly integrative multi-omics approaches encompassing genomics, transcriptomics, proteomics, and metabolomics-has profoundly reshaped sepsis research by enabling comprehensive profiling of molecular perturbations across biological layers. However, the unprecedented scale, dimensionality, and heterogeneity of multi-omics datasets exceed the analytical capacity of conventional statistical methods, necessitating more advanced computational strategies to derive biologically meaningful and clinically actionable insights. In this context, artificial intelligence (AI) has emerged as a powerful paradigm for decoding the complexity of sepsis. By leveraging machine learning and deep learning algorithms, AI can efficiently process ultra-high-dimensional and heterogeneous multi-omics data, uncover latent molecular patterns, and integrate multilayered biological information into unified predictive frameworks. These capabilities have driven substantial advances in early sepsis detection, molecular subtyping, prognosis prediction, and therapeutic target identification, thereby narrowing the gap between molecular mechanisms and clinical application. As a result, the convergence of AI and multi-omics is redefining sepsis research, shifting the field from descriptive analyses toward predictive, mechanistic, and precision-oriented medicine. Despite these advances, the clinical translation of AI-driven multi-omics approaches in sepsis remains constrained by several challenges, including limited data availability, cohort heterogeneity, restricted interpretability and causal inference, high computational demands, difficulties in integrating static molecular profiles with dynamic clinical data, ethical and governance concerns, and limited generalizability across populations and platforms. Addressing these barriers will require the establishment of standardized, multicenter datasets, the development of explainable and robust AI frameworks, and sustained interdisciplinary collaboration between computational scientists and clinicians. Through these efforts, AI-enabled multi-omics research may progress toward reproducible, interpretable, and equitable clinical implementation. Ultimately, the synergy between artificial intelligence and multi-omics heralds a new era of intelligent discovery and precision medicine in sepsis, with the potential to transform both research paradigms and bedside practice.","author":[{"family":"Shen","given":"Youxie"},{"family":"Zhang","given":"Peidong"},{"family":"Luo","given":"Jialiu"},{"family":"Chen","given":"Shunyao"},{"family":"Gu","given":"Shuaipeng"},{"family":"Lin","given":"Zhiqiang"},{"family":"Tang","given":"Zhaohui"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/biomedicines14020261","URL":"https://doi.org/10.3390/biomedicines14020261","source":"openalex"},{"id":"oa:W4411151151","type":"article-journal","title":"Evaluation of the Performance of Artificial Intelligence Based Chatbots in Providing First Aid Information on Dental Trauma According to the ToothSOS Application","abstract":"AIM: The aim of this study was to evaluate the performance of ChatGPT-4o and Gemini Advanced artificial intelligence-based chatbots (AI-based chatbots) in providing emergency intervention recommendations for dental trauma with intraoral photographs of patients diagnosed with traumatic dental injuries, and to assess their compatibility with emergency intervention recommendations in the ToothSOS application. MATERIAL AND METHODS: In this study, 80 intraoral photographs obtained from patients presenting with dental trauma were uploaded to two different AI-based chatbots (ChatGPT-4o and Gemini Advanced) and the responses generated by these systems were evaluated by four paediatric dentists. The evaluators scored the responses with a Modified Global Quality Score (GQS), referring to the English instructions of the ToothSOS application. In order to analyse the reliability of the responses, a total of three evaluation sessions were conducted 1 week apart. RESULTS: The ChatGPT-4o performed better when all injury types were considered together (p = 0.012). It was found that the ChatGPT-4o performed much better in complicated crown fracture cases (p = 0.004) and the Gemini Advanced chatbot performed much better in critical dental injuries such as avulsion (p < 0.001). CONCLUSIONS: AI-based chatbots can be a helpful tool in the assessment of dental trauma. However, further development and expert validation are needed to improve their accuracy and consistency, especially in complex cases. Incorporating the International Association of Dental Traumatology (IADT) guidelines into the databases of these systems could improve the reliability of their recommendations. In addition, given the widespread use of AI-based chatbots in many fields, particularly health, they could contribute to public health by supporting access to accurate information.","author":[{"family":"Çege","given":"Ecem"},{"family":"Cömert","given":"Hamide"},{"family":"Akal","given":"Neşe"},{"family":"Ölmez","given":"Ayşegül"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/edt.13078","URL":"https://doi.org/10.1111/edt.13078","source":"openalex"},{"id":"oa:W4412685719","type":"article-journal","title":"Enhancing emotional intelligence in medical education: a systematic review of interventions","abstract":"Introduction: Emotional intelligence (EI) is a crucial competency for medical professionals, facilitating effective interpersonal relationships between physicians and patients. The ability to evaluate, regulate, and apply emotional understanding plays a significant role in fostering empathy, communication, and stress management. This systematic review aimed to determine the impact of various interventions on medical students' EI development, academic performance, and overall patient care. Methods: A comprehensive literature search was conducted for studies published from 2021 until 2024, inclusion criteria focused on studies on medical students, employed validated EI assessment tools, and utilized appropriate research designs. Results: 44 articles met the inclusion criteria. The Joanna Briggs Institute (JBI) Critical Appraisal Checklist was applied to assess the quality of included studies. Although a meta-analysis was initially planned, substantial heterogeneity across the studies limited the pooling of quantitative data. Using an inductive coding approach, eight major themes were identified: Narrative and Storytelling Interventions, Reflective Practices and Writing, Communication Skills Training, Emotional Intelligence Enhancement, Experiential Learning and Patient Exposure, Stress, Burnout, and Coping Interventions, Assessment Tools and Structural Interventions, and Personalized Interventions and Diversity Considerations. These themes were subsequently mapped onto Daniel Goleman's model of Emotional Intelligence to provide a structured theoretical framework. Discussion: The findings of this review highlight that various interventions hold promise in enhancing EI among medical students, leading to improvements in personal well-being, communication skills, and professional development. The most effective approach appears to be a multifaceted, longitudinal integration of EI-focused strategies throughout medical training, incorporating repeated practice, guided reflection, faculty mentorship, and structured debriefing. The broader implications extend to improved doctor-patient relationships, reduced burnout, and enhanced clinical decision-making, ultimately contributing to higher patient satisfaction and more compassionate healthcare delivery. Future research should focus on refining intervention methodologies and assessing their long-term impact on medical education and practice.","author":[{"family":"Maity","given":"Sabyasachi"},{"family":"Filippis","given":"Samantha"},{"family":"Aldanese","given":"Alexander"},{"family":"Mcculloch","given":"Melissa"},{"family":"Sandor","given":"Alexis"},{"family":"Cajigas","given":"Jan"},{"family":"Antoniadis","given":"Yiorgos"},{"family":"Rochester","given":"Te"},{"family":"Carter","given":"Lauren"},{"family":"Preisig","given":"Alexander"},{"family":"Kobeissi","given":"Julia"},{"family":"Nayak","given":"Narendra"},{"family":"Mendoza","given":"Jaime"},{"family":"Nauhria","given":"Samal"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fmed.2025.1587090","URL":"https://doi.org/10.3389/fmed.2025.1587090","source":"openalex"},{"id":"oa:W4406334317","type":"article-journal","title":"The Epistemic Cost of Opacity: How the Use of Artificial Intelligence Undermines the Knowledge of Medical Doctors in High-Stakes Contexts","abstract":"Abstract Artificial intelligent (AI) systems used in medicine are often very reliable and accurate, but at the price of their being increasingly opaque. This raises the question whether a system’s opacity undermines the ability of medical doctors to acquire knowledge on the basis of its outputs. We investigate this question by focusing on a case in which a patient’s risk of recurring breast cancer is predicted by an opaque AI system. We argue that, given the system’s opacity, as well as the possibility of malfunctioning AI systems, practitioners’ inability to check the correctness of their outputs, and the high stakes of such cases, the knowledge of medical practitioners is indeed undermined. They are lucky to form true beliefs based on the AI systems’ outputs, and knowledge is incompatible with luck. We supplement this claim with a specific version of the safety condition on knowledge, Safety*. We argue that, relative to the perspective of the medical doctor in our example case, his relevant beliefs could easily be false, and this despite his evidence that the AI system functions reliably. Assuming that Safety* is necessary for knowledge, the practitioner therefore doesn’t know. We address three objections to our proposal before turning to practical suggestions for improving the epistemic situation of medical doctors.","author":[{"family":"Schmidt","given":"Eva"},{"family":"Putora","given":"Paul"},{"family":"Fijten","given":"Rianne"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s13347-024-00834-9","URL":"https://doi.org/10.1007/s13347-024-00834-9","source":"openalex"},{"id":"oa:W4414805087","type":"article-journal","title":"The role of artificial intelligence for early warning systems: Status, applicability, guardrails, and ways forward","abstract":"Artificial intelligence (AI) is gaining momentum in earth sciences as a tool to analyze complex natural hazards and their impacts. Such analyses are critical for effective Early Warning Systems (EWSs), which is aiming to generate timely and actionable risk information to protect sectors, systems, and people. Despite advancements in AI, its role in EWS remains underexplored across the four pillars of the Early Warning for All (EW4All) framework; risk knowledge, forecasting, warning dissemination and communication and response preparedness. This study draws on a systematic literature review to assess AI methods utilized in the context of EWS, examines their challenges and opportunities and discusses guiding questions for responsible use. Our study highlights key gaps across knowledge, application and policy. Moreover, we call for coordinated efforts to develop responsible AI frameworks that enhance EWS while ensuring they remain inclusive, accessible, and people-centred that ultimately supports the goal of EW4All by 2027.","author":[{"family":"Tiggeloven","given":"Timothy"},{"family":"Pfeiffer","given":"Samira"},{"family":"Matanó","given":"Alessia"},{"family":"Homberg","given":"Marc"},{"family":"Thalheimer","given":"Lisa"},{"family":"Reichstein","given":"Markus"},{"family":"Torresan","given":"Silvia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.isci.2025.113689","URL":"https://doi.org/10.1016/j.isci.2025.113689","source":"openalex"},{"id":"oa:W7138063471","type":"article-journal","title":"Artificial Intelligence-Driven Development and Characterization of Nanomedicine","abstract":"Abstract Nanomedicine has enabled major advances in targeted therapeutics by improving drug bioavailability, precision delivery, and safety profiles. However, the rational design and reproducible synthesis of nanoparticles with tightly controlled physicochemical attributes such as size, morphology, and surface characteristics remain significant challenges due to the complex, nonlinear interplay of formulation and process parameters. Artificial intelligence (AI) and machine learning (ML) have emerged as powerful tools to address these limitations by enabling data-driven optimization, predictive modeling, and automated analysis across nanoparticle synthesis and characterization workflows. Recent advances demonstrate that AI-based models can accurately predict nanoparticle properties, optimize synthesis conditions, interpret high-dimensional characterization data, and forecast biological performance, thereby reducing experimental burden and accelerating translation. This review critically examines current AI and ML strategies applied to nanoparticle synthesis, optimization of key physicochemical attributes, characterization, and biological evaluation for nanomedicine applications. Emphasis is placed on comparative model performance, integration of experimental and computational pipelines, and emerging challenges related to data quality, interpretability, and generalizability. Collectively, this work highlights the transformative potential of AI-enabled nanotechnology while outlining key directions required for its reliable clinical translation. Graphical Abstract Automated Synthesis Platforms: Integrating AI and ML for Next-Generation Nanomaterials","author":[{"family":"Okafor","given":"Nnamdi"},{"family":"Igbokwe","given":"Nkeiruka"},{"family":"Onohuean","given":"Hope"},{"family":"Faya","given":"Mbuso"},{"family":"Choonara","given":"Yahya"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1007/s12668-026-02476-x","URL":"https://doi.org/10.1007/s12668-026-02476-x","source":"openalex"},{"id":"oa:W4407361583","type":"article-journal","title":"Artificial intelligence support improves diagnosis accuracy in anterior segment eye diseases","abstract":"CorneAI, a deep learning model designed for diagnosing cataracts and corneal diseases, was assessed for its impact on ophthalmologists' diagnostic accuracy. In the study, 40 ophthalmologists (20 specialists and 20 residents) classified 100 images, including iPhone 13 Pro photos (50 images) and diffuser slit-lamp photos (50 images), into nine categories (normal condition, infectious keratitis, immunological keratitis, corneal scar, corneal deposit, bullous keratopathy, ocular surface tumor, cataract/intraocular lens opacity, and primary angle-closure glaucoma). The iPhone and slit-lamp images represented the same cases. After initially answering without CorneAI, the same ophthalmologists responded to the same cases with CorneAI 2-4 weeks later. With CorneAI's support, the overall accuracy of ophthalmologists increased significantly from 79.2 to 88.8% (P < 0.001). Specialists' accuracy rose from 82.8 to 90.0%, and residents' from 75.6 to 86.2% (P < 0.001). Smartphone image accuracy improved from 78.7 to 85.5% and slit-lamp image accuracy from 81.2 to 90.6% (both, P < 0.001). In this study, CorneAI's own accuracy was 86%, but its support enhanced ophthalmologists' accuracy beyond the CorneAI's baseline. This study demonstrated that CorneAI, despite being trained on diffuser slit-lamp images, effectively improved diagnostic accuracy, even with smartphone images.","author":[{"family":"Maehara","given":"Hiroki"},{"family":"Ueno","given":"Yuta"},{"family":"Yamaguchi","given":"Takefumi"},{"family":"Kitaguchi","given":"Yoshiyuki"},{"family":"Miyazaki","given":"Dai"},{"family":"Nejima","given":"Ryohei"},{"family":"Inomata","given":"Takenori"},{"family":"Kato","given":"Naoko"},{"family":"Chikama","given":"Tai"},{"family":"Ominato","given":"Jun"},{"family":"Yunoki","given":"Tatsuya"},{"family":"Tsubota","given":"Kinya"},{"family":"Oda","given":"Masahiro"},{"family":"Suzutani","given":"Manabu"},{"family":"Sekiryu","given":"Tetsuju"},{"family":"Oshika","given":"Tetsuro"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-89768-6","URL":"https://doi.org/10.1038/s41598-025-89768-6","source":"openalex"},{"id":"oa:W4413312755","type":"article-journal","title":"Multimodal Large Language Models in Medical Imaging: Current State and Future Directions","abstract":"Multimodal large language models (MLLMs) are emerging as powerful tools in medicine, particularly in radiology, with the potential to serve as trusted artificial intelligence (AI) partners for clinicians. In radiology, these models integrate large language models (LLMs) with diverse multimodal data sources by combining clinical information and text with radiologic images of various modalities, ranging from 2D chest X-rays to 3D CT/MRI. Methods for achieving this multimodal integration are rapidly evolving, and the high performance of freely available LLMs may further accelerate MLLM development. Current applications of MLLMs now span automatic generation of preliminary radiology report, visual question answering, and interactive diagnostic support. Despite these promising capabilities, several significant challenges hinder widespread clinical adoption. MLLMs require access to large-scale, high-quality multimodal datasets, which are scarce in the medical domain. Risks of hallucinated findings, lack of transparency in decision-making processes, and high computational demands further complicate implementation. This review summarizes the current capabilities and limitations of MLLMs in medicine-particularly in radiology-and outlines key directions for future research. Critical areas include incorporating region-grounded reasoning to link model outputs to specific image regions, developing robust foundation models pre-trained on large-scale medical datasets, and establishing strategies for the safe and effective integration of MLLMs into clinical practice.","author":[{"family":"Nam","given":"Yoojin"},{"family":"Kim","given":"D"},{"family":"Kyung","given":"Sunggu"},{"family":"Seo","given":"Jinyoung"},{"family":"Song","given":"Jeong"},{"family":"Kwon","given":"Jimin"},{"family":"Kim","given":"Jihyun"},{"family":"Jo","given":"Wooyoung"},{"family":"Park","given":"HJ"},{"family":"Sung","given":"Jimin"},{"family":"Park","given":"Sangah"},{"family":"Kwon","given":"Heeyeon"},{"family":"Kwon","given":"TH"},{"family":"Kim","given":"Kanghyun"},{"family":"Kim","given":"Namkug"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3348/kjr.2025.0599","URL":"https://doi.org/10.3348/kjr.2025.0599","source":"openalex"},{"id":"oa:W4406371336","type":"article-journal","title":"Large Language Models lack essential metacognition for reliable medical reasoning","abstract":"Large Language Models have demonstrated expert-level accuracy on medical board examinations, suggesting potential for clinical decision support systems. However, their metacognitive abilities, crucial for medical decision-making, remain largely unexplored. To address this gap, we developed MetaMedQA, a benchmark incorporating confidence scores and metacognitive tasks into multiple-choice medical questions. We evaluated twelve models on dimensions including confidence-based accuracy, missing answer recall, and unknown recall. Despite high accuracy on multiple-choice questions, our study revealed significant metacognitive deficiencies across all tested models. Models consistently failed to recognize their knowledge limitations and provided confident answers even when correct options were absent. In this work, we show that current models exhibit a critical disconnect between perceived and actual capabilities in medical reasoning, posing significant risks in clinical settings. Our findings emphasize the need for more robust evaluation frameworks that incorporate metacognitive abilities, essential for developing reliable Large Language Model enhanced clinical decision support systems. Large Language Models demonstrate expert-level accuracy in medical exams, supporting their potential inclusion in healthcare settings. Here, authors reveal that their metacognitive abilities are underexplored, showing significant gaps in recognizing knowledge limitations, difficulties in modulating their confidence, and challenges in identifying when a problem cannot be answered due to insufficient information.","author":[{"family":"Griot","given":"Maxime"},{"family":"Hemptinne","given":"Coralie"},{"family":"Vanderdonckt","given":"Jean"},{"family":"Yüksel","given":"Demet"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41467-024-55628-6","URL":"https://doi.org/10.1038/s41467-024-55628-6","source":"openalex"},{"id":"oa:W4412505013","type":"article-journal","title":"The synergy of neuromarketing and artificial intelligence: A comprehensive literature review in the last decade","abstract":"Abstract This paper conducts a systematic literature analysis on \"artificial intelligence, ethical artificial intelligence, neuromarketing, consumer neuroscience, neuroethics, and neurotechnology.\" This study followed the systematic literature review methodology to select and extract the relevant documents from the Scopus database (2013–2023). The findings revealed the valuable transformative impact of integrating artificial intelligence (AI) into neuromarketing (NM) and consumer neuroscience (Cons-Neuro), redefining the understanding and influence of consumer behavior. Emotion, attention, and memory have become vital in NM and AI studies. AI algorithms analyze vast neural and physiological datasets, offering marketers insights into the emotional impact of campaigns, granular insights into consumer focus, and optimizing content for maximum impact. Furthermore, memory plays a vital role in increasing brand recall and fostering lasting relationships. In addition, integrating brain-computer interfaces (BCI) into consumer neuroscience provides direct insights, with AI interpreting BCI data for real-time adjustments. The synergy of NM and AI offers insights into consumer behavior's cognitive and emotional aspects. While enabling targeted campaigns and improved customer experiences, this integration raises ethical concerns necessitating transparency and responsible neural data use. This paper offers valuable insights into the intersection of AI and NM, exploring innovative applications and ethical considerations in these evolving fields.","author":[{"family":"Alsharif","given":"Ahmed"},{"family":"Wang","given":"Junhai"},{"family":"Isa","given":"Salmi"},{"family":"Salleh","given":"Nor"},{"family":"Dawas","given":"Husam"},{"family":"Alsharif","given":"Mohammed"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s43093-025-00591-x","URL":"https://doi.org/10.1186/s43093-025-00591-x","source":"openalex"},{"id":"oa:W4410934331","type":"article-journal","title":"Artificial Intelligence in Glioblastoma—Transforming Diagnosis and Treatment","abstract":"Glioblastoma (GBM) is the most aggressive and common primary brain malignancy in adults, characterized by poor prognosis and treatment resistance. Despite advancements in treatment options, the median survival is roughly 15 months, underlining the need for novel and effective treatments. Artificial intelligence (AI) has emerged as a transformative technology in healthcare, offering outstanding capabilities in data analysis, pattern recognition, and helping in decision-making. This review explores the current and potential role of AI in GBM care, focusing on its applications in diagnosis, treatment planning, prognostication, and drug discovery. AI-based algorithms have demonstrated promising potential in enhancing diagnostics through imaging analysis, radiomics, and tumor segmentation. These technologies could enable non-invasive molecular profiling and early detection of GBM. In treatment planning, AI could improve approaches by optimizing surgical resection, radiotherapy regimen, and chemotherapy protocols. Furthermore, machine learning models can integrate multimodal data to develop personalized treatments. They can also enhance prognostication by predicting survival, recurrence, and treatment responses, helping clinicians to make more informed decisions. AI is also revolutionizing pharmacotherapy by identifying novel molecular targets and optimizing combination therapies. Despite notable progress, challenges persist. Limited data quality and quantity, algorithm interpretability, integration problems, and ethical considerations, remain significant challenges to clinical implementation. This review emphasizes the need for continued research and interdisciplinary collaboration to overcome many barriers and realize the transformative potential of AI in GBM care.","author":[{"family":"Rončević","given":"Alen"},{"family":"Koruga","given":"Nenad"},{"family":"Koruga","given":"Anamarija"},{"family":"Rončević","given":"Robert"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s41016-025-00399-2","URL":"https://doi.org/10.1186/s41016-025-00399-2","source":"openalex"},{"id":"oa:W4409571599","type":"article-journal","title":"Mammographic classification of interval breast cancers and artificial intelligence performance","abstract":"BACKGROUND: European studies suggest that artificial intelligence (AI) can reduce interval breast cancers. Research on interval breast cancer classification and AI's effectiveness in the United States, however, particularly using digital breast tomosynthesis and annual screening, is limited. We aimed to mammographically classify interval breast cancers and assess AI performance using a 12-month screening interval. METHODS: From digital mammography and digital breast tomosynthesis screening mammograms acquired between 2010 and 2019 at a US tertiary-care academic center, we identified interval breast cancers diagnosed less than 12 months after a negative mammogram. At least 3 breast radiologists retrospectively classified interval breast cancers as missed-reading error, minimal signs-actionable, minimal signs-nonactionable, true interval, occult, or missed-technical error. A deep-learning AI tool assigned risk scores ranging from 1 to 10 to the negative index screening mammograms, with scores of 8 or higher considered \"flagged.\" Statistical analysis evaluated associations among interval breast cancer types and AI exam scores, AI markings, and patient and tumor characteristics. RESULTS: From 184 935 screening mammograms (65% digital mammography, 35% digital breast tomosynthesis), we identified 148 interval breast cancers in 148 women (mean [SD] age = 61 [12] years). Of these, 26% were minimal signs-actionable, 24% were occult, 22% were minimal signs-nonactionable, 17% were missed-reading error, 6% were true interval, and 5% were missed-technical error (P < .001). AI scored 131 mammograms (17 errors excluded); it most frequently flagged exams with missed-reading error (90%), minimal signs-actionable (89%), and minimal signs-nonactionable (72%) (P = .02). AI localized mammographically visible types more accurately (35%-68%) than nonvisible types (0%-50%; P = .02). CONCLUSION: AI more frequently flagged and accurately localized interval breast cancer types that were mammographically visible at screening (missed or minimal signs) compared with true interval or occult cancers.","author":[{"family":"Yu","given":"Tiffany"},{"family":"Hoyt","given":"Anne"},{"family":"Joines","given":"Melissa"},{"family":"Fischer","given":"Cheryce"},{"family":"Yaghmai","given":"Nazanin"},{"family":"Chalfant","given":"James"},{"family":"Chow","given":"Lucy"},{"family":"Mortazavi","given":"Shabnam"},{"family":"Sears","given":"Christopher"},{"family":"Sayre","given":"James"},{"family":"Elmore","given":"Joann"},{"family":"Hsu","given":"William"},{"family":"Milch","given":"Hannah"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/jnci/djaf103","URL":"https://doi.org/10.1093/jnci/djaf103","source":"openalex"},{"id":"oa:W4407289259","type":"article-journal","title":"Artificial Intelligence in Multi-Disease Medical Diagnostics: An Integrative Approach","abstract":"With advanced algorithms, artificial intelligence (AI) has revolutionized the medical diagnostic field where diseases can be predicted simultaneously. The integrative nature of this approach is novel because it can better encompass the complexity of comorbid conditions that are so common in patients; thus, addressing them in a more holistic diagnostic tone that is lacking in previous works. In this study, the investigation of the usage of AI models for simultaneously diagnosing diseases like diabetes, cardiovascular conditions, and neurological disorders is done. Therefore, based on AI techniques i.e. artificial neural networks (ANNs) and ensemble learning methods, a multi-disease diagnostic framework was developed to achieve this. A variety of features, related to each condition, were captured from multi-modal datasets including imaging, laboratory test results, and patient histories. The system was developed to manage the big flow of aggregated data and offer detailed diagnostic views of many diseases. Sensitivity, specificity, and overall diagnostic accuracy were used to evaluate the framework's performance. The results showed that the AI framework has high diagnostic accuracy for all targeted conditions an overall sensitivity of 93% and a specificity of 91%. Importantly, the combination of multi-modal data proved to substantially improve the system’s ability to identify and distinguish comorbid conditions. It makes the importance of using various data sources to benefit from the reliability and comprehensiveness of AI diagnostics obvious. Overall, AI-driven multi-disease diagnostic systems provide great promise for the role of delivering potentially transformative clinical healthcare workflow improvements, reducing errors, and improving patient outcomes. These frameworks will need to be scaled and tested in various healthcare settings and also across more varied diseases to help make medical diagnosis more available and effective.","author":[{"family":"Sultana","given":"Nigar"},{"family":"Saimon","given":"Shariar"},{"family":"Islam","given":"Ishraq"},{"family":"Abir","given":"Shake"},{"family":"Hossain","given":"Md"},{"family":"Shiam","given":"Sarder"},{"family":"Khan","given":"Nazrul"}],"issued":{"date-parts":[[2025]]},"DOI":"10.32996/jcsts.2025.7.1.12","URL":"https://doi.org/10.32996/jcsts.2025.7.1.12","source":"openalex"},{"id":"oa:W4406690454","type":"article-journal","title":"MR-linac: role of artificial intelligence and automation","abstract":"The integration of artificial intelligence (AI) into radiotherapy has advanced significantly during the past 5 years, especially in terms of automating key processes like organ at risk delineation and treatment planning. These innovations have enhanced consistency, accuracy, and efficiency in clinical practice. Magnetic resonance (MR)-guided linear accelerators (MR-linacs) have greatly improved treatment accuracy and real-time plan adaptation, particularly for tumors near radiosensitive organs. Despite these improvements, MR-guided radiotherapy (MRgRT) remains labor intensive and time consuming, highlighting the need for AI to streamline workflows and support rapid decision-making. Synthetic CTs from MR images and automated contouring and treatment planning will reduce manual processes, thus optimizing treatment times and expanding access to MR-linac technology. AI-driven quality assurance will ensure patient safety by predicting machine errors and validating treatment delivery. Advances in intrafractional motion management will increase the accuracy of treatment, and the integration of imaging biomarkers for outcome prediction and early toxicity assessment will enable more precise and effective treatment strategies.","author":[{"family":"Psoroulas","given":"S"},{"family":"Paunoiu","given":"Alina"},{"family":"Corradini","given":"Stefanie"},{"family":"Hörnerrieber","given":"Juliane"},{"family":"Tanadinilang","given":"Stephanie"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00066-024-02358-9","URL":"https://doi.org/10.1007/s00066-024-02358-9","source":"openalex"},{"id":"oa:W4406072657","type":"article-journal","title":"Artificial intelligence assisted real-time recognition of intra-abdominal metastasis during laparoscopic gastric cancer surgery","abstract":"Laparoscopic exploration (LE) is crucial for diagnosing intra-abdominal metastasis (IAM) in advanced gastric cancer (GC). However, overlooking single, tiny, and occult IAM lesions during LE can severely affect the treatment and prognosis due to surgeons' visual misinterpretations. To address this, we developed the artificial intelligence laparoscopic exploration system (AiLES) to recognize IAM lesions with various metastatic extents and locations. The AiLES was developed based on a dataset consisting of 5111 frames from 100 videos, using 4130 frames for model development and 981 frames for evaluation. The AiLES achieved a Dice score of 0.76 and a recognition speed of 11 frames per second, demonstrating robust performance in different metastatic extents (0.74-0.76) and locations (0.63-0.90). Furthermore, AiLES performed comparably to novice surgeons in IAM recognition and excelled in recognizing tiny and occult lesions. Our results demonstrate that the implementation of AiLES could enhance accurate tumor staging and assist individualized treatment decisions.","author":[{"family":"Chen","given":"Hao"},{"family":"Gou","given":"Longfei"},{"family":"Fang","given":"Zhiwen"},{"family":"Dou","given":"Qi"},{"family":"Chen","given":"Haobin"},{"family":"Chen","given":"Chang"},{"family":"Qiu","given":"Yuqing"},{"family":"Zhang","given":"Jinglin"},{"family":"Ning","given":"Chenglin"},{"family":"Hu","given":"Yanfeng"},{"family":"Deng","given":"Haijun"},{"family":"Yu","given":"Jiang"},{"family":"Li","given":"Guoxin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-024-01372-6","URL":"https://doi.org/10.1038/s41746-024-01372-6","source":"openalex"},{"id":"oa:W4412915069","type":"article-journal","title":"A Review of Artificial Intelligence and Deep Learning Approaches for Resource Management in Smart Buildings","abstract":"This comprehensive review maps the fast-evolving landscape in which artificial intelligence (AI) and deep-learning (DL) techniques converge with the Internet of Things (IoT) to manage energy, comfort, and sustainability across smart environments. A PRISMA-guided search of four databases retrieved 1358 records; after applying inclusion criteria, 143 peer-reviewed studies published between January 2019 and April 2025 were analyzed. This review shows that AI-driven controllers—especially deep-reinforcement-learning agents—deliver median energy savings of 18–35% for HVAC and other major loads, consistently outperforming rule-based and model-predictive baselines. The evidence further reveals a rapid diversification of methods: graph-neural-network models now capture spatial interdependencies in dense sensor grids, federated-learning pilots address data-privacy constraints, and early integrations of large language models hint at natural-language analytics and control interfaces for heterogeneous IoT devices. Yet large-scale deployment remains hindered by fragmented and proprietary datasets, unresolved privacy and cybersecurity risks associated with continuous IoT telemetry, the growing carbon and compute footprints of ever-larger models, and poor interoperability among legacy equipment and modern edge nodes. The authors of researches therefore converges on several priorities: open, high-fidelity benchmarks that marry multivariate IoT sensor data with standardized metadata and occupant feedback; energy-aware, edge-optimized architectures that lower latency and power draw; privacy-centric learning frameworks that satisfy tightening regulations; hybrid physics-informed and explainable models that shorten commissioning time; and digital-twin platforms enriched by language-model reasoning to translate raw telemetry into actionable insights for facility managers and end users. Addressing these gaps will be pivotal to transforming isolated pilots into ubiquitous, trustworthy, and human-centered IoT ecosystems capable of delivering measurable gains in efficiency, resilience, and occupant wellbeing at scale.","author":[{"family":"Amangeldy","given":"Bibars"},{"family":"Imankulov","given":"Timur"},{"family":"Tasmurzayev","given":"Nurdaulet"},{"family":"Dikhanbayeva","given":"Gulmira"},{"family":"Nurakhov","given":"Yedil"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/buildings15152631","URL":"https://doi.org/10.3390/buildings15152631","source":"openalex"},{"id":"oa:W4411283308","type":"article-journal","title":"Artificial Intelligence Empowering Dynamic Spectrum Access in Advanced Wireless Communications: A Comprehensive Overview","abstract":"This review paper examines the integration of artificial intelligence (AI) in wireless communication, focusing on cognitive radio (CR), spectrum sensing, and dynamic spectrum access (DSA). As the demand for spectrum continues to rise with the expansion of mobile users and connected devices, cognitive radio networks (CRNs), leveraging AI-driven spectrum sensing and dynamic access, provide a promising solution to improve spectrum utilization. The paper reviews various deep learning (DL)-based spectrum-sensing methods, highlighting their advantages and challenges. It also explores the use of multi-agent reinforcement learning (MARL) for distributed DSA networks, where agents autonomously optimize power allocation (PA) to minimize interference and enhance quality of service. Additionally, the paper discusses the role of machine learning (ML) in predicting spectrum requirements, which is crucial for efficient frequency management in the fifth generation (5G) networks and beyond. Case studies show how ML can help self-optimize networks, reducing energy consumption while improving performance. The review also introduces the potential of generative AI (GenAI) for demand-planning and network optimization, enhancing spectrum efficiency and energy conservation in wireless networks (WNs). Finally, the paper highlights future research directions, including improving AI-driven network resilience, refining predictive models, and addressing ethical considerations. Overall, AI is poised to transform wireless communication, offering innovative solutions for spectrum management (SM), security, and network performance.","author":[{"family":"Gbengailori","given":"Abiodun"},{"family":"Imoize","given":"Agbotiname"},{"family":"Noor","given":"Kinzah"},{"family":"Adebolu-Ololade","given":"Paul"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ai6060126","URL":"https://doi.org/10.3390/ai6060126","source":"openalex"},{"id":"oa:W4405986581","type":"article-journal","title":"Influence of next-generation artificial intelligence on headache research, diagnosis and treatment: the junior editorial board members’ vision – part 2","abstract":"Part 2 explores the transformative potential of artificial intelligence (AI) in addressing the complexities of headache disorders through innovative approaches, including digital twin models, wearable healthcare technologies and biosensors, and AI-driven drug discovery. Digital twins, as dynamic digital representations of patients, offer opportunities for personalized headache management by integrating diverse datasets such as neuroimaging, multiomics, and wearable sensor data to advance headache research, optimize treatment, and enable virtual trials. In addition, AI-driven wearable devices equipped with next-generation biosensors combined with multi-agent chatbots could enable real-time physiological and biochemical monitoring, diagnosing, facilitating early headache attack forecasting and prevention, disease tracking, and personalized interventions. Furthermore, AI-driven advances in drug discovery leverage machine learning and generative AI to accelerate the identification of novel therapeutic targets and optimize treatment strategies for migraine and other headache disorders. Despite these advances, challenges such as data standardization, model explainability, and ethical considerations remain pivotal. Collaborative efforts between clinicians, biomedical and biotechnological engineers, AI scientists, legal representatives and bioethics experts are essential to overcoming these barriers and unlocking AI's full potential in transforming headache research and healthcare. This is a call to action in proposing novel frameworks for integrating AI-based technologies into headache care.","author":[{"family":"Petrušić","given":"Igor"},{"family":"Chiang","given":"Chia‐chun"},{"family":"Garcíaazorín","given":"David"},{"family":"Ha","given":"Woo‐seok"},{"family":"Ornello","given":"Raffaele"},{"family":"Pellesi","given":"Lanfranco"},{"family":"Rubiobeltrán","given":"Eloísa"},{"family":"Ruscheweyh","given":"Ruth"},{"family":"Waliszewskaprosół","given":"Marta"},{"family":"Wells-Gatnik","given":"William"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s10194-024-01944-7","URL":"https://doi.org/10.1186/s10194-024-01944-7","source":"openalex"},{"id":"oa:W4406288592","type":"article-journal","title":"Diagnostic accuracy in coronary CT angiography analysis: artificial intelligence versus human assessment","abstract":"BACKGROUND: Visual assessment of coronary CT angiography (CCTA) is time-consuming, influenced by reader experience and prone to interobserver variability. This study evaluated a novel algorithm for coronary stenosis quantification (atherosclerosis imaging quantitative CT, AI-QCT). METHODS: The study included 208 patients with suspected coronary artery disease (CAD) undergoing CCTA in Perfusion Imaging and CT Coronary Angiography With Invasive Coronary Angiography-1. AI-QCT and blinded readers assessed coronary artery stenosis following the Coronary Artery Disease Reporting and Data System consensus. Accuracy of AI-QCT was compared with a level 3 and two level 2 clinical readers against an invasive quantitative coronary angiography (QCA) reference standard (≥50% stenosis) in an area under the curve (AUC) analysis, evaluated per-patient and per-vessel and stratified by plaque volume. RESULTS: Among 208 patients with a mean age of 58±9 years and 37% women, AI-QCT demonstrated superior concordance with QCA compared with clinical CCTA assessments. For the detection of obstructive stenosis (≥50%), AI-QCT achieved an AUC of 0.91 on a per-patient level, outperforming level 3 (AUC 0.77; p<0.002) and level 2 readers (AUC 0.79; p<0.001 and AUC 0.76; p<0.001). The advantage of AI-QCT was most prominent in those with above median plaque volume. At the per-vessel level, AI-QCT achieved an AUC of 0.86, similar to level 3 (AUC 0.82; p=0.098) stenosis, but superior to level 2 readers (both AUC 0.69; p<0.001). CONCLUSIONS: AI-QCT demonstrated superior agreement with invasive QCA compared to clinical CCTA assessments, particularly compared to level 2 readers in those with extensive CAD. Integrating AI-QCT into routine clinical practice holds promise for improving the accuracy of stenosis quantification through CCTA.","author":[{"family":"Bernardo","given":"Rachel"},{"family":"Nurmohamed","given":"Nick"},{"family":"Bom","given":"Michiel"},{"family":"Jukema","given":"Ruurt"},{"family":"Winter","given":"Ruben"},{"family":"Sprengers","given":"Ralf"},{"family":"Stroes","given":"Erik"},{"family":"Min","given":"James"},{"family":"Earls","given":"James"},{"family":"Danad","given":"Ibrahim"},{"family":"Choi","given":"Andrew"},{"family":"Knaapen","given":"Paul"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/openhrt-2024-003115","URL":"https://doi.org/10.1136/openhrt-2024-003115","source":"openalex"},{"id":"oa:W4409700773","type":"article-journal","title":"The Rise of Transformers – Redefining the Landscape of Artificial Intelligence","abstract":"The 2017 paper 'Attention Is All You Need' by Vaswani et al. marked a major paradigm shift in AI. Rather than clinging to tired methods like recurrence or convolutions, its Transformer design boldly flipped the script melding self-attention into a fresh take that not only remade natural language processing but also trickled over into computer vision, robotics, quirky multi-modal setups, and more [1]. At its very core lies a refreshingly simple idea self-attention, which lets models dynamically figure out which bits of the input deserve extra focus instead of being bound by strict, linear routines. This clever tweak kicked the old, rigid rules to the curb and opened up levels of parallel processing that we hadn’t seen before. In many cases, this shift not only ramped up the training speed dramatically but also proved essential in our data-swamped world where speed and nimbleness really do make all the difference.","author":[{"family":"Ladu","given":"Nyagong"},{"family":"Turyasingura","given":"Benson"},{"family":"Willbroad","given":"Byamukama"},{"family":"Atuhaire","given":"Abraham"}],"issued":{"date-parts":[[2025]]},"DOI":"10.58496/bjai/2025/007","URL":"https://doi.org/10.58496/bjai/2025/007","source":"openalex"},{"id":"oa:W4414610304","type":"article-journal","title":"Artificial Intelligence-Enhanced Liquid Biopsy and Radiomics in Early-Stage Lung Cancer Detection: A Precision Oncology Paradigm","abstract":"BACKGROUND: Lung cancer remains the leading cause of cancer-related mortality globally, largely due to delayed diagnosis in its early stages. While conventional diagnostic tools like low-dose CT and tissue biopsy are routinely used, they suffer from limitations including invasiveness, radiation exposure, cost, and limited sensitivity for early-stage detection. Liquid biopsy, a minimally invasive alternative that captures circulating tumor-derived biomarkers such as ctDNA, cfRNA, and exosomes from body fluids, offers promising diagnostic potential-yet its sensitivity in early disease remains suboptimal. Recent advances in Artificial Intelligence (AI) and radiomics are poised to bridge this gap. OBJECTIVE: This review aims to explore how AI, in combination with radiomics, enhances the diagnostic capabilities of liquid biopsy for early detection of lung cancer and facilitates personalized monitoring strategies. Content Overview: We begin by outlining the molecular heterogeneity of lung cancer, emphasizing the need for earlier, more accurate detection strategies. The discussion then transitions into liquid biopsy and its key analytes, followed by an in-depth overview of AI techniques-including machine learning (e.g., SVMs, Random Forest) and deep learning models (e.g., CNNs, RNNs, GANs)-that enable robust pattern recognition across multi-omics datasets. The role of radiomics, which quantitatively extracts spatial and morphological features from imaging modalities such as CT and PET, is explored in conjunction with AI to provide an integrative, multimodal approach. This convergence supports the broader vision of precision medicine by integrating omics data, imaging, and electronic health records. DISCUSSION: The synergy between AI, liquid biopsy, and radiomics signifies a shift from traditional diagnostics toward dynamic, patient-specific decision-making. Radiomics contributes spatial information, while AI improves pattern detection and predictive modeling. Despite these advancements, challenges remain-including data standardization, limited annotated datasets, the interpretability of deep learning models, and ethical considerations. A push toward rigorous validation and multimodal AI frameworks is necessary to facilitate clinical adoption. CONCLUSION: The integration of AI with liquid biopsy and radiomics holds transformative potential for early lung cancer detection. This non-invasive, scalable, and individualized diagnostic paradigm could significantly reduce lung cancer mortality through timely and targeted interventions. As technology and regulatory pathways mature, collaborative research is crucial to standardize methodologies and translate this innovation into routine clinical practice.","author":[{"family":"Cherukuri","given":"Swathi"},{"family":"Kaur","given":"Anmolpreet"},{"family":"Goyal","given":"Bipasha"},{"family":"Kukunoor","given":"Hanisha"},{"family":"Sahito","given":"Areesh"},{"family":"Sachdeva","given":"Pratyush"},{"family":"Yerrapragada","given":"Gayathri"},{"family":"Poonguzhali","given":"E"},{"family":"Shariff","given":"Mohammed"},{"family":"Natarajan","given":"TK"},{"family":"Janarthanan","given":"Jayarajasekaran"},{"family":"Richard","given":"Samuel"},{"family":"Venkatesaprasath","given":"Shakthidevi"},{"family":"Karuppiah","given":"Shiva"},{"family":"Iyer","given":"Vivek"},{"family":"Helgeson","given":"Scott"},{"family":"Arunachalam","given":"Shivaram"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/cancers17193165","URL":"https://doi.org/10.3390/cancers17193165","source":"openalex"},{"id":"oa:W4415250556","type":"article-journal","title":"Artificial Intelligence in the Management of Infectious Diseases in Older Adults: Diagnostic, Prognostic, and Therapeutic Applications","abstract":"Background: Older adults are highly vulnerable to infectious diseases due to immunosenescence, multimorbidity, and atypical presentations. Artificial intelligence (AI) offers promising opportunities to improve diagnosis, prognosis, treatment, and continuity of care in this population. This review summarizes current applications of AI in the management of infections in older adults across diagnostic, prognostic, therapeutic, and preventive domains. Methods: We conducted a narrative review of peer-reviewed studies retrieved from PubMed, Scopus, and Web of Science, focusing on AI-based tools for infection diagnosis, risk prediction, antimicrobial stewardship, prevention of healthcare-associated infections, and post-discharge care in individuals aged ≥65 years. Results: AI models, including machine learning, deep learning, and natural language processing techniques, have demonstrated high performance in detecting infections such as sepsis, pneumonia, and healthcare-associated infections (Area Under the Curve AUC up to 0.98). Prognostic algorithms integrating frailty and functional status enhance the prediction of mortality, complications, and readmission. AI-driven clinical decision support systems contribute to optimized antimicrobial therapy and timely interventions, while remote monitoring and telemedicine applications support safer hospital-to-home transitions and reduced 30-day readmissions. However, the implementation of these technologies is limited by the underrepresentation of frail older adults in training datasets, lack of real-world validation in geriatric settings, and the insufficient explainability of many models. Additional barriers include system interoperability issues and variable digital infrastructure, particularly in long-term care and community settings. Conclusions: AI has strong potential to support predictive and personalized infection management in older adults. Future research should focus on developing geriatric-specific, interpretable models, improving system integration, and fostering interdisciplinary collaboration to ensure safe and equitable implementation.","author":[{"family":"Pinto","given":"Antonio"},{"family":"Pennisi","given":"Flavia"},{"family":"Odelli","given":"Stefano"},{"family":"Ponti","given":"Emanuele"},{"family":"Veronese","given":"Nicola"},{"family":"Signorelli","given":"Carlo"},{"family":"Baldo","given":"Vincenzo"},{"family":"Gianfredi","given":"Vincenza"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/biomedicines13102525","URL":"https://doi.org/10.3390/biomedicines13102525","source":"openalex"},{"id":"oa:W4410915363","type":"article-journal","title":"A general framework for governing marketed AI/ML medical devices","abstract":"This project represents the first systematic assessment of the US Food and Drug Administration's postmarket surveillance of legally marketed artificial intelligence and machine learning based medical devices. We focus on the Manufacturer and User Facility Device Experience database-the FDA's central tool for tracking the safety of marketed AI/ML devices. In particular, we evaluate the data pertaining to adverse events associated with approximately 950 medical devices incorporating AI/ML functions for devices approved between 2010 through 2023, and we find that the existing system is insufficient for properly assessing the safety and effectiveness of AI/ML devices. In particular, we make three contributions: (1) characterize the adverse event reports for such devices, (2) examine the ways in which the existing FDA adverse reporting system for medical devices falls short, and (3) suggest changes FDA might consider in its approach to adverse event reporting for devices incorporating AI/ML functions.","author":[{"family":"Babic","given":"Boris"},{"family":"Cohen","given":"IG"},{"family":"Stern","given":"Ariel"},{"family":"Li","given":"Yiwen"},{"family":"Ouellet","given":"Melissa"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01717-9","URL":"https://doi.org/10.1038/s41746-025-01717-9","source":"openalex"},{"id":"oa:W4412097424","type":"article-journal","title":"Artificial Intelligence-Enabled Point-of-Care Echocardiography: Bringing Precision Imaging to the Bedside","abstract":"PURPOSE OF REVIEW: The integration of artificial intelligence (AI) with point-of-care ultrasound (POCUS) is transforming cardiovascular diagnostics by enhancing image acquisition, interpretation, and workflow efficiency. These advancements hold promise in expanding access to cardiovascular imaging in resource-limited settings and enabling early disease detection through screening applications. This review explores the opportunities and challenges of AI-enabled POCUS as it reshapes the landscape of cardiovascular imaging. RECENT FINDINGS: AI-enabled systems can reduce operator dependency, improve image quality, and support clinicians-both novice and experienced-in capturing diagnostically valuable images, ultimately promoting consistency across diverse clinical environments. However, widespread adoption faces significant challenges, including concerns around algorithm generalizability, bias, explainability, clinician trust, and data privacy. Addressing these issues through standardized development, ethical oversight, and clinician-AI collaboration will be critical to safe and effective implementation. Looking ahead, emerging innovations-such as autonomous scanning, real-time predictive analytics, tele-ultrasound, and patient-performed imaging-underscore the transformative potential of AI-enabled POCUS in reshaping cardiovascular care and advancing equitable healthcare delivery worldwide.","author":[{"family":"East","given":"Sasha"},{"family":"Wang","given":"Yanting"},{"family":"Yanamala","given":"Naveena"},{"family":"Maganti","given":"Kameswari"},{"family":"Sengupta","given":"Partho"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s11883-025-01316-9","URL":"https://doi.org/10.1007/s11883-025-01316-9","source":"openalex"},{"id":"oa:W4409252703","type":"article-journal","title":"Systematic literature review on the application of explainable artificial intelligence in palliative care studies","abstract":"BACKGROUND: As machine learning models become increasingly prevalent in palliative care, explainability has become a critical factor in their successful deployment in this sensitive field, where decisions can profoundly impact patient health and quality of life. To address these concerns, Explainable AI (XAI) aims to make complex AI models more understandable and trustworthy. OBJECTIVE: This study aims to assess the current state of machine learning models in palliative care, specifically focusing on their compliance with the principles of XAI. METHODS: A comprehensive literature search in four databases was conducted to identify articles on machine learning in palliative care studies published until May 2024, followed by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guideline. The Checklist for Assessment of Medical Artificial Intelligence was used to evaluate the quality of the studies. RESULTS: Mortality and survival prediction were the primary focus areas in 15 (54%) of the included 28 studies. Regarding data explainability, 20 studies (71%) documented their data preprocessing methods. However, a notable concern is that 45% of the studies did not address handling missing data. Across these studies, 74 machine learning algorithms were employed. Complex models, including Random Forest, Support Vector Machines, Gradient Boosting Machines, and Deep Neural Networks, were predominantly used (64%) due to their high predictive power, achieving AUC values between 0.82 and 0.96. Post-hoc explanation techniques were applied in only 11 studies, using seven different XAI techniques, focusing on global explanations to enhance understanding of model behavior. CONCLUSION: Given the critical role of AI-driven decisions in patient care, adopting XAI techniques is essential for fostering trust and usability. Although progress has been made, significant gaps persist. A main challenge remains the trade-off between model performance and interpretability, as highly accurate models often lack the transparency required to build trust in clinical settings. Additionally, complex models frequently provide inadequate explanations for their outputs, lack consistent documentation, and have limited XAI applications, reducing the interpretability of machine learning studies for clinicians and decision-makers.","author":[{"family":"Migiddorj","given":"Battushig"},{"family":"Batterham","given":"Marijka"},{"family":"Win","given":"Khin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.ijmedinf.2025.105914","URL":"https://doi.org/10.1016/j.ijmedinf.2025.105914","source":"openalex"},{"id":"oa:W4409339512","type":"article-journal","title":"Validity and accuracy of artificial intelligence-based dietary intake assessment methods: a systematic review","abstract":"Abstract One of the most significant challenges in research related to nutritional epidemiology is the achievement of high accuracy and validity of dietary data to establish an adequate link between dietary exposure and health outcomes. Recently, the emergence of artificial intelligence (AI) in various fields has filled this gap with advanced statistical models and techniques for nutrient and food analysis. We aimed to systematically review available evidence regarding the validity and accuracy of AI-based dietary intake assessment methods (AI-DIA). In accordance with PRISMA guidelines, an exhaustive search of the EMBASE, PubMed, Scopus and Web of Science databases was conducted to identify relevant publications from their inception to 1 December 2024. Thirteen studies that met the inclusion criteria were included in this analysis. Of the studies identified, 61·5 % were conducted in preclinical settings. Likewise, 46·2 % used AI techniques based on deep learning and 15·3 % on machine learning. Correlation coefficients of over 0·7 were reported in six articles concerning the estimation of calories between the AI and traditional assessment methods. Similarly, six studies obtained a correlation above 0·7 for macronutrients. In the case of micronutrients, four studies achieved the correlation mentioned above. A moderate risk of bias was observed in 61·5 % ( n 8) of the articles analysed, with confounding bias being the most frequently observed. AI-DIA methods are promising, reliable and valid alternatives for nutrient and food estimations. However, more research comparing different populations is needed, as well as larger sample sizes, to ensure the validity of the experimental designs.","author":[{"family":"Cofre","given":"Sebastián"},{"family":"Sánchez","given":"Camila"},{"family":"Quezada-Figueroa","given":"Gladys"},{"family":"López-Cortés","given":"Xaviera"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1017/s0007114525000522","URL":"https://doi.org/10.1017/s0007114525000522","source":"openalex"},{"id":"oa:W4409157497","type":"article-journal","title":"Retrieval augmented generation for 10 large language models and its generalizability in assessing medical fitness","abstract":"Large Language Models (LLMs) hold promise for medical applications but often lack domain-specific expertise. Retrieval Augmented Generation (RAG) enables customization by integrating specialized knowledge. This study assessed the accuracy, consistency, and safety of LLM-RAG models in determining surgical fitness and delivering preoperative instructions using 35 local and 23 international guidelines. Ten LLMs (e.g., GPT3.5, GPT4, GPT4o, Gemini, Llama2, and Llama3, Claude) were tested across 14 clinical scenarios. A total of 3234 responses were generated and compared to 448 human-generated answers. The GPT4 LLM-RAG model with international guidelines generated answers within 20 s and achieved the highest accuracy, which was significantly better than human-generated responses (96.4% vs. 86.6%, p = 0.016). Additionally, the model exhibited an absence of hallucinations and produced more consistent output than humans. This study underscores the potential of GPT-4-based LLM-RAG models to deliver highly accurate, efficient, and consistent preoperative assessments.","author":[{"family":"Ke","given":"Yu"},{"family":"Jin","given":"Liyuan"},{"family":"Elangovan","given":"Kabilan"},{"family":"Abdullah","given":"Hairil"},{"family":"Liu","given":"Nan"},{"family":"Sia","given":"Alex"},{"family":"Soh","given":"Chai"},{"family":"Tung","given":"Joshua"},{"family":"Ong","given":"Jasmine"},{"family":"Kuo","given":"Chang‐fu"},{"family":"Wu","given":"Shaochun"},{"family":"Kovacheva","given":"Vesela"},{"family":"Ting","given":"Daniel"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01519-z","URL":"https://doi.org/10.1038/s41746-025-01519-z","source":"openalex"},{"id":"oa:W4407183498","type":"article-journal","title":"Artificial intelligence–based clinical decision support in the emergency department: A scoping review","abstract":"OBJECTIVE: Artificial intelligence (AI)-based clinical decision support (CDS) has the potential to augment high-stakes clinical decisions in the emergency department (ED). However, its current usage and translation to implementation remains poorly understood. We asked: (1) What is the current landscape of AI-CDS for individual patient care in the ED? and (2) What phases of development have AI-CDS tools achieved? METHODS: We performed a scoping review of AI for prognostic, diagnostic, and treatment decisions regarding individual ED patient care. We searched five databases (MEDLINE, EMBASE, Cochrane Central, Scopus, Web of Science) and gray literature sources from January 1, 2010, to December 11, 2023. We adhered to guidelines from the Joanna Briggs Institute and PRISMA Extension for Scoping Reviews. We published our protocol on Open Science Framework (DOI 10.17605/OSF.IO/FDZ3Y). RESULTS: Of 5168 unique records identified, we selected 605 studies for inclusion. The majority (369, 61%) were published in 2021-2023. The studies ranged over a variety of clinical applications, patient populations, and AI model types. Prognostic outcomes were most commonly assessed (270, 44.6%), followed by diagnostic (193, 31.9%) and disposition (115, 19%). Most studies remained in the earliest phase of preclinical development (572, 94.5%) with few advancing to later phases (33, 5.5%). CONCLUSIONS: By thoroughly mapping the landscape of AI-CDS in the ED, we demonstrate a rapidly increasing volume of studies covering a breadth of clinical applications, yet few have achieved advanced phases of testing or implementation. A more granular understanding of the barriers and facilitators to implementing AI-CDS in the ED is needed.","author":[{"family":"Kareemi","given":"Hashim"},{"family":"Yadav","given":"Krishan"},{"family":"Price","given":"Courtney"},{"family":"Bobrovitz","given":"Niklas"},{"family":"Meehan","given":"Andrew"},{"family":"Li","given":"Henry"},{"family":"Goel","given":"Gautam"},{"family":"Masood","given":"Sameer"},{"family":"Grant","given":"Lars"},{"family":"Benyakov","given":"Maxim"},{"family":"Michalowski","given":"Wojtek"},{"family":"Vaillancourt","given":"Christian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/acem.15099","URL":"https://doi.org/10.1111/acem.15099","source":"openalex"},{"id":"oa:W4410043903","type":"article-journal","title":"The mediating digital literacy and the moderating role of academic support in the relationship between artificial intelligence usage and creative thinking in nursing students","abstract":"Artificial intelligence can be an important tool in developing creative thinking while supporting individualized learning and problem-solving skills in educational processes. However, the effect of artificial intelligence on creative thinking may differ depending on factors such as students’ digital literacy level and the academic support they receive. In this context, how artificial intelligence can encourage creative thinking in educational processes and how this process is shaped by digital literacy and academic support stands out as an issue to be investigated. This study artificial intelligence to determine the effect of artificial intelligence use on creative thinking skills in nursing students, the mediating role of digital literacy in this effect, and the situational mediating effect of academic support. This cross-sectional and descriptive correlational study was conducted with 426 nursing students from three universities in different regions of Turkey during the fall semester of 2024–2025. Simple random sampling method was used. Data were collected through demographic information form and four valid and reliable scales. Research findings show that using artificial intelligence substantially and significantly affects Creative Thinking (β = 0.70, p < 0.001). Digital literacy plays mediating role in this relationship, indirectly strengthening the effect of artificial intelligence use on creative thinking (β = 0.422, LLCI = 0.357, ULCI = 0.490). As a moderating factor of this relationship, the academic support creates a more limited effect at low levels of academic support (Effect = 0.224). At the same time, it maximizes this effect at high levels of support (Effect = 0.298). The effect of artificial intelligence use on Creative Thinking is significantly strengthened by the mediating role of digital literacy and the moderating role of academic support. These findings suggest that developing digital skills and strengthening academic support mechanisms are critical to integrating artificial intelligence technologies into educational processes. Artificial intelligence significantly increases the creative thinking tendencies of nursing students both directly and through digital literacy. The academic support is a critical factor that strengthens the development of creative thinking skills by regulating the effect of artificial intelligence usage on digital literacy. Integrating artificial intelligence applications into the curriculum and developing digital literacy skills in nursing education programs can potentially increase the creative thinking and problem-solving competencies of 21st-century health professionals.","author":[{"family":"Ağaoğlu","given":"Ferhat"},{"family":"Baş","given":"Murat"},{"family":"Tarsuslu","given":"Sinan"},{"family":"Ekinci","given":"Lokman"},{"family":"Ağaoğlu","given":"Nihat"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12912-025-03128-3","URL":"https://doi.org/10.1186/s12912-025-03128-3","source":"openalex"},{"id":"oa:W4410632614","type":"article-journal","title":"Deep learning and artificial intelligence for drug discovery, application, challenge, and future perspectives","abstract":"This review will examine how artificial intelligence, profound learning technologies, has affected drug discovery. Deep learning technology (DLT), a sub-field of AI that uses intricate algorithms and enormous datasets, is transforming every point along the road to drug development. Integrating clinical trial data, target identification or lead optimization, and personalized medicine have all become possible thanks to DLT. Given the explosion in IUPAC-compliant compounds registered with PubChem or derived from existing ones, DLT has given the pharmaceutical industry a massive booster shot. We will explore the key role generative models play in creating new drug compounds and why interdisciplinary collaboration is essential to entirely using AI's potential for drug discovery. In addition, the purpose of this article is to consider further perspectives concerning what problems exist at present in deep learning and AI-driven drug discovery. We focus on its potential as an accelerated, more effectiveeven tailored healthcare technology. As AI technology advances, a new field emerges in drug development, tipping the global balance between 'well' and 'ill.'","author":[{"family":"Ali","given":"Nouman"},{"family":"Hanif","given":"Nimra"},{"family":"Khan","given":"Hassan"},{"family":"Waseem","given":"Muhammad"},{"family":"Saeed","given":"Afshan"},{"family":"Zakir","given":"Sadia"},{"family":"Khan","given":"Abeeha"},{"family":"Aamir","given":"Mejerrah"},{"family":"Ali","given":"Adeeba"},{"family":"Ali","given":"Aamir"},{"family":"Saleem","given":"A"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s42452-025-06991-6","URL":"https://doi.org/10.1007/s42452-025-06991-6","source":"openalex"},{"id":"oa:W4410444203","type":"article-journal","title":"Performance of the Large Language Models in African rheumatology: a diagnostic test accuracy study of ChatGPT-4, Gemini, Copilot, and Claude artificial intelligence","abstract":"BACKGROUND: Artificial intelligence (AI) tools, particularly Large Language Models (LLMs), are revolutionizing medical practice, including rheumatology. However, their diagnostic capabilities remain underexplored in the African context. To assess the diagnostic accuracy of ChatGPT-4, Gemini, Copilot, and Claude AI in rheumatology within an African population. METHODS: This was a cross-sectional analytical study with retrospective data collection, conducted at the Rheumatology Department of Bogodogo University Hospital Center (Burkina Faso) from January 1 to June 30, 2024. Standardized clinical and paraclinical data from 103 patients were submitted to the four AI models. The diagnoses proposed by the AIs were compared to expert-confirmed diagnoses established by a panel of senior rheumatologists. Diagnostic accuracy, sensitivity, specificity, and predictive values were calculated for each AI model. RESULTS: Among the patients enrolled in the study period, infectious diseases constituted the most common diagnostic category, representing 47.57% (n = 49). ChatGPT-4 achieved the highest diagnostic accuracy (86.41%), followed by Claude AI (85.44%), Copilot (75.73%), and Gemini (71.84%). The inter-model agreement was moderate, with Cohen's kappa coefficients ranging from 0.43 to 0.59. ChatGPT-4 and Claude AI demonstrated high sensitivity (> 90%) for most conditions but had lower performance for neoplastic diseases (sensitivity < 67%). Patients under 50 years old had a significantly higher probability of receiving a correct diagnosis with Copilot (OR = 3.36; 95% CI [1.16-9.71]; p = 0.025). CONCLUSION: LLMs, particularly ChatGPT-4 and Claude AI, show high diagnostic capabilities in rheumatology, despite some limitations in specific disease categories. CLINICAL TRIAL NUMBER: Not applicable.","author":[{"family":"Bayala","given":"Yannick"},{"family":"Tiendrébeogo","given":"Wendlassida"},{"family":"Ouedraogo","given":"Dieu‐donné"},{"family":"Kaboré","given":"Fulgence"},{"family":"Sougué","given":"Charles"},{"family":"Yaméogo","given":"Rélwendé"},{"family":"Nacanabo","given":"Wendlassida"},{"family":"Tinni","given":"Ismaël"},{"family":"Ouédraogo","given":"Aboubakar"},{"family":"Zongo","given":"Enselme"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s41927-025-00512-z","URL":"https://doi.org/10.1186/s41927-025-00512-z","source":"openalex"},{"id":"oa:W4407363677","type":"article-journal","title":"Ethical implications of artificial intelligence integration in nursing practice in arab countries: literature review","abstract":"BACKGROUND: Applying artificial intelligence (AI) to nursing practice has dramatically enhanced healthcare delivery in Arab countries. However, AI application also raises complex moral issues, including patient privacy, data security, responsibility, transparency, and equity in decision-making. AIM: A systematic analysis of the ethical issues surrounding the application of AI in nursing practice in Arab nations is carried out in this review, highlighting the most important ethical issues and recommending responsible AI integration. METHODS: A comprehensive literature search was conducted across major databases. Following the initial identification of 150 articles, 120 were selected for full-text review based on the title and abstract screening. Subsequently, 50 pertinent studies were incorporated into this review. RESULTS: Numerous significant ethical concerns regarding AI application in decision-making processes were identified. The assessment also highlighted the possible effects of AI on the nurse-patient interaction and the critical role played by the ethics committees and regulatory frameworks in resolving these issues. CONCLUSION: Ethical frameworks must be established to guarantee AI integration into nursing practice, safeguard patients' welfare, and strengthen the trust between healthcare providers and patients. CLINICAL TRIAL: No clinical Trial.","author":[{"family":"Ibrahim","given":"Ateya"},{"family":"Zoromba","given":"Mohamed"},{"family":"Abousoliman","given":"Ali"},{"family":"Zaghamir","given":"Donia"},{"family":"Alenezi","given":"Ibrahim"},{"family":"Elsayed","given":"Ebtesam"},{"family":"Mohamed","given":"Heba"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12912-025-02798-3","URL":"https://doi.org/10.1186/s12912-025-02798-3","source":"openalex"},{"id":"oa:W4406149781","type":"article-journal","title":"A Comprehensive Review of Artificial Intelligence (AI) Applications in Pulmonary Hypertension (PH)","abstract":"Background: Pulmonary hypertension (PH) is a complex condition associated with significant morbidity and mortality. Traditional diagnostic and management approaches for PH often face limitations, leading to delays in diagnosis and potentially suboptimal treatment outcomes. Artificial intelligence (AI), encompassing machine learning (ML) and deep learning (DL) offers a transformative approach to PH care. Materials and Methods: We systematically searched PubMed, Scopus, and Web of Science for original studies on AI applications in PH, using predefined keywords. Out of more than 500 initial articles, 45 relevant studies were selected. Risk of bias was evaluated using PROBAST (Prediction model Risk of Bias Assessment Tool). Results: This review examines the potential applications of AI in PH, focusing on its role in enhancing diagnosis, disease classification, and prognostication. We discuss how AI-powered analysis of medical data can improve the accuracy and efficiency of detecting PH. Furthermore, we explore the potential of AI in risk stratification, leading to treatment optimization for PH. Conclusions: While acknowledging the existing challenges and limitations and the need for continued exploration and refinement of AI-driven tools, this review highlights the significant promise of AI in revolutionizing PH management to improve patient outcomes.","author":[{"family":"Esfahani","given":"Sogol"},{"family":"Ali","given":"Nima"},{"family":"Farina","given":"Juan"},{"family":"Scalia","given":"Isabel"},{"family":"Pereyra","given":"Milagros"},{"family":"Abbas","given":"Mohammed"},{"family":"Javadi","given":"Niloofar"},{"family":"Bismee","given":"Nadera"},{"family":"Abdelfattah","given":"Fatmaelzahraa"},{"family":"Awad","given":"Kamal"},{"family":"Ibrahim","given":"Omar"},{"family":"Sheashaa","given":"Hesham"},{"family":"Barry","given":"Timothy"},{"family":"Scott","given":"Robert"},{"family":"Ayoub","given":"Chadi"},{"family":"Arsanjani","given":"Reza"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/medicina61010085","URL":"https://doi.org/10.3390/medicina61010085","source":"openalex"},{"id":"oa:W4409286413","type":"article-journal","title":"Application of artificial intelligence in the diagnosis of malignant digestive tract tumors: focusing on opportunities and challenges in endoscopy and pathology","abstract":"BACKGROUND: Malignant digestive tract tumors are highly prevalent and fatal tumor types globally, often diagnosed at advanced stages due to atypical early symptoms, causing patients to miss optimal treatment opportunities. Traditional endoscopic and pathological diagnostic processes are highly dependent on expert experience, facing problems such as high misdiagnosis rates and significant inter-observer variations. With the development of artificial intelligence (AI) technologies such as deep learning, real-time lesion detection with endoscopic assistance and automated pathological image analysis have shown potential in improving diagnostic accuracy and efficiency. However, relevant applications still face challenges including insufficient data standardization, inadequate interpretability, and weak clinical validation. OBJECTIVE: This study aims to systematically review the current applications of artificial intelligence in diagnosing malignant digestive tract tumors, focusing on the progress and bottlenecks in two key areas: endoscopic examination and pathological diagnosis, and to provide feasible ideas and suggestions for subsequent research and clinical translation. METHODS: A systematic literature search strategy was adopted to screen relevant studies published between 2017 and 2024 from databases including PubMed, Web of Science, Scopus, and IEEE Xplore, supplemented with searches of early classical literature. Inclusion criteria included studies on malignant digestive tract tumors such as esophageal cancer, gastric cancer, or colorectal cancer, involving the application of artificial intelligence technology in endoscopic diagnosis or pathological analysis. The effects and main limitations of AI diagnosis were summarized through comprehensive analysis of research design, algorithmic methods, and experimental results from relevant literature. RESULTS: In the field of endoscopy, multiple deep learning models have significantly improved detection rates in real-time polyp detection, early gastric cancer, and esophageal cancer screening, with some commercialized systems successfully entering clinical trials. However, the scale and quality of data across different studies vary widely, and the generalizability of models to multi-center, multi-device environments remains to be verified. In pathological analysis, using convolutional neural networks, multimodal pre-training models, etc., automatic tissue segmentation, tumor grading, and assisted diagnosis can be achieved, showing good scalability in interactive question-answering. Nevertheless, clinical implementation still faces obstacles such as non-uniform data standards, lack of large-scale prospective validation, and insufficient model interpretability and continuous learning mechanisms. CONCLUSION: Artificial intelligence provides new technological opportunities for endoscopic and pathological diagnosis of malignant digestive tract tumors, achieving positive results in early lesion identification and assisted decision-making. However, to achieve the transition from research to widespread clinical application, data standardization, model reliability, and interpretability still need to be improved through multi-center joint research, and a complete regulatory and ethical system needs to be established. In the future, artificial intelligence will play a more important role in the standardization and precision management of diagnosis and treatment of digestive tract tumors.","author":[{"family":"Gao","given":"Yinhu"},{"family":"Wen","given":"Peizhen"},{"family":"Liu","given":"Yuan"},{"family":"Sun","given":"Yahuang"},{"family":"Qian","given":"Hui"},{"family":"Zhang","given":"Xin"},{"family":"Peng","given":"Huan"},{"family":"Gao","given":"Yanli"},{"family":"Li","given":"Cuiyu"},{"family":"Gu","given":"Zhangyuan"},{"family":"Zeng","given":"Hua‐jin"},{"family":"Hong","given":"Zhijun"},{"family":"Wang","given":"Weijun"},{"family":"Yan","given":"Ronglin"},{"family":"Hu","given":"Zunqi"},{"family":"Fu","given":"Hongbing"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12967-025-06428-z","URL":"https://doi.org/10.1186/s12967-025-06428-z","source":"openalex"},{"id":"oa:W7135205874","type":"article-journal","title":"Artificial Intelligence Applications in Gastric Cancer Surgery: Bridging Early Diagnosis and Responsible Precision Medicine","abstract":"Background: Artificial intelligence is emerging as a promising tool in surgical oncology, with growing evidence suggesting potential applications in diagnostic support, intraoperative guidance, and perioperative risk assessment. In gastric cancer surgery, emerging applications range from AI-assisted endoscopic detection to data-driven perioperative risk prediction, while some technological developments, particularly in robotic autonomy, derive from broader surgical or experimental models that may inform future gastric procedures. Methods: A narrative review was conducted following established methodological standards, including the Scale for the Assessment of Narrative Review Articles (SANRA) and the Search–Appraisal–Synthesis–Analysis (SALSA) framework. English-language studies indexed in PubMed, Scopus, Embase, and Web of Science up to October 2025 were included. Evidence was synthesized thematically across five domains: AI-assisted anatomical recognition and lymphadenectomy support, autonomous robotic systems, early cancer detection, perioperative predictive and frailty models, and ethical and regulatory considerations. Results: AI-based computer vision and deep learning algorithms have demonstrated promising capabilities for real-time anatomical recognition, surgical phase classification, and intraoperative guidance, although evidence of direct patient-level benefit remains limited. In diagnostic settings, AI-assisted endoscopy and Raman spectroscopy have been shown to improve early lesion detection and reduce dependence on operator experience. Predictive models, including MySurgeryRisk and AI-driven frailty assessments, may support individualized prehabilitation planning and perioperative risk stratification. Persistent limitations include small and heterogeneous datasets, insufficient external validation, and unresolved concerns related to data privacy, algorithmic interpretability, and medico-legal responsibility. Conclusions: Artificial intelligence is progressively emerging as a promising tool in gastric cancer surgery, integrating automation, advanced analytics, and human clinical reasoning. Its safe and ethical adoption requires robust validation, transparent governance, and continuous surgeon oversight. When developed within human-centered and ethically grounded frameworks, AI can augment, rather than replace, surgical expertise, potentially advancing precision, safety, and equity in oncologic care.","author":[{"family":"Malerba","given":"Silvia"},{"family":"Vladimirov","given":"Miljana"},{"family":"Goyal","given":"Aman"},{"family":"Dulskas","given":"Audrius"},{"family":"Baušys","given":"Augustinas"},{"family":"Cwalinski","given":"Tomasz"},{"family":"Girnyi","given":"Sergii"},{"family":"Skokowski","given":"Jaroslaw"},{"family":"Duka","given":"Ruslan"},{"family":"Molchanov","given":"Robert"},{"family":"Jovanovic","given":"Bojan"},{"family":"Ciarleglio","given":"Francesco"},{"family":"Brolese","given":"Alberto"},{"family":"Gonfa","given":"Kebebe"},{"family":"Demmo","given":"Abdi"},{"family":"Dambrauskas","given":"Žilvinas"},{"family":"Bonet","given":"Adolfo"},{"family":"Testini","given":"M"},{"family":"Prete","given":"Francesco"},{"family":"Calu","given":"Valentin"},{"family":"Calomino","given":"Natale"},{"family":"Jain","given":"Vikas"},{"family":"Karamarković","given":"Aleksandar"},{"family":"Połom","given":"Karol"},{"family":"Abou-Mrad","given":"Adel"},{"family":"Oviedo","given":"Rodolfo"},{"family":"Vashist","given":"Yogesh"},{"family":"Marano","given":"Luigi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/jcm15062208","URL":"https://doi.org/10.3390/jcm15062208","source":"openalex"},{"id":"oa:W4409473464","type":"article-journal","title":"Role of Generative Artificial Intelligence in Personalized Medicine: A Systematic Review","abstract":"Precision medicine presents challenges in data collection, cost, and privacy as it tailors treatments to each patient's unique genetic and clinical profile. With its ability to produce realistic and confidential patient data, generative artificial intelligence (AI) offers a promising avenue that could revolutionize patient-centric healthcare. This systematic review aims to assess the role of generative AI in personalized medicine. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we searched PubMed, Web of Science, Scopus, CINAHL, and Google Scholar, identifying 549 studies. After removing duplicates and applying eligibility criteria, 27 studies were found relevant and were included in this systematic review. Generative adversarial networks (GANs) were the most commonly used models (16 studies), followed by variational autoencoders (VAEs; seven studies). These models were primarily applied to drug response prediction, treatment effect estimation, biomarker discovery, and patient stratification. Generative AI models have shown significant promise in revolutionizing personalized medicine by enabling precise treatment predictions and patient-specific therapeutic insights. Despite their potential, challenges related to model validation, interpretability, and bias remain. Future research should prioritize large-scale validation studies using diverse datasets to enhance the clinical applicability and reliability of these AI-driven approaches.","author":[{"family":"Mishra","given":"Aashish"},{"family":"Majumder","given":"Anirban"},{"family":"Kommineni","given":"Dheeraj"},{"family":"Joseph","given":"Christopher"},{"family":"Chowdhury","given":"Tanay"},{"family":"Anumula","given":"Sathish"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.82310","URL":"https://doi.org/10.7759/cureus.82310","source":"openalex"},{"id":"oa:W4409278465","type":"article-journal","title":"A critical assessment of artificial intelligence in magnetic resonance imaging of cancer","abstract":"Given the enormous output and pace of development of artificial intelligence (AI) methods in medical imaging, it can be challenging to identify the true success stories to determine the state-of-the-art of the field. This report seeks to provide the magnetic resonance imaging (MRI) community with an initial guide into the major areas in which the methods of AI are contributing to MRI in oncology. After a general introduction to artificial intelligence, we proceed to discuss the successes and current limitations of AI in MRI when used for image acquisition, reconstruction, registration, and segmentation, as well as its utility for assisting in diagnostic and prognostic settings. Within each section, we attempt to present a balanced summary by first presenting common techniques, state of readiness, current clinical needs, and barriers to practical deployment in the clinical setting. We conclude by presenting areas in which new advances must be realized to address questions regarding generalizability, quality assurance and control, and uncertainty quantification when applying MRI to cancer to maintain patient safety and practical utility.","author":[{"family":"Wu","given":"Chengyue"},{"family":"Andaloussi","given":"Meryem"},{"family":"Hormuth","given":"David"},{"family":"Lima","given":"Ernesto"},{"family":"Lorenzo","given":"Guillermo"},{"family":"Stowers","given":"Casey"},{"family":"Ravula","given":"Sriram"},{"family":"Levac","given":"Brett"},{"family":"Dimakis","given":"Alexandros"},{"family":"Tamir","given":"Jonathan"},{"family":"Brock","given":"Kristy"},{"family":"Chung","given":"Caroline"},{"family":"Yankeelov","given":"Thomas"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s44303-025-00076-0","URL":"https://doi.org/10.1038/s44303-025-00076-0","source":"openalex"},{"id":"oa:W4410791589","type":"article-journal","title":"Insights Into the Future: Assessing Medical Students' Artificial Intelligence Readiness ‐ A Cross‐Sectional Study at Kerman University of Medical Sciences (2022)","abstract":"ABSTRACT Background Artificial intelligence (AI) has recently advanced in medicine globally, transforming healthcare delivery and medical education. While AI integration into medical curricula is gaining momentum worldwide, research on medical students' preparedness remains limited, particularly in developing countries. This paper aims to investigate the readiness of medical students at the Kerman University of Medical Sciences to employ AI in medicine in 2022. Methods This cross‐sectional research was carried out by distributing the validated 20‐item Medical Artificial Intelligence Readiness Scale for Medical Students (MAIRS‐MS) among 360 medical students, with a response rate of 94% ( n = 340). The MAIRS‐MS assessed four domains, including cognition (8 items), ability (7 items), vision (2 items), and ethics (3 items), using a 5‐point Likert scale. Data analysis was conducted by descriptive statistics and independent sample t ‐tests in SPSS v24.0, considering p < 0.05 significant. Results Participants demonstrated below‐average readiness scores across all domains: ability ( M = 21.88 ± 6.74, 62.5% of the maximum possible score), cognition ( M = 20.30 ± 7.04, 50.8%), ethics ( M = 10.94 ± 3.04, 72.9%), and vision ( M = 6.09 ± 1.94, 60.9%). The total mean readiness score was 59.21 ± 16.12 (59.2% of the maximum). The highest and lowest‐rated items were “value of AI in education” (3.96 ± 1.18) and “explaining AI system training” (2.10 ± 1.01), respectively. No significant differences were found across demographic factors ( p > 0.05). Conclusion Iranian medical students currently show limited readiness for AI integration in healthcare practice. Therefore, the study recommends: (1) implementing structured introductory AI courses in medical curricula, focusing particularly on technical fundamentals and practical applications, and (2) developing hands‐on training programs that combine AI concepts with clinical scenarios. These findings provide valuable insights for curriculum development and educational policy in medical education.","author":[{"family":"Rezazadeh","given":"Hossein"},{"family":"Mahani","given":"Ali"},{"family":"Salajegheh","given":"Mahla"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/hsr2.70870","URL":"https://doi.org/10.1002/hsr2.70870","source":"openalex"},{"id":"oa:W4408291037","type":"article-journal","title":"Evaluating the Use of Generative Artificial Intelligence to Support Genetic Counseling for Rare Diseases","abstract":"Background/Objectives: Rare diseases often present challenges in obtaining reliable and accurate information than common diseases owing to their low prevalence. Patients and families often rely on self-directed learning, but understanding complex medical information can be difficult, increasing the risk of misinformation. This study aimed to evaluate whether generative artificial intelligence (AI) provides accurate and non-harmful answers to rare disease-related questions and assesses its utility in supporting patients and families requiring genetic counseling. Methods: We evaluated four generative AI models available between 22 September and 4 October 2024: ChatGPT o1-Preview, Gemini advanced, Claude 3.5 sonnet, and Perplexity sonar huge. A total of 102 questions targeting four rare diseases, covering general information, diagnosis, treatment, prognosis, and counseling, were prepared. Four evaluators scored the responses for professionalism and accuracy using the Likert scale (1: poor, 5: excellent). Results: The average scores ranked the AI models as: ChatGPT (4.24 ± 0.73), Gemini (4.15 ± 0.74), Claude (4.13 ± 0.82), and Perplexity (3.35 ± 0.80; p < 0.001). Perplexity had the highest proportion of scores of 1 (very poor) and 2 (poor) (7.6%, 31/408), followed by Gemini (2.0%, 8/408), Claude (1.5%, 6/408), and ChatGPT (1.5%, 6/408). The accuracy of responses in the counseling part across all four diseases was significantly different (p < 0.001). Conclusions: The four generative AI models generally provided reliable information. However, occasional inaccuracies and ambiguous references may lead to confusion and anxiety among patients and their families. To ensure its effective use, recognizing the limitations of generative AI and providing guidance from experts regarding its proper utilization is essential.","author":[{"family":"Jeon","given":"Suok"},{"family":"Lee","given":"Sua"},{"family":"Chung","given":"Hae‐sun"},{"family":"Yun","given":"Ji"},{"family":"Park","given":"Eun"},{"family":"So","given":"Min‐kyung"},{"family":"Huh","given":"Jungwon"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/diagnostics15060672","URL":"https://doi.org/10.3390/diagnostics15060672","source":"openalex"},{"id":"oa:W4410592464","type":"article-journal","title":"Artificial intelligence-based predictive models for shear wave velocity of soils: A comprehensive review","abstract":"Shear wave velocity (V s ) of soils is a crucial property in geotechnical engineering practice, affecting seismic site response analysis, seismic hazard assessment , and dynamic soil-structure interaction. The precise determination of V s is crucial in assessing the dynamic behavior of soils during seismic events, as it markedly influences the amplification and attenuation of ground motions. While several empirical equations have been proposed thus far for estimating V s in earthen materials, the majority of them lack the required accuracy and predictive capability. As a result, there has been a growing tendency among practicing engineers towards utilizing Artificial Intelligence (AI) for V s prediction. This paper presents an extensive overview of the developments in deploying AI and its subsets, including Machine Learning (ML) and Deep Learning (DL) techniques, for the precise estimation of V s in soil deposits. Notably, despite the importance of shear wave velocity as a key geotechnical parameter, no prior review study has exclusively focused on evaluating it using AI-based models. This review systematically examines various AI-based methodologies employed by researchers to enhance the reliability and precision of V s predictions using soil properties and in-situ test data. The advantages of ML techniques over conventional empirical correlations are thoroughly analyzed and critically compared. Additionally, the paper discusses the relative performance of different AI-based approaches, outlining their strengths and limitations in V s estimation. The review also provides a general qualitative assessment of V s measurement methods, offering guidance on selecting the most appropriate approach based on project-specific requirements and constraints. Finally, through a critical evaluation of existing literature, key knowledge gaps are identified, and potential directions for future research in this domain are proposed.","author":[{"family":"Payan","given":"Meghdad"},{"family":"Asadi","given":"Parsa"},{"family":"Jamaldar","given":"Amirhossein"},{"family":"Salimi","given":"Mahdi"},{"family":"Ranjbar","given":"Payam"},{"family":"Armaghani","given":"Danial"},{"family":"He","given":"Xuzhen"},{"family":"Sheng","given":"Daichao"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.engappai.2025.111095","URL":"https://doi.org/10.1016/j.engappai.2025.111095","source":"openalex"},{"id":"oa:W4416470794","type":"article-journal","title":"Leveraging artificial intelligence in antibody-drug conjugate development: from target identification to clinical translation in oncology","abstract":"Artificial intelligence (AI) is opening new frontiers in the development of antibody-drug conjugates (ADCs), offering unprecedented opportunities for precision therapy. This review outlines how AI empowers each stage of the ADC pipeline. In target discovery, multi-omics integration and graph-based learning prioritize tumor-selective and internalizing antigens. In antibody engineering, structure prediction, affinity optimization, and developability modeling streamline candidate selection. For linker-payload design, generative models and multi-objective optimization approaches support the rational design of conjugates that balance potency, stability, and immunogenicity. In absorption, distribution, metabolism, excretion, and toxicity (ADMET) modeling, deep learning and transformer-based frameworks predict pharmacokinetics and toxicity with increasing accuracy and mechanistic clarity. In clinical development, AI facilitates patient stratification, response prediction, and trial simulation through digital twin models, adaptive dosing algorithms, and real-world data integration. These capabilities support a more personalized and efficient pathway from bench to bedside. To further realize the impact of AI in ADC development, we highlight strategic priorities including the creation of curated, multimodal datasets, interpretable model architectures, and closed-loop experimental platforms. Together, these advances will be essential for realizing the full potential of AI to support rational, scalable, and personalized ADC-based therapies in oncology.","author":[{"family":"Lu","given":"Ye"},{"family":"Huang","given":"Weijun"},{"family":"Li","given":"Yuxuan"},{"family":"Xu","given":"Yanzhi"},{"family":"Wei","given":"Qingyang"},{"family":"Sha","given":"Chulin"},{"family":"Guo","given":"Peng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41698-025-01159-2","URL":"https://doi.org/10.1038/s41698-025-01159-2","source":"openalex"},{"id":"oa:W4409045115","type":"article-journal","title":"Applications of Generative Artificial Intelligence in Electronic Medical Records: A Scoping Review","abstract":"Electronic Medical Records (EMRs) are central to the modern healthcare system. Recent advances in artificial intelligence (AI), particularly generative artificial intelligence (GenAI), have opened new opportunities for the advancement of EMRs. This scoping review aims to explore the current real-world applications of GenAI within EMRs to support an understanding of AI applications in healthcare. A literature search was conducted following PRISMA-ScR guidelines. The search was conducted using Ovid MEDLINE, up to 28 October 2024, using a peer-reviewed search strategy. Overall, 55 studies were included. A list of five themes was generated by human reviewers based on the literature review: data manipulation (24), patient communication (9), clinical decision making (8), clinical prediction (8), summarization (4), and other (2). The majority of studies originated from the United States (35). Both proprietary and commercially available models were tested, with ChatGPT being the most commonly referenced LLM. As these models continue to be developed, their diverse use cases within EMRs have the potential to improve patient outcomes, enhance access to medical data, streamline hospital workflows, and reduce physician workload. However, continued problems surrounding data privacy, trust, bias, model hallucinations, and the need for robust evaluation remain. Further research considering the ethical, medical, and societal implications of GenAI applications in EMRs is essential to validate these findings and address existing limitations to support healthcare advancement.","author":[{"family":"Morjaria","given":"Leo"},{"family":"Gandhi","given":"B"},{"family":"Haider","given":"Nabil"},{"family":"Mellon","given":"Matthew"},{"family":"Sibbald","given":"Matthew"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/info16040284","URL":"https://doi.org/10.3390/info16040284","source":"openalex"},{"id":"oa:W7139943021","type":"article-journal","title":"From prediction to decision: Advancing medical artificial intelligence toward real clinical practice","abstract":"Artificial intelligence (AI) in medicine is advancing steadily toward real clinical practice, not only through improved predictive performance but also through more decision-relevant modeling, evaluation, and interaction. The studies highlighted here illustrate three complementary directions. First, contemporary imaging models—exemplified by 3D vision transformer-based analysis of preoperative chest computed tomography (CT)—are being used to infer clinically consequential phenotypes, advancing from image recognition toward decision support in high-stakes settings such as surgical planning. Second, image-to-biomarker pipelines such as automated quantification of retinal vascular fractal dimension demonstrate how medical images can be transformed into reproducible quantitative markers suitable for population-level analysis and risk stratification. Third, large language models (LLMs) are increasingly evaluated and positioned as clinical communication and interpretation components: structured assessments in telepharmacy and bilingual patient education move beyond fluency to safety, actionability, empathy, and readability, while emerging perspectives consider LLMs as interpretive interfaces for complex, temporally evolving health data, including wearable sensing. At the same time, these advances also expose persistent bottlenecks that limit real-world deployment: the continued dominance of single-task and static formulations, fragmented systems driven by task-specific fine-tuning, limited reasoning over disease trajectories and evolving clinical contexts, and evaluation practices that remain insufficiently coupled to clinical workflows and downstream consequences. We argue that the next stage of medical AI should shift from accuracy-centered prediction toward decision-oriented, practice-ready systems that are robust across settings, clinically aligned in evaluation, and deployable at scale in routine care.","author":[{"family":"Chen","given":"Qi"},{"family":"Lyu","given":"Han"},{"family":"Li","given":"Dong"},{"family":"Wang","given":"Zhenchang"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1016/j.imed.2026.03.003","URL":"https://doi.org/10.1016/j.imed.2026.03.003","source":"openalex"},{"id":"oa:W4413969609","type":"article-journal","title":"Physical foundations for trustworthy medical imaging: A survey for artificial intelligence researchers","abstract":"Artificial intelligence in medical imaging has grown rapidly in the past decade, driven by advances in deep learning and widespread access to computing resources. Applications cover diverse imaging modalities, including those based on electromagnetic radiation (e.g., X-rays), subatomic particles (e.g., nuclear imaging), and acoustic waves (ultrasound). Each modality features and limitations are defined by its underlying physics. However, many artificial intelligence practitioners lack a solid understanding of the physical principles involved in medical image acquisition. This gap hinders leveraging the full potential of deep learning, as incorporating physics knowledge into artificial intelligence systems promotes trustworthiness, especially in limited data scenarios. This work reviews the fundamental physical concepts behind medical imaging and examines their influence on recent developments in artificial intelligence, particularly, generative models and reconstruction algorithms. Finally, we describe physics-informed machine learning approaches to improve feature learning in medical imaging.","author":[{"family":"Cobo","given":"Miriam"},{"family":"Fontecha","given":"David"},{"family":"Silva","given":"Wilson"},{"family":"Iglesias","given":"LL"},{"family":"Iglesias","given":"Lara"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.artmed.2025.103251","URL":"https://doi.org/10.1016/j.artmed.2025.103251","source":"openalex"},{"id":"oa:W4408850043","type":"article-journal","title":"An Academic Viewpoint (2025) on the Integration of Generative Artificial Intelligence in Medical Education: Transforming Learning and Practices","abstract":"Generative artificial intelligence (GAI) has introduced a new era of medical education by offering innovative solutions to critical challenges in teaching, assessment, and clinical training. This expanded review explores the current and potential applications of GAI across multiple domains, including personalized tutoring, enhanced academic administrative efficiency, and improved preparedness for daily learning interactions. Utilizing a narrative review methodology combined with expert analysis, this study involved a structured literature search in January 2025 across PubMed, Scopus, and Google Scholar, followed by iterative brainstorming sessions and expert evaluations to assess the feasibility and impact of various GAI applications. Six domain experts then appraised the feasibility and impact of GAI technologies across educational settings, resulting in 10 identified domains of application: Quality and Administration, Curriculum Development, Teaching and Learning, Assessment and Evaluation, Clinical Training, Academic Guidance, Student Research, Student Affairs, Internship Management, and Student Activities. Our findings highlight how GAI supports personalized learning - through adaptive tutoring and automated performance dashboards - while optimizing administrative tasks such as course registration and policy oversight. In addition, immersive simulations and virtual patient encounters reinforce clinical decision-making and practical skills. GAI-driven tools also streamline research processes via automated literature reviews and proposal refinement, ultimately fostering greater efficiency across academic environments. Despite these opportunities, ethical considerations remain a priority. Issues pertaining to data privacy, algorithmic bias, and equitable access must be addressed through robust regulatory frameworks and institution-wide policies. Overall, by embracing targeted, ethically guided implementations, GAI has the evolving potential to enhance educational quality, improve operational effectiveness, and equip future healthcare professionals with the adaptive skills needed in a patient-centered clinical landscape.","author":[{"family":"Almansour","given":"Mohammad"},{"family":"Soliman","given":"Mona"},{"family":"Aldekhyyel","given":"Raniah"},{"family":"Binkheder","given":"Samar"},{"family":"Temsah","given":"Mohamad"},{"family":"Malki","given":"Khalid"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.81145","URL":"https://doi.org/10.7759/cureus.81145","source":"openalex"},{"id":"oa:W4406062609","type":"article-journal","title":"Introduction to Artificial Intelligence and Machine Learning in Pathology and Medicine: Generative and Nongenerative Artificial Intelligence Basics","abstract":"This manuscript serves as an introduction to a comprehensive 7-part review article series on artificial intelligence (AI) and machine learning (ML) and their current and future influence within pathology and medicine. This introductory review provides a comprehensive grasp of this fast-expanding realm and its potential to transform medical diagnosis, workflow, research, and education. Fundamental terminology employed in AI-ML is covered using an extensive dictionary. The article also provides a broad overview of the main domains in the AI-ML field, encompassing both generative and nongenerative (traditional) AI, thereby serving as a primer to the other 6 review articles in this series that describe the details about statistics, regulations, bias, ethical dilemmas, and ML-Ops in AI-ML. The intent of these review articles is to better equip individuals who are or will be working in an AI-enabled health care system.","author":[{"family":"Rashidi","given":"Hooman"},{"family":"Pantanowitz","given":"Joshua"},{"family":"Hanna","given":"Matthew"},{"family":"Tafti","given":"Ahmad"},{"family":"Sanghani","given":"Parth"},{"family":"Buchinsky","given":"Adam"},{"family":"Fennell","given":"Brandon"},{"family":"Deebajah","given":"Mustafa"},{"family":"Wheeler","given":"Sarah"},{"family":"Pearce","given":"Thomas"},{"family":"Abukhiran","given":"Ibrahim"},{"family":"Robertson","given":"Scott"},{"family":"Palmer","given":"Octavia"},{"family":"Gur","given":"Mert"},{"family":"Tran","given":"Nam"},{"family":"Pantanowitz","given":"Liron"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.modpat.2024.100688","URL":"https://doi.org/10.1016/j.modpat.2024.100688","source":"openalex"},{"id":"oa:W4413141291","type":"article-journal","title":"Machine Learning and Artificial Intelligence in Nanomedicine","abstract":"Nanomedicine harnesses nanoscale materials, such as lipid, polymeric, and inorganic nanoparticles, to deliver diagnostic or therapeutic agents for cancer, infectious disease, and neurological disorders, among others. However, translating promising nanoparticle designs into clinically approved products remains a challenge. Factors such as particle size, surface chemistry, and payload interactions must be optimized, and preclinical results often fail to predict human efficacy. In recent years, artificial intelligence (AI) and machine learning (ML) have emerged as transformative tools to address these hurdles at every stage of nanomedicine development. By rapidly screening extensive libraries and extracting structure-function relationships, AI-driven models can rationalize nanoparticle formulation, predict biodistribution, and guide optimal design. Techniques like high-throughput DNA barcoding and automated liquid handling facilitate robust, large-scale data collection, feeding into computational pipelines that expedite discovery while reducing reliance on resource-intensive trial-and-error experiments. AI-based platforms also enable improved modeling of protein corona formation, which profoundly affects nanoparticle immunogenicity and cellular uptake. Despite these advances, challenges persist in data standardization, model generalizability, and establishing a clear regulatory framework since no dedicated U.S. Food and Drug Administration (FDA) guidance addresses the intersection of AI and nanomedicine. Overcoming these limitations requires harmonized data sharing, rigorous in vivo validation, and clear ethical and regulatory guidelines. This review summarizes the rapidly evolving landscape of AI in nanomedicine, highlighting key successes in design and preclinical prediction, as well as persistent obstacles to full-scale clinical integration. By illuminating these dynamics, we aim to chart a more efficient path forward in developing next-generation nanomedicine.","author":[{"family":"Chou","given":"Wei‐chun"},{"family":"Canchola","given":"Alexa"},{"family":"Zhang","given":"Fan"},{"family":"Lin","given":"Zhoumeng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/wnan.70027","URL":"https://doi.org/10.1002/wnan.70027","source":"openalex"},{"id":"oa:W4406152279","type":"article-journal","title":"Toward expert-level medical question answering with large language models","abstract":"Large language models (LLMs) have shown promise in medical question answering, with Med-PaLM being the first to exceed a 'passing' score in United States Medical Licensing Examination style questions. However, challenges remain in long-form medical question answering and handling real-world workflows. Here, we present Med-PaLM 2, which bridges these gaps with a combination of base LLM improvements, medical domain fine-tuning and new strategies for improving reasoning and grounding through ensemble refinement and chain of retrieval. Med-PaLM 2 scores up to 86.5% on the MedQA dataset, improving upon Med-PaLM by over 19%, and demonstrates dramatic performance increases across MedMCQA, PubMedQA and MMLU clinical topics datasets. Our detailed human evaluations framework shows that physicians prefer Med-PaLM 2 answers to those from other physicians on eight of nine clinical axes. Med-PaLM 2 also demonstrates significant improvements over its predecessor across all evaluation metrics, particularly on new adversarial datasets designed to probe LLM limitations (P < 0.001). In a pilot study using real-world medical questions, specialists preferred Med-PaLM 2 answers to generalist physician answers 65% of the time. While specialist answers were still preferred overall, both specialists and generalists rated Med-PaLM 2 to be as safe as physician answers, demonstrating its growing potential in real-world medical applications.","author":[{"family":"Singhal","given":"KK"},{"family":"Tu","given":"Tao"},{"family":"Gottweis","given":"Juraj"},{"family":"Sayres","given":"Rory"},{"family":"Wulczyn","given":"Ellery"},{"family":"Amin","given":"Mohamed"},{"family":"Hou","given":"Le"},{"family":"Clark","given":"Kevin"},{"family":"Pfohl","given":"Stephen"},{"family":"Cole-Lewis","given":"Heather"},{"family":"Neal","given":"Darlene"},{"family":"Rashid","given":"Qazi"},{"family":"Schaekermann","given":"Mike"},{"family":"Wang","given":"Amy"},{"family":"Dash","given":"Dev"},{"family":"Chen","given":"Jonathan"},{"family":"Shah","given":"Nigam"},{"family":"Lachgar","given":"Sami"},{"family":"Mansfield","given":"P"},{"family":"Prakash","given":"Sushant"},{"family":"Green","given":"Bradley"},{"family":"Dominowska","given":"Ewa"},{"family":"Arcas","given":"Blaise"},{"family":"Tomašev","given":"Nenad"},{"family":"Liu","given":"Yun"},{"family":"Wong","given":"Renee"},{"family":"Semturs","given":"Christopher"},{"family":"Mahdavi","given":"SS"},{"family":"Barral","given":"Joëlle"},{"family":"Webster","given":"Dale"},{"family":"Corrado","given":"Greg"},{"family":"Matias","given":"Yossi"},{"family":"Azizi","given":"Shekoofeh"},{"family":"Karthikesalingam","given":"Alan"},{"family":"Natarajan","given":"Vivek"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41591-024-03423-7","URL":"https://doi.org/10.1038/s41591-024-03423-7","source":"openalex"},{"id":"oa:W4415306199","type":"article-journal","title":"Artificial intelligence and computer-aided diagnosis in diagnostic decisions: 5 questions for medical informatics and human-computer interface research","abstract":"OBJECTIVES: Artificial intelligence (AI) has the potential to transform medical informatics by supporting clinical decision-making, reducing diagnostic errors, and improving workflows and efficiency. However, successful integration of AI-based decision support systems depends on careful consideration of human-AI collaboration, trust, skill maintenance, and automation bias. This work proposes five central questions to guide future research in medical informatics and human-computer interface (HCI). MATERIALS AND METHODS: We focus on AI-based clinical decision support systems, including computer vision algorithms for medical imaging (radiology, pathology), natural language processing for structured and unstructured electronic health record (EHR) data, and rule-based systems. Relevant data modalities include clinician-acquired images, EHR text, and increasingly, patient-generated content in telehealth contexts. We review existing evidence regarding diagnostic errors across specialties, the effectiveness and risks of AI tools in reducing perceptual and interpretive errors, and the human factors influencing diagnostic decision-making in AI-enabled contexts. We synthesize insights from medicine, cognitive science, and HCI to identify gaps in knowledge and propose five key questions for continued research. RESULTS: Diagnostic errors remain common across medicine, with AI offering potential to reduce both perceptual and interpretive errors. However, the impact of AI depends critically on how and when information is presented. Studies indicate that delayed or toggleable cues may outperform immediate ones, but attentional capture, overreliance, and bias remain significant risks. Explainable AI provides transparency but can also bias decisions. Long-term reliance on AI may erode clinician skills, particularly for trainees and in low-prevalence contexts. Historical failures of computer-aided diagnosis in mammography highlight these challenges. DISCUSSION AND CONCLUSION: Effective AI integration requires human-centered and adaptive design. Five central research questions address: (1) what type and format of information AI should provide; (2) when information should be presented; (3) how explainable AI affects diagnostic decisions; (4) how AI influences automation bias and complacency; and (5) the risks of skill decay due to reliance on AI. Each question underscores the importance of balancing efficiency, accuracy, and clinician expertise while mitigating bias and skill degradation. AI holds promise for improving diagnostic accuracy and efficiency, but realizing its potential requires post-deployment evaluation, equitable access, clinician oversight, and targeted training. AI must complement, rather than replace, human expertise, ensuring safe, effective, and sustainable integration into diagnostic decision-making. Addressing these challenges proactively can maximize AI's potential across healthcare and other high-stakes domains.","author":[{"family":"Brunyé","given":"Tad"},{"family":"Mitroff","given":"Stephen"},{"family":"Elmore","given":"Joann"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/jamia/ocaf123","URL":"https://doi.org/10.1093/jamia/ocaf123","source":"openalex"},{"id":"oa:W4407136469","type":"article-journal","title":"Perceptions of Medical Students towards Artificial Intelligence","abstract":"The incorporation of technological advancements, particularly Artificial Intelligence has transformed healthcare systems globally, especially post-COVID-19. Medical education faces challenges in incorporating AI due to instructor shortages and high software costs. Understanding medical students' attitudes towards AI is crucial for its successful integration into medical practice and education. Objective: To evaluate the attitude of medical undergraduate students towards AI in medicine. Methods: A descriptive, online cross-sectional study was executed among undergraduate medical students utilizing a non-probability convenience sampling. The questionnaire, distributed to 340 participants, included demographic details, perceptions towards artificial intelligence, and its effect on medical education. A total of 252 responses were received, receiving a 74% response rate. Data analysis was executed through SPSS version 26.0. Results: Demographic characteristics of 252 subjects revealed a mean age of 23.5 years, with a majority being female (74.2%) and in their first to third year of study (58.3%). Participants generally had intermediate computer literacy (75.7%) and used technology consistently for learning (57.5%). Regarding perceptions of AI, most students strongly agreed that AI will significantly impact healthcare (48.8%) and that all medical students should be educated about it (31.3%). Additionally, a substantial majority believed that integrating AI into medical education would enhance its quality (66.6%) and facilitate the learning experience (57.9%). Conclusions: It was concluded that students have positive perceptions regarding AI systems, demonstrating enthusiasm for expanding their knowledge of AI within their medical education.","author":[{"family":"Rizwan","given":"Shazia"},{"family":"Rizwan","given":"Shahveir"},{"family":"Rizwan","given":"Muhammad"},{"family":"Ali","given":"Hashim"},{"family":"Nawal"},{"family":"Batool","given":"Saima"}],"issued":{"date-parts":[[2025]]},"DOI":"10.54393/pjhs.v6i1.2364","URL":"https://doi.org/10.54393/pjhs.v6i1.2364","source":"openalex"},{"id":"doi:10.17632/xxkd3kjfdd.1","type":"article-journal","title":"Multi-Center Clinical Laboratory Dataset (MCLD)","abstract":"The Multi-Center Clinical Laboratory Dataset (MCLD) is a large-scale, anonymized clinical laboratory dataset collected from three accredited medical laboratories in the Kurdistan Region of Iraq. The dataset contains 145,288 laboratory test records corresponding to 51,029 anonymized laboratory visits, with patients aged 0 to 106 years, collected between 10 April 2023 and 25 March 2025. The dataset includes records from both female (91,143; 62.7%) and male (54,145; 37.3%) patients and covers 168 clinical laboratory tests. The standardized dataset is intended to facilitate research in artificial intelligence, machine learning, medical data mining, clinical decision support, epidemiological analysis, and healthcare analytics.","author":[{"family":"Latif Mahmood","given":"Mohammed"},{"family":"Rawf","given":"Karwan"},{"family":"Mohammed","given":"Hardi"},{"family":"Mohammed Salih","given":"Wria"},{"family":"Hassan Abdalqadir","given":"Akar"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/xxkd3kjfdd.1","URL":"https://doi.org/10.17632/xxkd3kjfdd.1","source":"datacite"},{"id":"doi:10.17632/xxkd3kjfdd","type":"article-journal","title":"Multi-Center Clinical Laboratory Dataset (MCLD)","abstract":"The Multi-Center Clinical Laboratory Dataset (MCLD) is a large-scale, anonymized clinical laboratory dataset collected from three accredited medical laboratories in the Kurdistan Region of Iraq. The dataset contains 145,288 laboratory test records corresponding to 51,029 anonymized laboratory visits, with patients aged 0 to 106 years, collected between 10 April 2023 and 25 March 2025. The dataset includes records from both female (91,143; 62.7%) and male (54,145; 37.3%) patients and covers 168 clinical laboratory tests. The standardized dataset is intended to facilitate research in artificial intelligence, machine learning, medical data mining, clinical decision support, epidemiological analysis, and healthcare analytics.","author":[{"family":"Latif Mahmood","given":"Mohammed"},{"family":"Rawf","given":"Karwan"},{"family":"Mohammed","given":"Hardi"},{"family":"Mohammed Salih","given":"Wria"},{"family":"Hassan Abdalqadir","given":"Akar"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/xxkd3kjfdd","URL":"https://doi.org/10.17632/xxkd3kjfdd","source":"datacite"},{"id":"doi:10.5281/zenodo.18213727","type":"article-journal","title":"Source Code for a Study on the Use of Artificial Intelligence in Predicting the Life Expectancy of Liver Graft Recipients","abstract":"The code developed in this research aims to implement, train, evaluate, and interpret Artificial Intelligence models capable of predicting whether liver transplant recipients will survive for more or less than twelve months, using exclusively the clinical data available to medical teams at the time an organ is offered. The entire development process was structured in a modular and reproducible manner, in accordance with the principles of the Design Science Research (DSR) methodology. The core component of the code corresponds to the implementation of supervised/unsupervised machine learning models, including linear models, decision trees and ensemble-based methods. Each model is trained to solve a binary classification problem, in which the target variable represents recipient survival beyond twelve months. The training process includes cross-validation and hyperparameter tuning to ensure robustness and to mitigate overfitting. To assess model performance, the code computes quantitative evaluation metrics the PR-ROC curve, Brier Sciore and Mean Net Benefit. These metrics enable systematic comparison across models and support the identification of those with the greatest potential for practical clinical application. In addition, the code incorporates interpretability and explainability techniques to identify the most relevant features for each trained model. This step is essential for aligning Artificial Intelligence outputs with clinical practice, allowing healthcare professionals to understand the factors that most strongly influence survival predictions and helping to reduce subjectivity in graft acceptance decisions. Please, do not forget to cite the database that are store in https://zenodo.org/records/18189988, with following DOI: https://doi.org/10.5281/zenodo.18189988. Full Citation (IEEE) [1]A. Galindo Leal, J. R. De Oliveira, B.-H. Ferraz Neto, R. L. Macacarie F. de A. Monteiro, “Anonymized Donor–Recipient Clinical Dataset for Artificial Intelligence Models in Liver Transplantation (São Paulo State, 2007–2022)”. Zenodo, jan. 08, 2026. doi: 10.5281/zenodo.18189988.","author":[{"family":"De Oliveira","given":"José"},{"family":"Galindo Leal","given":"Adriano"},{"family":"Macacari","given":"Rodrigo"},{"family":"Ferraz Neto","given":"Ben"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18213727","URL":"https://doi.org/10.5281/zenodo.18213727","source":"datacite"},{"id":"doi:10.5281/zenodo.19298958","type":"article-journal","title":"Source Code for a Study on the Use of Artificial Intelligence in Predicting the Life Expectancy of Liver Graft Recipients","abstract":"The code developed in this research aims to implement, train, evaluate, and interpret Artificial Intelligence models capable of predicting whether liver transplant recipients will survive for more or less than twelve months, using exclusively the clinical data available to medical teams at the time an organ is offered. The entire development process was structured in a modular and reproducible manner, in accordance with the principles of the Design Science Research (DSR) methodology. The core component of the code corresponds to the implementation of supervised/unsupervised machine learning models, including linear models, decision trees and ensemble-based methods. Each model is trained to solve a binary classification problem, in which the target variable represents recipient survival beyond twelve months. The training process includes cross-validation and hyperparameter tuning to ensure robustness and to mitigate overfitting. To assess model performance, the code computes quantitative evaluation metrics the PR-ROC curve, Brier Sciore and Mean Net Benefit. These metrics enable systematic comparison across models and support the identification of those with the greatest potential for practical clinical application. In addition, the code incorporates interpretability and explainability techniques to identify the most relevant features for each trained model. This step is essential for aligning Artificial Intelligence outputs with clinical practice, allowing healthcare professionals to understand the factors that most strongly influence survival predictions and helping to reduce subjectivity in graft acceptance decisions. Please, do not forget to cite the database that are store in https://zenodo.org/records/18189988, with following DOI: https://doi.org/10.5281/zenodo.18189988. Full Citation (IEEE) [1]A. Galindo Leal, J. R. De Oliveira, B.-H. Ferraz Neto, R. L. Macacarie F. de A. Monteiro, “Anonymized Donor–Recipient Clinical Dataset for Artificial Intelligence Models in Liver Transplantation (São Paulo State, 2007–2022)”. Zenodo, jan. 08, 2026. doi: 10.5281/zenodo.18189988.","author":[{"family":"De Oliveira","given":"José"},{"family":"Galindo Leal","given":"Adriano"},{"family":"Macacari","given":"Rodrigo"},{"family":"Ferraz Neto","given":"Ben"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19298958","URL":"https://doi.org/10.5281/zenodo.19298958","source":"datacite"},{"id":"doi:10.5281/zenodo.19740259","type":"article-journal","title":"A Literature Review on Artificial Intelligence Methods Related to Low Back Pain","abstract":"Abstract Low back pain (LBP) is a worldwide health problem caused by various diseases. It is difficult to establish standards in medical applications because of the differences in the causes of its occurrence and the individual effects in treatment. The LBP diagnosis and treatment processes generate different numerical and visual data. Today, artificial intelligence (AI) techniques have begun to be developed, aiming to improve the understanding of LBP's causes, treatment processes, and effectiveness using patient data. In our study, we aimed to systematically search the literature on the diagnosis and treatment processes of LBP using AI techniques. We conducted a systematic review of studies on LBP utilizing AI methods between 01.01.2000 and 01.05.2023, using the PubMed database. While searching the database, combinations of the terms \"Artificial Intelligence, \"\"Machine Learning, \"\"Deep Learning, \" and \"Low Back Pain\" were employed. A total of 369 articles were identified. According to the study inclusion criteria, 354 articles were excluded, and 15 studies were reviewed. Magnetic resonance images, biochemical parameters, kinematic variables, EMG signals, PET imaging, and other variables were used AI methods to diagnose LBP. The studies employing AI methods generally focused on classification and regression problems. The AI techniques developed for the diagnosis and treatment processes of LBP are promising. It is anticipated that multidisciplinary studies using artificial intelligence.","author":[{"family":"Ulku","given":"Veranyurt"},{"family":"Betul","given":"Akalin"},{"family":"Arzu","given":"Gerçek"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19740259","URL":"https://doi.org/10.5281/zenodo.19740259","source":"datacite"},{"id":"doi:10.5281/zenodo.19740260","type":"article-journal","title":"A Literature Review on Artificial Intelligence Methods Related to Low Back Pain","abstract":"Abstract Low back pain (LBP) is a worldwide health problem caused by various diseases. It is difficult to establish standards in medical applications because of the differences in the causes of its occurrence and the individual effects in treatment. The LBP diagnosis and treatment processes generate different numerical and visual data. Today, artificial intelligence (AI) techniques have begun to be developed, aiming to improve the understanding of LBP's causes, treatment processes, and effectiveness using patient data. In our study, we aimed to systematically search the literature on the diagnosis and treatment processes of LBP using AI techniques. We conducted a systematic review of studies on LBP utilizing AI methods between 01.01.2000 and 01.05.2023, using the PubMed database. While searching the database, combinations of the terms \"Artificial Intelligence, \"\"Machine Learning, \"\"Deep Learning, \" and \"Low Back Pain\" were employed. A total of 369 articles were identified. According to the study inclusion criteria, 354 articles were excluded, and 15 studies were reviewed. Magnetic resonance images, biochemical parameters, kinematic variables, EMG signals, PET imaging, and other variables were used AI methods to diagnose LBP. The studies employing AI methods generally focused on classification and regression problems. The AI techniques developed for the diagnosis and treatment processes of LBP are promising. It is anticipated that multidisciplinary studies using artificial intelligence.","author":[{"family":"Ulku","given":"Veranyurt"},{"family":"Betul","given":"Akalin"},{"family":"Arzu","given":"Gerçek"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19740260","URL":"https://doi.org/10.5281/zenodo.19740260","source":"datacite"},{"id":"oa:W4417121785","type":"article-journal","title":"A Survey of the Application of Explainable Artificial Intelligence in Biomedical Informatics","abstract":"This review investigates the application of Explainable Artificial Intelligence (XAI) in biomedical informatics, encompassing domains such as medical imaging, genomics, and electronic health records. Through a systematic analysis of 43 peer-reviewed articles, we examine current trends, as well as the strengths and limitations of methodologies currently used in real-world healthcare settings. Our findings highlight a growing interest in XAI, particularly in medical imaging, yet reveal persistent challenges in clinical adoption, including issues of trust, interpretability, and integration into decision-making workflows. We identify critical gaps in existing approaches and underscore the need for more robust, human-centred, and intrinsically interpretable models, with only 44% of the papers studied proposing human-centred validations. Furthermore, we argue that fairness and accountability, which are key to the acceptance of AI in clinical practice, can be supported by the use of post hoc tools for identifying potential biases but ultimately require the implementation of complementary fairness-aware or causal approaches alongside evaluation frameworks that prioritise clinical relevance and user trust. This review provides a foundation for advancing XAI research on the development of more transparent, equitable, and clinically meaningful AI systems for use in healthcare.","author":[{"family":"Eshkiki","given":"Hassan"},{"family":"Tanhaei","given":"Farinaz"},{"family":"Caraffini","given":"Fabio"},{"family":"Mora","given":"Benjamin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/app152412934","URL":"https://doi.org/10.3390/app152412934","source":"openalex"},{"id":"oa:W4411961370","type":"article-journal","title":"Artificial Intelligence and Machine Learning for Enhancing Resilience: Concepts, Applications, and Future Directions","abstract":"As contemporary societies face unprecedented challenges such as mounting mental health issues, environmental crises, and socioeconomic insecurity, the urgency of developing objective, scalable, and dynamic methodologies to study resilience has never been greater. This book arises at the intersection of cutting-edge technology and human insight. It focuses on the possibility for AI and ML to transform resilience assessment, prediction, and interventions across the individual, organizational, and ecological levels. The chapters included in this book represent an organized synthesis of cutting-edge science, pragmatic applications, and prospective potential. With machine learning algorithms to estimate psychological resilience and AI-based models for climate change adaptation and ecosystem management, this book demonstrates the rich innovations that are emerging at the cross-sector of technology and resilience science. Perhaps most importantly, this book does not gloss over the urgent ethical, technical, and regulatory issues that arise when AI is introduced to sensitive topics such as mental health and environmental management. Questions about data privacy, algorithmic bias, model interpretability, and equitable technology deployment are thoroughly investigated, providing lessons learned and suggestions for moving ahead. A significant strength of this work is its global focus. Showcasing work from contributors of various methodologies and regions provides the latest views on new methodologies, strategies for practical implementation, and on what still needs to be invented. This guarantees that the publication engages with the messy socio-cultural and environmental contexts in which these interventions work and that it doesn’t just mirror technological possibilities. For academicians, practitioners, technologists, and policymakers, this book is both a fundamental reference and an outlook resource. It provides: Holistic examination of AI and ML in the context of psychological, organizational, and ecological resilience. In-depth reviews on methodological innovations, such as deep learning, natural language processing, and sensor-based assessments. Unprecedented appraisals of barriers to implementation, with ethical and regulatory considerations. We trust that this book will inspire conversation, fuel innovation, and support a future in which technology supplements, rather than replaces, human ability to adapt, recover, and flourish. We encourage readers to critique the content, to reflect on how AI, ML, and resilience intersect in their particular contexts, and to join us in shaping a future where technological and human resilience evolve together.","author":[{"family":"Rane","given":"Nitin"},{"family":"Mallick","given":"Suraj"},{"family":"Rane","given":"Jayesh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.70593/978-93-7185-143-5","URL":"https://doi.org/10.70593/978-93-7185-143-5","source":"openalex"},{"id":"oa:W4413458885","type":"article-journal","title":"Comparing Artificial Intelligence Large Language Models in Medical Training: A Performance Analysis of ChatGPT and DeepSeek on United States Medical Licensing Examination (USMLE) Style Questions","abstract":"Introduction The integration of artificial intelligence (AI) into medical education is reshaping how students prepare for standardized examinations. Prior studies have shown that AI models can achieve high accuracy on United States Medical Licensing Examination (USMLE) questions, highlighting their potential for examination preparation. ChatGPT (GPT), especially the 4o model, is one of the most widely used AI models; however, its accessibility is limited by subscription costs and regional censorship. DeepSeek (DS), a newer AI model, offers free access and has demonstrated comparable performance in general tasks. In this study, we compared the performance of GPT-4o and DS DeepThink R1 on the AMBOSS medical board preparation question bank to evaluate their potential and limitations as supplementary tools in medical education. Methods We extracted 1,079 USMLE-style multiple-choice questions from the AMBOSS question bank. Questions were categorized by USMLE Step 1 and Step 2 examinations and further grouped by topic, resulting in 36 categories. Each question was assigned a difficulty level (easy, intermediate, or hard) based on AMBOSS grading criteria. To ensure balanced representation, we randomly selected 10 questions per difficulty level per category. Questions and answer choices were copied verbatim from the AMBOSS website and input into GPT-4o and DS R1 without any modification. Model responses were scored as correct or incorrect, and correctness rates were compared across GPT-4o, DS R1, and AMBOSS user performance. Results Both GPT and DS outperformed AMBOSS users, with overall accuracies of 88.79%, 78.68%, and 56.98%, respectively. Comparing GPT and DS, GPT performed significantly better overall (t=7.90, p<0.0001). When stratified by examination type, GPT achieved significantly higher accuracy than DS in both Step 1 (0.89 vs. 0.78, p < 0.0001) and Step 2 (0.88 vs. 0.80, p < 0.0001). GPT consistently showed higher accuracy than DS at all three difficulty levels. However, when further stratified by examination type, statistically significances were only observed in intermediate (p = 0.0002) and hard (p = 0.0021) questions in both Step 1 and Step 2. Conclusion Our findings demonstrated that both AI models outperformed human learners, with GPT-4o showing superior accuracy, particularly in intermediate and hard questions. While DS underperformed relative to GPT, its free accessibility and competitive accuracy in easy questions suggest that it may serve as a viable alternative, particularly in resource-limited settings.","author":[{"family":"Zhang","given":"Runze"},{"family":"Cai","given":"Qinyun"},{"family":"Sartori","given":"Ângela"},{"family":"Gayed","given":"Nasser"},{"family":"Collette","given":"Heather"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.90212","URL":"https://doi.org/10.7759/cureus.90212","source":"openalex"},{"id":"oa:W4405991965","type":"article-journal","title":"Optimizing Parkinson’s Disease Prediction: A Comparative Analysis of Data Aggregation Methods Using Multiple Voice Recordings via an Automated Artificial Intelligence Pipeline","abstract":"Patient-level grouped data are prevalent in public health and medical fields, and multiple instance learning (MIL) offers a framework to address the challenges associated with this type of data structure. This study compares four data aggregation methods designed to tackle the grouped structure in classification tasks: post-mean, post-max, post-min, and pre-mean aggregation. We developed a customized AI pipeline that incorporates twelve machine learning algorithms along with the four aggregation methods to detect Parkinson’s disease (PD) using multiple voice recordings from individuals available in the UCI Machine Learning Repository, which includes 756 voice recordings from 188 PD patients and 64 healthy individuals. Seven performance metrics—accuracy, precision, sensitivity, specificity, F1 score, AUC, and MCC—were utilized for model evaluation. Various techniques, such as Bag Over-Sampling (BOS), cross-validation, and grid search, were implemented to enhance classification performance. Among the four aggregation methods, post-mean aggregation combined with XGBoost achieved the highest accuracy (0.880), F1 score (0.922), and MCC (0.672). Furthermore, we identified potential trends in selecting aggregation methods that are suitable for imbalanced data, particularly based on their differences in sensitivity and specificity. These findings provide meaningful implications for the further exploration of grouped imbalanced data.","author":[{"family":"Yang","given":"Zhengxiao"},{"family":"Zhou","given":"Hao"},{"family":"Srivastav","given":"Sudesh"},{"family":"Shaffer","given":"Jeffrey"},{"family":"Abraham","given":"KT"},{"family":"Naandam","given":"Samuel"},{"family":"Kakraba","given":"Samuel"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/data10010004","URL":"https://doi.org/10.3390/data10010004","source":"openalex"},{"id":"oa:W4407930169","type":"article-journal","title":"The Role of Artificial Intelligence in Early Diagnosis and Management of Cardiovascular Diseases","abstract":"The increasing rate of cardiovascular diseases (CVDs) has posed a tremendous challenge to their early detection and personalized treatment. This research examines the potential of Artificial Intelligence (AI) for early detection and management of CVDs, in particular whether it can enhance diagnostic accuracy, personalize treatment guidelines, and reduce healthcare costs. A quantitative methodology was adopted and a survey strategy was employed for collecting primary data from 300 healthcare professionals consisting of cardiologists, general physicians, and professionals in AI fields from Punjab hospitals in Pakistan. The questionnaire was constructed to determine their knowledge, experiences, and perceptions regarding the use of AI in cardiovascular services. Data analysis revealed that application of AI had a strong correlation with increased diagnostic success, evident in a statistically significant chi-square test (p < 0.001). Furthermore, multiple regression analysis revealed that AI, together with years of experience and educational history, is an important contributor to personalizing cardiovascular treatment plans. The results indicate that AI has a key role in making more precise diagnoses and improving treatment methods, which can ultimately decrease the cost of healthcare and enhance patient outcomes. Yet, issues around data privacy, transparency, and clinician confidence in AI systems must be resolved in order for AI to be adopted more widely. Future research is suggested by the study into the integration of AI with other health technologies and the ethics of using AI in clinical practice.","author":[{"family":"Shabeer","given":"HA"},{"family":"Haider","given":"Hafiz"},{"family":"Khatri","given":"Tamana"},{"family":"Khan","given":"Nouman"},{"family":"Rafique","given":"Adnan"},{"family":"Anjam","given":"Fakhar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.70749/ijbr.v3i2.667","URL":"https://doi.org/10.70749/ijbr.v3i2.667","source":"openalex"},{"id":"oa:W4414939203","type":"article-journal","title":"Artificial intelligence for precision medicine","abstract":"INTRODUCTION: Precision medicine aims to tailor healthcare decisions and interventions to the unique biological and clinical characteristics of each patient. The recent convergence of artificial intelligence (AI) with advances in digital health, omics, and big data analytics has accelerated progress toward this goal. AI technologies - particularly machine learning, deep learning, natural language processing and generative large language models - enable the rapid and meaningful analysis of complex biomedical datasets, supporting more individualized care. PURPOSE OF REVIEW: In this narrative review, we provide an accessible overview of the core principles of AI for healthcare professionals and explore its practical applications across the spectrum of precision medicine. Real-world examples highlight how AI is being used to enhance early diagnosis, guide treatment selection, support disease prevention, and even contribute directly to therapeutic interventions. Alongside these advances, we discuss critical limitations and challenges, including ethical considerations, algorithmic bias, data privacy concerns, environmental impact, and practical barriers to clinical implementation. CONCLUSION: This review offers both an introduction to AI and a practical overview of how it is being used, and where its limitations lie, in precision medicine, with the goal of helping healthcare professionals understand these evolving tools and use them efficiently and responsibly in clinical practice.","author":[{"family":"Martel","given":"M"},{"family":"José-García","given":"Adán"},{"family":"Vens","given":"Celine"},{"family":"Vos","given":"Maarten"},{"family":"Sobanski","given":"Vincent"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.therap.2025.10.003","URL":"https://doi.org/10.1016/j.therap.2025.10.003","source":"openalex"},{"id":"oa:W4407069477","type":"article-journal","title":"Evaluating the Quality and Readability of Generative Artificial Intelligence (AI) Chatbot Responses in the Management of Achilles Tendon Rupture","abstract":"INTRODUCTION: The rise of artificial intelligence (AI), including generative chatbots like ChatGPT (OpenAI, San Francisco, CA, USA), has revolutionized many fields, including healthcare. Patients have gained the ability to prompt chatbots to generate purportedly accurate and individualized healthcare content. This study analyzed the readability and quality of answers to Achilles tendon rupture questions from six generative AI chatbots to evaluate and distinguish their potential as patient education resources. METHODS: The six AI models used were ChatGPT 3.5, ChatGPT 4, Gemini 1.0 (previously Bard; Google, Mountain View, CA, USA), Gemini 1.5 Pro, Claude (Anthropic, San Francisco, CA, USA) and Grok (xAI, Palo Alto, CA, USA) without prior prompting. Each was asked 10 common patient questions about Achilles tendon rupture, determined by five orthopaedic surgeons. The readability of generative responses was measured using Flesch-Kincaid Reading Grade Level, Gunning Fog, and SMOG (Simple Measure of Gobbledygook). The response quality was subsequently graded using the DISCERN criteria by five blinded orthopaedic surgeons. RESULTS: Gemini 1.0 generated statistically significant differences in ease of readability (closest to average American reading level) than responses from ChatGPT 3.5, ChatGPT 4, and Claude. Additionally, mean DISCERN scores demonstrated significantly higher quality of responses from Gemini 1.0 (63.0±5.1) and ChatGPT 4 (63.8±6.2) than ChatGPT 3.5 (53.8±3.8), Claude (55.0±3.8), and Grok (54.2±4.8). However, the overall quality (question 16, DISCERN) of each model was averaged and graded at an above-average level (range, 3.4-4.4). DISCUSSION AND CONCLUSION: Our results indicate that generative chatbots can potentially serve as patient education resources alongside physicians. Although some models lacked sufficient content, each performed above average in overall quality. With the lowest readability and highest DISCERN scores, Gemini 1.0 outperformed ChatGPT, Claude, and Grok and potentially emerged as the simplest and most reliable generative chatbot regarding management of Achilles tendon rupture.","author":[{"family":"Collins","given":"Christopher"},{"family":"Giammanco","given":"Peter"},{"family":"Guirgus","given":"Monica"},{"family":"Kricfalusi","given":"Mikayla"},{"family":"Rice","given":"Richard"},{"family":"Nayak","given":"Rusheel"},{"family":"Ruckle","given":"David"},{"family":"Filler","given":"Ryan"},{"family":"Elsissy","given":"Joseph"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.78313","URL":"https://doi.org/10.7759/cureus.78313","source":"openalex"},{"id":"oa:W4410745942","type":"article-journal","title":"Advances in Artificial Intelligence for Lung Cancer Detection and Diagnostic Accuracy: A Comprehensive Review","abstract":"Cancer is a deadly disease with a minimal probability of curing when detected in the later stages. Out of many different varieties of cancer diseases, lung cancer falls under the critical category as it is internal, invisible, and connected to the breathing control mechanism of human beings. Lung cancer has become the most frequent type in this generation and has intensified in the post-COVID-19 pandemic era, with a high degree of fatality. The histopathological images of lung cancer are very large in volume, so that analyzing such kind of data is monotonous and error-prone in a human- controlled method. The recent progress in the field of computer vision, along with machine learning techniques, has made the path of research smooth in the medical and healthcare domain, also achieved the feasibility of data analysis with easy detection of cancer cells. The advent of the deep learning concept made the automatic and accurate detection of cancer cells from the histopathological image data analysis possible. In this paper, an evaluation and a methodological survey on Cancer cell detection and the accuracy benchmark of Cancer tissue Segmentation of whole slide images (WSI) using Machine Learning (ML) and Deep Learning (DL) approaches have been carried out. The critical analysis, along with the exploration of more probable research trends in the accurate interpretation of the cancer cell images, has also been addressed towards achieving the greater potential.","author":[{"family":"Debnath","given":"Rupa"},{"family":"Mondal","given":"Rituparna"},{"family":"Chakraborty","given":"Arpita"},{"family":"Chatterjee","given":"Siddhartha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.38124/ijisrt/25may1339","URL":"https://doi.org/10.38124/ijisrt/25may1339","source":"openalex"},{"id":"oa:W4409607162","type":"article-journal","title":"Artificial Intelligence in Ovarian Cancer: A Systematic Review and Meta-Analysis of Predictive AI Models in Genomics, Radiomics, and Immunotherapy","abstract":"Background/Objectives: Artificial intelligence (AI) is increasingly influencing oncological research by enabling precision medicine in ovarian cancer through enhanced prediction of therapy response and patient stratification. This systematic review and meta-analysis was conducted to assess the performance of AI-driven models across three key domains: genomics and molecular profiling, radiomics-based imaging analysis, and prediction of immunotherapy response. Methods: Relevant studies were identified through a systematic search across multiple databases (2020–2025), adhering to PRISMA guidelines. Results: Thirteen studies met the inclusion criteria, involving over 10,000 ovarian cancer patients and encompassing diverse AI models such as machine learning classifiers and deep learning architectures. Pooled AUCs indicated strong predictive performance for genomics-based (0.78), radiomics-based (0.88), and immunotherapy-based (0.77) models. Notably, radiogenomics-based AI integrating imaging and molecular data yielded the highest accuracy (AUC = 0.975), highlighting the potential of multi-modal approaches. Heterogeneity and risk of bias were assessed, and evidence certainty was graded. Conclusions: Overall, AI demonstrated promise in predicting therapeutic outcomes in ovarian cancer, with radiomics and integrated radiogenomics emerging as leading strategies. Future efforts should prioritize explainability, prospective multi-center validation, and integration of immune and spatial transcriptomic data to support clinical implementation and individualized treatment strategies. Unlike earlier reviews, this study synthesizes a broader range of AI applications in ovarian cancer and provides pooled performance metrics across diverse models. It examines the methodological soundness of the selected studies and highlights current gaps and opportunities for clinical translation, offering a comprehensive and forward-looking perspective in the field.","author":[{"family":"Maiorano","given":"Mauro"},{"family":"Cormio","given":"Gennaro"},{"family":"Loizzi","given":"Vera"},{"family":"Maiorano","given":"Brigida"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ai6040084","URL":"https://doi.org/10.3390/ai6040084","source":"openalex"},{"id":"oa:W4414342860","type":"article-journal","title":"Analysis of Retracted Publications on Artificial Intelligence: Trends, Ethical Concerns, and Scientific Integrity","abstract":"BACKGROUND: Artificial intelligence (AI) has promoted progress across various fields. The number of papers regarding AI has risen in recent years. This study examines retracted publications regarding AI by analyzing trends, journals, and reasons. METHODS: This descriptive cross-sectional study thoroughly investigated retracted AI-related papers listed in PubMed. The data extraction comprised bibliographic data, reasons for retraction, citation metrics, journal indexing status, and Altmetric Attention Scores (AASs). Retraction notices were classified according to particular reasons. Descriptive statistics were employed to evaluate retraction trends, geographic distribution, and citation impact. RESULTS: A total of 764 retracted AI-related papers were examined, with the most retractions occurring in 2023 (n = 667). China had the highest number (n = 551), followed by India (n = 40) and Bangladesh (n = 23). Journals focusing on mathematical and computational biology, neurosciences, and healthcare sciences had the most retractions. The most common retraction reasons were peer review issues (n = 716) and data concerns (n = 714), followed by irrelevant citations (n = 571) and unethical AI use (n = 238). The median time to retraction was 510 days (18-4,200). The median citation and AAS scores were (0-167) and 0 (0-191). CONCLUSION: The high number of retractions from China highlights the need for higher research standards. Deficits in peer review and data issues emerged as the main reasons for retraction, underscoring persistent challenges in maintaining research integrity and quality assurance. For scientific literature integrity, academic institutions, publishers, and researchers should stress transparency, ethics, and rigorous post-publication inspection.","author":[{"family":"Koçyiğit","given":"Burhan"},{"family":"Okyay","given":"Ramazan"},{"family":"Seiil","given":"Birzhan"},{"family":"Qumar","given":"Ainur"},{"family":"Sümbül","given":"Hilmi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3346/jkms.2025.40.e280","URL":"https://doi.org/10.3346/jkms.2025.40.e280","source":"openalex"},{"id":"oa:W4409482562","type":"article-journal","title":"Multimodality imaging in prostate cancer diagnosis using artificial intelligence: basic concepts and current state-of-the-art","abstract":"Abstract The early diagnosis of prostate cancer (PCa) is highly recommended, as the tumor will not spread to other organs of the body and the bones. Moreover, a late diagnosis of PCa could lower the survival rate. The growing development of Artificial Intelligence (AI) and Machine Learning (ML) in medical images has led to significant improvement in PCa diagnosis. Multimodality is now commonly applied in medical imaging diagnosis, as it provides comprehensive information about a target (tissue or tumor). It has shown to be useful for advancing the clinical reliability of using medical images and ML for medical diagnostics and analysis. Hence, in this paper, a comprehensive survey is provided to explore the state-of-the-art Computer-Aided Diagnosis Systems (CADs) for PCa detection attributed to multimodality imaging, a background of PCa. different types of medical imaging used in PCa diagnosis, related clinical workflows, future perspectives, and some common limitations of related work. The review exhibits an extensive literature review done on multimodality imaging in PCa, highlights that multimodality imaging has the potential of wide applicability in diagnosis systems. It is expected that this study enhances the understanding necessary for developing CAD systems for PCa diagnosis. Additionally, it is expected to establish a great basis for developing multimodal images, the relevant datasets, some of the challenges, and future topics. Graphical Abstract","author":[{"family":"Ayyad","given":"Sarah"},{"family":"Abdel-Hamid","given":"Nahla"},{"family":"Ali","given":"Hesham"},{"family":"Labib","given":"Labib"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s11042-025-20786-2","URL":"https://doi.org/10.1007/s11042-025-20786-2","source":"openalex"},{"id":"oa:W4416459697","type":"article-journal","title":"Hybrid intelligence in medical image segmentation","abstract":"Medical image segmentation is vital for precise identification and analysis of anatomical structures and pathological regions, yet traditional models often fall short in aligning with clinical workflows, requiring extensive manual correction even when overall segmentation accuracy is high. To address this gap, we introduce HybridMS, a hybrid intelligence framework designed to maintain high segmentation accuracy while substantially reducing clinician workload through selective human intervention. HybridMS employs an uncertainty-driven feedback mechanism that selectively triggers clinician input only for cases predicted to be challenging, thereby avoiding unnecessary manual review. Corrected cases are prioritised during retraining through a weighted update strategy, enabling the model to adapt more effectively to clinically relevant errors. This design minimises intervention frequency while preserving segmentation quality. Evaluated on lung segmentation in chest X-rays for tuberculosis detection, HybridMS achieved comparable or improved performance over the baseline MedSAM model (Dice: 0.9538 vs. 0.9435; IoU: 0.9126 vs. 0.8941) with consistent boundary quality in difficult cases. For the subset of cases identified as challenging (baseline Dice < 0.92), HybridMS reduced mean Hausdorff Distance and Average Symmetric Surface Distance, demonstrating more stable anatomical boundaries. Workflow efficiency was markedly improved: in a preliminary timing study with radiologists, average annotation time was reduced by approximately 82% for standard cases and 60% for challenging cases, without compromising accuracy. By combining targeted human oversight with automated refinement, HybridMS demonstrates that stable segmentation performance can be achieved with significantly lower annotation effort, offering a clinically viable pathway for efficient and reliable deployment in diagnostic workflows.","author":[{"family":"Ali","given":"Namia"},{"family":"Oyelere","given":"Solomon"},{"family":"Jitani","given":"Nitya"},{"family":"Sarmah","given":"Rosy"},{"family":"Andrew","given":"S"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-24990-w","URL":"https://doi.org/10.1038/s41598-025-24990-w","source":"openalex"},{"id":"oa:W4411475682","type":"article-journal","title":"Algorithmic Bias and Data Justice: ethical challenges in Artificial Intelligence Systems","abstract":"This article examines the critical ethical challenges posed by algorithmic bias in artificial intelligence (AI) systems, focusing on its implications for social justice and data equity. Through a systematic review of case studies and theoretical frameworks, we analyze how biased datasets and algorithmic designs perpetuate structural inequalities, particularly affecting marginalized communities. The study highlights key examples, such as gender and racial biases in facial recognition and hiring algorithms, while exploring mitigation strategies rooted in data justice principles. Additionally, we evaluate regulatory responses, including the European Union's AI Act, which proposes a risk-based governance framework. The findings underscore the urgent need for interdisciplinary approaches to develop fairer AI systems that align with ethical standards and human rights.","author":[{"family":"Gonzálezargote","given":"Javier"},{"family":"Maldonado","given":"Emanuel"},{"family":"Maldonado","given":"Karina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.56294/ai2025159","URL":"https://doi.org/10.56294/ai2025159","source":"openalex"},{"id":"oa:W4412883239","type":"article-journal","title":"Application of artificial intelligence techniques for the profiling of visitors to tourist destinations","abstract":"Tourism in Peru represents an opportunity for local development; however, there is limited understanding of visitor profiles. The aim of this study was to characterize tourists using machine learning techniques in order to identify distinct segments that can inform planning and promotional strategies for the Alto Amazonas destination. The research followed the CRISP-DM methodology for data analysis, based on surveys administered to 882 visitors. The data were processed using the clustering algorithms K-Means, DBSCAN, HDBSCAN, and Agglomerative, with Principal Component Analysis applied beforehand for dimensionality reduction. The results showed that the Agglomerative Clustering model achieved the best performance in internal validation metrics, allowing for the identification of five distinct visitor profiles. These segments provide valuable insights for the design of more inclusive and personalized tourism products. In conclusion, the study demonstrates the value of machine learning as a tool for tourism segmentation, offering empirical evidence that can strengthen the management of emerging destinations such as Alto Amazonas. The practical contribution of this study lies in providing strategic information that enables destination managers to tailor services and experiences to the characteristics of each segment, thereby optimizing visitor satisfaction and strengthening the destination's competitiveness.","author":[{"family":"Schrader","given":"Juan"},{"family":"Pinedo","given":"Lloy"},{"family":"Vargas","given":"Fabiano"},{"family":"Martell","given":"Karla"},{"family":"Seijas-Díaz","given":"José"},{"family":"Amasifen","given":"Mtro"},{"family":"Orbe","given":"Rosa"},{"family":"Silva","given":"Mg"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/frai.2025.1632415","URL":"https://doi.org/10.3389/frai.2025.1632415","source":"openalex"},{"id":"oa:W4411087970","type":"article-journal","title":"Generative artificial intelligence for general practice; new potential ahead, but are we ready?","abstract":"BACKGROUND: Generative AI (Gen AI) is frequently cited as an innovation to address the current challenges in healthcare, also for primary care. Examples include automating tasks like voice-to-notes transcription or chatbots using large language models. Additionally, it may facilitate a learning healthcare system by generating personalised learning resources and real-time literature summaries. Yet - probably with the highest expectations - Gen AI may extend diagnostic and therapeutic capabilities in general practice by integrating complex, multimodal patient data for personalised care, enabling earlier disease detection, and providing real-time guidance for diagnostics, prognostics and treatments. METHOD & DISCUSSION: The authors of this opinion paper recently hosted a workshop at the WONCA Europe 2024 conference. From discussions at that workshop, three priorities emerge: practice support, education support, and clinical decision-making support. In this opinion paper, we argue that GPs and academic departments of primary care should lead in evaluating Gen AI across these three priorities. Primary care research must prioritise rigorous scientific evaluations, to ensure that developed tools actually work for GPs and their patients. CONCLUSION: Hereto, a coordinated effort, driven by the primary care academic community, is needed, starting with research agenda drafting. A broad, international follow-up is scheduled following this WONCA Europe 2024 workshop.","author":[{"family":"Geersing","given":"Geert‐jan"},{"family":"Wit","given":"Niek"},{"family":"Thompson","given":"Matthew"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/13814788.2025.2511645","URL":"https://doi.org/10.1080/13814788.2025.2511645","source":"openalex"},{"id":"oa:W4410236561","type":"article-journal","title":"Wearable sleep recording augmented by artificial intelligence for Alzheimer’s disease screening","abstract":"The recent emergence of wearable devices will enable large scale remote brain monitoring. This study investigated whether multimodal wearable sleep recordings could help screening for Alzheimer's disease (AD). Measurements were acquired simultaneously from polysomnography and a wearable device, measuring electroencephalography (EEG) and accelerometry (ACM) in 67 elderly without cognitive symptoms and 35 AD patients. Sleep staging was performed using an AI model (SeqSleepNet), followed by feature extraction from hypnograms and physiological signals. Using these features, a multi-layer perceptron was trained for AD detection, with elastic net identifying key features. The wearable AD detection model achieved an accuracy of 0.90 (0.76 for prodromal AD). Single-channel EEG and ACM physiological features captured sufficient information for AD detection and outperformed the hypnogram features, highlighting these physiological features as promising discriminative markers for AD. We conclude that wearable sleep monitoring augmented by AI shows promise towards non-invasive screening for AD in the older population.","author":[{"family":"Heremans","given":"Elisabeth"},{"family":"Devulder","given":"Astrid"},{"family":"Borzée","given":"Pascal"},{"family":"Vandenberghe","given":"Rik"},{"family":"Winter","given":"François‐laurent"},{"family":"Vandenbulcke","given":"Mathieu"},{"family":"Bossche","given":"Maarten"},{"family":"Buyse","given":"Bertien"},{"family":"Testelmans","given":"Dries"},{"family":"Paesschen","given":"Wim"},{"family":"Vos","given":"Maarten"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41514-025-00219-y","URL":"https://doi.org/10.1038/s41514-025-00219-y","source":"openalex"},{"id":"oa:W4407703947","type":"article-journal","title":"Accuracy of artificial intelligence in detecting tumor bone metastases: a systematic review and meta-analysis","abstract":"BACKGROUND: Bone metastases (BM) represent a prevalent complication of tumors. Early and accurate diagnosis, however, is a significant hurdle for radiologists. Recently, artificial intelligence (AI) has emerged as a valuable tool to assist radiologists in the detection of BM. This meta-analysis was undertaken to evaluate the AI diagnostic accuracy for BM. METHODS: Two reviewers performed an exhaustive search of several databases, including Wei Pu (VIP) database, China National Knowledge Infrastructure (CNKI), Web of Science, Cochrane Library, Ovid-Embase, Ovid-Medline, Wan Fang database, and China Biology Medicine (CBM), from their inception to December 2024. This search focused on studies that developed and/or validated AI techniques for detecting BM in magnetic resonance imaging (MRI) or computed tomography (CT). A hierarchical model was used in the meta-analysis to calculate diagnostic odds ratio (DOR), negative likelihood ratio (NLR), positive likelihood ratio (PLR), area under the curve (AUC), specificity (SP), and pooled sensitivity (SE). The risk of bias and applicability were assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST), while the Transparent Reporting of a multivariable prediction model for individual prognosis or diagnosis-artificial intelligence (TRIPOD-AI) was employed for evaluating the quality of evidence. RESULT: This review covered 20 articles, among them, 16 studies were included in the meta-analysis. The results revealed a pooled SE of 0.88 (0.82-0.92), a pooled SP of 0.89 (0.84-0.93), a pooled AUC of 0.95 (0.92-0.96), PLR of 8.1 (5.57-11.80), NLR of 0.14 (0.09-0.21) and DOR of 58 (31-109). When focusing on imaging algorithms. Based on ML, a pooled SE of 0.88 (0.77-0.92), SP 0.88 (0.82-0.92), and AUC 0.93 (0.91-0.95). Based on DL, a pooled SE of 0.89 (0.81-0.95), SP 0.89 (0.81-0.94), and AUC 0.95 (0.93-0.97). CONCLUSION: This meta-analysis underscores the substantial diagnostic value of AI in identifying BM. Nevertheless, in-depth large-scale prospective research should be carried out for confirming AI's clinical utility in BM management.","author":[{"family":"Tao","given":"Huimin"},{"family":"Xu","given":"Hui"},{"family":"Zhang","given":"Zhi"},{"family":"Zhu","given":"Rongrong"},{"family":"Wang","given":"Ping"},{"family":"Zhou","given":"Sheng"},{"family":"Yang","given":"Kehu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12885-025-13631-0","URL":"https://doi.org/10.1186/s12885-025-13631-0","source":"openalex"},{"id":"oa:W4406555789","type":"article-journal","title":"Leveraging Artificial Intelligence for Advancing Key Sectors of National Growth and Development","abstract":"This study explores the transformative potential of artificial intelligence (AI) in national development across key sectors, including economic growth, healthcare, education, infrastructure, and security. Using a qualitative doctrinal research approach, it synthesizes insights from peer-reviewed literature, policy analyses, and case studies to assess AI's contributions and challenges in fostering sustainable progress. Key findings highlight AI’s role in driving economic productivity through automation and innovation while underscoring its impact on the workforce and job market transformation. In healthcare, AI enhances diagnostics, treatment planning, and public health interventions, though concerns about data privacy and algorithmic bias persist. Education systems benefit from AI-enabled personalized learning and skill development, preparing future-ready workforces. Additionally, AI supports sustainable infrastructure through smart city initiatives, improving resource management and urban planning. Despite these advancements, the study identifies challenges such as ethical dilemmas, digital divides, and governance gaps that hinder equitable AI adoption. Recommendations include establishing transparent regulatory frameworks, fostering international collaboration, and investing in digital literacy to ensure inclusive growth. By addressing these issues, AI can become a cornerstone of national development, contributing to the United Nations' Sustainable Development Goals. This research underscores the importance of balanced policies and proactive governance to harness AI's benefits while mitigating associated risks.","author":[{"family":"Ogunseye","given":"Olusola"},{"family":"Ajayi","given":"Ola"},{"family":"Fabusoro","given":"Adetutu"},{"family":"Abba","given":"Amina"},{"family":"Adepoju","given":"Benjamin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.56557/ajocr/2025/v10i19056","URL":"https://doi.org/10.56557/ajocr/2025/v10i19056","source":"openalex"},{"id":"oa:W4414585996","type":"article-journal","title":"Enhancing Customer Engagement Through Artificial Intelligence Authenticity","abstract":"ABSTRACT Given the limited research on the factors and mechanisms underlying artificial intelligence (AI) authenticity, we examine its use in fostering breakthrough knowledge and enhancing customer engagement. We devised a robust model grounded in mind perception and social exchange theories, with a focus on the outcomes of AI authenticity. Tested across 452 virtual health home stations, the findings reveal that both performance expectation and effort expectation serve as mediators between AI authenticity and customer engagement. This research provides managers with comprehensive insights into the defining attributes and operational mechanics of AI authenticity, thereby highlighting its critical importance in boosting customer engagement.","author":[{"family":"Foroudi","given":"Pantea"},{"family":"Robson","given":"Matthew"},{"family":"Marvi","given":"Reza"},{"family":"Spyropoulou","given":"Stavroula"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/jpim.70008","URL":"https://doi.org/10.1111/jpim.70008","source":"openalex"},{"id":"oa:W4413766334","type":"article-journal","title":"Modelling STEM students’ intention to learn artificial intelligence (AI) in Ghana: a PLS-SEM and fsQCA approach","abstract":"Artificial intelligence (AI) is globally transforming industries, demanding AI literacy for future Science, Technology, Engineering and Mathematics (STEM) professionals. Ghana is adapting its educational and technological landscape to meet this need. This study investigates factors influencing Ghanaian STEM students’ intentions to learn AI, providing insights for educational policy. A descriptive cross-sectional survey of 233 AI-familiar STEM students examined subjective norm, facilitating conditions, self-efficacy, perceived usefulness, perceived social good, AI anxiety, AI literacy, and career relevance. Using partial least squares structural equation modelling and fuzzy-set qualitative comparative analysis (fsQCA), the analysis revealed multiple combinations of individual and contextual factors that drive students’ intentions to learn AI. Results indicate that subjective norm, facilitating conditions, self-efficacy, social good, AI literacy, and career relevance positively influence AI learning intentions. Also, AI anxiety negatively influence AI learning intentions. fsQCA revealed six configurations for high intention. Overall, individual and contextual factors shape Ghanaian STEM students’ AI learning intentions. Fostering AI literacy and career relevance boosts motivation. We recommend integrating AI literacy into STEM curriculum and increasing AI resource access as a national policy priority.","author":[{"family":"Abreh","given":"Might"},{"family":"Arthur","given":"Francis"},{"family":"Akwetey","given":"Freda"},{"family":"Nortey","given":"Sharon"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s44163-025-00466-8","URL":"https://doi.org/10.1007/s44163-025-00466-8","source":"openalex"},{"id":"oa:W4406703007","type":"article-journal","title":"Artificial Intelligence in Pediatric Epilepsy Detection: Balancing Effectiveness With Ethical Considerations for Welfare","abstract":"Background and Aim: Epilepsy is a major neurological challenge, especially for pediatric populations. It profoundly impacts both developmental progress and quality of life in affected children. With the advent of artificial intelligence (AI), there's a growing interest in leveraging its capabilities to improve the diagnosis and management of pediatric epilepsy. This review aims to assess the effectiveness of AI in pediatric epilepsy detection while considering the ethical implications surrounding its implementation. Methodology: A comprehensive systematic review was conducted across multiple databases including PubMed, EMBASE, Google Scholar, Scopus, and Medline. Search terms encompassed \"pediatric epilepsy,\" \"artificial intelligence,\" \"machine learning,\" \"ethical considerations,\" and \"data security.\" Publications from the past decade were scrutinized for methodological rigor, with a focus on studies evaluating AI's efficacy in pediatric epilepsy detection and management. Results: AI systems have demonstrated strong potential in diagnosing and monitoring pediatric epilepsy, often matching clinical accuracy. For example, AI-driven decision support achieved 93.4% accuracy in diagnosis, closely aligning with expert assessments. Specific methods, like EEG-based AI for detecting interictal discharges, showed high specificity (93.33%-96.67%) and sensitivity (76.67%-93.33%), while neuroimaging approaches using rs-fMRI and DTI reached up to 97.5% accuracy in identifying microstructural abnormalities. Deep learning models, such as CNN-LSTM, have also enhanced seizure detection from video by capturing subtle movement and expression cues. Non-EEG sensor-based methods effectively identified nocturnal seizures, offering promising support for pediatric care. However, ethical considerations around privacy, data security, and model bias remain crucial for responsible AI integration. Conclusion: While AI holds immense potential to enhance pediatric epilepsy management, ethical considerations surrounding transparency, fairness, and data security must be rigorously addressed. Collaborative efforts among stakeholders are imperative to navigate these ethical challenges effectively, ensuring responsible AI integration and optimizing patient outcomes in pediatric epilepsy care.","author":[{"family":"Mourid","given":"Marina"},{"family":"Irfan","given":"Hamza"},{"family":"Oduoye","given":"Malik"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/hsr2.70372","URL":"https://doi.org/10.1002/hsr2.70372","source":"openalex"},{"id":"oa:W4410517780","type":"article-journal","title":"Development and validation of an artificial intelligence-based pipeline for predicting oral epithelial dysplasia malignant transformation","abstract":"BACKGROUND: Oral epithelial dysplasia (OED) is a potentially malignant histopathological diagnosis given to lesions of the oral cavity that are at risk of progression to malignancy. Manual grading of OED is subject to substantial variability and does not reliably predict prognosis, potentially resulting in sub-optimal treatment decisions. METHOD: We developed a Transformer-based artificial intelligence (AI) pipeline for the prediction of malignant transformation from whole-slide images (WSIs) of Haematoxylin and Eosin (H&E) stained OED tissue slides, named ODYN (Oral Dysplasia Network). ODYN can simultaneously classify OED and assign a predictive score (ODYN-score) to quantify the risk of malignant transformation. The model was trained on a large cohort using three different scanners (Sheffield, 358 OED WSIs, 105 control WSIs) and externally validated on cases from three independent centres (Birmingham and Belfast, UK, and Piracicaba, Brazil; 108 OED WSIs). RESULTS: Model testing yielded an F1-score of 0.96 for classification of dysplastic vs non-dysplastic slides, and an AUROC of 0.73 for malignancy prediction, gaining comparable results to clinical grading systems. CONCLUSIONS: With further large-scale prospective validation, ODYN promises to offer an objective and reliable solution for assessing OED cases, ultimately improving early detection and treatment of oral cancer.","author":[{"family":"Shephard","given":"Adam"},{"family":"Mahmood","given":"Hanya"},{"family":"Raza","given":"Shan"},{"family":"Araújo","given":"Anna"},{"family":"Santossilva","given":"Alan"},{"family":"Lopes","given":"Márcio"},{"family":"Vargas","given":"Pablo"},{"family":"Mccombe","given":"Kris"},{"family":"Craig","given":"Stephanie"},{"family":"James","given":"Jacqueline"},{"family":"Brooks","given":"Jill"},{"family":"Nankivell","given":"Paul"},{"family":"Mehanna","given":"Hisham"},{"family":"Khurram","given":"Syed"},{"family":"Rajpoot","given":"Nasir"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s43856-025-00873-z","URL":"https://doi.org/10.1038/s43856-025-00873-z","source":"openalex"},{"id":"oa:W4414320896","type":"article-journal","title":"Artificial intelligence accelerates the interpretation of measurable residual B lymphoblastic leukemia by flow cytometry","abstract":"ABSTRACT: Measurable residual disease (MRD) assessment by flow cytometry (FC) plays an essential role in prognosis and therapy escalation of B-cell acute lymphoblastic leukemia (B-ALL). However, the high degree of expertise and manual analysis time required limits the availability of this assay. To overcome this limitation, we developed a data-enhancing artificial intelligence (AI) pipeline that accelerates and simplifies MRD analysis. Unaltered FC files from 171 B-ALL MRD-positive and 89 MRD-negative cases were processed through an AI pipeline trained with 31 expert-gated negative controls. Cluster-informed downsampling reduced FC files from 1.2 million to 155 884 cells per case, on average, (87% cellularity reduction), whereas preserving small MRD populations (median, 100% retention for MRD of <1%) and allowing for true percentage MRD estimates using a correction factor. A deep neural network cell classifier automatically identified normal hematopoietic subsets (macro-averaged F1 score of 0.86); and an AI measure of anomaly discriminated B-ALL from benign mononuclear (area under the curve [AUC] of 0.98) or B-lymphoid cells (AUC of 0.94). Manual analysis of AI-enhanced files was completed in only 1.01 minutes per case, on average (standard deviation of ±0.57); with 100% positive agreement with conventional analysis (for MRD of ≥0.01%), 100% negative agreement, and excellent quantitative correlation (R2 = 0.92). Our cloud-based AI-enhancement solution accelerates B-ALL MRD identification without compromising test performance and has the potential of facilitating B-ALL MRD analysis by more clinical laboratories.","author":[{"family":"Seheult","given":"Jansen"},{"family":"Otteson","given":"Gregory"},{"family":"Timm","given":"Michael"},{"family":"Weybright","given":"Matthew"},{"family":"Shi","given":"Min"},{"family":"Olteanu","given":"Horatiu"},{"family":"Jevremović","given":"Dragan"},{"family":"Chen","given":"Chuan"},{"family":"Chiu","given":"April"},{"family":"Horna","given":"Pedro"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1182/bloodadvances.2025016126","URL":"https://doi.org/10.1182/bloodadvances.2025016126","source":"openalex"},{"id":"oa:W4411194802","type":"article-journal","title":"Enhancing Dental Students' History‐Taking Skills With a Generative Artificial Intelligence Chatbot","abstract":"PURPOSE: Accurate history-taking is crucial for diagnosis and treatment planning in dentistry yet there is insufficient time in pre-clinical years to practice this essential skill. This study evaluated dental student motivations and enjoyment using a generative artificial intelligence (AI) chatbot to practice these skills. METHODS: In 2024, first- and second-year Doctor of Dental Medicine students trialed a virtual patient (VP) chatbot using generative AI based on GPT-4 and then completed surveys to assess expectations, motivations, and feedback on the chatbot and compare their experience with in-person learning. RESULTS: A total of 31 students participated in the trial. The cohort saw a significant improvement in perceived competence after using the chatbot. Second-year students found more value in the chatbot than first-year students. While the students felt the chatbot increased their participation and provided more practice opportunities, most found it less interesting compared to in-person teaching. Overall, students were supportive of the integration of the chatbot into the curriculum. CONCLUSION: Educational chatbots can act as VPs, allowing users to practice patient history-taking. This chatbot was seen as a beneficial, supplementary resource for pre-clinical dental students, allowing them to supplement the currently limited opportunities to practice this skill.","author":[{"family":"Or","given":"Aidan"},{"family":"Sukumar","given":"Smitha"},{"family":"Ma","given":"Andrew"},{"family":"Ang","given":"Dong"},{"family":"Liu","given":"Max"},{"family":"Ritchie","given":"Helen"},{"family":"Sarrafpour","given":"Babak"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/jdd.13952","URL":"https://doi.org/10.1002/jdd.13952","source":"openalex"},{"id":"oa:W4410028173","type":"article-journal","title":"Small language models learn enhanced reasoning skills from medical textbooks","abstract":"Small language models (SLM) offer promise for medical applications by addressing the privacy and hardware constraints of large language models; however, their limited parameters (often fewer than ten billion) hinder multi-step reasoning for complex medical tasks. This study presents Meerkat, a new family of medical SLMs designed to be lightweight while enhancing reasoning capabilities. We begin by designing an effective and efficient training method. This involves extracting high-quality chain-of-thought reasoning paths from 18 medical textbooks, which are then combined with diverse instruction-following datasets within the medical domain, totaling 441K training examples. Fine-tuning was conducted on open-source SLMs using this curated dataset. Our Meerkat-7B and Meerkat-8B models outperformed their counterparts by 22.3% and 10.6% across six exam datasets, respectively. They also improved scores on the NEJM Case Challenge from 7 to 16 and from 13 to 20, surpassing the human score of 13.7. Additionally, they demonstrated superiority in expert evaluations, excelling in all metrics-completeness, factuality, clarity, and logical consistency-of reasoning abilities.","author":[{"family":"Kim","given":"Hyunjae"},{"family":"Hwang","given":"Hyeon"},{"family":"Lee","given":"Ji"},{"family":"Park","given":"Sihyeon"},{"family":"Kim","given":"Dain"},{"family":"Lee","given":"Taewhoo"},{"family":"Yoon","given":"Chanwoong"},{"family":"Sohn","given":"Jiwoong"},{"family":"Park","given":"Jungwoo"},{"family":"Reykhart","given":"Olga"},{"family":"Fetherston","given":"Thomas"},{"family":"Choi","given":"Donghee"},{"family":"Kwak","given":"Soo"},{"family":"Chen","given":"Qingyu"},{"family":"Kang","given":"Jaewoo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01653-8","URL":"https://doi.org/10.1038/s41746-025-01653-8","source":"openalex"},{"id":"oa:W4414510983","type":"article-journal","title":"AI policy in healthcare: a checklist-based methodology for structured implementation","abstract":"INTRODUCTION: Artificial Intelligence (AI) is transforming anaesthesia and intensive care medicine, enhancing diagnostic precision, workflow efficiency, and patient safety. However, deploying AI in high-acuity environments involves regulatory, ethical, and operational challenges. The European Union Artificial Intelligence Act (AI Act), effective 2025, imposes binding obligations on healthcare organizations, creating an urgent need for structured, governance-focused AI policies. This work presents a checklist-based methodology for responsible, safe, ethical, and regulation-aligned AI adoption in clinical units. THE NEED FOR A METHODOLOGY TO DEVELOP AN AI POLICY: Effective AI policies must ensure transparency, safety, fairness, and regulatory compliance while remaining adaptable to rapid technological and legislative changes. The proposed methodology employs a domain-specific checklist to generate critical evaluative questions, enabling healthcare professionals to systematically assess AI systems' appropriateness, reliability, and legal implications without relying on rigid, quickly outdated prescriptive rules. THE AI ACT AND ITS RELEVANCE: Regulation (EU) 2024/1689 establishes the first comprehensive AI legal framework, introducing risk-based classification, imposing stringent requirements for high-risk AI, often including medical devices. Compliance obligations extend to both AI-system providers and deployers, making operational compliance instruments and AI literacy programmes essential for lawful implementation. AI LITERACY: OBLIGATION AND PLANNING: From February 2025, the AI Act mandates AI literacy for all personnel interacting with AI-systems. Training should cover baseline competencies for all staff, advanced modules for specialists, continuous professional development, and integration of ethical, legal, and governance principles. Competency acquisition and updates must be systematically documented to meet institutional and EU compliance standards. OPERATIONAL CHECKLIST FOR THE ADOPTION OF AI POLICY: The checklist has two integrated domains: clinical and technical validation, including evidence-based performance assessment, real-world validation, MDR compliance, GDPR adherence, and post-deployment monitoring; and governance and compliance, covering AI Act conformity, organizational accountability, decision traceability, human oversight, AI literacy, and structured audit and update mechanisms. FUTURE PERSPECTIVES: The checklist methodology offers a scalable, adaptable, regulation-ready framework for AI policy development. By embedding legal compliance, clinical safety, governance, and continuous staff training, it supports sustainable AI integration. Future updates will incorporate regulatory changes, real-world feedback, and impact metrics, enhancing AI's contribution to quality, safety, and equity in patient care.","author":[{"family":"Bignami","given":"Elena"},{"family":"Darhour","given":"Luigino"},{"family":"Franco","given":"Gabriele"},{"family":"Guarnieri","given":"Matteo"},{"family":"Bellini","given":"Valentina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s44158-025-00278-3","URL":"https://doi.org/10.1186/s44158-025-00278-3","source":"openalex"},{"id":"oa:W4411494854","type":"article-journal","title":"Current and future applications of artificial intelligence in lung cancer and mesothelioma","abstract":"BACKGROUND: Considerable challenges exist in managing lung cancer and mesothelioma, including diagnostic complexity, treatment stratification, early detection and imaging quantification. Variable incidence in mesothelioma also makes equitable provision of high-quality care difficult. In this context, artificial intelligence (AI) offers a range of assistive/automated functions that can potentially enhance clinical decision-making, while reducing inequality and pathway delay. AIMS: In this state-of-the-art narrative review, we synthesise evidence on this topic, focusing particularly on tools that ingest routine pathology and radiology images. We summarise the strengths and weaknesses of AI applied to common multidisciplinary team (MDT) functions, including histological diagnosis, therapeutic response prediction, radiological detection and quantification, and survival estimation. We also review emerging methods capable of generating novel biological insights and current barriers to implementation, including access to high-quality training data and suitable regulatory and technical infrastructure. NARRATIVE: Neural networks trained on pathology images have proven utility in histological classification, prognostication, response prediction and survival. Self-supervised models can also generate new insights into biological features responsible for adverse outcomes. Radiology applications include lung nodule tools, which offer critical pathway support for imminent lung cancer screening and urgent referrals. Tumour segmentation AI offers particular advantages in mesothelioma, where response assessment and volumetric staging are difficult using human readers due to tumour size and morphological complexity. AI is also critical for radiogenomics, permitting effective integration of molecular and radiomic features for discovery of non-invasive markers for molecular subtyping and enhanced stratification. CONCLUSIONS: AI solutions offer considerable potential benefits across the MDT, particularly in repetitive or time-consuming tasks based on pathology and radiology images. Effective leveraging of this technology is critical for lung cancer screening and efficient delivery of increasingly complex diagnostic and predictive MDT functions. Future AI research should involve transparent and interpretable outputs that assist in explaining the basis of AI-supported decision making.","author":[{"family":"Roche","given":"Joshua"},{"family":"Seyedshahi","given":"Farzaneh"},{"family":"Rakovic","given":"Kai"},{"family":"Thu","given":"Akari"},{"family":"Quesne","given":"John"},{"family":"Blyth","given":"Kevin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/thorax-2024-222054","URL":"https://doi.org/10.1136/thorax-2024-222054","source":"openalex"},{"id":"oa:W4415106116","type":"article-journal","title":"Artificial Intelligence in Cardiac Electrophysiology: A Clinically Oriented Review with Engineering Primers","abstract":"Artificial intelligence (AI) is transforming cardiac electrophysiology across the entire care pathway, from arrhythmia detection on 12-lead electrocardiograms (ECGs) and wearables to the guidance of catheter ablation procedures, through to outcome prediction and therapeutic personalization. End-to-end deep learning (DL) models have achieved cardiologist-level performance in rhythm classification and prognostic estimation on standard ECGs, with a reported arrhythmia classification accuracy of ≥95% and an atrial fibrillation detection sensitivity/specificity of ≥96%. The application of AI to wearable devices enables population-scale screening and digital triage pathways. In the electrophysiology (EP) laboratory, AI standardizes the interpretation of intracardiac electrograms (EGMs) and supports target selection, and machine learning (ML)-guided strategies have improved ablation outcomes. In patients with cardiac implantable electronic devices (CIEDs), remote monitoring feeds multiparametric models capable of anticipating heart-failure decompensation and arrhythmic risk. This review outlines the principal modeling paradigms of supervised learning (regression models, support vector machines, neural networks, and random forests) and unsupervised learning (clustering, dimensionality reduction, association rule learning) and examines emerging technologies in electrophysiology (digital twins, physics-informed neural networks, DL for imaging, graph neural networks, and on-device AI). However, major challenges remain for clinical translation, including an external validation rate below 30% and workflow integration below 20%, which represent core obstacles to real-world adoption. A joint clinical engineering roadmap is essential to translate prototypes into reliable, bedside tools.","author":[{"family":"Canino","given":"Giovanni"},{"family":"Costanzo","given":"Assunta"},{"family":"Salerno","given":"Nadia"},{"family":"Leo","given":"Isabella"},{"family":"Cannataro","given":"Mario"},{"family":"Guzzi","given":"Pietro"},{"family":"Veltri","given":"Pierangelo"},{"family":"Sorrentino","given":"Sabato"},{"family":"Rosa","given":"Salvatore"},{"family":"Torella","given":"Daniele"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bioengineering12101102","URL":"https://doi.org/10.3390/bioengineering12101102","source":"openalex"},{"id":"oa:W7131836311","type":"article-journal","title":"Artificial intelligence to investigate metabolomics data for precision medicine","abstract":"BACKGROUND: Metabolomic data offers insights into disease mechanisms, diagnostics, and therapeutic targets by analyzing metabolic profiles. In analyzing these profiles, traditional bioinformatic and statistical approaches, while valuable, often struggle to process high-dimensional and nonlinear metabolic data, lacking the sensitivity and adaptability that artificial intelligence (AI) and machine learning (ML) techniques provide. The integration of AI/ML has greatly enhanced the metabolomics field, enabling biomarker identification, disease prediction, and classification of metabolic patterns at an unprecedented level. AIM OF REVIEW: This study analyses and compares the scientific goals, methodologies, datasets, and sources of AI/ML approaches applied to metabolomic data, as well as assessing their implications in precision medicine. We systematically reviewed recent advancements in AI/ML applications to metabolomic data, focusing on peer-reviewed research indexed in PubMed. Significant number of studies were analyzed, covering diseases such as cancer, cardiovascular diseases, and diabetes. Our results showed that the most used AI/ML techniques were SVM, RF, Gradient Boosting, and Logistic Regression, highlighting their effectiveness in processing complex metabolic data. Despite these advancements, key challenges persist in AI/ML applications to metabolomics data, including small cohort sizes, data heterogeneity, and the need for improved model interpretability, and these challenges must be considered for future use. KEY SCIENTIFIC CONCEPTS OF REVIEW: Ultimately, our findings underscore the transformative potential of AI/ML in metabolomics and its critical role in advancing precision medicine by uncovering novel metabolic pathways, improving treatment strategies, and enabling the earlier diagnosis of diseases through predictive metabolic profiling.","author":[{"family":"Shenouda","given":"Antony"},{"family":"Senthilkumar","given":"Sahana"},{"family":"Mourad","given":"Youssef"},{"family":"Xie","given":"Joy"},{"family":"Peker","given":"Elizabeth"},{"family":"Zeeshan","given":"Saman"},{"family":"Ahmed","given":"Zeeshan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1007/s11306-026-02401-z","URL":"https://doi.org/10.1007/s11306-026-02401-z","source":"openalex"},{"id":"oa:W4409142663","type":"article-journal","title":"Validation of artificial intelligence spirometry diagnostic support software in primary care: a blinded diagnostic accuracy study","abstract":"Objective and design: The objective of the present study was to assess the discriminative accuracy of artificial intelligence (AI) software to identify COPD and other chronic respiratory diseases from primary care spirometry. This was a diagnostic study with blinded analysis. Methods: Retrospective hand-held spirometry data from consecutive patients attending primary care clinics in Hillingdon (London, UK) between September 2015 and March 2019 were used. The index diagnosis was the \"preferred\" diagnosis determined by AI software (highest probability) using supervised random-forest machine learning to interpret raw spirometry data and basic demographics. The reference diagnosis was based on the consensus of expert pulmonologists with access to primary and secondary care medical notes and results of relevant investigations. Cross-tabulation of the index test results by the results of the reference standard for COPD and other respiratory disease categories provided the main outcome measures. Results: In this primary care spirometry dataset from 1113 patients, 543 (48.8%) had a reference diagnosis of COPD. AI preferred diagnosis detected 456, achieving a sensitivity of 84.0% (95% CI 80.6-87.0%), specificity of 86.8% (83.8-89.5%), accuracy of 85.4% (83.2-87.5%) with area under curve (AUC) of 0.914 (0.896-0.930). AI preferred diagnosis identified 187 out of 249 patients with reference diagnosis of interstitial lung disease and 59 out of 107 patients with asthma, with AUCs of 0.900 (0.880-0.916) and 0.814 (0.790-0.836), respectively. Conclusion: AI software achieved high sensitivity and specificity in identifying COPD using spirometry and basic demographic data and may support accurate diagnosis of COPD in primary care. AI software performed less well for other chronic respiratory disease categories.","author":[{"family":"Sunjaya","given":"Anthony"},{"family":"Edwards","given":"George"},{"family":"Harvey","given":"Jennifer"},{"family":"Sylvester","given":"Karl"},{"family":"Purvis","given":"Joanna"},{"family":"Rutter","given":"Matthew"},{"family":"Shakespeare","given":"Joanna"},{"family":"Moore","given":"Vicky"},{"family":"El-Emir","given":"Ethaar"},{"family":"Doe","given":"Gillian"},{"family":"Orshoven","given":"Karolien"},{"family":"Patel","given":"Suhani"},{"family":"Vos","given":"Maarten"},{"family":"Elmahy","given":"Ahmed"},{"family":"Cuyvers","given":"Benoit"},{"family":"Desbordes","given":"Paul"},{"family":"Sehdev","given":"Satesh"},{"family":"Evans","given":"Rachael"},{"family":"Morgan","given":"Michael"},{"family":"Russell","given":"Richard"},{"family":"Jarrold","given":"Ian"},{"family":"Spain","given":"Nannette"},{"family":"Taylor","given":"Stephanie"},{"family":"Scott","given":"David"},{"family":"Prevost","given":"AT"},{"family":"Hopkinson","given":"Nicholas"},{"family":"Kon","given":"Samantha"},{"family":"Topalovic","given":"Marko"},{"family":"Man","given":"William"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1183/23120541.00116-2025","URL":"https://doi.org/10.1183/23120541.00116-2025","source":"openalex"},{"id":"oa:W4416782818","type":"article-journal","title":"The Emerging Role of Multimodal Artificial Intelligence in Urological Surgery","abstract":"BACKGROUND: Multimodal artificial intelligence (MMAI) is transforming urological oncology by enabling the seamless integration of diverse data sources, including imaging, clinical records and robotic telemetry to facilitate patient-specific decision-making. METHODS: This narrative review summarizes the current developments, applications, opportunities and risks of multimodal AI systems throughout the entire perioperative process in uro-oncologic surgery. RESULTS: MMAI demonstrates quantifiable benefits across the entire perioperative pathway. Preoperatively, it improves diagnostics and surgical planning via multimodal data fusion. Intraoperatively, AI-assisted systems provide real-time context-based decision support, risk prediction and skill assessment within the operating theater. Postoperatively, MMAI facilitates automated documentation, early complication detection and personalized follow-up. Generative AI further revolutionizes surgical training through adaptive feedback and simulations. However, critical limitations must be addressed, including data bias, the barrier of closed robotic platforms, insufficient model validation, data security issues, hallucinations and ethical concerns regarding liability and transparency. CONCLUSIONS: MMAI significantly enhances the precision, efficiency and patient-centeredness of uro-oncological care. To ensure safe and widespread implementation, resolving the technical and regulatory barriers to real-time integration into robotic platforms is paramount. This must be coupled with standardized quality controls, transparent decision-making processes and responsible integration that fully preserves physician autonomy.","author":[{"family":"Buck","given":"Leonhard"},{"family":"Kohler","given":"Jakob"},{"family":"Risch","given":"Julian"},{"family":"Incesu","given":"Reha‐baris"},{"family":"Hügelmann","given":"Konrad"},{"family":"Weiß","given":"Marie"},{"family":"Weische","given":"Oscar"},{"family":"Schließer","given":"Patricia"},{"family":"Knobloch","given":"Hans"},{"family":"Blessin","given":"Niclas"},{"family":"Bach","given":"Thorsten"},{"family":"Jarczyk","given":"Jonas"},{"family":"Nuhn","given":"Philipp"},{"family":"Rodler","given":"Severin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/curroncol32120665","URL":"https://doi.org/10.3390/curroncol32120665","source":"openalex"},{"id":"oa:W4415165763","type":"article-journal","title":"Analyzing enablers of artificial intelligence for decarbonization: implications for circular supply chains","abstract":"Abstract This study comprehensively explores the pivotal position that Artificial Intelligence (AI) enables on the advancement of decarbonization efforts, mainly in the context of Circular Supply Chains (CSCs). Employing a two-stage methodology, this study delves into identifying and analyzing the enablers essential for leveraging AI in the pursuit of decarbonization objectives. In the first stage, a literature review and an exploratory factor analysis are performed to discern the key enablers of AI for decarbonization initiatives. This process resulted in the identification of 15 significant enablers and categorization of enablers into environmental, organizational, institutional, and technological categories. Building upon the findings from the first stage, this study progresses to its second stage, wherein the Grey-Ordinal Priority Approach (G-OPA) is applied to analyze the identified enablers. The results indicate that adopting recyclable materials to enhance the efficiency of supply chains, emphasizing local production for recovery practices through advanced technology, and managing product life-cycle through intelligent and additive manufacturing technologies are the top three enablers. The application of the G-OPA enriches the robustness and comprehensiveness of the analysis, enabling an understanding of the complex interplay among the enablers. By clarifying the key enablers, business planners and designers can migrate from traditional linear supply chains to more sustainable CSCs through the careful implementation of enablers for decarbonization.","author":[{"family":"Srivastava","given":"Shefali"},{"family":"Shardeo","given":"Vipulesh"},{"family":"Dwivedi","given":"Ashish"},{"family":"Paul","given":"Sanjoy"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s10479-025-06843-x","URL":"https://doi.org/10.1007/s10479-025-06843-x","source":"openalex"},{"id":"oa:W4406243158","type":"article-journal","title":"Integration of Functional Materials in Photonic and Optoelectronic Technologies for Advanced Medical Diagnostics","abstract":"Integrating functional materials with photonic and optoelectronic technologies has revolutionized medical diagnostics, enhancing imaging and sensing capabilities. This review provides a comprehensive overview of recent innovations in functional materials, such as quantum dots, perovskites, plasmonic nanomaterials, and organic semiconductors, which have been instrumental in the development of diagnostic devices characterized by high sensitivity, specificity, and resolution. Their unique optical properties enable real-time monitoring of biological processes, advancing early disease detection and personalized treatment. However, challenges such as material stability, reproducibility, scalability, and environmental sustainability remain critical barriers to their clinical translation. Breakthroughs such as green synthesis, continuous flow production, and advanced surface engineering are addressing these limitations, paving the way for next-generation diagnostic tools. This article highlights the transformative potential of interdisciplinary research in overcoming these challenges and emphasizes the importance of sustainable and scalable strategies for harnessing functional materials in medical diagnostics. The ultimate goal is to inspire further innovation in the field, enabling the creation of practical, cost-effective, and environmentally friendly diagnostic solutions.","author":[{"family":"Thanjavur","given":"Naveen"},{"family":"Bugude","given":"Laxmi"},{"family":"Kim","given":"Young‐joon"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bios15010038","URL":"https://doi.org/10.3390/bios15010038","source":"openalex"},{"id":"oa:W4413140583","type":"article-journal","title":"The “Artificial Intelligence Statistician”: Utilizing Generative Artificial Intelligence to Select an Appropriate Model and Execute Network Meta-Analyses","abstract":"OBJECTIVES: This exploratory study aimed to develop a large language model (LLM)-based process to automate components of network meta-analysis (NMA), including model selection, analysis, output evaluation, and results interpretation. Automating these tasks with LLMs can enhance efficiency, consistency, and scalability in health economics and outcomes research, while ensuring that analyses adhere to established guidelines required by health technology assessment agencies. Improvements in efficiency and scalability may potentially become relevant as the European Union Health Technology Assessment Regulation comes into force, given anticipated analysis requirements and timelines. METHODS: Using Claude 3.5 Sonnet (V2), a process was designed to automate statistical model selection, NMA output evaluation, and results interpretation based on an \"analysis-ready\" data set. Validation was assessed by replicating examples from the National Institute for Health and Care Excellence Technical Support Document (TSD2), replicating results of non-Decision Support Unit-published NMAs, and generating comprehensive outputs (eg, heterogeneity, inconsistency, and convergence). RESULTS: The automated LLM-based process produced accurate results. Compared with TSD2 examples, differences were minimal, within expectations (given differences in sampling frameworks used), and comparable to those observed between estimates produced by the R vignettes against TSD2. Similar consistency was noted for non-Decision Support Unit-published NMA examples. Additionally, the LLM process generated and interpreted comprehensive NMA outputs. CONCLUSIONS: This exploratory study demonstrates the feasibility of LLMs to automate key components of NMAs, determining the requisite NMA framework based only on input data. Further exploring these capabilities could clarify their role in streamlining NMA workflows.","author":[{"family":"Reason","given":"Tim"},{"family":"Wu","given":"Yunchou"},{"family":"Jones","given":"Cheryl"},{"family":"Benbow","given":"Emma"},{"family":"Johannesen","given":"Kasper"},{"family":"Malcolm","given":"Bill"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jval.2025.08.001","URL":"https://doi.org/10.1016/j.jval.2025.08.001","source":"openalex"},{"id":"oa:W4408905879","type":"article-journal","title":"Artificial intelligence-based virtual staining platform for identifying tumor-associated macrophages from hematoxylin and eosin-stained images","abstract":"BACKGROUND: Virtual staining is an artificial intelligence-based approach that transforms pathology images between stain types, such as hematoxylin and eosin (H&E) to immunohistochemistry (IHC), providing a tissue-preserving and efficient alternative to traditional IHC staining. However, existing methods for translating H&E to virtual IHC often fail to generate images of sufficient quality for accurately delineating cell nuclei and IHC+ regions. To address these limitations, we introduce VISTA, an artificial intelligence-based virtual staining platform designed to translate H&E into virtual IHC. METHODS: We applied VISTA to identify M2-subtype tumor-associated macrophages (M2-TAMs) in H&E images from 968 patients with HPV+ oropharyngeal squamous cell carcinoma across six institutional cohorts. M2-TAMs are a critical component of the tumor microenvironment, and their increased presence has been linked to poor survival. Co-registered H&E and CD163 + IHC tissue microarrays were used to train (D1, N = 102) and test (D2, N = 50) the VISTA platform. M2-TAM density, defined as the ratio of M2-TAMs to total nuclei, was derived from VISTA-generated CD163 + IHC images and evaluated for prognostic significance in additional training (D3, N = 360) and testing (D4, N = 456) cohorts using biopsy or resection H&E whole slide images. RESULTS: High M2-TAM density was associated with worse overall survival in D4 (p = 0.0152, Hazard Ratio=1.63 [1.1-2.42]). VISTA outperformed existing methods, generating higher-quality virtual CD163 + IHC images in D2, with a Structural Similarity Index of 0.72, a Peak Signal-to-Noise Ratio of 21.5, and a Fréchet Inception Distance of 41.4. Additionally, VISTA demonstrated superior performance in segmenting M2-TAMs in D2 (Dice=0.74). CONCLUSION: These findings establish VISTA as a computational platform for generating virtual IHC and facilitating the discovery of novel biomarkers from H&E images.","author":[{"family":"Aggarwal","given":"Arpit"},{"family":"Jana","given":"Mayukhmala"},{"family":"Singh","given":"Amritpal"},{"family":"Dam","given":"Tanmoy"},{"family":"Maurya","given":"Himanshu"},{"family":"Pathak","given":"Tilak"},{"family":"Oršulić","given":"Sandra"},{"family":"Yang","given":"Kailin"},{"family":"Chute","given":"Deborah"},{"family":"Bishop","given":"Justin"},{"family":"Faraji","given":"Farhoud"},{"family":"Thorstad","given":"Wade"},{"family":"Koyfman","given":"Shlomo"},{"family":"Steward-Tharp","given":"Scott"},{"family":"Shi","given":"Qiuying"},{"family":"Sandulache","given":"Vlad"},{"family":"Saba","given":"Nabil"},{"family":"Lewis","given":"James"},{"family":"Corredor","given":"Germán"},{"family":"Madabhushi","given":"Anant"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.ejca.2025.115390","URL":"https://doi.org/10.1016/j.ejca.2025.115390","source":"openalex"},{"id":"oa:W4406357731","type":"article-journal","title":"Patients’ attitudes toward artificial intelligence (AI) in cancer care: A scoping review protocol","abstract":"BACKGROUND: Artificial intelligence broadly refers to computer systems that simulate intelligent behaviour with minimal human intervention. Emphasizing patient-centered care, research has explored patients' perspectives on artificial intelligence in medical care, indicating general acceptance of the technology but also concerns about supervision. However, these views have not been systematically examined from the perspective of patients with cancer, whose opinions may differ given the distinct psychosocial toll of the disease. OBJECTIVES: This protocol describes a scoping review aimed at summarizing the existing literature on the attitudes of patients with cancer toward the use of artificial intelligence in their medical care. The primary goal is to identify knowledge gaps and highlight opportunities for future research. METHODS: This scoping review protocol will adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PRISMA-ScR). The electronic databases MEDLINE (OVID), EMBASE, PsycINFO, and CINAHL will be searched for peer-reviewed primary research articles published in academic journals. We will have two independent reviewers screen the articles retrieved from the literature search and select relevant studies based on our inclusion criteria, with a third reviewer resolving any disagreements. We will then compile the data from the included articles into a narrative summary and discuss the implications for clinical practice and future research. DISCUSSION: To our knowledge, this will be the first scoping review to map the existing literature on the attitudes of patients with cancer regarding artificial intelligence in their medical care.","author":[{"family":"Hilbers","given":"Daniel"},{"family":"Nekain","given":"Navid"},{"family":"Bates","given":"Alan"},{"family":"Nuñez","given":"John"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1371/journal.pone.0317276","URL":"https://doi.org/10.1371/journal.pone.0317276","source":"openalex"},{"id":"oa:W4408781895","type":"article-journal","title":"A Review of the State of the Art for the Internet of Medical Things","abstract":"The technological developments in the Internet of Things (IoT), data science, artificial intelligence, wearable sensors, remote monitoring, decision support systems, fog, and edge systems have transformed digital healthcare. Especially after the pandemic, there has been a rapid transformation of healthcare infrastructure from a conventional to a digital approach. Now, specifically, technologies such as the Internet of Things play a vital role in the transformation of the healthcare system. In this paper, an effort has been made to encompass the transformation of healthcare with a focus on the Internet of Medical Things (IoMT). In particular, it provides a detailed overview of the Internet of Medical Things whilst discussing the design goals and challenges, the resource constraints and limitations of the complex healthcare systems. The paper also provides a detailed account of the research initiatives as well as off-the-shelf wireless motes, internet-enabled sensors and open-source platforms. A thorough account of the next-generation digital healthcare technologies and future research opportunities is provided. This work not only covers the state-of-the-art but also offers critical insight into the digital healthcare challenges. The work attempts to summarise the extensive literature in the domain and present a new perspective on the internet of medical things, affiliate technologies and their role in healthcare.","author":[{"family":"Matthew","given":"Peter"},{"family":"Mchale","given":"Sarah"},{"family":"Deng","given":"XT"},{"family":"Nakhla","given":"Ghada"},{"family":"Trovati","given":"Marcello"},{"family":"Nnamoko","given":"Nonso"},{"family":"Pereira","given":"Ella"},{"family":"Zhang","given":"Huaizhong"},{"family":"Raza","given":"Mohsin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/sci7020036","URL":"https://doi.org/10.3390/sci7020036","source":"openalex"},{"id":"oa:W4406045007","type":"article-journal","title":"Responsible Artificial Intelligence for Mental Health Disorders: Current Applications and Future Challenges","abstract":"Mental health disorders (MHDs) have significant medical and financial impacts on patients and society. Despite the potential opportunities for artificial intelligence (AI) in the mental health field, there are no noticeable roles of these systems in real medical environments. The main reason for these limitations is the lack of trust by domain experts in the decisions of AI-based systems. Recently, trustworthy AI (TAI) guidelines have been proposed to support the building of responsible AI (RAI) systems that are robust, fair, and transparent. This review aims to investigate the literature of TAI for machine learning (ML) and deep learning (DL) architectures in the MHD domain. To the best of our knowledge, this is the first study that analyzes the literature of trustworthiness of ML and DL models in the MHD domain. The review identifies the advances in the literature of RAI models in the MHD domain and investigates how this is related to the current limitations of the applicability of these models in real medical environments. We discover that the current literature on AI-based models in MHD has severe limitations compared to other domains regarding TAI standards and implementations. We discuss these limitations and suggest possible future research directions that could handle these challenges.","author":[{"family":"Elsappagh","given":"Shaker"},{"family":"Nazih","given":"Waleed"},{"family":"Alharbi","given":"Meshal"},{"family":"Abuhmed","given":"Tamer"}],"issued":{"date-parts":[[2025]]},"DOI":"10.57197/jdr-2024-0101","URL":"https://doi.org/10.57197/jdr-2024-0101","source":"openalex"},{"id":"oa:W4414255620","type":"article-journal","title":"Applications of artificial intelligence in early childhood health management: a systematic review from fetal to pediatric periods","abstract":"Background: The integration of artificial intelligence (AI) into early childhood health management has expanded rapidly, with applications spanning the fetal, neonatal, and pediatric periods. While numerous studies report promising results, a comprehensive synthesis of AI's performance, methodological quality, and translational readiness in child health is needed. Objectives: This systematic review aims to evaluate the current landscape of AI applications in fetal and pediatric care, assess their diagnostic accuracy and clinical utility, and identify key barriers to real-world implementation. Methods: A systematic literature search was conducted in PubMed, Scopus, and Web of Science for studies published between January 2021 and March 2025. Eligible studies involved AI-driven models for diagnosis, prediction, or decision support in individuals aged 0-18 years. Study selection followed the PRISMA 2020 guidelines. Data were extracted on application domain, AI methodology, performance metrics, validation strategy, and clinical integration level. Results: From 4,938 screened records, 133 studies were included. AI models demonstrated high performance in prenatal anomaly detection (mean AUC: 0.91-0.95), neonatal intensive care (e.g., sepsis prediction with sensitivity up to 89%), and pediatric genetic diagnosis (accuracy: 85%-93% using facial analysis). Deep learning enhanced consistency in fetal echocardiography and ultrasound interpretation. However, 76% of studies used single-center retrospective data, and only 21% reported external validation. Performance dropped by 15%-20% in cross-institutional settings. Fewer than 5% of models have been integrated into routine clinical workflows, with limited reporting on data privacy, algorithmic bias, and clinician trust. Conclusion: AI holds transformative potential across the pediatric continuum of care-from fetal screening to chronic disease management. However, most applications remain in the research phase, constrained by data heterogeneity, lack of prospective validation, and insufficient regulatory alignment. To advance clinical adoption, future efforts should focus on multicenter collaboration, standardized data sharing frameworks, explainable AI, and pediatric-specific regulatory pathways. This review provides a roadmap for clinicians, researchers, and policymakers to guide the responsible translation of AI in child health.","author":[{"family":"Wang","given":"Qingsong"},{"family":"Yin","given":"Jun"},{"family":"Zhang","given":"Xiaomeng"},{"family":"Ou","given":"Huang‐tz"},{"family":"Li","given":"Fuyan"},{"family":"Zhang","given":"Yundong"},{"family":"Wan","given":"Weiyi"},{"family":"Guo","given":"Caiyu"},{"family":"Cao","given":"Yongyu"},{"family":"Luo","given":"Tongyong"},{"family":"Wang","given":"Xianmin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fped.2025.1613150","URL":"https://doi.org/10.3389/fped.2025.1613150","source":"openalex"},{"id":"oa:W4413113793","type":"article-journal","title":"An analysis of the real world performance of an artificial intelligence based autism diagnostic","abstract":"Rapidly rising demand for pediatric autism evaluations has outpaced specialist capacity and created a crisis of delayed diagnoses and treatment. Streamlining the diagnostic process could reduce wait times and optimize use of limited specialist resources. Following strong clinical trial results, Canvas Dx, an AI-based diagnostic, was FDA authorized to support accurate diagnosis or rule-out of autism in children 18-72 months with caregiver or healthcare provider concern for developmental delay. To gain insight into real-world device performance, a de-identified aggregate data analysis of the initial 254 Canvas Dx prescriptions fulfilled post-market authorization was conducted to determine: accuracy of autism predictions compared to clinical reference standard diagnosis and prior clinical trial data, key real-world prescriber and patient characteristics, proportion of determinate device outputs (positive or negative for autism) and impact of decision threshold settings on device performance. In this sample of 254 children with a 54.7% autism prevalence rate (29.1% female, average age 39.99 months), Canvas Dx had a NPV of 97.6% (CI- 92.8% -100.0%) and a PPV of 92.4% (CI-87.7%-97.2%). A majority of cases (63.0%) received a determinate result. Sensitivity and specificity of determinate results were 99.1% (CI-97.3%-100.0%) and 81.6% (CI-70.8%-92.5%) respectively. The median age of children who received a positive for autism output was 37.2 months, which is more than 2 years earlier than the current median age of autism diagnosis. No performance differences were noted based on patients' sex. Compared to clinical trial results, real world performance was equivalent for all key metrics, with the exception of the determinate rate and the PPV which were significantly improved in real world performance. Analysis of real-world Canvas Dx data highlights its feasibility and utility in supporting accurate, equitable and early diagnosis or rule out of autism. With medical coverage and broader clinical adoption, innovative solutions such as Canvas Dx can play an important role in helping to address the growing specialist waitlist crisis, ensuring that more children gain access to targeted therapies during the critical window of neurodevelopment where they have the greatest life-changing impact.","author":[{"family":"Salomon","given":"Carmela"},{"family":"Heinz","given":"K"},{"family":"Aronson-Ramos","given":"Judith"},{"family":"Wall","given":"Dennis"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-15575-8","URL":"https://doi.org/10.1038/s41598-025-15575-8","source":"openalex"},{"id":"oa:W4409743032","type":"article-journal","title":"Overview of South Korean Guidelines for Approval of Large Language or Multimodal Models as Medical Devices: Key Features and Areas for Improvement","abstract":"The Ministry of Food and Drug Safety (MFDS) of the Republic of Korea, similar to the United States Food and Drug Administration and the United Kingdom's Medicines and Healthcare products Regulatory Agency, issued specific regulatory guidelines on January 24, 2025, for approving generative artificial intelligence (AI) technologies as medical devices [1].Although these guidelines use the term 'generative AI,' they predominantly focus on the approval of AI software tools based on large language models (LLMs) and large multimodal models (LMMs) [2], the latter of which can process various types of input data, such as texts, images, videos, audio, and bio-signals.Unlike the broader guidelines for approving AI models as medical devices, specific regulatory guidelines for LLMs/LMMs have arguably not yet been proposed in other countries.","author":[{"family":"Park","given":"Seong"},{"family":"Dean","given":"Geraldine"},{"family":"Ortiz","given":"Ernest"},{"family":"Choi","given":"Joon‐il"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3348/kjr.2025.0257","URL":"https://doi.org/10.3348/kjr.2025.0257","source":"openalex"},{"id":"oa:W4412692164","type":"article-journal","title":"Artificial Intelligence Approach for Waste-Printed Circuit Board Recycling: A Systematic Review","abstract":"The rapid advancement of technology has led to a substantial increase in Waste Electrical and Electronic Equipment (WEEE), which poses significant environmental threats and increases pressure on the planet’s limited natural resources. In response, Artificial Intelligence (AI) has emerged as a key enabler of the Circular Economy (CE), particularly in improving the speed and precision of waste sorting through machine learning and computer vision techniques. Despite this progress, to our knowledge, no comprehensive, systematic review has focused specifically on the role of AI in disassembling and recycling Waste-Printed Circuit Boards (WPCBs). This paper addresses this gap by systematically reviewing recent advancements in AI-driven disassembly and sorting approaches with a focus on machine learning and vision-based methodologies. The review is structured around three areas: (1) the availability and use of datasets for AI-based WPCB recycling; (2) state-of-the-art techniques for selective disassembly and component recognition to enable fast WPCB recycling; and (3) key challenges and possible solutions aimed at enhancing the recovery of critical raw materials (CRMs) from WPCBs.","author":[{"family":"Mohsin","given":"Muhammad"},{"family":"Rovetta","given":"Stefano"},{"family":"Masulli","given":"Francesco"},{"family":"Cabri","given":"Alberto"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/computers14080304","URL":"https://doi.org/10.3390/computers14080304","source":"openalex"},{"id":"oa:W4413285356","type":"article-journal","title":"Student engagement with artificial intelligence tools in academia: a survey of Jordanian universities","abstract":"The rapid advancement of artificial intelligence (AI) has led to its increasing integration into academic environments, raising critical questions about its educational implications. This study investigates the use of AI tools among university students in Jordan, focusing on platforms such as ChatGPT, Google Bard, Microsoft Bing, and Meta AI. A convergent-parallel mixed-methods design was employed, with quantitative (closed-ended) and qualitative (open-ended) data collected concurrently through an online survey distributed over two months. A total of 337 valid responses were obtained from students across 27 universities. The survey explored demographic characteristics, chatbot awareness and use, perceived benefits and challenges, ethical concerns, and future intentions. Results indicate that ChatGPT is the most recognized (94.3%) and widely used (90.4%) tool, while Meta AI is the least utilized (7.8%). Approximately 89% of students reported using AI tools for academic tasks, and 86.6% perceived them as educationally beneficial. However, only 39.7% believed these tools significantly improved their understanding, while 57.6% reported a positive impact on academic performance. These findings reveal a growing trend of AI integration into student study practices in Jordan, highlighting both its practical advantages and the need for further inquiry into its pedagogical value and ethical use.","author":[{"family":"Mashagbeh","given":"Mohammad"},{"family":"Alsharqawi","given":"Malak"},{"family":"Tudevdagva","given":"Uranchimeg"},{"family":"Khasawneh","given":"Hussam"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/feduc.2025.1550147","URL":"https://doi.org/10.3389/feduc.2025.1550147","source":"openalex"},{"id":"oa:W4414985172","type":"article-journal","title":"Attitudes and perceptions of dental students towards artificial intelligence","abstract":"BACKGROUND: Artificial intelligence (AI) is rapidly transforming healthcare, including dentistry, through its applications in diagnosis, prosthetic planning, and oral disease detection. As future professionals, dental students play a vital role in integrating AI into clinical practice. However, little is known about their attitudes toward AI, particularly in low-resource settings such as Palestine. METHODS: A cross-sectional survey was conducted among 305 dental students from four Palestinian universities using a validated 22-item questionnaire. Data were analyzed using descriptive statistics and chi-square tests (p < 0.05). RESULTS: Among the 305 participants (232 females, 73 males), 77% reported basic knowledge of AI, with social media being the most common source (66.2%). Female students were significantly more likely than males to believe that AI will bring major advancements to dentistry (p = 0.04), and that it can be used for diagnostic (p = 0.030), prognostic (p = 0.045), and treatment planning purposes (p = 0.015), as well as in postgraduate training (p = 0.017). Those with prior AI knowledge or awareness of its dental applications showed greater enthusiasm for its diagnostic use (p = 0.004) and integration into dental education (p = 0.004 and p = 0.009, respectively). CONCLUSION: Palestinian dental students demonstrate strong awareness and positive attitudes toward the use of AI in dentistry. Gender-based differences and ethical concerns emphasize the need for structured, inclusive, and responsible AI training within dental curricula.","author":[{"family":"Shrateh","given":"Oadi"},{"family":"Al-Batat","given":"Siwar"},{"family":"Alqudimat","given":"Ahmad"},{"family":"Ghannam","given":"Lara"},{"family":"Abuhanoud","given":"Lina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12909-025-07854-9","URL":"https://doi.org/10.1186/s12909-025-07854-9","source":"openalex"},{"id":"oa:W4410849931","type":"article-journal","title":"Artificial intelligence in focus: assessing awareness and perceptions among medical students in three private Syrian universities","abstract":"BACKGROUND: Artificial intelligence (AI) has gained significant attention and progress in various scientific fields, especially medicine. Since its introduction in the 1950s, AI has advanced remarkably, supporting innovations like diagnostic tools and healthcare technologies. Despite these developments, challenges such as ethical concerns and limited integration in regions like Syria emphasize the importance of increasing awareness and conducting more targeted studies. METHODS: A cross-sectional study was conducted to evaluate medical students' preparedness and readiness to use AI technologies in the medical field using the Medical Artificial Intelligence Readiness Scale for Medical Students (MAIRS_MS). The scale comprises 22 items divided into 4 domains: ethics, vision, ability, and cognition, with responses rated on a five-point Likert scale, higher scores indicate greater readiness. Data were collected through electronic and paper questionnaires distributed over a period of 20 days. RESULTS: The study included 564 medical students from various Syrian universities, of whom 77.8% demonstrated awareness of AI in the medical field. Significant differences in AI awareness were observed based on academic GPA (p = 0.035) and income level (p = 0.016), with higher awareness among students with higher GPA and income levels. Statistically significant differences were found between students aware of AI and those unaware, as well as between students with experience using AI and those without, across all domains of readiness, including cognition (t = -10.319, p < 0.001), ability (t = -11.519, p < 0.001), vision (t = -6.387, p < 0.001), ethics (t = -7.821, p < 0.001), and the overall readiness score (t = -11.354, p < 0.001). CONCLUSION: Integrating AI into medical education is essential for advancing healthcare in developing countries like Syria. Providing incentives and fostering a culture of continuous learning will equip medical students to leverage AI's benefits while mitigating its drawbacks.","author":[{"family":"Hanifa","given":"Hamdah"},{"family":"Atia","given":"Mohammad"},{"family":"Daboul","given":"Rawan"},{"family":"Alhamid","given":"Ahmad"},{"family":"Alayyoubi","given":"Aya"},{"family":"Naima","given":"Hiam"},{"family":"Alkassar","given":"Deema"},{"family":"Nabhan","given":"Murhaf"},{"family":"Alsaleh","given":"Basil"},{"family":"Abdula","given":"Farris"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12909-025-07396-0","URL":"https://doi.org/10.1186/s12909-025-07396-0","source":"openalex"},{"id":"oa:W7119087571","type":"article-journal","title":"The impact of generative AI on academic reading and writing: a synthesis of recent evidence (2023–2025)","abstract":"Introduction The aim of this systematic review is to examine the scientific literature published on digital reading and writing in higher education within the field of social sciences, assisted by generative artificial intelligence. Methods The PRISMA methodology and the SALSA Framework were applied, based on a bibliographic search conducted in the Scopus and Web of Science databases. Journal articles that explicitly addressed the established topic, published between 1 January 2023 and 7 March 2025, in open access, in Spanish or English, and within the field of Social Sciences, were included. After a rigorous screening and selection process, a final sample of 136 articles was compiled and used as the basis for the study. Results The findings indicate that the reviewed research employs a range of methodologies, encompassing quantitative (surveys, experimental studies, psychometric evaluations), qualitative (case studies, semi-structured interviews, thematic analysis), and mixed-method approaches. The results also reveal a clear trend toward the integration of artificial intelligence tools –particularly ChatGPT– into academic writing processes. A significant improvement is observed in the quality of students’ texts, especially regarding coherence, discursive organization, lexical richness, and argumentation. Furthermore, the role of AI in formative feedback, idea generation, paraphrasing, and fostering student autonomy in self-editing their texts is highlighted. The research also identifies key challenges, such as students’ overreliance on AI, diminished metacognitive engagement, and ethical dilemmas related to plagiarism and authorship. Discussion The emergence of AI in higher education is transforming teaching and learning processes, creating opportunities for personalization and enhanced support in academic writing. However, the scientific literature also exposes tensions between its potential benefits and associated risks, such as student dependency, loss of critical thinking, and ethical concerns regarding authorship and plagiarism. These findings call for a rethinking of pedagogical, assessment, and institutional practices, as well as the development of critical and digital literacy skills among both teachers and students.","author":[{"family":"Tejeda","given":"Aránzazu"},{"family":"Oller","given":"Juana"},{"family":"Baldaquí-Escandell","given":"Josep"},{"family":"Gómez-Díaz","given":"Raquel"},{"family":"García-Rodríguez","given":"Araceli"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/feduc.2025.1711718","URL":"https://doi.org/10.3389/feduc.2025.1711718","source":"openalex"},{"id":"oa:W4409894239","type":"article-journal","title":"LoRa Communications Spectrum Sensing Based on Artificial Intelligence: IoT Sensing","abstract":"The backbone of the Internet of Things ecosystem relies heavily on wireless sensor networks and low-power wide area network technologies, such as LoRa modulation, to provide the long-range, energy-efficient communications essential for applications as diverse as smart homes, healthcare, agriculture, smart grids, and transportation. With the number of IoT devices expected to reach approximately 41 billion by 2034, managing radio spectrum resources becomes a critical issue. However, as these devices are deployed at an increasing rate, the limited spectral resources will result in increased interference, packet collisions, and degraded quality of service. Current methods for increasing network capacity have limitations and require advanced solutions. This paper proposes a novel hybrid spectrum sensing framework that combines traditional signal processing and artificial intelligence techniques specifically designed for LoRa spreading factor detection and communication channel analytics. Our proposed framework processes wideband signals directly from IQ samples to identify and classify multiple concurrent LoRa transmissions. The results show that the framework is highly effective, achieving a detection accuracy of 96.2%, a precision of 99.16%, and a recall of 95.4%. The proposed framework's flexible architecture separates the AI processing pipeline from the channel analytics pipeline, ensuring adaptability to various communication protocols beyond LoRa.","author":[{"family":"Mutescu","given":"Partemie"},{"family":"Popa","given":"Valentin"},{"family":"Lavric","given":"Alexandru"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/s25092748","URL":"https://doi.org/10.3390/s25092748","source":"openalex"},{"id":"oa:W7154839439","type":"article-journal","title":"Artificial intelligence in additive Manufacturing: advances in smart materials, lattice optimization, and process intelligence","abstract":"Artificial intelligence (AI) and additive manufacturing (AM) have propelled the next wave of technological innovation by integrating data-driven intelligence with design freedom, thereby enabling adaptive, efficient, and multifunctional systems. This review highlights the transformative role of AI and machine learning (ML) in addressing the key challenges associated with process complexity, parameter tuning, and multifunctional design in AM. Starting with the historical evolution of AI-AM integration, the progression from rule-based modeling to contemporary deep learning, reinforcement learning, and physics-informed frameworks that enable autonomous and self-optimizing manufacturing systems was summarized. Attention is directed toward ML-driven topology and lattice optimization, data-driven methods for predicting process and structural properties, and Multiphysics optimization, demonstrating how AI replaces labor-intensive experimentation with predictive and adaptive models. In parallel, the integration of AI into smart materials (SMs) and 4D printing has been explored, with emphasis on property tuning for piezoelectric, shape-memory, and self-healing systems. The review concludes by highlighting key challenges, including data scarcity, limited interpretability, and the lack of standardized datasets, while pointing toward hybrid physics-informed ML and digital twin approaches for closed-loop and intelligent manufacturing. Collectively, this study provides a comprehensive roadmap illustrating how AI enables AM to evolve from empirical fabrication to autonomous, multifunctional manufacturing paradigms.","author":[{"family":"Zaman","given":"Saqlain"},{"family":"Mahmud","given":"Md"},{"family":"Mollick","given":"Ali"},{"family":"Lhaden","given":"Tenzin"},{"family":"Dantzler","given":"Joshua"},{"family":"Arroyo","given":"Sabina"},{"family":"Goona","given":"Nithin"},{"family":"Mesbah","given":"Maisha"},{"family":"Ahsan","given":"Md"},{"family":"Lin","given":"Yirong"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1007/s00170-026-18072-y","URL":"https://doi.org/10.1007/s00170-026-18072-y","source":"openalex"},{"id":"oa:W4412490723","type":"article-journal","title":"Artificial intelligence for endoscopic grading of gastric intestinal metaplasia: advancing risk stratification for gastric cancer","abstract":"Background: The Endoscopic Grading of Gastric Intestinal Metaplasia (EGGIM) classification correlates with histological assessment of gastric intestinal metaplasia and enables stratification of gastric cancer risk. We developed and evaluated an artificial intelligence (AI) approach for EGGIM estimation. Methods: Two datasets (A and B) with 1280 narrow-band imaging images were used for per-image analysis. Still images with manually selected patches of 224 × 224 pixels, annotated by experts, were used. Dataset A was retrospectively collected from clinical routine; Dataset B (used for per-patient analysis) was prospectively collected and included 65 fully documented patients. To mimic clinical practice, a deep neural network classified image patches into three EGGIM classes (0, 1, 2) and calculated the total per-patient EGGIM score (0–10). Results: On per-image analysis, an accuracy of 87% (95%CI 71%–100%) was obtained. Per-patient EGGIM estimation had an average error of 1.15 (out of 10) and showed 88% (95%CI 80%–96%) accurate clinical decisions for surveillance (EGGIM ≥5), with 85% (95%CI 75%–94%) specificity, no false negatives, and positive and negative predictive values of 62% (95%CI 32%–92%) and 100% (95%CI 100%–100%), respectively. Conclusions: EGGIM was estimated with high accuracy using AI tools in endoscopic image analyses. Automated assessment of EGGIM may provide a greener strategy for gastric cancer risk stratification, prospective studies, and interventional trials.","author":[{"family":"Almeida","given":"Eduarda"},{"family":"Martins","given":"Miguel"},{"family":"Marques","given":"David"},{"family":"Delas","given":"Rose"},{"family":"Almeida","given":"Tatiana"},{"family":"Chaves","given":"Jéssica"},{"family":"Libânio","given":"Diogo"},{"family":"Renna","given":"Francesco"},{"family":"Coimbra","given":"Miguel"},{"family":"Dinisribeiro","given":"Mário"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1055/a-2657-9906","URL":"https://doi.org/10.1055/a-2657-9906","source":"openalex"},{"id":"oa:W4410239408","type":"article-journal","title":"Artificial Intelligence-Assisted Muscular Ultrasonography for Assessing Inflammation and Muscle Mass in Patients at Risk of Malnutrition","abstract":"Background: Malnutrition, influenced by inflammation, is associated with muscle depletion and body composition changes. This study aimed to evaluate muscle mass and quality using Artificial Intelligence (AI)-enhanced ultrasonography in patients with inflammation. Methods: This observational, cross-sectional study included 502 malnourished patients, assessed through anthropometry, electrical bioimpedanciometry, and ultrasonography of the quadriceps rectus femoris (QRF). AI-assisted ultrasonography was used to segment regions of interest (ROI) from transversal QRF images to measure muscle thickness (RFMT) and area (RFMA), while a Multi-Otsu algorithm was used to extract biomarkers for muscle mass (MiT) and fat mass (FatiT). Inflammation was defined as C-reactive protein (CRP) levels above 3 mg/L. Results: The results showed a mean patient age of 63.72 (15.95) years, with malnutrition present in 82.3% and inflammation in 44.8%. Oncological diseases were prevalent (46.8%). The 44.8% of patients with inflammation (CRP > 3) exhibited reduced RFMA (2.91 (1.11) vs. 3.20 (1.19) cm2, p < 0.01) and RFMT (0.94 (0.28) vs. 1.01 (0.30) cm, p < 0.01). Muscle quality was reduced, with lower MiT (45.32 (9.98%) vs. 49.10 (1.22%), p < 0.01) and higher FatiT (40.03 (6.72%) vs. 37.58 (5.63%), p < 0.01). Adjusted for age and sex, inflammation increased the risks of low muscle area (OR = 1.59, CI: 1.10–2.31), low MiT (OR = 1.49, CI: 1.04–2.15), and high FatiT (OR = 1.44, CI: 1.00–2.06). Conclusions: AI-assisted ultrasonography revealed that malnourished patients with inflammation had reduced muscle area, thickness, and quality (higher fat content and lower muscle percentage). Elevated inflammation levels were associated with increased risks of poor muscle metrics. Future research should focus on exploring the impact of inflammation on muscles across various patient groups and developing AI-driven biomarkers to enhance the diagnosis, monitoring, and treatment of malnutrition and sarcopenia.","author":[{"family":"Gómez","given":"Juan"},{"family":"Estévez-Asensio","given":"Lucía"},{"family":"Cebriá","given":"Ángela"},{"family":"Izaola-Jáuregui","given":"Olatz"},{"family":"López","given":"Paloma"},{"family":"González-Gutiérrez","given":"Jaime"},{"family":"Primo","given":"David"},{"family":"Jiménez-Sahagún","given":"Rebeca"},{"family":"Hoyos","given":"Emilia"},{"family":"Rico","given":"Daniel"},{"family":"Godoy","given":"Eduardo"},{"family":"Luis","given":"Daniel"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/nu17101620","URL":"https://doi.org/10.3390/nu17101620","source":"openalex"},{"id":"oa:W4417327445","type":"article-journal","title":"Specialised Competencies and Artificial Intelligence in Perioperative Care: Contributions Toward Safer Practice","abstract":"This narrative review explores how specialised clinical competencies and artificial intelligence (AI) technologies converge in the context of perioperative care, with a focus on their combined potential to improve patient safety. Considering the growing complexity of surgical care and rising demands on healthcare professionals, the study aims to understand how human expertise and digital tools can complement each other in this high-stakes environment. Methods: A narrative review methodology was adopted to integrate insights from diverse sources, including empirical studies, policy documents, and expert analyses published over the last decade. Findings reveal that AI can support clinical decision-making, streamline workflows, and enable earlier identification of complications across all perioperative phases. These technologies enhance, rather than replace, the roles of nurses, anesthetists, and surgeons. However, their effective use depends on critical factors such as digital literacy, interdisciplinary collaboration, and ethical awareness. Issues related to data privacy, algorithmic bias, and unequal access to technology highlight the need for thoughtful, inclusive implementation. The future of perioperative care will likely depend on hybrid models where human judgment and AI-based tools are integrated in ways that uphold safety, equity, and person-centred values.","author":[{"family":"Raposo","given":"Sara"},{"family":"Mascarenhas","given":"Miguel"},{"family":"Correia","given":"Ricardo"},{"family":"Ferreira","given":"João"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/healthcare13243286","URL":"https://doi.org/10.3390/healthcare13243286","source":"openalex"},{"id":"oa:W4413018905","type":"article-journal","title":"Effectiveness of preliminary differential diagnosis of benign and malignant skin neoplasms using the Derma Onko Check artificial intelligence program","abstract":"Objective: to evaluate the effectiveness of preliminary differential diagnostics of benign and malignant skin tumors during initial medical consultations in primary health care using the Derma Onko Check artificial intelligence (AI) program for electronic computing devices (smartphone application). Material and methods. The effectiveness of the Derma Onko Check program for visual identification of benign and malignant skin tumors was evaluated in 135 patients aged 22 to 78 years with various skin lesions that appeared visually suspicious for malignancy. The conclusions generated by the program were compared with the results of dermatoscopic and morphological examinations. Results. The diagnostic accuracy of the Derma Onko Check program in determining the likelihood of a patient having a benign or malignant skin tumor was 96%, sensitivity was 98%, specificity was 96%, the proportion of false-positive results was 4.3%, and the propor Conclusion. The use of modern AI-based software for electronic computing devices enables early detection of malignant skin tumors during initial examinations in primary health care. This is particularly relevant for medical institutions and regions with a shortage or absence of dermatologists and oncologists. tion of falsenegative results was 2.4%.","author":[{"family":"Lamotkin","given":"AI"},{"family":"Korabelnikov","given":"DI"},{"family":"Olisova","given":"Olga"},{"family":"Lamotkin","given":"Igor"}],"issued":{"date-parts":[[2025]]},"DOI":"10.17749/2070-4909/farmakoekonomika.2025.294","URL":"https://doi.org/10.17749/2070-4909/farmakoekonomika.2025.294","source":"openalex"},{"id":"oa:W4414610376","type":"article-journal","title":"Diagnostic performance of artificial intelligence for dermatological conditions: a systematic review focused on low- and middle-income countries to address resource constraints and improve access to specialist care","abstract":"BACKGROUND: Artificial Intelligence (AI) has emerged as a transformative tool in dermatology, particularly in Low- and Middle-Income Countries (LMICs), where healthcare systems face challenges such as a shortage of dermatologists and limited resources. AI technologies, including deep learning models like Convolutional Neural Networks (CNNs), have demonstrated potential in improving diagnostic accuracy for skin diseases, which contribute significantly to the global disease burden. However, most research has focused on High-Income Countries (HICs), leaving gaps in understanding AI's applicability and effectiveness in LMICs. AIM/OBJECTIVE: This systematic review critically evaluates the application of AI in dermatological practice within LMICs, assessing the performance of AI technologies across diverse geographic regions. METHODOLOGY: The review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and included 19 studies from databases including PubMed, Embase, and Cochrane. Eligible studies evaluated AI applications in dermatology within LMICs, reporting metrics like sensitivity, specificity, precision, and accuracy. Data extraction and quality assessment were performed independently by several reviewers using tools like PROBAST and QUADAS-2. A qualitative synthesis as per SWiM guidelines was conducted due to heterogeneity in study designs and outcomes. CONCLUSION: AI shows significant promise in enhancing dermatological diagnostics and expanding access to dermatologic care in LMICs, with models achieving high accuracy (up to 99%) in tasks like skin cancer and infectious disease detection. However, challenges such as underrepresented skin tones in datasets, limited clinical validation, and infrastructural barriers currently hinder equitable implementation. Future efforts should prioritize creating and utilizing diverse datasets, lightweight models for mobile deployment, and human-AI collaboration to ensure context-specific and scalable solutions. Addressing these gaps can help leverage AI to mitigate global health disparities in dermatological care.","author":[{"family":"Uwishema","given":"Olivier"},{"family":"Ghezzawi","given":"Malak"},{"family":"Charbel","given":"Nicole"},{"family":"Alawieh","given":"Shireen"},{"family":"Roy","given":"S"},{"family":"Wojtara","given":"Magda"},{"family":"Hakayuwa","given":"Clyde"},{"family":"Jaafar","given":"Ibrahim"},{"family":"Nkurunziza","given":"Gerard"},{"family":"Prasad","given":"Manya"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12245-025-00975-4","URL":"https://doi.org/10.1186/s12245-025-00975-4","source":"openalex"},{"id":"oa:W4414752405","type":"article-journal","title":"The Effect of Artificial Intelligence in Promoting Positive Nursing Practice Environments: Mixed Methods Systematic Review","abstract":"AIM: To synthesise the available evidence on the effect of artificial intelligence in promoting positive nursing practice environments, exploring outcomes for professionals, clients, and institutions. BACKGROUND: Artificial intelligence has undergone significant advancements and shows great potential to transform nursing practice. However, this technological evolution is not without challenges, which must be identified and addressed. METHODS: A systematic mixed-methods review following the PRISMA 2020 guidelines and the methodology proposed by JBI. The search strategy was conducted in PubMed, CINAHL, Scopus, and Web of Science, including grey literature. Quantitative, qualitative, and mixed-methods studies were included, and the selection process involved screening by two independent reviewers, who assessed all studies, their methodological quality and extracted their data. RESULTS: From the conducted search, 11 studies were included, addressing how artificial intelligence has transformed nursing practice by optimising decision-making, task execution, and patient safety. Artificial intelligence, through predictive models and assistants such as ChatGPT, can enhance nursing management. However, challenges such as privacy concerns, resistance to change, and the need for professional training must be addressed to maximise its effectiveness. CONCLUSION: Artificial intelligence has the potential to positively transform the nursing practice environment, optimising decision-making, enhancing patient safety, and improving operational efficiency, with clear benefits for professionals, patients, and healthcare institutions. RELEVANCE TO CLINICAL PRACTICE: This study analysed the impact of artificial intelligence on nursing, highlighting improvements in clinical decision-making, patient safety, and institutional efficiency. Despite the identified benefits, the implementation of artificial intelligence in nursing is not without challenges and risks, which must be identified and addressed to ensure safe and effective adoption. REPORTING METHOD: The review followed the PRISMA 2020 checklist. PATIENT OR PUBLIC CONTRIBUTION: No.","author":[{"family":"Pereira","given":"Soraia"},{"family":"Ferreira","given":"Rosilene"},{"family":"Venturasilva","given":"João"},{"family":"Santos","given":"Eduardo"},{"family":"Fassarella","given":"Cí­ntia"},{"family":"Ribeiro","given":"Olga"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/jocn.70127","URL":"https://doi.org/10.1111/jocn.70127","source":"openalex"},{"id":"oa:W4408551077","type":"article-journal","title":"Role of Artificial Intelligence in Congenital Heart Disease and Interventions","abstract":"Artificial intelligence has promising impact on patients with congenital heart disease, a vulnerable population with life-long health care needs and, often, a substantially higher risk of death than the general population. This review explores the role artificial intelligence has had on cardiac imaging, electrophysiology, interventional procedures, and intensive care monitoring as it relates to children and adults with congenital heart disease. Machine learning and deep learning algorithms have enhanced not only imaging segmentation and processing but also diagnostic accuracy namely reducing interobserver variability. This has a meaningful impact in complex congenital heart disease improving anatomic diagnosis, assessment of cardiac function, and predicting long-term outcomes. Image processing has benefited procedural planning for interventional cardiology, allowing for a higher quality and density of information to be extracted from the same imaging modalities. In electrophysiology, deep learning models have enhanced the diagnostic potential of electrocardiograms, detecting subtle yet meaningful variation in signals that enable early diagnosis of cardiac dysfunction, risk stratification of mortality, and more accurate diagnosis and prediction of arrhythmias. In the congenital heart disease population, this has the potential for meaningful prolongation of life. Postoperative care in the cardiac intensive care unit is a data-rich environment that is often overwhelming. Detection of subtle data trends in this environment for early detection of morbidity is a ripe avenue for artificial intelligence algorithms to be used. Examples like early detection of catheter-induced thrombosis have already been published. Despite their great promise, artificial intelligence algorithms are still limited by hurdles such as data standardization, algorithm validation, drift, and explainability.","author":[{"family":"Holt","given":"Dudley"},{"family":"Elbokl","given":"Amr"},{"family":"Stromberg","given":"Daniel"},{"family":"Taylor","given":"Michael"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jscai.2025.102567","URL":"https://doi.org/10.1016/j.jscai.2025.102567","source":"openalex"},{"id":"oa:W4413399312","type":"article-journal","title":"Integrating artificial intelligence and optogenetics for Parkinson’s disease diagnosis and therapeutics in male mice","abstract":"Parkinson's disease (PD), a progressive neurodegenerative disorder, presents complex motor symptoms and lacks effective disease-modifying treatments. Here we show that integrating artificial intelligence (AI) with optogenetic intervention, termed optoRET, modulating c-RET (REarranged during Transfection) signalling, enables task-independent behavioural assessments and therapeutic benefits in freely moving male AAV-hA53T mice. Utilising a 3D pose estimation technique, we developed tree-based AI models that detect PD severity cohorts earlier and with higher accuracy than conventional methods. Employing an explainable AI technique, we identified a comprehensive array of PD behavioural markers, encompassing gait and spectro-temporal features. Moreover, our AI-driven analysis highlights that optoRET effectively alleviates PD progression by improving limb coordination and locomotion and reducing chest tremor. Our study demonstrates the synergy of integrating AI and optogenetic techniques to provide an efficient diagnostic method with extensive behavioural evaluations and sets the stage for an innovative treatment strategy for PD.","author":[{"family":"Hyeon","given":"Bobae"},{"family":"Shin","given":"Jaehyun"},{"family":"Lee","given":"Jae‐hun"},{"family":"Kim","given":"Woori"},{"family":"Kwon","given":"Jea"},{"family":"Lee","given":"Hee‐young"},{"family":"Kim","given":"Dae‐gun"},{"family":"Kim","given":"Choong"},{"family":"Kim","given":"Choong"},{"family":"Choi","given":"Sian"},{"family":"Jeong","given":"Jae‐woong"},{"family":"Kim","given":"Kwang‐soo"},{"family":"Lee","given":"CJ"},{"family":"Kim","given":"Daesoo"},{"family":"Kim","given":"Daesoo"},{"family":"Heo","given":"Won"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41467-025-63025-w","URL":"https://doi.org/10.1038/s41467-025-63025-w","source":"openalex"},{"id":"oa:W4416665897","type":"article-journal","title":"Intelligent Biosensors Based on Hyaluronic Acid Hydrogels for Monitoring Chronic Wound Healing with the Involvement of Artificial Intelligence","abstract":"Chronic wounds, such as those caused by diabetes, burns, and pressure ulcers, pose significant healthcare challenges due to their slow healing and potential for infections. Traditional methods for monitoring wound healing are often intrusive, slow, and lack real-time data. To overcome these limitations, innovative biosensors using hyaluronic acid hydrogels have emerged as a promising solution. As a non-intrusive, biocompatible platform, these biosensors can track pH, glucose levels, inflammatory proteins, and other key biomarkers as wounds heal. With the integration of artificial intelligence (AI), they enable personalized treatment adjustments and early complication detection through real-time data analysis and predictive modeling. This review discusses the recent progress of hyaluronic acid hydrogel biosensors for long-term wound healing, evaluating their strengths, challenges, and potential future improvements. This work aims to enhance chronic wound management and improve multiple clinical outcomes by focusing on the intersection of biomaterial innovation and AI.","author":[{"family":"Nicolae","given":"Antonia"},{"family":"Badea","given":"Mihaela"},{"family":"Bucurica","given":"Săndica"},{"family":"Rasaliu","given":"Florina"},{"family":"Constantinescu","given":"E"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bios15120773","URL":"https://doi.org/10.3390/bios15120773","source":"openalex"},{"id":"oa:W4413179134","type":"article-journal","title":"Application of artificial intelligence-based stemness index in cancer","abstract":"Cancer stem cells (CSCs) exhibit self-renewal and multidirectional differentiation capacities. The stemness of CSCs is the fundamental cause of tumor progression and treatment resistance. The stemness index, evaluating the number and activity of CSCs, is a crucial indicator predicting various aspects of tumor behavior such as growth, metastasis, and prognosis. With the advancements in artificial intelligence (AI), particularly in data analysis and machine learning, the identification and understanding of CSCs' stemness characteristics have improved. The AI-based analysis allows for processing vast datasets and recognizing patterns that assist in comprehending the role of CSCs in cancer development. The utilization of AI to analyze and compute the stemness index holds significant clinical relevance in tumor diagnosis and treatment. This approach provides more precise and personalized information, potentially influencing treatment strategies. Therefore, tailoring treatments specifically targeting CSCs is highly imperative and may enhance therapeutic efficacy and outcomes in cancer patients.","author":[{"family":"Liu","given":"Liyuan"},{"family":"Pei","given":"Qin"},{"family":"Qadir","given":"Javeria"},{"family":"Chen","given":"Yiyu"},{"family":"Li","given":"Jingyuan"},{"family":"Luo","given":"Yanan"},{"family":"Xian","given":"Jiawen"},{"family":"Du","given":"Rongrong"},{"family":"Ye","given":"Ting"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fonc.2025.1608712","URL":"https://doi.org/10.3389/fonc.2025.1608712","source":"openalex"},{"id":"oa:W4413313824","type":"article-journal","title":"Artificial intelligence in electroencephalography analysis for epilepsy diagnosis and management","abstract":"Introduction: Epilepsy is a prevalent chronic neurological disorder primarily diagnosed using electroencephalography (EEG). Traditional EEG interpretation relies on manual analysis, which suffers from high misdiagnosis rates and inefficiency. Methods: This review systematically evaluates the integration of artificial intelligence (AI), particularly deep learning (DL) and machine learning (ML), into EEG analysis for epilepsy management. We focus on two dominant AI-EEG application models: supportive AI (augmenting clinical decisions) and predictive AI (anticipating seizures or outcomes). Results: AI-based EEG analysis demonstrates significant potential in improving epilepsy detection, monitoring, and therapeutic evaluation. Key advancements include enhanced precision, efficiency, and capabilities for multimodal data fusion and personalized diagnosis. However, challenges persist, such as limited model interpretability, data quality constraints, and barriers to clinical translation. Crucially, AI outputs require clinician verification alongside multidimensional clinical data. Discussion: Future research must prioritize algorithm optimization, data quality improvement, and enhanced AI transparency. Interdisciplinary collaboration is essential to bridge the gap between technical innovation and clinical implementation. This review highlights both the transformative potential and current limitations of AI-EEG in epilepsy care, providing a roadmap for future developments.","author":[{"family":"Wang","given":"Chenxi"},{"family":"Yuan","given":"Xinyue"},{"family":"Jing","given":"Wei"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fneur.2025.1615120","URL":"https://doi.org/10.3389/fneur.2025.1615120","source":"openalex"},{"id":"oa:W7125421805","type":"article-journal","title":"Artificial Intelligence in Pediatric Dentistry: A Systematic Review and Meta-Analysis","abstract":"BACKGROUND/OBJECTIVES: Artificial intelligence (AI) has gained substantial prominence in pediatric dentistry, offering new opportunities to enhance diagnostic precision and clinical decision-making. AI-based systems are increasingly applied in caries detection, early childhood caries (ECC) risk prediction, tooth development assessment, mesiodens identification, and other key diagnostic tasks. This systematic review and meta-analysis aimed to synthesize evidence on the diagnostic performance of AI models developed specifically for pediatric dental applications. METHODS: A systematic search was conducted in PubMed, Scopus, Web of Science, and Embase following PRISMA-DTA guidelines. Studies evaluating AI-based diagnostic or predictive models in pediatric populations (≤18 years) were included. Reference screening, data extraction, and quality assessment were performed independently by two reviewers. Pooled sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated using random-effects models. Sources of heterogeneity related to imaging modality, annotation strategy, and dataset characteristics were examined. RESULTS: Thirty-two studies met the inclusion criteria for qualitative synthesis, and fifteen were eligible for quantitative analysis. For radiographic caries detection, pooled sensitivity, specificity, and AUC were 0.91, 0.97, and 0.98, respectively. Prediction models demonstrated good diagnostic performance, with pooled sensitivity of 0.86, specificity of 0.82, and AUC of 0.89. Deep learning architectures, particularly convolutional neural networks, consistently outperformed traditional machine learning approaches. Considerable heterogeneity was identified across studies, primarily driven by differences in imaging protocols, dataset balance, and annotation procedures. Beyond quantitative accuracy estimates, this review critically evaluates whether current evidence supports meaningful clinical translation and identifies pediatric domains that remain underrepresented in AI-driven diagnostic innovation. CONCLUSIONS: AI technologies exhibit strong potential to improve diagnostic accuracy in pediatric dentistry. However, limited external validation, methodological variability, and the scarcity of prospective real-world studies restrict immediate clinical implementation. Future research should prioritize the development of multicenter pediatric datasets, harmonized annotation workflows, and transparent, explainable AI (XAI) models to support safe and effective clinical translation.","author":[{"family":"Karamüftüoğlu","given":"Nevra"},{"family":"Üçpunar","given":"Büşra"},{"family":"Bi̇rben","given":"İrem"},{"family":"Altundağ","given":"Asya"},{"family":"Mullaoğlu","given":"Kübra"},{"family":"Bal","given":"Cenkhan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/children13010152","URL":"https://doi.org/10.3390/children13010152","source":"openalex"},{"id":"oa:W7134261441","type":"article-journal","title":"Navigation of drug discovery via artificial intelligence","abstract":"Abstract Background The growing global health burden necessitates precise therapeutics to mitigate pathogenesis and severe complications, which are increasing daily. Recently emerging and re-emerging viral infectious diseases, along with other ongoing complications from diverse infections and pathogens, contribute to global outbreaks. To address this, immediate and accurate therapeutic developments that can help manage this crisis are needed. Concerning the development of therapeutics, the conventional method-based drug design is time-consuming and requires a substantial investment of time to develop a drug against the pathogen successfully. Main body of the abstract To overcome these present obstacles, artificial intelligence (AI) came as a hope of revolutionizing the detection and advancement of pioneering, precise, cost- and time-effective drugs. AI uses advanced algorithms to improve the accuracy regarding target identification and further inhibitor selection. The pathogens were re-emerging daily, simultaneously, generating a huge amount of data with various specific properties and other essential details. Among them, some data can be helpful for therapeutic development. Using AI-based pipelines, tools, servers, databases, and useful resources to aid drug discovery, and employing different algorithms to examine the data, it was possible to identify a potential target that could aid therapeutic development; similarly, it also helped revolutionize the clinical aspects of drug discovery and the pharmaceutical industry by enabling more specific data handling. Moreover, it can help utilize available drugs and their significant details to address emerging and ongoing diseases through a drug repurposing-based approach using advanced AI-based computational analysis. Short conclusion Herein, this study offers the AI insight toward the drug discovery and development, how these approaches were utilized, and their advancements and challenges.","author":[{"family":"Mishra","given":"Saurav"},{"family":"Mamoudou","given":"Hamadou"},{"family":"Subba","given":"Akansha"},{"family":"Georrge","given":"John"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1186/s43094-026-00954-3","URL":"https://doi.org/10.1186/s43094-026-00954-3","source":"openalex"},{"id":"oa:W4408502459","type":"article-journal","title":"Systematic Review of Radiomics and Artificial Intelligence in Intracranial Aneurysm Management","abstract":"Intracranial aneurysms, with an annual incidence of 2%-3%, reflect a rare disease associated with significant mortality and morbidity risks when ruptured. Early detection, risk stratification of high-risk subgroups, and prediction of patient outcomes are important to treatment. Radiomics is an emerging field using the quantification of medical imaging to identify parameters beyond traditional radiology interpretation that may offer diagnostic or prognostic significance. The general radiomic workflow involves image normalization and segmentation, feature extraction, feature selection or dimensional reduction, training of a predictive model, and validation of the said model. Artificial intelligence (AI) techniques have shown increasing interest in applications toward vascular pathologies, with some commercially successful software including AiDoc, RapidAI, and Viz.AI, as well as the more recent Viz Aneurysm. We performed a systematic review of 684 articles and identified 84 articles exploring the applications of radiomics and AI in aneurysm treatment. Most studies were published between 2018 and 2024, with over half of articles in 2022 and 2023. Studies included categories such as aneurysm diagnosis (25.0%), rupture risk prediction (50.0%), growth rate prediction (4.8%), hemodynamic assessment (2.4%), clinical outcome prediction (11.9%), and occlusion or stenosis assessment (6.0%). Studies utilized molecular data (2.4%), radiologic data alone (51.2%), clinical data alone (28.6%), and combined radiologic and clinical data (17.9%). These results demonstrate the current status of this emerging and exciting field. An increased pace of innovation in this space is likely with the expansion of clinical applications of radiomics and AI in multiple vascular pathologies.","author":[{"family":"Owens","given":"Monica‐rae"},{"family":"Tenhoeve","given":"Samuel"},{"family":"Rawson","given":"Clayton"},{"family":"Azab","given":"Mohammed"},{"family":"Karsy","given":"Michael"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/jon.70037","URL":"https://doi.org/10.1111/jon.70037","source":"openalex"},{"id":"oa:W4412191972","type":"article-journal","title":"Do occupational health and safety tools that utilize artificial intelligence have a measurable impact on worker injury or illness? Findings from a systematic review","abstract":"BACKGROUND: Artificial intelligence (AI) holds promise as a tool that can be used by practitioners in the field of occupational health and safety (OHS). This study aimed to identify AI applications specifically used for OHS and examine their impact on worker morbidity or mortality outcomes. METHODS: We conducted a comprehensive systematic review. We searched six databases to identify published quantitative studies of OHS AI applications across the hierarchy of controls that were published between years 2018 to 2024. Title/abstract and full-text screening was conducted to identify eligible studies which were then assessed for quality and risk of bias and synthesized. RESULTS: Of the 1255 articles identified by our search, only two met eligibility criteria; one of which was appraised as medium quality and the other as low quality. The one medium quality study identified by our review was an AI-based chatbot health promotion tool which was shown to improve musculoskeletal symptoms. Our systematic review shows that we are at the early stages of understanding the role AI can play in OHS and it may be premature to recommend the wide-spread use of AI for health and safety practice within workplaces. CONCLUSION: There is a critical need for future research to unpack how considerations taken in the development and adoption of workplace AI tools for OHS can determine their effectiveness in addressing worker injury or illness. SYSTEMATIC REVIEW REGISTRATION: PROSPERO CRD42023414422.","author":[{"family":"Jetha","given":"Arif"},{"family":"Bakhtari","given":"Hela"},{"family":"Irvin","given":"Emma"},{"family":"Biswas","given":"Aviroop"},{"family":"Smith","given":"Maxwell"},{"family":"Mustard","given":"Cameron"},{"family":"Arrandale","given":"Victoria"},{"family":"Dennerlein","given":"Jack"},{"family":"Smith","given":"Peter"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s13643-025-02869-1","URL":"https://doi.org/10.1186/s13643-025-02869-1","source":"openalex"},{"id":"oa:W4411415408","type":"article-journal","title":"Evaluating the Role of Artificial Intelligence in Making Clinical Decisions for Treating Acute Pancreatitis","abstract":"Background/Objectives: Acute pancreatitis (AP) is an illness that requires prompt diagnosis and treatment since it has the potential to become life-threatening. The American College of Gastroenterology 2024 (ACG24) guidelines offer a framework for diagnosis, severity, and treatment criteria. To assess Google Gemini application of ACG24 guidelines to Medical Information Mart for Intensive Care-III AP cases for risk, nutrition, and complication management. Methods: This observational cross-sectional study was based on 512 patients with AP who were treated in the Medical Information Mart for Intensive Care-III database from 2001 to 2012. The study compared the efficiency of Gemini in relation to the ACG24 guidelines in the three main areas of risk stratification, enteral nutrition timing, and necrotizing pancreatitis management. Enteral nutrition, according to the ACG24 guidelines, should be started within 48 h for patients who are capable, and antibiotics should only be used for confirmed infected necrosis. Results: The study included 512 patients who were divided into two groups: 213 patients with mild pancreatitis (41.6%) and 299 patients with severe pancreatitis (58.4%). The model achieved 85% accuracy for mild cases and 82% accuracy for severe cases of pancreatitis. The Acute Physiology and Chronic Health Evaluation II and Ranson scores matched the predictions of Gemini for both mild cases (p = 0.28 and p = 0.33, respectively) and severe cases (p = 0.31 and p = 0.27, respectively). The recommendations for early enteral nutrition and delayed feeding in mild cases were correct for 78% of patients, but the system suggested oral intake prematurely in 8% of severe cases. The antibiotic guideline compliance reached 82% among 156 patients with necrotizing pancreatitis, and the procedure for draining infected necrosis was correct 85% of the time. Conclusions: The Gemini model achieved 78–85% accuracy in determining pancreatitis severity and adherence to treatment guidelines but showed lower accuracy in nutrition timing compared to other parameters. Core Tip: This study evaluated the Google Gemini model in applying the American College of Gastroenterology 2024 guidelines for acute pancreatitis across 512 Medical Information Mart for Intensive Care-III cases. Results demonstrated 85% accuracy in severity classification, precise prediction of Acute Physiology and Chronic Health Evaluation II and Ranson scores, and 78–85% compliance with nutritional and necrotizing pancreatitis management guidelines. These findings suggest that artificial intelligence-based clinical decision support systems can provide rapid, consistent, and guideline-concordant recommendations, which are particularly valuable in settings with limited specialist expertise.","author":[{"family":"Üçdal","given":"Mete"},{"family":"Bakhshandehpour","given":"Amir"},{"family":"Durak","given":"Muhammed"},{"family":"Balaban","given":"Yasemin"},{"family":"Kekilli","given":"Murat"},{"family":"Şimşek","given":"Cem"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jcm14124347","URL":"https://doi.org/10.3390/jcm14124347","source":"openalex"},{"id":"oa:W4412521550","type":"article-journal","title":"Demographic inaccuracies and biases in the depiction of patients by artificial intelligence text-to-image generators","abstract":"The wide usage of artificial intelligence (AI) text-to-image generators raises concerns about the role of AI in amplifying misconceptions in healthcare. This study therefore evaluated the demographic accuracy and potential biases in the depiction of patients by four commonly used text-to-image generators. A total of 9060 images of patients with 29 different diseases was generated using Adobe Firefly, Bing Image Generator, Meta Imagine, and Midjourney. Twelve independent raters determined the sex, age, weight, and race and ethnicity of the patients depicted. Comparison to the real-world epidemiology showed that the generated images failed to depict demographical characteristics such as sex, age, and race and ethnicity accurately. In addition, we observed an over-representation of White and normal weight individuals. Inaccuracies and biases may stem from non-representative and non-specific training data as well as insufficient or misdirected bias mitigation strategies. In consequence, new strategies to counteract such inaccuracies and biases are needed.","author":[{"family":"Wiegand","given":"Tim"},{"family":"Jung","given":"Leonard"},{"family":"Gudera","given":"Jonas"},{"family":"Schuhmacher","given":"Luisa"},{"family":"Moehrle","given":"Paulina"},{"family":"Rischewski","given":"Jon"},{"family":"Mehrzad","given":"Pardiss"},{"family":"Jeong","given":"Subin"},{"family":"Nguyen","given":"Lisa"},{"family":"Poeschla","given":"Michael"},{"family":"Velezmoro","given":"Laura"},{"family":"Kruk","given":"Linus"},{"family":"Dimitriadis","given":"Konstantinos"},{"family":"Koerte","given":"Inga"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01817-6","URL":"https://doi.org/10.1038/s41746-025-01817-6","source":"openalex"},{"id":"oa:W4406103327","type":"article-journal","title":"Intelligent Oil Production Management System Based on Artificial Intelligence Technology","abstract":"Production management serves as a pivotal component in the operational activities of oilfield sites, with the effectiveness of management practices directly influencing the success of developmental outcomes. To enhance the maintenance-free operational period of oil production systems, elevate management standards, and reduce overall operational costs, advanced technologies such as artificial intelligence (AI) and big data analytics have been strategically integrated into oilfield operations. These technologies are able to incorporate data resources from all stages of oilfield production, thus providing a comprehensive view of oilfield production and guidance for production. This study uses a series of diagnostic and predictive methods to construct a management system that allows for the comprehensive monitoring and fault diagnosis of oil production systems, which can ensure the intelligent management of oil production systems at multiple levels throughout their life cycle. Automated monitoring workflows and proactive analytical processes are at the heart of the framework, enabling real-time monitoring and predictive decision-making. This not only minimizes the likelihood of system failure but also optimizes resource allocation and operational efficiency.","author":[{"family":"Sui","given":"Xianfu"},{"family":"Lu","given":"Xin"},{"family":"Ji","given":"Yuchen"},{"family":"Yang","given":"Yang"},{"family":"Peng","given":"Jianlin"},{"family":"Li","given":"Menglong"},{"family":"Han","given":"Guoqing"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/pr13010133","URL":"https://doi.org/10.3390/pr13010133","source":"openalex"},{"id":"oa:W4409280922","type":"article-journal","title":"Pressure Injury Prediction in Intensive Care Units Using Artificial Intelligence: A Scoping Review","abstract":"Background/Objetives: Pressure injuries pose a significant challenge in healthcare, adversely impacting individuals’ quality of life and healthcare systems, particularly in intensive care units. The effective identification of at-risk individuals is crucial, but traditional scales have limitations, prompting the development of new tools. Artificial intelligence offers a promising approach to identifying and preventing pressure injuries in critical care settings. This review aimed to assess the extent of the literature regarding the use of artificial intelligence technologies in the prediction of pressure injuries in critically ill patients in intensive care units to identify gaps in current knowledge and direct future research. Methods: The review followed the Joanna Briggs Institute’s methodology for scoping reviews, and the study protocol was prospectively registered on the Open Science Framework platform. Results: This review included 14 studies, primarily highlighting the use of machine learning models trained on electronic health records data for predicting pressure injuries. Between 6 and 86 variables were used to train these models. Only two studies reported the clinical deployment of these models, reporting results such as reduced nursing workload, decreased prevalence of hospital-acquired pressure injuries, and decreased intensive care unit length of stay. Conclusions: Artificial intelligence technologies present themselves as a dynamic and innovative approach, with the ability to identify risk factors and predict pressure injuries effectively and promptly. This review synthesizes information about the use of these technologies and guides future directions and motivations.","author":[{"family":"Alves","given":"José"},{"family":"Azevedo","given":"RMD"},{"family":"Marques","given":"Ana"},{"family":"Encarnação","given":"Rúben"},{"family":"Alves","given":"Paulo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/nursrep15040126","URL":"https://doi.org/10.3390/nursrep15040126","source":"openalex"},{"id":"oa:W4408025939","type":"article-journal","title":"Can artificial intelligence diagnose seizures based on patients' descriptions? A study of GPT ‐4","abstract":"OBJECTIVE: Generalist large language models (LLMs) have shown diagnostic potential in various medical contexts but have not been explored extensively in relation to epilepsy. This paper aims to test the performance of an LLM (OpenAI's GPT-4) on the differential diagnosis of epileptic and functional/dissociative seizures (FDS) based on patients' descriptions. METHODS: GPT-4 was asked to diagnose 41 cases of epilepsy (n = 16) or FDS (n = 25) based on transcripts of patients describing their symptoms (median word count = 399). It was first asked to perform this task without additional training examples (zero-shot) before being asked to perform it having been given one, two, and three examples of each condition (one-, two, and three-shot). As a benchmark, three experienced neurologists performed this task without access to any additional clinical or demographic information (e.g., age, gender, socioeconomic status). RESULTS: In the zero-shot condition, GPT-4's average balanced accuracy was 57% (κ = .15). Balanced accuracy improved in the one-shot condition (64%, κ = .27), but did not improve any further in the two-shot (62%, κ = .24) and three-shot (62%, κ = .23) conditions. Performance in all four conditions was worse than the mean balanced accuracy of the experienced neurologists (71%, κ = .42). However, in the subset of 18 cases that all three neurologists had \"diagnosed\" correctly (median word count = 684), GPT-4's balanced accuracy was 81% (κ = .66). SIGNIFICANCE: Although its \"raw\" performance was poor, GPT-4 showed noticeable improvement having been given just one example of a patient describing epilepsy and FDS. Giving two and three examples did not further improve performance, but the finding that GPT-4 did much better in those cases correctly diagnosed by all three neurologists suggests that providing more extensive clinical data and more elaborate approaches (e.g., more refined prompt engineering, fine-tuning, or retrieval augmented generation) could unlock the full diagnostic potential of LLMs.","author":[{"family":"Ford","given":"Joseph"},{"family":"Pevy","given":"Nathan"},{"family":"Grünewald","given":"Richard"},{"family":"Howell","given":"Stephen"},{"family":"Reuber","given":"Markus"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/epi.18322","URL":"https://doi.org/10.1111/epi.18322","source":"openalex"},{"id":"oa:W7128476819","type":"article-journal","title":"The future of fundamental science led by generative closed-loop artificial intelligence","abstract":"Artificial intelligence is approaching the point at which it can complete the scientific cycle, from hypothesis generation to experimental design and validation, within a closed loop that requires little human intervention. Yet, the loop is not fully autonomous: humans still curate data, set hyperparameters, adjudicate interpretability, and decide what counts as a satisfactory explanation. As models scale, they begin to explore regions of hypothesis and solution space that are inaccessible to human reasoning because they are too intricate or alien to our intuitions. Scientists may soon rely on AI strategies they do not fully understand, trusting goals and empirical payoffs rather than derivations. This prospect forces a choice about how much control to relinquish to accelerate discovery while keeping outputs human relevant. The answer cannot be a blanket policy to deploy LLMs or any single paradigm everywhere. It demands principled matching of methods to domains, hybrid causal and neurosymbolic scaffolds around generative models, and governance that preserves plurality and counters recursive bias. Otherwise, recursive training and uncritical reuse risk model collapse in AI and an epistemic collapse in science, as statistical inertia amplifies flaws and narrows the investigation. We argue for graded autonomy in AI-conducted science: systems that can close the loop at machine speed, while remaining anchored to human priorities, verifiable mechanisms, and domain-appropriate forms of understanding.","author":[{"family":"Zenil","given":"Hector"},{"family":"Tegner","given":"Jesper"},{"family":"Abrahão","given":"Felipe"},{"family":"Lavin","given":"Alexander"},{"family":"Kumar","given":"Vipin"},{"family":"Frey","given":"Jeremy"},{"family":"Weller","given":"Adrian"},{"family":"Soldatova","given":"Larisa"},{"family":"Bundy","given":"Alan"},{"family":"Jennings","given":"Nicholas"},{"family":"Takahashi","given":"Koichi"},{"family":"Hunter","given":"Le"},{"family":"Džeroski","given":"Sašo"},{"family":"Briggs","given":"Andrew"},{"family":"Gregory","given":"Frederick"},{"family":"Gomes","given":"Carla"},{"family":"Rowe","given":"J"},{"family":"Evans","given":"James"},{"family":"Kitano","given":"H"},{"family":"King","given":"Ross"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/frai.2026.1678539","URL":"https://doi.org/10.3389/frai.2026.1678539","source":"openalex"},{"id":"oa:W7117679704","type":"article-journal","title":"Legal, ethical, and policy challenges of artificial intelligence translation tools in healthcare","abstract":"Artificial intelligence (AI) translation tools, such as Google Translate and ChatGPT, are increasingly used in healthcare for medical communication to overcome language barriers between patients and providers. While these tools offer accessible and efficient translation, their use raises significant legal, ethical, and policy concerns. Key patients' rights, including the rights to privacy, informed consent, and equitable access to care, may be compromised. Current European regulations, including the EU AI Act, General Data Protection Regulation, and Medical Devices Regulation, offer only partial protection, leaving important regulatory gaps. This study employs a mixed-methods approach combining legal doctrinal analysis of EU regulatory frameworks with manual content analysis of platform terms of service. It integrates interdisciplinary perspectives from bioethics, digital health, and health communication to evaluate the implications of AI-mediated translation in clinical care. Findings reveal persistent and overlapping risks: violations of data privacy, inaccuracies in translation, bias and discrimination, and unclear liability when errors occur. To mitigate these risks, we propose targeted policy interventions, including developing guidelines for AI translation use in healthcare settings. This article contributes to digital health policy debates by identifying legal pathways to regulate AI translation tools in healthcare, ensuring their use supports patients' rights and promotes health equity.","author":[{"family":"Kolfschooten","given":"Hannah"},{"family":"Goosen","given":"Simone"},{"family":"Oirschot","given":"Janneke"},{"family":"Schouten","given":"Barbara"},{"family":"Vajda","given":"Ildikó"},{"family":"Willems","given":"Luna"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12982-025-01277-z","URL":"https://doi.org/10.1186/s12982-025-01277-z","source":"openalex"},{"id":"oa:W4408412354","type":"article-journal","title":"Project-work Artificial Intelligence Integration Framework (PAIIF): Developing a CDIO-based framework for educational integration","abstract":"Artificial intelligence (AI) and generative AI (GenAI) have sparked confusion and concern regarding their impact on education. Beyond the assessment integrity risks that currently draw the most attention, technologies such as ChatGPT, Copilot, and Gemini have also been identified as tools that can support learning. Project work, especially when there is no single correct solution, provides a great opportunity for integration, fostering technology knowledge and higher learning standards. However, no AI-integration framework for project-based work is available, resulting in a limited understanding of how AI integration can occur or be maximized. To address this, a collaborative effort of 16 educators from 9 Australian universities has led to the development of a generic AI implementation framework, built upon the CDIO approach. With a focus on engineering education, this framework can be adapted to other project-based learning contexts, where educators can pick and choose the relevant implementation items as needed. This framework is called the Project-work Artificial Intelligence Integration Framework (PAIIF), and its development and structure are outlined here. Initial implementations have shown the effectiveness of promoting reflection and guidance on where and how AI integration can occur.","author":[{"family":"Nikolic","given":"Sasha"},{"family":"Quince","given":"Zach"},{"family":"Lindqvist","given":"Anna"},{"family":"Neal","given":"Peter"},{"family":"Grundy","given":"Sarah"},{"family":"Lim","given":"May"},{"family":"Tahmasebinia","given":"Faham"},{"family":"Rios","given":"Shannon"},{"family":"Burridge","given":"Josh"},{"family":"Petkoff","given":"Kathy"},{"family":"Chowdhury","given":"Ashfaque"},{"family":"Lee","given":"Wendy"},{"family":"Prestigiacomo","given":"Rita"},{"family":"Fernando","given":"HS"},{"family":"Lok","given":"Peter"},{"family":"Symes","given":"Mark"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3934/steme.2025016","URL":"https://doi.org/10.3934/steme.2025016","source":"openalex"},{"id":"oa:W4411223682","type":"article-journal","title":"Artificial intelligence for difficult airway assessment: a protocol for a systematic review with meta-analysis","abstract":"INTRODUCTION: Identifying difficult airways and avoiding unanticipated difficult airways through difficult airway assessment are crucial for patient safety prior to airway management. Therefore, accurately predicting difficult airways through airway assessment is a fundamental and significant technique in airway management by clinicians. Artificial intelligence (AI) is a rapidly evolving science with greater data processing ability than humans. AI, given its ever-expanding applications in medical diagnosis and disease prediction, has been employed to predict cases with difficult airways. Nevertheless, the diagnostic performance of AI algorithms for difficult airway assessment remains unclear due to the small sample sizes, insufficient image acquisition standards and poor predictive accuracies. Consequently, this study aims to formulate a protocol for a systematic review and meta-analysis to ascertain the diagnostic value of AI in assessing difficult airways. METHODS AND ANALYSIS: English-language databases (Cochrane Library, Web of Science, PubMed, Ovid Medline and Embase), Chinese electronic databases (China National Knowledge Infrastructure, VIP and Wanfang ] and clinical trial registry databases will be searched from their inception until January 2025 to identify clinical trials of AI for difficult airway assessment. Sensitivities, specificities, areas under the receiver operating characteristic curve, diagnostic likelihood ratios and diagnostic ORs with 95% CIs will be presented as indicators of AI's diagnostic accuracy in assessing difficult airways. Depending on the level of statistical heterogeneity evaluated by the I-square test, the fixed-effects or random-effects model will be employed. The risk of bias will be evaluated using the Quality Assessment of Diagnostic Accuracy Studies 2 tool. Furthermore, the quality of evidence concerning the outcomes will be assessed based on the Grading of Recommendations Assessment, Development and Evaluation criteria for diagnostic tests. Heterogeneity will be investigated through sensitivity, meta-regression and subgroup analyses. Additionally, Deeks' funnel plot asymmetry test will be used to detect publication bias. ETHICS AND DISSEMINATION: Ethical approval is not required for this systematic review protocol. The results will be disseminated through peer-reviewed publications. PROSPERO REGISTRATION NUMBER: CRD42023462926.","author":[{"family":"Zhang","given":"Weiyi"},{"family":"Du","given":"Li"},{"family":"Huang","given":"Yujie"},{"family":"Liu","given":"Dan"},{"family":"Li","given":"Tingting"},{"family":"Zheng","given":"Jianqiao"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/bmjopen-2024-096744","URL":"https://doi.org/10.1136/bmjopen-2024-096744","source":"openalex"},{"id":"oa:W4410992751","type":"article-journal","title":"Accuracy of Artificial Intelligence for Gatekeeping in Referrals to Specialized Care","abstract":"Importance: Integrating artificial intelligence (AI) technologies into gatekeeping holds significant potential, as it efficiently handles repetitive tasks and can process large amounts of information quickly. Objective: To develop and assess the accuracy of an AI model that enhances the gatekeeping process for referrals to specialized care. Design, Setting, and Participants: This diagnostic study comprised referrals from primary care to endocrinology, gastroenterology, proctology, rheumatology, and urology from a retrospective administrative database of patients in Brazil between June 2016 and April 2019. Analysis was performed between December 2022 and July 2024. Main Outcomes and Measures: The algorithm's development and testing comprised 2 stages. Multiple AI models were initially evaluated to train and test the algorithm for categorizing referrals as authorizing or requiring additional information. Subsequently, the model's performance was assessed against an independent set of referrals. Additionally, the current (human) evaluations of gatekeepers were evaluated against the standard. The reference standard was the consensus of 2 physicians with extensive experience. Accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC-ROC) were assessed. Results: The electronic system retrieved 45 039 eligible referrals for the development stage (mean [SD] patient age, 51.9 [15.8] years; 25 458 women [56.5%]). An algorithm utilizing word embeddings and a neural network proved the most effective. In the second phase, 1750 referrals (350 for each specialty) showed a 32% authorization rate according to the reference standard. The AI model achieved an overall accuracy of 0.716 (95% IC, 0.694-0.737), with a sensitivity of 0.542 (95% CI, 0.501 to 0.582) and specificity of 0.801 (95% CI, 0.777 to 0.822). Regarding individual specialties, rheumatology exhibited the highest accuracy (0.811; 95% IC, 0.767-0.849), while proctology had the lowest (0.649; 95% IC, 0.597-0.697). The overall AUC-ROC was 0.765 (95% IC, 0.742-0.788). When compared against the consensus standard, the AI model had higher accuracy and specificity and lower sensitivity than the current approach. Conclusions and Relevance: In this diagnostic study of referral data, a novel AI model effectively distinguished between referrals that warranted immediate authorization and those that required further information with moderate accuracy; it had higher specificity and lower sensitivity than gatekeepers decisions. Implementing this AI model in the gatekeeping process should combine human judgment and AI support to optimize the referral process.","author":[{"family":"Vergara","given":"Piter"},{"family":"Oliveira","given":"Jerônimo"},{"family":"Mattiello","given":"Rita"},{"family":"Montelongo","given":"Alfredo"},{"family":"Roman","given":"Rudi"},{"family":"Katz","given":"Natan"},{"family":"Wives","given":"Leandro"},{"family":"Rados","given":"Dimitris"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1001/jamanetworkopen.2025.13285","URL":"https://doi.org/10.1001/jamanetworkopen.2025.13285","source":"openalex"},{"id":"oa:W4406725043","type":"article-journal","title":"Cyber Espionage in the Age of Artificial Intelligence: A Comparative Study of State-Sponsored Campaign","abstract":"This study investigates the transformative role of artificial intelligence (AI) in state-sponsored cyber espionage, focusing on its dual use in offensive and defensive operations. Using data from the MITRE ATT&CK Framework, FireEye APT Groups Database, UNSW-NB15 Intrusion Detection Dataset, and the Cyber Conflict Tracker by CFR, this research applied network graph analysis, multi-criteria decision analysis (MCDA), ensemble classification models, and Difference-in-Differences (DiD) analysis. Results revealed that AI-driven offensive techniques, phishing (degree centrality 0.85), and adaptive malware (betweenness centrality 0.81) significantly enhance operational precision and scalability. Defensively, ensemble classification models achieved up to 95.8% accuracy, highlighting AI's efficacy in intrusion detection. AI regulatory frameworks reduced misattribution rates by 20% and escalation incidents by 10%, demonstrating their critical role in mitigating geopolitical risks. The findings impress AI's transformative potential in advancing cyber operations and shaping international policy and governance. By addressing challenges such as attribution, escalation risks, and ethical dilemmas, this study highlights the necessity for stronger global cooperation and regulatory frameworks to navigate the dual-use nature of AI, providing actionable insights for policymakers, cybersecurity professionals, and researchers, emphasizing the urgency of aligning technological advancements with strategies for enhancing global cybersecurity resilience.","author":[{"family":"Obioha-Val","given":"Onyinye"},{"family":"Olaniyi","given":"Oluwaseun"},{"family":"Gbadebo","given":"Michael"},{"family":"Balogun","given":"Adebayo"},{"family":"Olisa","given":"Anthony"}],"issued":{"date-parts":[[2025]]},"DOI":"10.9734/ajrcos/2025/v18i1557","URL":"https://doi.org/10.9734/ajrcos/2025/v18i1557","source":"openalex"},{"id":"oa:W4413764808","type":"article-journal","title":"AI-induced Deskilling in Medicine: A Mixed-Method Review and Research Agenda for Healthcare and Beyond","abstract":"Abstract The integration of Artificial Intelligence (AI) in healthcare is reshaping clinical practice, offering both opportunities for enhanced decision-making and risks of skill degradation among medical professionals. This growing impact calls for a comprehensive evaluation of its effects on medical expertise. This study presents a mixed-method literature review, combining systematic analysis with narrative synthesis to examine AI-induced deskilling and upskilling inhibition-the erosion of medical expertise and the reduction of opportunities for skill acquisition due to AI-driven decision support systems. Anchoring the discussion in the core medical competencies outlined by the Federation of Royal Colleges of Physicians of the UK-Practical Assessment of Clinical Examination Skills (PACES-MRCPUK), the systematic review identifies key vulnerabilities in physical examination, differential diagnosis, clinical judgment, and physician-patient communication. The narrative review explores broader themes related to Human–AI Interaction and the Impact of AI on Human Skills in Organizations. In response to concerns about the Second Singularity -a scenario in which decision-making autonomy is increasingly ceded to AI, weakening human oversight-this review advocates for a research agenda that prioritizes longitudinal studies, real-time monitoring of AI’s impact, and the development of frameworks to mitigate skill erosion, ensuring the preservation of professional autonomy and the safeguarding of the irreplaceable elements of human judgment in medicine and beyond.","author":[{"family":"Natali","given":"Chiara"},{"family":"Marconi","given":"Luca"},{"family":"Duran","given":"Leslye"},{"family":"Cabitza","given":"Federico"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s10462-025-11352-1","URL":"https://doi.org/10.1007/s10462-025-11352-1","source":"openalex"},{"id":"oa:W7135227096","type":"article-journal","title":"Artificial Intelligence and the Transformation of Cell and Gene Therapy Development","abstract":"Cell and Gene Therapy (CGT) represents a paradigm shift in medicine, offering curative potential for previously intractable diseases. However, the complexity, high cost, and manufacturing challenges inherent in developing, producing, and administering these therapies hinder their widespread accessibility. This review examines the critical and increasingly synergistic role of Artificial Intelligence (AI) and Machine Learning (ML) in overcoming these barriers across the entire CGT lifecycle, from discovery and construct design to smart manufacturing, clinical translation, and regulatory applications. We analyze how AI-driven approaches fundamentally differ from conventional methods, facilitating rapid construct optimization, generating highly predictive translational models, enabling the vision of autonomous, digital-twin-driven manufacturing, and establishing new paradigms for pharmacovigilance and regulatory oversight. The integration of AI is not merely an incremental improvement but a foundational transformation, positioning CGT to move from niche, bespoke treatments to scalable, accessible, and highly personalized medical modalities. We conclude by discussing current gaps, particularly data scarcity and regulatory uncertainty, and outlining a roadmap to realize the full potential of AI-enabled CGT.","author":[{"family":"Auclair","given":"Jared"},{"family":"Joung","given":"Jeewon"},{"family":"Singh","given":"Maya"},{"family":"Debauve","given":"Gaël"},{"family":"Singh","given":"Rominder"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/pharmaceutics18030356","URL":"https://doi.org/10.3390/pharmaceutics18030356","source":"openalex"},{"id":"oa:W4410125230","type":"article-journal","title":"The Unexpected Harms of Artificial Intelligence in Healthcare: Reflections on Four Real-World Cases","abstract":"INTRODUCTION: Rapid advances in Artificial Intelligence (AI), especially with large language models, present both opportunities and challenges in healthcare. This article analyzes real-world AI-related harms in healthcare. METHODS: We selected four recent AI-related incidents from the AIAAIC Repository. RESULTS: The incidents discussed include: Whisper's harmful hallucinations; UNOS's algorithm delaying transplants for black patients; the WHO's S.A.R.A.H. chatbot providing inaccurate health information; and Character AI's chatbot promoting disordered eating among teens. DISCUSSION AND CONCLUSION: These incidents highlight diverse risks, from misinformation to safety concerns, involving both industry and institutional providers. The article emphasizes the need for systematic reporting of AI-related harms, concerns about security, privacy, and ethics, and calls for a centralized health-specific database to enhance patient safety and understanding.","author":[{"family":"Denecke","given":"Kerstin"},{"family":"Lópezcampos","given":"Guillermo"},{"family":"Rivera-Romero","given":"Octavio"},{"family":"Gabarrón","given":"Elia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3233/shti250219","URL":"https://doi.org/10.3233/shti250219","source":"openalex"},{"id":"oa:W4410300494","type":"article-journal","title":"Artificial Intelligence Applications in Obstetric Risk Prediction: A Systematic Review of Machine Learning Models for Preeclampsia","abstract":"Preeclampsia remains a leading cause of maternal and perinatal morbidity and mortality worldwide. While traditional prediction models have shown limited accuracy, machine learning (ML) approaches offer promising alternatives by handling complex, non-linear relationships in multidimensional datasets. This systematic review evaluates the performance, methodological quality, and clinical applicability of ML models for preeclampsia prediction. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines, we searched five databases (PubMed, Embase, Scopus, Web of Science, and Cochrane Library) for studies published until April 15, 2025, and included studies that developed or validated ML models predicting preeclampsia. Risk of bias was assessed using the Prediction Model Risk-of-Bias Assessment Tool (PROBAST). Eleven studies (n = 11, comprising 116,253 pregnancies) were included. Ensemble methods (XGBoost, Random Forest) demonstrated superior performance, with area under the curve (AUCs) ranging from 0.84 to 0.973. Key predictors included mean arterial pressure, prior preeclampsia, and the biomarkers placental growth factor (PlGF) and pregnancy-associated plasma protein A (PAPP-A). Seven studies (63.6%) showed low overall risk of bias, while three (27.3%) had high risk due to analytical limitations. Only three studies (27.3%) conducted external validation. ML models, particularly ensemble methods, show excellent discriminative ability for preeclampsia prediction. However, heterogeneity in predictors and limited external validation constrain clinical translation. Future research should prioritize prospective validation studies with standardized outcome definitions and predictor sets.","author":[{"family":"Dkeen","given":"Nagla"},{"family":"Radwan","given":"Madina"},{"family":"Zumam","given":"Israa"},{"family":"Mohamed","given":"Nihal"},{"family":"Abdelmahmoud","given":"Eman"},{"family":"Magboul","given":"Nisrin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.83961","URL":"https://doi.org/10.7759/cureus.83961","source":"openalex"},{"id":"oa:W4416390169","type":"article-journal","title":"Applications of Machine Learning and Artificial Intelligence in Tropospheric Ozone Research","abstract":"Abstract. Machine learning (ML) is transforming atmospheric chemistry, offering powerful tools to address challenges in tropospheric ozone research, a critical area for climate resilience and public health. As in adjacent fields, ML approaches complement existing research by learning patterns from ever-increasing volumes of atmospheric and environmental data relevant to ozone. We highlight the rapid progress made in the field since Phase 1 of the Tropospheric Ozone Assessment Report (TOAR), focussing particularly on the most active areas of research, namely short-term ozone forecasting, emulation of atmospheric chemistry and the use of remote sensing for ozone estimation. This review provides a comprehensive synthesis of recent advancements, highlights critical challenges, and proposes actionable pathways to develop ML in ozone research. Further advances hinge on addressing domain-specific issues such as the dependence of ozone concentrations on several poorly observed precursor species, as well as making progress on generic ML challenges such as the definition of suitable benchmarks and developing robust, explainable models. Reaping the full potential of ML for ozone research and operational applications will require close collaborations across atmospheric chemistry, ML and computational science and vigilant pursuit of the rapid developments in adjacent fields.","author":[{"family":"Hickman","given":"Sebastian"},{"family":"Kelp","given":"Makoto"},{"family":"Griffiths","given":"Paul"},{"family":"Doerksen","given":"Kelsey"},{"family":"Miyazaki","given":"Kazuyuki"},{"family":"Pennington","given":"Elyse"},{"family":"Koren","given":"Gerbrand"},{"family":"Iglesiassuarez","given":"Fernando"},{"family":"Schultz","given":"Martin"},{"family":"Chang","given":"Kai‐lan"},{"family":"Cooper","given":"Owen"},{"family":"Archibald","given":"Alexander"},{"family":"Sommariva","given":"Roberto"},{"family":"Carlson","given":"David"},{"family":"Wang","given":"Hantao"},{"family":"West","given":"JJ"},{"family":"Liu","given":"Zhenze"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5194/gmd-18-8777-2025","URL":"https://doi.org/10.5194/gmd-18-8777-2025","source":"openalex"},{"id":"oa:W4415564840","type":"article-journal","title":"Artificial Intelligence in Digestive Endoscopy Training—The Past, Present, and Future","abstract":"BACKGROUND AND OBJECTIVE: Artificial intelligence (AI) is reshaping gastrointestinal endoscopy, yet its role in training remains unexplored. This narrative review summarizes current evidence on AI-assisted endoscopy training, addresses potential drawbacks, and envisions future directions. METHODS: This narrative review was performed via a systematic MEDLINE search (including articles from inception to January 2025), with search terms covering 'AI', 'endoscopy,' and 'training.' Studies were excluded if they were reviews, letters, editorials or comments; focused solely on model development; lacked a training component; or were limited to simple comparisons between the performance of endoscopists and AI systems. After screening 1443 records, 27 articles were included in this review. RESULTS: AI demonstrates potential in enhancing the training of various types of endoscopy (including luminal, hepatobiliary, capsule, and therapeutic endoscopy) by improving quality metrics, enhancing lesion detection, and guiding anatomical landmark recognition, yet the current applications are mainly task-based. Future AI must evolve to provide comprehensive training and personalized performance tracking to endoscopists of different levels of experience. Further studies are needed to assess objective educational outcomes and cost-effectiveness. Key concerns for AI adoption, including deskilling, over-reliance, ethical considerations, and practicality, should be addressed through structured implementation, quality assurance, and regulatory framework. CONCLUSION: In conclusion, AI can augment endoscopy training by improving skill acquisition and procedural quality, yet significant gaps remain. More research is needed to support its widespread integration.","author":[{"family":"Ho","given":"Jacky"},{"family":"Qian","given":"Zhouyao"},{"family":"Lau","given":"Louis"},{"family":"Yip","given":"Hon"},{"family":"Chiu","given":"Philip"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/den.70047","URL":"https://doi.org/10.1111/den.70047","source":"openalex"},{"id":"oa:W4415147093","type":"article-journal","title":"Artificial intelligence in osteoarthritis research: summary of the 2025 OARSI pre-congress workshop","abstract":"Objective: Artificial intelligence (AI) is transforming musculoskeletal research, offering new approaches to diagnosis, prognosis, and patient management in osteoarthritis (OA). However, implementation and ethical challenges persist. This manuscript summarizes insights from the OARSI 2025 Pre-Congress Workshop on Artificial Intelligence in Osteoarthritis Research, highlighting opportunities and challenges in applying AI across biomechanics, imaging, and clinical research domains. Design: The workshop, organized by the OARSI Early Career Investigator Committee and co-chaired by Drs. Matthew Harkey and Brooke Patterson, convened experts to discuss the use of AI in real-world biomechanics data collection, radiomics for imaging-based biomarkers, and large language models (LLMs) for clinical and research applications. Emphasis was placed on the need for interdisciplinary collaboration and ethical oversight. Results: In biomechanics, AI-driven markerless motion capture and wearable sensors enable scalable, ecologically valid data collection, though issues such as class imbalance, data privacy, and model interpretability remain. In imaging, radiomics and deep learning models show promise for early OA detection and progression prediction but face challenges in domain adaptation and external validation. In clinical research, LLMs can streamline documentation and thematic analysis but must address concerns around bias, data security, and transparency. Across domains, transparency, reproducibility, and ethical use of AI were emphasized as critical for maintaining scientific rigor. Conclusions: Cross-disciplinary collaboration and AI literacy are essential to responsibly advance AI integration in OA research. The workshop's collective insights call for ethical, patient-centered approaches that leverage AI's strengths while preserving research integrity and trust.","author":[{"family":"Harkey","given":"Matthew"},{"family":"Costello","given":"KE"},{"family":"Mehta","given":"Bella"},{"family":"Wen","given":"Chunyi"},{"family":"Malfait","given":"Anne‐marie"},{"family":"Madry","given":"Henning"},{"family":"Patterson","given":"Brooke"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.ocarto.2025.100687","URL":"https://doi.org/10.1016/j.ocarto.2025.100687","source":"openalex"},{"id":"oa:W4409169023","type":"article-journal","title":"Artificial Intelligence in Placental Pathology: New Diagnostic Imaging Tools in Evolution and in Perspective","abstract":"Artificial intelligence (AI) has emerged as a transformative tool in placental pathology, offering novel diagnostic methods that promise to improve accuracy, reduce inter-observer variability, and positively impact pregnancy outcomes. The primary objective of this review is to summarize recent developments in AI applications tailored specifically to placental histopathology. Current AI-driven approaches include advanced digital image analysis, three-dimensional placental reconstruction, and deep learning models such as GestAltNet for precise gestational age estimation and automated identification of histological lesions, including decidual vasculopathy and maternal vascular malperfusion. Despite these advancements, significant challenges remain, notably dataset heterogeneity, interpretative limitations of current AI algorithms, and issues regarding model transparency. We critically address these limitations by proposing targeted solutions, such as augmenting training datasets with annotated artifacts, promoting explainable AI methods, and enhancing cross-institutional collaborations. Finally, we outline future research directions, emphasizing the refinement of AI algorithms for routine clinical integration and fostering interdisciplinary cooperation among pathologists, computational researchers, and clinical specialists.","author":[{"family":"Damati","given":"Antonio"},{"family":"Baldini","given":"Giorgio"},{"family":"Difonzo","given":"Tommaso"},{"family":"Santoro","given":"Angela"},{"family":"Dellino","given":"Miriam"},{"family":"Cazzato","given":"Gerardo"},{"family":"Malvasi","given":"Antonio"},{"family":"Vimercati","given":"Antonella"},{"family":"Resta","given":"Leonardo"},{"family":"Zannoni","given":"Gian"},{"family":"Cascardi","given":"Eliano"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jimaging11040110","URL":"https://doi.org/10.3390/jimaging11040110","source":"openalex"},{"id":"oa:W4416873224","type":"article-journal","title":"Breast Cancer Diagnosis With Explainable Artificial Intelligence (XAI): Uncovering Strengths and Biases","abstract":"Breast cancer is one of the most common malignancies afflicting women globally, necessitating the use of cutting-edge AI approaches in diagnostic procedures to improve patient outcomes drastically. However, one key difficulty with the most recent AI models is a lack of transparency, making it difficult for medical practitioners to implement these technologies to increase diagnostic accuracy. Many explainable AI (XAI) solutions have been developed to overcome this issue. This survey focuses on applying XAI approaches in breast cancer detection and diagnosis, particularly emphasizing their role in increasing model transparency and clinical decision-making. The article also provides insights into the inherent biases in the most recent machine learning models towards specific XAI approaches, such as the compatibility of Convolutional Neural Networks (CNNs) with visual explanation methods and tree-based models with feature significance evaluations. Finally, the article covers the obstacles to using XAI technologies in clinical practice and the significance of defining standard measures for assessing their performance.","author":[{"family":"Bai","given":"Samita"},{"family":"Nasir","given":"Sidra"},{"family":"Khan","given":"Rizwan"},{"family":"Meyer","given":"Alexandre"},{"family":"Konik","given":"Hubert"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1109/access.2025.3639184","URL":"https://doi.org/10.1109/access.2025.3639184","source":"openalex"},{"id":"oa:W4411710523","type":"article-journal","title":"Artificial Intelligence‐Based Detection of Central Retinal Artery Occlusion Within 4.5 Hours on Standard Fundus Photographs","abstract":"BACKGROUND: Prompt diagnosis of acute central retinal artery occlusion (CRAO) is crucial for therapeutic management and stroke prevention. However, most stroke centers lack onsite ophthalmic expertise before considering fibrinolytic treatment. This study aimed to develop, train, and test a deep learning system to detect hyperacute CRAO on retinal fundus photographs within the critical 4.5-hour treatment window and up to 24 hours after visual loss to aid in secondary stroke prevention. METHODS: Our retrospective, cross-sectional study included 1322 color fundus photographs from 771 patients with acute visual loss due to CRAO, central retinal vein occlusion, nonarteritic anterior ischemic optic neuropathy, and healthy controls. Photographs were collected from 9 expert neuro-ophthalmology centers in 6 countries, including 3 randomized clinical trials. Training included 1039 photographs (517 patients), followed by testing on 2 data sets: (1) hyperacute CRAO (54 photographs, 54 patients) and (2) CRAO within 24 hours after visual loss (110 photographs, 109 patients). RESULTS: The deep learning system achieved an area under the receiver operating characteristic curve of 0.96 (95% confidence interval (CI), 0.95-0.98), a sensitivity of 92.6% (95% CI, 87.0-98.0), and a specificity of 85.0% (95% CI, 81.8-92.8) for detecting CRAO at hyperacute stage, with similar results within 24 hours. The deep learning system outperformed stroke neurologists on a subset of hyperacute testing data set (120 photographs, 120 patients). CONCLUSIONS: A deep learning system can accurately detect hyperacute CRAO on retinal photographs within a time window compatible with urgent fibrinolysis. If further validated, such systems could improve patient selection for fibrinolytic trials and optimize secondary stroke prevention. REGISTRATION: URL: https://www.clinicaltrials.gov; Unique identifier: NCT06390579.","author":[{"family":"Gungor","given":"Ayse"},{"family":"Sarbout","given":"Ilias"},{"family":"Gilbert","given":"Aubrey"},{"family":"Hamann","given":"Steffen"},{"family":"Lebranchu","given":"Pierre"},{"family":"Hobeanu","given":"Cristina"},{"family":"Gohier","given":"Philippe"},{"family":"Vignalclermont","given":"Catherine"},{"family":"Dumitrascu","given":"Oana"},{"family":"Cohen","given":"Salomon"},{"family":"Lagrèze","given":"Wolf"},{"family":"Feltgen","given":"Nicolas"},{"family":"Heide","given":"Frank"},{"family":"Lamirel","given":"C"},{"family":"Jonas","given":"Jost"},{"family":"Obadia","given":"Michaël"},{"family":"Racoceanu","given":"Daniel"},{"family":"Miléa","given":"Dan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1161/jaha.124.041441","URL":"https://doi.org/10.1161/jaha.124.041441","source":"openalex"},{"id":"oa:W4413499075","type":"article-journal","title":"Artificial intelligence for the science of evidence synthesis: how good are AI-powered tools for automatic literature screening?","abstract":"BACKGROUND: Literature screening constitutes a critical component in evidence synthesis; however, it typically requires substantial time and human resources. Artificial intelligence (AI) has shown promise in this field, yet the accuracy and effectiveness of AI tools for literature screening remain uncertain. This study aims to evaluate the performance of several existing AI-powered automated tools for literature screening. METHODS: This diagnostic accuracy study employed a cohort to evaluate the performance of five AI tools-ChatGPT 4.0, Claude 3.5, Gemini 1.5, DeepSeek-V3, and RobotSearch-in literature screening. We selected a random sample of 1,000 publications from a well-established literature cohort, with 500 as randomized controlled trials (RCTs) group and 500 as others group. Diagnostic accuracy was measured using several metrics, including the false negative fraction (FNF), time used for screening, false positive fraction (FPF), and the redundancy number needed to screen. RESULTS: We reported the FNF for the RCTs group and the FPF for the others group. In the RCTs group, RobotSearch exhibited the lowest FNF at 6.4% (95% CI: 4.6% to 8.9%), whereas Gemini exhibited the highest at 13.0% (95% CI: 10.3% to 16.3%). In the others group, the FPF of the four large language models ranged from 2.8% (95% CI: 1.7% to 4.7%) to 3.8% (95% CI: 2.4% to 5.9%), both of which were significantly lower than RobotSearch's rate of 22.2% (95% CI: 18.8% to 26.1%). In terms of screening efficiency, the mean time used for screening per article was 1.3 s for ChatGPT, 6.0 s for Claude, 1.2 s for Gemini, and 2.6 s for DeepSeek. CONCLUSIONS: The AI tools assessed in this study demonstrated commendable performance in literature screening; however, they are not yet suitable as standalone solutions. These tools can serve as effective auxiliary aids, and a hybrid approach that integrates human expertise with AI may enhance both the efficiency and accuracy of the literature screening process.","author":[{"family":"Ruan","given":"Minghao"},{"family":"Fan","given":"Junhao"},{"family":"Liu","given":"Mingkai"},{"family":"Meng","given":"Zhefeng"},{"family":"Zhang","given":"Xiaohai"},{"family":"Zhang","given":"Chengjing"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12874-025-02644-9","URL":"https://doi.org/10.1186/s12874-025-02644-9","source":"openalex"},{"id":"oa:W4415132431","type":"article-journal","title":"Explainable artificial intelligence in the talent recruitment process-a literature review","abstract":"With the widespread application of artificial intelligence in talent recruitment, the ‘black-box’ nature of AI, which leads to insufficient transparency and interpretability in decision-making, has gradually become a key challenge. This paper reviews the application of Explainable AI technologies throughout the entire talent recruitment process, aiming to analyze the role of XAI in enhancing decision-making transparency and traceability. The study finds that XAI, through interpretable algorithms such as LIME, SHAP, knowledge graphs, and causal reasoning, has significantly improved semantic understanding in resume parsing, precision in person-job recommendations, and analytical capabilities in interview evaluations. However, potential biases in generative large models, insufficient cross-scenario interpretability, and computational efficiency issues remain major bottlenecks. Future research should focus on dynamic fairness constraints and the integration of lightweight interpretability tools to promote the coordinated development of XAI in terms of ethical compliance, user trust, and practicality.","author":[{"family":"Zhang","given":"Gening"},{"family":"Pan","given":"Lin"},{"family":"Tang","given":"Fang"},{"family":"Yao","given":"Feng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/23311975.2025.2570881","URL":"https://doi.org/10.1080/23311975.2025.2570881","source":"openalex"},{"id":"oa:W7118808027","type":"article-journal","title":"Research That Matters: A Call for Enhancing Rigour and Relevance in Artificial Intelligence Research in Endodontics","abstract":"Artificial intelligence (AI) is emerging as a powerful tool in dentistry, where endodontics can benefit significantly from its potential applications. From enhanced diagnostic accuracy to treatment planning and decision-making, optimisation, and outcome prediction, AI may significantly improve the clinical practice and the teaching of endodontics (Mohammad-Rahimi et al. 2024; Ourang et al. 2024). However, this still remains a largely unfulfilled promise, because many studies suffer from significant limitations or methodological flaws. Several factors contribute to this condition. First, as often seen with the advent of novel technology, there is a lack of scientific expertise in AI methodology by editors and reviewers, which may result in the eventual publication of studies that display limitations or, sometimes, even fundamental flaws in design, implementation, or validation. Conversely, technical publications may present sophisticated algorithms that solve problems with limited clinical utility or fail to address the nuanced challenges of real-world endodontic practice, mainly due to a lack of endodontic domain experts who could bridge the gap between AI engineers and the field of endodontics. Consequently, literature becomes populated with low-quality studies with oversold conclusions, but no conceivable clinical applicability. In this paper, the authors critically present the common pitfalls observed in current AI research in dentistry, and more specifically, in endodontics and then offer practical solutions to address these pitfalls in future research. The goal is to provide constructive guidance that can help establish rigorous standards for a field that is transitioning from research to real-life application. While we aim to present ideal scenarios that highlight these challenges, it is essential to acknowledge that every research study has limitations and shortcomings. These observations are not a critique of the researchers' intent but a call for consideration of these factors to enhance future work in the field. Foundational research has been conducted, and basic AI applications have been developed for endodontic applications. The next steps in endodontic AI research will have to address clinical translation and relevance which may often have been overlooked. This pitfall includes two interconnected challenges, but in distinct ways: developing AI models that solve problems of limited clinical significance and conducting studies under conditions that cannot translate to the real world. Some AI studies focus on tasks that, while technically interesting, have limited practical value in endodontic practice. Thus, the fundamental question researchers must ask is: Does this AI application solve a problem that clinicians face in daily practice, and would its solution meaningfully improve patient care or clinical workflow? Without this foundation, even technically sophisticated and highly performant models may risk being solutions in search of problems. Another core problem in AI studies is using non-representative training datasets. Some studies developed AI models under idealised or artificial conditions and then reported results as if they were clinically meaningful. This gap between experimental success and clinical utility may lead to a false sense of success. It impairs translation to practice, as models that perform well in controlled settings can fail or mislead in real clinical settings (Ibrahim et al. 2021). For example, there are AI models developed on ex vivo samples, phantom images, or synthetic entities (e.g., artificial lesions or artificial root cracks) that simplify the task. Training and testing an AI model on the detection of artificial root cracks in radiographic images of extracted teeth does not imply that the model can detect vertical root fractures in radiographic images of real patients. While convenient for initial experimentation, like pilot or proof-of-concept studies, such data lack the complexit","author":[{"family":"Mohammadrahimi","given":"Hossein"},{"family":"Ramani","given":"Rishi"},{"family":"Setzer","given":"Frank"},{"family":"Schwendicke","given":"Falk"},{"family":"Pauwels","given":"Ruben"},{"family":"Nosrat","given":"Ali"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1111/iej.70094","URL":"https://doi.org/10.1111/iej.70094","source":"openalex"},{"id":"oa:W4411905413","type":"article-journal","title":"Exploring the experiences and perceptions of nursing students in utilizing artificial intelligence: a descriptive phenomenological study","abstract":"BACKGROUND: Artificial intelligence is transforming healthcare by enhancing diagnostics, efficiency, and outcomes. Evaluating nursing students' knowledge helps identify educational gaps, as they directly engage in patient care and decision-making. This study explored their experiences with using artificial intelligence in clinical settings. METHODS: A descriptive phenomenological approach was employed to recruit eight nursing students from the School of Nursing and Midwifery in Kermanshah-Iran, all of whom had prior experience with artificial intelligence applications in clinical settings. Participants were selected purposefully and underwent in-depth, semi-structured interviews. Eight interview sessions lasting 60-70 min each were conducted, with data saturation achieved after the eighth interview. The data were analyzed using Colaizzi's seven-step method, ensuring adherence to Guba and Lincoln's criteria for trustworthiness throughout the research process. Data management was facilitated using MAXQDA 20 software. RESULTS: Data analysis yielded 450 initial codes, 5 sub-themes, and 4 themes. The themes identified were: (1) Novel Potentials and Broad Horizons of Artificial Intelligence, (2) Ethical and Security Challenges In The World of Artificial Intelligence, (3) Changes in Human Identity and Social Relationships In The Age of Artificial Intelligence, and (4) Enhancing and Reinventing Skills for Coexistence With Artificial Intelligence. CONCLUSION: The study revealed that nursing students recognize the potential of artificial intelligence to enhance precision and efficiency in patient care. However, they emphasized the need for further education and the removal of barriers to its effective implementation. CLINICAL TRIAL NUMBER: Not applicable.","author":[{"family":"Nezhad","given":"Maysam"},{"family":"Abdi","given":"Alireza"},{"family":"Ahmadi","given":"Mahnaz"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12912-025-03392-3","URL":"https://doi.org/10.1186/s12912-025-03392-3","source":"openalex"},{"id":"oa:W4416061820","type":"article-journal","title":"Explainable Artificial Intelligence in Echocardiography","abstract":"Recent advancements in artificial intelligence (AI) have generated novel opportunities and challenges in ultrasound imaging. Deep learning algorithms exhibit significant potential in analyzing echocardiographic images, encompassing tasks such as view classification, quantification of cardiac function, and the diagnosis and risk assessment of cardiac diseases. The “black box” nature of AI models limits their clinical applications. Adopting explainable artificial intelligence (XAI) methods is crucial for improving the transparency and understanding of model predictions. This paper reviews the progress of AI applications in echocardiography, with a particular emphasis on XAI as a technical solution to enhance the transparency of model decision-making and its benefits compared to traditional AI models. This review outlines recent advancements in XAI applications for echocardiography and their clinical implications.","author":[{"family":"Hu","given":"Xinghong"},{"family":"Zhu","given":"Ye"},{"family":"Zhang","given":"Zisang"},{"family":"Quan","given":"Yuanting"},{"family":"Chen","given":"Wenwen"},{"family":"Chen","given":"Leichong"},{"family":"Xu","given":"Guangyu"},{"family":"Qin","given":"Luning"},{"family":"Xie","given":"Mingxing"},{"family":"Zhang","given":"Li"}],"issued":{"date-parts":[[2025]]},"DOI":"10.26599/audt.2025.250089","URL":"https://doi.org/10.26599/audt.2025.250089","source":"openalex"},{"id":"oa:W4414072562","type":"article-journal","title":"Exploring the Role of Artificial Intelligence in Evidence Synthesis: Insights From the CORE Information Retrieval Forum 2025","abstract":"ABSTRACT Introduction Information retrieval is essential for evidence synthesis, but developing search strategies can be labor‐intensive and time‐consuming. Automating these processes would be of benefit and interest, though it is unclear if Information Specialists (IS) are willing to adopt artificial intelligence (AI) methodologies or how they currently use them. In January 2025, the NIHR Innovation Observatory and NIHR Methodology Incubator for Applied Health and Care Research co‐sponsored the inaugural CORE Information Retrieval Forum, where attendees discussed AI's role in information retrieval. Methods The CORE Information Retrieval Forum hosted a Knowledge Café. Participation was voluntary, and attendees could choose one of six event‐themed discussion tables including AI. To support each discussion, a QR code linking to a virtual collaboration tool (Padlet; padlet.com ) and a poster in the exhibition space were available throughout the day for attendee contributions. Results The CORE Information Retrieval Forum was attended by 131 IS from nine different types of organizations, with most from the UK and ten countries represented overall. Among the six discussion points available in the Knowledge Café, the AI table was the most popular, receiving the highest number of contributions ( n = 49). Following the Forum, contributions to the AI topic were categorized into four themes: critical perception ( n = 21), current uses ( n = 19), specific tools ( n = 2), and training wants/needs ( n = 7). Conclusions While there are critical perspectives on the integration of AI in the IS space, this is not due to a reluctance to adapt and adopt but from a need for structure, education, training, ethical guidance, and systems to support the responsible use and transparency of AI. There is interest in automating repetitive and time‐consuming tasks, but attendees reported a lack of appropriate supporting tools. More work is required to identify the suitability of currently available tools and their potential to complement the work conducted by IS.","author":[{"family":"Eastaugh","given":"Claire"},{"family":"Still","given":"Madeleine"},{"family":"Beyer","given":"Fiona"},{"family":"Wallace","given":"S"},{"family":"Okeefe","given":"Hannah"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/cesm.70049","URL":"https://doi.org/10.1002/cesm.70049","source":"openalex"},{"id":"oa:W7115948080","type":"article-journal","title":"Development and Validation of Artificial Intelligence Addiction Scale for Researchers: A Methodological Study","abstract":"Background The integration of artificial intelligence (AI) tools into research has brought significant advancements, enhancing efficiency, innovation, and productivity across various academic disciplines. However, alongside these transformative benefits, the growing dependence on AI tools has raised concerns regarding overreliance and the potential for addictive behaviors among researchers. Despite the widespread adoption of AI among the researchers, there remains a notable gap in the availability of validated instruments specifically designed to assess AI addiction within this context. Objective To develop a scale to measure AI addiction among researchers and evaluate its psychometric properties. Design A methodological design was employed, consisting of two phases: scale development and psychometric evaluation. Methods Items were generated through a comprehensive literature review and semistructured interviews to capture AI addiction attributes. The scale’s psychometric properties—including content validity, face validity, construct validity, and internal consistency reliability—were assessed. Data from a convenience sample of 718 nursing researchers were randomly divided into two independent subsamples for exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Reliability was evaluated using Cronbach’s alpha, McDonald’s omega, split‐half reliability, and corrected item–total correlations. Results The finalized scale comprises 22 items across five dimensions: compulsive behavior, overdependency, functional impairment, withdrawal, and tolerance. EFA identified a five‐factor structure explaining 73.66% of the variance. CFA validated the structure with robust fit indices for first order ( χ 2 /DF = 2.289, CFI = 0.962, and RMSEA = 0.06) and second order ( χ 2 /DF = 2.243, CFI = 0.962, and RMSEA = 0.059) models. The scale demonstrated excellent internal consistency and reliability, with Cronbach’s alpha ( α = 0.924), McDonald’s omega ( ω = 0.870), and a Spearman–Brown split‐half coefficient of 0.814. Moderate interfactor correlations ( r = 0.41–0.62) confirmed its multidimensionality. Conclusion The researchers’ AI addiction scale is a valid and reliable tool for assessing AI addiction among researchers, providing a robust framework to evaluate compulsive behavior, dependency, functional disruption, withdrawal symptoms, and tolerance.","author":[{"family":"El-Sayed","given":"Ahmed"},{"family":"Alsenany","given":"Samira"},{"family":"Asal","given":"Maha"},{"family":"Alasqah","given":"Ibrahim"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1155/jonm/8458533","URL":"https://doi.org/10.1155/jonm/8458533","source":"openalex"},{"id":"oa:W4413024962","type":"article-journal","title":"Rethinking Artificial Intelligence (AI) in Qualitative Research","abstract":"Artificial intelligence (AI) is increasingly being integrated into qualitative research. While AI provides tools that claim to promise greater efficiency and productivity, it also raises vital concerns about methodological integrity, ethical conduct, and the maintenance of contextual depth. This paper explores the implications of AI for qualitative research. It seeks to address the ethical, methodological, and epistemological concerns associated with AI integration. It also aims to promote critical engagement with AI while upholding the foundational values of context-rich qualitative research. The article initially addresses the growing publicity that AI has received in studies and public discourse, focusing on inflated hopes and misconceptions. The central debate concerns the impacts of AI on qualitative research, such as the way it could improve analysis and transcription efficiency. However, the use of AI is not without issues related to ethics, bias, loss of interpretive depth, and over-reliance on automation. The article argues in favour of responsible, critical, and ethically aware use of AI in qualitative research. AI can be a valuable tool to support, but not replace, qualitative researchers. Its use must be governed by reflexivity, ethical sensitivity, and contextual knowledge to uphold the foundational values of qualitative inquiry.","author":[{"family":"Jamil","given":"Nurul"},{"family":"Rashid","given":"Nor’ain"},{"family":"Tan","given":"Woei"}],"issued":{"date-parts":[[2025]]},"DOI":"10.31436/ijcs.v8i2.468","URL":"https://doi.org/10.31436/ijcs.v8i2.468","source":"openalex"},{"id":"oa:W4412694387","type":"article-journal","title":"Artificial intelligence-driven pathomics in hepatocellular carcinoma: current developments, challenges and perspectives","abstract":"Hepatocellular carcinoma (HCC) is a highly malignant tumor with elevated incidence and mortality rates globally. Its complex etiology and pronounced heterogeneity present significant challenges in diagnosis and treatment. Recent advancements in artificial intelligence (AI) have demonstrated transformative potential to usher a new wave of precision oncology. Pathomics, an AI-based digital pathology technique, facilitates the extraction of extensive datasets from whole-slide histopathological images, enabling quantitative analyses to improve diagnosis, treatment, and prognostic prediction for HCC. Furthermore, emerging pathological foundation models are revolutionizing traditional paradigms and providing a robust framework for the development of specialized pathomics models tailored to specific clinical tasks in HCC. Despite its promise, pathomics research in HCC remains in its infancy, with clinical implementation hindered by challenges such as data heterogeneity, model interpretability, ethical concerns, regulatory issues, and the absence of standardized industry protocols. Future initiatives should prioritize the conduction of prospective multi-center studies, the integration of multi-modal data, the enhancement of regulatory frameworks, and the establishment of industry-wide standardized guidelines and compliant platform infrastructures to accelerate the clinical adoption of pathomics for personalized HCC treatment.","author":[{"family":"Ding","given":"Wei"},{"family":"Zhang","given":"Jin"},{"family":"Jin","given":"Zhi‐cheng"},{"family":"Hua","given":"Hongjin"},{"family":"Zu","given":"Qing‐quan"},{"family":"Yang","given":"Shudong"},{"family":"Wang","given":"W"},{"family":"Liu","given":"Sheng"},{"family":"Zhou","given":"Hai"},{"family":"Shi","given":"Hai"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s12672-025-03254-z","URL":"https://doi.org/10.1007/s12672-025-03254-z","source":"openalex"},{"id":"oa:W4408504098","type":"article-journal","title":"Exploring University Staff’s Perceptions of Using Generative Artificial Intelligence at University","abstract":"This study aimed to understand university staff’s perspectives and approaches regarding students’ use of generative artificial intelligence (GenAI) in an academic setting. Currently, there is a lack of social media analyses exploring this area. For the present study, a qualitative content analysis was conducted on posts about ChatGPT shared via X (formerly Twitter). This enabled a sample of n = 194 perspectives to be captured. Three main themes were generated: (1) perceptions of GenAI’s impact on higher education and skepticism towards its management; (2) GenAI in assessment: prevention and detection approaches; and (3) future-focused approaches to GenAI-enhanced learning and assessment. Some university staff see GenAI as a threat to their profession and have stressed the need for university guidance. Staff discussed both the positive and negative impacts of GenAI on student learning. Some staff want to prevent the use of GenAI in assessments, whilst others embrace the tool. These findings can inform university guidance on the use of GenAI in the future.","author":[{"family":"Whitbread","given":"M"},{"family":"Hayes","given":"Charles"},{"family":"Prabhakar","given":"Sundaresan"},{"family":"Upsher","given":"Rebecca"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/educsci15030367","URL":"https://doi.org/10.3390/educsci15030367","source":"openalex"},{"id":"oa:W7118072316","type":"article-journal","title":"Artificial intelligence for breast cancer management","abstract":"Artificial intelligence is transforming breast cancer management through various machine learning applications. Artificial intelligence supports precision medicine by enhancing detection, diagnosis, prognosis, and treatment response prediction. It achieves this by analysing data from medical imaging, histopathology, genomics and multi-omics sources to improve patient recovery. This review summarises AI-driven advancements across the entire continuum of breast cancer management, spanning detection, diagnosis, prognosis, treatment and recovery. It evaluates their efficacy and limitations, explores their impact on healthcare costs and clinical practice, and addresses key challenges including generalisability, reproducibility and regulatory barriers. Evidence from recent studies highlights AI’s role in improving breast cancer detection, molecular subtyping and prognostic accuracy. It also facilitates more patient-tailored therapeutic strategies and supports quality of life interventions. Nonetheless, the translation of these benefits into clinical practice requires rigorous validation, transparent model development, and equitable implementation. Chua et al., discuss how artificial intelligence is transforming breast cancer care by improving detection, diagnosis, prognosis, treatment planning, and patient recovery through advanced machine learning and deep learning applications. They emphasise that widespread adoption faces challenges such as data diversity, reproducibility, regulatory hurdles, infrastructure limitations, and ethical concerns around transparency and bias.","author":[{"family":"Chua","given":"Bryan"},{"family":"Thng","given":"Dexter"},{"family":"Toh","given":"Tan"},{"family":"Ho","given":"Dean"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1038/s43856-025-01342-3","URL":"https://doi.org/10.1038/s43856-025-01342-3","source":"openalex"},{"id":"oa:W4412522062","type":"article-journal","title":"Artificial intelligence-based models for quantification of intra-pancreatic fat deposition and their clinical relevance: a systematic review of imaging studies","abstract":"High intra-pancreatic fat deposition (IPFD) plays an important role in diseases of the pancreas. The intricate anatomy of the pancreas and the surrounding structures has historically made IPFD quantification a challenging measurement to make accurately on radiological images. To take on the challenge, automated IPFD quantification methods using artificial intelligence (AI) have recently been deployed. The aim was to benchmark the current knowledge on the use of AI-based models to measure IPFD automatedly. The search was conducted in the MEDLINE, Embase, Scopus, and IEEE Xplore databases. Studies were eligible if they used AI for both segmentation of the pancreas and quantification of IPFD. The ground truth was manual segmentation by radiologists. When possible, data were pooled statistically using a random-effects model. A total of 12 studies (10 cross-sectional and 2 longitudinal) encompassing more than 50 thousand people were included. Eight of the 12 studies used MRI, whereas four studies employed CT. U-Net model and nnU-Net model were the most frequently used AI-based models. The pooled Dice similarity coefficient of AI-based models in quantifying IPFD was 82.3% (95% confidence interval, 73.5 to 91.1%). The clinical application of AI-based models showed the relevance of high IPFD to acute pancreatitis, pancreatic cancer, and type 2 diabetes mellitus. Current AI-based models for IPFD quantification are suboptimal, as the dissimilarity between AI-based and manual quantification of IPFD is not negligible. Future advancements in fully automated measurements of IPFD will accelerate the accumulation of robust, large-scale evidence on the role of high IPFD in pancreatic diseases. KEY POINTS: Question What is the current evidence on the performance and clinical applicability of artificial intelligence-based models for automated quantification of intra-pancreatic fat deposition? Findings The nnU-Net model achieved the highest Dice similarity coefficient among MRI-based studies, whereas the nnTransfer model demonstrated the highest Dice similarity coefficient in CT-based studies. Clinical relevance Standardisation of reporting on artificial intelligence-based models for the quantification of intra-pancreatic fat deposition will be essential to enhancing the clinical applicability and reliability of artificial intelligence in imaging patients with diseases of the pancreas.","author":[{"family":"Joshi","given":"Tej"},{"family":"Virostko","given":"John"},{"family":"Petrov","given":"Maxim"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00330-025-11808-6","URL":"https://doi.org/10.1007/s00330-025-11808-6","source":"openalex"},{"id":"oa:W4417326160","type":"article-journal","title":"Artificial Intelligence in Anaesthesiology: Current Applications, Challenges, and Future Directions","abstract":"Artificial intelligence (AI) is rapidly transforming anaesthesiology through advances in machine learning, deep learning, and large language models. AI-driven tools now contribute to nearly every phase of perioperative care, including preoperative risk stratification, intraoperative monitoring, imaging interpretation, airway assessment, regional anaesthesia, and critical care. Applications such as automated American Society of Anesthesiologists classification, prediction of postoperative complications and intensive care unit needs, electroencephalography-based depth-of-anaesthesia estimation, and proactive haemodynamic management are reshaping clinical decision-making. AI-augmented echocardiography enhances chamber recognition and functional measurements, whereas computer vision systems support airway evaluation and ultrasound-guided regional anaesthesia by providing real-time anatomical identification and facilitating training. In critical care, AI models facilitate the early detection of sepsis, organ dysfunction, and haemodynamic instability, while improving workflow efficiency and resource allocation. AI is increasingly used in academic writing, data processing, and medical education, offering opportunities for personalised learning and simulation but raising concerns about accuracy and hallucinations. In this review, we aimed to summarise the current applications of AI in anaesthesiology, highlight the methodological, ethical, and practical challenges that limit its integration, and discuss future directions for its safe and effective adoption in perioperative care.","author":[{"family":"Dost","given":"Burhan"},{"family":"Turan","given":"Engin"},{"family":"Aydın","given":"Muhammed"},{"family":"Ahişkalıoğlu","given":"Ali"},{"family":"Narayanan","given":"Madan"},{"family":"Yılmaz","given":"Resül"},{"family":"Cassai","given":"Alessandro"}],"issued":{"date-parts":[[2025]]},"DOI":"10.4274/tjar.2025.252320","URL":"https://doi.org/10.4274/tjar.2025.252320","source":"openalex"},{"id":"oa:W4409187867","type":"article-journal","title":"The Role of Artificial Intelligence and Machine Learning in Polymer Characterization: Emerging Trends and Perspectives","abstract":"The application of artificial intelligence (AI) and machine learning (ML) is rapidly expanding and has begun to make a significant impact on polymer development and characterization. This perspective article explores the current state of AI in this field and highlights areas where its potential remains underutilized. While the optimization of polymer synthesis to achieve desired properties and the classification of polymer types are well-established, opportunities for AI integration in detailed characterization, analytical method development, and data processing remain largely untapped. Greater automation of the analytical laboratory, whether through dedicated algorithms or AI-driven solutions, will enable analytical chemists to focus more on addressing research questions and interpreting results, rather than on method development and routine measurements.","author":[{"family":"Hurk","given":"Rick"},{"family":"Pirok","given":"Bob"},{"family":"Bos","given":"Tijmen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s10337-025-04406-7","URL":"https://doi.org/10.1007/s10337-025-04406-7","source":"openalex"},{"id":"oa:W4407232216","type":"article-journal","title":"Applications of Artificial Intelligence for Metastatic Gastrointestinal Cancer: A Systematic Literature Review","abstract":"BACKGROUND/OBJECTIVES: This systematic literature review examines the application of Artificial Intelligence (AI) in the diagnosis, treatment, and follow-up of metastatic gastrointestinal cancers. METHODS: The databases PubMed, Scopus, Embase (Ovid), and Google Scholar were searched for published articles in English from January 2010 to January 2022, focusing on AI models in metastatic gastrointestinal cancers. RESULTS: forty-six studies were included in the final set of reviewed papers. The critical appraisal and data extraction followed the checklist for systematic reviews of prediction modeling studies. The risk of bias in the included papers was assessed using the prediction risk of bias assessment tool. CONCLUSIONS: AI techniques, including machine learning and deep learning models, have shown promise in improving diagnostic accuracy, predicting treatment outcomes, and identifying prognostic biomarkers. Despite these advancements, challenges persist, such as reliance on retrospective data, variability in imaging protocols, small sample sizes, and data preprocessing and model interpretability issues. These challenges limit the generalizability, clinical application, and integration of AI models.","author":[{"family":"Naemi","given":"Amin"},{"family":"Tashk","given":"Ashkan"},{"family":"Azar","given":"Amir"},{"family":"Samimi","given":"Tahereh"},{"family":"Tavassoli","given":"Ghanbar"},{"family":"Mohasefi","given":"Anita"},{"family":"Khanshan","given":"Elaheh"},{"family":"Najafabad","given":"Mehrdad"},{"family":"Tarighi","given":"Vafa"},{"family":"Wiil","given":"Uffe"},{"family":"Bagherzadeh","given":"Jamshid"},{"family":"Pirnejad","given":"Habibollah"},{"family":"Niazkhani","given":"Zahra"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/cancers17030558","URL":"https://doi.org/10.3390/cancers17030558","source":"openalex"},{"id":"oa:W4406408916","type":"article-journal","title":"The role of large language models in the peer-review process: opportunities and challenges for medical journal reviewers and editors","abstract":"The peer review process ensures the integrity of scientific research. This is particularly important in the medical field, where research findings directly impact patient care. However, the rapid growth of publications has strained reviewers, causing delays and potential declines in quality. Generative artificial intelligence, especially large language models (LLMs) such as ChatGPT, may assist researchers with efficient, high-quality reviews. This review explores the integration of LLMs into peer review, highlighting their strengths in linguistic tasks and challenges in assessing scientific validity, particularly in clinical medicine. Key points for integration include initial screening, reviewer matching, feedback support, and language review. However, implementing LLMs for these purposes will necessitate addressing biases, privacy concerns, and data confidentiality. We recommend using LLMs as complementary tools under clear guidelines to support, not replace, human expertise in maintaining rigorous peer review standards.","author":[{"family":"Lee","given":"Jisoo"},{"family":"Lee","given":"Jieun"},{"family":"Yoo","given":"Jeong‐ju"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3352/jeehp.2025.22.4","URL":"https://doi.org/10.3352/jeehp.2025.22.4","source":"openalex"},{"id":"oa:W4413998662","type":"article-journal","title":"eHealth literacy and attitudes towards use of artificial intelligence among university students in the United Arab Emirates, a cross-sectional study","abstract":"Introduction: With the rapid digitalization of healthcare information and the increasing dependability on online health resources, it has become crucial to understand digital health literacy and the use of emerging AI technologies like ChatGPT among stakeholders. This is of particular importance in the United Arab Emirates which has the highest internet penetration rates. Method: This study aimed to assess eHealth literacy and the factors influencing it among university students in the United Arab Emirates. Their attitudes towards ChatGPT use were also explored. Data from participants, studying in the public universities of UAE, was collected between April-July 2024 using eHEALS and TAME Chat GPT instruments. Results: Results indicated a mean eHealth literacy score of 29.3 out of 40, with higher scores among females and those in health-related disciplines. It was also found that students with higher eHealth literacy perceived ChatGPT as more useful in healthcare, despite their concerns about its risks and potential to replace healthcare professionals. Discussion: The findings from the study underscore the need of development of tailored digital health curricula, to enhance eHealth literacy particularly in subgroups showing lower literacy scores. Moreover, it is also imperative to develop guidelines for responsible and ethical AI use in health information seeking.","author":[{"family":"Alam","given":"Zufishan"},{"family":"Abdullahi","given":"Aminu"},{"family":"Alnuaimi","given":"Shamma"},{"family":"Shaka","given":"Hanouf"},{"family":"Alderei","given":"Saif"},{"family":"Alhemeiri","given":"Ahmed"},{"family":"Khorzom","given":"Hayma"},{"family":"Almaskari","given":"Hamad"},{"family":"Almaamari","given":"Khalid"},{"family":"Seiari","given":"Khalifa"},{"family":"Saadi","given":"Mohammed"},{"family":"Shamsi","given":"Nasser"},{"family":"Zaabi","given":"Omar"},{"family":"Altamimi","given":"Saoud"},{"family":"Rahma","given":"Azhar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fdgth.2025.1574263","URL":"https://doi.org/10.3389/fdgth.2025.1574263","source":"openalex"},{"id":"oa:W4407176841","type":"article-journal","title":"Artificial Intelligence in Logistics and Distribution: The function of AI in dynamic route planning for transportation, including self-driving trucks and drone delivery systems","abstract":"Artificial Intelligence (AI) is revolutionizing logistics and distribution by enhancing efficiency, reducing costs, and improving the overall delivery experience. One of the key applications of AI in this sector is dynamic route optimization, where machine learning algorithms analyze real-time data such as traffic patterns, weather conditions, and road closures to continuously adjust delivery routes. This reduces fuel consumption, optimizes delivery times, and mitigates operational disruptions. In parallel, autonomous delivery systems, including drones and self-driving trucks, are transforming last-mile and long-haul transportation by minimizing human intervention and increasing speed and reliability. AI-powered autonomous delivery systems leverage advanced technologies such as computer vision, sensors, and machine learning to navigate and make real-time decisions. Drones, for example, are already being utilized for time-sensitive deliveries, especially in remote or underserved areas, while self-driving trucks promise to revolutionize long-haul freight transportation with the potential for 24/7 operation, cost reduction, and increased safety. Despite the clear benefits, challenges such as regulatory frameworks, safety concerns, and public acceptance remain as significant barriers to large-scale adoption. This review explores the role of AI in dynamic route optimization, drones, and self-driving trucks within logistics, providing insights into how these technologies are currently being implemented and the challenges that lie ahead. It further discusses the potential for AI to reshape the future of logistics and transportation by driving innovation in automation, reducing operational inefficiencies, and contributing to the development of smarter, more sustainable supply chains. The future of AI in logistics promises a more integrated, cost-effective, and agile system that can adapt to global challenges and evolving consumer demands.","author":[{"family":"Adeoye","given":"Yetunde"},{"family":"Onotole","given":"Erumusele"},{"family":"Ogunyankinnu","given":"Tunde"},{"family":"Aipoh","given":"Godwin"},{"family":"Osunkanmibi","given":"Akintunde"},{"family":"Egbemhenghe","given":"Joseph"}],"issued":{"date-parts":[[2025]]},"DOI":"10.30574/wjarr.2025.25.2.0214","URL":"https://doi.org/10.30574/wjarr.2025.25.2.0214","source":"openalex"},{"id":"oa:W4412517485","type":"article-journal","title":"The Role of Artificial Intelligence in the Diagnosis and Management of Diabetic Retinopathy","abstract":"Background/Objectives: Diabetic retinopathy (DR) is a progressive microvascular complication of diabetes mellitus and a leading cause of vision impairment worldwide. Early detection and timely management are critical in preventing vision loss, yet current screening programs face challenges, including limited specialist availability and variability in diagnoses, particularly in underserved areas. This literature review explores the evolving role of artificial intelligence (AI) in enhancing the diagnosis, screening, and management of diabetic retinopathy. It examines AI’s potential to improve diagnostic accuracy, accessibility, and patient outcomes through advanced machine-learning and deep-learning algorithms. Methods: We conducted a non-systematic review of the published literature to explore advancements in the diagnostics of diabetic retinopathy. Relevant articles were identified by searching the PubMed and Google Scholar databases. Studies focusing on the application of artificial intelligence in screening, diagnosis, and improving healthcare accessibility for diabetic retinopathy were included. Key information was extracted and synthesized to provide an overview of recent progress and clinical implications. Conclusions: Artificial intelligence holds transformative potential in diabetic retinopathy care by enabling earlier detection, improving screening coverage, and supporting individualized disease management. Continued research and ethical deployment will be essential to maximize AI’s benefits and address challenges in real-world applications, ultimately improving global vision health outcomes.","author":[{"family":"Ansari","given":"A"},{"family":"Ansari","given":"Nabiha"},{"family":"Khalid","given":"Usman"},{"family":"Markov","given":"Daniel"},{"family":"Bechev","given":"Kristian"},{"family":"Aleksiev","given":"Vladimir"},{"family":"Markov","given":"Galabin"},{"family":"Poryazova","given":"E"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jcm14145150","URL":"https://doi.org/10.3390/jcm14145150","source":"openalex"},{"id":"oa:W4407823694","type":"article-journal","title":"From data to artificial intelligence: evaluating the readiness of gastrointestinal endoscopy datasets","abstract":"The incorporation of artificial intelligence (AI) into gastrointestinal (GI) endoscopy represents a promising advancement in gastroenterology. With over 40 published randomized controlled trials and numerous ongoing clinical trials, gastroenterology leads other medical disciplines in AI research. Computer-aided detection algorithms for identifying colorectal polyps have achieved regulatory approval and are in routine clinical use, while other AI applications for GI endoscopy are in advanced development stages. Near-term opportunities include the potential for computer-aided diagnosis to replace conventional histopathology for diagnosing small colon polyps and increased AI automation in capsule endoscopy. Despite significant development in research settings, the generalizability and robustness of AI models in real clinical practice remain inconsistent. The GI field lags behind other medical disciplines in the breadth of novel AI algorithms, with only 13 out of 882 Food and Drug Administration (FDA)-approved AI models focussed on GI endoscopy as of June 2024. Additionally, existing GI endoscopy image databases are disproportionately focussed on colon polyps, lacking representation of the diversity of other endoscopic findings. High-quality datasets, encompassing a wide range of patient demographics, endoscopic equipment types, and disease states, are crucial for developing effective AI models for GI endoscopy. This article reviews the current state of GI endoscopy datasets, barriers to progress, including dataset size, data diversity, annotation quality, and ethical issues in data collection and usage, and future needs for advancing AI in GI endoscopy.","author":[{"family":"Elamin","given":"Sami"},{"family":"Johri","given":"Shreya"},{"family":"Rajpurkar","given":"Pranav"},{"family":"Geisler","given":"Enrik"},{"family":"Berzin","given":"Tyler"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/jcag/gwae041","URL":"https://doi.org/10.1093/jcag/gwae041","source":"openalex"},{"id":"oa:W4414038162","type":"article-journal","title":"Frankenstein, thematic analysis and generative artificial intelligence: Quality appraisal methods and considerations for qualitative research","abstract":"OBJECTIVE: To determine accuracy and efficiency of using generative artificial intelligence (GenAI) to undertake thematic analysis. INTRODUCTION: With the increasing use of GenAI in data analysis, testing the reliability and suitability of using GenAI to conduct qualitative data analysis is needed. We propose a method for researchers to assess reliability of GenAI outputs using deidentified qualitative datasets. METHODS: We searched three databases (United Kingdom Data Service, Figshare, and Google Scholar) and five journals (PlosOne, Social Science and Medicine, Qualitative Inquiry, Qualitative Research, Sociology Health Review) to identify studies on health-related topics, published prior to whereby: humans undertook thematic analysis and published both their analysis in a peer-reviewed journal and the associated dataset. We prompted a closed system GenAI (Microsoft Copilot) to undertake thematic analysis of these datasets and analysed the GenAI outputs in comparison with human outputs. Measures include time (GenAI only), accuracy, overlap with human analysis, and reliability of selected data and quotes. RESULTS: Five studies were identified that met our inclusion criteria. The themes identified by human researchers and Copilot showed minimal overlap, with human researchers often using discursive thematic analyses (40%) and Copilot focusing on thematic analysis (100%). Copilot's outputs often included fabricated quotes (58% SD = 45%) and none of the Copilot outputs provided participant spread by theme. Additionally, Copilot's outputs primarily drew themes and quotes from the first 2-3 pages of textual data, rather than from the entire dataset. Human researchers provided broader representation and accurate quotes (79% quotes were correct, SD = 27%). CONCLUSIONS: Based on these results, we cannot recommend the current version of Copilot for undertaking thematic analyses. This study raises concerns about the validity of both human-generated and GenAI-generated qualitative data analysis and reporting.","author":[{"family":"Jowsey","given":"Tanisha"},{"family":"Stapleton","given":"Peta"},{"family":"Campbell","given":"Shawna"},{"family":"Davidson","given":"Alexandra"},{"family":"Mcgillivray","given":"Cher"},{"family":"Maugeri","given":"Isabella"},{"family":"Lee","given":"Megan"},{"family":"Keogh","given":"Justin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1371/journal.pone.0330217","URL":"https://doi.org/10.1371/journal.pone.0330217","source":"openalex"},{"id":"oa:W4412004643","type":"article-journal","title":"Matrix of technical solutions based on artificial intelligence in the professional training of future lawyers","abstract":"Importance. The current stage of technological development of society is characterized by the intensive integration of artificial intelligence (AI) technologies into professional spheres. AI-based technical solutions make it possible to automate some routine processes and free up time for humans to solve other more important and complex issues. Gradually, the interaction of specialistswith AI tools to solve professional problems is becoming a daily practice. In this regard, the training of qualified personnel at the university for the realities of today is impossible without integrating professionally oriented AI tools into the student learning process. Law is one of the activity fields in which modern AI technologies are able to take on many professional tasks. At the same time, the systematic integration of AI-based technical solutions into the university’s law student training process is impossible without a comprehensive study of the entire range of AI tools and their professionally oriented potential. The purpose of the work is to develop a matrix of AI-based technical solutions used in the professional training of future lawyers. Materials and Methods. The study is conducted on the expert assessment method basis. This allowed the authors to: a) identify a list of professional tasks solved by lawyers in the field of professional activity; b) based on the identified tasks, develop a matrix of AI-based technical solutions used in the professional training of future lawyers. The materials are scientific papers on pedagogy, methods of teaching foreign languages and specialized disciplines, published in scientific journals indexed in the Ministry of National Security (Scopus and Web of Science), as well as those included in the list of the Higher Attestation Commission of the Russian Federation (K1, K2), the Federal State Educational Standard for Higher Education in the field of Law. The AI tools widely used among current lawyers, which they use in their professional activities to solve professional problems, are used as practical materials. Results and Discussion. A matrix of AI-based technical solutions used in the professional training of future lawyers has been developed. The matrix is presented according to twelve professional tasks that lawyers solve in the course of their professional activities. The main and most accessible AI-based technical solutions for teachers of specialized disciplines that can help lawyers solve professional problems are the following: Legal AI tools, Legal Document Generator and DocZilla AI are used to draw up contracts (lease, sale, employment agreements, etc.), DocZilla AI and Genie AI – for the analysis and comparison of document editions, Mistral AI and LexisNexis – for checking documents for errors and contradictions, ROSS Intelligence and WestLaw – to search for relevant court decisions and analyze use cases, TrademarkVision and PatentPal – to search for similar trademarks, Perplexity AI – to analyze license agreements, Legalese Decoder, ChatGPT, YandexGPT, GigaChat and DeepSeek – to simplify legal terms for clients (colleagues, students), Canva and MidJourney – to visualize processes (for example, judicial meetings), LegalAI and Jasper AI – for legal advice, Perplexity AI, ChatGPT, YandexGPT, GigaChat and DeepSeek – for mathematical calculations (taxes, insurance payments, etc.), MidJourney – to create images of suspects, Legalese Decoder and Mistral AI are used for conducting examinations (handwriting, ballistic, etc.). Conclusion. The research novelty is the development of a matrix of AI-based technical solutions used in the professional training of future lawyers. The perspective of the conducted research lies in the development of step-by-step methods of teaching aspects of specialized disciplines based on the students’practice with specific technical solutions based on AI.","author":[{"family":"Sysoyev","given":"Pavel"},{"family":"Gavrilov","given":"MV"},{"family":"Bulochnikov","given":"Stanislav"}],"issued":{"date-parts":[[2025]]},"DOI":"10.20310/1810-0201-2025-30-2-336-351","URL":"https://doi.org/10.20310/1810-0201-2025-30-2-336-351","source":"openalex"},{"id":"oa:W7125689582","type":"article-journal","title":"Advancing the modernization of traditional Chinese medicine through artificial intelligence and multimodal data integration","abstract":"Traditional Chinese Medicine (TCM) is a valuable medical treasure trove that not only demonstrated unique advantages in treating complex and refractory diseases but also left behind a rich legacy of ancient texts and valuable evidence-based medical data based on its human experience for future generations. Nevertheless, the extensive data within TCM has been plagued by challenges, including inadequate data standardization, inconsistent data quality, limited data structuring, and obstacles in interdisciplinary integration. Recent advancements in artificial intelligence (AI) techniques have markedly improved the efficiency and effectiveness with which multimodal data in TCM, including machine learning (ML), deep learning (DL), knowledge graphs (KG), and natural language processing (NLP), particularly large language models (LLMs). These advancements have facilitated more precise data analysis, enhanced clinical decision-making, and improved research outcomes in TCM, such as target discovery, virtual screening of natural products (NPs), symptom differentiation and auxiliary prescription. This article presents a comprehensive review of the progress in applying AI across four dimensions: multiscale data in TCM, TCM research and development, TCM diagnosis and treatment, and LLMs. In summary, the application of AI technology in the modernization of TCM is expected to motivate researchers to achieve a deeper understanding of state-of-the-art applications in data-driven TCM complex systems, fundamental scientific research, and precision medicine, thereby bringing more opportunities and innovations for the modernization of TCM.","author":[{"family":"Guo","given":"Pengfei"},{"family":"Jiang","given":"Mengmeng"},{"family":"Hu","given":"Shaowu"},{"family":"Jiang","given":"Qianqian"},{"family":"Li","given":"Limin"},{"family":"Wu","given":"Junhong"},{"family":"Ma","given":"Yucui"},{"family":"Wu","given":"Zhengzhi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1186/s13020-025-01194-y","URL":"https://doi.org/10.1186/s13020-025-01194-y","source":"openalex"},{"id":"oa:W4407289123","type":"article-journal","title":"Integration of Artificial Intelligence and Robotics into the industrial sector","abstract":"The 4th industrial revolution is driven by the implementation of automated robots and artificial intelligence (AI) to enhance efficiency, accuracy, and safety. This integration encompasses several vital domains like optimizing the supply chain, interaction between human and robots on the shop floor, predictive maintenance, automation of repetitive tasks, customisation, behaviour design, and safety management, data analysis, etc. AI-enabled robots perform repetitive tasks at very high precision, reducing the chances of human error and allowing workers to focus on more complex tasks. Automated upkeep utilizes AI to determine the time machinery will likely fail, which minimizes downtime and maintenance costs. Automated testing and AI-driven vision systems support quality control by ensuring a balanced quality of the product. AI improves supply chain processes, optimizing logistics and inventory management. Collaboration between humans and collaborative robot’s results in safer and more productive environments with people working alongside each other. Artificial Intelligence plays an important role in making smarter decisions, analysing data more effectively, and providing valuable information that can be used to improve operations. Manufacturing customization and flexibility are reliant on adaptive systems and the ability to manufacture personalized products by means of productivity. Safe and Risk Management is consolidated because robots work in dangerous scenarios and artificial intelligence models assess potential dangers. Despite challenges including labour displacement, cybersecurity, ethics, and data integration stemming from this technology, these are all potentially available on your terms. This article reviews the broader impacts that robots and artificial Intelligence have had on the industrial sector, placing emphasis on the revolution it could lead towards as well as the key elements to consider before implementing it.","author":[{"family":"Abdullayev","given":"Vugar"},{"family":"Faizal","given":"Ajesh"},{"family":"Seyidova","given":"Irada"},{"family":"Mikayilov","given":"Seymur"},{"family":"Mammadova","given":"Rubaba"},{"family":"Pirverdiyeva","given":"Lala"},{"family":"Guliyev","given":"Etibar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.56294/dm2025209","URL":"https://doi.org/10.56294/dm2025209","source":"openalex"},{"id":"oa:W4413805190","type":"article-journal","title":"Artificial Intelligence in Climate Change Mitigation and Adaptation: A Review of Emerging Technologies and Real-World Applications","abstract":"Artificial Intelligence (AI) is increasingly recognized as a transformative tool in addressing the dual imperatives of climate change mitigation and adaptation. This review provides a comprehensive synthesis of the current state of AI applications that contribute to reducing greenhouse gas emissions and enhancing resilience to climate-related hazards. It systematically examines advances in machine learning, optimization, and data-driven decision support across key domains including renewable energy forecasting, energy system optimization, land-use planning, disaster risk management, precision agriculture, and water resource allocation. The paper also analyzes the enabling infrastructure required for scalable and ethical deployment, such as data interoperability, model interpretability, and integration with physical system models. The findings from existing literature indicate that AI has significantly improved predictive capacity, operational efficiency, and adaptive planning in climate-related contexts. However, persistent challenges ranging from data scarcity and geographic bias to the carbon footprint of AI systems and governance limitations continue to constrain equitable implementation. The review concludes by identifying critical research gaps and proposing a strategic roadmap focused on interdisciplinary collaboration, equitable data frameworks, and policy alignment with global climate objectives. By critically appraising both the potential and limitations of AI, this review contributes to the research on how intelligent systems can be leveraged to support sustainable, inclusive, and scientifically grounded climate action.","author":[{"family":"Eze","given":"Favour"},{"family":"Sanusi","given":"Adepeju"},{"family":"Iheoma","given":"Lsrael"},{"family":"Ekechi","given":"Chijioke"},{"family":"Olatunbosun","given":"Micheal"},{"family":"Ukasoanya","given":"Favour"},{"family":"Eleshin","given":"Muhdawwal"}],"issued":{"date-parts":[[2025]]},"DOI":"10.30574/gjeta.2025.24.2.0247","URL":"https://doi.org/10.30574/gjeta.2025.24.2.0247","source":"openalex"},{"id":"oa:W4412941949","type":"article-journal","title":"Bridging technology and medicine: artificial intelligence in targeted anticancer drug delivery","abstract":"The integration of artificial intelligence (AI) in targeted anticancer drug delivery represents a significant advancement in oncology, offering innovative solutions to enhance the precision and effectiveness of cancer treatments. This review explores the various AI methodologies that are transforming the landscape of targeted drug delivery systems. By leveraging machine learning algorithms, researchers can analyze extensive datasets, including genomic, proteomic, and clinical data, to identify patient-specific factors that influence therapeutic responses. Supervised learning techniques, such as support vector machines and random forests, enable the classification of cancer types and the prediction of treatment outcomes based on historical data. Deep learning approaches, particularly convolutional neural networks, facilitate improved tumor detection and characterization through advanced imaging analysis. Moreover, reinforcement learning optimizes treatment protocols by dynamically adjusting drug dosages and administration schedules based on real-time patient responses. The convergence of AI and targeted anticancer drug delivery holds the promise of advancing cancer therapy by providing tailored treatment strategies that enhance efficacy while minimizing side effects. By improving the understanding of tumor biology and patient variability, AI-driven methods can facilitate the transition from traditional treatment paradigms to more personalized and effective cancer care. This review discusses the challenges and limitations of implementing AI in targeted anticancer drug delivery, including data quality, interpretability of AI models, and the need for robust validation in clinical settings.","author":[{"family":"Khorsandi","given":"Danial"},{"family":"Farahani","given":"Amin"},{"family":"Zarepour","given":"Atefeh"},{"family":"Khosravi","given":"Arezoo"},{"family":"Iravani","given":"Siavash"},{"family":"Zarrabi","given":"Ali"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1039/d5ra03747f","URL":"https://doi.org/10.1039/d5ra03747f","source":"openalex"},{"id":"oa:W4412502018","type":"article-journal","title":"Artificial Intelligence and English as a Foreign Language (EFL) Teachers’ Competencies: A Systematic Review","abstract":"The integration of Artificial Intelligence (AI) into English as a Foreign Language (EFL) education has transformed teaching practices, necessitating a re-evaluation of teacher competencies in the digital age. This systematic review examines the intersection of AI technologies and EFL teachers' competencies, exploring how AI tools influence pedagogical skills, technological proficiency, and professional development. Through an analysis of recent literature, the study identifies key competencies required for EFL teachers to effectively leverage AI, including adaptive teaching strategies, data literacy, and ethical considerations in AI usage. The review also highlights challenges such as resistance to technological adoption, the digital divide, and the need for continuous upskilling. Findings suggest that while AI offers significant opportunities for personalized learning and efficiency, EFL teachers must develop a balanced skill set that integrates traditional teaching expertise with emerging technological demands. The results of the research showed that 1) EFL teachers’ competencies in artificial intelligence field consist of 10 competencies, namely: 1) AI-Assisted Lesson Planning, 2) AI-Powered Language Practice & Feedback, 3) Speech Recognition & Pronunciation Tools, 4) AI for Differentiated Instruction, 5) Automated Assessment & Grading, 6) Data-Driven Student Insights, 7) Ethical & Critical Use of AI, 8) AI for Content Creation & Gamification, 9) AI-Powered Translation & Comprehension Support, and 10) AI and Virtual/Augmented Reality in EFL. The paper concludes with recommendations for teacher training programs and policy frameworks to support the evolving role of EFL educators in an AI-driven educational landscape.","author":[{"family":"Laoha","given":"Rukthin"},{"family":"Chomthong","given":"Wichittra"},{"family":"Pongpanich","given":"Weerapa"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5539/hes.v15n3p262","URL":"https://doi.org/10.5539/hes.v15n3p262","source":"openalex"},{"id":"oa:W4412618920","type":"article-journal","title":"Advances in cardiac devices and bioelectronics augmented with artificial intelligence","abstract":"Diagnostic bioelectronics such as the electrocardiogram have become increasingly prevalent in the management of cardiovascular diseases like atrial fibrillation. While these devices provide meaningful clinical value, their complex and profuse data output requires considerable labour and training from clinicians to diagnose disease. This abundant production of complex data has made diagnostic bioelectronics a prime target for artificial intelligence (AI) integration. AI-integrated diagnostic bioelectronics have already left the clinic as widely prevalent wearable smartwatches equipped with single-lead electrocardiography sensors. Meanwhile, substantial innovation is also taking place at the intersection of other sensing modalities in the form of edge computing. Here, we overview the implementation and embedding of AI into diagnostic bioelectronics of multiple sensing modalities, including electrocardiography, photoplethysmography, echocardiography, and others, and discuss the recent advances made by medical device companies and researchers alike at the interface between the heart and AI.","author":[{"family":"Stark","given":"Charles"},{"family":"Rytkin","given":"Eric"},{"family":"Mircéa","given":"A"},{"family":"Efimov","given":"Igor"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1113/jp287135","URL":"https://doi.org/10.1113/jp287135","source":"openalex"},{"id":"oa:W4407410690","type":"article-journal","title":"The use of artificial intelligence in public administration: Bibliometric analysis","abstract":"Artificial intelligence in public administration is critical for the modernization of the public sector and adaptation to the challenges of modern society. The paper analyzes studies dedicated to the impact of artificial intelligence on the efficiency, innovation, and transparency of management processes in the public sector using meta- and bibliometric analysis. The goal is to identify the main areas and keywords that highlight theoretical and practical aspects of artificial intelligence in public administration. In total, 879 scientific articles were analyzed, of which 598 works are devoted to artificial intelligence. Dynamic time analysis revealed a significant surge in scientific interest in artificial intelligence in public administration: from 2010 to 2019, 135 publications were devoted to this issue, and from 2020 to 2024, 421. Bibliographic maps of keywords and publication maps showed the main thematic areas of research on artificial intelligence: the application of AI in public administration and the public sector, decision-making in public administration, data management and digital technologies in public administration, and the strategic use of AI to forecast socio-economic trends.The obtained data became the basis for expanding the scientific and practical potential of using artificial intelligence in public administration. The main areas of future research will concern the regulation of ethical issues to ensure the trust of citizens, the development of a regulatory framework and standards, increasing the efficiency of public services, the integration of artificial intelligence into strategic planning, and the use of artificial intelligence to achieve sustainable development goals. AcknowledgmentThe analysis was carried out within the framework of the implementation of the perspective plan for the development of the scientific area “Social Sciences” of Sumy State University, number d/r 0121U112685.","author":[{"family":"Rekunenko","given":"Іhor"},{"family":"Кобушко","given":"Яна"},{"family":"Dzydzyguri","given":"Oleksii"},{"family":"Balahurovska","given":"Іnna"},{"family":"Yurynets","given":"Oksana"},{"family":"Жук","given":"Олександр"}],"issued":{"date-parts":[[2025]]},"DOI":"10.21511/ppm.23(1).2025.16","URL":"https://doi.org/10.21511/ppm.23(1).2025.16","source":"openalex"},{"id":"oa:W7133960409","type":"article-journal","title":"Artificial intelligence in immunotherapy: revolutionizing diagnostic and therapeutic applications in cancer and autoimmune diseases","abstract":"Artificial intelligence (AI) is increasingly advancing precision immunotherapy by integrating high-dimensional biomedical data to support diagnosis, treatment selection, and longitudinal monitoring in both cancer and autoimmune diseases. This review summarizes AI applications in biomarker discovery, prediction of immune checkpoint inhibitor (ICI) response and toxicity, neoantigen prioritization, CAR-T cell optimization, and therapeutic antibody engineering. In oncology, multimodal models combining multi-omics, medical imaging, and clinical variables improve patient stratification and non-invasive response assessment, with several imaging- and pathology-based prediction tasks reporting clinically meaningful performance (frequently AUC ~ 0.70–0.95 across tumor types and endpoints). In autoimmune diseases, AI enables earlier diagnosis, molecular subtyping, treatment-response prediction, and real-time disease activity tracking using EHR, laboratory, imaging, and wearable data—supporting precision management in conditions such as rheumatoid arthritis and type 1 diabetes. Key challenges include data heterogeneity, model interpretability, and governance; however, explainable AI, federated learning, and digital twin frameworks offer practical routes toward trustworthy clinical translation. Overall, AI is emerging as a foundational technology for next-generation, patient-specific immunotherapy across oncology and autoimmune medicine.","author":[{"family":"Alshorman","given":"Jamal"},{"family":"Mehran","given":"Mohammad"},{"family":"Bahrami","given":"Yadollah"},{"family":"Mohammadzadeh","given":"Sara"},{"family":"Barzigar","given":"Rambod"},{"family":"Morshedi","given":"Mahdi"},{"family":"Haider","given":"Khawaja"},{"family":"Tembo","given":"Kingsley"},{"family":"Rong","given":"Shan"},{"family":"Jadgal","given":"Nasir"},{"family":"Altahla","given":"Ruba"},{"family":"Bolideei","given":"Mansoor"},{"family":"Wang","given":"Yongping"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1007/s10238-026-02107-5","URL":"https://doi.org/10.1007/s10238-026-02107-5","source":"openalex"},{"id":"oa:W4414078552","type":"article-journal","title":"ESCMID workshop: Artificial intelligence and machine learning in medical microbiology diagnostics","abstract":"Rapid advancements in artificial intelligence (AI) and machine learning (ML) offer significant potential to transform medical microbiology diagnostics, improving pathogen identification, antimicrobial susceptibility prediction and outbreak detection. To address these opportunities and challenges, the ESCMID workshop, \"Artificial Intelligence and Machine Learning in Medical Microbiology Diagnostics\", was held in Zurich, Switzerland, from June 2-5, 2025. The course featured expert lectures, practical sessions and panel discussions covering foundational ML concepts and deep learning architectures, data interoperability, quality control processes, model development and validation strategies. Key applications discussed included whole-genome sequencing for antimicrobial resistance detection, AI-enhanced digital microscopy automation and MALDI-TOF mass spectrometry-based diagnostics. Participants gained hands-on experience with essential AI tools and platforms. Special emphasis was placed on standardised laboratory protocols, regulatory compliance and ethical considerations, including data governance and patient privacy. Panel sessions further highlighted critical issues of equity, global disparities in AI access, sustainability and environmental impacts related to AI infrastructure. The workshop concluded by underscoring a necessity for ongoing interdisciplinary collaboration, continued education, and substantial investment in equitable AI infrastructure to realise the full potential of AI in clinical diagnostics.","author":[{"family":"Greutmann","given":"Mariella"},{"family":"Borgwardt","given":"Karsten"},{"family":"Brüningk","given":"Sarah"},{"family":"Franzeck","given":"Fabian"},{"family":"Giske","given":"Christian"},{"family":"Green","given":"Anna"},{"family":"Guerrero-López","given":"Alejandro"},{"family":"Ip","given":"Margaret"},{"family":"Jutzeler","given":"Catherine"},{"family":"Kahles","given":"André"},{"family":"Krauthammer","given":"Michael"},{"family":"Maćešić","given":"Nenad"},{"family":"Mcfadden","given":"Benjamin"},{"family":"Meijer","given":"Eline"},{"family":"Moore","given":"Nathan"},{"family":"Morangilad","given":"Jacob"},{"family":"Lboukili","given":"Imane"},{"family":"Nolte","given":"Oliver"},{"family":"Patel","given":"Robin"},{"family":"Schneider","given":"Gerold"},{"family":"Seeger","given":"Markus"},{"family":"Sethi","given":"Tavpritesh"},{"family":"Skov","given":"Robert"},{"family":"Yoon","given":"Chang"},{"family":"Rodríguezsánchez","given":"Belén"},{"family":"Egli","given":"Adrian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.micinf.2025.105562","URL":"https://doi.org/10.1016/j.micinf.2025.105562","source":"openalex"},{"id":"oa:W4409526623","type":"article-journal","title":"Artificial Intelligence in Frontline Service Encounters: A Systematic Review and Research Agenda","abstract":"ABSTRACT Over recent years, the proliferation of artificial intelligence (AI) has enabled businesses worldwide to employ AI‐driven service agents to deliver frontline services to their customers. This paradigm shift has also increased scholarly attention to consumer behavior research in AI‐driven frontline service encounters. Nevertheless, the existing body of knowledge in this domain lacks coherence and consistency, with disparate findings scattered across numerous disciplines. In this context, a critical and comprehensive overview of the existing literature in this domain is essential. Therefore, to provide an updated and comprehensive understanding of consumer research in this fast‐growing domain, we systematically analyzed 157 articles. Using the popular TCCM framework, we offer a detailed overview of the theories, contexts, characteristics, and methodologies used in the prior studies in this domain. The research also presents an integrated framework considering the independent variables, mediators, and moderators influencing customer outcomes. This analysis identifies several research gaps and suggests potential opportunities for further investigation that pertain to major emerging topics and overlooked areas. This review enhances the understanding of consumer reactions to AI‐driven frontline service encounters and offers novel insights for both the literature and managerial practice regarding the implementation of AI in frontline services.","author":[{"family":"George","given":"Sneha"},{"family":"Manu","given":"C"},{"family":"Edward","given":"Manoj"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/ijcs.70048","URL":"https://doi.org/10.1111/ijcs.70048","source":"openalex"},{"id":"oa:W4409454622","type":"article-journal","title":"Integrating artificial intelligence and machine learning with numerical simulation for enhanced thermal performance of ternary nanofluid","abstract":"Abstract The motivation for this investigation stems from a perceived gap in the vast literature on nanofluids, specifically in relation to their interactions with different surfaces and their numerical simulation. The main objective of this study is to effectively utilize novel machine learning (ML) and artificial intelligence (AI) techniques to investigate the thermal behavior of magnetohydrodynamic ternary nanofluids via an impermeable cylinder subject to activation energy and chemical reactions. We adopt the Levenberg–Marquardt algorithm with backpropagation artificial neural network technique (LMA-ANN), an AI-based scheme, to achieve this goal. The transition of governing equations to ordinary differential equations is accomplished through the use of similarity scaling. Obtained equations are then numerically evaluated using modified finite difference discretization (the Keller-Box approach). Regression scores equal to 1 indicate an excellent match between the numerical data and the predictions. The results demonstrate that temperature diminishes with the activation energy component, but it escalates with the chemical reaction. The activation energy parameter enhances both heat and mass transport processes. The results produced by this framework possess significant significance and usefulness in the field of biotechnology, drug delivery, cancer treatment, biological engineering, and bio-imaging.","author":[{"family":"Hussain","given":"Mohib"},{"family":"Du","given":"Lin"},{"family":"Waqas","given":"Hassan"},{"family":"Almdallal","given":"Qasem"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/jcde/qwaf041","URL":"https://doi.org/10.1093/jcde/qwaf041","source":"openalex"},{"id":"oa:W7160503958","type":"article-journal","title":"Evolving surgical teams in the age of artificial intelligence and robotics","abstract":"Surgery is a critical function of the healthcare system, key to addressing a substantial portion of the global disease burden. The integration of advanced artificial intelligence (AI) and robotics ecosystems into the operating room (OR) promises to radically transform surgery, with profound implications. This article analyzes the current state of surgical AI and robotic systems; presents a vision for their future, highlighting technological and research challenges and their associated impact on surgical teams; and discusses the ethical and regulatory implications. AI systems will use complex, multimodal data streams collected from patients, surgical teams, robots, and the OR environment to become increasingly capable of situational awareness, workflow recognition, performance benchmarking, causal inference, outcome prediction, and intraoperative decision-making to optimize surgical actions. Robotics will move from passive instrument-handling tools to autonomous systems with human-in-the-loop control, with embodied AI and enhanced sensor-based perception providing comprehensive spatial–temporal understanding, anticipatory behaviors, and adaptive learning. The surgeon’s role will shift toward supervision, coordination, and high-level decision-making, while nurses, assistants, and anesthesiologists will have additional competencies complemented by clinical data scientists and AI and robotic integration engineers. Ethical challenges will include liability and the implications of diluted authority chains, the potential for AI bias to exacerbate health inequalities, and the concentration of research and industry in resource-rich nations. New regulatory and compliance frameworks, trial methods, reporting standards, and training approaches will be needed to ensure the safety and effectiveness of these systems. Ultimately, AI and robotics should sustain, rather than disrupt, surgical practices by refining the skills of care providers to achieve true personalized surgery and propel procedural and technological advancements.","author":[{"family":"Granados","given":"Alejandro"},{"family":"Khanna","given":"Raghav"},{"family":"Fischer","given":"Nikola"},{"family":"Raison","given":"Nicholas"},{"family":"Ciabattini","given":"Margarita"},{"family":"Robertshaw","given":"Harry"},{"family":"Boels","given":"Maxence"},{"family":"Malik","given":"Mohsan"},{"family":"Granados","given":"Verónica"},{"family":"Vercauteren","given":"Tom"},{"family":"Shapey","given":"Jonathan"},{"family":"Booth","given":"T"},{"family":"Arora","given":"Asit"},{"family":"Gandaglia","given":"Giorgio"},{"family":"Briganti","given":"Alberto"},{"family":"Montorsi","given":"Francesco"},{"family":"Bergeles","given":"Christos"},{"family":"Ourselin","given":"Sebastien"},{"family":"Dasgupta","given":"Prokar"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/fsci.2026.1783803","URL":"https://doi.org/10.3389/fsci.2026.1783803","source":"openalex"},{"id":"oa:W4412043430","type":"article-journal","title":"Integrating artificial intelligence in healthcare: applications, challenges, and future directions","abstract":"Artificial intelligence (AI) has demonstrated remarkable potential in transforming medical diagnostics across various healthcare domains. This paper explores AI applications in cancer detection, dental medicine, brain tumor database management, and personalized treatment planning. AI technologies such as machine learning and deep learning have enhanced diagnostic accuracy, improved data management, and facilitated personalized treatment strategies. In cancer detection, AI-driven imaging analysis aids in early diagnosis and precise treatment decisions. In dental healthcare, AI applications improve oral disease detection, treatment planning, and workflow efficiency. AI-powered brain tumor databases streamline medical data management, enhancing diagnostic precision and research outcomes. Personalized treatment planning benefits from AI algorithms that analyze genetic, clinical, and lifestyle data to recommend tailored interventions. Despite these advancements, AI integration faces challenges related to data privacy, algorithm bias, and regulatory concerns. Addressing these issues requires improved data governance, ethical frameworks, and interdisciplinary collaboration among healthcare professionals, researchers, and policymakers. Through comprehensive validation, educational initiatives, and standardized protocols, AI adoption in healthcare can enhance patient outcomes and optimize clinical decision-making, advancing the future of precision medicine and personalized care.","author":[{"family":"Chong","given":"Peng"},{"family":"Vaigeshwari","given":"Vikneswaran"},{"family":"Reyasudin","given":"Basir"},{"family":"Hidayah","given":"Binti"},{"family":"Tatchanaamoorti","given":"Purnshatman"},{"family":"Yeow","given":"Jian"},{"family":"Kong","given":"Feng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/20565623.2025.2527505","URL":"https://doi.org/10.1080/20565623.2025.2527505","source":"openalex"},{"id":"oa:W4413051177","type":"article-journal","title":"Patient Preferences for Artificial Intelligence in Medical Imaging: A Single-Center Cross-Sectional Survey","abstract":"Artificial Intelligence (AI) is rapidly being implemented into clinical practice to improve diagnostic accuracy and reduce provider burnout. However, patient self-perceived knowledge and perceptions of AI's role in their care remain unclear. This study aims to explore patient preferences regarding the use of and communication of AI in their care for patients undergoing cross-sectional imaging exams. This single-center cross-sectional study, a structured questionnaire recruited patients undergoing outpatient CT or MRI examinations between June and July 2024 to assess baseline self-perceived knowledge of AI, perspectives on AI in clinical care, preferences regarding AI-generated results, and economic considerations related to AI, using Likert scales and categorical questions. A total of 226 participants (143 females; mean age 53 years) were surveyed with 67.4% (151/224) reporting having minimal to no knowledge of AI in medicine, with lower knowledge levels associated with lower socioeconomic status (p < .001). 90.3% (204/226) believed they should be informed about the use of AI in their care, and 91.1% (204/224) supported the right to opt out. Additionally, 91.1% (204/224) of participants expressed a strong preference for being informed when AI was involved in interpreting their medical images. 65.6% (143/218) indicated that they would not accept a screening imaging exam exclusively interpreted by an AI algorithm. Finally, 91.1% (204/224) of participants wanted disclosure when AI was used and 89.1% (196/220) felt such disclosure and clarification of discrepancies should be considered standard care. To align AI adoption with patient preferences and expectations, radiology practices must prioritize disclosure, patient engagement, and standardized documentation of AI use without being overly burdensome to the diagnostic workflow. Patients prefer transparency for AI utilization in their care, and our study highlights the discrepancy between patient preferences and current clinical practice. Patients are not expected to determine the technical aspects of an image examination such as acquisition parameters or reconstruction kernel and must trust their providers to act in their best interest. Clear communication of how AI is being used in their care should be provided in ways that do not overly burden the radiologist.","author":[{"family":"Mcghee","given":"Kennedye"},{"family":"Barrett","given":"DJ"},{"family":"Safarini","given":"Omar"},{"family":"Elkassem","given":"Asser"},{"family":"Eddins","given":"John"},{"family":"Smith","given":"Andrew"},{"family":"Rothenberg","given":"Steven"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s10278-025-01629-w","URL":"https://doi.org/10.1007/s10278-025-01629-w","source":"openalex"},{"id":"oa:W4412037398","type":"article-journal","title":"A Comprehensive Review of Explainable Artificial Intelligence (XAI) in Computer Vision","abstract":"Explainable Artificial Intelligence (XAI) is increasingly important in computer vision, aiming to connect complex model outputs with human understanding. This review provides a focused comparative analysis of representative XAI methods in four main categories, attribution-based, activation-based, perturbation-based, and transformer-based approaches, selected from a broader literature landscape. Attribution-based methods like Grad-CAM highlight key input regions using gradients and feature activation. Activation-based methods analyze the responses of internal neurons or feature maps to identify which parts of the input activate specific layers or units, helping to reveal hierarchical feature representations. Perturbation-based techniques, such as RISE, assess feature importance through input modifications without accessing internal model details. Transformer-based methods, which use self-attention, offer global interpretability by tracing information flow across layers. We evaluate these methods using metrics such as faithfulness, localization accuracy, efficiency, and overlap with medical annotations. We also propose a hierarchical taxonomy to classify these methods, reflecting the diversity of XAI techniques. Results show that RISE has the highest faithfulness but is computationally expensive, limiting its use in real-time scenarios. Transformer-based methods perform well in medical imaging, with high IoU scores, though interpreting attention maps requires care. These findings emphasize the need for context-aware evaluation and hybrid XAI methods balancing interpretability and efficiency. The review ends by discussing ethical and practical challenges, stressing the need for standard benchmarks and domain-specific tuning.","author":[{"family":"Cheng","given":"Zhu"},{"family":"Wu","given":"Yue"},{"family":"Li","given":"Yule"},{"family":"Cai","given":"Lingfeng"},{"family":"Ihnaini","given":"Baha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/s25134166","URL":"https://doi.org/10.3390/s25134166","source":"openalex"},{"id":"oa:W4407344156","type":"article-journal","title":"Adherence to the Checklist for Artificial Intelligence in Medical Imaging (CLAIM): an umbrella review with a comprehensive two-level analysis","abstract":"To comprehensively assess Checklist for Artificial Intelligence in Medical Imaging (CLAIM) adherence in medical imaging artificial intelligence (AI) literature by aggregating data from previous systematic and non-systematic reviews. METHODSA systematic search of PubMed, Scopus, and Google Scholar identified reviews using the CLAIM to evaluate medical imaging AI studies.Reviews were analyzed at two levels: review level (33 reviews; 1,458 studies) and study level (421 unique studies from 15 reviews).The CLAIM adherence metrics (scores and compliance rates), baseline characteristics, factors influencing adherence, and critiques of the CLAIM were analyzed. RESULTSA review-level analysis of 26 reviews (874 studies) found a weighted mean CLAIM score of 25 [standard deviation (SD): 4] and a median of 26 [interquartile range (IQR): 8; 25 th -75 th percentiles: 20-28].In a separate review-level analysis involving 18 reviews (993 studies), the weighted mean CLAIM compliance was 63% (SD: 11%), with a median of 66% (IQR: 4%; 25 th -75 th percentiles: 63%-67%).A study-level analysis of 421 unique studies published between 1997 and 2024 found a median CLAIM score of 26 (IQR: 6; 25 th -75 th percentiles: 23-29) and a median compliance of 68% (IQR: 16%; 25 th -75 th percentiles: 59%-75%).Adherence was independently associated with the journal impact factor quartile, publication year, and specific radiology subfields.After guideline publication, CLAIM compliance improved (P = 0.004).Multiple readers provided an evaluation in 85% (28/33) of reviews, but only 11% (3/28) included a reliability analysis.An item-wise evaluation identified 11 underreported items (missing in ≥50% of studies).Among the 10 identified critiques, the most common were item inapplicability to diverse study types and subjective interpretations of fulfillment. CONCLUSIONOur two-level analysis revealed considerable reporting gaps, underreported items, factors related to adherence, and common CLAIM critiques, providing actionable insights for researchers and journals to improve transparency, reproducibility, and reporting quality in AI studies. CLINICAL SIGNIFICANCEBy combining data from systematic and non-systematic reviews on CLAIM adherence, our comprehensive findings may serve as targets to help researchers and journals improve transparency, reproducibility, and reporting quality in AI studies.","author":[{"family":"Koçak","given":"Burak"},{"family":"Köse","given":"Fadime"},{"family":"Keleş","given":"Ali"},{"family":"Şendur","given":"Abdurrezzak"},{"family":"Meşe","given":"İsmail"},{"family":"Karagülle","given":"Mehmet"}],"issued":{"date-parts":[[2025]]},"DOI":"10.4274/dir.2025.243182","URL":"https://doi.org/10.4274/dir.2025.243182","source":"openalex"},{"id":"oa:W4409906372","type":"article-journal","title":"Readiness towards artificial intelligence among medical and dental undergraduate students in Peshawar, Pakistan: a cross-sectional survey","abstract":"INTRODUCTION: Artificial intelligence is a transformative tool for improving healthcare delivery and diagnostic accuracy in the medical and dental fields. This study aims to assess the readiness of future healthcare workers for artificial intelligence and address this gap by examining students' perceptions, attitudes, and knowledge related to AI in Peshawar, Pakistan. METHODS: A quantitative cross-sectional survey was conducted on 423 students from randomly chosen medical and dental colleges. The Medical AI Readiness Scale (MAIRS-MS) was used to perform a self-administered online questionnaire that was used to gather data. Using SPSS software, descriptive statistics and chi-square tests were used to evaluate the data. The level of significance was set at p ≤ 0.05. RESULTS: From multiple medical and dental colleges, 407 students participated in this survey. The survey showed that 29.7% of students had low, 62.2% had moderate, and only 8.1% had high readiness levels. Most medical and dental students in Peshawar, Pakistan, showed moderate readiness. There were significant gender discrepancies, showing males dominating females in readiness scores. There were only slight differences in the AI readiness scores and the academic years from the 1st to 5th year. Only a few non-Pakistani students responded, which may hinder conclusive determinations regarding national disparities. CONCLUSION: The study revealed moderate AI readiness among participants, with significant gender disparities favouring males. Overall, there were no significant differences between dentistry and medical fields. In-depth analysis by domain and knowledge areas might uncover further distinctions. CLINICAL TRIAL NUMBER: Not Applicable.","author":[{"family":"Baseer","given":"Saman"},{"family":"Jamil","given":"Brekhna"},{"family":"Khan","given":"Shehzad"},{"family":"Khan","given":"MOF"},{"family":"Syed","given":"Ambreen"},{"family":"Ali","given":"Liaqat"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12909-025-06911-7","URL":"https://doi.org/10.1186/s12909-025-06911-7","source":"openalex"},{"id":"doi:10.5281/zenodo.17406849","type":"article-journal","title":"CODE, TRUST AND FUTURE: THE ENGINEERING DIMENSIONS OF AI","abstract":"This volume explores how artificial intelligence is transforming healthcare, transportation, and cybersecurity. Each chapter offers a focused study on AI's practical applications, challenges, and future potential across different global contexts. The first chapter presents a hybrid CNN–ML model for breast cancer histopathology, aiming to improve diagnostic accuracy while ensuring trustworthy AI in medical settings. The second examines Lagos State's transportation system, identifying barriers to AI adoption and proposing strategies for smarter urban mobility. The final chapter analyzes cybersecurity practices in Slovak SMEs, highlighting vulnerabilities and offering tailored recommendations. Together, these studies underscore the need for responsible, context-aware AI to drive innovation and resilience across sectors.","author":[{"family":"Gubalova","given":"Jolana"},{"family":"Ripon","given":"Shamim"},{"family":"Rahman","given":"Rubaiya"},{"family":"Piash","given":"Moshiur"},{"family":"Chijioke","given":"Christian"},{"family":"Adigun","given":"Gbolahan"},{"family":"Loye","given":"Omoboriowo"},{"family":"Akosile","given":"Samuel"},{"family":"Gubalova","given":"Jolana"},{"family":"Hlavac","given":"Robert"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.17406849","URL":"https://doi.org/10.5281/zenodo.17406849","source":"datacite"},{"id":"doi:10.5281/zenodo.17406850","type":"article-journal","title":"CODE, TRUST AND FUTURE: THE ENGINEERING DIMENSIONS OF AI","abstract":"This volume explores how artificial intelligence is transforming healthcare, transportation, and cybersecurity. Each chapter offers a focused study on AI's practical applications, challenges, and future potential across different global contexts. The first chapter presents a hybrid CNN–ML model for breast cancer histopathology, aiming to improve diagnostic accuracy while ensuring trustworthy AI in medical settings. The second examines Lagos State's transportation system, identifying barriers to AI adoption and proposing strategies for smarter urban mobility. The final chapter analyzes cybersecurity practices in Slovak SMEs, highlighting vulnerabilities and offering tailored recommendations. Together, these studies underscore the need for responsible, context-aware AI to drive innovation and resilience across sectors.","author":[{"family":"Gubalova","given":"Jolana"},{"family":"Ripon","given":"Shamim"},{"family":"Rahman","given":"Rubaiya"},{"family":"Piash","given":"Moshiur"},{"family":"Chijioke","given":"Christian"},{"family":"Adigun","given":"Gbolahan"},{"family":"Loye","given":"Omoboriowo"},{"family":"Akosile","given":"Samuel"},{"family":"Gubalova","given":"Jolana"},{"family":"Hlavac","given":"Robert"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.17406850","URL":"https://doi.org/10.5281/zenodo.17406850","source":"datacite"},{"id":"doi:10.82451/k51185","type":"article-journal","title":"Insights and Inputs: Analyzing ChatGPT Prompt Design by Future Physician","abstract":"As artificial intelligence (AI) becomes increasingly integrated into healthcare, effective prompting has emerged as a key factor in optimizing large language model (LLM) performance. While strategies such as chain-of-thought prompting can enhance reasoning, little is known about how medical students naturally construct prompts when engaging with LLMs. Given ChatGPT’s ability to pass the USMLE, the research question guiding this study was “How do medical students interact with LLMs, and how does prior digital health training influence these approaches?” To address this, we examined how medical students at Rocky Vista University used ChatGPT-4.0 to answer medically related questions. The primary objective was to characterize prompt themes and elements; the secondary objective was to compare students enrolled in a longitudinal digital health curriculum—including training in prompt engineering—with peers lacking formal instruction. An 11-question Qualtrics survey was distributed across the Colorado and Utah campuses, including six demographic items and five medical questions. Of 108 responses, 60 met eligibility criteria and were analyzed. Responses were evaluated for prompting styles, AI interactions, and digital health participation. Findings revealed challenges for both students and ChatGPT in osteopathic principles and practice (OPP), particularly with sacral landmarks and axis application (correct response rate ~52%). In contrast, ChatGPT answered an ethics-based question correctly that many students misinterpreted, highlighting differences in reasoning rather than model performance. Prompting strategy influenced outcomes: students using targeted prompts with copy-and-paste achieved the highest accuracy (75% fully correct), while most relied on unmodified copy-and-paste. Limitations include the small sample size and recruitment from a single medical program, which may limit generalizability. This study suggests that targeted prompt design improves LLM accuracy and that incorporating structured prompting instruction may be critical to preparing medical students to engage with AI responsibly and productively in medical education.","author":[{"family":"Szeto","given":"Tiffani"},{"family":"Ahmad","given":"Hamza"},{"family":"Hamzavi","given":"Zaakir"},{"family":"Walters","given":"Hannah"},{"family":"Stiegmann","given":"Regan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.82451/k51185","URL":"https://doi.org/10.82451/k51185","source":"datacite"},{"id":"doi:10.82497/aitsde.2026.1246736","type":"article-journal","title":"Alzheimer Disease Classification Based on Phase Transfer Entropy Method of EEG Signal","abstract":"Alzheimer's disease (AD) is the most common type of dementia with a progressive neurodegenerative course, characterized by the accumulation of amyloid-beta plaques and tau tangles in the brain. The global prevalence of this disease is rapidly increasing with the aging population, and it is projected to reach over 152 million cases by 2050. Despite decades of research, there is still no definitive cure for AD, and current pharmacological interventions are limited to symptomatic relief with modest efficacy. In this regard, the development of automated methods based on biological signals (such as EEG) for early diagnosis and non-invasive screening of patients has emerged as a critical research approach.Due to the non-stationary nature of biological signals, most of the useful information lies in the frequency domain, and a complete representation of this information can be observed in time-frequency features. Numerous methods have been proposed for the automated classification of signals and medical images, and there is a continuous need for methods with higher accuracy and greater automation. On the other hand, reducing the number of channels used in processing decreases computation time and system complexity; however, optimal channel selection is of particular importance. In this study, based on the Phase Transfer Entropy method for calculating inter-channel connectivity, channels that showed the greatest difference between the two groups were selected as the chosen channels. In this paper, two selected channels are used for Alzheimer's diagnosis to reduce system complexity. In this research, 10-fold cross-validation was employed for the results. The obtained accuracy, sensitivity, and specificity were 97.9%, 98%, and 98%, respectively. The proposed method offers a novel approach to improve classification performance and reduce computational burden, and it is a suitable method for Alzheimer's diagnosis using an optimized convolutional neural network.","author":[{"family":"Gholami","given":"Maryam"},{"family":"Fatehi","given":"Mohammad"},{"family":"Pirbonyeh","given":"Mohammad"},{"family":"Taghizadeh","given":"Mehdi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.82497/aitsde.2026.1246736","URL":"https://doi.org/10.82497/aitsde.2026.1246736","source":"datacite"},{"id":"doi:10.5281/zenodo.20702145","type":"article-journal","title":"Medicura: AI-Based Medical Report Analysis and Healthcare Recommendation Platform","abstract":"The growing demand for accessible and intelligent healthcare solutions has increased the need for systems that can provide medical assistance in a simple, efficient, and user-friendly manner. Many individuals face difficulties in understanding medical reports, obtaining personalized healthcare guidance, and accessing healthcare services due to language and communication barriers. To address these challenges, this research presents Medicura, an AI-enabled healthcare management and assistance platform developed to support users through automated analysis and personalized healthcare services. Medicura integrates Artificial Intelligence (AI), Machine Learning (ML), Natural Language Processing (NLP), and location-aware technologies to deliver multiple healthcare functionalities within a unified system. The platform performs medical report analysis by extracting important information and converting complex medical content into simplified summaries for easier understanding. The system also includes an interactive healthcare chatbot that enables users to communicate naturally and receive immediate assistance. Additionally, Medicura provides personalized diet recommendations and medicine-related guidance based on analyzed information to encourage better health management. To improve usability and accessibility, the platform supports multilingual communication and offers location-based doctor recommendations that help users connect with nearby healthcare services.","author":[{"family":"Shaikh","given":"Mohd"},{"family":"Tamboli","given":"Mohd"},{"family":"Meshram","given":"Parth"},{"family":"Chandelkar","given":"Shivam"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20702145","URL":"https://doi.org/10.5281/zenodo.20702145","source":"datacite"},{"id":"doi:10.5281/zenodo.20702146","type":"article-journal","title":"Medicura: AI-Based Medical Report Analysis and Healthcare Recommendation Platform","abstract":"The growing demand for accessible and intelligent healthcare solutions has increased the need for systems that can provide medical assistance in a simple, efficient, and user-friendly manner. Many individuals face difficulties in understanding medical reports, obtaining personalized healthcare guidance, and accessing healthcare services due to language and communication barriers. To address these challenges, this research presents Medicura, an AI-enabled healthcare management and assistance platform developed to support users through automated analysis and personalized healthcare services. Medicura integrates Artificial Intelligence (AI), Machine Learning (ML), Natural Language Processing (NLP), and location-aware technologies to deliver multiple healthcare functionalities within a unified system. The platform performs medical report analysis by extracting important information and converting complex medical content into simplified summaries for easier understanding. The system also includes an interactive healthcare chatbot that enables users to communicate naturally and receive immediate assistance. Additionally, Medicura provides personalized diet recommendations and medicine-related guidance based on analyzed information to encourage better health management. To improve usability and accessibility, the platform supports multilingual communication and offers location-based doctor recommendations that help users connect with nearby healthcare services.","author":[{"family":"Shaikh","given":"Mohd"},{"family":"Tamboli","given":"Mohd"},{"family":"Meshram","given":"Parth"},{"family":"Chandelkar","given":"Shivam"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20702146","URL":"https://doi.org/10.5281/zenodo.20702146","source":"datacite"},{"id":"oa:W4408764619","type":"article-journal","title":"Artificial intelligence-driven forecasting and shift optimization for pediatric emergency department crowding","abstract":"Objective: This study aimed to develop and evaluate an artificial intelligence (AI)-driven system for forecasting Pediatric Emergency Department (PED) overcrowding and optimizing physician shift schedules using machine learning operations (MLOps). Materials and Methods: Data from 352 843 PED admissions between January 2018 and May 2023 were analyzed. Twenty time-series forecasting models-including classical methods and advanced deep learning architectures like Temporal Convolutional Network, Time-series Dense Encoder and Reversible Instance Normalization, Neural High-order Time Series model, and Neural Basis Expansion Analysis-were developed and compared using Python 3.8. Starting in January 2023, an MLOps simulation automated data updates and model retraining. Shift schedules were optimized based on forecasted patient volumes using integer linear programming. Results: improved to 60%. The MLOps architecture facilitated continuous model updates, enhancing forecast accuracy. Shift optimization adjusted staffing in 69 out of 84 shifts, increasing physician allocation by up to 30.4% during peak hours. This adjustment reduced the patient-to-physician ratio by an average of 4.32 patients during the 8-16 shift and 4.40 patients during the 16-24 shift. Discussion: The integration of advanced deep learning models with MLOps architecture allowed for continuous model updates, enhancing the accuracy of PED overcrowding forecasts and outperforming traditional methods. The AI-driven system demonstrated resilience against data drift caused by events like the COVID-19 pandemic, adapting to changing conditions. Optimizing physician shifts based on these forecasts improved workforce distribution without increasing staff numbers, reducing patient load per physician during peak hours. However, limitations include the single-center design and a fixed staffing model, indicating the need for multicenter validation and implementation in settings with dynamic staffing practices. Future research should focus on expanding datasets through multicenter collaborations and developing forecasting models that provide longer lead times without compromising accuracy. Conclusions: The AI-driven forecasting and shift optimization system demonstrated the efficacy of integrating AI and MLOps in predicting PED overcrowding and optimizing physician shifts. This approach outperformed traditional methods, highlighting its potential for managing overcrowding in emergency departments. Future research should focus on multicenter validation and real-world implementation to fully leverage the benefits of this innovative system.","author":[{"family":"Akbasli","given":"Izzet"},{"family":"Bırbılen","given":"Ahmet"},{"family":"Tekşam","given":"Özlem"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/jamiaopen/ooae138","URL":"https://doi.org/10.1093/jamiaopen/ooae138","source":"openalex"},{"id":"oa:W4408186594","type":"article-journal","title":"Generative AI in Education: Perspectives Through an Academic Lens","abstract":"In this paper, we investigated the role of generative AI in education in academic publications extracted from Web of Science (3506 records; 2019–2024). The proposed methodology included three main streams: (1) Monthly analysis trends; top-ranking research areas, keywords and universities; frequency of keywords over time; a keyword co-occurrence map; collaboration networks; and a Sankey diagram illustrating the relationship between AI-related terms, publication years and research areas; (2) Sentiment analysis using a custom list of words, VADER and TextBlob; (3) Topic modeling using Latent Dirichlet Allocation (LDA). Terms such as “artificial intelligence” and “generative artificial intelligence” were predominant, but they diverged and evolved over time. By 2024, AI applications had branched into specialized fields, including education and educational research, computer science, engineering, psychology, medical informatics, healthcare sciences, general medicine and surgery. The sentiment analysis reveals a growing optimism in academic publications regarding generative AI in education, with a steady increase in positive sentiment from 2023 to 2024, while maintaining a predominantly neutral tone. Five main topics were derived from AI applications in education, based on an analysis of the most relevant terms extracted by LDA: (1) Gen-AI’s impact in education and research; (2) ChatGPT as a tool for university students and teachers; (3) Large language models (LLMs) and prompting in computing education; (4) Applications of ChatGPT in patient education; (5) ChatGPT’s performance in medical examinations. The research identified several emerging topics: discipline-specific application of LLMs, multimodal gen-AI, personalized learning, AI as a peer or tutor and cross-cultural and multilingual tools aimed at developing culturally relevant educational content and supporting the teaching of lesser-known languages. Further, gamification with generative AI involves designing interactive storytelling and adaptive educational games to enhance engagement and hybrid human–AI classrooms explore co-teaching dynamics, teacher–student relationships and the impact on classroom authority.","author":[{"family":"Întorsureanu","given":"Iulian"},{"family":"Oprea","given":"Simona‐vasilica"},{"family":"Bârã","given":"Adela"},{"family":"Vespan","given":"Dragoș"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/electronics14051053","URL":"https://doi.org/10.3390/electronics14051053","source":"openalex"},{"id":"oa:W4406853748","type":"article-journal","title":"Avances en el uso de inteligencia artificial en la educación médica latinoamericana","abstract":"Artificial intelligence is the ability of a system to emulate cognitive functions. In healthcare, it is used to support complex decision-making and medical skills training. It is a tool for creating virtual simulation scenarios and evaluating the performance of medical students. This literature review aims to describe the advances in artificial intelligence in medical education in Latin America. The databases PubMed, SciELO, and Google Scholar were consulted; publications in Spanish and English from 2019 to 2024 were included, and keywords and Boolean operators were applied. Artificial intelligence in medical training seeks to replicate cognitive skills in problem-solving and is classified into narrow artificial intelligence and general artificial intelligence. It is a transformative tool that empowers virtual reality, optimizes outcomes, offers opportunities to strengthen the effectiveness of healthcare, and makes improvements in personalizing the learning process. However, its implementation requires addressing ethical and legal challenges for its full exploitation. In Latin America, there is a steady increase in the adoption of artificial intelligence-based tools for medical staff training.","author":[{"family":"Domínguez","given":"Casto"},{"family":"Somoza","given":"Graciamaría"},{"family":"Guzmán","given":"Naara"},{"family":"Trinidad","given":"Marta"},{"family":"Reyes","given":"Alejandro"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5377/alerta.v8i1.19194","URL":"https://doi.org/10.5377/alerta.v8i1.19194","source":"openalex"},{"id":"oa:W4410206004","type":"article-journal","title":"Application of Artificial Intelligence to Deliver Healthcare From the Eye","abstract":"Importance: Oculomics is the science of analyzing ocular data to identify, diagnose, and manage systemic disease. This article focuses on prescreening, its use with retinal images analyzed by artificial intelligence (AI), to identify ocular or systemic disease or potential disease in asymptomatic individuals. The implementation of prescreening in a coordinated care system, defined as Healthcare From the Eye prescreening, has the potential to improve access, affordability, equity, quality, and safety of health care on a global level. Stakeholders include physicians, payers, policymakers, regulators and representatives from industry, government, and data privacy sectors. Observations: The combination of AI analysis of ocular data with automated technologies that capture images during routine eye examinations enables prescreening of large populations for chronic disease. Retinal images can be acquired during either a routine eye examination or in settings outside of eye care with readily accessible, safe, quick, and noninvasive retinal imaging devices. The outcome of such an examination can then be digitally communicated across relevant stakeholders in a coordinated fashion to direct a patient to screening and monitoring services. Such an approach offers the opportunity to transform health care delivery and improve early disease detection, improve access to care, enhance equity especially in rural and underserved communities, and reduce costs. Conclusions and Relevance: With effective implementation and collaboration among key stakeholders, this approach has the potential to contribute to an equitable and effective health care system.","author":[{"family":"Weinreb","given":"Robert"},{"family":"Lee","given":"Aaron"},{"family":"Baxter","given":"Sally"},{"family":"Lee","given":"Richard"},{"family":"Leng","given":"Theodore"},{"family":"Mcconnell","given":"Michael"},{"family":"El-Nimri","given":"Nevin"},{"family":"Rhew","given":"David"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1001/jamaophthalmol.2025.0881","URL":"https://doi.org/10.1001/jamaophthalmol.2025.0881","source":"openalex"},{"id":"oa:W4409557347","type":"article-journal","title":"Research advancements in the Use of artificial intelligence for prenatal diagnosis of neural tube defects","abstract":"Artificial Intelligence is revolutionizing prenatal diagnostics by enhancing the accuracy and efficiency of procedures. This review explores AI and machine learning (ML) in the early detection, prediction, and assessment of neural tube defects (NTDs) through prenatal ultrasound imaging. Recent studies highlight the effectiveness of AI techniques, such as convolutional neural networks (CNNs) and support vector machines (SVMs), achieving detection accuracy rates of up to 95% across various datasets, including fetal ultrasound images, genetic data, and maternal health records. SVM models have demonstrated 71.50% accuracy on training datasets and 68.57% on testing datasets for NTD classification, while advanced deep learning (DL) methods report patient-level prediction accuracy of 94.5% and an area under the receiver operating characteristic curve (AUROC) of 99.3%. AI integration with genomic analysis has identified key biomarkers associated with NTDs, such as Growth Associated Protein 43 (GAP43) and Glial Fibrillary Acidic Protein (GFAP), with logistic regression models achieving 86.67% accuracy. Current AI-assisted ultrasound technologies have improved diagnostic accuracy, yielding sensitivity and specificity rates of 88.9% and 98.0%, respectively, compared to traditional methods with 81.5% sensitivity and 92.2% specificity. AI systems have also streamlined workflows, reducing median scan times from 19.7 min to 11.4 min, allowing sonographers to prioritize critical patient care. Advancements in DL algorithms, including Oct-U-Net and PAICS, have achieved recall and precision rates of 0.93 and 0.96, respectively, in identifying fetal abnormalities. Moreover, AI's evolving role in genetic research supports personalized NTD prevention strategies and enhances public awareness through AI-generated health messages. In conclusion, the integration of AI in prenatal diagnostics significantly improves the detection and assessment of NTDs, leading to greater accuracy and efficiency in ultrasound imaging. As AI continues to advance, it has the potential to further enhance personalized healthcare strategies and raise public awareness about NTDs, ultimately contributing to better maternal and fetal outcomes.","author":[{"family":"Yeganegi","given":"Maryam"},{"family":"Danaei","given":"Mahsa"},{"family":"Azizi","given":"Sepideh"},{"family":"Jayervand","given":"Fatemeh"},{"family":"Bahrami","given":"Reza"},{"family":"Dastgheib","given":"Seyed"},{"family":"Rashnavadi","given":"Heewa"},{"family":"Masoudi","given":"Ali"},{"family":"Shiri","given":"Amirmasoud"},{"family":"Aghili","given":"Kazem"},{"family":"Noorishadkam","given":"Mahmood"},{"family":"Neámatzadeh","given":"Hossein"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fped.2025.1514447","URL":"https://doi.org/10.3389/fped.2025.1514447","source":"openalex"},{"id":"oa:W7128024495","type":"article-journal","title":"Recent Advances in Microfluidic Chip Technology for Laboratory Medicine: Innovations and Artificial Intelligence Integration","abstract":"Microfluidic chip technologies, also known as lab-on-a-chip systems, have profoundly transformed laboratory medicine by enabling the miniaturization, automation, and rapid processing of complex diagnostic assays using minimal sample volumes. Recent advances in chip design, fabrication methods-including 3D printing, modular and flexible substrates-and biosensor integration have significantly enhanced the performance, sensitivity, and clinical applicability of these devices. Integration of advanced biosensors allows for real-time detection of circulating tumor cells, nucleic acids, and exosomes, supporting innovative applications in cancer diagnostics, infectious disease detection, point-of-care testing (POCT), personalized medicine, and therapeutic monitoring. Notably, the convergence of microfluidics with artificial intelligence (AI) and machine learning has amplified device automation, reliability, and analytical power, resulting in \"smart\" diagnostic platforms capable of self-optimization, automated analysis, and clinical decision support. Emerging applications in fields such as neuroscience diagnostics and microbiome profiling further highlight the broad potential of microfluidic technology. Here, we present findings from a comprehensive review of recent innovations in microfluidic chip design and fabrication, advances in biosensor and AI integration, and their clinical applications in laboratory medicine. We also discuss current challenges in manufacturing, clinical validation, and system integration, as well as future directions for translating next-generation microfluidic technologies into routine clinical and public health practice.","author":[{"family":"Cai","given":"Hong"},{"family":"Wang","given":"Dongxia"},{"family":"Zhao","given":"Yiqun"},{"family":"Yang","given":"Chunhui"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/bios16020104","URL":"https://doi.org/10.3390/bios16020104","source":"openalex"},{"id":"oa:W4409896229","type":"article-journal","title":"Artificial intelligence based multispecialty mortality prediction models for septic shock in a multicenter retrospective study","abstract":"Septic shock is one of the most lethal conditions in ICU, and early risk prediction may help reduce mortality. We developed a TOPSIS-based Classification Fusion (TCF) model to predict mortality risk in septic shock patients using data from 4872 ICU patients from February 2003 to November 2023 across three hospitals. The model integrates seven machine learning models via the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS), achieving AUCs of 0.733 in internal validation, 0.808 in the pediatric ICU, 0.662 in the respiratory ICU, with external validation AUCs of 0.784 and 0.786, respectively. It demonstrated high stability and accuracy in cross-specialty and multi-center validation. This interpretable model provides clinicians with a reliable early-warning tool for septic shock mortality risk, facilitating early intervention to reduce mortality.","author":[{"family":"Wang","given":"Shurui"},{"family":"Liu","given":"Xinyi"},{"family":"Yuan","given":"Shaohua"},{"family":"Bian","given":"Yi"},{"family":"Wu","given":"Hong"},{"family":"Ye","given":"Qing"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01643-w","URL":"https://doi.org/10.1038/s41746-025-01643-w","source":"openalex"},{"id":"oa:W4412020208","type":"article-journal","title":"Assessing risk of bias in toxicological studies in the era of artificial intelligence","abstract":"Risk of bias is a critical factor influencing the reliability and validity of toxicological studies, impacting evidence synthesis and decision-making in regulatory and public health contexts. The traditional approaches for assessing risk of bias are often subjective and time-consuming. Recent advancements in artificial intelligence (AI) offer promising solutions for automating and enhancing bias detection and evaluation. This article reviews key types of biases-such as selection, performance, detection, attrition, and reporting biases-in in vivo, in vitro, and in silico studies. It further discusses specialized tools, including the SYRCLE and OHAT frameworks, designed to address such biases. The integration of AI-based tools into risk of bias assessments can significantly improve the efficiency, consistency, and accuracy of evaluations. However, AI models are themselves susceptible to algorithmic and data biases, necessitating robust validation and transparency in their development. The article highlights the need for standardized, AI-enabled risk of bias assessment methodologies, training, and policy implementation to mitigate biases in AI-driven analyses. The strategies for leveraging AI to screen studies, detect anomalies, and support systematic reviews are explored. By adopting these advanced methodologies, toxicologists and regulators can enhance the quality and reliability of toxicological evidence, promoting evidence-based practices and ensuring more informed decision-making. The way forward includes fostering interdisciplinary collaboration, developing bias-resilient AI models, and creating a research culture that actively addresses bias through transparent and rigorous practices.","author":[{"family":"Härtung","given":"Thomas"},{"family":"Hoffmann","given":"Sebastian"},{"family":"Whaley","given":"Paul"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00204-025-03978-5","URL":"https://doi.org/10.1007/s00204-025-03978-5","source":"openalex"},{"id":"oa:W4414836009","type":"article-journal","title":"Artificial Intelligence Threatens Critical Thinking in Education Systems","abstract":"We examine how the increasing use of artificial intelligence (AI) in education—through tools that generate and summarize text, translate languages, and produce visual content—impacts students' critical thinking. While these technologies enhance personalized learning, broaden assessment strategies, and support data-driven policy decisions, we argue that their integration into the learning process carries unintended cognitive consequences. Specifically, we show that when students offload key tasks to AI systems, their cognitive load decreases in ways that weaken memory retention and reduce active engagement with content. This shift fosters a pattern of overreliance, as students increasingly depend on AI to perform intellectual tasks in their place. As a result, their ability to think critically, question information, and evaluate sources diminishes over time. We highlight this emerging dependency as a medium- to long-term threat to critical thinking and call for a more careful evaluation of how generative AI is used in education—not only in terms of its benefits, but also its influence on core cognitive processes. Finally, we propose targeted strategies to mitigate these effects and preserve students' critical capacities in AI-rich learning environments","author":[{"family":"Özer","given":"Mahmut"},{"family":"Tanberkan","given":"Hande"},{"family":"Perc","given":"Matjaž"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5961/higheredusci.1747885","URL":"https://doi.org/10.5961/higheredusci.1747885","source":"openalex"},{"id":"oa:W4406287889","type":"article-journal","title":"Exploring Health Sciences Students' Perspectives on Using Generative Artificial Intelligence in Higher Education: A Qualitative Study","abstract":"The widespread adoption of artificial intelligence (AI) tools in academic settings has the potential to revolutionize learning experiences, enhance educational outcomes, and streamline academic processes. The aim of this research was to explore the perceptions of Lebanese health sciences students regarding the use of generative AI in higher education. A qualitative descriptive research design informed by descriptive phenomenology was employed. Semi-structured interviews were carried out among 23 health sciences students at one major private university in Beirut. Inductive thematic analysis was conducted over the period of 3 months. The inductive thematic analysis generated two themes, and eight subthemes highlighting the benefits and concerns in using AI; customized, self-paced, and autonomous learning, improved language and writing skills, development of innovative concepts, enhanced efficiency, accuracy of information, overreliance on AI, equitable access, unclear policies. Results from this study emphasized the importance of combined efforts across sectors to close access gaps, encourage inclusiveness, and develop well framed policies that enable students to utilize these new technologies for their maximum benefits.","author":[{"family":"Fawaz","given":"Mirna"},{"family":"Malti","given":"Wassim"},{"family":"Alreshidi","given":"Salman"},{"family":"Kavuran","given":"Esin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/nhs.70030","URL":"https://doi.org/10.1111/nhs.70030","source":"openalex"},{"id":"oa:W4414349719","type":"article-journal","title":"Using Artificial Intelligence to Develop Clinical Decision Support Systems—The Evolving Road of Personalized Oncologic Therapy","abstract":"Background/Objectives: The use of artificial intelligence (AI) in oncology has the potential to improve decision making, particularly in managing the risk associated with targeted therapies. This study aimed to develop and validate a machine learning-based clinical decision support system (CDSS) capable of predicting complications associated with Bevacizumab or its biosimilars and to translate the resulting predictive model into a clinically applicable tool. Methods: A prospective observational study was conducted on 395 records from patients treated with Bevacizumab or biosimilars for solid tumors. Pretherapeutic variables, such as demographic data, medical history, tumor characteristics and laboratory findings, were retrieved from medical records. Several machine learning models (logistic regression, Random Forest, XGBoost) were trained using 70/30 and 80/20 data splits. Their predictive performances were compared using accuracy, AUC-ROC, sensitivity, specificity, F1-scores and error rate. The best-performing model was used to derive a logistic-based risk score, which was further implemented as an interactive HTML form. Results: The optimized Random Forest model trained on the 80/20 split demonstrated the best balance between accuracy (70.63%), sensitivity (66.67%), specificity (73.85%), and AUC-ROC (0.75). The derived logistic risk score showed good performance (AUC-ROC = 0.720) and calibration. It identified variables, such as age ≥ 65, anemia, elevated urea, leukocytosis, tumor differentiation, and stage, as significant predictors of complications. The final tool provides clinicians with an easy-to-use, offline form that estimates individual risk levels and stratifies patients into low-, intermediate-, or high-risk categories. Conclusions: This study offers a proof of concept for developing AI-supported predictive tools in oncology using real-world data. The resulting logistic risk score and interactive form can assist clinicians in tailoring therapeutic decisions for patients receiving targeted therapies, enhancing the personalization of care without replacing clinical judgment.","author":[{"family":"Chitoran","given":"Elena"},{"family":"Rotaru","given":"Vlad"},{"family":"Gelal","given":"Aisa"},{"family":"Ionescu","given":"S"},{"family":"Gullo","given":"Giuseppe"},{"family":"Stefan","given":"Daniela"},{"family":"Simion","given":"Laurenţiu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/diagnostics15182391","URL":"https://doi.org/10.3390/diagnostics15182391","source":"openalex"},{"id":"oa:W4410035392","type":"article-journal","title":"Artificial Intelligence Approaches for Geographic Atrophy Segmentation: A Systematic Review and Meta-Analysis","abstract":"Geographic atrophy (GA) is a progressive retinal disease associated with late-stage age-related macular degeneration (AMD), a significant cause of visual impairment in senior adults. GA lesion segmentation is important for disease monitoring in clinical trials and routine ophthalmic practice; however, its manual delineation is time-consuming, laborious, and subject to inter-grader variability. The use of artificial intelligence (AI) is rapidly expanding within the medical field and could potentially improve accuracy while reducing the workload by facilitating this task. This systematic review evaluates the performance of AI algorithms for GA segmentation and highlights their key limitations from the literature. Five databases and two registries were searched from inception until 23 March 2024, following the PRISMA methodology. Twenty-four studies met the prespecified eligibility criteria, and fifteen were included in this meta-analysis. The pooled Dice similarity coefficient (DSC) was 0.91 (95% CI 0.88-0.95), signifying a high agreement between the reference standards and model predictions. The risk of bias and reporting quality were assessed using QUADAS-2 and CLAIM tools. This review provides a comprehensive evaluation of AI applications for GA segmentation and identifies areas for improvement. The findings support the potential of AI to enhance clinical workflows and highlight pathways for improved future models that could bridge the gap between research settings and real-world clinical practice.","author":[{"family":"Chatzara","given":"Aikaterini"},{"family":"Maliagkani","given":"Eirini"},{"family":"Mitsopoulou","given":"Dimitra"},{"family":"Katsimpris","given":"Andreas"},{"family":"Apostolopoulos","given":"Ioannis"},{"family":"Papageorgiou","given":"Elpiniki"},{"family":"Georgalas","given":"Ilias"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bioengineering12050475","URL":"https://doi.org/10.3390/bioengineering12050475","source":"openalex"},{"id":"oa:W4412831613","type":"article-journal","title":"Liability Risks of Ambient Clinical Workflows With Artificial Intelligence for Clinicians, Hospitals, and Manufacturers","abstract":"In August 2024, the nation's largest nonprofit integrated health care provider, Kaiser Permanente, announced that clinicians would have access to an ambient clinical documentation scribe: an assisted clinical documentation tool that uses artificial intelligence (AI) to securely summarize relevant medical information from spoken, natural conversations (also called ambient clinical documentation or AI scribes). After automatically summarizing the encounter, the AI scribe sends the summary to the clinician for review. Ambient clinical documentation scribes are now offered by some of the fastest-growing AI companies in health care, with significant venture capital funding and an impressive roster of health system customers. Technologies such as ambient clinical documentation and other generative AI tools may improve care and lessen clinician burnout by reducing documentation burdens. But they also raise the question of who is responsible when AI-generated patient information is inaccurate, especially when those errors cause injury to a patient. This question is particularly acute in cancer care, where there is a unique set of terminology for each of the more than 400 types of cancer, leading to an increased chance of documentation error, and where decisions on the basis of the assumption of information accuracy can be life-altering. AI transcription tools in their current versions are not considered regulated medical devices under the US Federal Food, Drug, and Cosmetic Act. Unless this changes, the responsibility falls to stakeholders other than the US Food and Drug Administration (FDA) to ensure the technology's safety and efficacy. In this article, we analyze the AI governance responsibilities and potential tort liability for clinicians, hospitals, and manufacturers using AI for clinical note-taking and suggest several potential ways to address them.","author":[{"family":"Gerke","given":"Sara"},{"family":"Simon","given":"David"},{"family":"Roman","given":"Benjamin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1200/op-24-01060","URL":"https://doi.org/10.1200/op-24-01060","source":"openalex"},{"id":"oa:W4406363268","type":"article-journal","title":"Survey of Artificial Intelligence Model Marketplace","abstract":"The rapid advancement and widespread adoption of artificial intelligence (AI) across diverse industries, including healthcare, finance, manufacturing, and retail, underscore the transformative potential of AI technologies. This necessitates the development of viable AI model marketplaces that facilitate the development, trading, and sharing of AI models across the pervasive industrial domains to harness and streamline their daily activities. These marketplaces act as centralized hubs, enabling stakeholders such as developers, data owners, brokers, and buyers to collaborate and exchange resources seamlessly. However, existing AI marketplaces often fail to address the demands of modern and next-generation application domains. Limitations in pricing models, standardization, and transparency hinder their efficiency, leading to a lack of scalability and user adoption. This paper aims to target researchers, industry professionals, and policymakers involved in AI development and deployment, providing actionable insights for designing robust, secure, and transparent AI marketplaces. By examining the evolving landscape of AI marketplaces, this paper identifies critical gaps in current practices, such as inadequate pricing schemes, insufficient standardization, and fragmented policy enforcement mechanisms. It further explores the AI model life-cycle, highlighting pricing, trading, tracking, security, and compliance challenges. This detailed analysis is intended for an audience with a foundational understanding of AI systems, marketplaces, and their operational ecosystems. The findings aim to inform stakeholders about the pressing need for innovation and customization in AI marketplaces while emphasizing the importance of balancing efficiency, security, and trust. This paper serves as a blueprint for the development of next-generation AI marketplaces that meet the demands of both current and future application domains, ensuring sustainable growth and widespread adoption.","author":[{"family":"Qian","given":"Mian"},{"family":"Musa","given":"Abubakar"},{"family":"Biswas","given":"Milon"},{"family":"Guo","given":"Yifan"},{"family":"Liao","given":"Weixian"},{"family":"Yu","given":"Wei"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/fi17010035","URL":"https://doi.org/10.3390/fi17010035","source":"openalex"},{"id":"oa:W4410877449","type":"article-journal","title":"Artificial intelligence-enabled prenatal ultrasound for the detection of fetal cardiac abnormalities: a systematic review and meta-analysis","abstract":"Background Advances in artificial intelligence (AI) have triggered interest in using intelligent systems to improve prenatal detection of fetal congenital heart defects (CHDs). Our aim is to systematically examine the current literature on diagnostic performance of AI-enabled prenatal cardiac ultrasound. Methods This systematic review and meta-analysis was registered with PROSPERO (CRD42024549601). Embase, Medline, Cochrane Central Database of Controlled Trials, and CINAHL were searched from inception until February 2025. Studies evaluating AI performance in prenatal detection of fetal CHDs were eligible for inclusion, and studies focusing on the application of AI before 16 weeks of gestation, or using three- or four-dimensional ultrasound, were excluded. Pooled sensitivity and specificity were obtained using random-effect method, and pooled proportions using the Freeman-Tukey arcsine square root transformation. Heterogeneity was assessed with I 2 statistics. Risk of bias and adherence to reporting standards were assessed using QUADAS-2 and TRIPOD+AI, respectively. Risk of publication bias was assessed with Deek's test and certainty of evidence for outcomes with GRADE approach. Findings Fifteen studies were included, of which fourteen developed and evaluated a model and one externally evaluated a previously trained model. Images and videos obtained during cardiac screening or fetal echocardiography of 30.121 fetuses were used for training, validation and testing. For the binary task of classifying heart as normal or abnormal, AI models achieved a pooled sensitivity of 0.89 (95% CI 0.83–0.93, I 2 = 77.92%) and specificity of 0.91 (95% CI 0.84–0.95, I 2 = 77.92%). The subgroup analysis showed that models tested on various CHDs exhibited lower sensitivity compared to those tested for a specific cardiac abnormality (0.85; 95% CI 0.75–0.91 vs 0.92; 95% CI 0.87–0.96), while specificity remained comparable (0.90; 95% CI 0.79–0.96 vs 0.91; 95% CI 0.81–0.97). Overall, AI models performed better than operators with lower expertise and were nearly comparable to experts; however, the human comparator group (median six clinicians, IQR 3–10) was usually small and non-blinded. Relevant sources of heterogeneity were the types of cardiac views collected, the prevalence of CHDs across different datasets, and the types of CHDs examined. The risk of bias was moderate-high and adherence to reporting standards low (>70% in 18/51 TRIPOD+AI items). The risk of publication bias was not statistically significant (Deek's test p=0.474). Interpretation These findings suggest that AI models perform better than clinicians with lower expertise, but this must be interpreted with caution due to the high risk of bias and sources of heterogeneity. Funding This study was partly supported by the InnoHK-funded Hong Kong Centre for Cerebro-cardiovascular Health Engineering (COCHE) Project 2.1 (Cardiovascular risks in early life and fetal echocardiography). ATP and JAN are supported by the National Institute for Health and Care Research (NIHR) Oxford Biomedical Research Centre (BRC).","author":[{"family":"Dalberti","given":"Elena"},{"family":"Patey","given":"Olga"},{"family":"Smith","given":"Carolyn"},{"family":"Šalović","given":"Bojana"},{"family":"Hernandez-Cruz","given":"Netzahualcoyotl"},{"family":"Noble","given":"JA"},{"family":"Papageorghiou","given":"Aris"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.eclinm.2025.103250","URL":"https://doi.org/10.1016/j.eclinm.2025.103250","source":"openalex"},{"id":"oa:W4412454706","type":"article-journal","title":"Decoding Trust in Artificial Intelligence: A Systematic Review of Quantitative Measures and Related Variables","abstract":"As artificial intelligence (AI) becomes ubiquitous across various fields, understanding people’s acceptance and trust in AI systems becomes essential. This review aims to identify quantitative measures used to measure trust in AI and the associated studied elements. Following the PRISMA guidelines, three databases were consulted, selecting articles published before December 2023. Ultimately, 45 articles out of 1283 were selected. Articles were included if they were peer-reviewed journal publications in English reporting empirical studies measuring trust in AI systems with multi-item questionnaires. Studies were analyzed through the lenses of cognitive and affective trust. We investigated trust definitions, questionnaires employed, types of AI systems, and trust-related constructs. Results reveal diverse trust conceptualizations and measurements. In addition, the studies covered a wide range of AI system types, including virtual assistants, content detection tools, chatbots, medical AI, robots, and educational AI. Overall, the studies show compatibility of cognitive or affective trust focus between theorization, items, experimental stimuli, and level of anthropomorphism of the systems. The review underlines the need to adapt measurement of trust in the specific characteristics of human–AI interaction, accounting for both the cognitive and affective sides. Trust definitions and measurement could be chosen depending also on the level of anthropomorphism of the systems and the context of application.","author":[{"family":"Aquilino","given":"Letizia"},{"family":"Dio","given":"Cinzia"},{"family":"Manzi","given":"Federico"},{"family":"Massaro","given":"Davide"},{"family":"Bisconti","given":"Piercosma"},{"family":"Marchetti","given":"Antonella"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/informatics12030070","URL":"https://doi.org/10.3390/informatics12030070","source":"openalex"},{"id":"oa:W4412645877","type":"article-journal","title":"The application of artificial intelligence in forensic pathology: a systematic literature review","abstract":"Introduction: Recent advancements in Artificial Intelligence have shown immense potential across various domains of healthcare, including forensic pathology. This systematic review aims to evaluate the latest innovations brought by Artificial Intelligence in forensic pathology and provide insights into future directions in this evolving field. Methods: A systematic literature search was conducted using databases for papers published from 1990 to 2025. The search strategy combined terms related to artificial intelligence, forensic odontology, forensic psychiatry and forensic medicine/pathology. Following PRISMA guidelines, 65 articles were initially identified, of which 18 met the inclusion criteria after applying exclusion criteria. Results: Artificial Intelligence applications demonstrated significant success across multiple forensic domains. In post-mortem analysis, deep learning achieved 70-94% accuracy in neurological forensics. Wound analysis systems showed high accuracy rates (87.99-98%) in gunshot wound classification. Artificial Intelligence-enhanced diatom testing for drowning cases achieved precision scores of 0.9 and recall scores of 0.95. Microbiome analysis applications reached accuracy rates up to 90% for individual identification and geographical origin determination. AI shows promise in forensic age estimation, psychiatric risk assessment, and insanity evaluations. Discussion: While Artificial Intelligence shows promise as a supportive tool in forensic pathology, several limitations persist, including small sample sizes and variable performance across different applications. Artificial Intelligence serves best as an enhancement rather than a replacement for human expertise. Future development should focus on larger datasets, specialized systems for different forensic applications, and improved interpretability of Artificial Intelligence decisions for legal contexts. The integration of Artificial Intelligence in forensic pathology represents a significant advancement, requiring careful balance between technological innovation and human expertise for optimal implementation.","author":[{"family":"Orsini","given":"Francesco"},{"family":"Cioffi","given":"Andrea"},{"family":"Cipolloni","given":"Luigi"},{"family":"Bibbò","given":"Roberta"},{"family":"Montana","given":"Angelo"},{"family":"Simone","given":"Stefania"},{"family":"Cecannecchia","given":"Camilla"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fmed.2025.1583743","URL":"https://doi.org/10.3389/fmed.2025.1583743","source":"openalex"},{"id":"oa:W4416716717","type":"article-journal","title":"Artificial intelligence for natural product drug discovery and development: current landscape, applications, and future directions","abstract":"Artificial intelligence accelerates natural product discovery in oncology, infection, inflammation, and neuroprotection by enabling activity prediction, mechanism inference, and prioritization. These approaches include tree ensembles, graph neural networks, and self-supervised molecular embeddings for mixtures, isolated metabolites, and peptide analogs, while network pharmacology models herb–ingredient–target–pathway graphs to propose synergistic effects. Operational multi-omics gates (transcriptomic signature reversal, proteome-scale target engagement, and untargeted metabolomics with feature-based molecular networking) move ranked candidates into reproducible validations. Population-level analyses map formulation-derived ingredient–target signatures to clinical outcomes, and large language models are beginning to standardize herbal prescriptions and their curation. Persistent barriers include mixture and batch variability, incomplete provenance, small and imbalanced datasets, domain shift, off-target liability, limited interpretability, and bias. Practical solutions include minimal information for AI on natural product metadata for provenance and safety, scaffold and time-split benchmarks with cross-lab replication, uncertainty and applicability-domain gating, mechanistic add-back experiments, constrained generative and semi-synthetic design, micro-physiological systems with digital twins, and provenance-aware pharmacovigilance under evolving FDA, EMA, and WHO expectations. This review integrates these advances, evidence gaps, and governance requirements into a roadmap for the mechanistically grounded, prospectively validated translation of AI in natural product research. • AI tools accelerate natural product–based drug discovery and development. • Machine learning and deep learning models predict anticancer, anti-inflammatory, and antimicrobial actions. • Several AI-predicted natural compounds were validated in vitro, confirming translational potential. • Key challenges include small datasets, data imbalance, and limited experimental validation.","author":[{"family":"Othman","given":"Zhinya"},{"family":"Ahmed","given":"Mohamed"},{"family":"Kasimieh","given":"Omar"},{"family":"Musa","given":"Shuaibu"},{"family":"Branda","given":"Francesco"},{"family":"Cue","given":"Edgar"},{"family":"Ocampo","given":"Justine"},{"family":"Luceroprisno","given":"Don"},{"family":"Vimolmangkang","given":"Sornkanok"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.ibmed.2025.100316","URL":"https://doi.org/10.1016/j.ibmed.2025.100316","source":"openalex"},{"id":"oa:W4409507896","type":"article-journal","title":"Artificial Intelligence-Powered Quality Assurance: Transforming Diagnostics, Surgery, and Patient Care—Innovations, Limitations, and Future Directions","abstract":"Artificial intelligence is rapidly transforming quality assurance in healthcare, driving advancements in diagnostics, surgery, and patient care. This review presents a comprehensive analysis of artificial intelligence integration-particularly convolutional and recurrent neural networks-across key clinical domains, significantly enhancing diagnostic accuracy, surgical performance, and pathology evaluation. Artificial intelligence-based approaches have demonstrated clear superiority over conventional methods: convolutional neural networks achieved 91.56% accuracy in scanner fault detection, surpassing manual inspections; endoscopic lesion detection sensitivity rose from 2.3% to 6.1% with artificial intelligence assistance; and gastric cancer invasion depth classification reached 89.16% accuracy, outperforming human endoscopists by 17.25%. In pathology, artificial intelligence achieved 93.2% accuracy in identifying out-of-focus regions and an F1 score of 0.94 in lymphocyte quantification, promoting faster and more reliable diagnostics. Similarly, artificial intelligence improved surgical workflow recognition with over 81% accuracy and exceeded 95% accuracy in skill assessment classification. Beyond traditional diagnostics and surgical support, AI-powered wearable sensors, drug delivery systems, and biointegrated devices are advancing personalized treatment by optimizing physiological monitoring, automating care protocols, and enhancing therapeutic precision. Despite these achievements, challenges remain in areas such as data standardization, ethical governance, and model generalizability. Overall, the findings underscore artificial intelligence's potential to outperform traditional techniques across multiple parameters, emphasizing the need for continued development, rigorous clinical validation, and interdisciplinary collaboration to fully realize its role in precision medicine and patient safety.","author":[{"family":"Shin","given":"Yoojin"},{"family":"Lee","given":"Mingyu"},{"family":"Lee","given":"YK"},{"family":"Kim","given":"Kyuri"},{"family":"Kim","given":"Tae"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/life15040654","URL":"https://doi.org/10.3390/life15040654","source":"openalex"},{"id":"oa:W4414664064","type":"article-journal","title":"Artificial Intelligence as an Organizing Capability Arising from Human‐Algorithm Relations","abstract":"Abstract In this article, we move beyond the prevailing view of artificial intelligence (AI) as an independent entity within organizations, which, we argue, risks obscuring potential explanations of the effects of AI on organizing. Drawing on posthumanism, we propose an ontological shift in conceptualizing AI. We theorize that, instead of residing within algorithmic actors, AI arises from the relations among human and algorithmic actors as an organizing capability. This capability is characterized by connectivity, codependence, and emergence as core properties, and contributes to organizational analysing, learning, and acting in pursuit of organizational goals. The shift from the entity view to the organizing capability view of AI has significant implications for understanding its organizational effects and opens new avenues for research in human‐algorithm collaboration, algorithmic management, and organizational intelligence, while counterbalancing tendencies to treat AI as autonomous agents.","author":[{"family":"Stelmaszak","given":"Marta"},{"family":"Joshi","given":"Mayur"},{"family":"Constantiou","given":"Ioanna"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/joms.70003","URL":"https://doi.org/10.1111/joms.70003","source":"openalex"},{"id":"oa:W4407558261","type":"article-journal","title":"Application of artificial intelligence in Alzheimer’s disease: a bibliometric analysis","abstract":"Background Understanding how artificial intelligence (AI) is employed to predict, diagnose, and perform relevant analyses in Alzheimer’s disease research is a rapidly evolving field. This study integrated and analyzed the relevant literature from the Science Citation Index (SCI) and Social Science Citation Index (SSCI) on the application of AI in Alzheimer’s disease (AD), covering publications from 2004 to 2023. Objective This study aims to identify the key research hotspots and trends of the application of AI in AD over the past 20 years through a bibliometric analysis. Methods Using the Web of Science Core Collection database, we conducted a comprehensive visual analysis of literature on AI and AD published between January 1, 2004, and December 31, 2023. The study utilized Excel, Scimago Graphica, VOSviewer, and CiteSpace software to visualize trends in annual publications and the distribution of research by countries, institutions, journals, references, authors, and keywords related to this topic. Results A total of 2,316 papers were obtained through the research process, with a significant increase in publications observed since 2018, signaling notable growth in this field. The United States, China, and the United Kingdom made notable contributions to this research area. The University of London led in institutional productivity with 80 publications, followed by the University of California System with 74 publications. Regarding total publications, the Journal of Alzheimer’s Disease was the most prolific while Neuroimage ranked as the most cited journal. Shen Dinggang was the top author in both total publications and average citations. Analysis of reference and keyword highlighted research hotspots, including the identification of various stages of AD, early diagnostic screening, risk prediction, and prediction of disease progression. The “task analysis” keyword emerged as a research frontier from 2021 to 2023. Conclusion Research on AI applications in AD holds significant potential for practical advancements, attracting increasing attention from scholars. Deep learning (DL) techniques have emerged as a key research focus for AD diagnosis. Future research will explore AI methods, particularly task analysis, emphasizing integrating multimodal data and utilizing deep neural networks. These approaches aim to identify emerging risk factors, such as environmental influences on AD onset, predict disease progression with high accuracy, and support the development of prevention strategies. Ultimately, AI-driven innovations will transform AD management from a progressive, incurable state to a more manageable and potentially reversible condition, thereby improving healthcare, rehabilitation, and long-term care solutions.","author":[{"family":"Song","given":"Sijia"},{"family":"Li","given":"Tong"},{"family":"Lin","given":"Wei"},{"family":"Liu","given":"Ran"},{"family":"Zhang","given":"Yujie"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fnins.2025.1511350","URL":"https://doi.org/10.3389/fnins.2025.1511350","source":"openalex"},{"id":"oa:W4414547178","type":"article-journal","title":"Harnessing Geospatial Artificial Intelligence (GeoAI) for Environmental Epidemiology: A Narrative Review","abstract":"PURPOSE OF REVIEW: Geospatial analysis is an essential tool for research on the role of environmental exposures and health, and critical for understanding impacts of environmental risk factors on diseases with long latency (e.g. cardiovascular disease, dementia, cancers) as well as upstream behaviors including sleep, physical activity, and cognition. There is emerging interest in leveraging machine learning and artificial intelligence (AI) for environmental epidemiology research. In this review, we provide an accessible overview of recent advances. RECENT FINDINGS: There have been two major recent shifts in geospatial data types and analytic methods. First, novel methods for statistical prediction, combining geospatial analysis with machine learning and artificial intelligence (GeoAI), allow for scalable geospatial exposure assessment within large population health databases (e.g. cohorts, administrative claims). Second, the widespread adoption of smartphones and wearables with global positioning systems and other sensors has allowed for passive data collection from people, and when combined with geographic information systems, enables exposure assessment at finer spatial scales and temporal resolution than ever before. Illustrative examples include refining models for predicting outdoor air pollution exposure, characterizing populations susceptible to water pollution, and use of deep learning to classify Street View image-derived measures of greenspace. While these tools and approaches may facilitate more rapid, higher quality objective exposure measures, they pose challenges with respect to participant privacy, representativeness of collected data, and curation of high quality validation sets for training of GeoAI algorithms. GeoAI approaches are beginning to be used for environmental exposure assessment and behavioral outcome ascertainment with higher spatial and temporal precision than before. Epidemiologists should continue to apply critical assessment of measurement accuracy and design validity when incorporating these new tools into their work.","author":[{"family":"Iyer","given":"Hari"},{"family":"Karasaki","given":"Seigi"},{"family":"Li","given":"Yi"},{"family":"Hswen","given":"Yulin"},{"family":"James","given":"Peter"},{"family":"Vopham","given":"Trang"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s40572-025-00497-4","URL":"https://doi.org/10.1007/s40572-025-00497-4","source":"openalex"},{"id":"oa:W4411453131","type":"article-journal","title":"Edge Intelligence: A Review of Deep Neural Network Inference in Resource-Limited Environments","abstract":"Deploying deep neural networks (DNNs) in resource-limited environments—such as smartwatches, IoT nodes, and intelligent sensors—poses significant challenges due to constraints in memory, computing power, and energy budgets. This paper presents a comprehensive review of recent advances in accelerating DNN inference on edge platforms, with a focus on model compression, compiler optimizations, and hardware–software co-design. We analyze the trade-offs between latency, energy, and accuracy across various techniques, highlighting practical deployment strategies on real-world devices. In particular, we categorize existing frameworks based on their architectural targets and adaptation mechanisms and discuss open challenges such as runtime adaptability and hardware-aware scheduling. This review aims to guide the development of efficient and scalable edge intelligence solutions.","author":[{"family":"Ngo","given":"Dat"},{"family":"Park","given":"Hyun"},{"family":"Kang","given":"Bongsoon"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/electronics14122495","URL":"https://doi.org/10.3390/electronics14122495","source":"openalex"},{"id":"oa:W4416204925","type":"article-journal","title":"What makes university students accept generative artificial intelligence? A moderated mediation model","abstract":"Artificial Intelligence (AI) technology developments are increasing the importance of accepting and utilizing generative AI. Higher Education is one of the most common areas where AI tools are used. University students use AI tools such as ChatGPT for various purposes (e.g., homework and projects). However, there is limited research on the factors affecting university students' acceptance of AI. AI-related technology and literacy skills are effective in promoting the acceptance of new technologies. Additionally, attitude towards AI and AI self-efficacy can promote the acceptance of AI. Within this context, this study tested a hypothetical model to examine the relationships among university students' attitude towards AI, AI literacy, AI self-efficacy, AI learning anxiety, and AI acceptance. Data were collected from 356 participants (265 females) with a mean of 22.71 (SD = ± 3.72). Mediation and moderation analyses were used to examine the role of AI literacy, AI self-efficacy, and AI learning anxiety in the relationship between attitude towards AI and AI acceptance among Turkish university students. The research results showed that the relationship between attitude toward AI and acceptance of AI can be explained by AI literacy and AI self-efficacy. Moreover, AI learning anxiety can moderate the predictive role of students' attitudes towards AI on AI acceptance. The study broadens and enhances the educational AI literature related to the factors that complicate and facilitate AI acceptance.","author":[{"family":"Türk","given":"Nuri"},{"family":"Batuk","given":"Barzan"},{"family":"Kaya","given":"Alican"},{"family":"Yildirim","given":"Oğuzhan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s40359-025-03559-2","URL":"https://doi.org/10.1186/s40359-025-03559-2","source":"openalex"},{"id":"oa:W4407295321","type":"article-journal","title":"Artificial Intelligence in Biomedical Engineering and Its Influence on Healthcare Structure: Current and Future Prospects","abstract":"Artificial intelligence (AI) is a growing area of computer science that combines technologies with data science to develop intelligent, highly computation-able systems. Its ability to automatically analyze and query huge sets of data has rendered it essential to many fields such as healthcare. This article introduces you to artificial intelligence, how it works, and what its central role in biomedical engineering is. It brings to light new developments in medical science, why it is being applied in biomedicine, key problems in computer vision and AI, medical applications, diagnostics, and live health monitoring. This paper starts with an introduction to artificial intelligence and its major subfields before moving into how AI is revolutionizing healthcare technology. There is a lot of emphasis on how it will transform biomedical engineering through the use of AI-based devices like biosensors. Not only can these machines detect abnormalities in a patient's physiology, but they also allow for chronic health tracking. Further, this review also provides an overview of the trends of AI-enabled healthcare technologies and concludes that the adoption of artificial intelligence in healthcare will be very high. The most promising are in diagnostics, with highly accurate, non-invasive diagnostics such as advanced imaging and vocal biomarker analyzers leading medicine into the future.","author":[{"family":"Tripathi","given":"Divya"},{"family":"Hajra","given":"Kasturee"},{"family":"Mulukutla","given":"Aditya"},{"family":"Shreshtha","given":"Romi"},{"family":"Maity","given":"Dipak"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bioengineering12020163","URL":"https://doi.org/10.3390/bioengineering12020163","source":"openalex"},{"id":"oa:W4407509037","type":"article-journal","title":"Artificial Intelligence in Colonoscopy: Where Are We Now in 2024?","abstract":"INTRODUCTION: Colonoscopy has a crucial role in reducing colorectal cancer incidence and mortality. Different artificial intelligence (AI) systems were developed to further improve its quality assurance (computer-aided quality improvement [CAQ]), lesion detection (computer-aided detection [CADe]), and lesion characterization (computer-aided characterization [CADx]). There were studies investigating the roles of these AI systems in different domains of standard colonoscopies. METHODS: In this state-of-the-art narrative review, we summarize the current evidence, discuss existing limitations, as well as explore the future directions of AI in colonoscopy. RESULTS: CAQ enhances colonoscopy quality through real-time feedback and quality monitoring systems, but the studies have inconsistent results due to small training datasets and varied methodologies. CADe increases adenoma detection rate and reduces adenoma missed rates, but there are concerns about false positives, unnecessary polypectomies, potential deskilling of endoscopists, and cost-effectiveness. CADx systems have mixed results and accuracies in differentiating polyp types, and its use is further hindered by inadequate representation of sessile serrated lesions and a lack of rigorous trials comparing it with standard colonoscopy. CONCLUSION: Despite the emerging evidence of AI-assisted colonoscopy, its potential drawbacks and limitations may hinder the further implementation in real-world clinical practice. Long-term data on clinical efficacy, cost-effectiveness, liability, and data sharing are the key areas to be addressed.","author":[{"family":"Lai","given":"Wan"},{"family":"Lin","given":"Kenneth"},{"family":"Ling","given":"L"},{"family":"Li","given":"James"},{"family":"Lau","given":"Louis"},{"family":"Chiu","given":"Philip"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1159/000544030","URL":"https://doi.org/10.1159/000544030","source":"openalex"},{"id":"oa:W4406870591","type":"article-journal","title":"Artificial Intelligence–Based Psychotherapeutic Intervention on Psychological Outcomes: A Meta‐Analysis and Meta‐Regression","abstract":"Background: Artificial intelligence (AI)–based psychotherapeutic interventions may bring a new and viable approach to expanding psychiatric care. However, evidence of their effectiveness remains scarce. We evaluated the efficacy of AI‐based psychotherapeutic interventions on depressive, anxiety, and stress symptoms at postintervention and follow‐up assessments. Methods: A three‐step comprehensive search via nine electronic databases (PubMed, Embase, CINAHL, Cochrane Library, Scopus, IEEE Xplore, Web of Science, PsycINFO, and ProQuest Dissertations and Theses) was performed. Results: Thirty randomized controlled trials (RCTs) in 31 publications involving 6100 participants from nine countries were included. The majority (79.1%) of trials with intention‐to‐treat analysis but less than half (48.6%) of trials with perprotocol analysis were graded as low risk. Meta‐analyses showed that interventions significantly reduced depressive symptoms at the postintervention assessment ( t = −4.40, p = 0.001) with medium effect size ( g = −0.54, 95% CI: −0.79 to −0.29) and at 6–12 months of assessment ( t = −3.14, p < 0.016) with small effect size ( g = −0.23, 95% CI: −0.40 to −0.06) in comparison with comparators. Our subgroup analyses revealed that the depressed participants had a significantly larger effect size in reducing depressive symptoms than participants with stress and other conditions. At postintervention and follow‐up assessments, we discovered that AI‐based psychotherapeutic interventions did not significantly alter anxiety, stress, and the total scores of depressive, anxiety, and stress symptoms in comparison to comparators. The random‐effects univariate meta‐regression did not identify any significant covariates for depressive and anxiety symptoms at postintervention. The certainty of evidence ranged between moderate and very low. Conclusions: AI‐based psychotherapeutic interventions can be used in addition to usual treatments for reducing depressive symptoms. Well‐designed RCTs with long‐term follow‐up data are warranted. Trial Registration: CRD42022330228","author":[{"family":"Lau","given":"Ying"},{"family":"Ang","given":"Wei"},{"family":"Ang","given":"Wen"},{"family":"Pang","given":"Patrick"},{"family":"Wong","given":"Sai"},{"family":"Chan","given":"Kin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1155/da/8930012","URL":"https://doi.org/10.1155/da/8930012","source":"openalex"},{"id":"oa:W4412630965","type":"article-journal","title":"Artificial Intelligence in Cosmetic Formulation: Predictive Modeling for Safety, Tolerability, and Regulatory Perspectives","abstract":"Artificial intelligence (AI) and machine learning (ML) are increasingly transforming the landscape of cosmetic formulation, enabling the development of safer, more effective, and personalized products. This article explores how AI-driven predictive modeling is applied across various components of cosmetic products, including surfactants, polymers, fragrances, preservatives, antioxidants, and prebiotics. These technologies are employed to forecast critical properties such as texture, stability, and shelf-life, optimizing both product performance and user experience. The integration of computational toxicology and ML algorithms also allows for early prediction of skin sensitization risks, including the likelihood of adverse events such as allergic contact dermatitis. Furthermore, AI models can support efficacy assessment, bridging formulation science with dermatological outcomes. The article also addresses the ethical, regulatory, and safety challenges associated with AI in cosmetic science, underlining the need for transparency, accountability, and harmonized standards. The potential of AI to reshape dermocosmetic innovation is vast, but it must be approached with robust oversight and a commitment to user well-being.","author":[{"family":"Guardo","given":"Antonio"},{"family":"Trovato","given":"Federica"},{"family":"Cantisani","given":"Carmen"},{"family":"Dattola","given":"Annunziata"},{"family":"Nisticò","given":"Steven"},{"family":"Pellacani","given":"Giovanni"},{"family":"Paganelli","given":"Alessia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/cosmetics12040157","URL":"https://doi.org/10.3390/cosmetics12040157","source":"openalex"},{"id":"oa:W4413272174","type":"article-journal","title":"Artificial Intelligence in Primary Care: Support or Additional Burden on Physicians’ Healthcare Work?—A Qualitative Study","abstract":"Background: Artificial intelligence (AI) is being increasingly promoted as a means to enhance diagnostic accuracy, to streamline workflows, and to improve overall care quality in primary care. However, empirical evidence on how primary care physicians (PCPs) perceive, engage with, and emotionally respond to AI technologies in everyday clinical settings remains limited. Concerns persist regarding AI’s usability, transparency, and potential impact on professional identity, workload, and the physician–patient relationship. Methods: This qualitative study investigated the lived experiences and perceptions of 28 PCPs practicing in diverse outpatient settings across Germany. Participants were purposively sampled to ensure variation in age, practice characteristics, and digital proficiency. Data were collected through in-depth, semi-structured interviews, which were audio-recorded, transcribed verbatim, and subjected to rigorous thematic analysis employing Mayring’s qualitative content analysis framework. Results: Participants demonstrated a fundamentally ambivalent stance toward AI integration in primary care. Perceived advantages included enhanced diagnostic support, relief from administrative burdens, and facilitation of preventive care. Conversely, physicians reported concerns about workflow disruption due to excessive system prompts, lack of algorithmic transparency, increased cognitive and emotional strain, and perceived threats to clinical autonomy and accountability. The implications for the physician–patient relationship were seen as double-edged: while some believed AI could foster trust through transparent use, others feared depersonalization of care. Crucial prerequisites for successful implementation included transparent and explainable systems, structured training opportunities, clinician involvement in design processes, and seamless integration into clinical routines. Conclusions: Primary care physicians’ engagement with AI is marked by cautious optimism, shaped by both perceived utility and significant concerns. Effective and ethically sound implementation requires co-design approaches that embed clinical expertise, ensure algorithmic transparency, and align AI applications with the realities of primary care workflows. Moreover, foundational AI literacy should be incorporated into undergraduate health professional curricula to equip future clinicians with the competencies necessary for responsible and confident use. These strategies are essential to safeguard professional integrity, support clinician well-being, and maintain the humanistic core of primary care.","author":[{"family":"Mache","given":"Stefanie"},{"family":"Bernburg","given":"Monika"},{"family":"Würtenberger","given":"Annika"},{"family":"Groneberg","given":"David"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/clinpract15080138","URL":"https://doi.org/10.3390/clinpract15080138","source":"openalex"},{"id":"oa:W4407852883","type":"article-journal","title":"Evaluating the evidence-based potential of six large language models in paediatric dentistry: a comparative study on generative artificial intelligence","abstract":"PURPOSE: The use of large language models (LLMs) in generative artificial intelligence (AI) is rapidly increasing in dentistry. However, their reliability is yet to be fully founded. This study aims to evaluate the diagnostic accuracy, clinical applicability, and patient education potential of LLMs in paediatric dentistry, by evaluating the responses of six LLMs: Google AI's Gemini and Gemini Advanced, OpenAI's ChatGPT-3.5, -4o and -4, and Microsoft's Copilot. METHODS: Ten open-type clinical questions, relevant to paediatric dentistry were posed to the LLMs. The responses were graded by two independent evaluators from 0 to 10 using a detailed rubric. After 4 weeks, answers were reevaluated to assess intra-evaluator reliability. Statistical comparisons used Friedman's and Wilcoxon's and Kruskal-Wallis tests to assess the model that provided the most comprehensive, accurate, explicit and relevant answers. RESULTS: Variations of results were noted. Chat GPT 4 answers were scored as the best (average score 8.08), followed by the answers of Gemini Advanced (8.06), ChatGPT 4o (8.01), ChatGPT 3.5 (7.61), Gemini (7,32) and Copilot (5.41). Statistical analysis revealed that Chat GPT 4 outperformed all other LLMs, and the difference was statistically significant. Despite variations and different responses to the same queries, remarkable similarities were observed. Except for Copilot, all chatbots managed to achieve a score level above 6.5 on all queries. CONCLUSION: This study demonstrates the potential use of language models (LLMs) in supporting evidence-based paediatric dentistry. Nevertheless, they cannot be regarded as completely trustworthy. Dental professionals should critically use AI models as supportive tools and not as a substitute of overall scientific knowledge and critical thinking.","author":[{"family":"Dermata","given":"Anastasia"},{"family":"Arhakis","given":"Aristidis"},{"family":"Makrygiannakis","given":"Miltiadis"},{"family":"Giannakopoulos","given":"Kostis"},{"family":"Kaklamanos","given":"Eleftherios"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s40368-025-01012-x","URL":"https://doi.org/10.1007/s40368-025-01012-x","source":"openalex"},{"id":"oa:W4410633341","type":"article-journal","title":"Generative artificial intelligence-supported programming education: Effects on learning performance, self-efficacy and processes","abstract":"Recent advancements in generative artificial intelligence (GenAI) have drawn significant attention from educators and researchers. However, its effects on learners’ programming performance, self-efficacy and learning processes remain inconclusive, while the mechanisms underlying its efficiency-enhancing potential are underexplored. This study addresses these gaps through a quasi-experiment comparing an experimental group using GenAI for self-directed programming learning with a control group relying on alternative tools. Additionally, the experimental group was divided into high- and low-performance subgroups to examine the relationship between learning behaviour patterns and academic outcomes using process mining techniques. The findings reveal that (a) GenAI demonstrates strong potential to enhance learning outcomes and self-efficacy but negatively affects long-term knowledge transfer; (b) excessive reliance on GenAI and cognitive outsourcing impede effective knowledge acquisition; (c) high-performing learners exhibit greater epistemic agency, actively critiquing and engaging with AI-generated content to construct knowledge proactively. This study underscores the risks of over-reliance on GenAI and the detrimental effects of cognitive offloading, highlighting the critical role of cognitive engagement and epistemic agency in fostering hybrid intelligence. It provides empirical and theoretical insights to inform the design of GenAI-supported programming education strategies and interventions. Implications for practice or policy: Instructors can enhance programming self-efficacy by integrating GenAI tools like ChatGPT into self-learning activities, particularly for reinforcing academic performance. Course leaders should emphasise GenAI use in programming courses to support student engagement, though they should also prepare students for problem-solving without external resources. Educational institutions may consider developing guidelines for balanced GenAI usage to maximise learning benefits while addressing potential limitations in problem-solving skills.","author":[{"family":"Li","given":"Siran"},{"family":"Liu","given":"Jiangyue"},{"family":"Dong","given":"Qianyan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.14742/ajet.9932","URL":"https://doi.org/10.14742/ajet.9932","source":"openalex"},{"id":"oa:W4413025933","type":"article-journal","title":"From Perception to Practice: Artificial Intelligence as a Pathway to Enhancing Digital Literacy in Higher Education Teaching","abstract":"In the context of increasing Artificial Intelligence integration in higher education, understanding the factors influencing university teachers’ adoption of AI tools is critical for effective implementation. This study adopts a perception–intention–behavior framework to explores the roles of perceived usefulness, perceived ease of use, perceived trust, perceived substitution crisis, and perceived risk in shaping teachers’ behavioral intention and actual usage of AI tools. It also investigates the moderating effects of peer influence and organizational support on these relationships. Using a comprehensive survey instrument, data was collected from 487 university teachers across four major regions in China. The results reveal that perceived usefulness and perceived ease of use are strong predictors of behavioral intention, with perceived ease of use also significantly influencing perceived usefulness. Perceived trust serves as a key mediator, enhancing the relationship between perceived usefulness, perceived ease of use, and behavioral intention. While perceived substitution crisis negatively influenced perceived trust, it showed no significant direct effect on behavioral intention, suggesting a complex relationship between job displacement concerns and AI adoption. In contrast, perceived risk was found to negatively impact behavioral intention, though it was mitigated by perceived ease of use. Peer influence significantly moderated the relationship between perceived trust and behavioral intention, highlighting the importance of peer influence in AI adoption, while organizational support amplified the effect of perceived ease of use on behavioral intention. These findings inform practical strategies such as co-developing user-centered AI tools, enhancing institutional trust through transparent governance, leveraging peer support, providing structured training and technical assistance, and advancing policy-level initiatives to guide digital transformation in universities.","author":[{"family":"Zuo","given":"Zhili"},{"family":"Luo","given":"Yanqi"},{"family":"Yan","given":"Shiyu"},{"family":"Jiang","given":"Lisheng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/systems13080664","URL":"https://doi.org/10.3390/systems13080664","source":"openalex"},{"id":"oa:W4411927032","type":"article-journal","title":"Navigating the Complexity of Generative Artificial Intelligence in Higher Education: A Systematic Literature Review","abstract":"Technological innovation has transformed educational settings, enabling artificial intelligence (AI)-driven teaching and learning processes. While AI is still in its embryonic stage in education, generative artificial intelligence has evolved rapidly, significantly shifting the teaching and learning context. With no clarity about the impacts of generative artificial intelligence on education, there is a need to synthesise research findings to demystify generative artificial intelligence and address concerns regarding its application in the teaching and learning process. This paper systematically synthesises studies on generative artificial intelligence in teaching and learning to understand key arguments and stakeholders’ perceptions of generative artificial intelligence in teaching and learning. The systematic review reveals five main domains of research within the field: (i) current awareness (understanding) of generative artificial intelligence, (ii) stakeholder perceptions, (iii) mechanisms for adopting generative artificial intelligence, (iv) issues and challenges of implementing generative artificial intelligence, and (v) contributions of generative artificial intelligence to student performance. This review examines the practical and policy implications of generative artificial intelligence, providing recommendations to address the concerns and challenges associated with generative artificial intelligence-driven teaching and learning processes.","author":[{"family":"Amofa","given":"Birago"},{"family":"Kamudyariwa","given":"Xebiso"},{"family":"Fernandes","given":"Fatima"},{"family":"Osobajo","given":"Oluyomi"},{"family":"Jeremiah","given":"Faith"},{"family":"Oke","given":"Adekunle"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/educsci15070826","URL":"https://doi.org/10.3390/educsci15070826","source":"openalex"},{"id":"oa:W4406490791","type":"article-journal","title":"Navigating AI Convergence in Human–Artificial Intelligence Teams: A Signaling Theory Approach","abstract":"ABSTRACT Teams that combine human intelligence with artificial intelligence (AI) have become indispensable for solving complex tasks in various decision‐making contexts in modern organizations. However, the factors that contribute to AI convergence, where human team members align their decisions with those of their AI counterparts, still remain unclear. This study integrates signaling theory with self‐determination theory to investigate how specific signals—such as signal fit, optional AI advice, and signal set congruence—affect employees' AI convergence in human–AI teams. Based on four experimental studies conducted in facial recognition and hiring contexts with approximately 1100 participants, the findings highlight the significant positive impact of congruent signals from both human and AI team members on AI convergence. Moreover, providing an option for employees to solicit AI advice also enhances AI convergence; when AI signals are chosen by employees rather than forced upon them, participants are more likely to accept AI advice. This research advances knowledge on human–AI teaming by (1) expanding signaling theory into the human–AI team context; (2) developing a deeper understanding of AI convergence and its drivers in human–AI teams; (3) providing actionable insights for designing teams and tasks to optimize decision‐making in high‐stakes, uncertain environments; and (4) introducing facial recognition as an innovative context for human–AI teaming.","author":[{"family":"Smith","given":"Andria"},{"family":"Wagoner","given":"Hunter"},{"family":"Keplinger","given":"Ksenia"},{"family":"Celebi","given":"Can"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/job.2856","URL":"https://doi.org/10.1002/job.2856","source":"openalex"},{"id":"oa:W4411620636","type":"article-journal","title":"Generative Artificial Intelligence in Primary Care: Qualitative Study of UK General Practitioners’ Views","abstract":"Background: The potential for generative artificial intelligence (GenAI) to assist with clinical tasks is the subject of ongoing debate within biomedical informatics and related fields. Objective: This study aimed to explore general practitioners' (GPs') opinions about GenAI on primary care. Methods: In January 2025, we conducted a web-based survey of 1005 UK GPs' experiences and opinions of GenAI in clinical practice. This study involved a qualitative inductive descriptive analysis of a written response (\"comments\") to an open-ended question in the survey. After analysis, the interpretation of themes was also informed by the technology acceptance model. Results: Out of 1005 respondents, 611 GPs (61%) provided written comments in response to the free text question, totaling 7990 words. Comments were classified into 3 major themes and 8 subthemes in relation to GenAI in clinical practice. The major themes were (1) unfamiliarity, (2) ambivalence and anxiety, and (3) role in clinical tasks. \"Unfamiliarity\" encompassed a lack of experience and knowledge, and the need for training on GenAI. \"Ambivalence and anxiety\" included mixed expectations among GPs in relation to these tools, beliefs about diminished human connection, and skepticism about AI accountability. Finally, commenting on the role of GenAI in clinical tasks, GPs believed it would help with documentation. However, respondents questioned AI's clinical judgment and raised concerns about operational uncertainty concerning these tools. Female GPs were more likely to leave comments than male GPs, with 53% (324/611) of female GPs providing feedback compared to 41.1% (162/394) who did not. Chi-square tests confirmed this difference ((χ²₂= 14.6, P=.001). In addition, doctors who left comments were significantly more likely to have used GenAI in clinical practice compared with those who did not. Among all respondents, 71.7% (438/611) had not used GenAI. However, noncommenters were even less likely to have used it, with 80.7% (318/394) reporting no use. A chi-square test confirmed this difference (χ²₁=10.0, P=.002). Conclusions: This study provides timely insights into UK GPs' perspectives on the role, impact, and limitations of GenAI in primary care. However, the study has limitations. The qualitative data analyzed originates from a self-selected subset of respondents who chose to provide free-text comments, and these participants were more likely to have used GenAI tools in clinical practice. However, the substantial number of comments offers valuable insights into the diverse views held by GPs regarding GenAI. Furthermore, the majority of our respondents reported limited experience and training with these tools; however, many GPs perceived potential benefits of GenAI and ambient AI for documentation. Notably, 2 years after the widespread introduction of GenAI, GPs' persistent lack of understanding and training remains a critical concern. More extensive qualitative work would provide a more in-depth understanding of GPs' views.","author":[{"family":"Blease","given":"Charlotte"},{"family":"Kharko","given":"Anna"},{"family":"Sanchez","given":"Carolina"},{"family":"Alderman","given":"Joseph"},{"family":"Kumpunen","given":"Stephanie"},{"family":"Sundemo","given":"David"},{"family":"Torous","given":"John"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/74428","URL":"https://doi.org/10.2196/74428","source":"openalex"},{"id":"oa:W4410236541","type":"article-journal","title":"Transformative impact of explainable artificial intelligence: bridging complexity and trust","abstract":"Artificial Intelligence and Deep Learning have gained widespread popularity in all sectors and industries from healthcare to finance and industrial management. Explainable Artificial Intelligence (XAI) is urgent need to bridge the gap between the needs of society interpretability, and trust while maximizing AI benefits. This review XAI methodologies is presented as a comprehensive analysis of three different types model including model-specific, model-agnostic, and hybrid, along with their applications. The review discussed sectors of healthcare, finance, and industrial management etc. where XAI can be utilized for better results and gain trust. The generic and prominent key challenges in terms of trade-offs between accuracy and interpretability, the existing scalability issues, and ethical considerations were focused. The paper also discussed future directions, such as domain-specific frameworks interdisciplinary collaborations and standardized evaluation metrics, to be proposed for advancing XAI research and applications. The review highlighted the potential of XAI for upbringing a society equipped with modern AI with precise results, high responsibility and more transparency.","author":[{"family":"Paliwal","given":"Girish"},{"family":"Kumar","given":"Ashish"},{"family":"Prasad","given":"S"},{"family":"Bhargava","given":"Deepshikha"},{"family":"Shrimal","given":"Vijay"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s44163-025-00281-1","URL":"https://doi.org/10.1007/s44163-025-00281-1","source":"openalex"},{"id":"oa:W4406014077","type":"article-journal","title":"Artificial Intelligence-Guided Inverse Design of Deployable Thermo-Metamaterial Implants","abstract":"Current limitations in implant design often lead to trade-offs between minimally invasive surgery and achieving the desired post-implantation functionality. Here, we present an artificial intelligence inverse design paradigm for creating deployable implants as planar and tubular thermal mechanical metamaterials (thermo-metamaterials). These thermo-metamaterial implants exhibit tunable mechanical properties and volume change in response to temperature changes, enabling minimally invasive and personalized surgery. We begin by generating a large database of corrugated thermo-metamaterials with various cell structures and bending stiffnesses. An artificial intelligence inverse design model is subsequently developed by integrating an evolutionary algorithm with a neural network. This model allows for the automatic determination of the optimal microstructure for thermo-metamaterials with desired performance,i.e., target bending stiffness. We validate this approach by designing patient-specific spinal fusion implants and tracheal stents. The results demonstrate that the deployable thermo-metamaterial implants can achieve over a 200% increase in volume or cross-sectional area in their fully deployed states. Finally, we propose a broader vision for a clinically informed artificial intelligence design process that prioritizes biocompatibility, feasibility, and precision simultaneously for the development of high-performing and clinically viable implants. The feasibility of this proposed vision is demonstrated using a fuzzy analytic hierarchy process to customize thermo-metamaterial implants based on clinically relevant factors.","author":[{"family":"Jiao","given":"Pengcheng"},{"family":"Zhang","given":"Chenjie"},{"family":"Meng","given":"Wenxuan"},{"family":"Wang","given":"Jiajun"},{"family":"Jang","given":"Daeik"},{"family":"Wu","given":"Zhangming"},{"family":"Agarwal","given":"Nitin"},{"family":"Alavi","given":"Amir"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1021/acsami.4c17625","URL":"https://doi.org/10.1021/acsami.4c17625","source":"openalex"},{"id":"oa:W4409961726","type":"article-journal","title":"Adoption challenges to artificial intelligence literacy in public healthcare: an evidence based study in Saudi Arabia","abstract":"In recent years, Artificial Intelligence (AI) is transforming healthcare systems globally and improved the efficiency of its delivery. Countries like Saudi Arabia are facing unique adoption challenges in their public healthcare, these challenges are specific to AI literacy, understanding and effective usage of AI technologies. In addition, cultural, regulatory and operational barriers increase the complication of integrating AI literacy into public healthcare operations. In spite of its critical contribution in enabling sustainable healthcare development, limited studies have addressed these adoption challenges. Our study explores the AI literacy adoption barriers in context to Saudi Arabian public healthcare sector, focusing on its relevance for advancing healthcare operations and achieving Sustainable Development Goals (SDGs). The research aims to identifying and addressing the adoption challenges of Artificial Intelligence literacy within the public healthcare in Saudi Arabia. The research aims to enhance the understanding of AI literacy, its necessity for enhancing healthcare operations, and the specific hurdles that impede its successful AI adoption in Saudi Arabia's public healthcare ecosystem. The research employs a qualitative analysis using the T-O-E framework to explore the adoption challenges of AI literacy. Additionally, the Best-Worse Method (BWM) is applied to evaluate the adoption challenges to AI literacy adoption across various operational levels within Saudi Arabia's public healthcare supply chain. The study uncovers substantial adoption challenges at operational, tactical, and strategic level, including institutional readiness, data privacy, and compliance with regulatory frameworks. These challenges complicate the adoption of AI literacy in the Saudi public healthcare supply chains. The research offers critical insights into the various issues affecting the promotion of AI literacy in Saudi Arabia's public healthcare sector. This evidence-based study provides essential commendations for healthcare professionals and policymakers to effectively address the identified challenges, nurturing an environment beneficial to the integration of AI literacy and advancing the goals of sustainable healthcare development.","author":[{"family":"Kumar","given":"Rakesh"},{"family":"Singh","given":"Ajay"},{"family":"Kassar","given":"Ahmed"},{"family":"Humaida","given":"Mohammed"},{"family":"Joshi","given":"Sudhanshu"},{"family":"Sharma","given":"Manu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fpubh.2025.1558772","URL":"https://doi.org/10.3389/fpubh.2025.1558772","source":"openalex"},{"id":"oa:W4407925494","type":"article-journal","title":"Artificial Intelligence-Enabled 4D Printed Hydrogel Wearables: Temperature and Ultraviolet Monitoring","abstract":"Integrating smart wearable technology into daily life is becoming increasingly important for monitoring environmental factors such as temperature and ultraviolet (UV) radiation, both of which can impact health.Prolonged UV exposure is linked to skin cancer and eye damage, while major temperature changes can affect comfort and well-being.This study focuses on the manufacturing process of a novel smart wearable made from a hydrogel composite to monitor these environmental changes.The composite, consisting of hydroxyethyl methacrylate (HEMA) and polyethylene glycol diacrylate (PEGDA) with triphenylphosphine oxide (TPO) as the photoinitiator, is manufactured using digital light processing (DLP) 3-dimensional (3D)printing.The HEMA to PEGDA ratio was optimized to achieve plastic-like durability.Tensile tests showed that both the hydrogel composite and nylon samples exhibited almost identical stress-strain behavior, with a tensile strength of approximately 40 MPa.Thermochromic and photochromic powders were added to provide dynamic color responses to temperature and UV light.The thermal color transitions are linked to an artificial intelligence model specifically trained to decode these hues into temperature measurements.Additionally, the photochromic aspect of the wearables acts as a visual alarm against UV exposure and thus advises on the potential requirements of protective measures.","author":[{"family":"El-Nemr","given":"Mohamed"},{"family":"Halawani","given":"Yasmin"},{"family":"Elkaffas","given":"Ragi"},{"family":"Elkaffas","given":"Rami"},{"family":"Samad","given":"Yarjan"},{"family":"Hisham","given":"Muhammed"},{"family":"Mohammad","given":"Baker"},{"family":"Butt","given":"Haider"}],"issued":{"date-parts":[[2025]]},"DOI":"10.30919/mm1428","URL":"https://doi.org/10.30919/mm1428","source":"openalex"},{"id":"oa:W4406015538","type":"article-journal","title":"Differential diagnosis of iron deficiency anemia from aplastic anemia using machine learning and explainable Artificial Intelligence utilizing blood attributes","abstract":"As per world health organization, Anemia is a most prevalent blood disorder all over the world. Reduced number of Red Blood Cells or decrease in the number of healthy red blood cells is considered as Anemia. This condition also leads to the decrease in the oxygen carrying capacity of the blood. The main goal of this research is to develop a dependable method for diagnosing Aplastic Anemia and Iron Deficiency Anemia by examining the blood test attributes. As of today, there are no studies which use Interpretable Artificial Intelligence to perform the above differential diagnosis. The dataset used in this study is collected from Kasturba Medical College, Manipal. The dataset consisted of various blood test attributes such as Red Blood cell count, Hemoglobin level, Mean Corpuscular Volume, etc. One of the trending topics in Machine Learning is Explainable Artificial Intelligence. They are known to demystify the machine learning outputs to all its stakeholders. Hence, Five XAI tools including SHAP, LIME, Eli5, Qlattice and Anchor are used to understand the model's predictions. The importance characteristics according to XAI models are PLT, PCT, MCV, PDW, HGB, ABS LYMP, WBC, MCH, and MCHC. are employed to train and test the data. The goal of using data analytic techniques is to give medical professionals a useful tool that improves decision-making, enhances resource management, and eventually raises the standard of patient care. By considering the unique qualities of each patient, medical professionals who must rely on AI-assisted diagnosis and treatment suggestions, XAI offers arguments to strengthen their faith in the model outcomes.","author":[{"family":"Darshan","given":"BS"},{"family":"Sampathila","given":"Niranjana"},{"family":"Bairy","given":"GM"},{"family":"Prabhu","given":"Srikanth"},{"family":"Belurkar","given":"Sushma"},{"family":"Chadaga","given":"Krishnaraj"},{"family":"Nandish","given":"S"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-024-84120-w","URL":"https://doi.org/10.1038/s41598-024-84120-w","source":"openalex"},{"id":"oa:W4409418757","type":"article-journal","title":"Ethical implications related to processing of personal data and artificial intelligence in humanitarian crises: a scoping review","abstract":"BACKGROUND: Humanitarian organizations are rapidly expanding their use of data in the pursuit of operational gains in effectiveness and efficiency. Ethical risks, particularly from artificial intelligence (AI) data processing, are increasingly recognized yet inadequately addressed by current humanitarian data protection guidelines. This study reports on a scoping review that maps the range of ethical issues that have been raised in the academic literature regarding data processing of people affected by humanitarian crises. METHODS: We systematically searched databases to identify peer-reviewed studies published since 2010. Data and findings were standardized, grouping ethical issues into the value categories of autonomy, beneficence, non-maleficence, and justice. The study protocol followed Arksey and O'Malley's approach and PRISMA reporting guidelines. RESULTS: We identified 16,200 unique records and retained 218 relevant studies. Nearly one in three (n = 66) discussed technologies related to AI. Seventeen studies included an author from a lower-middle income country while four included an author from a low-income country. We identified 22 ethical issues which were then grouped along the four ethical value categories of autonomy, beneficence, non-maleficence, and justice. Slightly over half of included studies (n = 113) identified ethical issues based on real-world examples. The most-cited ethical issue (n = 134) was a concern for privacy in cases where personal or sensitive data might be inadvertently shared with third parties. Aside from AI, the technologies most frequently discussed in these studies included social media, crowdsourcing, and mapping tools. CONCLUSIONS: Studies highlight significant concerns that data processing in humanitarian contexts can cause additional harm, may not provide direct benefits, may limit affected populations' autonomy, and can lead to the unfair distribution of scarce resources. The increase in AI tool deployment for humanitarian assistance amplifies these concerns. Urgent development of specific, comprehensive guidelines, training, and auditing methods is required to address these ethical challenges. Moreover, empirical research from low and middle-income countries, disproportionally affected by humanitarian crises, is vital to ensure inclusive and diverse perspectives. This research should focus on the ethical implications of both emerging AI systems, as well as established humanitarian data management practices. TRIAL REGISTRATION: Not applicable.","author":[{"family":"Kreutzer","given":"Tino"},{"family":"Orbinski","given":"James"},{"family":"Appel","given":"Lora"},{"family":"An","given":"Aijun"},{"family":"Marston","given":"Jerome"},{"family":"Boone","given":"Ella"},{"family":"Vinck","given":"Patrick"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12910-025-01189-2","URL":"https://doi.org/10.1186/s12910-025-01189-2","source":"openalex"},{"id":"oa:W4412707688","type":"article-journal","title":"Clinical prediction models using artificial intelligence approaches in dementia","abstract":"BACKGROUND: While nearly half of all dementia cases are potentially preventable, early detection and targeted interventions are critical. Artificial intelligence (AI)-enhanced clinical prediction models offer promising tools to improve diagnostic and prognostic accuracy by leveraging machine learning (ML) to integrate diverse data sources. This systematic review evaluates the development, performance, and clinical applicability of AI-based prediction models in dementia. METHODS: Searches of PubMed, Embase, and Web of Science identified peer-reviewed studies up to October 2024, focusing on AI-based models predicting dementia onset. Included studies were assessed for model accuracy, bias, and generalizability using the PROBAST tool. Data extraction adhered to the TRIPOD and CHARMS frameworks, capturing study design, participant demographics, predictor variables, and performance metrics. RESULTS: Among 2699 articles initially screened, 21 studies were included, encompassing over 1 million participants. AI models, extremely heterogenous for their nature, demonstrated good predictive accuracy, with a mean area under the curve of 0.845. While internal validation was conducted in all studies, external validation was limited. Models incorporating ML methods like random forests and support vector machines outperformed traditional approaches. The most used parameters were clinical and cognitive data, whilst data about biomarkers were the less used. Risk of bias was generally low, though calibration and generalizability remained challenges. CONCLUSIONS: AI-based prediction models show strong potential for early dementia detection and personalized care. However, their integration into clinical practice requires addressing issues of external validation, data representativeness, and model interpretability. Further research should focus on robust validation and ethical implementation to optimize their utility in dementia care.","author":[{"family":"Veronese","given":"Nicola"},{"family":"Bolzetta","given":"Francesco"},{"family":"Gallo","given":"Livia"},{"family":"Durante","given":"Giorgia"},{"family":"Vernuccio","given":"Laura"},{"family":"Saccaro","given":"Carlo"},{"family":"Gambino","given":"Caterina"},{"family":"Custodero","given":"Carlo"},{"family":"Portincasa","given":"Piero"},{"family":"Morotti","given":"Andrea"},{"family":"Galli","given":"Alice"},{"family":"Trasciatti","given":"Chiara"},{"family":"Padovani","given":"Alessandro"},{"family":"Pilotto","given":"Andrea"},{"family":"Barbagallo","given":"Mario"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s40520-025-03112-6","URL":"https://doi.org/10.1007/s40520-025-03112-6","source":"openalex"},{"id":"oa:W4409457779","type":"article-journal","title":"University Students’ Usage of Generative Artificial Intelligence for Sustainability: A Cross-Sectional Survey from China","abstract":"The rapid development of generative artificial intelligence (GenAI) technology has triggered extensive discussions about its potential applications in sustainable higher education. Based on the technology acceptance model (TAM) and task–technology fit (TTF) theory, this research aimed to investigate the current situations and challenges of Chinese university students using GenAI in four typical task scenarios. This was performed using a cross-sectional research design. The data were collected via questionnaire, with 486 undergraduates from a Chinese university participating. The data analysis methods include descriptive statistics, inferential statistics, and content analysis. The results show that more than 70% of university students actively use GenAI, but nearly half of them are not very proficient in its use. Doubao and ERNIE Bot are the GenAI tools they prefer most. The primary functions they use are text production and information retrieval. They mainly learn the relevant knowledge and skills through self-media and knowledge-sharing platforms. Among the four typical task scenarios, GenAI is widely used in course learning and research activities, while its application in daily life and job search is relatively limited. The analysis of demographic variables shows that grade and major have a significant impact on university students’ use of GenAI. In addition, university students suggest that universities should offer relevant courses or lectures and provide comprehensive technical support to improve the popularity and operability of GenAI. This study provides suggestions for universities, education administration departments, and technology development departments to improve GenAI services. It will help universities optimize the allocation of educational resources and promote educational equity for sustainability.","author":[{"family":"Xiao","given":"Lin"},{"family":"Pyng","given":"How"},{"family":"Ayub","given":"Ahmad"},{"family":"Zhu","given":"Zhihui"},{"family":"Gao","given":"Jianping"},{"family":"Qing","given":"Zehu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/su17083541","URL":"https://doi.org/10.3390/su17083541","source":"openalex"},{"id":"oa:W4411403091","type":"article-journal","title":"The impact of generative AI on health professional education: A systematic review in the context of student learning","abstract":"BACKGROUND: Generative Artificial Intelligence (GenAI) is increasingly integrated into health professions education (HPE), offering new opportunities for student learning. However, current research lacks a comprehensive understanding of how HPE students actually use GenAI in practice. Laurillard's Conversational Framework outlines six learning types-acquisition, inquiry, practice, production, discussion and collaboration-commonly used to categorise learning activities supported by conventional and digital technologies. Gaining insight into how GenAI aligns with these six learning types could assist HPE academics in integrating it more thoughtfully and effectively into teaching and learning. PURPOSE: This systematic review investigates how HPE students utilise GenAI and examines how these uses align with Laurillard's six learning types compared to conventional and digital technologies. MATERIAL AND METHODS: A systematic review searching five major databases-ERIC, Education Database, Ovid Medline, Ovid Embase and Scopus including articles on HPE students' use of GenAI until 15th September 2024. Studies were included if they were conducted within formal HPE training programs in HPE and specifically mentioned how students interact with GenAI. Data were mapped to the six learning modes of the Laurillard's Framework. Study quality was assessed using the Medical Education Research Study Quality Instrument (MERSQI). RESULTS: Thirty-three studies met inclusion criteria. GenAI supported learning most frequently in practice (73%), inquiry (70%), production (67%) and acquisition (55%). These studies highlight GenAI's varied educational applications, from clarifying complex concepts to simulating clinical scenarios and generating practice materials. Discussion and collaboration were less represented (12% each), suggesting a shift toward more individualised learning with GenAI. The findings highlight benefits such as efficiency and accessibility, alongside concerns about critical thinking, academic integrity and reduced peer interaction. CONCLUSION: This review has provided insights into HPE students' learning aligned with Laurillard's existing six learning types. Although GenAI supports personalised and self-directed learning, its role in collaborative modes is under-explored.","author":[{"family":"Pham","given":"Thai"},{"family":"Karunaratne","given":"Nilushi"},{"family":"Exintaris","given":"Betty"},{"family":"Liu","given":"Danny"},{"family":"Lay","given":"Travis"},{"family":"Yuriev","given":"Elizabeth"},{"family":"Lim","given":"Angelina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/medu.15746","URL":"https://doi.org/10.1111/medu.15746","source":"openalex"},{"id":"oa:W4415041425","type":"article-journal","title":"Effect of artificial intelligence-assisted personalized feedback on radiographic diagnostic performance of dental students: a controlled study","abstract":"BACKGROUND: This study aimed to evaluate the impact of MeSH based personalized learning guides generated by ChatGPT-4o on the radiographic diagnostic performance of dental students and to compare it with the traditional correct/incorrect feedback method. METHODS: This randomized controlled study was conducted among fifth-year dental students at Afyonkarahisar Health Sciences University. A total of 110 students were randomly assigned to either the experimental or control group. The experimental group received personalized study guides targeting their learning gaps, generated by ChatGPT-4o based on Medical Subject Headings (MeSH). The control group received only a standard correct/incorrect feedback analysis. One month after the intervention, a post-test was administered to assess diagnostic accuracy and student satisfaction. RESULTS: The increase in test scores from pre- to post-test was significantly higher in the experimental group (3.6 ± 1.0) compared to the control group (1.3 ± 1.2; p < 0.001). Final test scores were also significantly higher in the experimental group (p < 0.001). Survey responses indicated that the experimental group rated the feedback as more understandable, beneficial, and motivating compared to the control group. CONCLUSIONS: ChatGPT-4o based personalized feedback proved to be an effective tool for enhancing diagnostic performance and supporting learning in dental education. The findings suggest that AI-driven individualized educational strategies hold significant potential in the future of dental training.","author":[{"family":"Yılmaz","given":"Büşra"},{"family":"Yılmaz","given":"Büşra"},{"family":"Ozbey","given":"Furkan"},{"family":"Yilmaz","given":"Baki"},{"family":"Yilmaz","given":"Baki"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12909-025-07875-4","URL":"https://doi.org/10.1186/s12909-025-07875-4","source":"openalex"},{"id":"oa:W4410164686","type":"article-journal","title":"Artificial intelligence demonstrates potential to enhance orthopaedic imaging across multiple modalities: A systematic review","abstract":"Purpose: While several artificial intelligence (AI)-assisted medical imaging applications are reported in the recent orthopaedic literature, comparison of the clinical efficacy and utility of these applications is currently lacking. The aim of this systematic review is to evaluate the effectiveness and reliability of AI applications in orthopaedic imaging, focusing on their impact on diagnostic accuracy, image segmentation and operational efficiency across various imaging modalities. Methods: Based on the PRISMA guidelines, a comprehensive literature search of PubMed, Cochrane and Scopus databases was performed, using combinations of keywords and MeSH descriptors ('AI', 'ML', 'deep learning', 'orthopaedic surgery' and 'imaging') from inception to March 2024. Included were studies published between September 2018 and February 2024, which evaluated machine learning (ML) model effectiveness in improving orthopaedic imaging. Studies with insufficient data regarding the output variable used to assess the reliability of the ML model, those applying deterministic algorithms, unrelated topics, protocol studies, and other systematic reviews were excluded from the final synthesis. The Joanna Briggs Institute (JBI) Critical Appraisal tool and the Risk Of Bias In Non-randomised Studies-of Interventions (ROBINS-I) tool were applied for the assessment of bias among the included studies. Results: The 53 included studies reported the use of 11.990.643 images from several diagnostic instruments. A total of 39 studies reported details in terms of the Dice Similarity Coefficient (DSC), while both accuracy and sensitivity were documented across 15 studies. Precision was reported by 14, specificity by nine, and the F1 score by four of the included studies. Three studies applied the area under the curve (AUC) method to evaluate ML model performance. Among the studies included in the final synthesis, Convolutional Neural Networks (CNN) emerged as the most frequently applied category of ML models, present in 17 studies (32%). Conclusion: The systematic review highlights the diverse application of AI in orthopaedic imaging, demonstrating the capability of various machine learning models in accurately segmenting and analysing orthopaedic images. The results indicate that AI models achieve high performance metrics across different imaging modalities. However, the current body of literature lacks comprehensive statistical analysis and randomized controlled trials, underscoring the need for further research to validate these findings in clinical settings. Level of evidence: Systematic Review; Level of evidence IV.","author":[{"family":"Longo","given":"Umile"},{"family":"Lalli","given":"Alberto"},{"family":"Nicodemi","given":"Guido"},{"family":"Pisani","given":"Matteo"},{"family":"Sire","given":"Alessandro"},{"family":"Dhooghe","given":"Pieter"},{"family":"Nazarian","given":"Ara"},{"family":"Oeding","given":"Jacob"},{"family":"Zsidai","given":"Bálint"},{"family":"Samuelsson","given":"Kristian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/jeo2.70259","URL":"https://doi.org/10.1002/jeo2.70259","source":"openalex"},{"id":"oa:W4408551026","type":"article-journal","title":"Artificial Intelligence in Coronary Artery Interventions: Preprocedural Planning and Procedural Assistance","abstract":"Artificial intelligence (AI) has profoundly influenced the field of cardiovascular interventions and coronary artery procedures in particular. AI has enhanced diagnostic accuracy in coronary artery disease through advanced invasive and noninvasive imaging modalities, facilitating more precise diagnosis and personalized interventional strategies. AI integration in coronary interventions has streamlined diagnostic and procedural workflows, improved procedural accuracy, increased clinician efficiency, and enhanced patient safety and outcomes. Despite its potential, AI still faces significant challenges, including concerns regarding algorithmic biases, lack of transparency in AI-driven decision making, and ethical challenges. This review explores the latest advancements of AI applications in coronary artery interventions, focusing on preprocedural planning and real-time procedural guidance. It also addresses the major limitations and obstacles that hinder the widespread clinical adoption of AI technologies in this field.","author":[{"family":"Samant","given":"Saurabhi"},{"family":"Panagopoulos","given":"Anastasios"},{"family":"Wu","given":"Wei"},{"family":"Zhao","given":"Shijia"},{"family":"Chatzizisis","given":"Yiannis"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jscai.2024.102519","URL":"https://doi.org/10.1016/j.jscai.2024.102519","source":"openalex"},{"id":"oa:W7127575260","type":"article-journal","title":"Artificial intelligence in airway management: a narrative review","abstract":"This narrative review examines artificial intelligence (AI) applications in airway management through a structured literature search completed in July 2025. AI shows promise in predicting difficult airways through facial recognition, voice analysis, and multiparametric assessment. AI models demonstrate improved positive predictive values compared with conventional bedside tests, though prediction remains imperfect. The fundamental challenge of low positive predictive values persists, albeit at reduced levels. For videolaryngoscopy, AI-powered systems provide real-time structure identification, procedural guidance, tracheal tube placement verification, and potentially fewer complications. AI also offers cognitive support during critical scenarios by mitigating decision biases, providing intelligent alarms, and ensuring guideline adherence when human performance might be compromised by stress. In education, AI-enhanced virtual reality simulations create realistic practice environments with personalised feedback tailored to learners' needs. Experimental robotic intubation systems guided by AI algorithms demonstrate performance comparable with experts in controlled settings, though primarily for training and remote applications. Despite these advances, AI should complement rather than replace human expertise. Important challenges remain, including the risk of clinician deskilling, the 'black box' nature of some algorithms limiting transparency, and the need for robust validation before widespread implementation. The ideal approach involves human-machine collaboration where AI compensates for cognitive limitations while physicians contribute clinical judgement, context awareness, and empathy. As AI continues evolving, appropriate ethical frameworks and regulatory oversight must be developed in parallel to ensure these technologies enhance patient safety while preserving the essential human element of healthcare.","author":[{"family":"Sorbello","given":"Massimiliano"},{"family":"Via","given":"Luigi"},{"family":"Paternò","given":"Daniele"},{"family":"Tutino","given":"Simona"},{"family":"Giudice","given":"Emilia"},{"family":"Lentini","given":"Mario"},{"family":"Maniaci","given":"Antonino"},{"family":"Pappalardo","given":"Federico"},{"family":"Ds","given":"Paternò"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1016/j.bja.2025.12.052","URL":"https://doi.org/10.1016/j.bja.2025.12.052","source":"pubmed"},{"id":"oa:W4407613670","type":"article-journal","title":"An artificial intelligence and machine learning-driven CFD simulation for optimizing thermal performance of blood-integrated ternary nano-fluid","abstract":"Optimising heat transfer in biomedical systems, especially in blood-mediated liquids, is essential for precise medication administration and thermal ablation treatments. However, conventional methods for modelling and optimizing these frameworks frequently encounter challenges owing to their intricacy and the multitude of interconnected variables. In this work, we propose a computational fluid dynamics (CFD), machine learning (ML), and an artificial intelligence (AI) based computational framework for hemodynamics simulation of couple-stressed hybrid nano-integrated blood flow through parallel plates under external squeezing. The aim of this study is to enhance the thermal conductivity of blood-integrated tri-hybrid nanofluids, thus increasing the transfer of heat and maintaining temperature in biomedical systems. An AI-integrated, the Levenberg-Marquardt algorithm is employed with a neural network back propagation approach (ANN-LMA) for comprehensive analysis of viscous dissipation and the Lorentz force effects influenced tri-hybrid nano-fluid mixture. Non-linear, coupled partial differential equations are transformed into ordinary differential equations with similarity scaling to characterize heat transfer and fluid flow, which are then numerically solved using the modified finite difference method (the Keller-Box method). The heat transfer ability of ternary nano-fluid is enhanced with an increase in the couple stress parameter while, a rising Hartmann number results in more thermal diffusion. Regression scores equal to 1 indicate a good match between the actual data and the predictions. Conclusively, the proposed investigation provides insightful AI, ML and CFD-proposed analysis of blood-based nano-particles which can improve imaging techniques, provide tailored drug delivery, reduce hyperthermia, improve blood flow, and show potential for application in medicine.Highlights Artificial intelligence and machine learning-based CFD simulation of the blood-mediated tri-hybrid nano-fluid flow is presented.An improved finite difference scheme (the Keller-Box method), is utilized to numerically evaluate the problem.The LMA-ANN forecasts with an absolute error range of 10−11 to 10−8 relative to the actual data.Regression scores equal to 1 indicate a strong correlation between forecasts and actual data.","author":[{"family":"Hussain","given":"Mohib"},{"family":"Du","given":"Lin"},{"family":"Waqas","given":"Hassan"},{"family":"Almdallal","given":"Qasem"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/19942060.2025.2459664","URL":"https://doi.org/10.1080/19942060.2025.2459664","source":"openalex"},{"id":"oa:W4414280387","type":"article-journal","title":"Accuracy of Artificial Intelligence vs Professionally Translated Discharge Instructions","abstract":"Importance: Patients using languages other than English are a group at risk of poor health outcomes and encounter barriers to access of translated written materials. Although artificial intelligence (AI) may offer an opportunity to improve access, few studies have evaluated the accuracy and safety of AI translation for clinical care under routine practice conditions. Objective: To investigate the accuracy of AI translation compared with professional human translation of patient-specific issued pediatric inpatient discharge instructions. Design, Setting, and Participants: This comparative effectiveness analysis compared translations by a neural machine translation model vs professional translators using patient-specific pediatric inpatient discharge instructions received by families between May 18, 2023, and May 18, 2024, at a single center academic pediatric hospital. Instructions were translated to Simplified Chinese, Somali, Spanish, and Vietnamese by professional translators and the Azure AI system and then broken into scoring sections. Two professional translators per language evaluated translations (blinded to source) on an established 5-point scale for fluency, adequacy, meaning, and error severity, with 1 indicating worst performance and 5 indicating best performance. Exposure: AI vs professional translation. Main Outcome and Measure: Quality of discharge instruction translation, including fluency, adequacy, meaning, and severity of errors. Results: A total of 148 sections from 34 discharge instructions were analyzed. When considering all 4 languages together, average fluency, adequacy, and meaning were lower among AI compared with professional human translations. Among all tested languages, mean (SD) fluency for AI translations was 2.98 (1.12) compared with 3.90 (0.96) for professional translations (difference, 0.92; 95% CI, 0.83-1.01; P < .001), adequacy was 3.81 (1.14) compared with 4.56 (0.70) (difference, 0.74; 95% CI, 0.65-0.83; P < .001), meaning was 3.38 (1.15) compared with 4.28 (0.84) (difference, 0.90; 95% CI, 0.80-0.99; P < .001), and error severity was 3.53 (1.28) compared with 4.48 (0.88) (difference, 0.95; 95% CI, 0.85-1.06; P < .001). Compared with professional translations, the Spanish AI translations were noninferior in adequacy (difference, 0.08; 95% CI, -0.02 to 0.19) and error severity (difference, 0.03; 95% CI, -0.09 to 0.14) but inferior in fluency (difference, 0.38; 95% CI, 0.23-0.53) and just crossed the inferiority threshold in meaning (difference, 0.08; 95% CI, -0.04 to 0.20). The Chinese, Vietnamese, and Somali AI translations were inferior to the professional translations across all metrics, with the greatest differences for Somali. Conclusions and Relevance: In this comparative effectiveness analysis of AI- vs professionally translated issued discharge instructions, AI-translated instructions performed similarly for Spanish but worse for other languages tested. Validation and clinical implementation of AI-based translation will require special attention to languages of lesser diffusion to prevent creating new inequities.","author":[{"family":"Martos","given":"MM"},{"family":"Fields","given":"Blanca"},{"family":"Finlayson","given":"Samuel"},{"family":"Hartell","given":"Nigel"},{"family":"Kim","given":"Theresa"},{"family":"Larimer","given":"Emily"},{"family":"Lau","given":"Jason"},{"family":"Lin","given":"Yu"},{"family":"Salaguinto","given":"Taylor"},{"family":"Tran-Ngoc","given":"Nguyen"},{"family":"Lion","given":"KC"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1001/jamanetworkopen.2025.32312","URL":"https://doi.org/10.1001/jamanetworkopen.2025.32312","source":"openalex"},{"id":"oa:W4409488333","type":"article-journal","title":"Artificial intelligence tools in supporting healthcare professionals for tailored patient care","abstract":"Artificial intelligence (AI) tools to support clinicians in providing patient-centered care can contribute to patient empowerment and care efficiency. We aimed to draft potential AI tools for tailored patient support corresponding to patients' needs and assess clinicians' perceptions about the usefulness of those AI tools. To define patients' issues, we analyzed 528,199 patient messages of 11,123 patients with diabetes by harnessing natural language processing and AI. Applying multiple prompt-engineering techniques, we drafted a series of AI tools, and five endocrinologists evaluated them for perceived usefulness and risk. Patient education and administrative support for timely and streamlined interaction were perceived as highly useful, yet deeper integration of AI tools into patient data was perceived as risky. This study proposes assorted AI applications as clinical assistance tailored to patients' needs substantiated by clinicians' evaluations. Findings could offer essential ramifications for developing potential AI tools for precision patient care for diabetes and beyond.","author":[{"family":"Kim","given":"Jiyeong"},{"family":"Chen","given":"Michael"},{"family":"Rezaei","given":"Shawheen"},{"family":"Hernandezboussard","given":"Tina"},{"family":"Chen","given":"Jonathan"},{"family":"Rodríguez","given":"Fátima"},{"family":"Han","given":"Summer"},{"family":"Lal","given":"Rayhan"},{"family":"Kim","given":"Sun"},{"family":"Dosiou","given":"Chrysoula"},{"family":"Seav","given":"Susan"},{"family":"Akcan","given":"Tugce"},{"family":"Rodríguez","given":"Carolyn"},{"family":"Asch","given":"Steven"},{"family":"Linos","given":"Eleni"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01604-3","URL":"https://doi.org/10.1038/s41746-025-01604-3","source":"openalex"},{"id":"oa:W4415646726","type":"article-journal","title":"The Synergy of Artificial Intelligence and 3D Bioprinting: Unlocking New Frontiers in Precision and Tissue Fabrication","abstract":"This Review examines the transformative role of artificial intelligence (AI) in 3D bioprinting, focusing on how advanced AI technologies enhance its precision, functionality, and scalability. AI, through branches, such as machine learning (ML), computer vision (CV), robotics, natural language processing and expert systems, provides critical improvements in real-time process monitoring, error correction, and optimization of bioprinting parameters. The integration of AI enables automated quality control and predictive maintenance, improving bioprinting outcomes by increasing cell viability and structural fidelity, and reducing the amount of bioink wasted. Specifically, ML algorithms are employed to predict optimal bioprinting conditions and streamline the bioprinting workflow, while deep learning enhances the ability to process complex datasets for precision tissue biofabrication. Furthermore, AI-powered robotics and CV systems ensure accurate bioink placement and facilitate the construction of complex tissues. Despite the remarkable progress, challenges remain, particularly in the areas of process monitoring, quality control, and the scalability of bioprinting systems. This Review also aims to guide scientists, engineers, and healthcare providers in understanding the complexities and potential of AI-enhanced bioprinting, fostering a deeper appreciation of its role in the future of regenerative medicine and personalized healthcare.","author":[{"family":"Robazzi","given":"João"},{"family":"Derman","given":"İrem"},{"family":"Gupta","given":"Deepak"},{"family":"Haugh","given":"Logan"},{"family":"Singh","given":"Yogendra"},{"family":"Pal","given":"Vaibhav"},{"family":"Yilmaz","given":"Yasar"},{"family":"Liu","given":"Suihong"},{"family":"Dias","given":"André"},{"family":"Flauzino","given":"Rogério"},{"family":"Özbolat","given":"İbrahim"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/adfm.202509530","URL":"https://doi.org/10.1002/adfm.202509530","source":"openalex"},{"id":"oa:W4415815680","type":"article-journal","title":"From Black Box to Glass Box: A Practical Review of Explainable Artificial Intelligence (XAI)","abstract":"Explainable Artificial Intelligence (XAI) has become essential as machine learning systems are deployed in high-stakes domains such as security, finance, and healthcare. Traditional models often act as “black boxes”, limiting trust and accountability. Traditional models often act as “black boxes”, limiting trust and accountability. However, most existing reviews treat explainability either as a technical problem or a philosophical issue, without connecting interpretability techniques to their real-world implications for security, privacy, and governance. This review fills that gap by integrating theoretical foundations with practical applications and societal perspectives. define transparency and interpretability as core concepts and introduce new economics-inspired notions of marginal transparency and marginal interpretability to highlight diminishing returns in disclosure and explanation. Methodologically, we examine model-agnostic approaches such as LIME and SHAP, alongside model-specific methods including decision trees and interpretable neural networks. We also address ante-hoc vs. post hoc strategies, local vs. global explanations, and emerging privacy-preserving techniques. To contextualize XAI’s growth, we integrate capital investment and publication trends, showing that research momentum has remained resilient despite market fluctuations. Finally, we propose a roadmap for 2025–2030, emphasizing evaluation standards, adaptive explanations, integration with Zero Trust architectures, and the development of self-explaining agents supported by global standards. By combining technical insights with societal implications, this article provides both a scholarly contribution and a practical reference for advancing trustworthy AI.","author":[{"family":"Liu","given":"Xiaoming"},{"family":"Huang","given":"Danni"},{"family":"Yao","given":"Jingyu"},{"family":"Dong","given":"Jing"},{"family":"Song","given":"Litong"},{"family":"Wang","given":"Hui"},{"family":"Yao","given":"Chao"},{"family":"Chu","given":"Weishen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ai6110285","URL":"https://doi.org/10.3390/ai6110285","source":"openalex"},{"id":"oa:W4411078597","type":"article-journal","title":"Macy Foundation Innovation Report Part II: From Hype to Reality: Innovators’ Visions for Navigating AI Integration Challenges in Medical Education","abstract":"PURPOSE: Artificial intelligence (AI) promises to significantly impact medical education, yet its implementation raises important questions about educational effectiveness, ethical use, and equity. In the second part of a 2-part innovation report, which was commissioned by the Josiah Macy Jr. Foundation to inform discussions at a conference on AI in medical education, the authors explore the perspectives of innovators actively integrating AI into medical education, examining their perceptions regarding the impacts, opportunities, challenges, and strategies for successful AI adoption and risk mitigation. METHOD: Semistructured interviews were conducted with 25 medical education AI innovators-including learners, educators, institutional leaders, and industry representatives-from June to August 2024. Interviews explored participants' perceptions of AI's influence on medical education, challenges to integration, and strategies for mitigating challenges. Transcripts were analyzed using thematic analysis to identify themes and synthesize participants' recommendations for AI integration. RESULTS: Innovators' responses were synthesized into 2 main thematic areas: (1) AI's impact on teaching, learning, and assessment, and (2) perceived threats and strategies for mitigating them. Participants identified AI's potential to enact precision education through virtual tutors and standardized patients, support active learning formats, enable centralized teaching, and facilitate cognitive offloading. AI-enhanced assessments could automate grading, predict learner trajectories, and integrate performance data from clinical interactions. Yet, innovators expressed concerns over threats to transparency and validity, potential propagation of biases, risks of over-reliance and deskilling, and institutional disparities. Proposed mitigation strategies emphasized validating AI outputs, establishing foundational competencies, fostering collaboration and open-source sharing, enhancing AI literacy, and maintaining robust ethical standards. CONCLUSIONS: AI innovators in medical education envision transformative opportunities for individualized learning and precision education, balanced against critical threats. Realizing these benefits requires proactive, collaborative efforts to establish rigorous validation frameworks; uphold foundational medical competencies; and prioritize ethical, equitable AI integration.","author":[{"family":"Gin","given":"Brian"},{"family":"Laforge","given":"Kate"},{"family":"Burkrafel","given":"Jesse"},{"family":"Boscardin","given":"Christy"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1097/acm.0000000000006117","URL":"https://doi.org/10.1097/acm.0000000000006117","source":"openalex"},{"id":"oa:W4409546651","type":"article-journal","title":"Artificial Intelligence in Oral Diagnosis: Detecting Coated Tongue with Convolutional Neural Networks","abstract":"Background/Objectives: Coated tongue is a common oral condition with notable clinical relevance, often overlooked due to its asymptomatic nature. Its presence may reflect poor oral hygiene and can serve as an early indicator of underlying systemic diseases. This study aimed to develop a robust diagnostic model utilizing convolutional neural networks and machine learning classifiers to improve the detection of coated tongue lesions. Methods: A total of 200 tongue images (100 coated and 100 healthy) were analyzed. Images were acquired using a DSLR camera (Nikon D5500 with Sigma Macro 105 mm lens, Nikon, Tokyo, Japan) under standardized daylight conditions. Following preprocessing, feature vectors were extracted using CNN architectures (VGG16, VGG19, ResNet, MobileNet, and NasNet) and classified using Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Multi-Layer Perceptron (MLP) classifiers. Performance metrics included sensitivity, specificity, accuracy, and F1 score. Results: The SVM + VGG19 hybrid model achieved the best performance among all tested configurations, with a sensitivity of 82.6%, specificity of 88.23%, accuracy of 85%, and an F1 score of 86.36%. Conclusions: The SVM + VGG19 model demonstrated high accuracy and reliability in diagnosing coated tongue lesions, highlighting its potential as an effective clinical decision support tool. Future research with larger datasets may further enhance model robustness and applicability in diverse populations.","author":[{"family":"Baybars","given":"Sümeyye"},{"family":"Talu","given":"Merve"},{"family":"Danacı","given":"Çağla"},{"family":"Tuncer","given":"Seda"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/diagnostics15081024","URL":"https://doi.org/10.3390/diagnostics15081024","source":"openalex"},{"id":"oa:W4416250669","type":"article-journal","title":"Reshaping Higher Education Designs and Futures: Postdigital Co-design with Generative Artificial Intelligence","abstract":"Abstract This article examines how collaborative design practices in higher education are reshaped through postdigital entanglement with generative artificial intelligence (GenAI). We collectively explore how co-design, an inclusive, iterative, and relational approach to educational design and transformation, expands in meaning, practice, and ontology when GenAI is approached as a collaborator. The article brings together 19 authors and three open reviewers to engage with postdigital inquiry, structured in three parts: (1) a review of literature on co-design, GenAI, and postdigital theory; (2) 11 situated contributions from educators, researchers, and designers worldwide, each offering practice-based accounts of co-design with GenAI; and (3) an explorative discussion of implications for higher education designs and futures. Across these sections, we show how GenAI unsettles assumptions of collaboration, knowing, and agency, foregrounding co-design as a site of ongoing material, ethical, and epistemic negotiation. We argue that postdigital co-design with GenAI reframes educational design as a collective practice of imagining, contesting, and shaping futures that extend beyond human knowing.","author":[{"family":"Zeivots","given":"Sandris"},{"family":"Casey","given":"Alison"},{"family":"Winchester","given":"Tiffany"},{"family":"Webster","given":"Jack"},{"family":"Wang","given":"Xin"},{"family":"Tan","given":"Linus"},{"family":"Smeenk","given":"Wina"},{"family":"Schulte","given":"Frank"},{"family":"Scholkmann","given":"Antonia"},{"family":"Paulovich","given":"Belinda"},{"family":"Muñoz","given":"Diego"},{"family":"Mignone","given":"Joanne"},{"family":"Mantai","given":"Lilia"},{"family":"Hrastinski","given":"Stefan"},{"family":"Godwin","given":"Rebecca"},{"family":"Engwall","given":"Olov"},{"family":"Dindas","given":"Henrik"},{"family":"Dijk","given":"Marieke"},{"family":"Chubb","given":"Laura"},{"family":"Rapanta","given":"Chrysi"},{"family":"Jaldemark","given":"Jimmy"},{"family":"Hayes","given":"Sarah"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s42438-025-00595-4","URL":"https://doi.org/10.1007/s42438-025-00595-4","source":"openalex"},{"id":"oa:W4411457325","type":"article-journal","title":"Segmentation of Pulp and Pulp Stones with Automatic Deep Learning in Panoramic Radiographs: An Artificial Intelligence Study","abstract":"Background/Objectives: Different sized calcified masses called pulp stones are often detected in dental pulp and can impact dental procedures. The current research was conducted with the aim of measuring the ability of artificial intelligence algorithms to accurately diagnose pulp and pulp stone calcifications on panoramic radiographs. Methods: We used 713 panoramic radiographs, on which a minimum of one pulp stone was detected, identified retrospectively, and included in the study—4675 pulp stones and 5085 pulps were marked on these radiographs using CVAT v1.7.0 labeling software. Results: In the test dataset, the AI model segmented 462 panoramic radiographs for pulp stone and 220 panoramic radiographs for pulp. The dice coefficient and Intersection over Union (IoU) recorded for the Pulp Segmentation model were 0.84 and 0.758, respectively. Precision and recall were computed to be 0.858 and 0.827, respectively. The Pulp Stone Segmentation model achieved a dice coefficient of 0.759 and an IoU of 0.686, with precision and recall of 0.792 and 0.773, respectively. Conclusions: Pulp and pulp stones can successfully be identified using artificial intelligence algorithms. This study provides evidence that artificial intelligence software using deep learning algorithms can be valuable adjunct tools in aiding clinicians in radiographic diagnosis. Further research in which larger datasets are examined are needed to enhance the capability of artificial intelligence models to make accurate diagnoses.","author":[{"family":"Firincioglulari","given":"Mujgan"},{"family":"Boztuna","given":"Mehmet"},{"family":"Mırzaeı","given":"Omid"},{"family":"Karanfiller","given":"Tolgay"},{"family":"Akkaya","given":"Nurullah"},{"family":"Orhan","given":"Kaan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/dj13060274","URL":"https://doi.org/10.3390/dj13060274","source":"openalex"},{"id":"oa:W4413955462","type":"article-journal","title":"The Future of Enhanced Recovery After Surgery in General Surgery: Integrating Artificial Intelligence, Personalized Care, and Technological Advances","abstract":"Enhanced Recovery After Surgery (ERAS) protocols have changed and improved surgical care and practice, transforming classic approaches toward more evidence-based practices. ERAS protocols demonstrate consistent benefits, including reduced hospital length of stay, decreased postoperative complications, and significant cost savings. Advances include the development and integration of AI and digital health technologies that promise to personalize and optimize recovery pathways. However, the global application of ERAS may face great challenges, starting with resource limitations, resistance to change, and variable compliance rates. This review discusses and highlights the application of ERAS, identifies barriers to implementation, and proposes evidence-based recommendations for optimizing ERAS adoption and sustainability in modern surgical practice.","author":[{"family":"Abosheisha","given":"Mohamed"},{"family":"Nasr","given":"É"},{"family":"Abdel-Latif","given":"Mohamed"},{"family":"Swealem","given":"Ahmed"},{"family":"Ammar","given":"Ahmed"},{"family":"Hasan","given":"Md"},{"family":"Abdelglil","given":"Momen"},{"family":"Tamanna","given":"Rezuana"},{"family":"Ismaiel","given":"Mohamed"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.91528","URL":"https://doi.org/10.7759/cureus.91528","source":"openalex"},{"id":"oa:W4411367739","type":"article-journal","title":"Code of ethics for the use of artificial intelligence in the Russian Federation healthcare","abstract":"The article discusses the process of development and approval of the first Code of Ethics of Artificial Intelligence (AI) application in the Russian Federation Healthcare. Against the backdrop of the active integration of AI technologies into medical practice (39 relevant medical devices have been registered), the emphasis is placed on the importance of establishing ethical standards that ensure the protection of patients' rights, increasing trust in technologies, and standardization processes. International approaches to AI ethics in healthcare (EU, USA, UK, Canada, Australia, China, India) are analyzed and the need to harmonize the domestic code with international initiatives is outlined. The stages of development of the document, in which employees of specialized departments of the Ministry of Health of Russia, chief freelance specialists and experts took part, as well as the structure and main provisions of the approved version of the Code are presented. The key principles emphasized include transparency, confidentiality, fairness, limited autonomy, oversight, and accountability of AI systems. The final version of the document was published in March 2025 on the Unified State Information System in Healthcare (EGISZ) portal after approval by the Interdepartmental Working Group under the Russian Ministry of Health. The Code is intended to serve as a foundation for the sustainable and safe implementation of AI in Russia's healthcare system.","author":[{"family":"Koroleva","given":"Julia"},{"family":"Хохлов","given":"АЛ"},{"family":"Artemova","given":"OR"},{"family":"Kostina","given":"E"},{"family":"Зарубина","given":"ТВ"}],"issued":{"date-parts":[[2025]]},"DOI":"10.25881/18110193_2025_2_98","URL":"https://doi.org/10.25881/18110193_2025_2_98","source":"openalex"},{"id":"oa:W4414830255","type":"article-journal","title":"Opportunities and Challenges of Using Artificial Intelligence in Predicting Clinical Outcomes and Length of Stay in Neonatal Intensive Care Units: Systematic Review","abstract":"BACKGROUND: The use of artificial intelligence (AI) in health care has been steadily increasing for over 2 decades. Integrating AI into neonatal intensive care units (NICUs) has promise as it has the potential to reshape neonatal care and improve outcomes. However, challenges such as data quality, clinical interpretation, and ethical considerations may hinder AI's practical implementation in NICUs. OBJECTIVE: This study aims (1) to analyze the current AI research landscape for predicting clinical outcomes and length of stay in the NICU and (2) to explore the benefits and challenges of using AI in the NICU for these predictions. METHODS: A systematic review was conducted across 6 databases-PubMed, Embase, CINAHL, Cochrane Library, Informit, and La Trobe Library-to identify English-language peer-reviewed articles published between January 2017 and March 2023 that focused on the use of AI for predicting length of stay and clinical outcomes for NICU patients. Eligibility criteria excluded studies outside the NICU context or lacking predictive focus. Both prospective and retrospective designs were included. A thematic analysis of AI applications in NICUs from the articles identified was conducted. RESULTS: A total of 24 studies were included in the review, comprising 15 retrospective and 9 prospective designs. These studies primarily originated from the United States (13 studies), with others from Austria, Taiwan, and other countries. The studies evaluated AI applications in NICU settings to predict comorbidities (18/24), mortality (4/24), and length of stay (2/24). Sixteen studies were in the exploration stage, lacking cohesive AI strategies, while 8 demonstrated systematic exploration but no fully integrated solutions. The synthesis of results identified key applications of AI in NICU care, including data-driven insights and predictive models, advancements in medical imaging, improved risk stratification, and personalized neonatal care. AI showed promise in enhancing diagnostic accuracy and care planning, but significant challenges persist, such as data quality, model generalization, and ethical concerns. No studies reported a fully integrated AI ecosystem, highlighting the need for further research to bridge gaps and realize AI's transformative potential in neonatal care. CONCLUSIONS: This review highlights the potential of AI in improving NICU care, particularly through predictive models, medical imaging, and personalized interventions. However, the evidence is limited by significant methodological variability, small sample sizes, risk of bias, and a lack of external validation in included studies. Many studies remain in exploratory phases without cohesive AI strategies or integration into clinical practice, limiting the practical applicability of findings. These results underscore the importance of addressing challenges such as data quality, model generalization, and ethical considerations to fully realize AI's potential in neonatal care. Future research should focus on robust validation, comprehensive implementation strategies, and ethical frameworks to ensure AI's effective and responsible integration into NICU settings.","author":[{"family":"Tudor","given":"Samantha"},{"family":"Bhatia","given":"Risha"},{"family":"Liem","given":"Michael"},{"family":"Wani","given":"Tafheem"},{"family":"Boyd","given":"James"},{"family":"Khan","given":"Urooj"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/63175","URL":"https://doi.org/10.2196/63175","source":"openalex"},{"id":"oa:W4411886972","type":"article-journal","title":"Exploring value co-creation and co-destruction between consumers & generative artificial intelligence (GAI) in travel","abstract":"Little is known about the (dis)benefits of using generative artificial intelligence (GAI) with travel-related purposes, which hinders an understanding of the value co-created and co-destructed in the process of its use by tourists. This mixed methods study explored and examined the key factors in value co-creation and co-destruction when using a popular GAI's conversational interface, ChatGPT, in tourism. The results indicate that the key perceived utility of ChatGPT is in travel planning and time saving, and the main perceived shortcomings are its limited knowledge and inaccurate responses. The study pinpoints the importance of refining and developing GAI collaboratively by all tourism stakeholders given that perceived value co-creation outweighs value co-destruction.","author":[{"family":"Bui","given":"Hien"},{"family":"Filimonau","given":"Viachaslau"},{"family":"Sezerel","given":"Hakan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.tmp.2025.101392","URL":"https://doi.org/10.1016/j.tmp.2025.101392","source":"openalex"},{"id":"oa:W4411801875","type":"article-journal","title":"Framing and Evaluating Task-Centered Generative Artificial Intelligence Literacy for Higher Education Students","abstract":"The rise in generative artificial intelligence (GenAI) demands new forms of literacy among higher education students. This paper introduces a novel task-centered generative artificial intelligence literacy framework, which was developed collaboratively with academic and administrative staff at a large research university in Israel. The framework identifies eight skills which are informed by the six cognitive domains of Bloom’s Taxonomy. Based on this framework, we developed a measuring tool for students’ GenAI literacy and surveyed 1667 students. Findings from the empirical phase show moderate GenAI use and medium–high literacy levels, with significant variations by gender, discipline, and age. Notably, 82% of students support formal GenAI instruction, favoring integration within curricula to prepare for broader digital society participation. The study offers actionable insights for educators and policymakers aiming to integrate GenAI into higher education responsibly and effectively.","author":[{"family":"Hershkovitz","given":"Arnon"},{"family":"Tabach","given":"Michal"},{"family":"Reich","given":"Yoram"},{"family":"Lurie","given":"Lilach"},{"family":"Cholcman","given":"Tamar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/systems13070518","URL":"https://doi.org/10.3390/systems13070518","source":"openalex"},{"id":"oa:W4412077961","type":"article-journal","title":"Development and retrospective validation of an artificial intelligence system for diagnostic assessment of prostate biopsies: study protocol","abstract":"INTRODUCTION: Histopathological evaluation of prostate biopsies using the Gleason scoring system is critical for prostate cancer diagnosis and treatment selection. However, grading variability among pathologists can lead to inconsistent assessments, risking inappropriate treatment. Similar challenges complicate the assessment of other prognostic features like cribriform cancer morphology and perineural invasion. Many pathology departments are also facing an increasingly unsustainable workload due to rising prostate cancer incidence and a decreasing pathologist workforce coinciding with increasing requirements for more complex assessments and reporting. Digital pathology and artificial intelligence (AI) algorithms for analysing whole slide images show promise in improving the accuracy and efficiency of histopathological assessments. Studies have demonstrated AI's capability to diagnose and grade prostate cancer comparably to expert pathologists. However, external validations on diverse data sets have been limited and often show reduced performance. Historically, there have been no well-established guidelines for AI study designs and validation methods. Diagnostic assessments of AI systems often lack preregistered protocols and rigorous external cohort sampling, essential for reliable evidence of their safety and accuracy. METHODS AND ANALYSIS: This study protocol covers the retrospective validation of an AI system for prostate biopsy assessment. The primary objective of the study is to develop a high-performing and robust AI model for diagnosis and Gleason scoring of prostate cancer in core needle biopsies, and at scale evaluate whether it can generalise to fully external data from independent patients, pathology laboratories and digitalisation platforms. The secondary objectives cover AI performance in estimating cancer extent and detecting cribriform prostate cancer and perineural invasion. This protocol outlines the steps for data collection, predefined partitioning of data cohorts for AI model training and validation, model development and predetermined statistical analyses, ensuring systematic development and comprehensive validation of the system. The protocol adheres to Transparent Reporting of a multivariable prediction model of Individual Prognosis Or Diagnosis+AI (TRIPOD+AI), Protocol Items for External Cohort Evaluation of a Deep Learning System in Cancer Diagnostics (PIECES), Checklist for AI in Medical Imaging (CLAIM) and other relevant best practices. ETHICS AND DISSEMINATION: Data collection and usage were approved by the respective ethical review boards of each participating clinical laboratory, and centralised anonymised data handling was approved by the Swedish Ethical Review Authority. The study will be conducted in agreement with the Helsinki Declaration. The findings will be disseminated in peer-reviewed publications (open access).","author":[{"family":"Mulliqi","given":"Nita"},{"family":"Blilie","given":"Anders"},{"family":"Ji","given":"Xiaoyi"},{"family":"Szolnoky","given":"Kelvin"},{"family":"Olsson","given":"Henrik"},{"family":"Titus","given":"Matteo"},{"family":"Gonzalez","given":"Geraldine"},{"family":"Boman","given":"Sol"},{"family":"Valkonen","given":"Masi"},{"family":"Gudlaugsson","given":"Einar"},{"family":"Kjosavik","given":"Svein"},{"family":"Asenjo","given":"J"},{"family":"Gambacorta","given":"Marcello"},{"family":"Libretti","given":"Paolo"},{"family":"Braun","given":"Marcin"},{"family":"Kordek","given":"Radzisław"},{"family":"Łowicki","given":"Roman"},{"family":"Hotakainen","given":"Kristina"},{"family":"Väre","given":"Päivi"},{"family":"Pedersen","given":"Bodil"},{"family":"Sørensen","given":"Karina"},{"family":"Ulhøi","given":"Benedicte"},{"family":"Rantalainen","given":"Mattias"},{"family":"Ruusuvuori","given":"Pekka"},{"family":"Delahunt","given":"Brett"},{"family":"Samaratunga","given":"Hemamali"},{"family":"Tsuzuki","given":"Toyonori"},{"family":"Janssen","given":"Emilius"},{"family":"Egevad","given":"Lars"},{"family":"Kartasalo","given":"Kimmo"},{"family":"Eklund","given":"Martin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/bmjopen-2024-097591","URL":"https://doi.org/10.1136/bmjopen-2024-097591","source":"openalex"},{"id":"oa:W4409434819","type":"article-journal","title":"The Role of Artificial Intelligence in the Evaluation of Prostate Pathology","abstract":"Artificial intelligence (AI) is an emerging tool in diagnostic pathology, including prostate pathology. This review summarizes the possibilities offered by AI and also discusses the challenges and risks. AI has the potential to assist in the diagnosis and grading of prostate cancer. Diagnostic safety can be enhanced by avoiding the accidental underdiagnosis of small lesions. Another possible benefit is a greater degree of standardization of grading. AI for clinical use needs to be trained on large, high-quality data sets that have been assessed by experienced pathologists. A problem with the use of AI in prostate pathology is the plethora of benign mimics of prostate cancer and morphological variants of cancer that are too unusual to allow sufficient training of AI. AI systems need to be able to account for variations in local routines for cutting, staining, and scanning of slides. We also need to be aware of the risk that users will rely too much on the output of an AI system, leading to diagnostic errors and loss of clinical competence. The reporting pathologist must ultimately be responsible for accepting or rejecting the diagnosis proposed by AI.","author":[{"family":"Egevad","given":"Lars"},{"family":"Camilloni","given":"Andrea"},{"family":"Delahunt","given":"Brett"},{"family":"Samaratunga","given":"Hemamali"},{"family":"Eklund","given":"Martin"},{"family":"Kartasalo","given":"Kimmo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/pin.70015","URL":"https://doi.org/10.1111/pin.70015","source":"openalex"},{"id":"oa:W4409769501","type":"article-journal","title":"Digital and artificial intelligence-assisted cephalometric training effectively enhanced students’ landmarking accuracy in preclinical orthodontic education","abstract":"BACKGROUND: Digital cephalometric analyses, including those assisted by artificial intelligence (AI), are widely used in clinical practice. Similarly, computer-assisted learning has demonstrated teaching outcomes comparable to those of traditional methods in orthodontic education. However, the potential application of digital and AI-assisted cephalometric training in the preclinical education of orthodontic students remains unexplored. Cephalometric analysis is a fundamental skill for orthodontic students and practitioners. Therefore, this study aimed to integrate digital and AI-assisted cephalometric training into preclinical orthodontic education and evaluate its educational effectiveness. METHODS: Forty undergraduate students were grouped into pairs to use digital cephalometric training software. The students' landmarking abilities were evaluated by comparing their total scores before and after training on the same two lateral radiographs using digital cephalometric training software. The effectiveness of the software in improving landmarking accuracy was assessed objectively. Lateral radiographs of eight common patient types were selected. Twenty-four clinical training students from different grades used an AI-assisted cephalometric platform to analyze skeletal, dental, and soft tissue indicators. The accuracy of the measurements was compared among students in different grades. RESULTS: Digital cephalometric training, through real-time feedback and visual error-correction mechanisms, enabled students to quickly identify and correct errors in landmarking, significantly improving their accuracy. There was no significant difference in AI-assisted cephalometric analysis ability among students with varying levels of clinical experience. CONCLUSIONS: Digital cephalometric training effectively enhances students' landmarking accuracy in preclinical orthodontic education. AI-assisted cephalometry has the potential to minimize performance disparities among students with varying levels of clinical experience. Owing to the real-time feedback and self-directed learning features of digital tools, these technologies serve as valuable supplements to instructor-led training, potentially reducing educators' workload and accelerating skill acquisition in novice orthodontic students. However, these preliminary findings require further multicenter validation and long-term educational assessments while also considering the ethical implications of these technologies.","author":[{"family":"Lin","given":"Jiayu"},{"family":"Liao","given":"Zhihao"},{"family":"Dai","given":"Jingtao"},{"family":"Wang","given":"Manyi"},{"family":"Yu","given":"Robert"},{"family":"Yang","given":"Hong"},{"family":"Liu","given":"Chufeng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12903-025-05978-4","URL":"https://doi.org/10.1186/s12903-025-05978-4","source":"openalex"},{"id":"oa:W4406603444","type":"article-journal","title":"Performance of a medical smartband with photoplethysmography technology and artificial intelligence algorithm to detect atrial fibrillation","abstract":"Background: Atrial fibrillation (AF) is a prevalent arrhythmia with significant public health implications, including increased risk of stroke and mortality. Early detection is challenging but crucial for managing complications. Wearable technology with photoplethysmography (PPG) offers a potential solution for long-term, non-invasive monitoring. This study aims to evaluate the performance of three artificial intelligence (AI) algorithms (Happitech, Preventicus, and Philips Biosensing AF) in detecting AF using PPG signals from a medical smartband and compare it with the gold standard electrocardiogram (ECG). Methods: A medical smartband equipped with PPG technology was used to collect cardiovascular data from patients with and without AF. The sensitivity and specificity of the algorithm for detecting AF were determined by comparing their output to a trained technician's examination of concurrent ECG recordings. Results: Seventy two participants (42% female, 57±17 years old) were included in this study. The medical smartband provided continuous PPG signals, with AI algorithms evaluating the data for AF episodes. The accuracy of AF detection by the algorithms was compared with that of the concurrent ECG recordings. Sensitivity varied between 80.0% (62.5-97.5%) and 97.6% (97.6-97.6%), specificity between 90.6% (80.5-100%) and 96.9% (90.8-100%). Conclusions: This study demonstrates the potential of medical smartbands combined with PPG technology and AI algorithms for reliable AF detection. The findings suggest a promising direction for remote AF monitoring and early intervention, potentially reducing AF-related complications and healthcare costs.","author":[{"family":"Blok","given":"Sebastiaan"},{"family":"Gielen","given":"Willem"},{"family":"Piek","given":"M"},{"family":"Hoeksema","given":"Wiert"},{"family":"Tulevski","given":"Igor"},{"family":"Somsen","given":"Geert"},{"family":"Winter","given":"Michiel"}],"issued":{"date-parts":[[2025]]},"DOI":"10.21037/mhealth-24-10","URL":"https://doi.org/10.21037/mhealth-24-10","source":"openalex"},{"id":"oa:W4410652481","type":"article-journal","title":"Analyzing the impact of artificial intelligence on the online purchase decision-making process through the lens of the UTAUT 2 model","abstract":"Abstract Consumer perceptions and purchasing decision-making have been significantly impacted by the rapid use of artificial intelligence (AI) in the online retail sector. The online shopping experience has been transformed by AI technologies, which include sophisticated product evaluation algorithms, individualized purchasing recommendations, and improved customer care. To better understand the deep interactions between factors linked to artificial intelligence and customer decision-making processes, this study investigates the complex adoption landscape of AI within the Delhi-NCR online retail industry using the Unified Theory of Acceptance and Use of Technology 2 (UTAUT 2) framework. The research employs a quantitative methodology by applying Structural Equation Modelling (SEM) through SmartPLS 4 to evaluate key behavioral variables. The results reveal that Behavioral Intention (BI) has a strong and significant impact on Usage Behavior (UB). Trust in AI systems emerged as a critical factor affecting BI. Significant factors like Habit, Hedonic Motivation, Trust, and Personal Innovativeness emerged as pivotal in shaping consumers’ Behavioral Intentions towards AI in online shopping. Perceived Risk showed a marginally negative effect on BI, suggesting consumers’ hesitation in adopting AI for online shopping. However, factors such as Social Influence and Effort Expectancy had negligible impacts on BI, contrasting with prior research. These findings suggest that consumer trust, delight, and habitual AI usage are more influential than usability or social pressures for AI enabled e-commerce landscape.","author":[{"family":"Dixit","given":"Rinku"},{"family":"Choudhary","given":"Shailee"},{"family":"Govil","given":"Nikhil"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s10791-025-09575-5","URL":"https://doi.org/10.1007/s10791-025-09575-5","source":"openalex"},{"id":"oa:W4413796400","type":"article-journal","title":"Embracing Artificial Intelligence in Dental Practice: An Exploratory Study of Romanian Clinicians’ Perspectives and Experiences","abstract":"Background/Objectives: Standard dental practice is being reshaped by digital technologies, and artificial intelligence (AI) is emerging as one of the most challenging recent innovations. Methods: The present study assessed the interest of Romanian dentists in the integration of AI into their current practice through an anonymous questionnaire distributed to 200 respondents. The questionnaire addressed the integration of AI in dentistry by analyzing the following areas of intervention: stages of patient care, perceived impact on the doctor–patient relationship, data security, implementation costs, and the legislative framework. Results: The results showed that 53.6% of dentists reported low difficulties, 37.3% reported moderate difficulties, and 9.1% reported high difficulties with using digital tools. Dentists’ reported willingness to adopt AI-based solutions was as follows: 58.6% were very willing, 30% were moderately willing, and only 11.4% were not very willing. Currently, 80.5% already use digital techniques in their daily practice. The participants emphasized the need to maintain a strong doctor–patient relationship while recognizing the benefits of increased efficiency. They were aware of the risk of diminishing human connection and trust. Also, data security and the financial stress associated with implementing and maintaining new systems were seen as major obstacles. Conclusions: The dentists surveyed showed an increased interest in modern digital technologies, provided that there is a clear legal framework, a strong data protection system, and the preservation of the doctor–patient relationship based on trust and confidentiality that defines the medical profession.","author":[{"family":"Cozmescu","given":"Alin"},{"family":"Cernega","given":"Ana"},{"family":"Mincă","given":"Dana"},{"family":"Didilescu","given":"Andreea"},{"family":"Imre","given":"Marina"},{"family":"Totan","given":"Alexandra"},{"family":"Pârvu","given":"Simona"},{"family":"Pițuru","given":"Silviu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/dj13090390","URL":"https://doi.org/10.3390/dj13090390","source":"openalex"},{"id":"oa:W4413221133","type":"article-journal","title":"Enhancing Coordination and Decision Making in Humanitarian Logistics Through Artificial Intelligence: A Grounded Theory Approach","abstract":"Background: The adoption of artificial intelligence (AI) in humanitarian logistics is essential for improving coordination and decision making, especially in the challenging landscape of disaster-relief settings. However, the current literature offers limited empirical evidence with respect to the specific impact of AI on coordination and decision making for real-life humanitarian problems. Based on evidence from the humanitarian sector, this paper focuses on how AI could help humanitarian organizations collaborate better, streamline relief supply-chain operations and use resources more effectively. Methods: Twelve key themes influencing AI integration are identified by the study using a Grounded Theory (GT) approach based on interviews with experts from the humanitarian sector. These themes include data reliability, operational limitations, ethical considerations and cultural sensitivities, among others. Results: The findings suggest that AI improves forecasting, planning and inter-organizational coordination and is especially useful during the preparedness and mitigation stages of relief operations. Successful adoption, however, depends on adjusting tools to actual field conditions, building trust and training and striking a balance between algorithmic support and human expertise. Conclusions: The paper offers useful and practical advice for humanitarian organizations looking to use AI technologies in an ethical way while taking into account workforce capabilities, cross-agency cooperation and field-level realities.","author":[{"family":"Pantiris","given":"Panagiotis"},{"family":"Pallıs","given":"Petros"},{"family":"Chountalas","given":"Panos"},{"family":"Dasaklis","given":"Thomas"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/logistics9030113","URL":"https://doi.org/10.3390/logistics9030113","source":"openalex"},{"id":"oa:W7104178444","type":"article-journal","title":"Design and development of an intelligent system based on artificial intelligence and machine learning using customs digital indicators","abstract":"This paper aims to evaluate the role of AI and ML-driven innovative technologies in enhancing customs operations in Jordan. This research employed a quantitative approach to develop an overall conceptual model that encompasses both the technical and behavioral aspects of intelligent system adoption. The target population consisted of customs officers, border security personnel, and IT personnel responsible for customs clearance and trade facilitation in Jordan. The structured questionnaires were administered to the respondents to measure their perceptions of system effectiveness, satisfaction, performance outcomes, and evasion behavior and yielded a total of 358 valid responses. The research was conducted with proper statistical analysis, and the statistical techniques employed included primary data collected via SPSS Version 29 and advanced modeling using Structural Equation Modeling-Partial Least Squares (SEM-PLS) through the use of SmartPLS 4.0. The results indicated that the measurement model proved to be both valid and reliable, with Cronbach's alpha values exceeding 0.82 and AVE values above 0.50, indicating good internal consistency and convergent validity. Moreover, the structural model achieved good explanatory power, with R² values of 59% for Customs Evasion, 43% for User Satisfaction, and 100% for Digital Performance Indicators. These findings underscore the significance of user satisfaction as a key outcome of system effectiveness and a valuable tool for enhancing performance and deterrence. More specifically, the results shown how Intelligent System Effectiveness presents a positive and significant impact on User Satisfaction (β = 0.656, p < 0.001), which in turn has a high positive effect on both Digital Performance Indicators (β = 1.000, p < 0.001) and Customs Evasion reduction (β = 0.770, p < 0.001). The mediation analysis also confirmed that User Satisfaction fully mediates the relationship between system effectiveness and performance outcome, as well as between system effectiveness and evasion reduction. This research contributes to theory and practice by demystifying the design and implementation of AI-driven customs systems. It illustrates the importance of valuing both technical system quality and user-centric values in achieving and maintaining optimal performance in the digital space, as well as conformance with the law.","author":[{"family":"Qahman","given":"Ashraf"},{"family":"Alzaqebah","given":"Malek"},{"family":"Jawarneh","given":"Sana"},{"family":"Al-Zaqeba","given":"Murad"},{"family":"Al-Taani","given":"Attallah"},{"family":"Aloqaily","given":"Ahmad"},{"family":"Almatrooshi","given":"Maryam"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5267/j.ijdns.2025.9.022","URL":"https://doi.org/10.5267/j.ijdns.2025.9.022","source":"openalex"},{"id":"oa:W4409655236","type":"article-journal","title":"Artificial intelligence model for the assessment of unstained live sperm morphology","abstract":"Abstract: Traditional sperm morphology assessment requires staining and high magnification (100×), rendering sperm unsuitable for further use. We aimed to determine whether an in-house artificial intelligence (AI) model could reliably assess normal sperm morphology in living sperm and compare its performance with that of computer-aided semen analysis and conventional semen analysis methods. In this experimental study, we enrolled 30 healthy male volunteers aged 18-40 years at the Songklanagarind Assisted Reproductive Centre, Songklanagarind Hospital. We developed a novel dataset of sperm morphological images captured with confocal laser scanning microscopy at low magnification and high resolution to train and validate an AI model. Semen samples were divided into three aliquots and assessed for unstained live sperm morphology using the AI model, whereas computer-aided and conventional semen analysis methods evaluated fixed sperm morphology. The performance of our in-house AI model for evaluating unstained live sperm morphology was compared with that of the other two methods. The in-house AI model showed the strongest correlation with computer-aided semen analysis (r = 0.88), followed by conventional semen analysis (r = 0.76). The correlation between computer-aided semen analysis and conventional semen analysis was weaker (r = 0.57). Both the in-house AI and conventional semen analysis methods detected normal sperm morphology at significantly higher rates than computer-aided semen analysis. The in-house AI model could enhance assisted reproductive technology outcomes by improving the selection of high-quality sperm with normal morphology. This could lead to better outcomes of intracytoplasmic sperm injections and other fertility treatments. Lay summary: We evaluated a new in-house AI model for assessing the shape and size (morphology) of live sperm without staining and performed comparisons with computer-aided semen analysis and conventional semen analysis, which require sperm to be fixed and stained before analysis. This new method of assessing unstained, live sperm is significant because it facilitates viable sperm selection for use in assisted reproductive technology immediately after assessment, ultimately contributing to improved fertility outcomes. The AI model allowed sperm morphology assessments with significantly improved accuracy and reliability. By using high-resolution images and advanced microscopy, the AI model could detect subcellular features. This AI model could be an effective tool in clinical settings, because it minimizes subjectivity and improves sperm selection for assisted reproductive technologies, potentially leading to higher success rates in infertility treatments. Further research can refine the model and validate its effectiveness in diverse clinical environments.","author":[{"family":"Jaruenpunyasak","given":"Jermphiphut"},{"family":"Maneelert","given":"Prawai"},{"family":"Nawae","given":"Marwan"},{"family":"Choksuchat","given":"Chainarong"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1530/raf-25-0014","URL":"https://doi.org/10.1530/raf-25-0014","source":"openalex"},{"id":"oa:W4407408800","type":"article-journal","title":"Applications of Artificial Intelligence for the Prediction and Diagnosis of Cancer Therapy-Related Cardiac Dysfunction in Oncology Patients","abstract":"Cardiovascular diseases and cancer are the leading causes of morbidity and mortality in modern society. Expanding cancer therapies that have improved prognosis may also be associated with cardiotoxicity, and extended life span after survivorship is associated with the increasing prevalence of cardiovascular disease. As such, the field of cardio-oncology has been rapidly expanding, with an aim to identify cardiotoxicity and cardiac disease early in a patient who is receiving treatment for cancer or is in survivorship. Artificial intelligence is revolutionizing modern medicine with its ability to identify cardiac disease early. This article comprehensively reviews applications of artificial intelligence specifically applied to electrocardiograms, echocardiography, cardiac magnetic resonance imaging, and nuclear imaging to predict cardiac toxicity in the setting of cancer therapies, with a view to reduce early complications and cardiac side effects from cancer therapies such as chemotherapy, radiation therapy, or immunotherapy.","author":[{"family":"Scalia","given":"Isabel"},{"family":"Pathangey","given":"Girish"},{"family":"Abdelnabi","given":"Mahmoud"},{"family":"Ibrahim","given":"Omar"},{"family":"Abdelfattah","given":"Fatmaelzahraa"},{"family":"Pereyra","given":"Milagros"},{"family":"Ibrahim","given":"Ramzi"},{"family":"Farina","given":"Juan"},{"family":"Banerjee","given":"Imon"},{"family":"Tamarappoo","given":"Balaji"},{"family":"Arsanjani","given":"Reza"},{"family":"Ayoub","given":"Chadi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/cancers17040605","URL":"https://doi.org/10.3390/cancers17040605","source":"openalex"},{"id":"oa:W4409650920","type":"article-journal","title":"Acromegaly facial changes analysis using last generation artificial intelligence methodology: the AcroFace system","abstract":"PURPOSE: To describe the development of the AcroFace system, an AI-based system for early detection of acromegaly, based on facial photographs analysis. METHODS: Two types of features were explored: (1) the visual/texture of a set of 2D facial images, and (2) geometric information obtained from a reconstructed 3D model from a single image. We optimized acromegaly detection by integrating SVM for geometric features and CNNs for visual features, each chosen for their strength in processing distinct data types effectively. This combination enhances overall accuracy by leveraging SVM's capability to manage structured, quantitative data and CNNs' proficiency in interpreting complex image textures, thus providing a comprehensive analysis of both geometric alignment and textural anomalies. ResNet-50, VGG-16, MobileNet, Inception V3, DensNet121 and Xception models were trained with an expert endocrinologist-based score as a ground truth. RESULTS: ResNet-50 model as a feature extractor and Support Vector Regression (SVR) with a linear kernel showed the best performance (accuracy δ1 of 75% and δ3 of 89%), followed by the VGG-16 as a feature extractor and SVR with a linear kernel. Geometric features yield less accurate results than visual ones. The validation cohort showed the following performance: precision 0.90, accuracy 0.93, F1-Score 0.92, sensitivity 0.93 and specificity 0.93. CONCLUSION: AcroFace system shows a good performance to discriminate acromegaly and non-acromegaly facial traits that may serve for the detection of acromegaly at an early stage as a screening procedure at a population level.","author":[{"family":"Rashwan","given":"Hatem"},{"family":"Marqués-Pamies","given":"Montserrat"},{"family":"Ruiz","given":"Sabina"},{"family":"Gil","given":"Joan"},{"family":"Asensio-Wandosell","given":"Diego"},{"family":"Martínez-Momblán","given":"Maria"},{"family":"Vázquez","given":"Federico"},{"family":"Salinas","given":"Isabel"},{"family":"Ciriza","given":"Raquel"},{"family":"Jordà","given":"Mireia"},{"family":"Chanson","given":"Philippe"},{"family":"Valassi","given":"Elena"},{"family":"Abdelnasser","given":"Mohamed"},{"family":"Puig","given":"Domènec"},{"family":"Puigdomingo","given":"Manel"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s11102-025-01515-2","URL":"https://doi.org/10.1007/s11102-025-01515-2","source":"openalex"},{"id":"oa:W7127048854","type":"article-journal","title":"Mapping the landscape of AI-assisted formative feedback in medical education: A bibliometric analysis","abstract":"BACKGROUND: Artificial intelligence (AI) is transforming medical education, particularly in formative feedback. This study conducted a comprehensive bibliometric analysis to map the intellectual landscape, research trends, and future directions of AI-assisted formative feedback in medical education. METHODS: A systematic search was performed in the Web of Science Core Collection database for English-language articles published between January 1, 2021, and October 12, 2025. Bibliometric analysis was conducted using VOSviewer, CiteSpace, and the R-bibliometrix package to analyze publication trends, geographic distribution, institutional collaboration, journal impact, author contributions, co-citation patterns, and keyword occurrences. RESULTS: The analysis included 116 publications, revealing exponential growth in AI-assisted formative feedback research from 2021 to 2025. The United States dominated the research landscape, followed by China and other European nations. Institutional collaboration centered on the University of Michigan, connecting North American and Asian research clusters. BMC Medical Education emerged as the leading journal, while interdisciplinary knowledge flow originated from clinical medicine and drew on health sciences and educational psychology. Keyword bursts identified \"feedback,\" \"large language model,\" and \"medical education\" as the most prominent research hotspots in 2024 to 2025. CONCLUSION: AI-assisted formative feedback in medical education is a rapidly evolving field driven by advancements in large language models, immersive technologies, and personalized assessments. Future research should prioritize theory-informed, ethically grounded, and patient-oriented AI integration to augment human instruction and demonstrably improve learner competence and patient care. Increased international collaboration and interdisciplinary knowledge exchange are crucial for the responsible adoption of AI in medical education.","author":[{"family":"Yu","given":"Sha"},{"family":"Liu","given":"Jin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1097/md.0000000000047489","URL":"https://doi.org/10.1097/md.0000000000047489","source":"pubmed"},{"id":"oa:W4415300354","type":"article-journal","title":"Artificial intelligence for maxillofacial prosthodontics: A technological shift in craniofacial rehabilitation- a scoping review","abstract":"Introduction: Artificial intelligence (AI) transforms dentistry and holds considerable promise for maxillofacial prosthodontics (MFP). Applications in imaging, computer-aided design and manufacturing (CAD/CAM), and additive manufacturing are improving diagnosis, treatment planning, and prosthetic rehabilitation for patients with craniofacial abnormalities. Despite advances in materials and digital workflows, challenges remain in achieving optimal accuracy, efficiency, and customisation in prosthetic design. The integration of AI in maxillofacial prosthodontics is still in its early stages. Currently, there is no review detailing the scope, trends, potential, and limitations of AI in this field. A scoping review is therefore necessary to consolidate existing evidence, identify knowledge gaps, and suggest directions for future research and clinical application. This review objective is to systematically map and analyse the current literature on AI in maxillofacial prosthodontics, focusing on its role in craniofacial rehabilitation. Methods: This scoping review adhered to the methodological framework of Arksey and O'Malley (2005) and was guided by the Joanna Briggs Institute (JBI) Manual for Evidence Synthesis (2020). Reporting complied with PRISMA-ScR (Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews) guidelines to ensure clarity and reproducibility. The review was registered with the Open Science Framework (registration number: www.osf.io/3b9jr). Electronic databases, including Medline via PubMed, Scopus, Cochrane Database, Science Direct, Google Scholar, and Semantic Scholar, were searched up to 7 June 2025. Full-text English articles containing the keywords \"Artificial Intelligence and Maxillofacial Prosthodontics\" and related terms were included. Results: This scoping review included 35 articles from diverse geographic regions. The studies addressed several specific applications of AI in maxillofacial prosthodontics, including the production of implant-supported auricular prostheses, coloration of maxillofacial prostheses, evaluation of facial attractiveness in patients with clefts, capture of 3D impressions of cleft palates, identification of hypernasality, assessment of lip symmetry, and detection of teeth in cleft lip and palate cases Conclusion: Artificial intelligence offers significant opportunities for maxillofacial prosthodontics, especially in imaging, digital design, and prosthesis production. Progress in this area requires interdisciplinary teamwork, large-scale clinical trials, and the development of standardized validation methods to ensure safe and effective clinical application.","author":[{"family":"Aradya","given":"Anupama"},{"family":"Sravani","given":"Koduru"},{"family":"Ravi","given":"MB"},{"family":"Swamy","given":"Kn"},{"family":"Ganesh","given":"Sajaysurya"},{"family":"Sowmya","given":"HK"},{"family":"Jayashankar","given":"Bindhya"},{"family":"Kumar","given":"Nisarga"},{"family":"Sangeeta","given":"Kenchappanavarmaheshappa"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jobcr.2025.10.006","URL":"https://doi.org/10.1016/j.jobcr.2025.10.006","source":"openalex"},{"id":"oa:W4410723968","type":"article-journal","title":"Research-based clinical deployment of artificial intelligence algorithm for prostate MRI","abstract":"PURPOSE: A critical limitation to deployment and utilization of Artificial Intelligence (AI) algorithms in radiology practice is the actual integration of algorithms directly into the clinical Picture Archiving and Communications Systems (PACS). Here, we sought to integrate an AI-based pipeline for prostate organ and intraprostatic lesion segmentation within a clinical PACS environment to enable point-of-care utilization under a prospective clinical trial scenario. METHODS: A previously trained, publicly available AI model for segmentation of intra-prostatic findings on multiparametric Magnetic Resonance Imaging (mpMRI) was converted into a containerized environment compatible with MONAI Deploy Express. An inference server and dedicated clinical PACS workflow were established within our institution for evaluation of real-time use of the AI algorithm. PACS-based deployment was prospectively evaluated in two phases: first, a consecutive cohort of patients undergoing diagnostic imaging at our institution and second, a consecutive cohort of patients undergoing biopsy based on mpMRI findings. The AI pipeline was executed from within the PACS environment by the radiologist. AI findings were imported into clinical biopsy planning software for target definition. Metrics analyzing deployment success, timing, and detection performance were recorded and summarized. RESULTS: In phase one, clinical PACS deployment was successfully executed in 57/58 cases and were obtained within one minute of activation (median 33 s [range 21-50 s]). Comparison with expert radiologist annotation demonstrated stable model performance compared to independent validation studies. In phase 2, 40/40 cases were successfully executed via PACS deployment and results were imported for biopsy targeting. Cancer detection rates for prostate cancer were 82.1% for ROI targets detected by both AI and radiologist, 47.8% in targets proposed by AI and accepted by radiologist, and 33.3% in targets identified by the radiologist alone. CONCLUSIONS: Integration of novel AI algorithms requiring multi-parametric input into clinical PACS environment is feasible and model outputs can be used for downstream clinical tasks.","author":[{"family":"Harmon","given":"Stephanie"},{"family":"Tetreault","given":"Jesse"},{"family":"Esengür","given":"Ömer"},{"family":"Qin","given":"Ming"},{"family":"Yılmaz","given":"Enis"},{"family":"Chang","given":"Victor"},{"family":"Yang","given":"Dong"},{"family":"Xu","given":"Ziyue"},{"family":"Cohen","given":"Gregg"},{"family":"Plum","given":"Jeff"},{"family":"Sherif","given":"Testi"},{"family":"Levin","given":"RJ"},{"family":"Schmidt-Richberg","given":"Alexander"},{"family":"Thompson","given":"Scott"},{"family":"Coons","given":"Samuel"},{"family":"Chen","given":"Te"},{"family":"Choyke","given":"Peter"},{"family":"Xu","given":"Daguang"},{"family":"Gurram","given":"Sandeep"},{"family":"Wood","given":"Bradford"},{"family":"Pinto","given":"Peter"},{"family":"Turkbey","given":"Baris"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00261-025-05014-7","URL":"https://doi.org/10.1007/s00261-025-05014-7","source":"openalex"},{"id":"oa:W4411005770","type":"article-journal","title":"Detection of emergency department patients at risk of dementia through artificial intelligence","abstract":"INTRODUCTION: The study aimed to develop and validate the Emergency Department Dementia Algorithm (EDDA) to detect dementia among older adults (65+) and support clinical decision-making in the emergency department (ED). METHODS: In a multisite retrospective study of 759,665 ED visits, electronic health record data from Yale New Haven Health (2014-2022) were used to train three supervised and semi-unsupervised positive-unlabeled machine learning models (XGBoost, Random Forest, LASSO). A separate test set of 400 ED encounters underwent adjudicated chart review for validation. RESULTS: EDDA achieved an area under the receiver-operating characteristic curve (AUROC) of 0.85 in the test set and 0.93 in the validation set. Positive-unlabeled learning improved performance. Agreement between EDDA and clinician-adjudicated dementia diagnoses was moderate (kappa = 0.50), with 17% of EDDA-positive patients having undiagnosed probable dementia. DISCUSSION: EDDA enhances dementia detection in the ED, with potential for real-time implementation to improve patient outcomes and care transitions. HIGHLIGHTS: Developed a machine learning algorithm using electronic health record data to detect dementia in the emergency department (ED). Algorithm designed to balance detection accuracy with ease of ED implementation. Parsimonious model with limited but predictive variables selected for rapid ED use. Focused on real-time application, optimizing ED workflows, and clinician support. Aims to enhance ED dementia detection, patient safety, and care coordination.","author":[{"family":"Cohen","given":"Inessa"},{"family":"Taylor","given":"Richard"},{"family":"Xue","given":"Haipeng"},{"family":"Faustino","given":"Isaac"},{"family":"Festa","given":"Natalia"},{"family":"Brandt","given":"Cynthia"},{"family":"Gao","given":"Emily"},{"family":"Han","given":"Ling"},{"family":"Khasnavis","given":"Siddarth"},{"family":"Lai","given":"James"},{"family":"Mecca","given":"Adam"},{"family":"Sapre","given":"Atharva"},{"family":"Young","given":"Juan"},{"family":"Zanchelli","given":"Michael"},{"family":"Hwang","given":"Ula"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/alz.70334","URL":"https://doi.org/10.1002/alz.70334","source":"openalex"},{"id":"oa:W7122626431","type":"article-journal","title":"On the Possibility of Using Artificial Intelligence in Medicine: From Theory to Practice","abstract":"Modern life is inextricably linked with the latest technologies. Artificial intelligence (AI) poses a new challenge to humanity, the application of which affects all areas of life, including medicine. This article examines the potential application of AI in medical practice, particularly in ophthalmology. It presents an example of how AI can be used to determine risk factors for dry eye syndrome in patients undergoing cosmetic procedures in the periorbital area. It also analyzes the limitations of AI in medicine. An analysis of the literature demonstrates that the use of AI in scientific and medical practice has opened up a wide range of opportunities for conducting research at a new technological level, such as screening examinations, image-based diagnostics, and disease prediction; selection of optimal drug dosages; mitigating the threat of pandemics; and automation and precision of surgical interventions. When integrating AI technologies into medical practice, it’s important to consider a wide range of ethical issues, including potential breaches of confidentiality, transparency, and the reliability of information received. Frequent use of chatbots can lead to errors and the dequalification of physicians, especially those with limited clinical experience, as well as disruption of doctor-patient communication. Furthermore, it’s important to consider legal and forensic issues, primarily the question of who will bear responsibility for making decisions. Given the above, in our view, a personalized approach to treating each individual patient remains a priority in everyday clinical practice. This approach takes into account not only objective indicators but also anamnestic data, the body’s individual responses to treatment, and psycho-emotional aspects, as well as the physician’s fundamental knowledge and experience.","author":[{"family":"Trubilin","given":"VN"},{"family":"Polunina","given":"EG"},{"family":"Kurenkov","given":"VV"},{"family":"Trubilin","given":"AV"},{"family":"Kechin","given":"EV"},{"family":"Kasparova","given":"EA"},{"family":"Arabadzhyan","given":"SI"},{"family":"Filonenko","given":"AV"},{"family":"Ponomareva","given":"EN"},{"family":"Tsaregorodtseva","given":"MA"}],"issued":{"date-parts":[[2025]]},"DOI":"10.18008/1816-5095-2025-4-725-731","URL":"https://doi.org/10.18008/1816-5095-2025-4-725-731","source":"openalex"},{"id":"oa:W4412775165","type":"article-journal","title":"Artificial intelligence in acupuncture: bridging traditional knowledge and precision integrative medicine","abstract":"The integration of artificial intelligence (AI) into acupuncture research is accelerating the transformation of this traditional, experience-based practice into a data-driven, precision discipline. This review synthesizes recent advances in AI-enabled outcome prediction techniques, encompassing deep learning, meta-analytic modeling, natural language processing (NLP), computer vision, and neuroimaging-based analysis. For instance, convolutional neural networks (CNNs) have been successfully applied to classify tongue images and detect ZHENG patterns, while transformer-based NLP models enable automated extraction of clinical knowledge from classical texts. These technologies improve diagnostic objectivity, standardize treatment planning, and facilitate individualized care by enabling longitudinal efficacy modeling and real-time monitoring. Despite their potential, current implementations are constrained by limited and heterogeneous datasets, annotation variability, and gaps in clinical validation. We analyze key methodological innovations and challenges, and recommend future directions including the construction of federated multimodal data platforms, development of explainable AI frameworks, and promotion of open science practices. This convergence of AI and acupuncture presents a unique opportunity to enhance scientific rigor, clinical utility, and global integration of acupuncture within the paradigm of precision integrative medicine.","author":[{"family":"Hou","given":"Guoliang"},{"family":"Dong","given":"Bao"},{"family":"Yu","given":"Boyang"},{"family":"Dai","given":"Jian"},{"family":"Lin","given":"Xingxing"},{"family":"Cheng","given":"Zixin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fmed.2025.1633416","URL":"https://doi.org/10.3389/fmed.2025.1633416","source":"openalex"},{"id":"oa:W4416770028","type":"article-journal","title":"Utilization of artificial intelligence in prostate cancer detection: a comprehensive review of innovations in screening and diagnosis","abstract":"Prostate cancer management has long been challenged by the limitations of traditional screening tools like PSA testing, which contribute to significant rates of overdiagnosis and overtreatment. While advanced imaging such as multiparametric MRI (mpMRI) has improved the diagnostic pathway, the integration of Artificial Intelligence (AI) is now catalyzing a paradigm shift across the entire continuum of care. This comprehensive review details the transformative role of AI in prostate cancer. In diagnostics, deep learning algorithms enhance the interpretation of mpMRI by improving lesion detection, segmentation, and risk stratification, thereby reducing unnecessary biopsies. In digital pathology, AI provides automated and consistent Gleason grading, minimizing inter-observer variability and refining prognostication. In the therapeutic domain, AI is crucial for personalizing treatment by streamlining radiotherapy planning through automated contouring, predicting patient outcomes and toxicity, and enabling the development of adaptive therapy strategies for advanced disease. Multimodal AI models that synthesize imaging, biomarker, and clinical data are creating robust predictive tools for superior clinical decision support. Despite formidable challenges related to prospective validation, data equity, and regulatory approval, AI is paving the way for a new standard of care characterized by greater precision, efficiency, and personalization.","author":[{"family":"Rajih","given":"Emad"},{"family":"Bakhsh","given":"Abdulaziz"},{"family":"Borhan","given":"Walaa"},{"family":"Alqahtani","given":"Saeed"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fimmu.2025.1670671","URL":"https://doi.org/10.3389/fimmu.2025.1670671","source":"openalex"},{"id":"oa:W4410498502","type":"article-journal","title":"Detection of carotid artery calcifications using artificial intelligence in dental radiographs: a systematic review and meta-analysis","abstract":"BACKGROUND: Carotid artery calcifications are important markers of cardiovascular health, often associated with atherosclerosis and a higher risk of stroke. Recent research shows that dental radiographs can help identify these calcifications, allowing for earlier detection of vascular diseases. Advances in artificial intelligence (AI) have improved the ability to detect carotid calcifications in dental images, making it a useful screening tool. This systematic review and meta-analysis aimed to evaluate how accurately AI methods can identify carotid calcifications in dental radiographs. MATERIALS AND METHODS: A systematic search in databases including PubMed, Scopus, Embase, and Web of Science for studies on AI algorithms used to detect carotid calcifications in dental radiographs was conducted. Two independent reviewers collected data on study aims, imaging techniques, and statistical measures such as sensitivity and specificity. A meta-analysis using random effects was performed, and the risk of bias was evaluated with the QUADAS-2 tool. RESULTS: Nine studies were suitable for qualitative analysis, while five provided data for quantitative analysis. These studies assessed AI algorithms using cone beam computed tomography (n = 3) and panoramic radiographs (n = 6). The sensitivity of the included studies ranged from 0.67 to 0.98 and specificity varied between 0.85 and 0.99. The overall effect size, by considering only one AI method in each study, resulted in a sensitivity of 0.92 [95% CI 0.81 to 0.97] and a specificity of 0.96 [95% CI 0.92 to 0.97]. CONCLUSIONS: The high sensitivity and specificity indicate that AI methods could be effective screening tools, enhancing the early detection of stroke and related cardiovascular risks. CLINICAL TRIAL NUMBER: Not applicable.","author":[{"family":"Arzani","given":"Sarah"},{"family":"Soltani","given":"Parisa"},{"family":"Karimi","given":"Ali"},{"family":"Yazdi","given":"Maryam"},{"family":"Ayoub","given":"Ashraf"},{"family":"Khurshid","given":"Zohaib"},{"family":"Galderisi","given":"Domenico"},{"family":"Devlin","given":"Hugh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12880-025-01719-9","URL":"https://doi.org/10.1186/s12880-025-01719-9","source":"openalex"},{"id":"oa:W4406124015","type":"article-journal","title":"Implementation of Artificial Intelligence in nursing education: Α Νarrative Review","abstract":"Background: As technological advancements continue to reshape various industries, the integration of AI in healthcare education emerges as a crucial facet in preparing future nursing professionals. This narrative review aims to elucidate the numerous ways AI technologies are being utilized in nursing education. Methodology: A search in two internet databases was conducted for relevant studies, using keywords. The inclusion criteria encompassed studies published within the last 5 years, written in English, and focused on the integration of AI technologies in nursing education settings. The selected articles underwent a systematic screening process. Results: Of the 523 papers retrieved, 7 were included in the final synthesis. These studies evaluate the implementation of AI methods in undergraduate nursing students. The AI method usually used was a Chatbot. In 4 studies, a 3D avatar was incorporated into the AI tool to serve as a Virtual Patient. The studies focused on various learning objectives, with 4 studies emphasizing communication skills enhancement. The remaining 3 studies used the AI tool to assess students' knowledge and clinical skills. Clinical scenarios were predominantly used, and in studies with a 3D avatar, scenarios addressed theoretical knowledge, critical thinking, and decision-making in escalating clinical conditions. Endpoints of AI implementation were assessed using self-reported questionnaires, interview and direct feedback from the Chatbot. Consistent endpoints included students' self-efficacy, knowledge of the learning objective, students' satisfaction and attitudes toward the learning style. Conclusions: As technology continues to advance, the potential for AI in nursing education is becoming increasingly evident. Given that nursing is an interactive science, it seems that AI Chatbots are more useful in nursing education. Further AI implementation will enrich our understanding of how its integration will serve nursing education.","author":[{"family":"Kouka","given":"Aikaterini"},{"family":"Giannelou","given":"Evangelia"},{"family":"Konstantinidis","given":"Kleanthis"},{"family":"Apostolakis","given":"Ioannis"}],"issued":{"date-parts":[[2025]]},"DOI":"10.12681/healthresj.36795","URL":"https://doi.org/10.12681/healthresj.36795","source":"openalex"},{"id":"oa:W4409752426","type":"article-journal","title":"Artificial intelligence and physician burnout: A productivity paradox","abstract":"Introduction: Physician burnout persists in the American healthcare system. In part, this burnout is believed to be driven by the Electronic Health Record (EHR) and its fraught role in the clinical work of physicians. Artificial intelligence (AI)-enabled healthcare technologies are often promoted on the basis of their promise to reduce burnout by introducing efficiencies into clinical work, particularly related to EHR utilization and documentation. Where documentation is perceived as the problem, AI scribes are offered as the solution. Methods: This essay looks closely at existing studies of AI scribes in clinical context and draws upon experience and understanding of healthcare delivery and the EHR to anticipate how AI may related to provider burnout. Results: We find that it is premature to assert that AI tools will reduce physician burnout. Considering the integration of AI scribes into Learning Health Systems healthcare delivery becomes a starting point for understanding the challenges faced in safely adopting AI tools more generally, with attention to the healthcare workforce and patients. Conclusion: It is not a foregone conclusion that AI-enabled healthcare technologies, in their current state and application, will lead to improved healthcare delivery and reduced burnout. Instead, this is an open question that demands rigorous evaluation and high standards of evidence before we restructure the work of physicians and redefine the care of our patients.","author":[{"family":"Goodson","given":"David"},{"family":"Garcia","given":"Brittany"},{"family":"Hogarth","given":"Michael"},{"family":"Tu","given":"Shin‐ping"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/lrh2.70013","URL":"https://doi.org/10.1002/lrh2.70013","source":"openalex"},{"id":"oa:W4417416460","type":"article-journal","title":"Impact of generative AI in medical education in India: a systematic review","abstract":"Background: The advent of generative Artificial Intelligence (AI) has presented a fundamental change in the approach to medical education across the world. In India, where the medical education is facing a shortage in faculties and resources, generative AI (GenAI) has the potential of transforming this. This systematic review summarizes the current evidence on the impact, student readiness, and various ethical challenges and barriers of integration of AI into the medical curriculum. Methods: We followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines and searched published articles in PubMed and Google Scholar from 2020 to 2025. The search yielded 19,777 articles, from which 11 studies focusing on Indian medical students were selected. The findings of these studies were analyzed using Laurillard's six learning modes to gain a comprehensive pedagogical understanding. Result: Our study revealed a significant finding: while high awareness and positive perception towards AI have been shown by Indian medical students, most of the students lack formal training. These selected studies show that the students mostly use generative AI for clearing doubts, making assignments, and self-directed learning, shifting from the 'Acquisition' to 'Inquiry' and 'Production' modes of Laurillard's learning. Comparative Analysis showed that GenAI tools outperform students on standard exams, thus showing their potential. However, certain challenges also exist, including the risk of misinformation, over-reliance, potential decrease in critical thinking, and ethical concerns of data privacy. Conclusion: Indian medical students are enthusiastically adopting GenAI, but their engagement is mostly unstructured and informal. A significant gap exists between the readiness of the students and the medical institutions. To maximize the potential use of GenAI, our institutions have to develop a structured curriculum, invest in faculty training, and establish ethical guidelines. Teamwork between policymakers, educators, and researchers is the need of the hour so that our future physicians will be ready to integrate AI-enabled healthcare. Syestematic review registration: https://doi.org/10.17605/OSF.IO/2MJVK.","author":[{"family":"Mateen","given":"Azfar"},{"family":"Kumar","given":"Visesh"},{"family":"Singh","given":"Ajay"},{"family":"Yadav","given":"Berendra"},{"family":"Mahto","given":"Mala"},{"family":"Hassan","given":"Atiq"},{"family":"Nasir","given":"Nazim"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/frai.2025.1704785","URL":"https://doi.org/10.3389/frai.2025.1704785","source":"openalex"},{"id":"oa:W4413308457","type":"article-journal","title":"Artificial Intelligence in Assessing Reproductive Aging: Role of Mitochondria, Oxidative Stress, and Telomere Biology","abstract":"Fertility potential ever more diminishes due to the complex, multifactorial, and still not entirely clarified process of reproductive aging in women and men. Gamete quality and reproductive lifespan are compromised by biologic factors like mitochondrial dysfunction, increased oxidative stress (OS), and incremental telomere shortening. Clinically confirmed biomarkers, including follicle-stimulating hormone (FSH) and anti-Müllerian hormone (AMH), are used to estimate ovarian reserve and reproductive status, but these markers have limited predictive validity and an incomplete representation of the complexity of reproductive age. Recent advances in artificial intelligence (AI) have the capacity to address the integration and interpretation of disparate and complex sets of data, like imaging, molecular, and clinical, for consideration. AI methodologies that improve the accuracy of reproductive outcome predictions and permit the construction of personalized treatment programs are machine learning (ML) and deep learning. To promote fertility evaluations, here, as part of its critical discussion, the roles of mitochondria, OS, and telomere biology as latter-day biomarkers of reproductive aging are presented. We also address the current status of AI applications in reproductive medicine, promises for the future, and applications involving embryo selection, multi-omics set integration, and estimation of reproductive age. Finally, to ensure that AI technology is used ethically and responsibly for reproductive care, model explainability, heterogeneity of data, and other ethical issues remain as residual concerns.","author":[{"family":"Moustakli","given":"Efthalia"},{"family":"Grigoriadis","given":"Themos"},{"family":"Stavros","given":"Sofoklis"},{"family":"Potiris","given":"Anastasios"},{"family":"Zikopoulos","given":"Athanasios"},{"family":"Gerede","given":"Angeliki"},{"family":"Tsimpoukis","given":"Ioannis"},{"family":"Papageorgiou","given":"Charikleia"},{"family":"Louis","given":"Konstantinos"},{"family":"Domali","given":"Ekaterini"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/diagnostics15162075","URL":"https://doi.org/10.3390/diagnostics15162075","source":"openalex"},{"id":"oa:W7131788581","type":"article-journal","title":"Narrative review of the ethics of artificial intelligence: are we ready for artificial intelligence in surgery?","abstract":"Background and Objective: Artificial intelligence (AI) is transforming surgical care by enhancing clinical decision-making and providing intraoperative guidance. As its applications expand, ethical challenges arise, including algorithmic bias, transparency in AI reasoning, informed consent regarding AI involvement, and accountability surrounding AI-guided decisions. This review explores the readiness of the surgical community to address these issues at both the institutional and individual levels. Methods: A PubMed search identified literature on AI in surgery published between 2018-2025. Fourteen key studies were selected and reviewed to assess AI applications across the surgical continuum, with attention to ethical considerations and barriers to integration. Key Content and Findings: AI now supports surgical care from the preoperative diagnosis through postoperative recovery. AI can outperform or match clinician performance in tumor detection, disease diagnosis, and surgical risk stratification. Predictive models using deep learning can outperform traditional scoring systems for perioperative and postoperative complication risk. Intraoperatively, AI enables surgical phase recognition, augmented reality guidance, and detection of technical errors. Despite these benefits, ethical concerns remain. Algorithmic bias may underestimate the needs of marginalized populations. Furthermore, questions of legal liability arise when AI-guided decisions cause harm. Informed consent must now address AI's role, limitations, and potential consequences if declined. Surgeons must guard against \"automation bias\" to preserve human judgment and patient trust. Institutional readiness remains unsatisfactory, as many healthcare systems lack infrastructure for real-time data integration and governance over data ownership. Surgeon skepticism and the \"black box\" nature of models also hinder adoption of the technology. Education on AI's design, validation, and biases is essential for safe integration. Conclusions: While AI holds immense potential to enhance surgical care, its use should be grounded in ethical principles to ensure non-maleficence and justice. Adoption should aim to maximize beneficence while preserving patient autonomy through transparent consent and promoting equity in access and implementation. At the same time, surgeons must remain vigilant against automation bias such that AI supports, not replaces, clinical intuition and trust, which lies at the core of the surgeon-patient relationship.","author":[{"family":"Yu","given":"Erin"},{"family":"Rosenberg","given":"Graeme"},{"family":"Udelsman","given":"Brooks"},{"family":"Harano","given":"Takashi"},{"family":"Atay","given":"Scott"},{"family":"Kim","given":"Anthony"},{"family":"Shakhsheer","given":"Baddr"},{"family":"Wightman","given":"Sean"}],"issued":{"date-parts":[[2026]]},"DOI":"10.21037/jtd-2025-1814","URL":"https://doi.org/10.21037/jtd-2025-1814","source":"openalex"},{"id":"oa:W7117721922","type":"article-journal","title":"Artificial Intelligence Applications in the Diagnosis, Treatment, and Prognosis of Hepatocellular Carcinoma","abstract":"The global burden of hepatocellular carcinoma (HCC) has shifted from viral to nonviral etiologies. However, successful antiviral therapy does not fully eliminate the risk of HCC, underscoring the demand for more effective surveillance strategies. Current screening methods, such as semiannual ultrasonography and the measurement of α-fetoprotein levels, offer suboptimal sensitivity for early detection. A cost-effective, reliable surveillance approach remains an unmet need. The Barcelona Clinic Liver Cancer staging system provides a framework to guide HCC therapy; yet, some gray zone exists, particularly for patients with intermediate-stage disease. Although tyrosine kinase inhibitors and immunotherapies have transformed the therapeutic landscape, their efficacies vary among patients, highlighting the necessity for personalized treatment strategies. In response to these challenges, artificial intelligence (AI) approaches have emerged as transformative tools in healthcare. By processing complex, nonlinear relationships and uncovering hidden patterns in clinical data, AI methods offer capabilities beyond those of traditional statistical methods. Furthermore, AI-driven multi-omics analysis holds promise for identifying novel biomarkers, thereby advancing precision medicine for HCC patients. This review introduces the potential of AI applications in enhancing the diagnosis, treatment, and prognosis of HCC.","author":[{"family":"Lu","given":"Ming"},{"family":"Wu","given":"Jacky"},{"family":"Lu","given":"Henry"},{"family":"Eslam","given":"Mohammed"},{"family":"Ming-Lung","given":"Yu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5009/gnl250268","URL":"https://doi.org/10.5009/gnl250268","source":"openalex"},{"id":"oa:W4406802063","type":"article-journal","title":"An Effectiveness Study of Generative Artificial Intelligence Tools Used to Develop Multiple-Choice Test Items","abstract":"Generative artificial intelligence (GenAI) tools developed to support teaching and learning are widely available. Trustworthiness concerns, however, have prompted calls for researchers to study their effectiveness and for educators and educational researchers to be involved in their creation and piloting processes. This study investigated one type of GenAI created to support educators: multiple-choice question generators (MCQ GenAI). Among the nine MCQ GenAI tools investigated, a variety of useful options were available, but only one indicated teacher involvement and none mentioned testing experts in development processes. MCQ GenAI-created items (n = 270) were coded based on MCQ quality item-writing guidelines. Results showed 80.00% of items (n = 216) violated at least one guideline, with 73.70% (n = 199) likely to produce major measurement error (should not use without revision), 6.30% (n = 17) likely to elicit minor measurement error (consider modifying), and 20.00% (n = 54) acceptable (usable as created). Implications suggest multidisciplinary teams are needed in educational GenAI tool development.","author":[{"family":"May","given":"Toni"},{"family":"Fan","given":"Yiyun"},{"family":"Stone","given":"Gregory"},{"family":"Koskey","given":"Kristin"},{"family":"Sondergeld","given":"Connor"},{"family":"Folger","given":"Timothy"},{"family":"Archer","given":"James"},{"family":"Provinzano","given":"Kathleen"},{"family":"Johnson","given":"Carla"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/educsci15020144","URL":"https://doi.org/10.3390/educsci15020144","source":"openalex"},{"id":"oa:W4411330206","type":"article-journal","title":"Reproducible generative artificial intelligence evaluation for health care: a clinician-in-the-loop approach","abstract":"Objectives: To develop and apply a reproducible methodology for evaluating generative artificial intelligence (AI) powered systems in health care, addressing the gap between theoretical evaluation frameworks and practical implementation guidance. Materials and Methods: A 5-dimension evaluation framework was developed to assess query comprehension and response helpfulness, correctness, completeness, and potential clinical harm. The framework was applied to evaluate ClinicalKey AI using queries drawn from user logs, a benchmark dataset, and subject matter expert curated queries. Forty-one board-certified physicians and pharmacists were recruited to independently evaluate query-response pairs. An agreement protocol using the mode and modified Delphi method resolved disagreements in evaluation scores. Results: Of 633 queries, 614 (96.99%) produced evaluable responses, with subject matter experts completing evaluations of 426 query-response pairs. Results demonstrated high rates of response correctness (95.5%) and query comprehension (98.6%), with 94.4% of responses rated as helpful. Two responses (0.47%) received scores indicating potential clinical harm. Pairwise consensus occurred in 60.6% of evaluations, with remaining cases requiring third tie-breaker review. Discussion: The framework demonstrated effectiveness in quantifying performance through comprehensive evaluation dimensions and structured scoring resolution methods. Key strengths included representative query sampling, standardized rating scales, and robust subject matter expert agreement protocols. Challenges emerged in managing subjective assessments of open-ended responses and achieving consensus on potential harm classification. Conclusion: This framework offers a reproducible methodology for evaluating health-care generative AI systems, establishing foundational processes that can inform future efforts while supporting the implementation of generative AI applications in clinical settings.","author":[{"family":"Livingston","given":"Leah"},{"family":"Featherstone-Uwague","given":"Amber"},{"family":"Barry","given":"AP"},{"family":"Barretto","given":"Kenneth"},{"family":"Morey","given":"Tara"},{"family":"Herrmannová","given":"Drahomíra"},{"family":"Avula","given":"Venkatesh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/jamiaopen/ooaf054","URL":"https://doi.org/10.1093/jamiaopen/ooaf054","source":"openalex"},{"id":"oa:W4414704819","type":"article-journal","title":"Raindrop optimizer: a novel nature-inspired metaheuristic algorithm for artificial intelligence and engineering optimization","abstract":"This paper presents a novel meta-heuristic optimization method, the Raindrop Algorithm (RD), inspired by natural raindrop phenomena, and explores its applications in artificial intelligence. The raindrop algorithm comprises two primary phases: exploration and exploitation. During the exploration phase, mechanisms including splash, diversion, and evaporation are employed to enhance global search capabilities. In the exploitation phase, raindrop convergence and overflow behaviors are simulated to improve local search performance. The algorithm demonstrates rapid convergence characteristics, typically achieving optimal solutions within 500 iterations while maintaining computational efficiency. The effectiveness and competitiveness of the raindrop algorithm have been validated on 23 benchmark functions and the CEC-BC-2020 benchmark suite, achieving first-place rankings in 76% of test cases. Specifically, on the CEC-BC-2020 benchmark, Wilcoxon rank-sum tests ([Formula: see text]) demonstrate statistically significant superiority in 94.55% of comparative cases. The raindrop algorithm has been successfully applied to optimize state estimation filters and controller parameters in robotic engineering problems, achieving an 18.5% reduction in position estimation error and a 7.1% improvement in overall filtering accuracy compared to conventional methods. Experimental results across five distinct engineering scenarios confirm the competitiveness and versatility of the raindrop algorithm, consistently maintaining top-three rankings in complex, nonlinear, and constrained optimization problems, thereby providing a promising solution for challenging optimization tasks in artificial intelligence-driven engineering environments.","author":[{"family":"Chen","given":"Shengjin"},{"family":"Yang","given":"Guangyong"},{"family":"Cui","given":"Guanghai"},{"family":"Dong","given":"Xiaoli"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-15832-w","URL":"https://doi.org/10.1038/s41598-025-15832-w","source":"openalex"},{"id":"oa:W4410443235","type":"article-journal","title":"Facial Analysis for Plastic Surgery in the Era of Artificial Intelligence: A Comparative Evaluation of Multimodal Large Language Models","abstract":"Background/Objectives: Facial analysis is critical for preoperative planning in facial plastic surgery, but traditional methods can be time consuming and subjective. This study investigated the potential of Artificial Intelligence (AI) for objective and efficient facial analysis in plastic surgery, with a specific focus on Multimodal Large Language Models (MLLMs). We evaluated their ability to analyze facial skin quality, volume, symmetry, and adherence to aesthetic standards such as neoclassical facial canons and the golden ratio. Methods: We evaluated four MLLMs—ChatGPT-4o, ChatGPT-4, Gemini 1.5 Pro, and Claude 3.5 Sonnet—using two evaluation forms and 15 diverse facial images generated by a Generative Adversarial Network (GAN). The general analysis form evaluated qualitative skin features (texture, type, thickness, wrinkling, photoaging, and overall symmetry). The facial ratios form assessed quantitative structural proportions, including division into equal fifths, adherence to the rule of thirds, and compatibility with the golden ratio. MLLM assessments were compared with evaluations from a plastic surgeon and manual measurements of facial ratios. Results: The MLLMs showed promise in analyzing qualitative features, but they struggled with precise quantitative measurements of facial ratios. Mean accuracy for general analysis were ChatGPT-4o (0.61 ± 0.49), Gemini 1.5 Pro (0.60 ± 0.49), ChatGPT-4 (0.57 ± 0.50), and Claude 3.5 Sonnet (0.52 ± 0.50). In facial ratio assessments, scores were lower, with Gemini 1.5 Pro achieving the highest mean accuracy (0.39 ± 0.49). Inter-rater reliability, based on Cohen’s Kappa values, ranged from poor to high for qualitative assessments (κ > 0.7 for some questions) but was generally poor (near or below zero) for quantitative assessments. Conclusions: Current general purpose MLLMs are not yet ready to replace manual clinical assessments but may assist in general facial feature analysis. These findings are based on testing models not specifically trained for facial analysis and serve to raise awareness among clinicians regarding the current capabilities and inherent limitations of readily available MLLMs in this specialized domain. This limitation may stem from challenges with spatial reasoning and fine-grained detail extraction, which are inherent limitations of current MLLMs. Future research should focus on enhancing the numerical accuracy and reliability of MLLMs for broader application in plastic surgery, potentially through improved training methods and integration with other AI technologies such as specialized computer vision algorithms for precise landmark detection and measurement.","author":[{"family":"Haider","given":"Syed"},{"family":"Prabha","given":"Srinivasagam"},{"family":"Gomez-Cabello","given":"Cesar"},{"family":"Borna","given":"Sahar"},{"family":"Genovese","given":"Ariana"},{"family":"Trabilsy","given":"Maissa"},{"family":"Elegbede","given":"Adekunle"},{"family":"Yang","given":"Jenny"},{"family":"Galvao","given":"Andrea"},{"family":"Tao","given":"Cui"},{"family":"Forte","given":"Antonio"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jcm14103484","URL":"https://doi.org/10.3390/jcm14103484","source":"openalex"},{"id":"oa:W4414211804","type":"article-journal","title":"Artificial intelligence in traditional medicine: evidence, barriers, and a research roadmap for personalized care","abstract":"Background: Traditional medicine (TM) systems such as Ayurveda, Traditional Chinese Medicine (TCM), and Thai Traditional Medicine (TTM) are increasingly intersecting with artificial intelligence (AI). Objective: To synthesize how AI is currently applied to TM and to outline barriers and research needs for safe, equitable, and scalable adoption. Methods: We conducted a targeted narrative mini review of peer reviewed studies (2017-Aug 2025) retrieved from PubMed, Scopus, and Google Scholar using terms spanning TM (Ayurveda/TCM/TTM) and AI (machine learning (ML), natural language processing (NLP), computer vision, telemedicine. Inclusion favored studies with reported methods and, when available, performance metrics; commentary and preprints without data were excluded. Findings: Current evidence supports AI assisted diagnostic pattern recognition, personalization frameworks integrating multi source data, digital preservation of TM knowledge, telemedicine enablement, and AI supported herbal pharmacology and safety assessment. Reported performance varies and is context dependent, with limited prospective external validation. Limitations: Evidence heterogeneity, small datasets, inconsistent ontologies across TM systems, and nascent regulatory pathways constrain real world deployment. Conclusion: AI can augment TM education, research, and clinical services, but progress requires standards, culturally informed datasets, prospective trials, and clear governance. We propose a research roadmap to guide rigorous and ethical integration.","author":[{"family":"Jongjiamdee","given":"Ketmanee"},{"family":"Pornwonglert","given":"Pimnipa"},{"family":"Bangchang","given":"Nutnichar"},{"family":"Akarasereenont","given":"Pravit"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/frai.2025.1659338","URL":"https://doi.org/10.3389/frai.2025.1659338","source":"openalex"},{"id":"oa:W4413817249","type":"article-journal","title":"An artificial intelligence cloud platform for OCT-based retinal anomalies screening system in real clinical environments","abstract":"Millions of individuals worldwide suffer from retinal anomalies, which can lead to irreversible vision loss. However, the number of ophthalmologists is highly mismatched with the population base in China, especially in many rural and underdeveloped towns. To tackle these challenges, this paper developed an Artificial Intelligence Cloud Platform for OCT-based Retinal Anomalies Screening (AI-PORAS), which is capable of detecting 15 retinal anomalies in OCT images to enhance remote diagnostic efficiency. AI-PORAS has been trained, validated, and deployed to 207 medical institutions in 29 provinces in China. The validation on 165,384 eyes with 3,551,959 OCT B-scan slices, AI-PORAS achieved an average accuracy of 93.16%, an AUC of 93.64%, a FPR of 6.82%, a FNR of 7.87%, matching the average performance level of attending ophthalmologists. Additionally, Statistical analysis of the 116,717 remotely diagnosed patients provided insightful guidance for healthcare decision-making and the development of tailored treatment plans.","author":[{"family":"Chen","given":"Xinjian"},{"family":"Wang","given":"Jiangtao"},{"family":"Qian","given":"Tianwei"},{"family":"Wang","given":"Jingcheng"},{"family":"Ding","given":"Yiming"},{"family":"Zhang","given":"Su"},{"family":"Liao","given":"Jingjing"},{"family":"Qian","given":"Cheng"},{"family":"Yang","given":"Ting"},{"family":"Mateen","given":"Muhammad"},{"family":"Fan","given":"Yu"},{"family":"Song","given":"Zongming"},{"family":"Chen","given":"Jili"},{"family":"Li","given":"Suyan"},{"family":"Hu","given":"Juejun"},{"family":"Yan","given":"Wentao"},{"family":"Chen","given":"Haoyu"},{"family":"Wu","given":"Wencan"},{"family":"Jing","given":"Huang"},{"family":"Wong","given":"Tien"},{"family":"Xu","given":"Xun"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01959-7","URL":"https://doi.org/10.1038/s41746-025-01959-7","source":"openalex"},{"id":"oa:W4412804424","type":"article-journal","title":"Measuring public opinion towards artificial intelligence: development and validation of a general AI attitude short scale","abstract":"Abstract The rapid proliferation of artificial intelligence (AI) has sparked both enthusiasm and ethical concerns in societies. As AI continues to permeate daily life, policymakers need to understand how it is perceived by diverse stakeholders and communities. To reliably measure attitudes towards AI of the general public, a short scale is essential for universal application. Existing scales face limitations in applicability due to their length, sub-standard internal consistency, or a focus on only negative attitudes. In response, we built up on existing scales and developed a unidimensional six-item general AI attitude short scale. First tests on internet panel data from Germany ( n = 1001) and the US ( n = 3091) obtained favorable results for classical test theory (CTT) and item response theory (IRT). Confirmatory factor analysis indicated an excellent fit for a single-factor structure, while the scale also exhibited strong criterion-related validity, correlating positively with digital competency and predicting acceptance of several AI applications. Additional IRT analyses suggested high item discrimination, broad coverage of the attitude spectrum and no meaningful differential item functioning (DIF). Thus, we propose a psychometrically sound short scale for measuring general AI attitude and provide insights into the antecedents and consequences of the construct.","author":[{"family":"Novotny","given":"Marcus"},{"family":"Weber","given":"Wiebke"},{"family":"Kern","given":"Christoph"},{"family":"Kreuter","given":"Frauke"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00146-025-02478-5","URL":"https://doi.org/10.1007/s00146-025-02478-5","source":"openalex"},{"id":"oa:W4417056660","type":"article-journal","title":"Artificial Intelligence in Prostate MRI : Addressing Current Limitations Through Emerging Technologies","abstract":"Prostate MRI has transformed lesion detection and risk stratification in prostate cancer, but its impact is constrained by the high cost of the exam, variability in interpretation, and limited scalability. False negatives, false positives, and moderate inter-reader agreement undermine reliability, while long acquisition times restrict throughput. Artificial intelligence (AI) offers potential solutions to address many of the limitations of prostate MRI in the clinical management pathway. Machine learning-based triage can refine patient selection to optimize resources. Deep learning reconstruction enables accelerated acquisition while preserving diagnostic quality, with multiple FDA-cleared products now in clinical use. Ongoing development of automated quality assessment and artifact correction aims to improve reliability by reducing nondiagnostic exams. In image interpretation, AI models for lesion detection and clinically significant prostate cancer prediction achieve performance comparable to radiologists, and the PI-CAI international reader study has provided the strongest evidence to date of non-inferiority at scale. More recent work extends MRI-derived features into prognostic modeling of recurrence, metastasis, and functional outcomes. This review synthesizes progress across five domains-triage, accelerated acquisition and reconstruction, image quality assurance, diagnosis, and prognosis-highlighting the level of evidence, validation status, and barriers to adoption. While acquisition and reconstruction are furthest along, with FDA-cleared tools and prospective evaluations, triage, quality control, and prognosis remain earlier in development. Ensuring equitable performance across populations, incorporating uncertainty estimation, and conducting prospective workflow trials will be essential to move from promising prototypes to routine practice. Ultimately, AI could accelerate the adoption of prostate MRI toward a scalable platform for earlier detection and population-level prostate cancer management. EVIDENCE LEVEL: N/A TECHNICAL EFFICACY: 3.","author":[{"family":"Johnson","given":"Patricia"},{"family":"Umapathy","given":"Lavanya"},{"family":"Gigax","given":"Bradley"},{"family":"Rossi","given":"Juan"},{"family":"Tong","given":"Angela"},{"family":"Bruno","given":"Mary"},{"family":"Sodickson","given":"Daniel"},{"family":"Nayan","given":"Madhur"},{"family":"Chandarana","given":"Hersh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/jmri.70189","URL":"https://doi.org/10.1002/jmri.70189","source":"openalex"},{"id":"oa:W4409584265","type":"article-journal","title":"Ethical challenges and regulatory pathways for artificial intelligence in rheumatology","abstract":"AI integration in rheumatology is facing unique ethical and regulatory challenges due to longitudinal patient data complexity. Dear Editor, The integration of artificial intelligence (AI) into rheumatology practice raises important ethical considerations that warrant careful attention [1–3]. The chronic and complex nature of rheumatologic conditions presents unique challenges for AI implementation that deserve focused discussion. Rheumatology stands apart from other medical specialties due to the longitudinal accumulation of patient data over years or decades of disease management. Patients with conditions like rheumatoid arthritis, systemic lupus erythematosus, and other autoimmune diseases generate vast amounts of data through regular monitoring of disease activity, medication responses and periodic flares. This longitudinal data presents both opportunities and ethical challenges for AI applications. The European Union (EU) AI Act [4] stands as the first comprehensive legislation specifically addressing AI systems in healthcare and other domains. While other jurisdictions rely on policy frameworks and guidelines, this legislation establishes binding requirements for AI development and deployment. This distinction is particularly relevant for rheumatology, where AI applications could involve complex decision support systems managing sensitive patient data over extended periods. Disease flare prediction exemplifies the complex interplay between AI capabilities and regulatory requirements in rheumatology. For instance, under the EU AI Act, such predictive systems would be classified as ‘high-risk’ AI since they influence medical decisions and patient outcomes [5, 6]. This classification brings significant regulatory obligations, including requirements for robust risk management systems, high-quality training data and detailed technical documentation. For flare prediction specifically, these requirements are crucial given the serious consequences of prediction errors. While early prediction could revolutionize pre-emptive treatment, a false positive prediction might lead to unnecessary treatment intensification with potential adverse effects, while a missed prediction could result in preventable organ damage. The high-risk designation means these systems must maintain rigorous logging capabilities and enable human oversight of predictions, allowing rheumatologists to understand and potentially override AI recommendations when clinically appropriate. The regulatory landscape for medical AI varies significantly across jurisdictions. The EU AI Act establishes a risk-based regulatory approach with binding legal requirements. In contrast, the US Food and Drug Administration implements a product-based classification system through regulatory guidelines, focusing on safety and effectiveness verification for AI-driven medical devices. The UK has adopted a distinctive sector-specific approach, combining targeted oversight through existing regulatory bodies like the Medicines and Healthcare products Regulatory Agency with new coordination mechanisms through the Digital Regulation Cooperation Forum. This is supported by the AI Safety Institute and upcoming AI Bill, maintaining flexibility while implementing mandatory requirements for high-risk systems. The implications of AI regulation for rheumatology extend beyond flare prediction. Any AI system used for medical diagnosis, patient triage or treatment planning falls under the high-risk category and must meet stringent requirements for accuracy, robustness and cybersecurity (Table 1). Healthcare providers deploying these systems must implement quality management systems, ensure ongoing monitoring and maintain detailed documentation of the AI system’s development and validation. Ethical and safety considerations for AI systems in rheumatology under the EU AI Act. Ethical and safety considerations for AI systems in rheumatology under the EU AI Act. Data security takes on heightened importanc","author":[{"family":"Venerito","given":"Vincenzo"},{"family":"Gupta","given":"Latika"},{"family":"Mileto","given":"Saverio"},{"family":"Iannone","given":"Florenzo"},{"family":"Bılgın","given":"Emre"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/rap/rkaf035","URL":"https://doi.org/10.1093/rap/rkaf035","source":"openalex"},{"id":"oa:W4411801984","type":"article-journal","title":"Assisted artificial intelligence in medical writing: a primer for humans","abstract":"Assisted artificial intelligence (A-AI) has rapidly become a gold standard approach for conducting data analysis and medical writing [1]. From executing complex statistical tasks to extracting relevant information and supporting manuscript drafting, A-AI is increasingly embedded in the daily workflow of clinical and translational researchers [2]. However, the thin red line between being assisted by AI and becoming reliant on it, or even fully driven by it, is becoming harder to define. This ambiguity raises essential questions about authorship, accountability and the integrity of scientific output in the AI era. A growing number of early-career researchers are engaging with high-dimensional machine learning (ML) methodologies, often without comprehensive training in classical statistics or computational foundations. While the integration of such advanced tools reflects a welcome shift towards data-driven research, it also raises concerns regarding methodological rigor. In certain instances, traditional statistical frameworks are overlooked, and AI-generated outputs, particularly those derived from large language models, are interpreted without sufficient critical appraisal. To better understand this, 2 examples of A-AI analysis follow. To assess the magnitude of post-procedural thrombocytopaenia after transcatheter aortic valve implantation (TAVI) and its potential interplay with subclinical leaflet thrombosis (SLT) as detected by 1-month multidetector computed tomography (MDCT), I turned to my ‘Assisted AI master’. Within seconds, the AI generated a synthetic dataset comprising 100 patients and 120 prespecified variables, including demographic, procedural and post-implantation metrics, based on the assumption that SLT would occur in 18% of the sample. To enhance realism, I requested a ‘real-world’ dataset with missing values and statistical outliers. My AI master not only accommodated these nuances but also proposed appropriate imputation strategies, including multiple imputation by chained equations. My assisted AI master returned the following dataset (synthetitic.csv) with the variables reported in Fig. 1 [3]. AI master was instructed to perform multiple imputation to address missing data. Variables’ name and description provided as synthetic data. Once the dataset was imputed, the AI executed an ML pipeline using the popular Extreme Gradient Boosting (XGBoost) algorithm to predict SLT [4]. Without any need for custom code or manual feature engineering, the model was trained and internally validated. To enhance interpretability, the AI automatically produced the R-code for a beeswarm plot of SHAP (SHapley Additive exPlanations) values, offering a visually compelling summary of the model’s top predictors and their direction of effect [5]. Within moments, a complete set of analytical outputs is provided and easily be mistaken for a results section ready for submission of for an abstract, for example (Fig. 2A). By interpreting the SHAP beeswarm plot, my A-AI master concluded that ‘Age’ was the most influential variable driving the model’s prediction of SLT. The entire process took only a few min. While the dataset is entirely synthetic, it remains a valuable resource for statistical training and method development. Both the dataset and the full R code are available for download in the public repository (https://github.com/mmlondon77/Biobook.). (A) SHAP plot showing which synthetic features have the biggest impact on the model’s predictions. (B) Forest plot of odds ratio by a multivariable logistic regression. BMI: body mass index; SHAP: SHapley Additive exPlanations. We aimed to explore whether colchicine is effective in preventing SLT at 1 month following TAVI. To do so, I turned to my AI master, and within min, a complete synthetic randomized clinical trial (RCT) was generated. The simulated RCT included 200 patients, randomized 1:1 to receive colchicine or placebo. Baseline and procedural variables were generated to reflec","author":[{"family":"Moscarelli","given":"Marco"},{"family":"Pollari","given":"Francesco"},{"family":"Franzese","given":"Ilaria"},{"family":"Barili","given":"Fabio"},{"family":"Marco","given":"Luca"},{"family":"Nenna","given":"Antonio"},{"family":"Salsano","given":"Antonio"},{"family":"Santarpino","given":"Giuseppe"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/ejcts/ezaf167","URL":"https://doi.org/10.1093/ejcts/ezaf167","source":"openalex"},{"id":"oa:W4410153852","type":"article-journal","title":"Exploring the complex nature of implementation of Artificial intelligence in clinical practice: an interview study with healthcare professionals, researchers and Policy and Governance Experts","abstract":"Artificial Intelligence (AI)-based tools have shown potential to optimize clinical workflows, enhance patient quality and safety, and facilitate personalized treatment. However, transitioning viable AI solutions to clinical implementation remains limited. To understand the challenges of bringing AI into clinical practice, we explored the experiences of healthcare professionals, researchers, and Policy and Governance Experts in hospitals. We conducted a qualitative study with thirteen semi-structured interviews (mean duration 52.1 ± 5.4 minutes) with healthcare professionals, researchers, and Policy and Governance Experts, with prior experience on AI development in hospitals. The interview guide was based on value, application, technology, governance, and ethics from the Innovation Funnel for Valuable AI in Healthcare, and the discussions were analyzed through thematic analysis. Six themes emerged: (1) demand-pull vs. tech-push: AI development focusing on innovative technologies may face limited success in large-scale clinical implementation. (2) Focus on generating knowledge, not solutions: Current AI initiatives often generate knowledge without a clear path for implementing AI models once proof-of-concept is achieved. (3) Lack of multidisciplinary collaboration: Successful AI initiatives require diverse stakeholder involvement, often hindered by late involvement and challenging communication. (4) Lack of appropriate skills: Stakeholders, including IT departments and healthcare professionals, often lack the required skills and knowledge for effective AI integration in clinical workflows. (5) The role of the hospital: Hospitals need a clear vision for integrating AI, including meeting preconditions in infrastructure and expertise. (6) Evolving laws and regulations: New regulations can hinder AI development due to unclear implications but also enforce standardization, emphasizing quality and safety in healthcare. In conclusion, this study highlights the complexity of AI implementation in clinical settings. Multidisciplinary collaboration is essential and requires facilitation. Balancing divergent perspectives is crucial for successful AI implementation. Hospitals need to assess their readiness for AI, develop clear strategies, standardize development processes, and foster better collaboration among stakeholders.","author":[{"family":"Leenen","given":"Jobbe"},{"family":"Hiemstra","given":"Ps"},{"family":"Hoeve","given":"Martine"},{"family":"Jansen","given":"Anouk"},{"family":"Dijk","given":"JDV"},{"family":"Vendel","given":"BN"},{"family":"Versteeg","given":"Guido"},{"family":"Hakvoort","given":"Gido"},{"family":"Hettinga","given":"Marike"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1371/journal.pdig.0000847","URL":"https://doi.org/10.1371/journal.pdig.0000847","source":"openalex"},{"id":"oa:W4411464271","type":"article-journal","title":"Artificial intelligence in imaging diagnosis of liver tumors: current status and future prospects","abstract":"Liver cancer remains a significant global health concern, ranking as the sixth most common malignancy and the third leading cause of cancer-related deaths worldwide. Medical imaging plays a vital role in managing liver tumors, particularly hepatocellular carcinoma (HCC) and metastatic lesions. However, the large volume and complexity of imaging data can make accurate and efficient interpretation challenging. Artificial intelligence (AI) is recognized as a promising tool to address these challenges. Therefore, this review aims to explore the recent advances in AI applications in liver tumor imaging, focusing on key areas such as image reconstruction, image quality enhancement, lesion detection, tumor characterization, segmentation, and radiomics. Among these, AI-based image reconstruction has already been widely integrated into clinical workflows, helping to enhance image quality while reducing radiation exposure. While the adoption of AI-assisted diagnostic tools in liver imaging has lagged behind other fields, such as chest imaging, recent developments are driving their increasing integration into clinical practice. In the future, AI is expected to play a central role in various aspects of liver cancer care, including comprehensive image analysis, treatment planning, response evaluation, and prognosis prediction. This review offers a comprehensive overview of the status and prospects of AI applications in liver tumor imaging.","author":[{"family":"Masatoshi","given":"Hori"},{"family":"Suzuki","given":"Yuki"},{"family":"Sofue","given":"Keitaro"},{"family":"Sato","given":"Junya"},{"family":"Nishigaki","given":"Daiki"},{"family":"Tomiyama","given":"Miyuki"},{"family":"Nakamoto","given":"Atsushi"},{"family":"Murakami","given":"Takamichi"},{"family":"Tomiyama","given":"Noriyuki"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00261-025-05059-8","URL":"https://doi.org/10.1007/s00261-025-05059-8","source":"openalex"},{"id":"oa:W4414900906","type":"article-journal","title":"Artificial intelligence–based quantification of breast arterial calcifications to predict cardiovascular morbidity and mortality","abstract":"BACKGROUND AND AIMS: Women are underdiagnosed and undertreated for cardiovascular disease (CVD). Automatic quantification of breast arterial calcification (BAC) on screening mammography can identify women at risk for CVD. This study aimed to determine whether artificial intelligence-based automatic quantification of BAC from screening mammograms predicts CVD and mortality beyond PREVENT scores in a large, racially diverse, multi-institutional population. METHODS: This retrospective cohort study included 123 762 women from two healthcare systems who had screening mammograms. Breast arterial calcification was quantified using a transformer-based neural network for segmentation. Breast arterial calcification severity was categorized as zero (0 mm2), mild (>0-10 mm2), moderate (>10-25 mm2), and severe (>25 mm2). Kaplan-Meier analysis, Cox proportional hazards, and Fine-Gray competing event models were used to examine the association between BAC and major adverse cardiovascular events (MACE). RESULTS: Breast arterial calcification was detected in 16.1% (internal cohort) and 20.6% (external cohort) of women and provided significant prognostic value incremental to the PREVENT score. In PREVENT adjusted models, a clear dose-response was observed. Compared with zero BAC, mild [internal: hazard ratio (HR) 1.32, 95% confidence interval (CI) 1.10-1.59; external: HR 1.28, 95% CI 1.17-1.39], moderate (internal: HR 1.75, 95% CI 1.23-2.50; external: HR 1.79, 95% CI 1.55-2.06), and severe BAC (internal: HR 3.29, 95% CI 2.15-5.05; external: HR 2.80, 95% CI 2.36-3.32) were all prognostic for any MACE. Each 1 mm2 increase in BAC conferred an additional 2%-3% risk for MACE (P < .001). CONCLUSIONS: Automatically quantified BAC is an independent predictor of MACE and mortality, adding prognostic value to the PREVENT score. This approach may provide an opportunistic cardiovascular risk assessment during routine mammography screening without additional radiation exposure to guide earlier and more effective preventive care for women.","author":[{"family":"Dapamede","given":"Theo"},{"family":"Khan","given":"Aisha"},{"family":"Joshi","given":"Vedant"},{"family":"Gershon","given":"Gabrielle"},{"family":"Li","given":"Frank"},{"family":"Chavoshi","given":"Mohammadreza"},{"family":"Brown-Mulry","given":"Beatrice"},{"family":"Isaac","given":"Rohan"},{"family":"Mansuri","given":"Aawez"},{"family":"Robichaux","given":"Chad"},{"family":"Ayoub","given":"Chadi"},{"family":"Arsanjani","given":"Reza"},{"family":"Sperling","given":"Laurence"},{"family":"Gichoya","given":"Judy"},{"family":"Assen","given":"Marly"},{"family":"Oneill","given":"WC"},{"family":"Banerjee","given":"Imon"},{"family":"Trivedi","given":"Hari"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1093/eurheartj/ehag128","URL":"https://doi.org/10.1093/eurheartj/ehag128","source":"openalex"},{"id":"oa:W4412141497","type":"article-journal","title":"New tools for diagnosis of primary immunodeficiencies: from awareness to artificial intelligence","abstract":"Primary immune deficiencies (PI) are rare diseases associated with frequent, severe infections, inflammatory and autoimmune diseases and/or cancer. Because of the variability in presentation, undiagnosed PI patients can be encountered by many different medical specialists. A lack of awareness of and the rarity of PI can lead to delayed diagnosis particularly among primary care physicians and non-immunology specialists. These delays can lead to irreversible sequelae, decreased quality of life and premature mortality. In this review, we describe two projects designed to decrease the time to diagnosis in PI patients: 1) the expert-driven PIDCAP project conducted in Spain to promote early diagnosis in the primary care setting, and 2) a multi-modal data-driven approach using artificial intelligence and machine learning to identify individuals at high risk for PI. Both approaches aim to create widely available tools to promote early diagnosis and treatment of PI. Initial results have been positive. Future directions include larger studies and potentially combining expert-driven and data-driven approaches.","author":[{"family":"Solerpalacín","given":"Pere"},{"family":"Rivière","given":"Jacques"},{"family":"Burns","given":"Siobhan"},{"family":"Rider","given":"Nicholas"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fimmu.2025.1593897","URL":"https://doi.org/10.3389/fimmu.2025.1593897","source":"openalex"},{"id":"oa:W4414206725","type":"article-journal","title":"Artificial Intelligence in Educational Technology: A Systematic Review of Datasets and Applications","abstract":"Artificial Intelligence (AI) has the potential to impact a diverse range of domains. For instance, AI for the education domain has received increasing interest with various applications, including predicting performance, curating learning materials, and automated assessment and feedback. Despite the developments, some imbalances appear in the literature; for example, traditional classrooms and non-scientific academic subjects received little attention. This survey provides a systematic review of the current trends in AI research for education, specifically addressing applications within secondary education (ages 11+) through to higher education (HE), and offers a detailed compilation of datasets and methods, facilitating a deeper understanding of the field and encouraging further investigation. It includes a thorough review of the datasets available to encourage and enable future research, development, and collaboration, as well as the establishment of performance benchmarks. Furthermore, this survey provides an overview of issues and problems arising from recent developments, which may aid policymakers in their decision-making and addressing ethical concerns and standards. For example, many AI in Education (AIEd) platforms are not grounded in educational theory. We also present several guidelines to aid future developments in AIEd, guiding long-term impactful projects and investments.","author":[{"family":"Topham","given":"Luke"},{"family":"Atherton","given":"Pete"},{"family":"Reynolds","given":"Tom"},{"family":"Hussain","given":"Yasir"},{"family":"Hussain","given":"Abir"},{"family":"Kolivand","given":"Hoshang"},{"family":"Khan","given":"Wasiq"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1145/3768312","URL":"https://doi.org/10.1145/3768312","source":"openalex"},{"id":"oa:W4413911574","type":"article-journal","title":"Recommendations for disclosure of artificial intelligence in scientific writing and publishing: a regional anesthesia and pain medicine modified Delphi study","abstract":"Introduction The use of artificial intelligence (AI) in the scientific process is advancing at a remarkable speed, thanks to continued innovations in large language models. While AI provides widespread benefits, including editing for fluency and clarity, it also has drawbacks, including fabricated content, perpetuation of bias, and lack of accountability. The editorial board of Regional Anesthesia & Pain Medicine (RAPM) therefore sought to develop best practices for AI usage and disclosure. Methods A steering committee from the American Society of Regional Anesthesia and Pain Medicine used a modified Delphi process to address definitions, disclosure requirements, authorship standards, and editorial oversight for AI use in publishing. The committee reviewed existing publication guidelines and identified areas of ambiguity, which were translated into questions and distributed to an expert workgroup of authors, reviewers, editors, and AI researchers. Results Two survey rounds, with 91% and 87% response rates, were followed by focused discussion and clarification to identify consensus recommendations. The workgroup achieved consensus on recommendations to authors about definitions of AI, required items to report, disclosure locations, authorship stipulations, and AI use during manuscript preparation. The workgroup formulated recommendations to reviewers about monitoring and evaluating the responsible use of AI in the review process, including the endorsement of AI-detection software, identification of concerns about undisclosed AI use, situations where AI use may necessitate the rejection of a manuscript, and use of checklists in the review process. Finally, there was consensus about AI-driven work, including required and optional disclosures and the use of checklists for AI-associated research. Discussion Our modified Delphi study identified practical recommendations on AI use during the scientific writing and editorial process. The workgroup highlighted the need for transparency, human accountability, protection of patient confidentiality, editorial oversight, and the need for iterative updates. The proposed framework enables authors and editors to harness AI’s efficiencies while maintaining the fundamental principles of responsible scientific communication and may serve as an example for other journals.","author":[{"family":"Fettiplace","given":"Michael"},{"family":"Bhatia","given":"Anuj"},{"family":"Chen","given":"Yian"},{"family":"Orebaugh","given":"Steven"},{"family":"Gofeld","given":"Michael"},{"family":"Gabriel","given":"Rodney"},{"family":"Sessler","given":"Daniel"},{"family":"Lonsdale","given":"Hannah"},{"family":"Bungart","given":"Brittani"},{"family":"Cheng","given":"Christopher"},{"family":"Burnett","given":"Garrett"},{"family":"Han","given":"Lichy"},{"family":"Wiles","given":"MD"},{"family":"Coppens","given":"Steve"},{"family":"Joseph","given":"Thomas"},{"family":"Schreiber","given":"Kristin"},{"family":"Volk","given":"Thomas"},{"family":"Urman","given":"Richard"},{"family":"Kovacheva","given":"Vesela"},{"family":"Wu","given":"Christopher"},{"family":"Mariano","given":"Edward"},{"family":"Ip","given":"Vivian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/rapm-2025-106852","URL":"https://doi.org/10.1136/rapm-2025-106852","source":"openalex"},{"id":"oa:W4416649115","type":"article-journal","title":"Changes in public perception of artificial intelligence in healthcare after exposure to ChatGPT","abstract":"The public perception of artificial intelligence (AI) in healthcare is key to its large-scale acceptance and implementation. This study investigated how exposure to ChatGPT changed public perception of AI in healthcare, using baseline and follow-up data from 5899 survey participants reporting their perception of AI in 2022 (before ChatGPT's launch) and 2024, and ChatGPT use in 2024. Multinomial multivariate logistic regression was used to model how exposure to ChatGPT use affected changes in perception of AI. At follow-up, 1195 individuals (20%) had been exposed to ChatGPT use, which was associated with higher odds of changing perception of AI to beneficial (OR 3.21 [95% CI: 2.34-4.40]) among individuals who were unsure at baseline, and lower odds of changing to uncertainty from more defined baseline perceptions. This study demonstrates the potential for reducing uncertainty and improving public perception of AI in healthcare through exposure to AI tools.","author":[{"family":"Isaksen","given":"Anders"},{"family":"Schaarup","given":"Jonas"},{"family":"Bjerg","given":"Lasse"},{"family":"Hulmán","given":"Ádám"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-02169-x","URL":"https://doi.org/10.1038/s41746-025-02169-x","source":"openalex"},{"id":"oa:W4414516182","type":"article-journal","title":"Ophthalmic drug discovery and development using artificial intelligence and digital health technologies","abstract":"Globally, drug discovery and development programs are complex, multi-decade long and prohibitively expensive. Artificial intelligence (AI) and other digital health technologies have the potential to enhance and accelerate each stage of drug discovery and development, from pre-clinical target identification to post-market repurposing, and even revolutionize the entire process. Using ophthalmology as an example, this review highlights recent AI and digital health innovations in different phases of drug discovery and development. By leveraging machine learning algorithms and vast clinical and multiomics datasets, AI can rapidly identify and validate new drug targets, optimize lead compounds, and predict pharmacokinetics, pharmacodynamics and toxicity. AI-assisted multi-modal ocular biomarkers may improve treatment monitoring and support personalized medicine. Integrating AI shortens development timelines, enhances efficiency, reduces costs, and increases the success rate of new drugs. Currently, standardized regulations for AI in ocular drug development are still lacking and urgently needed to ensure safe and equitable implementation.","author":[{"family":"Cheng","given":"Haoran"},{"family":"Wong","given":"Joy"},{"family":"Quek","given":"Chrystie"},{"family":"Goldberg","given":"Jeffrey"},{"family":"Mahajan","given":"Vinit"},{"family":"Wong","given":"Tien"},{"family":"Mehta","given":"Jodhbir"},{"family":"Ting","given":"Daniel"},{"family":"Ting","given":"Darren"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01954-y","URL":"https://doi.org/10.1038/s41746-025-01954-y","source":"openalex"},{"id":"oa:W4410902394","type":"article-journal","title":"Enhancing professional communication training in higher education through artificial intelligence(AI)-integrated exercises: study protocol for a randomised controlled trial","abstract":"BACKGROUND: Effective communication skills are fundamental for health care professionals, yet conventional training methods face challenges in scalability and accessibility due to resource constraints. The emergence of artificial intelligence (AI), particularly generative AI, offers innovative ways for enhancing communication skills training by simulating realistic conversational scenarios and providing personalised, adaptive feedback. This manuscript is presented as study protocol for a cluster-randomised controlled trial that aims at evaluating the efficacy of an AI-supported higher education training protocol incorporating generative AI exercises to enhance communication competencies among psychology students. METHODS: In this cluster-randomised controlled trial, psychology students enrolled in communication skill seminars at a medium sized university in a medium sized German city will participate. Classes will be assigned within a parallel group design to the AI condition (AI-enhanced exercises alongside teaching-as-usual, TAU, that includes classical exercises) or the control condition (TAU only). Additional non-randomised comparison classes will comprise students with TAU only, but not be part of main analyses. The primary outcome is the change in communication skills from baseline, assessed through questions reflecting the communication techniques emphasised in the training. Secondary outcomes include communication skills, self-efficacy and self-concept, motivation, attitudes toward AI, user experience with the AI tool, student evaluations of course quality, and feasibility aspects such as uptake and usability. Data will be collected via online surveys and the university's teaching platform. Statistical analyses will employ mixed models to evaluate the intervention's impact. DISCUSSION: This study will provide empirical evidence on the effectiveness and feasibility of integrating AI into higher education communication skills training. Successful integration of AI-enhanced training could revolutionise educational practices by offering scalable, accessible, and personalised learning experiences. The findings may have broader implications for incorporating AI tools in various educational and professional training contexts, while addressing ethical considerations and promoting responsible use of AI in education. TRIAL REGISTRATION: This trial has been pre-registered on the Open Science Framework (OSF) under identifier 'th6f4'.","author":[{"family":"Meinlschmidt","given":"Gunther"},{"family":"Koc","given":"Sara"},{"family":"Boerner","given":"Emma"},{"family":"Tegethoff","given":"Marion"},{"family":"Simacek","given":"Thomas"},{"family":"Schirmer","given":"Liam"},{"family":"Schneider","given":"Michael"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12909-025-07307-3","URL":"https://doi.org/10.1186/s12909-025-07307-3","source":"openalex"},{"id":"oa:W4409623003","type":"article-journal","title":"Artificial Intelligence Social Responsibility in the Consumer Market: Dimension Exploration and Scale Development","abstract":"ABSTRACT This study explored the conceptualization, dimensional structure, and measurement of artificial intelligence (AI) social responsibility in the consumer market. Data were collected through semi‐structured in‐depth interviews with 32 respondents. A grounded theory research approach was employed to construct a structural model of AI social responsibility that included the dimensions of ethics, safety, applicability, credibility, and reflexivity. Subsequently, an exploratory factor analysis was conducted on 305 questionnaire data collected through an online survey as well as a confirmatory factor analysis on 325 questionnaire data. The analyses led to the development of an AI social responsibility scale consisting of 18 items and demonstrating good reliability and validity. Moreover, using structural equation modeling, strong nomological validity was demonstrated. The results indicated that AI social responsibility and its dimensions significantly predicted flow experience and experience satisfaction. The findings enhance understanding of the conceptual meaning and dimensional structure of AI social responsibility in the consumer market, as well as provide a psychometrically reliable and valid measurement tool for use in future research. Furthermore, the findings not only facilitate the design and implementation of AI technologies, but they also offer crucial insights for companies and their stakeholders to devise and refine AI social responsibility strategies and other marketing tactics—thereby augmenting CSR 3.0 management practices.","author":[{"family":"Peng-Yi","given":"Shen"},{"family":"Li","given":"Jinxiong"},{"family":"Wan","given":"Demin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/ijcs.70054","URL":"https://doi.org/10.1111/ijcs.70054","source":"openalex"},{"id":"oa:W4410977560","type":"article-journal","title":"Artificial intelligence in bone metastasis analysis: Current advancements, opportunities and challenges","abstract":"BACKGROUND: Artificial Intelligence is transforming medical imaging, particularly in the analysis of bone metastases (BM), a serious complication of advanced cancers. Machine learning and deep learning techniques offer new opportunities to improve detection, recognition, and segmentation of bone metastasis. Yet, challenges such as limited data, interpretability, and clinical validation remain. METHODS: Following PRISMA guidelines, we reviewed artificial intelligence methods and applications for bone metastasis analysis across major imaging modalities including CT, MRI, PET, SPECT, and bone scintigraphy. The survey includes traditional machine learning models and modern deep learning architectures such as CNNs and transformers. We also examined available datasets and their effect in developing artificial intelligence in this field. RESULTS: Artificial intelligence models have achieved strong performance across tasks and modalities, with Convolutional Neural Network (CNN) and Transformer architectures showing particularly efficient performance across different tasks. However, limitations persist, including data imbalance, overfitting risks, and the need for greater transparency. Clinical translation is also challenged by regulatory and validation hurdles. CONCLUSION: Artificial intelligence holds strong potential to improve BM diagnosis and streamline radiology workflows. To reach clinical maturity, future work must address data diversity, model explainability, and large-scale validation, which are critical steps for being trusted to be integrated into the oncology care routines.","author":[{"family":"Afnouch","given":"Marwa"},{"family":"Bougourzi","given":"Fares"},{"family":"Gaddour","given":"Olfa"},{"family":"Dornaika","given":"Fadi"},{"family":"Ahmed","given":"Abdelmalik"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.compbiomed.2025.110372","URL":"https://doi.org/10.1016/j.compbiomed.2025.110372","source":"openalex"},{"id":"oa:W4406675593","type":"article-journal","title":"What generative Artificial Intelligence priorities and challenges do senior Australian educational policy makers identify (and why)?","abstract":"Abstract Free access to powerful generative Artificial Intelligence (AI) in schools has left educators and system leaders grappling with how to responsibly respond to the consequent challenges and opportunities that this new technology poses. This paper examines the priorities and challenges that senior Australian educational leaders identify with relation to responsible and ethical use of generative AI in school education, and the reasons for their beliefs. Members of the Australian generative Artificial Intelligence in Education working group as well as other senior policymakers throughout Australia participated in a two-phase data collection process involving survey responses and focus group discussions. Ranking activities revealed a large number of priorities and systemic challenges, with no unilateral consensuses emerging. The highest priorities for senior policymakers related to managing risks, educating teachers, and educating system leaders, while the main systemic and environmental challenges related to the pace of change, teacher capabilities and professional learning, and equitable access to the technology. Throughout the analysis, meta themes emerged that characterised the policy-setting environment as one involving urgency, uncertainty, interconnectedness, contextuality, and complexity, with the pivotal role of teachers highlighted throughout. Reflections on responsible and ethical policy-setting in response to rapid technological change are provided, including with relation to anticipatory and networked governance and the inter-relationship with the broader policy context. Recommendations for further research and practice are also proposed.","author":[{"family":"Bower","given":"Matt"},{"family":"Henderson","given":"Michael"},{"family":"Slade","given":"Christine"},{"family":"Southgate","given":"Erica"},{"family":"Gulson","given":"Kalervo"},{"family":"Lodge","given":"Jason"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s13384-025-00801-z","URL":"https://doi.org/10.1007/s13384-025-00801-z","source":"openalex"},{"id":"oa:W4413399056","type":"article-journal","title":"The application of artificial intelligence models in predicting the risk of diabetic foot: a multicenter study","abstract":"This study explores diabetic foot (DF), a severe complication in diabetes, by combining deep learning (DL) and machine learning (ML) to develop a multi-model prediction tool. Early identification of high-risk DF patients can reduce disability and mortality. The research also aims to create an integrated application to assist clinicians in precise, efficient risk assessment for early intervention. In this multicenter retrospective study, 6,180 elderly diabetic patients (aged 60-85) were enrolled from 11 community hospitals in Shanghai in 2024. Lasso regression was used to identify 16 key DF risk factors, including age, MMSE score, lower limb discomfort, ABI, and hematocrit. Fourteen ML models (RF, XGBoost, CART, MLP, etc.) and three DL models (DNN, CNN, Transformer) were trained, with hyperparameters optimized via cross-validation and grid search. An application was developed integrating these models, offering both single and batch prediction options with visualization tools for clinical use.Experimental results showed the Logistic regression ensemble model achieved robust performance, with AUC values of 0.943 (validation set, 95% CI: 0.935-0.951) and 0.938 (test set, 95% CI: 0.929-0.947), along with high accuracy, precision, recall, and F1 scores. SHAP analysis revealed key predictive features including ABI results, lower limb discomfort, and MMSE score. The developed app integrates multiple models, compares their predictions for different clinical scenarios, and enhances prediction transparency and reliability.The multi-model approach demonstrates strong predictive performance for DF risk, offering clinicians an intuitive and accurate assessment tool tailored to individual patients. By combining multiple models, we enhance result stability and clinical applicability compared to single-model approaches. Future work will focus on algorithm optimization, expanded datasets, and real-time monitoring integration to enable more precise, dynamic risk evaluation for improved DF prevention and early intervention.","author":[{"family":"Yao","given":"Li"},{"family":"Zhou","given":"Siyuan"},{"family":"Ren","given":"Bichen"},{"family":"Ju","given":"Shuai"},{"family":"Li","given":"Xiaoyan"},{"family":"Li","given":"Wenqiang"},{"family":"Li","given":"Bingzhe"},{"family":"Cai","given":"Yunmin"},{"family":"Chang","given":"Chunlei"},{"family":"Huang","given":"Lihong"},{"family":"Dong","given":"Zhihui"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s13040-025-00477-2","URL":"https://doi.org/10.1186/s13040-025-00477-2","source":"openalex"},{"id":"oa:W4406289994","type":"article-journal","title":"LungDiag: Empowering artificial intelligence for respiratory diseases diagnosis based on electronic health records, a multicenter study","abstract":"Abstract Respiratory diseases pose a significant global health burden, with challenges in early and accurate diagnosis due to overlapping clinical symptoms, which often leads to misdiagnosis or delayed treatment. To address this issue, we developed LungDiag, an artificial intelligence (AI)‐based diagnostic system that utilizes natural language processing (NLP) to extract key clinical features from electronic health records (EHRs) for the accurate classification of respiratory diseases. This study employed a large cohort of 31,267 EHRs from multiple centers for model training and internal testing. Additionally, prospective real‐world validation was conducted using 1142 EHRs from three external centers. LungDiag demonstrated superior diagnostic performance, achieving an F1 score of 0.711 for top 1 diagnosis and 0.927 for top 3 diagnoses. In real‐world testing, LungDiag outperformed both human experts and ChatGPT 4.0, achieving an F1 score of 0.651 for top 1 diagnosis. The study emphasizes the potential of LungDiag as an effective tool to support physicians in diagnosing respiratory diseases more accurately and efficiently. Despite the promising results, further large‐scale multicenter validation with larger sample sizes is still needed to confirm its clinical utility and generalizability.","author":[{"family":"Liang","given":"Hengrui"},{"family":"Yang","given":"Tao"},{"family":"Liu","given":"Zi"},{"family":"Jian","given":"Wenhua"},{"family":"Chen","given":"Yilong"},{"family":"Li","given":"Bingliang"},{"family":"Yan","given":"Zeping"},{"family":"Xu","given":"Weiqiang"},{"family":"Chen","given":"Luming"},{"family":"Qi","given":"Yifan"},{"family":"Wang","given":"Zhiwei"},{"family":"Liao","given":"Yajing"},{"family":"Lin","given":"Peixuan"},{"family":"Li","given":"Jiameng"},{"family":"Wang","given":"Wei"},{"family":"Li","given":"Li"},{"family":"Wang","given":"Meijia"},{"family":"Zhang","given":"Yunhui"},{"family":"Deng","given":"Lizong"},{"family":"Jiang","given":"Taijiao"},{"family":"He","given":"Jianxing"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/mco2.70043","URL":"https://doi.org/10.1002/mco2.70043","source":"openalex"},{"id":"oa:W4411256651","type":"article-journal","title":"Artificial Intelligence in Melanoma Detection: A Review of Current Technologies and Future Directions","abstract":"Early and accurate identification of malignant melanoma continues to be a major challenge for clinicians in the field. Traditional diagnostic approaches, including physical examination, histology, imaging, and nodal assessments, are frequently costly, require significant expertise, and can display large variations among clinicians. These factors may result in missed or misdiagnosis, which often significantly affects a patient’s prognosis. We examine in detail how the application of AI methods such as machine learning and deep learning can be used to advance early detection and identification of melanoma. We review various AI algorithms, including standard classifiers, ensemble techniques, and complex deep learning models. Hybrid models that combine convolutional neural networks (CNNs) and support vector machines (SVMs) are emphasized in this review, as they show enhanced performance and improved resistance to variations in the diagnostician’s input. Better utility of transfer learning and data augmentation approaches is discussed to overcome the challenges posed by small and unbalanced medical datasets. The authors consider the combination of various types of medical information for more effective cancer diagnosis. However, significant obstacles, including model explainability, privacy safeguarding, and clinical evaluation, still need to be addressed. Extensive efforts are needed to overcome these barriers if AI systems are to be effectively adopted within healthcare environments. We suggest that AI offers the opportunity to revolutionize melanoma care by enabling rapid decision support and individualized treatment plans. Realizing this opportunity will depend on effective partnerships between researchers, clinicians, and industry to bring together advances in technology and their effective implementation in the healthcare system.","author":[{"family":"Alam","given":"Fakhre"},{"family":"Ullah","given":"Asad"},{"family":"Shah","given":"Dilawar"},{"family":"Ali","given":"Shujaat"},{"family":"Tahir","given":"Muhammad"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1155/int/3164952","URL":"https://doi.org/10.1155/int/3164952","source":"openalex"},{"id":"oa:W4415273759","type":"article-journal","title":"Application and research progress on artificial intelligence in the quality of Traditional Chinese Medicine","abstract":"The clinical safety and therapeutic performance of Traditional Chinese Medicine (TCM) are closely tied to its quality. However, with the rapid expansion of the TCM industry, conventional quality control approaches based on empirical observations and single-metabolite quantification have become increasingly inadequate for addressing the complex and variable requirements of quality assessment. In recent years, artificial intelligence (AI)-with strong capabilities in data processing and pattern recognition-has emerged as a promising tool for establishing predictive models to efficiently handle heterogeneous, multi-source datasets (such as spectra, chromatograms, images, and textual information). This enables intelligent prediction of quality indicators and anomaly detection, and offering novel strategies for modernizing TCM quality control. This review provides a comprehensive synthesis of commonly applied machine learning and deep learning algorithms, systematically outlining recent advances in AI-enabled sensing applications such as image recognition, odor analysis, authenticity verification, origin tracing, quality grading, and storage-age determination. It further emphasizes the integration of AI with multi-omics and bioinformatics approaches for efficacy-oriented evaluation and safety assessment, including identification of Q-markers, elucidation of pharmacodynamic mechanisms, and predictive modeling of both endogenous and exogenous toxic metabolites. It also identifies key challenges and technical bottlenecks, and outlines priorities for building scalable, regulation-aware, data-driven quality-control systems that support the sustainable, high-quality development of the TCM industry.","author":[{"family":"Li","given":"Meiyu"},{"family":"Zhu","given":"Junqing"},{"family":"Liu","given":"Xiaonan"},{"family":"Wu","given":"Mengyue"},{"family":"Dong","given":"Kun"},{"family":"Li","given":"Xiaoyan"},{"family":"Gao","given":"Peng"},{"family":"Jiang","given":"Zhihui"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fphar.2025.1687681","URL":"https://doi.org/10.3389/fphar.2025.1687681","source":"openalex"},{"id":"oa:W4413351111","type":"article-journal","title":"Applications and Performance of Artificial Intelligence in Spinal Metastasis Imaging: A Systematic Review","abstract":"Background: Spinal metastasis is the third most common site for metastatic localization, following the lung and liver. Manual detection through imaging modalities such as CT, MRI, PET, and bone scintigraphy can be costly and inefficient. Preliminary artificial intelligence (AI) techniques and computer-aided detection (CAD) systems have attempted to improve lesion detection, segmentation, and treatment response in oncological imaging. The objective of this review is to evaluate the current applications of AI across multimodal imaging techniques in the diagnosis of spinal metastasis. Methods: Databases like PubMed, Scopus, Web of Science Advance, Cochrane, and Embase (Ovid) were searched using specific keywords like ‘spine metastases’, ‘artificial intelligence’, ‘machine learning’, ‘deep learning’, and ‘diagnosis’. The screening of studies adhered to the PRISMA guidelines. Relevant variables were extracted from each of the included articles such as the primary tumor type, cohort size, and prediction model performance metrics: area under the receiver operating curve (AUC), accuracy, sensitivity, specificity, internal validation and external validation. A random-effects meta-analysis model was used to account for variability between the studies. Quality assessment was performed using the PROBAST tool. Results: This review included 39 studies published between 2007 and 2024, encompassing a total of 6267 patients. The three most common primary tumors were lung cancer (56.4%), breast cancer (51.3%), and prostate cancer (41.0%). Four studies reported AUC values for model training, 16 for internal validation, and five for external validation. The weighted average AUCs were 0.971 (training), 0.947 (internal validation), and 0.819 (external validation). The risk of bias was the highest in the analysis domain, with 22 studies (56%) rated high risk, primarily due to inadequate external validation and overfitting. Conclusions: AI-based approaches show promise for enhancing the detection, segmentation, and characterization of spinal metastatic lesions across multiple imaging modalities. Future research should focus on developing more generalizable models through larger and more diverse training datasets, integrating clinical and imaging data, and conducting prospective validation studies to demonstrate meaningful clinical impact.","author":[{"family":"Sanker","given":"Vivek"},{"family":"Gowda","given":"Poorvikha"},{"family":"Thaller","given":"Alexander"},{"family":"Li","given":"Zhikai"},{"family":"Heesen","given":"Philip"},{"family":"Qiang","given":"Zekai"},{"family":"Hariharan","given":"S"},{"family":"Nordin","given":"Emil"},{"family":"Cavagnaro","given":"María"},{"family":"Ratliff","given":"John"},{"family":"Desai","given":"Atman"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jcm14165877","URL":"https://doi.org/10.3390/jcm14165877","source":"openalex"},{"id":"oa:W4406892722","type":"article-journal","title":"A foundation model for human-AI collaboration in medical literature mining","abstract":"Applying artificial intelligence (AI) for systematic literature review holds great potential for enhancing evidence-based medicine, yet has been limited by insufficient training and evaluation. Here, we present LEADS, an AI foundation model trained on 633,759 samples curated from 21,335 systematic reviews, 453,625 clinical trial publications, and 27,015 clinical trial registries. In experiments, LEADS demonstrates consistent improvements over four cutting-edge large language models (LLMs) on six literature mining tasks, e.g., study search, screening, and data extraction. We conduct a user study with 16 clinicians and researchers from 14 institutions to assess the utility of LEADS integrated into the expert workflow. In study selection, experts using LEADS achieve 0.81 recall vs. 0.78 without, saving 20.8% time. For data extraction, accuracy reached 0.85 vs. 0.80, with 26.9% time savings. These findings encourage future work on leveraging high-quality domain data to build specialized LLMs that outperform generic models and enhance expert productivity in literature mining.","author":[{"family":"Wang","given":"Zifeng"},{"family":"Cao","given":"Lang"},{"family":"Jin","given":"Qiao"},{"family":"Chan","given":"Joey"},{"family":"Wan","given":"Nicholas"},{"family":"Afzali","given":"Behdad"},{"family":"Cho","given":"Hyun"},{"family":"Choi","given":"CY"},{"family":"Emamverdi","given":"Mehdi"},{"family":"Gill","given":"Manjot"},{"family":"Kim","given":"Sunhyung"},{"family":"Li","given":"Yijia"},{"family":"Liu","given":"Yi"},{"family":"Luo","given":"Yiming"},{"family":"Ong","given":"Hanley"},{"family":"Rousseau","given":"Justin"},{"family":"Sheikh","given":"Irfan"},{"family":"Wei","given":"Jenny"},{"family":"Xu","given":"Ziyang"},{"family":"Zallek","given":"Christopher"},{"family":"Kim","given":"Kyungsang"},{"family":"Peng","given":"Yifan"},{"family":"Lu","given":"Zhiyong"},{"family":"Sun","given":"Jimeng"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41467-025-62058-5","URL":"https://doi.org/10.1038/s41467-025-62058-5","source":"openalex"},{"id":"oa:W4409290068","type":"article-journal","title":"Change of Heart: Can Artificial Intelligence Transform Infective Endocarditis Management?","abstract":"Artificial intelligence (AI) has emerged as a promising adjunct in the diagnosis and management of infective endocarditis (IE), a disease characterized by diagnostic complexity and significant morbidity. Machine learning (ML) models such as SABIER and SYSUPMIE have demonstrated strong predictive accuracy for early IE diagnosis, embolic risk stratification, and postoperative mortality, surpassing traditional clinical scoring systems. In imaging, AI-enhanced echocardiography and advanced modalities like FDG-PET/CT offer improved sensitivity, specificity, and reduced inter-observer variability, potentially transforming clinical decision making. Additionally, AI-powered microbiological techniques, including MALDI-TOF mass spectrometry combined with ML and neural network-based metagenomic classifiers, show promise in rapidly identifying pathogens and predicting antimicrobial resistance. Despite encouraging early results, widespread adoption faces barriers, including data limitations, interpretability issues, ethical concerns, and the need for robust validation. Future directions include leveraging generative AI as clinical consultative tools, provided their capabilities and limitations are carefully managed. Ultimately, collaborative efforts addressing these challenges could transform IE care, enhancing diagnostic accuracy, clinical outcomes, and patient safety.","author":[{"family":"Mchugh","given":"Jack"},{"family":"Challener","given":"Douglas"},{"family":"Tabaja","given":"Hussam"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/pathogens14040371","URL":"https://doi.org/10.3390/pathogens14040371","source":"openalex"},{"id":"oa:W4411234791","type":"article-journal","title":"Artificial Intelligence Models in Diagnosis and Treatment of Kidney Diseases: Current Status and Prospects","abstract":"Background: Artificial intelligence (AI) has made significant advances in nephrology, revolutionizing the diagnosis, prognosis, and treatment of kidney diseases. Summary: This review provides an overview of AI applications in nephrology, introducing the basic structures of each model, highlighting both traditional machine-learning approaches and neural networks, and providing model application comparisons along with selection recommendations. It discussed key challenges in deciding appropriate AI models for specific tasks and evaluated their advantages, limitations, and optimal use cases. Current applications of AI in nephrology mainly include diagnosis and disease outcome prediction, medical image analysis, treatment recommendations, and personalized health management, supported by massive electronic health records and multimodal data integration. Traditional machine learning models perform well on datasets of varying sizes and structures, while neural networks excel at handling complex and imaging data. Emerging hardware innovations are expected to improve the performance of neural network models, enabling more accurate diagnosis and automated analysis in clinical practice. In the future, AI will have great potential to advance individualized patient care and enable real-time data processing in nephrology. Key Messages: An overview of AI applications in nephrology is provided in this review.","author":[{"family":"Li","given":"Cheng"},{"family":"Liu","given":"Jing"},{"family":"Fu","given":"Ping"},{"family":"Zou","given":"Jie"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1159/000546397","URL":"https://doi.org/10.1159/000546397","source":"openalex"},{"id":"oa:W4413962523","type":"article-journal","title":"Artificial intelligence-oriented predictive model for the risk of postpartum depression: a systematic review","abstract":"Introduction Postpartum depression (PPD) is a significant mental health concern affecting 3.5-33.0% of mothers worldwide, with potentially severe consequences for both maternal and infant well-being. The emergence of artificial intelligence (AI) and machine learning (ML) technologies offers new opportunities for the early prediction of PPD risk, potentially enabling timely interventions to mitigate adverse outcomes. Methods This systematic review was conducted until October 31, 2024, using several electronic databases, including PubMed, Web of Science, CBM, VIP, CNKI, and Wanfang Data. All the studies predicted the occurrence of PPD using algorithms. The review process involved dual-independent screening by two authors using predefined criteria, with discrepancies resolved through consensus discussion involving a third investigator, and assessed the quality of the included models using the prediction model risk of bias assessment tool (PROBAST). Inter-rater agreement was quantified using Cohen’s κ. Results Eleven studies were included in the systematic review. The random forest, support vector machine, and logistic regression algorithms demonstrated high predictive performance (AUROC > 0.9). The main predictors of PPD were maternal age, pregnancy stress and adverse emotions, history of mental disorders, maternal education, marital relationship, and sleep status. The overall performance of the prediction model was excellent. However, the generalizability of the model was limited, and there was a certain risk of bias. Issues such as data quality, algorithm interpretability, and the cross-cultural and cross-population applicability of the model need to be addressed. Conclusion The model has the potential to predict the risk of PPD and provide support for early identification and intervention. Future research should optimize the model, improve its prediction accuracy, and test its applicability across cultures and populations to reduce the incidence of PPD and guarantee the mental health of pregnant and maternal women.","author":[{"family":"Xia","given":"Jie"},{"family":"Chen","given":"Chen"},{"family":"Lu","given":"Xiuqin"},{"family":"Zhang","given":"Tengfei"},{"family":"Wang","given":"Tingting"},{"family":"Wang","given":"Qingling"},{"family":"Zhou","given":"Qianqian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fpubh.2025.1631705","URL":"https://doi.org/10.3389/fpubh.2025.1631705","source":"openalex"},{"id":"oa:W4411750810","type":"article-journal","title":"Synergizing DeepSeek's artificial intelligence innovations with brain–computer interfaces","abstract":"Abstract The integration of artificial intelligence (AI) and brain–computer interfaces (BCIs) represents a significant advancement in neurotechnology, with broad potential applications in healthcare, communication, and human augmentation. This study examines the synergy between DeepSeek, a leader in efficient, open‐source AI models, and next‐generation BCI technologies. We analyze DeepSeek's contributions to model training efficiency, adaptive reasoning, and open‐source accessibility, and propose a framework for BCI development that incorporates these innovations. Additionally, we explore how AI‐driven neural signal processing, hardware optimization, and ethical AI–BCI systems can address the critical limitations of current BCI technologies, including signal fidelity, scalability, and real‐world applicability. Finally, we offer recommendations for interdisciplinary collaboration, regulatory improvements, and equitable technology dissemination to foster the sustainable development of AI–BCI technology.","author":[{"family":"Wu","given":"Canbiao"},{"family":"Chen","given":"Nayu"},{"family":"Sun","given":"Tuo"},{"family":"Tan","given":"Ping"},{"family":"Wang","given":"Peng"},{"family":"Li","given":"Guangli"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/brx2.70035","URL":"https://doi.org/10.1002/brx2.70035","source":"openalex"},{"id":"oa:W7118092294","type":"article-journal","title":"Artificial intelligence driven meat preservation technology: research progress, challenges, and future directions","abstract":"Abstract With the rapid advancement of artificial intelligence (AI) technology, its applications in food science have become increasingly widespread—particularly in the preservation of meat products—where it demonstrates substantial potential. This study systematically reviews the key applications of AI in meat product preservation, encompassing quality assessment, cold chain monitoring, shelf-life prediction, the development of microbial prediction models, intelligent packaging optimization, and consumer behavior analysis. AI-enabled preservation strategies are effective in maintaining meat product quality. Notably, machine learning algorithms offer distinct advantages in the real-time monitoring of critical quality indicators and the prediction of shelf life under diverse preservation conditions, thereby providing robust support for producers’ decision-making processes. Furthermore, this study explores in depth the application prospects and development trends of AI in meat product preservation, with the aim of offering novel insights to facilitate the development of more efficient and intelligent meat preservation processes. By integrating AI technology, the meat preservation sector is expected to undergo an intelligent transformation—shifting from a traditional experience-driven paradigm to a data-driven one—thereby elevating the overall development level of the industry.","author":[{"family":"Yu","given":"FR"},{"family":"Zhang","given":"D"},{"family":"Zhang","given":"Jiamin"},{"family":"Wang","given":"W"},{"family":"Cheng","given":"Jian"},{"family":"Zhang","given":"Chunjiang"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1093/fqsafe/fyaf083","URL":"https://doi.org/10.1093/fqsafe/fyaf083","source":"openalex"},{"id":"oa:W4409211505","type":"article-journal","title":"The role of artificial intelligence in cardiovascular research: Fear less and live bolder","abstract":"BACKGROUND: Artificial intelligence (AI) has captured the attention of everyone, including cardiovascular (CV) clinicians and scientists. Moving beyond philosophical debates, modern cardiology cannot overlook AI's growing influence but must actively explore its potential applications in clinical practice and research methodology. METHODS AND RESULTS: AI offers exciting possibilities for advancing CV medicine by uncovering disease heterogeneity, integrating complex multimodal data, and enhancing treatment strategies. In this review, we discuss the innovative applications of AI in cardiac electrophysiology, imaging, angiography, biomarkers, and genomic data, as well as emerging tools like face recognition and speech analysis. Furthermore, we focus on the expanding role of machine learning (ML) in predicting CV risk and outcomes, outlining a roadmap for the implementation of AI in CV care delivery. While the future of AI holds great promise, technical limitations and ethical challenges remain significant barriers to its widespread clinical adoption. CONCLUSIONS: Addressing these issues through the development of high-quality standards and involving key stakeholders will be essential for AI to transform cardiovascular care safely and effectively.","author":[{"family":"Scuricini","given":"Alessandro"},{"family":"Ramoni","given":"Davide"},{"family":"Liberale","given":"Luca"},{"family":"Montecucco","given":"Fabrizio"},{"family":"Carbone","given":"Federico"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/eci.14364","URL":"https://doi.org/10.1111/eci.14364","source":"openalex"},{"id":"oa:W4416447593","type":"article-journal","title":"Determination of the General Attitude to and Anxiety About Artificial Intelligence of Nurses Working in Internal Medicine Clinics: A Mixed‐Method Study","abstract":"AIM: To determine the concerns and general attitudes to artificial intelligence (AI) of nurses working in Internal Medicine clinics in Türkiye. BACKGROUND: In parallel with developments in technology in all areas, AI is actively used in the healthcare sector in the diagnosis, treatment, and follow-up of patients, with algorithms based on the analysis of medical data. METHODS: The study was conducted between January 1, 2025, and May 15, 2025, using a mixed-method model of nested design (descriptive relationship-seeking and qualitative research) in a high-capacity tertiary care university hospital in the southern region of Türkiye. The data were collected using a Descriptive Information Form, the General Attitudes to Artificial Intelligence Scale, the Artificial Intelligence Anxiety Scale, and a semi-structured interview form. In the examination of causality between the quantitative scales in the data analyses, structural equation modeling was used, and in the analysis of the qualitative data, AI-supported Max Qualitative Data Analysis (MAXQDA) was used. STROBE was used in the reporting. RESULTS: Quantitative findings revealed a moderate level of positive attitude, negative attitude, and AI anxiety. A significant weak negative correlation was found between positive attitude and AI anxiety, while a moderate positive correlation emerged between negative attitude and anxiety. Qualitative analysis identified three primary themes influencing these attitudes: ethical violation (subthemes: first do no harm and confidentiality of information/privacy), holistic care (empathy, trust, and emotion), and quality and patient satisfaction (medical error, missed nursing care, and satisfaction). DISCUSSION: Adopting AI fosters positive attitudes and alleviates concerns about its use. CONCLUSIONS: The study results showed that the levels of positive attitude and negative attitude to AI, and concerns of the nurses about the use of AI, were slightly above average. In the context of the use of AI in nursing services, it was seen that ethical neglect, holistic care, quality, and patient satisfaction affected the attitudes to AI and anxiety of nurses. IMPLICATIONS FOR NURSING AND HEALTH POLICY: AI technologies can reduce the workload of nurses if they are integrated into healthcare policies under state control, ensuring high reliability of software and hardware. They can also help as a reminder by serving as a decision support component, can facilitate the complex nursing process, and increase the time that the nurse can spend with the patient. Thus, the success of the institution will be positively affected by increased quality of care in the delivery of healthcare services, the prevention of malpractice, and increased patient safety and satisfaction.","author":[{"family":"Seven","given":"Ahmet"},{"family":"Soylu","given":"Ayşe"},{"family":"Soylu","given":"Dilek"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/inr.70134","URL":"https://doi.org/10.1111/inr.70134","source":"openalex"},{"id":"oa:W4414866642","type":"article-journal","title":"Designing artificial intelligence chatbots for self-regulated learning from a systematic review based on Habermas's three interests","abstract":"Self-regulated learning (SRL) is widely acknowledged as a key driver of academic success and lifelong learning, yet learners often struggle to plan, monitor, and reflect effectively on their progress. Concurrently, advances in artificial intelligence (AI) chatbots offer unprecedented opportunities to scale SRL. This paper reviews 78 studies from 2018 to 2025 on how AI chatbots can be created to support self-regulated learning (SRL) by leveraging technology, social interaction, and reflection, based on Habermas’s three types of knowledge interests (technical, practical, and emancipatory). We identify eight core chatbot functions that align with the forethought, performance, and self-reflection phases of Zimmerman’s SRL cycle. Socially, effective implementations embed chatbots within teacher–peer–chatbot triads to facilitate co-regulation, dialogue, and community building. Critically, chatbots can prompt deep self-evaluation and ethical inquiry, though designs must guard against overreliance and superficial reflection. We synthesise these findings into an integrative framework that maps chatbot features onto SRL processes and Habermasian interests, and we highlight underexplored areas – such as K-12 contexts, multimodal analytics, and emancipatory reflection – for future research. Our review offers actionable guidance for researchers, designers, and educators seeking to harness AI chatbots as versatile partners in fostering autonomous, socially grounded, and critically reflective learners.","author":[{"family":"Wu","given":"Xiu"},{"family":"Radloff","given":"Jeffrey"},{"family":"Yeter","given":"Ibrahim"},{"family":"Wang","given":"Lei"},{"family":"Chiu","given":"Thomas"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/10494820.2025.2563086","URL":"https://doi.org/10.1080/10494820.2025.2563086","source":"openalex"},{"id":"oa:W4415818264","type":"article-journal","title":"Next-Generation Advances in Prostate Cancer Imaging and Artificial Intelligence Applications","abstract":"Prostate cancer is one of the leading causes of cancer-related morbidity and mortality worldwide, and imaging plays a critical role in its detection, localization, staging, treatment, and management. The advent of artificial intelligence (AI) has introduced transformative possibilities in prostate imaging, offering enhanced accuracy, efficiency, and consistency. This review explores the integration of AI in prostate cancer diagnostics across key imaging modalities, including multiparametric MRI (mpMRI), PSMA PET/CT, and transrectal ultrasound (TRUS). Advanced AI technologies, such as machine learning, deep learning, and radiomics, are being applied for lesion detection, risk stratification, segmentation, biopsy targeting, and treatment planning. AI-augmented systems have demonstrated the ability to support PI-RADS scoring, automate prostate and tumor segmentation, guide targeted biopsies, and optimize radiation therapy. Despite promising performance, challenges persist regarding data heterogeneity, algorithm generalizability, ethical considerations, and clinical implementation. Looking ahead, multimodal AI models integrating imaging, genomics, and clinical data hold promise for advancing precision medicine in prostate cancer care and assisting clinicians, particularly in underserved regions with limited access to specialists. Continued multidisciplinary collaboration will be essential to translate these innovations into evidence-based practice. This article explores current AI applications and future directions that are transforming prostate imaging and patient care.","author":[{"family":"Miao","given":"Kathleen"},{"family":"Miao","given":"Julia"},{"family":"Finkelstein","given":"Mark"},{"family":"Chatterjee","given":"Aritrick"},{"family":"Oto","given":"Aytekin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jimaging11110390","URL":"https://doi.org/10.3390/jimaging11110390","source":"openalex"},{"id":"oa:W4407833429","type":"article-journal","title":"Exploring the Ethical Challenges of Conversational AI in Mental Health Care: Scoping Review","abstract":"BACKGROUND: Conversational artificial intelligence (CAI) is emerging as a promising digital technology for mental health care. CAI apps, such as psychotherapeutic chatbots, are available in app stores, but their use raises ethical concerns. OBJECTIVE: We aimed to provide a comprehensive overview of ethical considerations surrounding CAI as a therapist for individuals with mental health issues. METHODS: We conducted a systematic search across PubMed, Embase, APA PsycINFO, Web of Science, Scopus, the Philosopher's Index, and ACM Digital Library databases. Our search comprised 3 elements: embodied artificial intelligence, ethics, and mental health. We defined CAI as a conversational agent that interacts with a person and uses artificial intelligence to formulate output. We included articles discussing the ethical challenges of CAI functioning in the role of a therapist for individuals with mental health issues. We added additional articles through snowball searching. We included articles in English or Dutch. All types of articles were considered except abstracts of symposia. Screening for eligibility was done by 2 independent researchers (MRM and TS or AvB). An initial charting form was created based on the expected considerations and revised and complemented during the charting process. The ethical challenges were divided into themes. When a concern occurred in more than 2 articles, we identified it as a distinct theme. RESULTS: We included 101 articles, of which 95% (n=96) were published in 2018 or later. Most were reviews (n=22, 21.8%) followed by commentaries (n=17, 16.8%). The following 10 themes were distinguished: (1) safety and harm (discussed in 52/101, 51.5% of articles); the most common topics within this theme were suicidality and crisis management, harmful or wrong suggestions, and the risk of dependency on CAI; (2) explicability, transparency, and trust (n=26, 25.7%), including topics such as the effects of \"black box\" algorithms on trust; (3) responsibility and accountability (n=31, 30.7%); (4) empathy and humanness (n=29, 28.7%); (5) justice (n=41, 40.6%), including themes such as health inequalities due to differences in digital literacy; (6) anthropomorphization and deception (n=24, 23.8%); (7) autonomy (n=12, 11.9%); (8) effectiveness (n=38, 37.6%); (9) privacy and confidentiality (n=62, 61.4%); and (10) concerns for health care workers' jobs (n=16, 15.8%). Other themes were discussed in 9.9% (n=10) of the identified articles. CONCLUSIONS: Our scoping review has comprehensively covered ethical aspects of CAI in mental health care. While certain themes remain underexplored and stakeholders' perspectives are insufficiently represented, this study highlights critical areas for further research. These include evaluating the risks and benefits of CAI in comparison to human therapists, determining its appropriate roles in therapeutic contexts and its impact on care access, and addressing accountability. Addressing these gaps can inform normative analysis and guide the development of ethical guidelines for responsible CAI use in mental health care.","author":[{"family":"Meadi","given":"Mehrdad"},{"family":"Sillekens","given":"Tomas"},{"family":"Metselaar","given":"Suzanne"},{"family":"Balkom","given":"Anton"},{"family":"Bernstein","given":"Justin"},{"family":"Batelaan","given":"Neeltje"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/60432","URL":"https://doi.org/10.2196/60432","source":"openalex"},{"id":"oa:W4415152514","type":"article-journal","title":"Application of Artificial Intelligence Technologies as an Intervention for Promoting Healthy Eating and Nutrition in Older Adults: A Systematic Literature Review","abstract":"Background/Objectives: The aging population faces a multitude of health challenges, particularly when it comes to maintaining proper nutrition. Age-related physiological changes, such as decreased metabolism, diminished taste perception, and difficulty in chewing, can lead to insufficient nutrient intake, ultimately resulting in malnutrition. It is crucial to address these issues to promote not only physical health but also overall well-being. In this modern era, artificial intelligence (AI) technologies, including robots and machine learning algorithms, are being increasingly harnessed to encourage healthy eating habits among older adults. This is critical to support healthy aging and mitigate diet-related chronic diseases. However, little or no synthesis has established their effectiveness in delivering personalized, scalable, and adaptive interventions for older adults. This systematic review considers the state-of-the-art application of AI-based interventions aimed at improving dietary behaviors and nutritional outcomes in older adults. Methods: Following the PRISMA 2020 guidelines and a registered PROSPERO protocol (ID: CRD420241045268), we systematically analyzed 30 studies we collected from five databases, published between 2015 and 2025 based on different AI techniques, including machine learning, natural language processing, and recommender systems. We synthesized data collected from these studies to examine the intervention types, outcomes, and methodological approaches. Results: Findings from our review highlight the potential of AI-based interventions to promote engagement among older adults and improve adherence to healthy eating guidelines. Additionally, we found some challenges related to ethical concerns such as privacy and transparency, and limited evidence of their long-term effectiveness. Conclusions: AI-based interventions offer significant promise in promoting healthy eating among older adults through personalized, adaptive, and scalable interventions. Yet, current evidence is constrained by some methodological limitations and ethical concerns, which calls for future research to design inclusive, evidence-based AI interventions that address the unique physiological, psychological, and social needs of older adults.","author":[{"family":"Kalu","given":"Kingsley"},{"family":"Ataguba","given":"Grace"},{"family":"Onifade","given":"Oyepeju"},{"family":"Orji","given":"Fidelia"},{"family":"Giweli","given":"Nabil"},{"family":"Orji","given":"Rita"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/nu17203223","URL":"https://doi.org/10.3390/nu17203223","source":"openalex"},{"id":"oa:W4413962358","type":"article-journal","title":"Diagnostic performance of ultrasound characteristics-based artificial intelligence models for thyroid nodules: a systematic review and meta-analysis","abstract":"Background: Nowadays, artificial intelligence (AI) diagnostic models based on ultrasound features have been gradually integrated into the evaluation of thyroid nodules. However, the diagnostic effects of different AI-assisted diagnosis methods vary greatly. Objective: This study aims to systematically evaluate the performance of the ultrasound-based artificial intelligence diagnostic models in differentiating benign and malignant thyroid nodules and to determine the most effective diagnostic model. Methods: We conducted a comprehensive literature search in PubMed, Web of Science, and the Cochrane Library using subject-specific keywords to identify studies on AI-assisted thyroid nodule diagnosis. Study quality was assessed using Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2). Meta-analysis was performed using Meta-Disc 1.4, Review Manager 5.4, R 4.4.2, and Stata 17.0. Pooled sensitivity, specificity, diagnostic odds ratio (DOR), and area under the summary receiver operating characteristic curve (SROC-AUC) with 95% confidence intervals (CI) were calculated. Subgroup analyses and clinical applicability assessments were conducted. Results: Twenty-eight studies involving 134,028 patients, 158,161 thyroid nodules, and 529,479 ultrasound images were included. The AI-assisted diagnostic system demonstrated high diagnostic performance: pooled sensitivity = 0.89 (95% CI: 0.87-0.91), specificity = 0.84 (0.80-0.88), positive likelihood ratio (PLR) = 5.60 (4.40-7.20), negative likelihood ratio (NLR) = 0.13 (0.10-0.16), DOR = 43.94 (30.11-64.14), and SROC-AUC = 0.93 (0.91-0.95). The threshold effect analysis (Spearman correlation = -0.18, P > 0.05) indicated no significant heterogeneity. The diagnostic accuracy is higher in Asian countries, in prospective and multicenter designs, with external validation sets, without cross-validation, with deep learning, and in postoperative patient subgroups. Additionally, improved performance was observed in cohorts with smaller nodule diameters (<20 mm), higher malignancy rates, older patient age (≥50 years), and higher female proportions, though heterogeneity remained significant. Univariate and multivariate meta-regression analyses identified AI type, malignancy rate of nodules as significant sources of heterogeneity. Notably, the EDLC-TN model showed the highest diagnostic accuracy. Conclusion: AI-assisted diagnostic techniques demonstrate significant potentialin thyroid nodule evaluation, with the EDLC-TN model showing particularly high clinical utility. Optimal diagnostic performance was observed for nodules <20 mm in diameter and in patients aged ≥50 years. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD42024581421, identifier CRD42024581421.","author":[{"family":"Zhan","given":"Jianfeng"},{"family":"Zhang","given":"Jian"},{"family":"Zhu","given":"Shaoqi"},{"family":"Ni","given":"Lin"},{"family":"Zhang","given":"Chen"},{"family":"Hu","given":"Jia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fonc.2025.1614603","URL":"https://doi.org/10.3389/fonc.2025.1614603","source":"openalex"},{"id":"oa:W4413992240","type":"article-journal","title":"Artificial Intelligence-Driven Personalization in Breast Cancer Screening: From Population Models to Individualized Protocols","abstract":"Conventional breast cancer screening programs are predominantly age-based, applying uniform intervals and modalities across broad populations. While this model has reduced mortality, it entails harms-including overdiagnosis, false positives, and missed interval cancers-prompting interest in risk-stratified approaches. In recent years, artificial intelligence (AI) has emerged as a critical enabler of this paradigm shift. This narrative review examines how AI-driven tools are advancing breast cancer screening toward personalization, with a focus on mammographic risk models, multimodal risk prediction, and AI-enabled clinical decision support. We reviewed studies published from 2015 to 2025, prioritizing large cohorts, randomized trials, and prospective validations. AI-based mammographic risk models generally improve discrimination versus classical models and are being externally validated; however, evidence remains heterogeneous across subtypes and populations. Emerging multimodal models integrate genetics, clinical data, and imaging; AI is also being evaluated for triage and personalized intervals within clinical workflows. Barriers remain-explainability, regulatory validation, and equity. Widespread adoption will depend on prospective clinical benefit, regulatory alignment, and careful integration. Overall, AI-based mammographic risk models generally improve discrimination versus classical models and are being externally validated; however, evidence remains heterogeneous across molecular subtypes, with signals strongest for ER-positive disease and limited data for fast-growing and interval cancers. Prospective trials demonstrating outcome benefit and safe interval modification are still pending. Accordingly, adoption should proceed with safeguards, equity monitoring, and clear separation between risk prediction, lesion detection, triage, and decision-support roles.","author":[{"family":"Pesapane","given":"Filippo"},{"family":"Nicosia","given":"Luca"},{"family":"Damelio","given":"Lucrezia"},{"family":"Quercioli","given":"Giulia"},{"family":"Pannarale","given":"Mario"},{"family":"Priolo","given":"Francesca"},{"family":"Marinucci","given":"Irene"},{"family":"Farina","given":"Maria"},{"family":"Penco","given":"Silvia"},{"family":"Dominelli","given":"Valeria"},{"family":"Rotili","given":"Anna"},{"family":"Meneghetti","given":"Lorenza"},{"family":"Bozzini","given":"Anna"},{"family":"Santicchia","given":"Sonia"},{"family":"Cassano","given":"Enrico"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/cancers17172901","URL":"https://doi.org/10.3390/cancers17172901","source":"openalex"},{"id":"oa:W4412526861","type":"article-journal","title":"Scorecard for synthetic medical data evaluation","abstract":"Although the interest in synthetic medical data (SMD) for developing and testing artificial intelligence (AI) methods is growing, the absence of a comprehensive framework to evaluate the quality and applicability of SMD hinders its wider adoption. Here, we outline an evaluation framework designed to meet the unique requirements of medical applications. We also introduce SMD scorecard, a comprehensive report accompanying artificially generated datasets. This scorecard provides a quantitative assessment of SMD across seven criteria (7 Cs), complemented by a descriptive section that contains all relevant information about the dataset. The SMD scorecard provides a practical framework for evaluating and reporting the quality of synthetic data, which can benefit SMD developers and users. The use of synthetic medical data (SMD) in AI development is on the rise, but its broader application is limited by the lack of a comprehensive evaluation framework. Here, Ghada Zamzmi and colleagues present a novel evaluation framework tailored for medical applications, along with an SMD scorecard that quantitatively assesses synthetic datasets across seven key criteria.","author":[{"family":"Zamzmi","given":"Ghada"},{"family":"Subbaswamy","given":"Adarsh"},{"family":"Sizikova","given":"Elena"},{"family":"Margerrison","given":"Edward"},{"family":"Delfino","given":"Jana"},{"family":"Badano","given":"Aldo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s44172-025-00450-1","URL":"https://doi.org/10.1038/s44172-025-00450-1","source":"openalex"},{"id":"oa:W4411098123","type":"article-journal","title":"Diagnosis melanoma with artificial intelligence systems: A meta‐analysis study and systematic review","abstract":"BACKGROUND: One of the most promising and rapidly advancing research areas in recent years is using dermoscopic images for automatic diagnosis with artificial intelligence and machine learning methods. OBJECTIVES: This study aimed to synthesize the existing studies for the clinical use of applications made with artificial intelligence methods and to summarize the predictive performance of deep learning and hybrid models-based algorithms in all these studies with a large-scale meta-analysis. METHOD: The literature review was conducted between January 2006 and May 2024, and meta-analysis data were created by scanning the Web of Science (WOS), Scopus and MEDLINE databases. This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist. RESULTS: A total of 2722 articles were evaluated. Data from 78 diagnostic tests from 39 primary studies meeting the inclusion and exclusion criteria were assessed. The pooled SROC overall model AUC was 0.96 [95% CI: 0.94-0.98], sensitivity was 0.89 [95% CI: 0.85-0.91] and specificity was 0.92 [95% CI: 0.90-0.94]. In the subgroup analyses, the pooled AUC was 0.98 [95% CI: 0.96-0.99] for HYBRID models. CONCLUSIONS: Recent studies have suggested that artificial intelligence algorithms and machine learning methods should be used extensively in medicine to assist physicians, especially in diagnosing melanoma. The ability of HYBRID model algorithms to predict diseases is promising. In particular, the performance of HYBRID models was found to be high. This information can assist clinicians in interpreting the most appropriate algorithms for diagnosing melanoma.","author":[{"family":"Zararsız","given":"Gözde"},{"family":"Zararsız","given":"Gözde"},{"family":"Yerlitaş","given":"Serra"},{"family":"Çeli̇k","given":"Elif"},{"family":"Erakcaoğlu","given":"Aleyna"},{"family":"Işıkhan","given":"Selen"},{"family":"Demirbaş","given":"Abdullah"},{"family":"Ertaş","given":"Ragıp"},{"family":"Eroğlu","given":"İrem"},{"family":"Korkmaz","given":"Selçuk"},{"family":"Elmas","given":"Ömer"},{"family":"Zararsız","given":"Gökmen"},{"family":"Zararsız","given":"Gökmen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/jdv.20781","URL":"https://doi.org/10.1111/jdv.20781","source":"openalex"},{"id":"oa:W7125141043","type":"article-journal","title":"Revolutionizing endodontics: the impact and innovations of artificial intelligence","abstract":"BACKGROUND: Artificial intelligence (AI) has been increasingly applied in the diagnosis and treatment of endodontic conditions. This narrative review summarizes the current literature on AI applications in endodontics from a clinical workflow perspective, discussing challenges in translating these approaches to practice and highlighting areas where further research is needed to better understand their effectiveness and limitations. METHODS: This narrative review summarizes applications and innovations of AI in endodontics up to July 2025. Relevant literature was identified through searches of PubMed and Web of Science using a combination of MeSH terms and free-text keywords covering areas such as diagnosis, image analysis, treatment planning, and prognosis prediction. Studies were selected for their relevance to clinical or experimental applications, and a few early, foundational studies were also included to provide context and show how AI has developed in this field. RESULTS: AI, leveraging its efficient image processing and pattern recognition capabilities, has demonstrated considerable advantages in endodontic imaging analysis. Evidence indicates that AI can assist clinicians in disease diagnosis, treatment planning, and prognosis assessment, thereby enhancing the quality of endodontic care across multiple clinical steps. Moreover, applications of AI in root canal anatomy identification, lesion detection, and integration into digital workflows further optimize clinical decision-making and procedural efficiency. CONCLUSIONS: Despite its promising potential in endodontics, the routine clinical application of AI remains limited by several factors, including the scarcity of publicly available databases, insufficient interpretability of algorithms, and ethical and privacy concerns. To achieve comprehensive integration of AI into clinical practice, future efforts could focus on multicenter validation studies, data sharing initiatives, development of interpretable AI models, and establishment of ethical and regulatory frameworks through interdisciplinary collaboration.","author":[{"family":"Huang","given":"Xin"},{"family":"Xu","given":"Zijia"},{"family":"Chen","given":"Han"},{"family":"Guo","given":"Xin"},{"family":"Yang","given":"Xuebin"},{"family":"Zhao","given":"Yuan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1186/s12903-025-07632-5","URL":"https://doi.org/10.1186/s12903-025-07632-5","source":"openalex"},{"id":"oa:W4414524030","type":"article-journal","title":"Artificial intelligence-driven approaches for the rational design of peptides with predictable aggregation propensity","abstract":"The rapid development of artificial intelligence has enabled accurate and efficient de novo design of protein and peptide structures. However, applying AI to design sequences with specific aggregation tendencies remains challenging due to the need to span multiple spatial-temporal scales. Here, we combined deep learning strategies—including genetic algorithms and reinforcement learning—to generate decapeptides with tunable aggregation propensities. Coarse-grained molecular dynamics simulations were used to evaluate solvent-accessible surface area and define aggregation propensity (AP). A Transformer-based prediction model with self-attention achieved high accuracy in AP prediction with only a 6% error rate. Furthermore, Monte Carlo Tree Search enabled targeted optimization of peptide sequences while preserving desired functional features. This study demonstrates how integrating AI with molecular modeling can guide the rational design of peptides with controlled assembly behavior, providing a scalable strategy for applications in biotechnology and medicine.","author":[{"family":"Yang","given":"Shuo"},{"family":"Ren","given":"Jing"},{"family":"Gao","given":"Wenli"},{"family":"Cao","given":"Leitao"},{"family":"Ling","given":"Shengjie"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s44431-025-00005-6","URL":"https://doi.org/10.1038/s44431-025-00005-6","source":"openalex"},{"id":"oa:W7127650232","type":"article-journal","title":"Generative artificial intelligence for literature reviews","abstract":"Generative artificial intelligence (GenAI), based on large-language models (LLMs), such as ChatGPT, has taken organizations, academia, and the public by storm. In particular, impressive GenAI capabilities such as summarization of large text corpora, question-answering, data extraction, and translation, carry profound implications for the conduct of literature reviews. This impacts science, organizations and the general public, as all can benefit from GenAI-supported literature reviews. Building on the technical foundations of GenAI and grounded in established methodological discourse, this work outlines approaches for conducting literature reviews using both general-purpose (e.g., ChatGPT, Gemini, Claude) and specialized GenAI tools (e.g., Consensus, Elicit). We provide illustrative examples of prompts and suggest methodologically-sound literature review strategies. Throughout this perspective paper, we adopt a balanced approach considering both the opportunities and the risks of relying on GenAI in the conduct of literature reviews. We conclude by discussing philosophical questions related to the effects of GenAI on long-term scientific progress, and also present fruitful opportunities for research on improving the core of GenAI’s technology—its architecture and training data—and suggest open issues in GenAI-based literature reviews methodology.","author":[{"family":"Wagner","given":"Gerit"},{"family":"Prester","given":"Julian"},{"family":"Mousavi","given":"Reza"},{"family":"Lukyanenko","given":"Roman"},{"family":"Pare","given":"Guy"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1177/02683962261425675","URL":"https://doi.org/10.1177/02683962261425675","source":"openalex"},{"id":"oa:W4407639400","type":"article-journal","title":"Advancements in Machine Learning and Artificial Intelligence in Polymer Science: A Comprehensive Review","abstract":"Abstract Technology, health care, and transport are merely some of the industries that historically rely on polymer‐based materials. In past centuries, the creation of innovative polymer materials has been dependent upon extensive experiments and error procedures that require an extensive number of resources as well as time. With the objective to explore the transformative potential of machine learning (ML) and artificial intelligence (AI) in material discovery, design, and optimization, this paper explores the integration of ML and AI in polymer‐based materials research. Researchers are able to speed the development of new polymer‐based materials with improved properties and functionalities by utilizing sophisticated algorithms and computational models. The use of ML and AI in polymer research is examined, with a focus on how these technologies may stimulate innovation and expand material science research.","author":[{"family":"Mavi","given":"Sheetal"},{"family":"Kadian","given":"SP"},{"family":"Sarangi","given":"Pradeepta"},{"family":"Sahoo","given":"Ashok"},{"family":"Singh","given":"Shruti"},{"family":"Yahya","given":"Muhd"},{"family":"Rahman","given":"Nor"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/masy.202400185","URL":"https://doi.org/10.1002/masy.202400185","source":"openalex"},{"id":"oa:W4407364174","type":"article-journal","title":"Medical multimodal multitask foundation model for lung cancer screening","abstract":"Lung cancer screening (LCS) reduces mortality and involves vast multimodal data such as text, tables, and images. Fully mining such big data requires multitasking; otherwise, occult but important features may be overlooked, adversely affecting clinical management and healthcare quality. Here we propose a medical multimodal-multitask foundation model (M3FM) for three-dimensional low-dose computed tomography (CT) LCS. After curating a multimodal multitask dataset of 49 clinical data types, 163,725 chest CT series, and 17 tasks involved in LCS, we develop a scalable multimodal question-answering model architecture for synergistic multimodal multitasking. M3FM consistently outperforms the state-of-the-art models, improving lung cancer risk and cardiovascular disease mortality risk prediction by up to 20% and 10% respectively. M3FM processes multiscale high-dimensional images, handles various combinations of multimodal data, identifies informative data elements, and adapts to out-of-distribution tasks with minimal data. In this work, we show that M3FM advances various LCS tasks through large-scale multimodal and multitask learning. Lung cancer screening (LCS) requires effectively and efficiently mining big, multimodal datasets. Here, the authors develop a medical multimodal-multitask foundation model (M3FM) for LCS from 3D low-dose computed tomography and medical multimodal data, outperforming state-of-the-art methods and allowing the identification of informative data elements.","author":[{"family":"Niu","given":"Chuang"},{"family":"Lyu","given":"Qing"},{"family":"Carothers","given":"Christopher"},{"family":"Kaviani","given":"Parisa"},{"family":"Tan","given":"Josh"},{"family":"Yan","given":"Pingkun"},{"family":"Kalra","given":"Mannudeep"},{"family":"Whitlow","given":"Christopher"},{"family":"Wang","given":"Ge"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41467-025-56822-w","URL":"https://doi.org/10.1038/s41467-025-56822-w","source":"openalex"},{"id":"oa:W4411149338","type":"article-journal","title":"A Practical Guide to Evaluating Artificial Intelligence Imaging Models in Scientific Literature","abstract":"Objective: Recent advances in artificial intelligence (AI) are revolutionizing ophthalmology by enhancing diagnostic accuracy, treatment planning, and patient management. However, a significant gap remains in practical guidance for ophthalmologists who lack AI expertise to effectively analyze these technologies and assess their readiness for integration into clinical practice. This paper aims to bridge this gap by demystifying AI model design and providing practical recommendations for evaluating AI imaging models in research publications. Design: Educational review: synthesizing key considerations for evaluating AI papers in ophthalmology. Participants: This paper draws on insights from an interdisciplinary team of ophthalmologists and AI experts with experience in developing and evaluating AI models for clinical applications. Methods: A structured framework was developed based on expert discussions and a review of key methodological considerations in AI research. Main Outcome Measures: A stepwise approach to evaluating AI models in ophthalmology, providing clinicians with practical strategies for assessing AI research. Results: This guide offers broad recommendations applicable across ophthalmology and medicine. Conclusions: As the landscape of health care continues to evolve, proactive engagement with AI will empower clinicians to lead the way in innovation while concurrently prioritizing patient safety and quality of care. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.","author":[{"family":"Mccarthy","given":"Angela"},{"family":"Valenzuela","given":"Ives"},{"family":"Chen","given":"Royce"},{"family":"Glass","given":"Lora"},{"family":"Thakoor","given":"Kaveri"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.xops.2025.100847","URL":"https://doi.org/10.1016/j.xops.2025.100847","source":"openalex"},{"id":"oa:W4410310579","type":"article-journal","title":"Empowering Generalist Material Intelligence with Large Language Models","abstract":"Large language models (LLMs) are steering the development of generalist materials intelligence (GMI), a unified framework integrating conceptual reasoning, computational modeling, and experimental validation. Central to this framework is the agent-in-the-loop paradigm, where LLM-based agents function as dynamic orchestrators, synthesizing multimodal knowledge, specialized models, and experimental robotics to enable fully autonomous discovery. Drawing from a comprehensive review of LLMs' transformative impact across representative applications in materials science, including data extraction, property prediction, structure generation, synthesis planning, and self-driven labs, this study underscores how LLMs are revolutionizing traditional tasks, catalyzing the agent-in-the-loop paradigm, and bridging the ontology-concept-computation-experiment continuum. Then the unique challenges of scaling up LLM adoption are discussed, particularly those arising from the misalignment of foundation LLMs with materials-specific knowledge, emphasizing the need to enhance adaptability, efficiency, sustainability, interpretability, and trustworthiness in the pursuit of GMI. Nonetheless, it is important to recognize that LLMs are not universally efficient. Their substantial resource demands and inconsistent performance call for careful deployment based on demonstrated task suitability. To address these realities, actionable strategies and a progressive roadmap for equitably and democratically implementing materials-aware LLMs in real-world practices are proposed.","author":[{"family":"Yuan","given":"Wenhao"},{"family":"Chen","given":"Guangyao"},{"family":"Wang","given":"Zhilong"},{"family":"You","given":"Fengqi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/adma.202502771","URL":"https://doi.org/10.1002/adma.202502771","source":"openalex"},{"id":"oa:W4410708925","type":"article-journal","title":"Artificial Intelligence in telemedicine and remote patient monitoring: Enhancing virtual healthcare through AI-driven diagnostic and predictive technologies","abstract":"The integration of Artificial Intelligence (AI) in telemedicine and remote patient monitoring has significantly transformed modern healthcare by enhancing accessibility, efficiency, and diagnostic precision. AI-powered technologies, including machine learning algorithms, predictive analytics, and natural language processing, have facilitated real-time health monitoring, early disease detection, and personalized treatment recommendations. The incorporation of AI-driven chatbot and virtual assistants has streamlined remote consultations, enabling healthcare professionals to manage patient inquiries more efficiently while ensuring timely medical interventions. Additionally, wearable health devices embedded with AI capabilities provide continuous monitoring of vital signs, allowing for proactive management of chronic diseases such as diabetes, hypertension, and cardiovascular disorders. AI-driven predictive models are also being utilized to assess patient risk factors, forecast potential health complications, and optimize treatment plans. Furthermore, AI has enhanced telemedicine by automating administrative processes, reducing operational costs, and expanding healthcare services to remote and underserved populations. However, the adoption of AI in telemedicine presents challenges related to data security, ethical considerations, regulatory compliance, and patient privacy, which require careful evaluation. As AI continues to evolve, its integration with telemedicine and remote patient monitoring is expected to revolutionize digital healthcare, providing innovative solutions to improve patient outcomes, optimize healthcare workflows, and bridge the gap between patients and medical professionals in an increasingly digital world.","author":[{"family":"Sarkar","given":"Malay"},{"family":"Dey","given":"Raktim"},{"family":"Mia","given":"Md"}],"issued":{"date-parts":[[2025]]},"DOI":"10.30574/ijsra.2025.15.2.1402","URL":"https://doi.org/10.30574/ijsra.2025.15.2.1402","source":"openalex"},{"id":"oa:W4410456003","type":"article-journal","title":"Software with artificial intelligence-derived algorithms for detecting and analysing lung nodules in CT scans: systematic review and economic evaluation","abstract":"Background: Lung cancer is one of the most common types of cancer and the leading cause of cancer death in the United Kingdom. Artificial intelligence-based software has been developed to reduce the number of missed or misdiagnosed lung nodules on computed tomography images. Objective: To assess the accuracy, clinical effectiveness and cost-effectiveness of using software with artificial intelligence-derived algorithms to assist in the detection and analysis of lung nodules in computed tomography scans of the chest compared with unassisted reading. Design: Systematic review and de novo cost-effectiveness analysis. Methods: Searches were undertaken from 2012 to January 2022. Company submissions were accepted until 31 August 2022. Study quality was assessed using the revised tool for the quality assessment of diagnostic accuracy studies (QUADAS-2), the extension to QUADAS-2 for assessing risk of bias in comparative accuracy studies (QUADAS-C) and the COnsensus-based Standards for the selection of health status Measurement INstruments (COSMIN) checklist. Outcomes were synthesised narratively. Two decision trees were used for cost-effectiveness: (1) a simple decision tree for the detection of actionable nodules and (2) a decision tree reflecting the full clinical pathways for people undergoing chest computed tomography scans. Models estimated incremental cost-effectiveness ratios, cost per correct detection of an actionable nodule, and cost per cancer detected and treated. We undertook scenario and sensitivity analyses. Results: Twenty-seven studies were included. All were rated as being at high risk of bias. Twenty-four of the included studies used retrospective data sets. Seventeen compared readers with and without artificial intelligence software. One reported prospective screening experiences before and after artificial intelligence software implementation. The remaining studies either evaluated stand-alone artificial intelligence or provided only non-comparative evidence. (1) Artificial intelligence assistance generally improved the detection of any nodules compared with unaided reading (three studies; average per-person sensitivity 0.43-0.68 for unaided and 0.79-0.99 for artificial intelligence-assisted reading), with similar or lower specificity (three studies; 0.77-1.00 for unaided and 0.81-0.97 for artificial intelligence-assisted reading). Nodule diameters were similar or significantly larger with semiautomatic measurements than with manual measurements. Intra-reader and inter-reader agreement in nodule size measurement and in risk classification generally improved with artificial intelligence assistance or were comparable to those with unaided reading. However, the effect on measurement accuracy is unclear. (2) Radiologist reading time generally decreased with artificial intelligence assistance in research settings. (3) Artificial intelligence assistance tended to increase allocated risk categories as defined by clinical guidelines. (4) No relevant clinical effectiveness and cost-effectiveness studies were identified. (5) The de novo cost-effectiveness analysis suggested that for symptomatic and incidental populations, artificial intelligence-assisted computed tomography image analysis dominated the unaided radiologist in cost per correct detection of an actionable nodule. However, when relevant costs and quality-adjusted life-years from the full clinical pathway were included, artificial intelligence-assisted computed tomography reading was dominated by the unaided reader. For screening, artificial intelligence-assisted computed tomography image analysis was cost-effective in the base case and all sensitivity and scenario analyses. Limitations: Due to the heterogeneity, sparseness, low quality and low applicability of the clinical effectiveness evidence and the major challenges in linking test accuracy evidence to clinical and economic outcomes, the findings presented here are highly uncertain and provide indicators/fr","author":[{"family":"Geppert","given":"Julia"},{"family":"Auguste","given":"Peter"},{"family":"Asgharzadeh","given":"Asra"},{"family":"Ghiasvand","given":"Hesam"},{"family":"Patel","given":"Mubarak"},{"family":"Brown","given":"Anna"},{"family":"Jayakody","given":"Surangi"},{"family":"Helm","given":"Emma"},{"family":"Todkill","given":"Daniel"},{"family":"Madan","given":"Jason"},{"family":"Stinton","given":"Chris"},{"family":"Gallacher","given":"Daniel"},{"family":"Taylorphillips","given":"Sian"},{"family":"Chen","given":"Yen‐fu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3310/jytw8921","URL":"https://doi.org/10.3310/jytw8921","source":"openalex"},{"id":"oa:W4411647611","type":"article-journal","title":"Bioethical challenges in the integration of artificial intelligence in transplant surgery 4.0: A scoping review","abstract":"Introduction The integration of artificial intelligence (AI) into transplant surgery offers potential benefits, including enhanced precision and personalized patient care. However, these advancements raise critical ethical issues that must be addressed to ensure responsible implementation. This scoping review and bibliometric analysis explore the literature on the ethical considerations associated with the application of AI in transplant surgery. Methods Following Joanna Briggs Institute and PRISMA-ScR guidelines, we conducted a systematic search of databases, including PubMed, Scopus, Science Direct, JSTOR, LILACS, IEEE, and GreyNet from 2013 to 2025. We identified articles that discussed ethical implications in English or Spanish. Data were charted using Rayyan (Rayyan Systems Inc., Doha, Qatar) and analyzed with Biblioshiny . One reviewer reviewed each article twice to ensure accuracy. Results Our search identified 6824 records, of which 16 studies met the selection criteria. The bibliometric analysis revealed a significant increase in scholarly output on AI ethics in transplantation since 2020. Key ethical concerns comprehend dehumanization of medical care, limitations in AI interpretability, and erroneous decision-making. The potential benefits highlighted include improved donor-recipient matching and personalized patient care. However, the need for human oversight in AI applications is emphasized mitigating risks such as patient dehumanization and biased decision-making. Conclusion This review is the first to comprehensively map the ethical landscape of AI integration in transplant surgery. It identifies both the potential and the significant ethical challenges of these technologies. Future research should focus on developing frameworks for ethical AI implementation and ensuring that advancements in AI contribute to equitable and just healthcare practices.","author":[{"family":"Lozano-Suárez","given":"Nicolás"},{"family":"Gómez-Montero","given":"Andrea"},{"family":"Jiménez-Gómez","given":"Maritza"},{"family":"Cabas","given":"Santiago"},{"family":"Giron-Londoño","given":"Nicolás"},{"family":"García-López","given":"Andrea"},{"family":"Girón-Luque","given":"Fernando"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1177/20552076251351700","URL":"https://doi.org/10.1177/20552076251351700","source":"openalex"},{"id":"oa:W4417178901","type":"article-journal","title":"Artificial intelligence in predicting anti-VEGF treatment response in diabetic macular edema: current progress and future directions","abstract":"Diabetic macular edema (DME), a leading cause of vision impairment in diabetes, is primarily treated with intravitreal anti-vascular endothelial growth factor (anti-VEGF) injections. However, variable treatment response rates often lead to persistent edema and irreversible vision loss. Accurate prediction of treatment response is therefore critical for optimizing treatment strategies and preserving visual function. This review examines the application of artificial intelligence (AI) to predict anti-VEGF treatment outcomes in DME. While optical coherence tomography (OCT) imaging has shown significant progress, the potential of optical coherence tomography angiography (OCTA) and fluorescein angiography (FA) remains under-exploited. AI holds considerable promise for enhancing predictive accuracy. Future research should focus on multimodal imaging approaches integrating structural, ischemic, and vascular information to develop more accurate and reliable predictive models for personalized DME treatment.","author":[{"family":"Cao","given":"Dan"},{"family":"Yao","given":"Jie"},{"family":"Ting","given":"Daniel"},{"family":"Tan","given":"Gavin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.48130/vns-0025-0027","URL":"https://doi.org/10.48130/vns-0025-0027","source":"openalex"},{"id":"oa:W4410763127","type":"article-journal","title":"Seven Opportunities for Artificial Intelligence in Primary Care Electronic Visits: Qualitative Study of Staff and Patient Views","abstract":"PURPOSE: Increased workload associated with electronic visits (eVisits) in primary care could potentially be decreased by the use of artificial intelligence (AI); however, it is unknown whether this use of AI would be acceptable to staff and patients. We explored patient and primary care staff views on the use of and opportunities for AI during eVisits. METHODS: ) from May 2020 to September 2021. We analyzed verbatim transcripts using thematic analysis. RESULTS: Misconceptions regarding AI were common, which led to initial reservations on its use during eVisits. Perceived potential AI benefits included decreased staff workload and faster response times for patients. Safety concerns stemmed from the complexity of primary care and fears of depersonalized service. The following 7 opportunities for AI during eVisits were identified: workflow, directing, prioritization, asking questions, writing assistance, providing self-help information, and face-to-face appointment booking. Despite staff concerns regarding patient acceptability, most patients welcomed the use of AI if it were used as an adjunct to (not replacement for) clinical judgment and could support them in getting help more quickly. Retention of clinical oversight and ongoing evaluation was key to staff acceptability. CONCLUSIONS: Patients and staff welcomed the use of AI and identified 7 potential uses during eVisits to decrease staff workload and improve patient safety. Successful implementation will depend on clear communication from practices, demonstrating and monitoring safety, clarifying misconceptions, and reassuring that it will not replace humans.","author":[{"family":"Moschogianis","given":"Susan"},{"family":"Darley","given":"Sarah"},{"family":"Coulson","given":"Tessa"},{"family":"Peek","given":"Niels"},{"family":"Cheraghisohi","given":"Sudeh"},{"family":"Brown","given":"Benjamin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1370/afm.240292","URL":"https://doi.org/10.1370/afm.240292","source":"openalex"},{"id":"oa:W4411043515","type":"article-journal","title":"Generative artificial intelligence for sustainable development: predictive trend analysis in key sectors using natural language processing","abstract":"Generative Artificial Intelligence (GenAI) has revolutionized multiple industries by improving efficiency and promoting innovation in fields such as business, education, healthcare, and cybersecurity. This study evaluates the transformative impact of GenAI on advancing the United Nations' Sustainable Development Goals (SDGs) by analyzing research trends and applications. In this research, authors utilized Latent Dirichlet Allocation (LDA), a Natural Language Processing (NLP) technique. The work delineates emergent themes and sector-specific progressions in GenAI research. The dataset consists of 2162 research articles published from 2020 to 2025, obtained from the Scopus database. The abstracts of these publications provide the foundation for analysis, facilitating a thorough examination of literature. The authors provided 10 topics, which are recent trends that future researchers can explore. The results indicate GenAI's capability in processing automation, creative enhancement, and innovation promotion while addressing ethical concerns such as prejudice, privacy, and social effects. The study indicates potential directions for ethical technology adoption by analyzing trends in GenAI applications. This work highlights the essential requirement for interdisciplinary methods and ethical frameworks to optimize GenAI's contributions to innovation while assuring alignment with sustainable and equitable development goals.","author":[{"family":"Sharma","given":"Chetan"},{"family":"Sharma","given":"Shamneesh"},{"family":"Bhardwaj","given":"Vivek"},{"family":"Dhaliwal","given":"Balwinder"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s42452-025-07207-7","URL":"https://doi.org/10.1007/s42452-025-07207-7","source":"openalex"},{"id":"oa:W4413078261","type":"article-journal","title":"Context-Aware Retrieval-Augmented Generation for Artificial Intelligence in Urology","abstract":"Background Artificial intelligence (AI) is increasingly being used in healthcare, particularly for interpreting complex medical queries. However, conventional AI models often generate inaccurate or irrelevant responses that are commonly termed hallucinations, which may compromise patient safety. To address this, our study introduces a modified retrieval-augmented generation (RAG) framework tailored for the urology domain to enhance contextual relevance and accuracy in AI-generated responses. Methodology We developed a context-aware RAG system integrating PubMedBERT embeddings for encoding and retrieving urological literature stored in a Pinecone vector database. The system uses named entity recognition for domain-specific query filtering and incorporates dynamic memory to retain contextual flow during interactions. Response generation is powered by the LLaMA3-8B model via LangChain. A custom dataset of urology-related queries was used for evaluation, with a large language model-based scoring using the Deepseek-R1 model. Results The proposed framework demonstrated a significant reduction in hallucinations, with responses being more contextually relevant and evidence-based. Compared to baseline models, our system achieved an 89% performance improvement in generating medically appropriate answers. Integration of memory modules and named entity filtering further improved precision and reliability. Conclusions Our RAG-enhanced system shows strong potential for clinical use by producing trustworthy, context-aware responses in urology. It addresses key challenges in medical AI, including hallucination mitigation and domain relevance. Future work will focus on reducing inference latency and improving automated validation without manual oversight.","author":[{"family":"Sriram","given":"Aadhitya"},{"family":"Maheswaran","given":"N"},{"family":"Sundan","given":"Bose"},{"family":"Krishnamoorthy","given":"Sriram"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.88167","URL":"https://doi.org/10.7759/cureus.88167","source":"openalex"},{"id":"oa:W7125963307","type":"article-journal","title":"Multi‐model Artificial Intelligence Evaluation in Sudden Sensorineural Hearing Loss","abstract":"OBJECTIVE: To compare the diagnostic accuracy, linguistic clarity, and user satisfaction of three large language models (ChatGPT-4.0, Claude 3.7 Sonet, and OpenAI Mini 3) in managing sudden sensorineural hearing loss. STUDY DESIGN: Prospective, multi-domain comparative analysis using blinded expert evaluation. SETTING: Online artificial intelligence (AI) platforms accessed under standardized conditions. METHODS: Twenty-seven sudden sensorineural hearing loss-related questions-covering general knowledge, audiometric interpretation, and clinical case scenarios-were submitted to the three AI models. Responses were evaluated by 10 board-certified otolaryngologists using three validated tools: Quality Assessment of Medical Artificial Intelligence (QAMAI), Artificial Intelligence Performance Instrument (AIPI), and Artificial Intelligence Satisfaction and Performance Evaluation Questionnaire (AISPE-Q). Linguistic complexity was assessed using metrics such as word count, sentence length, lexical diversity, and clinical verb use. RESULTS: ChatGPT-4.0 demonstrated the highest scores in clinical accuracy (QAMAI: 4.57), completeness (4.53), and evaluator satisfaction (AISPE-Q: 94%). Claude 3.7 outperformed in clarity and sentence complexity, while OpenAI Mini 3 exhibited the highest lexical diversity and directive tone but scored lower overall. Inter-rater reliability was strong (intraclass correlation coefficient [ICC] > 0.85). Correlation analysis revealed a significant relationship between objective quality and subjective satisfaction (r > 0.76). CONCLUSION: ChatGPT-4.0 delivered the most clinically aligned and satisfactory responses, whereas Claude 3.7 provided linguistically refined outputs. Our findings support the context-specific application of hybrid large language model approaches in otolaryngology, particularly for patient education, diagnosis, and AI-driven triage. LEVEL OF EVIDENCE: 2-prospective comparative diagnostic accuracy study.","author":[{"family":"Aliyeva","given":"Aynur"},{"family":"Muradova","given":"Antiga"},{"family":"Hashimli","given":"Ramil"},{"family":"Müderris","given":"Togay"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1002/ohn.70143","URL":"https://doi.org/10.1002/ohn.70143","source":"openalex"},{"id":"oa:W4416307141","type":"article-journal","title":"Building Symbiotic Artificial Intelligence: Reviewing the AI Act for a Human-Centred, Principle-Based Framework","abstract":"Abstract Artificial Intelligence (AI) spreads quickly as new technologies and services take over modern society. The need to regulate AI design, development, and use is strictly necessary to avoid unethical and potentially dangerous consequences to humans. The European Union (EU) has released a new legal framework, the AI Act, to regulate AI by undertaking a risk-based approach to safeguard humans during interaction. At the same time, researchers offer a new perspective on AI systems, commonly known as Human-Centred AI (HCAI), highlighting the need for a human-centred approach to their design. In this context, Symbiotic AI (a subtype of HCAI) promises to enhance human capabilities through a deeper and continuous collaboration between human intelligence and AI. This article presents the results of a Systematic Literature Review (SLR) that aims to identify principles that characterise the design and development of Symbiotic AI systems while considering humans as the core of the process. Through content analysis, we elicit four principles that must be applied to create Human-Centred AI systems that can establish a symbiotic relationship with humans. In addition, current trends and challenges are presented to indicate open questions that may guide future research for the development of SAI systems that comply with the AI Act.","author":[{"family":"Calvano","given":"Miriana"},{"family":"Curci","given":"Antonio"},{"family":"Desolda","given":"Giuseppe"},{"family":"Esposito","given":"Andrea"},{"family":"Lanzilotti","given":"Rosa"},{"family":"Piccinno","given":"Antonio"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s11023-025-09753-w","URL":"https://doi.org/10.1007/s11023-025-09753-w","source":"openalex"},{"id":"oa:W7116985164","type":"article-journal","title":"Applications of artificial intelligence in non–small cell lung cancer: from precision diagnosis to personalized prognosis and therapy","abstract":"BACKGROUND: Non-small cell lung cancer (NSCLC) carries a major global burden. The rapid growth of multimodal medical data challenges conventional methods to deliver stable, transferable and interpretable decisions across heterogeneous longitudinal high dimensional inputs. METHODS: This review summarizes advances in artificial intelligence (AI) for NSCLC from 2023 to 2025 and outlines a translation-focused framework that links algorithmic progress to clinical utility. We survey thoracic imaging, digital pathology and multiomics together with evaluation practices and implementation guidance. We also adopt a critical perspective. RESULTS: Many high performing deep models remain black boxes, and popular post hoc explanations such as Grad CAM heatmaps are rarely validated for faithfulness or stability, which undermines clinician trust and limits use in high stakes decisions. To address this gap, we propose a minimum evidence package for explainability that comprises sanity checks, quantitative faithfulness tests such as deletion or insertion, ROAR or IROF and infidelity, stability analyses, concept level validation for example TCAV with statistical testing, and prospective human factors studies that demonstrate improved decisions without automation bias. Across modalities, evaluation has expanded beyond discrimination to include calibration, uncertainty quantification (UQ) and subgroup analyses across scanners, sites and populations. However, the evidence base remains constrained by retrospective single center designs, inconsistent external or temporal validation and limited decision curve analysis (DCA). Translational priorities include a staged validation ladder from technical to clinical to prospective deployment, alignment with Software as a Medical Device frameworks, interoperable governance, fairness and economic assessment, and validated explainability coupled with uncertainty aware selective workflows. CONCLUSIONS: Looking ahead, progress will depend on multimodal foundation models, causal and temporal modeling, and regulatory qualification of computable biomarkers with verified explanations, supported by multicenter prospective studies that demonstrate durable generalizability, clinical value and clinician trust.","author":[{"family":"Chang","given":"Luyuan"},{"family":"Li","given":"Haipeng"},{"family":"Wu","given":"Wei"},{"family":"Liu","given":"X"},{"family":"Yan","given":"Jiaqi"},{"family":"Chen","given":"Zuo"},{"family":"Wu","given":"Huan"},{"family":"Song","given":"Shilong"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12967-025-07591-z","URL":"https://doi.org/10.1186/s12967-025-07591-z","source":"openalex"},{"id":"oa:W4410176648","type":"article-journal","title":"From traditional to artificial intelligence-driven approaches: Revolutionizing personalized and precision nutrition in inflammatory bowel disease","abstract":"Inflammatory bowel disease (IBD), comprising ulcerative colitis and Crohn's disease, is a chronic inflammatory condition with global prevalence and varying incidence. The IBD pathogenesis involves intricate interactions among genetic, host and environmental factors, leading to dysregulated immune responses and chronic intestinal inflammation. Alongside elevated levels of inflammatory cytokines and altered miRNAs expression, more studies highlight significant dysbiosis in both fecal and ileal microbiota of IBD patients. This dysbiosis is characterized by an increase in pro-inflammatory and mucin-degrading bacteria (e.g., Fusobacterium spp., Escherichia spp.) and a decline in short-chain fatty acids (SCFAs) -producing microbes (e.g., Roseburia spp., Faecalibacterium spp.) which play a protective role in gut health. Diet emerges as a key environmental factor influencing IBD onset and progression and recent advancements in\"omics\" technologies, such as genomics, transcriptomics, and metabolomics, provide a deeper understanding of the molecular interactions between genes, gut microbiota (GM) and nutrition. Finally, new technologies like artificial intelligence (AI), further enhance findings by enabling data integration and personalized dietary strategies. In this scenario, this review aims to summarize accumulating data on the effects of dietary interventions in IBD patients and introduce the role of artificial intelligence (AI) in facilitating precision dietary approaches to improve IBD management.","author":[{"family":"Baldi","given":"Simone"},{"family":"Sarikaya","given":"Dilara"},{"family":"Lotti","given":"Sofia"},{"family":"Cuffaro","given":"Francesca"},{"family":"Fink","given":"Dorian"},{"family":"Colombini","given":"Barbara"},{"family":"Sofi","given":"Francesco"},{"family":"Amedei","given":"Amedeo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.clnesp.2025.05.012","URL":"https://doi.org/10.1016/j.clnesp.2025.05.012","source":"openalex"},{"id":"oa:W7124443394","type":"article-journal","title":"Can Generative Artificial Intelligence Effectively Enhance Students’ Mathematics Learning Outcomes?—A Meta-Analysis of Empirical Studies from 2023 to 2025","abstract":"Generative artificial intelligence (GenAI) shows transformative potential in mathematics education. However, empirical findings remain inconsistent, and a systematic synthesis of its effects across distinct engagement dimensions is lacking. This preregistered meta-analysis (INPLASY2025110051) systematically reviewed 22 empirical studies (46 independent samples, N = 5232) published between 2023 and 2025. The results indicated that GenAI has a moderate positive impact on students’ mathematics learning outcomes (g = 0.534). Moderation analysis further revealed that the level of GenAI integration in teaching, sample size, and learning content are the primary factors influencing this effect. The study found that the effect was most pronounced under the creative transformation (CT) integration mode, was significant when applied to geometry learning, and was stronger in studies with small samples or small class sizes; collaborative learning approaches also significantly enhance these mathematics learning outcomes. By contrast, educational stage and intervention duration did not show significant moderating effects. The GRADE assessment indicated that while the overall evidence is supportive, the certainty of evidence is stronger for cognitive outcomes than for non-cognitive domains. The findings also offer a reference for future research on constructing a human–machine collaborative learning environment.","author":[{"family":"Liu","given":"Baoxin"},{"family":"Zhang","given":"Wenlan"},{"family":"Wang","given":"Fangfang"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/educsci16010140","URL":"https://doi.org/10.3390/educsci16010140","source":"openalex"},{"id":"oa:W4414552918","type":"article-journal","title":"Towards responsible artificial intelligence in healthcare—getting real about real-world data and evidence","abstract":"BACKGROUND: The use of real-world data (RWD) in artificial intelligence (AI) applications for healthcare offers unique opportunities but also poses complex challenges related to interpretability, transparency, safety, efficacy, bias, equity, privacy, ethics, accountability, and stakeholder engagement. METHODS: A multi-stakeholder expert panel comprising healthcare professionals, AI developers, policymakers, and other stakeholders was assembled. Their task was to identify critical issues and formulate consensus recommendations, focusing on the responsible use of RWD in healthcare AI. The panel's work involved an in-person conference and workshop and extensive deliberations over several months. RESULTS: The panel's findings revealed several critical challenges, including the necessity for data literacy and documentation, the identification and mitigation of bias, privacy and ethics considerations, and the absence of an accountability structure for stakeholder management. To address these, the panel proposed a series of recommendations, such as the adoption of metadata standards for RWD sources, the development of transparency frameworks and instructional labels likened to \"nutrition labels\" for AI applications, the provision of cross-disciplinary training materials, the implementation of bias detection and mitigation strategies, and the establishment of ongoing monitoring and update processes. CONCLUSION: Guidelines and resources focused on the responsible use of RWD in healthcare AI are essential for developing safe, effective, equitable, and trustworthy applications. The proposed recommendations provide a foundation for a comprehensive framework addressing the entire lifecycle of healthcare AI, emphasizing the importance of documentation, training, transparency, accountability, and multi-stakeholder engagement.","author":[{"family":"Koski","given":"Eileen"},{"family":"Das","given":"Amar"},{"family":"Hsueh","given":"Pei"},{"family":"Solomonides","given":"Anthony"},{"family":"Joseph","given":"Amanda"},{"family":"Srivastava","given":"Gyana"},{"family":"Johnson","given":"Cheyenne"},{"family":"Kannry","given":"Joseph"},{"family":"Oladimeji","given":"Bilikis"},{"family":"Price","given":"Amy"},{"family":"Labkoff","given":"Steven"},{"family":"Bharathy","given":"Gnana"},{"family":"Lin","given":"Baihan"},{"family":"Fridsma","given":"Douglas"},{"family":"Fleisher","given":"Lee"},{"family":"López-González","given":"Mónica"},{"family":"Singh","given":"Reva"},{"family":"Weiner","given":"Mark"},{"family":"Stolper","given":"Robert"},{"family":"Baris","given":"Russell"},{"family":"Sincavage","given":"Suzanne"},{"family":"Naumann","given":"Tristan"},{"family":"Williams","given":"TA"},{"family":"Bui","given":"Tien"},{"family":"Quintana","given":"Yuri"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/jamia/ocaf133","URL":"https://doi.org/10.1093/jamia/ocaf133","source":"openalex"},{"id":"oa:W7117479576","type":"article-journal","title":"Artificial Intelligence in Intensive Care: An Overview of Systematic Reviews with Clinical Maturity and Readiness Mapping","abstract":"Background: ICU care is time critical and data dense, making it a promising but high-risk setting for AI decision support when tools are weakly validated. ICU AI evidence is heterogeneous, with limited external validation, inconsistent clinically actionable reporting, and scarce real-world impact data, yielding fragmented review conclusions. We mapped five prespecified ICU domains and assessed clinical and implementation maturity to identify key translational gaps. Methods: We performed a PRIOR-aligned overview of systematic reviews with prespecified maturity constructs. PubMed, Embase, and Web of Science were searched (title and abstract) on 13 December 2025, supplemented by backward citation searching. Two reviewers screened and extracted data with arbitration, assessed the review-level risk of bias using ROBIS, and synthesized findings without meta-analysis using a SWiM-guided narrative prioritizing AUROC ranges. Results: We included 34 systematic reviews (2017–2025) across five ICU domains, dominated by prognostic and early warning applications, mostly in adult populations and commonly using EHR and multimodal inputs. Reporting focused on discrimination, with AUROC ranges roughly 0.54–0.99 for prognostic tasks and 0.64–0.99 for diagnostic tasks, while calibration and clinical utility were rarely addressed and overlap suggested partial dependence. Maturity signals clustered at low-to-intermediate levels, with no evidence for routine, and regulated CDS deployment at the review level. Conclusions: Review-level evidence indicates a translational gap between retrospective performance and clinically mature, safely deployable ICU AI, supporting priorities for external validation, prospective impact studies, standardized reporting including calibration, and governance-focused implementation.","author":[{"family":"Żerdziński","given":"Krzysztof"},{"family":"Janiec","given":"Julita"},{"family":"Jóźwik","given":"Kamil"},{"family":"Łajczak","given":"Paweł"},{"family":"Krzych","given":"Łukasz"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jcm15010185","URL":"https://doi.org/10.3390/jcm15010185","source":"openalex"},{"id":"oa:W7129047016","type":"article-journal","title":"Current Applications and Future Perspectives of Artificial Intelligence in Face-Driven Orthodontics: A Scoping Review","abstract":"Artificial Intelligence (AI) has introduced transformative possibilities in orthodontics by enhancing diagnostic precision, treatment planning, and aesthetic outcomes. In face-driven orthodontics, treatment objectives extend beyond achieving proper occlusion to optimizing facial balance and harmony. With the growing patient demand for aesthetic improvements, AI technologies enable clinicians to integrate facial analysis and dynamic soft-tissue evaluation into personalized treatment approaches. Research in this scoping review analyzed current applications of AI in face-driven orthodontics, focusing on diagnosis, soft-tissue assessment, and individualized treatment planning. A comprehensive search was conducted in PubMed and Scopus for studies published between 2021 and 2025. The review followed the PRISMA-ScR guidelines. Of 54 initially identified studies, 24 met the inclusion criteria after title, abstract, and full-text screening. Extracted data were organized according to the main application areas of AI in face-driven orthodontics. Most studies focused on AI-assisted facial analysis, 3D reconstruction, and treatment simulation. Deep learning models demonstrated high performance in soft-tissue prediction, aesthetic evaluation, and diagnostic accuracy. However, heterogeneity in datasets, a lack of standardized validation protocols, limited external validation across included studies and limited clinical applicability were identified as key limitations. AI-based facial analysis supports a shift toward individualized, aesthetics-oriented orthodontic planning. Although current evidence highlights its potential for improving diagnostic precision and treatment outcomes, further validation through large-scale clinical studies is essential for broader implementation in everyday practice.","author":[{"family":"Heribanová","given":"Barbora"},{"family":"Janáková","given":"Katarína"},{"family":"Tomášik","given":"Juraj"},{"family":"Tichá","given":"Daniela"},{"family":"Harsányi","given":"Štefan"},{"family":"Thurzo","given":"Andrej"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/biomimetics11020146","URL":"https://doi.org/10.3390/biomimetics11020146","source":"openalex"},{"id":"oa:W4414068771","type":"article-journal","title":"Artificial Intelligence in Adult Congenital Heart Disease: Diagnostic and Therapeutic Applications and Future Directions","abstract":"Adult congenital heart disease (ACHD) constitutes a heterogeneous and expanding patient cohort with distinctive diagnostic and management challenges. Conventional detection methods are ineffective at reflecting lesion heterogeneity and the variability in risk profiles. Artificial intelligence (AI), including machine learning (ML) and deep learning (DL) models, has revolutionized the potential for improving diagnosis, risk stratification, and personalized care across the ACHD spectrum. This narrative review discusses the current and future applications of AI in ACHD, including imaging interpretation, electrocardiographic analysis, risk stratification, procedural planning, and long-term care management. AI has been demonstrated as being highly accurate in congenital anomaly detection by various imaging modalities, automating measurement, and improving diagnostic consistency. Moreover, AI has been utilized in electrocardiography to detect previously undetected defects and estimate arrhythmia risk. Risk-prediction models based on clinical and imaging information can estimate stroke, heart failure, and sudden cardiac death as outcomes, thereby informing personalized therapy choices. AI also contributes to surgery and interventional planning through three-dimensional (3D) modelling and image fusion, while AI-powered remote monitoring tools enable the detection of early signals of clinical deterioration. While these insights are encouraging, limitations in data availability, algorithmic bias, a lack of prospective validation, and integration issues remain to be addressed. Ethical considerations of transparency, privacy, and responsibility should also be highlighted. Thus, future initiatives should prioritize data sharing, explainability, and clinician training to facilitate the secure and effective use of AI. The appropriate integration of AI can enhance decision-making, improve efficiency, and deliver individualized, high-quality care to ACHD patients.","author":[{"family":"Antoun","given":"Ibrahim"},{"family":"Nizam","given":"Ali"},{"family":"Ebeid","given":"Armia"},{"family":"Rajesh","given":"MC"},{"family":"Abdelrazik","given":"Ahmed"},{"family":"Eldesouky","given":"Mahmoud"},{"family":"Thu","given":"Kaung"},{"family":"Barker","given":"Joseph"},{"family":"Layton","given":"Georgia"},{"family":"Zakkar","given":"Mustafa"},{"family":"Ibrahim","given":"Mokhtar"},{"family":"Safwan","given":"Kassem"},{"family":"Dibek","given":"Radek"},{"family":"Somani","given":"Riyaz"},{"family":"Ng","given":"GA"},{"family":"Bolger","given":"Aidan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.31083/rcm41523","URL":"https://doi.org/10.31083/rcm41523","source":"openalex"},{"id":"oa:W4410417117","type":"article-journal","title":"Economic implications of artificial intelligence-driven recommended systems in healthcare: a focus on neurological disorders","abstract":"Introduction: The rapid advancement of Artificial Intelligence (AI)-driven recommendation systems in healthcare presents significant economic implications, particularly in the context of neurological disorders. These systems offer opportunities to enhance diagnostic accuracy, optimize resource allocation, and improve patient outcomes. However, conventional economic models fail to address the dynamic complexities of AI integration in healthcare, including market inefficiencies and stakeholder behaviors. Methods: To bridge this gap, we propose a Dynamic Equilibrium Model for Health Economics (DEHE), incorporating reinforcement learning and stochastic optimization. This model captures uncertainty in healthcare decision-making and includes dynamic pricing, behavioral incentives, and adaptive insurance premium mechanisms. Results: Our experimental results demonstrate that DEHE improves economic efficiency by optimizing AI-driven recommendations while balancing healthcare cost and accessibility. Through multi-agent simulations, the model shows strong real-world applicability and stability. It effectively addresses asymmetric information, moral hazard, and market dynamics. Discussion: This study offers a novel economic framework for integrating AI-driven systems in neurological healthcare. We recommend the adoption of adaptive policy mechanisms and stakeholder-specific incentives to enhance cost-effectiveness and equitable access. These insights contribute to the development of more sustainable and inclusive AI-based healthcare policies.","author":[{"family":"Zhang","given":"Jing"},{"family":"Xiang","given":"Shihui"},{"family":"Li","given":"Li"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fpubh.2025.1588270","URL":"https://doi.org/10.3389/fpubh.2025.1588270","source":"openalex"},{"id":"oa:W4416538550","type":"article-journal","title":"Artificial Intelligence in Autism Spectrum Disorder Diagnosis: A Scoping Review of Face, Voice, and Text Analysis Methods","abstract":"Background: Autism is a complex neurodevelopmental condition affecting social interaction and behavior. Traditional diagnostic methods, relying on observational techniques and interviews conducted by trained professionals, remain the gold standard for ASD diagnosis. However, these methods can be time-consuming and may be influenced by subjective factors. Recent advancements in artificial intelligence (AI) offer promising approaches to augment existing methods, potentially enhancing efficiency and providing additional objective data through facial, vocal, and textual analysis. Objective: The objective of this study was to conduct a comprehensive review of artificial intelligence applications in autism spectrum disorder (ASD) diagnosis, specifically focusing on facial, vocal, and textual analysis methods. Methods: A comprehensive search was conducted in PubMed, Web of Science, Scopus, and Google Scholar. The findings were reported in accordance with the PRISMA checklist. Data were collated and summarized, and results were reported qualitatively, adopting a narrative synthesis approach. Results: In facial image analysis, deep learning algorithms demonstrated high accuracy in identifying autism-related facial features, algorithms such as Xception achieved 98% accuracy, while hybrid approaches like the combination of Random Forest (RF) and VGG16-MobileNet showed accuracy at 99%. Voice analysis studies utilized both traditional machine learning methods and advanced deep learning techniques, achieving accuracies between 70% and 98% in detecting atypical speech patterns and prosodic abnormalities associated with autism. Text-based analyses showed potential in identifying linguistic markers of autism through natural language processing techniques. Overall, Deep learning approaches were mainly employed in facial image analysis for autism diagnosis. In contrast, voice and text recognition studies utilized machine learning algorithms. Conclusions: This review demonstrates artificial intelligence's significant role in diagnosing autism spectrum disorder (ASD). These AI-driven approaches can complement traditional diagnostic methods, potentially leading to earlier interventions and improved outcomes for individuals with ASD.","author":[{"family":"Mohammadi","given":"Fatemeh"},{"family":"Shahrokhi","given":"Hassan"},{"family":"Asadzadeh","given":"Afsoon"},{"family":"Pirmoradi","given":"Saeed"},{"family":"Moghtader","given":"Ali"},{"family":"Rezaeihachesu","given":"Peyman"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/hsr2.71476","URL":"https://doi.org/10.1002/hsr2.71476","source":"openalex"},{"id":"oa:W7128543618","type":"article-journal","title":"Current challenges and the way forwards for regulatory databases of artificial intelligence as a medical device","abstract":"Effective regulatory oversight is a key step in ensuring that artificial intelligence as a medical device (AIaMD) is safe in real-world clinical settings. In this Perspective, we provide insights from our experience working with international regulatory databases, informed by our recent research and the expertise of the multidisciplinary authorship team. We highlight four key challenges, discuss attempts to circumvent these limitations, and highlight emerging initiatives. Nevertheless, the underlying issue of the quality and availability of input data from regulatory databases remains. We discuss considerations for improving accessibility and transparency, outline key aspects for a next-generation regulatory data ecosystem for AIaMDs, and call on global stakeholders to come together and align efforts to develop a clear roadmap to accelerate safe innovation and improve outcomes for patients worldwide.","author":[{"family":"Ong","given":"Ariel"},{"family":"Kale","given":"Aditya"},{"family":"Antoun","given":"Joe"},{"family":"Hogg","given":"Henry"},{"family":"Hammond","given":"Ben"},{"family":"Keane","given":"Pearse"},{"family":"Pearson","given":"Russell"},{"family":"Harvey","given":"Hugh"},{"family":"Denniston","given":"Alastair"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1038/s41746-026-02407-w","URL":"https://doi.org/10.1038/s41746-026-02407-w","source":"openalex"},{"id":"oa:W7123588148","type":"article-journal","title":"Explainable Artificial Intelligence for Pulmonary Disease Detection from Chest Radiographs","abstract":"The integration of Explainable Artificial Intelligence (XAI) with deep learning has revolutionized medical image analysis, which is now able to yield accurate and interpretable disease diagnosis. In spite of significant advances in the interpretation of chest X-ray (CXR) for diseases such as tuberculosis, pneumonia, and COVID-19, a number of issues remain to be solved in the fields of explainability, dataset diversity, and clinical trust. This review has comprehensively reviewed 27 research papers (2020–2025) that focus on the use of XAI in pulmonary imaging. It has mapped out the methods, datasets, visualization tools, interpretability results, and performance metrics of the papers. Moreover, it points out the gaps in research concerning the identification of diseases in multi-class, generalizability of models, and validation of the clinical environment, thus generating the next-stage ideas to move towards hybrid CNN–Transformer architectures and clinician-centered explainable AI.","author":[{"family":"Pabba","given":"Prasanna"},{"family":"Chevuturi","given":"Venkata"},{"family":"Sabbineni","given":"Jeevani"},{"family":"Samudrala","given":"Udvisha"},{"family":"Shaik","given":"Mohammad"},{"family":"Kotla","given":"Sreenidhi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1109/ictbig68706.2025.11323948","URL":"https://doi.org/10.1109/ictbig68706.2025.11323948","source":"openalex"},{"id":"oa:W4408719812","type":"article-journal","title":"Effectiveness of Artificial Intelligence–Based Platform in Administering Therapies for Children With Autism Spectrum Disorder: 12-Month Observational Study","abstract":"Abstract Background A 12-month longitudinal observational study was conducted on 43 children aged 2‐18 years to evaluate the effectiveness of the CognitiveBotics artificial intelligence (AI)–based platform in conjunction with continuous therapy in improving therapeutic outcomes for children with autism spectrum disorder (ASD). Objective This study evaluates the CognitiveBotics software’s effectiveness in supporting children with ASD through structured, technology-assisted learning. The primary objectives include assessing user engagement, tracking progress, and measuring efficacy using standardized clinical assessments. Methods A 12-month observational study was conducted on children diagnosed with ASD using the CognitiveBotics AI-based platform. Standardized assessments, include the Childhood Autism Rating Scale (CARS), Vineland Social Maturity Scale, Developmental Screening Test, and Receptive Expressive Emergent Language Test (REEL), were conducted at baseline (T1) and at the endpoint (T2). All participants meeting the inclusion criteria were provided access to the platform and received standard therapy. Participants who consistently adhered to platform use as per the study protocol were classified as the intervention group, while those who did not maintain continuous platform use were designated as the control group. Additionally, caregivers received structured training, including web-based parent teaching sessions, reinforcement strategy training, and home-based activity guidance. Results Participants in the intervention group demonstrated statistically significant improvements across multiple scales. CARS scores reduced from 33.41 (SD 1.89) at T1 to 28.34 (SD 3.80) at T2 ( P <.001). Social age increased from 22.80 (SD 7.33) to 35.76 (SD 9.09; mean change: 12.96, 56.84% increase; P <.001). Social quotient increased from 53.26 (SD 11.84) to 64.75 (SD 16.12; mean change: 11.49, 21.57% increase; P <.001). Developmental age showed an improvement from 30.93 (SD 9.91) to 45.31 (SD 11.20; mean change: 14.38, 46.49% increase; P <.001), while developmental quotient increased from 70.94 (SD 10.95) to 81.33 (SD 16.85; mean change: 10.39, 14.65% increase; P <.001). REEL scores showed substantial improvements, with receptive language increasing by 56.22% ( P <.001) and expressive language by 59.93% ( P <.001). In the control group, while most psychometric parameters showed some improvements, they were not statistically significant. CARS scores decreased by 10.62% ( P =.06), social age increased by 52.27% ( P =.06), social quotient increased by 19.62% ( P =.12), developmental age increased by 44.88% ( P =.06), and developmental quotient increased by 11.23% ( P =.19). REEL receptive and expressive language increased by 34.69% ( P =.10) and 40.48% ( P =.054), respectively. Conclusions Overall, the platform was an effective supplement in enhancing therapeutic outcomes for children with ASD. This platform holds promise as a valuable tool for augmenting ASD therapies across cognitive, social, and developmental domains. Future development should prioritize expanding the product’s accessibility across various languages, ensuring cultural sensitivity and enhancing user-friendliness.","author":[{"family":"Atturu","given":"H"},{"family":"Naraganti","given":"S"},{"family":"Rao","given":"Bharti"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/70589","URL":"https://doi.org/10.2196/70589","source":"openalex"},{"id":"oa:W4406658975","type":"article-journal","title":"Current applications and challenges in large language models for patient care: a systematic review","abstract":"BACKGROUND: The introduction of large language models (LLMs) into clinical practice promises to improve patient education and empowerment, thereby personalizing medical care and broadening access to medical knowledge. Despite the popularity of LLMs, there is a significant gap in systematized information on their use in patient care. Therefore, this systematic review aims to synthesize current applications and limitations of LLMs in patient care. METHODS: We systematically searched 5 databases for qualitative, quantitative, and mixed methods articles on LLMs in patient care published between 2022 and 2023. From 4349 initial records, 89 studies across 29 medical specialties were included. Quality assessment was performed using the Mixed Methods Appraisal Tool 2018. A data-driven convergent synthesis approach was applied for thematic syntheses of LLM applications and limitations using free line-by-line coding in Dedoose. RESULTS: We show that most studies investigate Generative Pre-trained Transformers (GPT)-3.5 (53.2%, n = 66 of 124 different LLMs examined) and GPT-4 (26.6%, n = 33/124) in answering medical questions, followed by patient information generation, including medical text summarization or translation, and clinical documentation. Our analysis delineates two primary domains of LLM limitations: design and output. Design limitations include 6 second-order and 12 third-order codes, such as lack of medical domain optimization, data transparency, and accessibility issues, while output limitations include 9 second-order and 32 third-order codes, for example, non-reproducibility, non-comprehensiveness, incorrectness, unsafety, and bias. CONCLUSIONS: This review systematically maps LLM applications and limitations in patient care, providing a foundational framework and taxonomy for their implementation and evaluation in healthcare settings.","author":[{"family":"Busch","given":"Felix"},{"family":"Hoffmann","given":"Lena"},{"family":"Rueger","given":"Christopher"},{"family":"Dijk","given":"Elon"},{"family":"Kader","given":"Rawen"},{"family":"Ortizprado","given":"Esteban"},{"family":"Makowski","given":"Marcus"},{"family":"Saba","given":"Luca"},{"family":"Hadamitzky","given":"Martin"},{"family":"Kather","given":"Jakob"},{"family":"Truhn","given":"Daniel"},{"family":"Cuocolo","given":"Renato"},{"family":"Adams","given":"Lisa"},{"family":"Bressem","given":"Keno"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s43856-024-00717-2","URL":"https://doi.org/10.1038/s43856-024-00717-2","source":"openalex"},{"id":"oa:W4410824364","type":"article-journal","title":"Artificial Intelligence-Based Models for Automated Bone Age Assessment from Posteroanterior Wrist X-Rays: A Systematic Review","abstract":"Introduction: Bone-age assessment using posteroanterior left hand–wrist radiographs is indispensable in pediatric endocrinology and forensic age determination. Traditional methods—Greulich–Pyle atlas and Tanner–Whitehouse scoring—are time-consuming, operator-dependent, and prone to inter- and intra-observer variability. Aim: To systematically review the performance of AI-based models for automated bone-age estimation from left PA hand–wrist radiographs. Materials and Methods: A systematic review was carried out and previously registered in PROSPERO (CRD42024619808) in MEDLINE (PubMed), Google Scholar, ELSEVIER (Scopus), EBSCOhost, Cochrane Library, Web of Science (WoS), IEEE Xplore, and ProQuest for original studies published between 2019 and 2024. Two independent reviewers extracted study characteristics and outcomes, assessed methodological quality via the Newcastle–Ottawa Scale, and evaluated bias using ROBINS-E. Results: Seventy-seven studies met inclusion criteria, encompassing convolutional neural networks, ensemble and hybrid models, and transfer-learning approaches. Commercial systems (e.g., BoneXpert®, Physis®, VUNO Med®-BoneAge) achieved mean absolute errors of 2–31.8 months—significantly surpassing Greulich–Pyle and Tanner–Whitehouse benchmarks—and reduced reading times by up to 87%. Common limitations included demographic bias, heterogeneous imaging protocols, and scarce external validation. Conclusions: AI-based approaches have substantially advanced automated bone-age estimation, delivering clinical-grade speed and mean absolute errors below 6 months. To ensure equitable, generalizable performance, future work must prioritize demographically diverse training cohorts, implement bias-mitigation strategies, and perform local calibration against region-specific standards.","author":[{"family":"Pérez","given":"Isidro"},{"family":"Bourhim","given":"Sofia"},{"family":"Pérez","given":"Sebastián"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/app15115978","URL":"https://doi.org/10.3390/app15115978","source":"openalex"},{"id":"oa:W7118197163","type":"article-journal","title":"Artificial Intelligence-Enhanced Wearable Blood Pressure Monitoring in Resource-Limited Settings: A Co-Design of Sensors, Model, and Deployment","abstract":"Accurate blood pressure (BP) monitoring is essential for preventing and managing cardiovascular disease. Advancements in materials science, medicine, flexible electronic, and artificial intelligence (AI) have enabled cuffless, unobtrusive BP monitoring systems, offering an alternative to traditional sphygmomanometers. However, extending these advances to real-world cardiovascular care particularly in resource-limited settings remains challenging due to constraints in computational resources, power efficiency, and deployment scalability. This review presents a comprehensive synthesis of AI-enhanced wearable BP monitoring, emphasizing its potential for personalized, scalable, and accessible healthcare. We systematically analyze the end-to-end system architecture, from mechano-electric sensing principles and AI-based estimation models to edge-aware deployment strategies tailored for low-resource environments. We further discuss clinical validation metrics and implementation barriers and prospective strategies. To bridge lab-to-field translation, we propose an innovative \"sensor-model-deployment-assessment\" co-design framework. This roadmap highlights how AI-enhanced BP technologies can support proactive hypertension control and promote cardiovascular health equity on a global scale.","author":[{"family":"Zhang","given":"Yiming"},{"family":"Qiu","given":"Shirong"},{"family":"Du","given":"Kai"},{"family":"Wu","given":"Shun"},{"family":"Xiang","given":"Ting"},{"family":"Zheng","given":"Kenghao"},{"family":"Liu","given":"Zijun"},{"family":"Chen","given":"Hanjie"},{"family":"Ji","given":"Nan"},{"family":"Wang","given":"Fa"},{"family":"Wu","given":"Weijia"},{"family":"Zhang","given":"Yuan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1007/s40820-025-02003-9","URL":"https://doi.org/10.1007/s40820-025-02003-9","source":"openalex"},{"id":"oa:W4409960845","type":"article-journal","title":"Sociotechnical imaginaries and public communication: Analytical framework and empirical illustration using the case of artificial intelligence","abstract":"The concept of sociotechnical imaginaries (SIs) has been widely used and proven fruitful to understand diverging trajectories of technologies. While scholars have acknowledged the multi-layered materialisation of SIs and highlighted the importance of the communicative layer therein, this aspect has remained under-conceptualised. Therefore, we propose an analytical framework to better understand Sociotechnical Imaginaries in Public Communication (SIPCs), defined as publicly constructed visions of (un)desirable sociotechnical futures that guide action, mobilise resources and lay out trajectories for the materialisation or prevention of those futures. In this article, we first discuss relevant strands of research on SIs and public communication. We then lay out the analytical framework of SIPCs that enables the rigorous reconstruction and comparison of sociotechnical imaginaries in public and mediated communication. Finally, we illustrate the framework with examples from public communication about artificial intelligence, an emerging key technology in contemporary societies.","author":[{"family":"Brause","given":"Saba"},{"family":"Schäfer","given":"Mike"},{"family":"Katzenbach","given":"Christian"},{"family":"Mao","given":"Yishu"},{"family":"Richter","given":"Vanessa"},{"family":"Zeng","given":"Jing"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1177/13548565251338192","URL":"https://doi.org/10.1177/13548565251338192","source":"openalex"},{"id":"oa:W4412994393","type":"article-journal","title":"Artificial Intelligence‐Based Pathology to Assist Prediction of Neoadjuvant Therapy Responses for Breast Cancer","abstract":"BACKGROUND: Neoadjuvant therapy (NAT) is a standard breast cancer treatment, but patient response varies significantly. Predictive markers can guide treatment decisions, yet their interpretation suffers from inter-pathologist variability due to breast cancer's complex histology and heterogeneity. Artificial intelligence (AI) applied to image-based omics offers potential to enhance pathological interpretation precision and consistency. METHODS: This review synthesizes existing literature on the application of AI in breast cancer pathology. We specifically focused on identifying and summarizing research that utilizes diverse histopathological features-including morphological characteristics, molecular markers, gene expression profiles, and multidimensional omics data-to predict NAT response in breast cancer patients. RESULTS: AI demonstrates significant capabilities in automatically recognizing histopathological patterns and predicting NAT efficacy. It shows promise as a tool for patient stratification in precision oncology. Research utilizing various pathological feature types (morphological, molecular, genomic, multi-omics) for NAT response prediction is actively evolving. While AI models integrating multi-omics features show potential, challenges remain in robustly predicting NAT outcomes. CONCLUSION: AI-based pathology represents a prospective and powerful decision-support tool for predicting breast cancer NAT response. Despite existing challenges, particularly with complex multi-omics models, AI holds great potential to assist clinical oncologists in optimizing future cancer treatment management.","author":[{"family":"Ji","given":"Juan"},{"family":"Duan","given":"Fanglei"},{"family":"Liao","given":"Qiong"},{"family":"Wang","given":"Hao"},{"family":"Liu","given":"Shiwei"},{"family":"Liu","given":"Yang"},{"family":"Huang","given":"Zongyao"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/cam4.71132","URL":"https://doi.org/10.1002/cam4.71132","source":"openalex"},{"id":"oa:W4411166218","type":"article-journal","title":"Artificial Intelligence in Epilepsy: A Systemic Review","abstract":"Diagnosing and managing epilepsy is difficult for doctors. Surgery can help some patients, but it often takes a long time to get there. This research looks at scientific studies to see if artificial intelligence and machine learning (ML) can be used to improve epilepsy treatment. In-depth research was conducted across PubMed, Google Scholar, Scopus, Wiley, Web of Science, and Microsoft Academia. This search focused on studies exploring the use of ML for diagnosing epilepsy, predicting treatment response, and predicting outcomes of epilepsy surgery. The search was limited to original English-language articles published between 2015 and 2022. This review examined 36 studies on using ML to predict epilepsy. The studies fell into four categories: general diagnosis (27), treatment outcome (3), identifying surgical candidates (2), and predicting surgical results (4). Researchers employed a diverse set of data, including symptoms and brain scans, alongside machine learning algorithms like support vector machines and convolutional neural networks, to construct their models. Some models achieved impressive results with areas under the curve reaching up to 0.99, but most studies were limited by small sample sizes and a lack of independent validation. ML shows potential for epilepsy treatment based on initial studies, but real-world use is restricted due to small sample sizes and the need for more validation from other studies. Large collaborative research efforts and data on long-term outcomes are essential before ML can be widely adopted by doctors and make a positive difference for epilepsy patients.","author":[{"family":"Al-Breiki","given":"Almuntasar"},{"family":"Al-Sinani","given":"Said"},{"family":"Elsharaawy","given":"Ahmed"},{"family":"Usama","given":"Mohamed"},{"family":"Alsaadi","given":"Tariq"}],"issued":{"date-parts":[[2025]]},"DOI":"10.14581/jer.25002","URL":"https://doi.org/10.14581/jer.25002","source":"openalex"},{"id":"oa:W4410420135","type":"article-journal","title":"The DRAGON benchmark for clinical NLP","abstract":"Artificial Intelligence can mitigate the global shortage of medical diagnostic personnel but requires large-scale annotated datasets to train clinical algorithms. Natural Language Processing (NLP), including Large Language Models (LLMs), shows great potential for annotating clinical data to facilitate algorithm development but remains underexplored due to a lack of public benchmarks. This study introduces the DRAGON challenge, a benchmark for clinical NLP with 28 tasks and 28,824 annotated medical reports from five Dutch care centers. It facilitates automated, large-scale, cost-effective data annotation. Foundational LLMs were pretrained using four million clinical reports from a sixth Dutch care center. Evaluations showed the superiority of domain-specific pretraining (DRAGON 2025 test score of 0.770) and mixed-domain pretraining (0.756), compared to general-domain pretraining (0.734, p < 0.005). While strong performance was achieved on 18/28 tasks, performance was subpar on 10/28 tasks, uncovering where innovations are needed. Benchmark, code, and foundational LLMs are publicly available.","author":[{"family":"Bosma","given":"Joeran"},{"family":"Dercksen","given":"Koen"},{"family":"Builtjes","given":"Luc"},{"family":"Andre","given":"RJ"},{"family":"Roest","given":"Christian"},{"family":"Fransen","given":"Stefan"},{"family":"Noordman","given":"Constant"},{"family":"Navarro-Padilla","given":"Mar"},{"family":"Lefkes","given":"Judith"},{"family":"Alves","given":"Natália"},{"family":"Grauw","given":"Max"},{"family":"Eekelen","given":"Leander"},{"family":"Spronck","given":"Joey"},{"family":"Schuurmans","given":"Megan"},{"family":"Wilde","given":"Bram"},{"family":"Hendrix","given":"Ward"},{"family":"Aswolinskiy","given":"Witali"},{"family":"Saha","given":"Anindo"},{"family":"Twilt","given":"Jasper"},{"family":"Geijs","given":"Daan"},{"family":"Veltman","given":"Jeroen"},{"family":"Yakar","given":"Derya"},{"family":"Rooij","given":"Maarten"},{"family":"Ciompi","given":"Francesco"},{"family":"Hering","given":"Alessa"},{"family":"Geerdink","given":"Jeroen"},{"family":"Huisman","given":"Henkjan"},{"family":"Grauw","given":"Max"},{"family":"Eekelen","given":"Leander"},{"family":"Wilde","given":"Bram"},{"family":"Lohuizen","given":"Quintin"},{"family":"Stegeman","given":"Michelle"},{"family":"Rutten","given":"Karlijn"},{"family":"Smit","given":"Inge"},{"family":"Stultiens","given":"Gijs"},{"family":"Overduin","given":"Christiaan"},{"family":"Rutten","given":"Matthieu"},{"family":"Scholten","given":"Ernst"},{"family":"Post","given":"Rachel"},{"family":"Grünberg","given":"Katrien"},{"family":"Vos","given":"Shoko"},{"family":"Taken","given":"Elise"},{"family":"Nagtegaal","given":"Iris"},{"family":"Mickan","given":"Anne"},{"family":"Groeneveld","given":"Miriam"},{"family":"Gerke","given":"Paul"},{"family":"Meakin","given":"James"},{"family":"Looijen-Salamon","given":"MG"},{"family":"Haas","given":"Tijmen"},{"family":"Hoitsma","given":"Fabian"},{"family":"Damato","given":"Marina"},{"family":"Rooij","given":"Maarten"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01626-x","URL":"https://doi.org/10.1038/s41746-025-01626-x","source":"openalex"},{"id":"oa:W7133535600","type":"article-journal","title":"Performance Validation of ORTHOSEG, a Novel Artificial Intelligence Tool for the Segmentation of Orthopantomographs and Intra-Oral X-Rays","abstract":"Background: Dental radiographs are essential for diagnosis and treatment planning in modern dentistry. However, their manual interpretation is time-consuming and subject to variability, highlighting the need for automated tools to improve efficiency and consistency. This study aims to validate ORTHOSEG, a deep learning-based system designed to automate the segmentation of anatomical, pathological, and non-pathological elements in radiographs, including orthopantomograms, bitewings, and periapical images. Methods: ORTHOSEG’s performance was evaluated using a rigorously curated dataset of 150 dental radiographs, including 50 orthopantomograms, 50 bitewings, and 50 periapical images, with manual annotations by expert clinicians serving as the ground truth. The system’s segmentation performance was assessed using standard evaluation metrics, including mean Dice Similarity Coefficient (mDSC) and mean Intersection over Union (mIoU), and inference time was also recorded. Results: The system achieved high accuracy, with mDSC and mIoU values of 0.635 ± 0.233 and 0.576 ± 0.214, respectively. In particular for orthopantomograms, it achieved an mDSC of 0.756 ± 0.174 and an mIoU of 0.684 ± 0.172, surpassing existing benchmarks. Its segmentation capabilities extend to approximately 70 distinct elements, underscoring its comprehensive utility. The system demonstrated efficient computational performance, with processing times of 19.745 ± 3.625 s for orthopantomograms, 8.467 ± 0.903 s for bitewings, and 5.653 ± 0.897 s for periapical radiographs on standard clinical hardware. Conclusions: ORTHOSEG demonstrates efficiency suitable for integration into routine workflows. This study confirms ORTHOSEG’s reliability and potential to improve diagnostic workflows, offering clinicians a valuable tool for faster and more detailed radiograph analysis. Future research will focus on extending validation across diverse clinical scenarios to ensure broader applicability. However, this study has limitations, including the use of a dataset derived from a European population and the absence of usability and clinical workflow evaluation, which should be addressed in future studies.","author":[{"family":"Cota","given":"Giuseppe"},{"family":"Scaramozzino","given":"Gaetano"},{"family":"Chiesa","given":"Marco"},{"family":"Gennaro","given":"Lelio"},{"family":"Pascadopoli","given":"Maurizio"},{"family":"Scribante","given":"Andrea"},{"family":"Colombo","given":"Marco"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/clinpract16030054","URL":"https://doi.org/10.3390/clinpract16030054","source":"openalex"},{"id":"oa:W4413922470","type":"article-journal","title":"Reinforcement Learning in Medical Imaging: Taxonomy, LLMs, and Clinical Challenges","abstract":"Reinforcement learning (RL) is being used more in medical imaging for segmentation, detection, registration, and classification. This survey provides a comprehensive overview of RL techniques applied in this domain, categorizing the literature based on clinical task, imaging modality, learning paradigm, and algorithmic design. We introduce a unified taxonomy that supports reproducibility, highlights design guidance, and identifies underexplored intersections. Furthermore, we examine the integration of Large Language Models (LLMs) for automation and interpretability, and discuss privacy-preserving extensions using Differential Privacy (DP) and Federated Learning (FL). Finally, we address deployment challenges and outline future research directions toward trustworthy and scalable medical RL systems.","author":[{"family":"Riad","given":"Abm"},{"family":"Barek","given":"Md"},{"family":"Shahriar","given":"Hossain"},{"family":"Francia","given":"Guillermo"},{"family":"Ahamed","given":"Sheikh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/fi17090396","URL":"https://doi.org/10.3390/fi17090396","source":"openalex"},{"id":"oa:W4406880779","type":"manuscript","title":"Advancing Generative Artificial Intelligence and Large Language Models for Demand Side Management with Internet of Electric Vehicles","abstract":"The energy optimization and demand side management (DSM) of Internet of Things (IoT)-enabled microgrids are being transformed by generative artificial intelligence, such as large language models (LLMs). This paper explores the integration of LLMs into energy management, and emphasizes their roles in automating the optimization of DSM strategies with Internet of Electric Vehicles (IoEV) as a representative example of the Internet of Vehicles (IoV). We investigate challenges and solutions associated with DSM and explore the new opportunities presented by leveraging LLMs. Then, we propose an innovative solution that enhances LLMs with retrieval-augmented generation for automatic problem formulation, code generation, and customizing optimization. The results demonstrate the effectiveness of our proposed solution in charging scheduling and optimization for electric vehicles, and highlight our solution's significant advancements in energy efficiency and user adaptability. This work shows LLMs' potential in energy optimization of the IoT-enabled microgrids and promotes intelligent DSM solutions.","author":[{"family":"Zhang","given":"Hanwen"},{"family":"Zhang","given":"Ruichen"},{"family":"Zhang","given":"Weidong"},{"family":"Niyato","given":"Dusit"},{"family":"Wen","given":"Yonggang"},{"family":"Miao","given":"Chunyan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.48550/arxiv.2501.15544","URL":"https://doi.org/10.48550/arxiv.2501.15544","source":"openalex"},{"id":"oa:W7161305024","type":"article-journal","title":"Academic transformation in the era of artificial intelligence: drivers of university faculty adoption of GenAI based on the UTAUT model","abstract":"Generative Artificial Intelligence (GenAI) technology is reshaping the traditional paradigms of research and teaching, with “AI for Science” and “AI-Enabled Education” profoundly transforming contemporary higher education. As the central force driving research innovation and education, university faculty members’ willingness to adopt GenAI technology is critical to educational innovation. This study employs the Unified Theory of Acceptance and Use of Technology (UTAUT), incorporating GenAI literacy and usage attitudes, and utilizes structural equation modeling to explore the key factors influencing faculty members’ intention to use GenAI technology in research and teaching contexts. A total of 238 valid responses were collected from university teachers. The findings reveal that: first, performance expectancy and usage attitudes within the UTAUT model significantly and positively impact faculty members’ usage intention, highlighting the importance of perceived benefits and personal attitudes in the technology acceptance process; second, faculty members’ overall level of GenAI literacy significantly and positively affects their attitudes toward GenAI technology, indicating that enhancing their GenAI literacy- as reflected in technical proficiency, critical evaluation, communication proficiency, creative application- is crucial for forming positive attitudes; and third, usage attitudes serve as a mediating factor, fully mediating the effect of GenAI literacy on usage intention, emphasizing the central role of positive attitudes in technology adoption. Based on these findings, this study provides theoretical and practical guidance for promoting GenAI technology in universities, suggesting that institutions should prioritize cultivating faculty members’ GenAI literacy and shaping positive attitudes toward technology to enhance their willingness to adopt it. Furthermore, the study offers new perspectives and frameworks for future research on the application of GenAI technology in education.","author":[{"family":"Wang","given":"Jingyao"},{"family":"Wang","given":"Haoming"},{"family":"Yang","given":"Junwu"},{"family":"Wang","given":"Chengliang"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1057/s41599-026-07606-0","URL":"https://doi.org/10.1057/s41599-026-07606-0","source":"openalex"},{"id":"oa:W7160419605","type":"article-journal","title":"Exploring Students’ Perceptions and Usage of Artificial Intelligence in Supporting Mental Health: A Preliminary Study in Higher Education in Qatar","abstract":"Background: Artificial intelligence (AI) is widely used in mental health care for screening, monitoring, and intervention. Notably, most studies of AI in mental health have been performed in Western contexts, with limited evidence from the Arab Gulf region, where cultural factors such as stigma, privacy, and help-seeking norms may influence acceptance. Objective: Investigating university students’ perceptions of AI in mental health support, including awareness, trust, readiness, and preferences in a Gulf context. Methods: A cross-sectional survey was administered to 220 university students in Qatar. Data were analyzed using descriptive statistics, Chi-square tests, and one-way ANOVA to explore associations and group differences. Results: Students showed low-to-moderate levels of awareness and trust in AI-based mental health tools. The majority of participants showed that they were prepared to employ AI for stress management, but they do not prefer to replace face-to-face therapy, suggesting a preference for complementary use. A significant association was found between readiness and expectations (p < 0.00001), which means ambivalence toward AI effectiveness. No significant differences were observed across gender or academic level (p > 0.05). Key concerns included loss of human interaction, overreliance on technology, and diagnostic accuracy, while perceived benefits included cost reduction and 24/7 accessibility. Conclusions: Students exhibit cautious adoption of AI in mental health services. Acceptance is influenced by trust, privacy issues, and apparent compassion. AI is optimally situated as a supplementary instrument within ethically regulated, culturally attuned hybrid care frameworks that maintain the fundamental importance of human connection.","author":[{"family":"Elbarazi","given":"Amani"},{"family":"Mohamed","given":"Hatem"},{"family":"Nasser","given":"Ramzi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/healthcare14091247","URL":"https://doi.org/10.3390/healthcare14091247","source":"openalex"},{"id":"oa:W7139022403","type":"article-journal","title":"Electrocardiographic Signatures of Dysglycaemia: Mechanistic Foundations, Digital Biomarkers, and Artificial Intelligence for Non-Invasive Diabetes Risk Stratification","abstract":"Diabetes mellitus is projected to affect more than 1.3 billion people worldwide by 2050, with millions remaining undiagnosed or in a prediabetic state. Cardiovascular complications account for nearly half of diabetes-related deaths, highlighting the need for scalable tools capable of identifying metabolic dysregulation before irreversible cardiac damage develops. This review synthesizes current mechanistic, clinical, and computational evidence linking diabetes to cardiac electrophysiological remodeling and examines electrocardiography (ECG) as a non-invasive modality for early detection of dysglycaemia. Chronic hyperglycaemia, insulin resistance, oxidative stress, microvascular dysfunction, and cardiac autonomic neuropathy collectively contribute to measurable ECG alterations, including QT/QTc prolongation, increased QT dispersion, changes in Tp–e indices, and reduced heart rate variability. These changes often precede overt cardiovascular disease and correlate with glycaemic burden and diabetes duration. Recent advances in signal processing and artificial intelligence have expanded the diagnostic potential of ECG. Both classical machine learning approaches and large-scale deep learning models demonstrate that ECG contains latent signatures associated with incident type 2 diabetes and glycaemic status. Despite promising results, heterogeneity in study design, limited representation of prediabetes, and lack of standardized validation frameworks remain major barriers to clinical translation. Prospective, multi-ethnic studies are needed to establish ECG-based screening as a reliable component of early diabetes detection strategies.","author":[{"family":"Alimbayev","given":"Chingiz"},{"family":"Alimbayeva","given":"Zhadyra"},{"family":"Ozhikenov","given":"Kassymbek"},{"family":"Karibayev","given":"Kairat"},{"family":"Orynbay","given":"Zhansila"},{"family":"Igembay","given":"Yerbolat"},{"family":"Daniyalov","given":"Madiyar"},{"family":"Nurdanali","given":"Akzhol"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/app16062902","URL":"https://doi.org/10.3390/app16062902","source":"openalex"},{"id":"oa:W7161630578","type":"article-journal","title":"Videomics and artificial intelligence in endoscopic diagnosis of laryngeal lesions: mapping current evidence through a scoping review","abstract":"Laryngeal lesions are common and despite advances like high-definition videolaryngoscopy and enhanced imaging modalities such as narrow-band imaging, laryngoscopy remains operator-dependent. In this setting, artificial intelligence (AI) represents a promising tool to support clinical evaluation. This scoping review evaluated the current applications of AI in the endoscopic diagnosis of laryngeal lesions. A comprehensive search of MEDLINE and Scopus databases included 35 studies addressing AI-based detection, classification, or segmentation of laryngeal pathologies. Detection models frequently achieved real-time inference speeds and strong performance metrics although external validation was limited. Classification studies showed particularly robust results for binary tasks distinguishing high-risk from low-risk lesions, with some models achieving sensitivity and accuracy exceeding 90%. Segmentation models demonstrated the potential for precise delineation of cancer margins, a capability of notable relevance for surgical planning and intraoperative decision-making. Despite promising advances, heterogeneity in study design, limited external validation, and reliance on single-centre datasets currently restrict broad clinical implementation. Nonetheless, the emerging integration of AI into laryngeal endoscopy represents a significant step toward reproducible and accessible diagnostic assessment.","author":[{"family":"Ioppi","given":"Alessandro"},{"family":"Bellini","given":"Elisa"},{"family":"Salvetta","given":"Maria"},{"family":"Marchi","given":"Filippo"},{"family":"Maria","given":"Domenico"},{"family":"Peretti","given":"Giorgio"},{"family":"Dalessio","given":"Pasquale"},{"family":"Perotti","given":"Pietro"},{"family":"Piccin","given":"Ottavio"},{"family":"Sampieri","given":"Claudio"}],"issued":{"date-parts":[[2026]]},"DOI":"10.14639/0392-100x-suppl.1-46-2026-a1967","URL":"https://doi.org/10.14639/0392-100x-suppl.1-46-2026-a1967","source":"openalex"},{"id":"oa:W7118927645","type":"article-journal","title":"Natural Language Processing (NLP)-Based Frameworks for Cyber Threat Intelligence and Early Prediction of Cyberattacks in Industry 4.0: A Systematic Literature Review","abstract":"This study provides a systematic overview of Natural Language Processing (NLP)-based frameworks for Cyber Threat Intelligence (CTI) and the early prediction of cyberattacks in Industry 4.0. As digital transformation accelerates through the integration of IoT, SCADA, and cyber-physical systems, manufacturing environments face an expanding and complex cyber threat landscape. Following the PRISMA 2020 systematic review protocol, 80 peer-reviewed studies published between 2015 and 2025 were analyzed across IEEE Xplore, Scopus, and Web of Science to identify methods that employ NLP for CTI extraction, reasoning, and predictive modelling. The review finds that transformer-based architectures, knowledge graph reasoning, and social media mining are increasingly used to convert unstructured data into actionable intelligence, thereby enabling earlier detection and forecasting of cyber threats. Large Language Models (LLMs) demonstrate strong potential for anticipating attack sequences, while domain-specific models enhance industrial relevance. Persistent challenges include data scarcity, domain adaptation, explainability, and real-time scalability in operational-technology environments. The review concludes that NLP is reshaping Industry 4.0 cybersecurity from reactive defense toward predictive, adaptive, and intelligence-driven protection, and it highlights the need for interpretable, domain-specific, and resource-efficient frameworks to secure Industry 4.0 ecosystems.","author":[{"family":"Albarrak","given":"Majed"},{"family":"Salonitis","given":"Konstantinos"},{"family":"Jagtap","given":"Sandeep"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/app16020619","URL":"https://doi.org/10.3390/app16020619","source":"openalex"},{"id":"oa:W7140187407","type":"article-journal","title":"AI-supported case-based learning in medical education: a comprehensive scoping review","abstract":"Introduction: Recent literature indicates that generative artificial intelligence (GenAI) is also being integrated into case-based learning (CBL) through activities such as clinical case generation, clinical reasoning support, and structured feedback. However, the evidence about GenAI's role in CBL remains fragmented. Given the diverse nature of available GenAI-CBL studies, we conducted a comprehensive scoping review to map and synthesize the evidence in this area, highlighting key themes, outcomes, challenges, and limitations to inform future research, curriculum development, and policies. Method: A comprehensive search was performed across multidisciplinary databases, including PubMed, ERIC, Scopus, Web of Science, and Google Scholar, covering publications from 2019 to 2025. Title and abstract screening, followed by full-text review and data extraction, were conducted independently by two reviewers using predefined eligibility criteria. The data synthesis involved thematic analysis to create an evidence map of GenAI-supported case-based and case-anchored learning in medical education. Results: The findings were organized into six key themes that showcase the role of GenAI in enhancing case-based learning, covering areas such as clinical reasoning and contextual thinking; efficient and scalable case creation; learner engagement, motivation, and perceived usefulness; accuracy, reliability, and ethical issues; faculty adaptation and pedagogical integration; and hybrid and reflective learning methods. Conclusion: Overall, the evidence indicates that GenAI can effectively support CBL in medical education, especially during early and intermediate stages. It also highlights the ongoing importance of faculty oversight and the need for further research to address advanced clinical judgment and ethical reasoning. Systematic review registration: https://doi.org/10.17605/OSF.IO/28E3G, identifier (28E3G).","author":[{"family":"Abidi","given":"Dr"},{"family":"Almazan","given":"Joseph"},{"family":"Fabiyi","given":"Olaoluwa"},{"family":"Zehra","given":"Fatin"},{"family":"Tariq","given":"Muhammad"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/fmed.2026.1798097","URL":"https://doi.org/10.3389/fmed.2026.1798097","source":"openalex"},{"id":"oa:W7138916174","type":"article-journal","title":"Systematic Review of Artificial Intelligence in Positive and Existential Psychiatry: Advancing Mental and Emotional Health Through Metacompetency Development","abstract":"Background: Positive and existential psychiatry are approaches to mental health that emphasize the promotion of well-being, resilience, and optimal functioning alongside the conventional management of mental illness. Research suggests that the development of self-regulatory metacompetencies is associated with positive mental health and well-being outcomes. Artificial intelligence (AI) technologies are increasingly being used as assistive tools in psychiatry. However, the integration of AI in therapeutic interventions remains underexplored. Objectives: Thus, this systematic review aimed to synthesize evidence from randomized controlled trials evaluating whether AI-based positive and existential psychiatry interventions contribute to improvements in mental and emotional health. A second objective was to examine whether the therapeutic components and psychological processes implemented in these interventions conceptually relate to self-regulatory metacompetencies that underpin sustainable mental health and human flourishing. Methods: The review was conducted according to PRISMA 2020 guidelines. Only experimental studies including randomized controlled trials (RCTs) published from 2015 to 2025 were included. Twenty-four studies met the inclusion criteria. Results: Across interventions using conversational AI chatbots, generative AI and AI-augmented reflective systems, embodied conversational agents, social and humanoid AI robots, consistent improvements were observed in depression, anxiety, negative affect, and loneliness. The interventions enhanced various metacompetencies such as emotional regulation, emotional awareness, self-reflection, and cognitive reappraisal. Conclusions: The findings suggest that AI-based positive and existential psychiatry interventions can support mental and emotional health, especially when fostering key metacompetencies. Although promising, further high-quality trials are needed to clarify long-term effects. The findings of this study can contribute to the discussion about the ways AI-supported interventions may promote sustainable mental health.","author":[{"family":"Mitsea","given":"Eleni"},{"family":"Drigas","given":"Athanasios"},{"family":"Skianis","given":"Charalabos"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/healthcare14060783","URL":"https://doi.org/10.3390/healthcare14060783","source":"openalex"},{"id":"oa:W7148281198","type":"article-journal","title":"The Evolution of Intelligent Digital Profiling: A Multi-Sectoral Synthesis of Explainable Artificial Intelligence and Federated Learning Frameworks","abstract":"Digital profiling has evolved into a cornerstone of computational intelligence, enabling the synthesis of complex user representations through the systematic analysis of multi-dimensional digital interactions. This study provides a comprehensive investigation into the conceptual and architectural frameworks of digital profiling, delineating its evolution across various strategic sectors. Utilizing a thematic synthesis methodology, we systematically analyzed 197 high-impact articles published between 2011 and 2025. Our findings categorize digital profiling into six fundamental computational stages: objective definition, multi-source data acquisition, feature selection, similarity modeling, representation synthesis, and iterative monitoring. While the analysis underscores the increasing deployment of profiling in finance, security, and healthcare, it reveals critical systemic risks, including algorithmic bias and privacy vulnerabilities. We propose a strategic transition toward \"Privacy-by-Design\" architectures, highlighting the integration of Federated Learning and Explainable Artificial Intelligence (XAI) as essential mechanisms for aligning profiling systems with global regulatory standards (e.g., GDPR). This research contributes a robust theoretical roadmap for developing transparent, accountable, and ethically-aligned intelligent systems, bridging the gap between technical efficiency and user-centric rights.","author":[{"family":"Darıcı","given":"Sefer"},{"family":"Şahin","given":"Zekeriya"},{"family":"Darıcı","given":"Ayla"}],"issued":{"date-parts":[[2026]]},"DOI":"10.59543/68svs968","URL":"https://doi.org/10.59543/68svs968","source":"openalex"},{"id":"oa:W7126422792","type":"article-journal","title":"Flight rules for clinical AI: lessons from aviation for human-AI collaboration in medicine","abstract":"The parallels between medicine and aviation are well-recognised. The aviation industry's early experience with automation improved safety and efficiency, but simultaneously introduced new vulnerabilities and occasionally created misplaced trust in complex systems. Aviation has developed a robust safety framework in response to these costly lessons. In this Perspective, which draws from the experiences of clinicians and aviation experts, we argue that it is now time for the medical community to consider how we can learn from these lessons as artificial intelligence (AI) becomes increasingly integrated into clinical care. We propose that this requires a shift in perspective from AI as \"autopilot\" to collaboration with a \"digital copilot\", as well as considerations of practicalities such as scenario-based training, clinician benchmarking, and minimum unaided practice, with the ultimate aim of optimising human-AI collaboration to improve patient care.","author":[{"family":"Ong","given":"Ariel"},{"family":"Merle","given":"David"},{"family":"Pollreisz","given":"Andreas"},{"family":"Wagner","given":"Siegfried"},{"family":"Sevgi","given":"Mertcan"},{"family":"Keane","given":"Pearse"},{"family":"Huemer","given":"Roman"},{"family":"Oehling","given":"Julian"},{"family":"Jäger","given":"Markus"},{"family":"Huemer","given":"Josef"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1038/s41746-026-02410-1","URL":"https://doi.org/10.1038/s41746-026-02410-1","source":"openalex"},{"id":"oa:W7133938113","type":"article-journal","title":"Artificial intelligence-assisted project-based learning: Examining the interaction with learning creativity on the students’ digital content outcomes","abstract":"Despite the rapid integration of artificial intelligence (AI) in education, limited empirical evidence compares the effectiveness of AI-assisted project-based learning (PjBL) and problem-based learning (PBL) while considering students’ learning creativity in digital content production. This study aims to examine the effects of AI-assisted PjBL and AI-assisted PBL, as well as their interaction with students’ learning creativity, on learning outcomes in digital content production. The study employs a 2×2 quasi-experimental factorial design, in which factor (A) represents the learning model (AI-assisted PjBL and AI-assisted PBL) and factor (B) represents learning creativity (high and low). The population consists of eighth-grade students at the junior high school level. The findings address four hypotheses: (1) there is a significant difference in learning outcomes between students receiving AI-assisted PjBL and those receiving AI-assisted PBL (Sig. 0.043 < 0.05); (2) there is a significant interaction effect between AI-assisted learning models and learning creativity (Sig. 0.000 < 0.05); (3) students with high creativity who receive AI-assisted PjBL achieve higher learning outcomes than those receiving AI-assisted PBL (Sig. 0.000 < 0.05); and (4) students with low creativity who receive AI-assisted PjBL demonstrate lower learning outcomes than those receiving AI-assisted PBL (Sig. 0.000 < 0.05). These findings indicate that AI-assisted PjBL and learning creativity enhance students’ learning outcomes in producing creative digital content, particularly for students with high learning creativity. This study contributes empirical evidence to the integration of AI in junior high school computer science education.","author":[{"family":"Hartono","given":"Rudi"},{"family":"Rifani","given":"Salsabila"},{"family":"Waspodo","given":"Muktiono"}],"issued":{"date-parts":[[2026]]},"DOI":"10.22515/jemin.v6i1.12775","URL":"https://doi.org/10.22515/jemin.v6i1.12775","source":"openalex"},{"id":"oa:W7140287256","type":"article-journal","title":"Factors Associated with Artificial Intelligence-Help-Seeking Behavior Among University Students in the UAE: A Cross-Sectional Study","abstract":"Artificial intelligence (AI)-mediated tools have rapidly penetrated student life and become a valuable resource for seeking help with academic assignments/tasks, psychological problems, and social interactions. This study aims to investigate the levels and associations of AI-help-seeking behavior (AI-HSB), anxiety, stress, and depression among university students in the United Arab Emirates (UAE). In addition, it examines the factors associated with AI-HSB based on the selected demographic (gender, marital status, age, academic year, employment status, major, and nationality), as well as anxiety, stress, and depression. This study employed a descriptive cross-sectional design among 433 university students, who were recruited via an online Google Form between 1 October 2025 and 10 December 2025. The study utilized validated Arabic versions of the AI-HSB scale and the anxiety, stress, and depression scale. Descriptive statistics, Pearson correlation, and predictive analyses were conducted using SPSS v 25. Results indicated that students reported moderate reliance on AI-HSB despite moderate to severe levels of psychological distress, with particular emphasis on anxiety. The AI-HSB was positively associated with anxiety, stress, and depression amongst the participants. Furthermore, both depression and the students’ academic year emerged as the only significant predictors of AI-HSB, explaining a modest but meaningful proportion of variance with an exact percentage of 18.1%. AI tools may partially circumvent stigma by offering privacy and anonymity; however, cultural expectations around interpersonal support, trust, and authority may simultaneously limit students’ willingness to rely on non-human agents for emotional care.","author":[{"family":"Alfuqaha","given":"Othman"},{"family":"Msall","given":"Kyle"},{"family":"Abdelrahman","given":"Rasha"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/educsci16040506","URL":"https://doi.org/10.3390/educsci16040506","source":"openalex"},{"id":"oa:W4407046752","type":"article-journal","title":"Configuration Testing of an Artificial Pancreas System Using a Digital Twin: An Evaluative Case Study","abstract":"ABSTRACT The recent growth in popularity of wearable medical devices has improved the quality of life of people with medical conditions. Testing such devices may require users to configure these systems using physical trials, putting themselves in potentially dangerous scenarios. Misconfiguration of such devices has caused disease misdiagnoses and incorrect drug prescriptions. Digital twins have been proposed as an opportunity to reduce such risks of testing system configurations in simulated environments, decoupling the user from the system under test. In this paper, we perform an evaluative case study to assess the use of a digital twin for configuration testing of an artificial pancreas system (APS) control algorithm. These systems regulate the blood glucose levels in people with type 1 diabetes mellitus, and so misconfigurations can cause severe hypoglycaemia or hyperglycaemia, which can be life‐threatening. We tested the OpenAPS control algorithm against 156 people's clinical data. We found that our digital twin provided an accurate simulation environment to perform configuration testing and accurately predict blood glucose–insulin behaviour. We evaluated different APS configurations, identifying a potentially unsafe configuration without the risks associated with a physical trial. We identified the challenges associated with modelling clinical data, which could lead to misinterpretations in configuration testing and the reduction of test reliability when modelling stochastic body dynamics.","author":[{"family":"Somers","given":"Richard"},{"family":"Walkinshaw","given":"Neil"},{"family":"Hierons","given":"Robert"},{"family":"Elliott","given":"Daisy"},{"family":"Iqbal","given":"Ahmed"},{"family":"Walkinshaw","given":"Emma"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/stvr.70000","URL":"https://doi.org/10.1002/stvr.70000","source":"openalex"},{"id":"oa:W7129067071","type":"article-journal","title":"Toward Timely Diagnosis of Pancreatic Cancer: Revolutionizing Early Detection Through Genomics, Artificial Intelligence, and Noninvasive Biomarkers","abstract":"BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) remains one of the most aggressive cancers, typically diagnosed at an advanced stage due to its subtle and often absent early symptoms. Despite representing only 3% of new cancer cases, it is projected to become the second leading cause of cancer-related deaths by 2030. Currently, early diagnosis remains a significant challenge, and survival rates remain poor due to the lack of effective screening tools. METHODS: We conducted a comprehensive literature review to explore the most recent advances in PDAC detection, focusing on novel biomarkers, liquid biopsies, artificial intelligence (AI)-enhanced imaging, and non-invasive surveillance strategies. We examined the role of circulating tumor DNA (ctDNA), microRNAs, and volatile organic compounds (VOCs) as diagnostic tools, alongside the integration of advanced imaging modalities like MRI, EUS, and MRCP in high-risk individuals, including those with hereditary cancer syndromes. RESULTS: Emerging technologies, such as AI-driven imaging and liquid biopsy, have shown promising improvements in detecting PDAC at earlier, potentially resectable stages. Surveillance strategies for high-risk populations, including BRCA1/2 mutation carriers and individuals with Lynch syndrome, have demonstrated increased detection of Stage I PDAC, offering a significant opportunity for curative intervention. AI and machine learning techniques are also enhancing the sensitivity and specificity of imaging, providing a new frontier in early-stage diagnosis. CONCLUSION: The integration of molecular diagnostics, advanced imaging technologies, and AI may enable a paradigm shift in PDAC detection, transitioning from late to early-stage diagnosis and potentially improving survival rates. However, further clinical validation and standardization of these technologies are essential to ensure their widespread clinical adoption. The future of PDAC detection lies in a multimodal, personalized approach, optimizing diagnostic accuracy and early intervention for high-risk individuals.","author":[{"family":"Ma","given":"Hussain"},{"family":"Qammar","given":"Sana"},{"family":"Wang","given":"Ju‐mei"},{"family":"Zhai","given":"Aoqiang"},{"family":"Li","given":"Fu‐yu"},{"family":"Hu","given":"Hai‐jie"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1111/jgh.70281","URL":"https://doi.org/10.1111/jgh.70281","source":"openalex"},{"id":"oa:W4415133359","type":"article-journal","title":"Robot Path Planning: from Analytical to Computer Intelligence Approaches","abstract":"Abstract In an era where robots are becoming an integral part of human quotidian activities, understanding how they function is crucial. Among the inherent building complexities, from electronics to mechanics, path planning emerges as a universal aspect of robotics. The primary contribution of this work is to provide an overview of the current state of robot path planning topics and a comparison between those same algorithms and its inherent characteristics. The path planning concept relies on the process by which an algorithm determines a collision-free path between a start and an end point, optimizing parameters such as energy consumption and distance. The quest for the most effective path planning method has been a long-standing discussion, as the choice of method is highly dependent on the specific application. This review consolidates and elucidates the categories of path planning methods, specifically classical or analytical methods, and computer intelligence methods. In addition, the operational principles of these categories will be explored, discussing their respective advantages and disadvantages, and reinforcing these discussions with relevant studies in the field. This work will focus on the most prevalent and recognized methods within the robotics path planning problem, being mobile robotics or manipulator arms, including Cell Decomposition, A*, Probabilistic Roadmaps, Rapidly-exploring Random Trees, Genetic Algorithms, Particle Swarm Optimization, Ant Colony Optimization, Artificial Potential Fields, Fuzzy, and Neural Networks. Following the detailed explanation of these methods, a comparative analysis of their advantages and drawbacks is organized in a comprehensive table. This comparison will be based on various quality metrics, such as the type of trajectory provided (global or local), the scenario implementation type (real or simulated scenarios), testing environments (static or dynamic), hybrid implementation possibilities, real-time implementation, completeness of the method, consideration of the robot’s kinodynamic constraints, use of smoothing techniques, and whether the implementation is online or offline.","author":[{"family":"Dias","given":"Pedro"},{"family":"Souza","given":"João"},{"family":"Pires","given":"EJS"},{"family":"Filipe","given":"Vítor"},{"family":"Figueiredo","given":"Daniel"},{"family":"Rocha","given":"Luís"},{"family":"Silva","given":"Manuel"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s10846-025-02322-4","URL":"https://doi.org/10.1007/s10846-025-02322-4","source":"openalex"},{"id":"oa:W7158242381","type":"article-journal","title":"External validation of ECG artificial intelligence for emergency and cardiac assessment across a large-scale U.S. healthcare system","abstract":"An ECG-based artificial intelligence (AI) model was previously developed to generate ten digital biomarkers for emergency and cardiac assessment and is currently deployed in clinical practice in Korea (ECG Buddy, ARPI Inc.). Its external validity within U.S. healthcare settings has not been established. This study evaluated model performance using a large-scale, multi-center U.S. dataset via the Mayo Clinic Platform (MCP) Discover. Two validation cohorts were assessed: an emergency-diagnosis cohort (mortality, AMI, STEMI and equivalents, hyperkalemia, pulmonary edema) using initial ED ECGs, and a cardiac-function cohort (left and right ventricular systolic dysfunction, pulmonary hypertension, hemodynamically significant pericardial effusion) anchored to echocardiography or right-heart catheterization. The primary objective was non-inferiority of the Area Under the Receiver Operating Characteristic Curve (AUC) relative to prespecified benchmarks. Across ten target conditions, AUCs ranged from 0.883 to 0.949; all met non-inferiority criteria. Performance was consistent across sex and age strata. In a paired subset (N = 1368), ECG-AI outperformed initial troponin T for AMI (AUC 0.920 vs. 0.878) and STEMI equivalents (AUC 0.932 vs. 0.736; all P < 0.001). These findings support the model's potential for ECG-based screening and triage, and provide a foundation for prospective evaluation of calibration, clinical integration, and impact across diverse populations.","author":[{"family":"Lee","given":"Haemin"},{"family":"Kim","given":"Yerin"},{"family":"Sykora","given":"Daniel"},{"family":"Ryu","given":"Alexander"},{"family":"Cho","given":"Y"},{"family":"Kim","given":"Joonghee"},{"family":"Song","given":"Joanne"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1038/s41746-026-02682-7","URL":"https://doi.org/10.1038/s41746-026-02682-7","source":"openalex"},{"id":"oa:W7147240224","type":"article-journal","title":"Heat for Healing: A Review of Infrared Thermography in Medical Diagnostics and Therapy","abstract":"Infrared thermography (IRT) is an emerging noninvasive imaging modality that provides real-time, contactless assessment of skin surface temperature, reflecting underlying vascular perfusion. This narrative review explores the principles, clinical utility, advantages, limitations, and future potential of IRT in vascular diagnostics and monitoring. IRT has demonstrated diagnostic relevance across a spectrum of vascular conditions, including peripheral arterial disease, diabetic foot complications, venous insufficiency, Raynaud's phenomenon, and postoperative vascular monitoring. Its key benefits - such as radiation-free imaging, portability, and dynamic functional assessment - make it especially valuable for use in vulnerable populations and resource-limited settings. However, challenges such as environmental sensitivity, lack of standardized imaging protocols, and limited specificity necessitate further validation. With the integration of artificial intelligence and wearable technology, IRT holds significant promise as a complementary tool in modern vascular medicine.","author":[{"family":"Kumar","given":"Bitesh"},{"family":"Jain","given":"Vishesh"},{"family":"Yadav","given":"Devendra"},{"family":"Dhua","given":"Anjan"},{"family":"Goel","given":"Prabudh"},{"family":"Jain","given":"Dhawal"},{"family":"Singh","given":"Shubhendu"}],"issued":{"date-parts":[[2026]]},"DOI":"10.4103/jmp.jmp_175_25","URL":"https://doi.org/10.4103/jmp.jmp_175_25","source":"openalex"},{"id":"oa:W7124240574","type":"article-journal","title":"Symbiotic intelligence in dental trauma diagnostics—an exploratory case study","abstract":"Dental trauma in children is common and requires prompt diagnosis, which can be challenging in remote or isolated settings with limited access to emergency dental care. This exploratory case study investigates whether OpenAI's o3 can support dental trauma diagnostics in primary incisors, building on prior pretesting of GPT-4 on summative dental education exams (2023) and multimodal dental trauma analyses (2024), and focusing on o3's multimodal capability and reliability in 2025 with expert assessment (“human in the loop”) prior to a supervisor seminar with students and supervisors ( N = 84). Preliminary findings indicate that GPT-4 performed well on sample exams (2023), and that 7/10 multimodal analyses of dental injuries were accurate (2024); in the 2025 case, o3 correctly identified pulp necrosis in tooth 51 and uncomplicated enamel/dentin fractures in teeth 51 and 61, consistent with IADT guidance. Human expert involvement contributed essential validation, particularly for treatment decisions and ethical considerations. Overall, the study illustrates how symbiotic intelligence—purposeful collaboration between human and AI—may enhance learning outcomes in scenario-based simulations in remote areas, while requiring active human involvement and multiple validation communities.","author":[{"family":"Krumsvik","given":"Rune"},{"family":"Klock","given":"Kristin"},{"family":"Bratteberg","given":"Magnus"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/froh.2025.1687841","URL":"https://doi.org/10.3389/froh.2025.1687841","source":"openalex"},{"id":"oa:W4416234224","type":"article-journal","title":"Achieving health equity in immune disease: leveraging big data and artificial intelligence in an evolving health system landscape","abstract":"Prevalence of immune diseases is rising, imposing burdens on patients, healthcare providers, and society. Addressing the future impact of immune diseases requires \"big data\" on global distribution/prevalence, patient demographics, risk factors, biomarkers, and prognosis to inform prevention, diagnosis, and treatment strategies. Big data offer promise by integrating diverse real-world data sources with artificial intelligence (AI) and big data analytics (BDA), yet cautious implementation is vital due to the potential to perpetuate and exacerbate biases. In this review, we outline some of the key challenges associated with achieving health equity through the use of big data, AI, and BDA in immune diseases and present potential solutions. For example, political/institutional will and stakeholder engagement are essential, requiring evidence of return on investment, a clear definition of success (including key metrics), and improved communication of unmet needs, disparities in treatments and outcomes, and the benefits of AI and BDA in achieving health equity. Broad representation and engagement are required to foster trust and inclusivity, involving patients and community organizations in study design, data collection, and decision-making processes. Enhancing technical capabilities and accountability with AI and BDA are also crucial to address data quality and diversity issues, ensuring datasets are of sufficient quality and representative of minoritized populations. Lastly, mitigating biases in AI and BDA is imperative, necessitating robust and iterative fairness assessments, continuous evaluation, and strong governance. Collaborative efforts to overcome these challenges are needed to leverage AI and BDA effectively, including an infrastructure for sharing harmonized big data, to advance health equity in immune diseases through transparent, fair, and impactful data-driven solutions.","author":[{"family":"Kachnowski","given":"Stan"},{"family":"Khan","given":"Asif"},{"family":"Floquet","given":"Shadé"},{"family":"Whitlock","given":"Kendal"},{"family":"Wisnivesky","given":"Juan"},{"family":"Neill","given":"Daniel"},{"family":"Dankwamullan","given":"Irene"},{"family":"Ortega","given":"Gezzer"},{"family":"Daoud","given":"Moataz"},{"family":"Zaheer","given":"Raza"},{"family":"Hightower","given":"Maia"},{"family":"Rowe","given":"Paul"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fdata.2025.1621526","URL":"https://doi.org/10.3389/fdata.2025.1621526","source":"openalex"},{"id":"oa:W4413919806","type":"article-journal","title":"Role of Artificial Intelligence in P2P Energy Trading for Transforming Smart Homes to Smart Cities: A Comprehensive Survey","abstract":"Smart infrastructures, electric cars, and DERs have hastened the transition from centralized power systems to decentralized, prosumer-driven energy ecosystems. P2P energy trading allows families, villages, and cities to trade energy directly without a third party, making it important to this shift. These markets demand decision-making skills beyond standard control approaches because to their complexity, stochasticity, and real-time monitoring. AI is becoming the key driver of P2P trading, offering prosumers and municipalities with intelligent forecasting, optimization, automation, and security solutions. This survey provides a comprehensive overview of the role of AI in P2P energy trading, outlining its applications in smart homes and smart cities. We survey over a decade of research and present a taxonomy of more than 40 AI techniques, including machine learning, deep learning, reinforcement learning, optimization heuristics, federated and transfer learning, hybrid AI-blockchain models, and edge-cloud intelligence. We analyze the capability, application domains, challenges, and case studies of each technique. The poll also consolidates key open issues including privacy, interoperability, explainability, real-time latency, and the lack of standardized AI benchmarking platforms. Based on these insights, we propose future directions such as federated multi-agent frameworks, behavioral modeling, massive language model integration, quantum AI, standardized digital twin platforms, AI-driven policy simulations, and debiasing of ethical bias. This work reveals how AI can transform smart homes into active prosumers and their interactions into city-scale decentralized energy ecosystems by rigorously connecting technological advancements to real-world challenges.Keyword: Artificial intelligence, P2P energy trading, smart homes, smart cities, distributed energy resource, intelligent energy management systems, energy market design, sustainable energy transition.","author":[{"family":"Raza","given":"Ali"},{"family":"Iqbal","given":"Muhammad"},{"family":"Adnan","given":"Muhammad"}],"issued":{"date-parts":[[2025]]},"DOI":"10.22541/au.175683829.98253473/v1","URL":"https://doi.org/10.22541/au.175683829.98253473/v1","source":"openalex"},{"id":"oa:W7135190630","type":"article-journal","title":"The Performance of Artificial Intelligence in Classifying Molecular Markers in Adult-Type Gliomas Using Histopathological Images: Systematic Review","abstract":"Background: Adult-type gliomas are among the most prevalent and lethal primary central nervous system tumors, where prompt and accurate diagnosis is essential for maximizing survival prospects. Molecular classification, particularly the detection of isocitrate dehydrogenase (IDH) mutations and 1p/19q codeletions, has become crucial for accurate diagnosis and prognosis. Artificial intelligence (AI) has emerged as a promising adjunct in enhancing diagnostic accuracy using histopathological images. Existing reviews mostly focused on radiology rather than histopathology, and no comprehensive systematic review has specifically evaluated AI performance exclusively from histopathological images for detecting these two molecular markers. Objective: This study aims to systematically evaluate the performance of AI models in detecting and classifying IDH mutation status and 1p/19q gene codeletion in adult-type gliomas using histopathological images. Methods: A systematic review was conducted in accordance with PRISMA-DTA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Extension for Diagnostic Test Accuracy) guidelines. Seven databases (MEDLINE, PsycINFO, Embase, IEEE Xplore, ACM Digital Library, Scopus, and Google Scholar) were searched for studies published between 2015 and 2025. Eligible studies used AI models on histopathological images for molecular classification of adult-type gliomas and reported performance metrics. Study selection, data extraction, and risk of bias assessment using a modified QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2) tool were conducted independently by two reviewers. Extracted data were synthesized narratively. Results: A total of 2453 reports were identified, with 22 studies meeting the inclusion criteria. The pooled average accuracy, sensitivity, specificity, and area under the curve (AUC) across studies were 85.46%, 84.55%, 86.03%, and 86.53%, respectively. Hybrid models demonstrated the highest diagnostic performance (accuracy 92.80% and sensitivity 89.62%). In general, AI models that used multimodal data outperformed those that used unimodal data in terms of sensitivity (90.15% vs 84.31%) and AUC (88.93% vs 86.29%). Furthermore, models had a better overall performance in identifying IDH mutations than 1p/19q codeletions, with higher accuracy (86.13% vs 81.63%), specificity (86.61% vs 78.11%), and AUC (86.74% vs 85.15%). Unexpectedly, AI models designed for binary classification exhibited lower performance than those for multiclass classification in terms of both accuracy (91.98% vs 84.02%) and sensitivity (93.41% vs 80.18%). However, these differences should be interpreted as descriptive trends rather than statistically validated superiority, as formal between-group comparisons were not feasible. Conclusions: AI models show strong potential as complementary tools for the molecular classification of adult-type gliomas using histopathology images, particularly for IDH mutation detection. However, these findings are constrained by the limited number of studies, the focus on adult-type gliomas, lack of meta-analysis, and restriction to English-language publications. While AI offers valuable diagnostic support, it must be integrated with expert clinical judgment. Future research should prioritize larger, more diverse datasets and multimodal AI frameworks and extend to other brain tumor types for broader applicability.","author":[{"family":"Almaabreh","given":"Obada"},{"family":"Al-Dafi","given":"Rukaya"},{"family":"Tabassum","given":"Aliya"},{"family":"Othman","given":"Ahmad"},{"family":"Abd-Alrazaq","given":"Alaa"}],"issued":{"date-parts":[[2026]]},"DOI":"10.2196/78377","URL":"https://doi.org/10.2196/78377","source":"openalex"},{"id":"oa:W7140192272","type":"article-journal","title":"Artificial intelligence-driven gastrointestinal functional assessment: multimodal imaging, digital biomarkers, and real-time monitoring","abstract":"Gastrointestinal (GI) functional disorders and chronic inflammatory diseases impose a substantial health burden, yet their assessment remains challenging because symptoms reflect dynamic interactions among motility, visceral sensation, immune-microbiome regulation, and brain-gut signaling. Artificial intelligence (AI) is rapidly reshaping GI functional medicine by enabling scalable, quantitative interpretation of complex data generated from multimodal imaging, physiological sensing, and real-world patient monitoring. This review synthesizes advances across three tightly connected pillars that map onto a physiology-informed \"assessment-to-action\" loop: (i) AI-assisted multimodal GI imaging for quantitative phenotyping and integrated diagnosis; (ii) AI-enabled discovery and validation of digital biomarkers that capture dynamic GI function in naturalistic settings; and (iii) real-time monitoring platforms that support early warning, longitudinal assessment, and adaptive management. We summarize representative applications in functional GI disorders, inflammatory bowel disease (IBD), and GI oncology, highlighting methodological themes including multimodal fusion, temporal modeling, uncertainty estimation, and explainable AI. We then discuss barriers to translation-standardization and interoperability, external validation under dataset shift, privacy and governance, and workflow integration-and outline practical directions for building clinically trustworthy AI systems for GI functional assessment. Collectively, physiology-centered AI approaches have the potential to transform GI care from episodic testing to longitudinal, mechanism-aware monitoring and personalized intervention.","author":[{"family":"Li","given":"Liucheng"},{"family":"Lv","given":"Fang"},{"family":"Du","given":"Chen"},{"family":"Yang","given":"Lianjun"},{"family":"Pa","given":"Chengzhou"},{"family":"Dai","given":"Yunrui"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/fphys.2026.1778235","URL":"https://doi.org/10.3389/fphys.2026.1778235","source":"openalex"},{"id":"oa:W7127149757","type":"article-journal","title":"Data pipeline quality: development and validation of a quality assessment tool for data-driven algorithms and artificial intelligence in healthcare","abstract":"OBJECTIVES: To develop and validate a tool for standardised quality assessment of data-driven algorithms in healthcare, focusing on the underlying data pipeline. METHODS: Data Assessment Tool for Algorithm Critical Appraisal and Robust Evidence (DATA-CARE) was iteratively developed from the established Quality In Prognosis Studies framework, selected after reviewing 10 existing quality assessment tools for observational and artificial intelligence studies. DATA-CARE evaluates five quality domains of the data pipeline: study population, data, algorithm, outcome and report transparency. Each domain comprises three to five quality criteria. With a total score of 75 points, study quality is categorised as low (<45), moderate (45-59) or high (≥60). DATA-CARE was validated during a systematic review on data-driven algorithms using continuous physiological monitoring data within the paediatric intensive care unit. Two independent reviewers performed quality assessment using DATA-CARE of included studies. Tool validation was evaluated using inter-rater agreement and intraclass correlation coefficient (ICC). RESULTS: DATA-CARE demonstrated robust inter-rater agreement (93.5%) with ICC 0.98 (95% CI 0.96 to 0.99). Of 3858 screened studies, 31 were reviewed in the use case, describing diverse algorithms. Studies were predominantly low (32.3%) to moderate (41.9%) and sporadically (25.8%) high quality. DISCUSSION: Predominance of low-to-moderate quality studies reveals critical barriers to clinical implementation of data-driven algorithms, including low quality data capture and processing, lacking validation strategies and non-transparent reporting of findings. CONCLUSIONS: DATA-CARE allows standardised and reliable critical appraisal for a wide variety of algorithms, addressing current gaps in standardised and reproducible algorithm development.","author":[{"family":"Twist","given":"Eris"},{"family":"Winden","given":"Brian"},{"family":"Jonge","given":"Rogier"},{"family":"Taal","given":"HR"},{"family":"Hoog","given":"Matthijs"},{"family":"Schouten","given":"Alfred"},{"family":"Tax","given":"David"},{"family":"Kuiper","given":"Jan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1136/bmjhci-2025-101608","URL":"https://doi.org/10.1136/bmjhci-2025-101608","source":"openalex"},{"id":"oa:W7147256499","type":"article-journal","title":"Artificial Intelligence-Based Exosome Analysis for Improving Diagnostic Performance of Breast Lesions on Ultrasound: Protocol of a Prospective, Multicenter Cohort Study","abstract":"PURPOSE: Exosome-surface enhanced Raman spectroscopy-artificial intelligence platform (exosome-SERS-AI) is an innovative liquid biopsy method that acquires SERS signals from plasma exosomes and analyzes them using deep learning models to diagnose cancer. This study aimed to evaluate whether exosome-SERS-AI could increase the diagnostic accuracy of ultrasonography (US) for suspicious breast lesions. METHODS: This prospective multicenter study enrolled 500 patients between November 2024 and December 2025. Eligible participants will be women aged ≥ 40 years who will undergo US performed by specialized breast radiologists and have suspicious breast lesions assigned to a Breast Imaging Reporting and Data System (BI-RADS) category 3-5 assessment. A 6 mL whole blood sample was collected from each participant. After plasma separation, SERS, which is highly sensitive to exosomes, was employed to measure Raman signals, and the acquired data were processed using artificial intelligence algorithms. Following sampling, all patients underwent US-guided core needle biopsy for breast lesions classified as BI-RADS category 4 and 5, and 12-months of follow-up US for lesions classified as BI-RADS category 3. Histopathological examination was used as the reference standard for BI-RADS 4 and 5 lesions, whereas stability on 12-month follow-up US was used as the reference standard for BI-RADS 3 lesions. The cohort is expected to have an equal distribution of benign and malignant cases. The following outcome measures were compared between US alone and the combination of exosome-SERS-AI with US: sensitivity, specificity, positive predictive value, negative predictive value, and the area under the receiver operating characteristic curve. Enrollment is expected to be completed by 2025, and the study results are expected to be presented in 2026. DISCUSSION: This prospective multicenter study will evaluate the performance of exosome-SERS-AI compared to US in women with BI-RADS categories 3-5. Participant enrollment is ongoing. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT06672302. Registered on November 4, 2024.","author":[{"family":"Song","given":"Sung"},{"family":"Shin","given":"Hyunku"},{"family":"Park","given":"Yong"},{"family":"Choi","given":"Yeonho"},{"family":"Jung","given":"Seung"}],"issued":{"date-parts":[[2026]]},"DOI":"10.4048/jbc.2025.0206","URL":"https://doi.org/10.4048/jbc.2025.0206","source":"openalex"},{"id":"oa:W7134076461","type":"article-journal","title":"Artificial Intelligence in Venous Thromboembolism Prevention: A Narrative Review of Machine Learning, Deep Learning, and Natural Language Processing","abstract":"Venous thromboembolism (VTE), which includes deep vein thrombosis and pulmonary embolism, is a significant and preventable cause of morbidity and mortality worldwide. Despite the existence of clinical prediction models, biomarker-based risk assessments, and imaging techniques, gaps remain in accurately identifying and managing high-risk patients. In recent years, artificial intelligence has emerged as a transformative tool in healthcare, offering promising applications for enhancing VTE prevention strategies. This narrative review synthesizes current evidence on the use of artificial intelligence (AI) technologies including machine learning (ML), deep learning (DL), and natural language processing (NLP). We explore how supervised ML algorithms, such as random forests, support vector machines, and gradient boosting, improve predictive performance compared to traditional models by capturing complex, nonlinear relationships within electronic health record data. We also examine the role of DL models, particularly convolutional neural networks, in interpreting imaging data, achieving diagnostic accuracies comparable to expert radiologists. Additionally, the review highlights NLP applications in extracting risk-relevant information from unstructured clinical notes and the emerging integration of wearable device data and time-series analysis for dynamic risk assessment. We argue that the successful integration of AI into routine VTE prevention workflows requires rigorous prospective validation, cross-institutional collaboration, and thoughtful implementation into clinical decision support systems.","author":[{"family":"Crisan","given":"Daniela"},{"family":"Cut","given":"Talida"},{"family":"Herlo","given":"Lucian"},{"family":"Ivanović","given":"N"},{"family":"Herlo","given":"Alexandra"},{"family":"Alexandrescu","given":"Luana"},{"family":"Sălcudean","given":"Andreea"},{"family":"Dumache","given":"Raluca"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/jcdd13030119","URL":"https://doi.org/10.3390/jcdd13030119","source":"openalex"},{"id":"oa:W4415603723","type":"article-journal","title":"Unveiling Dynamic Resilience on Sustainable Performance in Supply Chain: Artificial Intelligence, System Optimization and Information Transparency","abstract":"ABSTRACT Supply chain resilience (SCR) enhances the use of artificial intelligence, system optimization, and information transparency to facilitate the dynamics in sustainable supply chain performance improvement. Prior studies neglect system dynamics involvement to identify the dynamic changes. This study proposes a hybrid method for extracting real practices from the globally registered patents and extracts these variables by adopting bidirectional encoder representations of transformers. The results of comparing models based on different environmental data reveal that supply chain resilience does not reliably improve sustainable supply chain performance, especially during crises. Artificial intelligence is able to generate dynamics in the SCR. The resilient systems could be improved through system optimization and information transparency without artificial intelligence in both the before and after dynamic environments.","author":[{"family":"Wu","given":"Kuo‐jui"},{"family":"Han","given":"Ming‐yong"},{"family":"Huang","given":"Caiyan"},{"family":"Qiu","given":"Hailing"},{"family":"Sethanan","given":"Kanchana"},{"family":"Tseng","given":"Ming‐lang"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/jbl.70046","URL":"https://doi.org/10.1111/jbl.70046","source":"openalex"},{"id":"oa:W7131354288","type":"article-journal","title":"Artificial Intelligence in Adverse Outcome Pathways: A Review of Strategies for Automated Information Extraction, Quantitative Analysis, and Iterative Optimization","abstract":"The rapid emergence of novel chemical substances escalates the occupational and environmental health risks, posing significant challenges to the traditional toxicological risk assessment framework. While adverse outcome pathways (AOPs) have become a pivotal theoretical framework for alternative toxicity testing and future risk assessments, their development and optimization remain hindered by time-consuming and labor-intensive manual processing. This narrative review systematically elucidates how artificial intelligence (AI) facilitates the development and optimization of AOPs. Specifically, AI automates the extraction of knowledge modules for AOPs via natural language processing, quantifies key relationships through integrating methods like Bayesian networks, and supports continuous AOP refinement using machine learning platforms. Together, these technologies establish a modern, data-driven, and iterative framework. Furthermore, the review discusses the current limitations in applying AI to the AOP domain alongside its substantial potential to enhance chemical risk assessment and regulatory decision-making. Ultimately, this work aims to provide new insights and methodologies for advancing AOP development, thereby strengthening the risk assessment and regulation of chemical exposures in environmental and occupational settings.","author":[{"family":"Zhu","given":"Ziqi"},{"family":"Hu","given":"Guiping"},{"family":"Jia","given":"Guang"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/occuphealth1010009","URL":"https://doi.org/10.3390/occuphealth1010009","source":"openalex"},{"id":"oa:W7128705500","type":"article-journal","title":"Bioinspired Cross‐Modal Self‐Adaptive Machine Intelligence for Event‐Driven and Ultrahigh‐Precision Underwater Grasping","abstract":"ABSTRACT Embodied intelligent agents, which represent the future of robotics, demand precise perception and real‐time decision‐making capabilities to achieve natural environmental interactions. However current systems face inherent limitations in unimodal sensing and cross‐modal coordination, which hinder their performance in dynamic contact‐rich operations. Herein, we present a fabric‐based event‐driven tactile interface that features an innovative woven structure with cross‐fiber electrodes. It achieves breakthroughs in sensitivity (246.3 kPa −1 ), pressure detection (>450 kPa), and waterproof robustness. This interface enables millisecond‐level pressure/slip dual‐mode feedback for self‐adaptive grasping, thereby improving the dexterous manipulation of fragile or slippery objects. For underwater scenarios, a bio‐inspired visual–tactile fusion (VTF) architecture leverages tactile perception to compensate for visual limitations, demonstrating a high accuracy of 97.7% in complex tasks, including underwater transparent object manipulation and recognition of similar objects. Event‐driven tactile feedback is merged with visual semantics for decision‐level optimization, thereby enhancing the autonomy and adaptation of humanoid machine intelligence. It creates an innovative closed‐loop cross‐modal perception–decision system that builds a direct link between environmental interaction and autonomous decision‐making for intelligent agent development in open‐world scenarios. The superior performance of the VTF architecture dynamic interaction tasks represents a crucial step toward robotic systems with advanced intelligence.","author":[{"family":"Chen","given":"Hongyu"},{"family":"Huang","given":"Zijian"},{"family":"Luo","given":"Yanhao"},{"family":"Wang","given":"Yujin"},{"family":"Wang","given":"Huasen"},{"family":"Liu","given":"Lei"},{"family":"Sun","given":"Yu"},{"family":"Hu","given":"Yu"},{"family":"Lin","given":"Yuchen"},{"family":"Wei","given":"Chao"},{"family":"Lin","given":"Wenjun"},{"family":"Su","given":"Gantang"},{"family":"Guo","given":"Ziquan"},{"family":"Zheng","given":"Jianghui"},{"family":"Chen","given":"Zhiqi"},{"family":"Liao","given":"Qingliang"},{"family":"Zheng","given":"Yuanjin"},{"family":"Xinqin","given":"Liao"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1002/adma.202519665","URL":"https://doi.org/10.1002/adma.202519665","source":"openalex"},{"id":"oa:W4406271458","type":"article-journal","title":"Machine Learning‐Driven Prediction, Preparation, and Evaluation of Functional Nanomedicines Via Drug–Drug Self‐Assembly","abstract":"Small molecules as nanomedicine carriers offer advantages in drug loading and preparation. Selecting effective small molecules for stable nanomedicines is challenging. This study used artificial intelligence (AI) to screen drug combinations for self-assembling nanomedicines, employing physiochemical parameters to predict formation via machine learning. Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) are identified as effective carriers for antineoplastic drugs, with high drug loading. Nanomedicines, PEG-coated indomethacin/paclitaxel nanomedicine (PiPTX), and laminarin-modified indomethacin/paclitaxel nanomedicine (LiDOX), are developed with extended circulation and active targeting functions. Indomethacin/paclitaxel nanomedicine iDOX exhibits pH-responsive drug release in the tumor microenvironment. These nanomedicines enhance anti-tumor effects and reduce side effects, offering a rapid approach to clinical nanomedicine development.","author":[{"family":"Zhang","given":"Chengyuan"},{"family":"Yuan","given":"Yuchuan"},{"family":"Xia","given":"Qiong"},{"family":"Wang","given":"Junjie"},{"family":"Xu","given":"Kang"},{"family":"Gong","given":"Zhiwei"},{"family":"Lou","given":"Jie"},{"family":"Li","given":"Gen"},{"family":"Wang","given":"Lu"},{"family":"Zhou","given":"Li"},{"family":"Liu","given":"Zhirui"},{"family":"Luo","given":"Kui"},{"family":"Zhou","given":"Xing"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/advs.202415902","URL":"https://doi.org/10.1002/advs.202415902","source":"openalex"},{"id":"oa:W7164135050","type":"article-journal","title":"EFFECT OF AIR POLLUTION MITIGATION BY ARTIFICIAL INTELLIGENCE TECHNOLOGY ON URBAN RESIDENTS’ HEALTH: EVIDENCE FROM 286 CITIES IN CHINA","abstract":"The technological revolution represented by Artificial Intelligence (AI) not only drives substantial changes in productivity, but also provides powerful impetus for profoundly altering people’s lifestyle. This article explores the effect of AI on residents’ health with a fixed effects model, TVP-SVVAR model and data of 286 cities from 2011 to 2023 in China. The results indicate that AI can decrease mortality rates and improve resident health. Mechanism analysis shows that AI improves resident health by mitigating air pollution. Heterogeneity analysis reveals AI exerts a significant improvement effect in the eastern and central regions, while the positive effect is insignificant in the western region. Further analysis illustrates that long term effects of AI on resident health exhibit time-varying characteristics. The above findings could provide insights and suggestions into air pollution suppression and resident health enhancement for prefecture cities of China.","author":[{"family":"Dong","given":"K"},{"family":"Ge","given":"Y"},{"family":"Liu","given":"TT"},{"family":"Wang","given":"GL"},{"family":"Dai","given":"PP"},{"family":"Meng","given":"ZH"}],"issued":{"date-parts":[[2026]]},"DOI":"10.15666/aeer/2403_35573575","URL":"https://doi.org/10.15666/aeer/2403_35573575","source":"openalex"},{"id":"oa:W7130723889","type":"article-journal","title":"Optimization of Hybrid Renewable Energy Systems: Classical Optimization Methods, Artificial Intelligence, Recent Trend, and Software Tools","abstract":"This article synthesizes the state of the art in the optimization of Hybrid Renewable Energy Systems (HRES), emphasizing that robust HRES planning is inherently an integrated sizing-and-dispatch problem constrained by techno-economic, environmental, and reliability requirements. The review first consolidates classical optimization methods, highlighting the continued relevance of deterministic programming (LP/MILP/MINLP) for transparent and reproducible co-optimization of capacity investment and operational dispatch, alongside analytical, graphical, iterative, and probabilistic approaches for feasibility screening and baseline benchmarking. It then evaluates artificial intelligence–based optimization techniques, including evolutionary computation, swarm intelligence, and multi-objective evolutionary frameworks, noting their effectiveness in nonconvex, mixed-variable, and simulation-driven sizing problems while underscoring the need for rigorous constraint handling, statistical validation, and transparent reporting of computational budgets. The article further examines hybrid optimization strategies that integrate global search with exact dispatch solvers, surrogate-assisted learning, decomposition schemes, and control–co-design paradigms, identifying these as mature approaches that enhance feasibility, scalability, and operational realism. Recent trends in newly proposed AI optimizers are critically discussed, with emphasis on reproducibility, sensitivity analysis, and fair benchmarking against strong baselines. Finally, the article outlines the role of software tools in enabling practical HRES optimization, spanning packaged techno-economic platforms, solver-based modeling environments, and co-simulation workflows for network-constrained planning. Overall, the findings indicate a clear progression toward multi-objective, uncertainty-aware, degradation-informed formulations implemented through integrated toolchains and hybrid solver–AI architectures, with future work warranted on uncertainty quantification, network and resilience constraints, and reproducible evaluation protocols.","author":[{"family":"Imbayah","given":"Ibrahim"},{"family":"Khaleel","given":"Mohamed"},{"family":"Yusupov","given":"Zıyodulla"}],"issued":{"date-parts":[[2025]]},"DOI":"10.65998/ijees.v3i4.150","URL":"https://doi.org/10.65998/ijees.v3i4.150","source":"openalex"},{"id":"oa:W7118208329","type":"article-journal","title":"Model confrontation and collaboration: A debate intelligence framework for enhancing medical reasoning in large language models","abstract":"Medical reasoning is fundamental to clinical decision-making, underpinning tasks such as patient communication, diagnosis, and treatment planning. Inspired by psychological findings that peer interaction promotes self-correction, we introduce model confrontation and collaboration (MCC), a debate intelligence framework that transcends static ensemble methods by integrating critique and self-reflection to iteratively refine reasoning through structured, multi-round confrontation and collaboration among diverse large language models (LLMs). In multiple-choice benchmarks, MCC achieved mean accuracy on MedQA (92.6%) and PubMedQA (84.8%) and demonstrated strong performance on medical subsets of MMLU. In long-form medical question answering, MCC outperformed all individual LLMs and the domain-specific LLM Med-PaLM 2 in both physician and layperson evaluations. In diagnostic dialog tasks, MCC further excelled in both history-taking and diagnostic accuracy, reaching a top-1 diagnosis rate of 80%. These results position MCC as a scalable, model-agnostic framework that advances medical reasoning through collaborative deliberation.","author":[{"family":"Sun","given":"Xinti"},{"family":"Hong","given":"Qiyang"},{"family":"Zhang","given":"Mengyan"},{"family":"Li","given":"Yuyan"},{"family":"Chen","given":"Tingwei"},{"family":"Huang","given":"Zigeng"},{"family":"Liang","given":"Guihan"},{"family":"Tang","given":"Wenjun"},{"family":"Xu","given":"Sulin"},{"family":"Ni","given":"Xiaolin"},{"family":"Pang","given":"Junling"},{"family":"Wan","given":"Peixing"},{"family":"Long","given":"Erping"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1016/j.xcrm.2025.102547","URL":"https://doi.org/10.1016/j.xcrm.2025.102547","source":"openalex"},{"id":"oa:W4415076046","type":"article-journal","title":"Emotional Intelligence Training Correlates With Medical Students’ Apprehension of AI in Healthcare: A Single-Institution, Observational Study","abstract":"Background: The use of artificial intelligence (AI) in healthcare is becoming increasingly widespread. While AI has been touted to improve many aspects of the field, its potential to undermine the doctor-patient relationship has also been recognized. It is not known what future physicians believe regarding the potential effects of AI on physicians' professional relationships. It is also unknown whether formal training in emotional intelligence (EI) and resilience strategies influences medical students' opinions on the use of AI in healthcare. Objective: To ascertain medical students’ opinion on the potential impact of AI on various EI-related aspects of healthcare, such as its use in guiding doctor-patient interactions, and to determine whether formal training in emotional intelligence and resilience (EIR) is associated with their opinion on these topics. Methods: Approximately 700 medical students were asked to voluntarily and anonymously complete a 12-item survey. All survey items were required to be answered for students' responses to be accepted and analyzed. Agreement with survey items was measured via a Likert-type scale, with 1 = strongly disagree and 5 = strongly agree. Results were summarized as means and standard deviations, stratified by whether students did (EIR+) or did not (EIR-) take our institution's EIR elective course. Median responses per survey item were compared across EIR+/- groups via the Mann-Whitney U test (α < 0.05), a non-parametric test comparing two independent samples allowing for unequal group size. Results: A total of 50 EIR+ students (59%) and 35 EIR- students (41%) replied, making the survey response rate approximately 12.14%. Compared to their counterparts, EIR+ students indicated statistically significantly greater disagreement with “I believe AI technology will improve the doctor-patient relationship” (3.0 (± 1.2) vs. 3.5 (± 1.0); p = 0.0327). While not statistically significant, EIR+ students also reported a similar magnitude (~0.5 points) of greater disagreement with “AI has a place in guiding physicians on how to better interact with their patients and colleagues” (2.9 (± 1.2) vs. 3.4 (± 1.1); p = 0.0701) and “Overall, I think adopting the use of AI in healthcare will be beneficial” (3.6 (± 1.2) vs. 4.0 (± 1.1); p = 0.0521). Conclusions: The preliminary data from our single-institution, observational study suggest EIR-trained medical students may be more cautious of AI usage in healthcare due to the potential negative impact AI can have on the doctor-patient relationship. However, larger studies using validated surveys are required to confirm this conclusion.","author":[{"family":"Runde","given":"Austin"},{"family":"Mishra","given":"Shambhavi"},{"family":"Feffer","given":"Marina"},{"family":"Shahid","given":"Ramzan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.94349","URL":"https://doi.org/10.7759/cureus.94349","source":"openalex"},{"id":"oa:W7151479764","type":"article-journal","title":"Artificial Intelligence in MRI-Based Glioma Imaging: From Radiomics-Based Machine Learning to Deep Learning Approaches","abstract":"Gliomas are generally readily detected and broadly characterized using conventional MRI; however, substantial challenges remain in accurately delineating tumor extent, grading heterogeneous disease, and translating imaging findings into consistent, reproducible clinical decisions. Despite reported Dice coefficients of 0.85–0.91 for whole-tumor segmentation and classification AUC values exceeding 0.90 for glioma grading in curated datasets, most AI systems remain limited by validation design, dataset bias, and inadequate external generalizability. This narrative review synthesizes current AI applications for MRI-based glioma detection and segmentation, highlighting the evolution from radiomics-based classical machine learning approaches relying on handcrafted features to deep learning models capable of end-to-end representation learning. Commonly used MRI sequences, algorithmic paradigms, and reported performance trends are reviewed, with particular emphasis on tumor segmentation as a foundational enabling task. Key limitations that hinder clinical translation are examined, including limited dataset diversity, validation practices that inflate reported performance, domain shift across institutions, acquisition-related bias, and inadequate model interpretability. Emerging strategies to address these challenges, such as multi-institutional training, harmonization techniques, explainable AI frameworks, and workflow-integrated validation, are also discussed. While AI-based models demonstrate strong technical performance in research settings, their clinical impact will depend on rigorous external validation, transparency, and alignment with real-world neuro-oncology workflows.","author":[{"family":"Saloum","given":"Ammar"},{"family":"Zaher","given":"Israa"},{"family":"Stipho","given":"Christian"},{"family":"Demir","given":"Enes"},{"family":"Naravetla","given":"Varun"},{"family":"Pahlevani","given":"Mehrdad"},{"family":"Yaghi","given":"Nasser"},{"family":"Karsy","given":"Michael"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/biomedinformatics6020020","URL":"https://doi.org/10.3390/biomedinformatics6020020","source":"openalex"},{"id":"oa:W7125477330","type":"article-journal","title":"Managing Conflict of Interest in Clinical Practice Guidelines With Artificial Intelligence: Insights From Large Language Models and Beyond","abstract":"BACKGROUND: Conflict of interest (COI) management is critical for ensuring the scientific integrity and fairness of clinical practice guidelines (CPGs). Large language models (LLMs) have great potential in strengthening COI management, particularly in information collection, assessment, and supporting guideline development groups. OBJECTIVE: To explore LLMs' role in COI management during CPG development, focusing on applications, challenges, and future directions. METHODS: We examined how LLMs can support COI management by designing and testing a set of simulated COI scenarios based on established management principles. RESULTS: LLMs can improve efficiency in data collection (e.g., in analyzing disclosures), objectivity in risk assessment, and transparency in reporting. However, privacy risks (e.g., data breaches) and technical issues (e.g., model bias) hinder the adoption of LLM based approaches. Setting up policy frameworks, research collaboration, and enhanced security, such as differential privacy levels, can enhance reliability. CONCLUSION: LLMs can support COI management in CPG development if ethical issues are adequately considered, but validation in real-world settings is still needed.","author":[{"family":"Wang","given":"Ye"},{"family":"Wang","given":"Qi"},{"family":"Xun","given":"Yangqin"},{"family":"Zhou","given":"Qi"},{"family":"Zhang","given":"Huayu"},{"family":"Liu","given":"Hanxiang"},{"family":"Qin","given":"Yishan"},{"family":"Wu","given":"MY"},{"family":"Wang","given":"Zijing"},{"family":"Li","given":"Haodong"},{"family":"Estill","given":"Janne"},{"family":"Chen","given":"Yaolong"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1111/jebm.70114","URL":"https://doi.org/10.1111/jebm.70114","source":"openalex"},{"id":"oa:W7160282947","type":"article-journal","title":"Knowledge, Attitude, Benefits, Risks, Barriers, Professional Impact, and Preparedness of Nursing Students Toward the Utilization of Artificial Intelligence in Healthcare","abstract":"Background/Objectives: Artificial intelligence (AI) is increasingly used in healthcare to support clinical decision-making, patient monitoring, and administrative tasks. Nurses are expected to work with these technologies. However, the evidence suggests that their knowledge and preparedness remain limited. As future healthcare providers, nursing students must be prepared to integrate AI into their practice. This study aimed to assess nursing students’ knowledge, attitudes, perceived benefits and risks, barriers, professional impact, and preparedness toward AI in healthcare. Methods: This cross-sectional descriptive study was conducted between April and July 2024 at the College of Nursing, University of Hail, Saudi Arabia. A convenience sample of 320 undergraduate nursing students completed an online structured questionnaire that assessed their demographics, knowledge, attitudes, perceived barriers, benefits, risks, professional impact, and preparedness. Data were analyzed using IBM SPSS version 27 with descriptive statistics. Inferential analyses, including independent t-tests and one-way ANOVA, were performed to examine differences between groups. Pearson’s correlation was used to identify correlations between the study variables. Statistical significance was set at p < 0.05. Results: Most students (79.7%) had poor AI knowledge, whereas 52.5% reported positive attitudes. Older students (≥24 years) and internship students showed significantly more positive attitudes (p < 0.001). Knowledge was weakly correlated with attitudes (r = 0.147), benefits (r = 0.222), and risks (r = 0.152). Attitudes were weakly positively correlated with benefits (r = 0.243) and negatively correlated with barriers (r = −0.219). Conclusions: Despite their positive attitudes, nursing students showed limited knowledge and preparedness. Integrating AI education and practical training into nursing curricula is therefore essential. These findings should be interpreted cautiously given the cross-sectional design, single-institution sampling, and reliance on self-reported measures, which may limit generalizability.","author":[{"family":"Alrasheeday","given":"Awatif"},{"family":"Alhawsawi","given":"Aeshah"},{"family":"Alshammari","given":"Bushra"},{"family":"Alkubati","given":"Sameer"},{"family":"Aouicha","given":"Wiem"},{"family":"Tlili","given":"Mohamed"},{"family":"Alharbi","given":"Abdulhafith"},{"family":"Siam","given":"Bahia"},{"family":"Mahmoud","given":"Soha"},{"family":"Elamin","given":"Badria"},{"family":"Alshammari","given":"Layla"},{"family":"Motakef","given":"Hajer"},{"family":"Alkhammali","given":"Tahani"},{"family":"Alanazi","given":"Ahad"},{"family":"Alshammari","given":"Fatimah"},{"family":"Alshammari","given":"Huda"},{"family":"Almohammed","given":"Ruqayyah"},{"family":"Alomran","given":"Ruba"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/nursrep16050154","URL":"https://doi.org/10.3390/nursrep16050154","source":"openalex"},{"id":"oa:W7128029160","type":"article-journal","title":"A generalizable foundation model for analysis of human brain MRI","abstract":"Artificial intelligence applied to brain magnetic resonance imaging (MRI) holds potential to advance diagnosis, prognosis and treatment planning for neurological diseases. The field has been constrained, thus far, by limited training data and task-specific models that do not generalize well across patient populations and medical tasks. By leveraging self-supervised learning, pretraining and targeted adaptation, foundation models present a promising paradigm to overcome these limitations. Here we present Brain Imaging Adaptive Core (BrainIAC)-a foundation model designed to learn generalized representations from unlabeled brain MRI data and serve as a core basis for diverse downstream application adaptation. Trained and validated on 48,965 brain MRIs across a broad spectrum of tasks, we demonstrate that BrainIAC outperforms localized supervised training and other pretrained models, particularly in low-data, few-shot, settings and in high-difficulty prediction tasks, allowing for application in scenarios otherwise infeasible. BrainIAC can be integrated into imaging pipelines and multimodal frameworks and may lead to improved biomarker discovery and artificial intelligence clinical translation.","author":[{"family":"Tak","given":"Divyanshu"},{"family":"Garomsa","given":"Biniam"},{"family":"Zapaishchykova","given":"Anna"},{"family":"Chaunzwa","given":"Tafadzwa"},{"family":"Pardo","given":"Juan"},{"family":"Ye","given":"Zezhong"},{"family":"Zielke","given":"John"},{"family":"Ravipati","given":"Yashwanth"},{"family":"Pai","given":"Suraj"},{"family":"Vajapeyam","given":"Sri"},{"family":"Mahootiha","given":"Maryam"},{"family":"Parker","given":"Mitchell"},{"family":"Pike","given":"Luke"},{"family":"Smith","given":"Ceilidh"},{"family":"Familiar","given":"Ariana"},{"family":"Liu","given":"Kevin"},{"family":"Prabhu","given":"Sanjay"},{"family":"Arnaout","given":"Omar"},{"family":"Bandopadhayay","given":"Pratiti"},{"family":"Nabavizadeh","given":"Ali"},{"family":"Mueller","given":"Sabine"},{"family":"Aerts","given":"Hugo"},{"family":"Huang","given":"RS"},{"family":"Poussaint","given":"Tina"},{"family":"Kann","given":"Benjamin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1038/s41593-026-02202-6","URL":"https://doi.org/10.1038/s41593-026-02202-6","source":"openalex"},{"id":"oa:W7154226246","type":"article-journal","title":"The Evolving Role of AI in Simulation-Based Medical Education: A Narrative Review","abstract":"Purpose: A growing body of literature has emerged on the topic of Artificial Intelligence (AI) use in Simulation-Based Medical Education (SBME) in recent years, but most studies have focused on isolated applications of AI to components of simulation, making it difficult for educators and decision makers to make informed decisions on the use of AI. Therefore, this narrative review aims to condense the current literature on the use of AI in the SBME, its influence on experiential learning, explore challenges, and future directions in this rapidly evolving field. Methods: A targeted literature search was conducted for this review on PubMed and Google Scholar, with combinations of keywords. Articles were selected from 2019 to 2025, based on their relevance to the use of AI in SBME, in areas of teaching, learning, and assessment. Studies without educational outcomes were excluded. Results: The search produced 2019 papers, out of which 45 were analyzed after applying the exclusion criteria. These showed that AI has been applied across multiple dimensions of the SBME, including scenario development, enhancing realism, personalized and collaborative learning, developing communication and psychomotor skills, and automated and AI augmented feedback. Several challenges have been raised, like ethical and privacy concerns, lack of AI literacy among users, lack of transparency, undesired outcomes, infrastructure cost, and environmental effects. Conclusion: Benefits of AI in SBME stem from AI-augmented human teaching rather than unsupervised usage of AI tools. Additionally, the potential for personalized and accessible learning warrants further research.","author":[{"family":"Hasan","given":"SH"},{"family":"Ahmed","given":"Ayesha"},{"family":"Ismail","given":"Faisal"}],"issued":{"date-parts":[[2026]]},"DOI":"10.2147/amep.s581691","URL":"https://doi.org/10.2147/amep.s581691","source":"openalex"},{"id":"oa:W7135044204","type":"article-journal","title":"Artificial intelligence-assisted reader evaluation in acute CT head interpretation (AI-REACT): a multireader multicase study","abstract":"Objective To assess whether an artificial intelligence (AI) tool improves the accuracy, speed and confidence of general radiologists, emergency clinicians and radiographers in detecting critical non-contrast CT head (NCCTH) abnormalities and to evaluate its stand-alone performance and factors influencing diagnostic accuracy. Methods and analysis A retrospective dataset of 150 NCCTH (52 normal and 98 with critical abnormalities) was reviewed by 30 readers (10 radiologists, 15 emergency clinicians and 5 radiographers) from four National Health Service trusts. Each interpreted scan is performed unaided and then with the qER EU 2.0 AI tool, separated by a 2-week washout period. Ground truth was established by two neuroradiologists. We measured the AI’s stand-alone performance and its effect on reader accuracy, confidence and speed. Results The qER algorithm showed strong diagnostic performance (area under the receiver operator curve 0.821–0.976). With AI, pooled reader sensitivity for critical abnormalities increased from 82.8% to 89.7% (+6.9%, p<0.001) and for intracranial haemorrhage from 84.6% to 91.6% (+7.0%, p<0.001), while specificity decreased from 84.5% to 78.9% (–5.5%, p=0.046). Reader confidence did not change significantly. Emergency department (ED) clinicians with AI achieved sensitivity similar to unaided radiologists. Conclusion AI assistance increased sensitivity for detecting critical abnormalities on NCCTH but reduced specificity. AI-enabled ED clinicians to achieve diagnostic sensitivity comparable to radiologists, supporting its potential to enhance non-radiologist performance. Further studies are needed to confirm these findings in clinical practice. Trial registration number NCT06018545 .","author":[{"family":"Novak","given":"Alex"},{"family":"Shah","given":"Ruchir"},{"family":"Morgado","given":"Abdala"},{"family":"Robert","given":"Dennis"},{"family":"Kumar","given":"Shamie"},{"family":"Oke","given":"Jason"},{"family":"Bhatia","given":"Kanika"},{"family":"Romsauerova","given":"Andrea"},{"family":"Das","given":"Tilak"},{"family":"Group","given":"The"},{"family":"Narbone","given":"Mariapaola"},{"family":"Dharmadhikari","given":"Rahul"},{"family":"Harrison","given":"Mark"},{"family":"Vimalesvaran","given":"Kavitha"},{"family":"Gooch","given":"Jane"},{"family":"Woznitza","given":"N"},{"family":"Lowe","given":"David"},{"family":"Shuaib","given":"Haris"},{"family":"Ather","given":"Sarim"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1136/bmjdh-2026-000071","URL":"https://doi.org/10.1136/bmjdh-2026-000071","source":"openalex"},{"id":"oa:W7128720054","type":"article-journal","title":"Preoperative localization of pulmonary nodules using ultra-low-dose CT based on artificial intelligence iterative reconstruction","abstract":"Background: Preoperative localization of pulmonary nodules requires multiple computed tomography (CT) scans, making it an urgent problem to address how to effectively reduce radiation damage while maintaining image quality. Artificial intelligence iterative reconstruction (AIIR) can significantly improve the image quality of ultra-low-dose CT (ULDCT). This study aimed to examine the feasibility of using ULDCT-AIIR for the preoperative localization of pulmonary nodules. Methods: This prospective study enrolled 40 consecutive patients with pulmonary nodules who underwent preoperative hook-wire localization under low-dose CT (LDCT). Immediately following the LDCT, an additional ULDCT scan was performed. Images were reconstructed using filtered back projection (FBP) and a hybrid iterative reconstruction (HIR) for both LDCT and ULDCT scans; additionally, AIIR was applied solely to the ULDCT images. Objective parameters measured included image noise, contrast-to-noise ratio (CNR), and the distances between nodules and reference. Subjective image quality was assessed using a 5-point Likert scale, evaluating the visualization of pulmonary nodules, localization grids, needle tips, hook-wires, and complications. Quantitative and qualitative metrics were compared across the reconstruction groups using the Kruskal-Wallis test. Results: 0.01). The subjective visualization scores for nodules, localization grids, needle tips, hook-wires, and complications with ULDCT-AIIR were non-inferior to those with LDCT-HIR and significantly superior to most other reconstruction groups (P<0.01). Distance measurements demonstrated no significant differences between ULDCT-AIIR and other reconstruction methods (P>0.05). Conclusions: ULDCT-AIIR achieves image quality comparable to LDCT-HIR with significantly reduced radiation doses, suggesting its potential as an alternative to LDCT for preoperative pulmonary nodule localization.","author":[{"family":"Lan","given":"Huixiang"},{"family":"Liu","given":"Xiaobin"},{"family":"Ou","given":"Danlin"},{"family":"Zhong","given":"Sihua"},{"family":"Zhong","given":"Hongcheng"},{"family":"Liang","given":"Mingzhu"}],"issued":{"date-parts":[[2026]]},"DOI":"10.21037/qims-2025-1544","URL":"https://doi.org/10.21037/qims-2025-1544","source":"openalex"},{"id":"oa:W7125480217","type":"article-journal","title":"Impact of artificial intelligence on empowering the future of nursing professionalism, educational and clinical advancements: an umbrella review on AI-driven transformation","abstract":"Artificial Intelligence (AI) is rapidly transforming healthcare by augmenting clinical decision-making, streamlining workflows, and personalizing education and patient care. Nursing, as the largest healthcare workforce, stands at the forefront of this transformation. This review examines how AI-driven tools empower nursing professionalism, enhance educational models, and optimize clinical practice. A systematic umbrella concept analysis was conducted using PubMed, Scopus, CINAHL, and Web of Science databases. Literature published between 2010 and 2025 was reviewed. Eligible studies included original research, reviews, policy reports, and frameworks focusing on AI applications in nursing education, practice, and professional development. Data were synthesized thematically under three domains: professional identity, educational innovation, and clinical advancement. Sixty-five studies met inclusion criteria. Evidence suggests that AI supports professional autonomy through clinical decision support systems, predictive analytics, and digital documentation, reducing administrative burdens. In education, AI-enabled simulations, adaptive learning platforms, and virtual mentors enhance critical thinking and competency development. Clinically, AI improves patient monitoring, diagnostic accuracy, and personalized care delivery. However, ethical dilemmas, data privacy risks, and limited digital literacy remain significant barriers. AI offers transformative potential for strengthening nursing professionalism, integrating evidence-based education, and advancing patient-centered clinical practice. To harness these opportunities, investment in nurse-centered AI training, interdisciplinary collaboration, and policy frameworks is essential. Nursing must embrace AI as a partner technology to redefine future roles and leadership in digital healthcare ecosystems.","author":[{"family":"Umar","given":"Mohammed"},{"family":"Kalyani","given":"B"},{"family":"Upreti","given":"Ms"},{"family":"Saini","given":"Pooja"},{"family":"Tamang","given":"Reshma"},{"family":"Geetha","given":"Paramasivam"}],"issued":{"date-parts":[[2026]]},"DOI":"10.18203/2349-3259.ijct20260052","URL":"https://doi.org/10.18203/2349-3259.ijct20260052","source":"openalex"},{"id":"oa:W4412715930","type":"article-journal","title":"Application Areas of Computer Vision and AI in Intelligent Automation Systems","abstract":"The combination of artificial intelligence (AI) and computer vision (CV) has led to tremendous growth in industries such as manufacturing, healthcare, agriculture, transportation, and sports. BMW applications show how robots guided by vision can solve 97% of problems accurately, so product quality control drives it forward. AI-based medical imaging is expected to have a significant impact on healthcare. These technologies have led to streamlined assembly lines, and integrated analytics in manufacturing processes to increase efficiency and reduce costs. In healthcare, AI-based medical imaging is expected to play a major role in accurate diagnosis and personalized patient care. With the global market estimated to reach $45.13 billion by 2027, the safe, autonomous vehicle market will reach about $556.67 billion by 2026. Moreover, the impact of AI and CV technology is expected to affect the sports analytics market, which is estimated to reach $4.62 billion by 2025. Essentially, AI and CV are enabling intelligent automation and restructuring of businesses around the world.","author":[{"family":"Kumar","given":"Vinod"},{"family":"Prabha","given":"Chander"},{"family":"Singh","given":"Ajay"},{"family":"Kumar","given":"Raj"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/9781394302734.ch15","URL":"https://doi.org/10.1002/9781394302734.ch15","source":"openalex"},{"id":"oa:W7138843766","type":"article-journal","title":"Artificial intelligence in rehabilitation: a review of clinical effectiveness, real-world performance, safety, and equity across modalities and settings","abstract":"Background: Rehabilitation faces a scale problem: millions who could benefit lack timely, effective services. Artificial intelligence (AI) and device-based modalities (e.g., robotics and VR) can extend reach and personalise care when validated, yet decision-makers lack a consolidated view of clinical usefulness, translation to practice, safety, equity, and cost. Methods: We conducted an umbrella review of reviews using a Population-Exposure-Outcome framework. Searches span biomedical, allied health, and engineering databases from inception to September 1, 2025. We distinguished AI-enabled (ML/DL) interventions from technology-assisted (no ML demonstrated) modalities and synthesised outcomes across impairment, activity, independence, usability/safety, equity, and economics. Findings: The most reproducible clinical signal is activity improvement for post-stroke upper limb with technology-assisted training (robotics with or without VR) that increases task-specific practice; effects on impairment and independence are inconsistent once dose is matched and assessors are blinded. Claims of non-inferiority are not established when prespecified margins and confidence-interval testing are absent, so parity is interpreted as no between-group advantage under those conditions. Across AI-enabled domains, a development-to-deployment performance drop is evident most notably for brain-computer-interface classifiers and computer-vision movement evaluation limiting immediate clinical impact. Imaging-based decision support (radiomics/CNN) is closer to practice but varies by software and site, requiring local calibration and impact evaluation before pathway change. Reported adverse events are generally mild, yet usability, adherence, equity, and cost are under-measured, particularly in home and hybrid delivery. Prediction-model and trial reporting frequently fall short of contemporary AI standards; representation skews toward high-income settings, and subgroup performance is seldom reported. Conclusion: An adjunct-first posture is warranted. Adoption should be gated by minimum clinically important difference-anchored benefit under dose symmetry and blinded assessment; external, multi-site validation with declared lab-to-clinic performance loss; subgroup fairness with mitigation; decision-grade economic value; interoperability; and readiness for regulation, change control, and cybersecurity. Priorities include pragmatic, multi-site, assessor-blinded, dose-matched trials; standardised safety/usability capture for home use; and a public, living evidence atlas. AI can expand rehabilitation when held to clinical standards that matter to patients and services. With clear adoption gates and continuous post-market monitoring, systems can extend access and independence without sacrificing rigour, safety, equity, or fairness.","author":[{"family":"Abdalla","given":"Nafisa"},{"family":"Arab","given":"Rabie"},{"family":"Abdrbo","given":"Amany"},{"family":"Almari","given":"Mohammad"},{"family":"Ayoub","given":"Mohammed"},{"family":"Alsaaideh","given":"Bilal"},{"family":"Dagamseh","given":"Mohammad"},{"family":"Almagharbeh","given":"Wesam"},{"family":"Abuadas","given":"Fuad"},{"family":"Mahfouz","given":"Mohammad"},{"family":"Gaballah","given":"Mastoura"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/fdgth.2026.1737957","URL":"https://doi.org/10.3389/fdgth.2026.1737957","source":"openalex"},{"id":"oa:W7117648215","type":"article-journal","title":"Physics-informed artificial intelligence with splines for modeling advection–diffusion–reaction under dynamic boundaries","abstract":"This paper proposes Physics-Informed Deep Operator Networks (PI-DeepONets) for rapid simulation of Advection–Diffusion–Reaction (ADR) systems with time-variable boundary conditions. These are viewed as inputs of the system and each family of inputs can be represented by a set of parameters. It is shown that, in practice, PI-DeepONets may not be able to model accurately the system of Partial Differential Equations (PDEs) for any combination of those parameterized inputs. Therefore, a new distributed architecture is proposed, which combines specialized PI-DeepONets, each being dedicated to represent one PDE. Furthermore, a mixture of experts allows to simulate multiple PI-DeepONets for which the input is a combination of several parameterized base functions. A model of an adsorption column is used to evaluate the relevance of the method. The Mixture of Experts PI-DeepONets with complex inputs such as splines, reaches a Mean Arctangent Absolute Percentage Error around 0.2 comparable to that of physics informed neural networks with constant inputs. The accuracy is improved by 20 % compared to distributed DeepONets, whereas a conventional DeepONet is unable to converge to the true solution. The results highlight the ability of a mixture of experts PI-DepeOnets to deliver an accurate and fast simulations of chemical reactors described by partial differential equations under complex and time-variable boundary conditions. The simulations can be completed in less than 10 ms compared to several minutes for traditional simulations methods, therefore opening the door to real time predictions and optimization.","author":[{"family":"Belmonte","given":"Romain"},{"family":"Dieulot","given":"Jean‐yves"},{"family":"Galanti","given":"Mattia"},{"family":"Annaland","given":"Martin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.dte.2025.100083","URL":"https://doi.org/10.1016/j.dte.2025.100083","source":"openalex"},{"id":"oa:W7167360095","type":"article-journal","title":"Nurses’ knowledge, attitudes, and perceived challenges toward artificial intelligence applications in patient care: a descriptive-analytical cross-sectional study","abstract":"BACKGROUND: Artificial intelligence (AI) is increasingly being integrated into healthcare systems; however, nurses' knowledge, attitudes, and perceived challenges play a crucial role in its adoption in patient care. This study aimed to assess nurses' knowledge, attitudes, and perceived challenges toward AI, examine the relationships among these variables, and explore their associations with demographic characteristics and prior AI training. METHODS: A descriptive analytical cross-sectional study was conducted among 107 nurses working in intensive care, medical, and surgical units at Zagazig University Hospital. A purposive sampling technique was used. Data were collected over two months using structured instruments during morning shifts. RESULTS: Most participants were aged 25-34 years (55.1%), male (65.4%; reflecting the accessible sample composition), and held bachelor's degrees (68.2%), with nearly half (49.5%) having 5-10 years of clinical experience. Overall, 68.2% of nurses achieved satisfactory knowledge scores, whereas 88.8% demonstrated positive attitudes toward AI applications. Perceived challenges were mainly related to technical and ethical concerns, particularly the need for continuous system updates, cybersecurity risks, and implementation costs. A statistically significant weak negative correlation was found between nurses' knowledge and attitudes toward AI (r = -0.195, p = 0.044). No significant correlations were observed between knowledge and perceived challenges (r = -0.162, p = 0.095) or between attitudes and perceived challenges (r = 0.142, p = 0.145). Previous AI-related training was significantly associated with more positive attitudes toward AI (p = 0.019), whereas no significant associations were found with knowledge or perceived challenges. Educational level, workplace, and years of experience were not significantly associated with nurses' knowledge, attitudes, or perceived challenges. CONCLUSION: Nurses demonstrated a satisfactory knowledge and generally positive attitudes toward AI applications in patient care, while perceiving moderate implementation challenges. Although previous AI-related training was associated with more positive attitudes, no significant associations were found with knowledge or perceived challenges. The weak negative correlation between knowledge and attitudes suggests that greater awareness of AI may be accompanied by increased concerns regarding its use. Further educational initiatives are needed to enhance nurses' preparedness for AI integration in clinical practice.","author":[{"family":"Elsayed","given":"Rehab"},{"family":"Elsayed","given":"Rehab"},{"family":"Nagy","given":"Ahmed"},{"family":"Hussein","given":"Eman"},{"family":"Elsayed","given":"Reham"},{"family":"Elsayed","given":"Reham"},{"family":"Ramadan","given":"Reda"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1186/s12912-026-04966-5","URL":"https://doi.org/10.1186/s12912-026-04966-5","source":"openalex"},{"id":"oa:W7124242635","type":"article-journal","title":"Examining the Impact of Artificial Intelligence Technology on Sustainable Development in Highway Maintenance Industry: A Structural Equation Modelling Approach","abstract":"In the field of the road transportation industry, quantitative research on the relationship between artificial intelligence (AI) technology and corporate sustainable development is relatively scarce. This disparity has led to discussions about whether artificial intelligence technology can truly promote the sustainable development level of the highway maintenance industry. Therefore, this study aims to quantify the relationship between artificial intelligence technology and the sustainable development of the highway maintenance industry, and to analyze the reasons behind the current controversies. The research results show: (1) Each exogenous variable has an impact on sustainable development, although the degree of influence varies, especially the economic development level (ED) has the strongest direct effect on sustainable development, followed by the level of market demand (MD), the level of policy support (PS), and the level of enterprise capital (EC); (2) Moderating variables can enhance this direct impact, among which the moderating effect of ED on the relationship between ED and sustainable development is the strongest; (3) Artificial intelligence technology has different impacts on enterprises at different positions in the industrial chain, thereby explaining the controversy over whether to adopt it or not. These conclusions highlight the value of artificial intelligence technology and provide a reasonable explanation for the existing controversies in the industry and research field.","author":[{"family":"Zhou","given":"Jizhao"},{"family":"Wang","given":"Chenyang"},{"family":"Guo","given":"Jin"},{"family":"Qin","given":"Peng"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/su18020889","URL":"https://doi.org/10.3390/su18020889","source":"openalex"},{"id":"oa:W7133358115","type":"article-journal","title":"Generative artificial intelligence in optimizing the quality of cancer care: potential, limitations, and future directions of development for Large Language Models. A narrative literature review","abstract":"This literature review included scientific articles, published between 2022 and 2025, indexed in PubMed, Scopus, and Proquest. The article investigates the potential of generative artificial intelligence (GenAI), particularly Large Language Models (LLMs), to improve the quality of cancer care. LLMs have demonstrated effectiveness in patient education by simplifying complex medical terminology and tailoring content to the user’s level of understanding. LLMs also assist physicians in clinical decision-making by analyzing medical data and supporting adherence to the latest guidelines. However, expert oversight is still necessary due to the risk of error. For cancer prevention, LLMs promote healthy lifestyle adoption, participation in screening programs, and vaccination. They also play an important role in reducing inequities in access to information. Another key feature of LLMs is their ability to translate complex diagnostic reports into patient-friendly language. LLMs have also shown promise in counteracting cancer-related misinformation. However, the article identifies certain LLMs limitations, such as model hallucination, incomplete personalization, and unresolved legal liability concerns. The article emphasizes ethical dilemmas, particularly those related to patient autonomy and the risk of dehumanizing care. For future progress, the article emphasizes the need to integrate LLMs into e-health systems and to develop specialized models supported by interdisciplinary teams.","author":[{"family":"Gotlib-Małkowska","given":"Joanna"},{"family":"Włodarczyk","given":"Kinga"},{"family":"Koczkodaj","given":"Paweł"},{"family":"Hreńczuk","given":"Marta"},{"family":"Nowakowski","given":"Adrian"},{"family":"Pańczyk","given":"Mariusz"},{"family":"Cieślak","given":"Ilona"}],"issued":{"date-parts":[[2026]]},"DOI":"10.29316/hpc/218352","URL":"https://doi.org/10.29316/hpc/218352","source":"openalex"},{"id":"oa:W4414110320","type":"article-journal","title":"Printed sensing human-machine interface with individualized adaptive machine learning","abstract":"Developing intelligent robots with integrated sensing capabilities is critical for advanced manufacturing, medical robots, and embodied intelligence. Existing robotic sensing technologies are limited to recording of acceleration, driving torque, pressure feedback, and so on. Expanding and integrating with the multimodal sensors to mimic and even surpass the human feeling is substantially underdeveloped. Here, we introduce a printed soft human-machine interface consisting of an e-skin-enabled gesture recognitions with feedback stimulus and a soft robot with multimodal perception of contact pressure, temperature, thermal conductivity, and electrical conductivity. The sensing e-skin with adaptive machine learning was able to decode and classify the hand gestures with re-wearable convenience and individual's differences. The soft interface provides the bidirectional communications between robotics and human bodies in the close-loop. This work could substantially extend the robotic intelligence and pave the way for more practical applications.","author":[{"family":"Wang","given":"Guohui"},{"family":"Tang","given":"Yao"},{"family":"Luo","given":"Xinran"},{"family":"Lu","given":"Shengdi"},{"family":"Zhou","given":"Yiru"},{"family":"Lu","given":"Yi"},{"family":"Sun","given":"Guangyang"},{"family":"Liu","given":"Pei"},{"family":"Ning","given":"Jiayu"},{"family":"Jiang","given":"Hua"},{"family":"Hu","given":"Ke"},{"family":"Liu","given":"Hongzhen"},{"family":"Song","given":"Wenqi"},{"family":"Yu","given":"You"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1126/sciadv.adw3725","URL":"https://doi.org/10.1126/sciadv.adw3725","source":"openalex"},{"id":"oa:W4417488987","type":"article-journal","title":"Advanced 3D Modeling and Bioprinting of Human Anatomical Structures: A Novel Approach for Medical Education Enhancement","abstract":"Current challenges in anatomical teachings, such as cadaver shortages, ethical limitations, and restricted access to pathological specimens, are increasingly being mitigated by advancing medical technologies, and among these are three-dimensional modeling technology and multi-material bioprinting. These innovations could facilitate a deeper understanding of complex anatomical components while encouraging an interactive learning environment that accommodates diverse educational needs. These technologies have the capacity to transform anatomy education, yielding better-prepared healthcare practitioners. Combining artificial intelligence with acquired medical images makes it easier to reconstruct anatomy and saves time while still being very accurate. This review seeks to thoroughly assess the current landscape of advanced three-dimensional printing, multi-material bioprinting, and related technologies used in anatomical education. It aims to consolidate evidence related to their educational effectiveness and to outline potential pathways for clinical applications and research development.","author":[{"family":"Castorina","given":"Sergio"},{"family":"Puleo","given":"Stefano"},{"family":"Crescimanno","given":"Caterina"},{"family":"Pezzino","given":"Salvatore"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/app16010005","URL":"https://doi.org/10.3390/app16010005","source":"openalex"},{"id":"oa:W7143486389","type":"article-journal","title":"Artificial intelligence anxiety, digital well-being, and future career concerns among engineering and information technology students in Jordan","abstract":"Introduction The rapid advancement of artificial intelligence (AI) is fundamentally transforming educational and employment landscapes, generating increasing psychological concerns among students in technology-intensive fields. This study examines AI-related anxiety, digital well-being, and career uncertainty among engineering and information technology (IT) students, with a focus on their prevalence, interrelationships, and demographic variations. Methods A cross-sectional quantitative design was employed using a structured survey administered to 820 undergraduate students from four Jordanian universities. Standardized measures were used to assess AI anxiety, digital well-being, and career-related concerns. Statistical analyses included descriptive statistics, correlation analysis, and group comparisons based on gender and academic discipline. Results The findings indicated elevated levels of AI anxiety ( M = 5.26, SD = 0.32), low levels of digital well-being ( M = 1.75, SD = 0.20), and moderate levels of career concerns ( M = 4.07, SD = 0.34). AI anxiety was strongly negatively correlated with digital well-being ( r = −0.849, p < 0.01) and positively correlated with career concerns ( r = 0.680, p < 0.01). Female students reported significantly higher AI anxiety and career concerns than male students ( p < 0.001). Additionally, IT students exhibited higher levels of AI anxiety and career uncertainty compared to engineering students ( p < 0.001). Discussion These findings highlight the psychological impact of AI integration on students, emphasizing the need for targeted AI literacy programs, digital well-being interventions, and career guidance strategies. Addressing gender disparities and discipline-specific differences is essential to enhance students’ resilience, adaptability, and readiness for an AI-driven labor market.","author":[{"family":"Al-Nasah","given":"Mais"},{"family":"Al-Tarawneh","given":"Luae"},{"family":"Alhwayan","given":"Ola"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/frai.2026.1598741","URL":"https://doi.org/10.3389/frai.2026.1598741","source":"openalex"},{"id":"oa:W7116896729","type":"article-journal","title":"Understanding implementation science in medical radiation sciences","abstract":"OBJECTIVES: Radiography, like many allied health professions, faces persistent challenges in translating evidence and innovation into routine clinical practice. Despite a strong foundation in evidence-based practice, the adoption of new technologies, protocols, and models of care is often inconsistent, delayed, or un-sustained. This paper introduces Implementation science, which offers a key, yet underutilised approach for advancing radiographic practice by focusing on how evidence-based interventions are adopted, integrated, and sustained in real-world settings. KEY FINDINGS: We present a conceptual overview of implementation science frameworks with particular relevance to radiography e.g. technology, devices and service improvement. Key frameworks considered include the Consolidated Framework for Implementation Research (CFIR), Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM), the Non-adoption, Abandonment, Scale-up, Spread, and Sustainability (NASSS) framework, the Theoretical Domains Framework (TDF), and Normalisation Process Theory (NPT). Each is appraised for its focus, strengths, and applicability to common implementation challenges in radiography. The frameworks highlight different but complementary perspectives, for example CFIR and TDF emphasise multilevel determinants of behaviour and RE-AIM structures evaluation of implementation outcomes. Applied examples from radiography and allied health illustrate how these approaches can be used to diagnose barriers, design strategies, and evaluate implementation efforts. CONCLUSION: Implementation science provides a rich methodological and theoretical toolkit for strengthening radiography research. By applying these frameworks, studies can move beyond questions of clinical efficacy to address the practical realities of translation, adoption and sustainability. IMPLICATIONS FOR PRACTICE: Embedding implementation science within radiographic research, practice, and education can support more systematic and context-sensitive translation. This shift enables the profession to progress from demonstrating clinical potential to delivering sustained improvements in service delivery, patient safety, and professional practice.","author":[{"family":"Manning","given":"F"},{"family":"Hancock","given":"A"},{"family":"Meertens","given":"R"},{"family":"Ede","given":"J"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.radi.2025.103288","URL":"https://doi.org/10.1016/j.radi.2025.103288","source":"openalex"},{"id":"oa:W4415993116","type":"article-journal","title":"Leveraging imperfection with MEDLEY: a multi-model approach harnessing bias in medical AI","abstract":"Bias in medical artificial intelligence is conventionally viewed as a defect that requires elimination. However, human reasoning inherently incorporates biases shaped by education, culture, and experience, suggesting their presence may be inevitable and potentially valuable. We propose MEDLEY (Medical Ensemble Diagnostic system with Leveraged diversitY), a conceptual framework that orchestrates multiple AI models while preserving their diverse outputs rather than collapsing them into a consensus. Unlike traditional approaches that suppress disagreement, MEDLEY documents model-specific biases as potential strengths and treats hallucinations as provisional hypotheses for clinician verification. A proof-of-concept demonstrator for differential diagnosis was developed using over 30 large language models, preserving both consensus and minority views, rendering diagnostic uncertainty and latent biases transparent to support clinical oversight. While not yet a validated clinical tool, the demonstration illustrates how structured diversity can enhance medical reasoning under the supervision of clinicians. By reframing AI imperfection as a resource, MEDLEY offers a paradigm shift that opens new regulatory, ethical, and innovation pathways for developing trustworthy medical AI systems.","author":[{"family":"Abtahi","given":"Farhad"},{"family":"Astaraki","given":"Mehdi"},{"family":"Seoane","given":"Fernando"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/frai.2026.1701665","URL":"https://doi.org/10.3389/frai.2026.1701665","source":"openalex"},{"id":"oa:W7117486807","type":"article-journal","title":"Cyber–Physical Systems in Healthcare Based on Medical and Social Research Reflected in AI-Based Digital Twins of Patients","abstract":"Cyber–physical systems (CPS) in healthcare represent a deep integration of computational intelligence, physical medical devices, and human-centric data, enabling continuous, adaptive, and personalized care. These systems combine real-time measurements, artificial intelligence (AI)-based analytics, and networked medical devices to monitor, predict, and optimize patient health outcomes. A key development in the field of CPS is the emergence of patient digital twins (DTs), virtual models of individual patients that simulate biological, behavioral, and social parameters. Using AI, DTs analyze complex medical and social data (genetics, lifestyle, environment, etc.) to support precise diagnosis and treatment planning. The implications of the bibliometric findings suggest that the field emerges from the conceptual phase, justifying the article’s emphasis on both the proposed architectures and their clinical validation. However, most research was conducted in computer science, engineering, and mathematics, rather than medicine and healthcare, suggesting an early stage of technological maturity. Leading countries were India, the United States, and China, but these countries did not have a high number of publications, nor did they record leading researchers or affiliations, suggesting significant research fragmentation. The most frequently observed Sustainable Development Goals indicate an industrial context. Reflecting insights from medical and social research, AI-based DT systems provide a holistic view of the patient, taking into account not only physiological states but also psychological and social well-being. These systems promote personalized therapy by dynamically adapting treatment based on real-time feedback from wearable sensors and electronic medical records. More broadly, CPS and DT systems increase healthcare system efficiency by reducing hospitalizations and supporting remote preventive care. Their implementation poses significant ethical and privacy challenges, particularly regarding data ownership, algorithm transparency, and patient autonomy.","author":[{"family":"Mikołajewska","given":"Emilia"},{"family":"Rogalla-Ładniak","given":"Urszula"},{"family":"Masiak","given":"Jolanta"},{"family":"Panas","given":"Ewelina"},{"family":"Mikołajewski","given":"Dariusz"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/app16010318","URL":"https://doi.org/10.3390/app16010318","source":"openalex"},{"id":"oa:W7125688390","type":"article-journal","title":"The Carbon Cost of Intelligence: A Domain-Specific Framework for Measuring AI Energy and Emissions","abstract":"The accelerating energy demands from artificial intelligence (AI) deployment introduce systemic challenges for achieving carbon neutrality. Large language models (LLMs) represent a dominant driver of AI energy consumption, with inference operations constituting 80–90% of total energy usage. Current energy benchmarks report aggregate metrics without domain-level breakdowns, preventing accurate carbon footprint estimation for workloadspecific operations. This study addresses this critical gap by introducing a carbon-aware framework centered on the carbon cost of intelligence (CCI), a novel metric enabling workload-specific energy and carbon calculation that balances accuracy and efficiency across heterogeneous domains. This paper presents a comprehensive cross-domain energy benchmark using the massive multitask language understanding (MMLU) dataset, measuring accuracy and energy consumption in five representative domains: clinical knowledge (medicine), professional accounting (finance), professional law (legal), college computer science (technology), and general knowledge. Empirical analysis of GPT-4 across 100 MMLU questions, 20 per domain, reveals substantive variations: legal queries consume 4.3× more energy than general knowledge queries (222 J vs. 52 J per query), while energy consumption varies by domain due to input length differences. Our analysis demonstrates the evolution from simple ratio-based approaches (weighted accuracy divided by weighted energy) to harmonic mean aggregation, showing that the harmonic mean, by preventing bias from extreme values, provides more accurate carbon usage estimates. The CCI metric, calculated using weighted harmonic mean (analogous to P/E ratios in finance, where A/E represents accuracy-to-energy ratio), enables practitioners to accurately estimate energy and carbon emissions for specific workload mixes (e.g., 80% medicine + 15% general + 5% law). Results demonstrate that the domain workload mix significantly impacts carbon footprint: a law firm workload (60% law) consumes 96% more energy per query than a hospital workload (80% medicine), representing 49% potential savings through workload optimization. Carbon footprint analysis using US Northeast grid intensity (320 gCO2e/kWh) shows domain-specific emissions ranging from 0.0046–0.0197 gCO2 per query. CCI is validated through comparison with simple weighted average, demonstrating differences up to 12.1%, confirming that the harmonic mean provides more accurate and conservative carbon estimates essential for carbon reporting and neutrality planning. Our findings provide a novel cross-domain energy benchmark for GPT-4 and establish a practical carbon calculator framework for sustainable AI deployment aligned with carbon neutrality goals.","author":[{"family":"Kaur","given":"Rashanjot"},{"family":"Kundu","given":"Triparna"},{"family":"Park","given":"Kathleen"},{"family":"Pinsky","given":"Eugene"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/en19030642","URL":"https://doi.org/10.3390/en19030642","source":"openalex"},{"id":"oa:W4406272821","type":"article-journal","title":"Integrating Model‐Informed Drug Development With AI : A Synergistic Approach to Accelerating Pharmaceutical Innovation","abstract":"The pharmaceutical industry constantly strives to improve drug development processes to reduce costs, increase efficiencies, and enhance therapeutic outcomes for patients. Model-Informed Drug Development (MIDD) uses mathematical models to simulate intricate processes involved in drug absorption, distribution, metabolism, and excretion, as well as pharmacokinetics and pharmacodynamics. Artificial intelligence (AI), encompassing techniques such as machine learning, deep learning, and Generative AI, offers powerful tools and algorithms to efficiently identify meaningful patterns, correlations, and drug-target interactions from big data, enabling more accurate predictions and novel hypothesis generation. The union of MIDD with AI enables pharmaceutical researchers to optimize drug candidate selection, dosage regimens, and treatment strategies through virtual trials to help derisk drug candidates. However, several challenges, including the availability of relevant, labeled, high-quality datasets, data privacy concerns, model interpretability, and algorithmic bias, must be carefully managed. Standardization of model architectures, data formats, and validation processes is imperative to ensure reliable and reproducible results. Moreover, regulatory agencies have recognized the need to adapt their guidelines to evaluate recommendations from AI-enhanced MIDD methods. In conclusion, integrating model-driven drug development with AI offers a transformative paradigm for pharmaceutical innovation. By integrating the predictive power of computational models and the data-driven insights of AI, the synergy between these approaches has the potential to accelerate drug discovery, optimize treatment strategies, and usher in a new era of personalized medicine, benefiting patients, researchers, and the pharmaceutical industry as a whole.","author":[{"family":"Raman","given":"Karthik"},{"family":"Kumar","given":"Rukmini"},{"family":"Musante","given":"Cynthia"},{"family":"Madhavan","given":"Subha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/cts.70124","URL":"https://doi.org/10.1111/cts.70124","source":"openalex"},{"id":"oa:W7154861292","type":"article-journal","title":"Mapping and Quality Appraisal of Artificial Intelligence Preferential Reporting Checklists, Items, Guidelines, and Consensus in Healthcare: An Altmetric, Bibliometric, and Systematic Review","abstract":"Introduction: The integration of artificial intelligence (AI) in healthcare has garnered significant scholarly attention, particularly in areas such as medical image analysis, prognosis, and treatment. Despite its potential, concerns regarding AI's reliability and application persist, prompting the development of guidelines aimed at standardizing its use in medicine. This study aims to evaluate the current content and quality of AI guidelines in healthcare, focusing on identifying gaps and providing a critical appraisal of existing checklists. Methodology: Comprehensive bibliometric analysis, Altmetric analysis, and systematic review were conducted, utilizing the AGREE II Tool for quality appraisal. The systematic search spanned Scopus, PubMed, and Dimension AI databases, focusing on English-language, open-access articles related to AI reporting guidelines. Two reviewers independently evaluated the data, with manual extraction performed in Microsoft Excel. The AGREE II Tool assessed six domains of guideline quality. Results: The search yielded 2477 articles, ultimately identifying 27 AI-specific reporting guidelines published between 2020 and 2025. The analysis revealed significant variations in quality across the AGREE II domains. Among the very first and most impactful were SPIRIT-AI, CONSORT-AI, MINIMAR, and CLAIM, all prioritized structured reporting but were hampered by their timing-based when there were few AI trials-resulting in limited applicability and risk of missing older AI terminologies. However, STAR-machine learning (ML), APPRAISE-AI, and CLEAR exhibited more general, domain-specific frameworks, whereas checklists such as CHEERS-AI, CREMLS, and MI-CLEAR-large language model (LLM) showcased limited author diversity in contribution. However, many guidelines exhibited weaknesses in methodological rigor and stakeholder involvement, limiting their practical applicability. Conclusion: The findings emphasize the need for evidence-based updates to AI reporting guidelines to ensure methodological integrity amid rapid advancements. Increased expert involvement and stakeholder engagement are crucial for enhancing the guidelines' applicability and rigor, addressing AI complexities in health research, and adapting reporting frameworks to evolving AI technologies in healthcare.","author":[{"family":"Vinay","given":"Vineet"},{"family":"Jodalli","given":"Praveen"},{"family":"Chavan","given":"Mahesh"},{"family":"Satyarup","given":"Dharmashree"},{"family":"Bhor","given":"Ketaki"},{"family":"Buddhikot","given":"Chaitanya"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1155/ijod/6730710","URL":"https://doi.org/10.1155/ijod/6730710","source":"openalex"},{"id":"oa:W7154735908","type":"article-journal","title":"Global English-language-dominated discourse on artificial intelligence in healthcare: a three-year longitudinal analysis of the #AIinHealthcare movement on X","abstract":"Background: Social media platforms facilitate global discourse on the application of artificial intelligence (AI) in healthcare. Nevertheless, there is a paucity of longitudinal analyses of digitally mediated discussions. Objective: To investigate the evolution of global English-language-dominated discourse on #AIinHealthcare over a three-year period on X (formerly Twitter). Methods: Using Fedica analytics, we analysed 57,880 tweets by 17,991 distinct users across 141 countries from 1 November 2022 to 1 November 2025. This analysis focused on English-language-dominant discourse around #AIinHealthcare (96.9% English), acknowledging hashtag-specific selection bias and linguistic limitations. This study used publicly available anonymised data and followed the ethical guidelines for social media research. Results: The #AIinHealthcare garnered 39.2 million impressions, with significant contributions from high-income countries, notably the United States (40.7%) and Canada (21.0%), as well as India (13.4%; a rapidly expanding economy), collectively accounting for 75.1% of tweets and reflecting hashtag-specific, geographically concentrated engagement. This peaked in mid-2023 and stabilized lower by mid-2025. English was the predominant language of the discourse (96.9%). The community consisted of 74.9% grassroots users with fewer than 1,000 followers, suggesting genuine participation beyond elite influencers. Total engagement reached 72,625 interactions, primarily passive, comprising 68.1% likes, 19.4% retweets, 10.3% replies, and 2.1% quote tweets. Hashtag co-occurrence patterns, supported by qualitative inspection of exemplar tweets, indicated majorly five distinct clusters: foundational technical topics (#GenerativeAI, #ChatGPT, #LLMs) peaked after November 2022; clinical application themes emerged across disease-specific specialties (#Oncology, #Cardiology, #MentalHealth); healthcare implementation themes addressed practical integration (#DigitalHealth, #Telemedicine, #EHR); governance and ethics themes gained prominence (#ResponsibleAI, #AIEthics, #ExplainableAI, #DataPrivacy); and professional integration themes fostered learning communities (#MedTwitter, #MedicalEducation). Sentiment was predominantly neutral (95%), with positive (3%) and negative (2%). Monthly tweets peaked in mid-2023 at 1,600-1,800 before declining to 750-900 per month by June 2025. High-engagement content linked AI to practical applications, governmental initiatives, and clinical breakthroughs. Conclusion: English-language-dominated discourse around #AIinHealthcare reveals hashtag-specific maturation from technical enthusiasm to governance and implementation focus. However, platform access restrictions in countries such as China and Russia may skew geographic representation. Disparities in sustainability discourse remain prevalent.","author":[{"family":"Wochelethoma","given":"Thomas"},{"family":"Ijinu","given":"Thadiyan"},{"family":"Sasidharan","given":"Sreejith"},{"family":"Manakkadan","given":"Anoop"},{"family":"Shine","given":"Lathikakumariamma"},{"family":"Krishnakumar","given":"Neenthamadathil"},{"family":"Aruna","given":"Selvaraj"},{"family":"Pasupuleti","given":"Nagarjuna"},{"family":"Aswany","given":"Thomas"},{"family":"Deepthi","given":"Divakaran"},{"family":"Ma","given":"Zilin"},{"family":"Hua","given":"Yining"},{"family":"Ławiński","given":"Michał"},{"family":"Litvinova","given":"Olena"},{"family":"Kletečka-Pulker","given":"Maria"},{"family":"Atanasov","given":"Atanas"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/fdgth.2026.1795488","URL":"https://doi.org/10.3389/fdgth.2026.1795488","source":"openalex"},{"id":"oa:W4417272080","type":"article-journal","title":"Antimicrobial use and resistance","abstract":"Antimicrobial resistance affects the delivery of safe and effective healthcare. Antimicrobial resistance has attracted strong political focus, with the 2024 United Nations General Assembly high level meeting providing a clear commitment to reducing mortality and improving antibiotic use. This review summarises recent political action, policy prioritisation, and identification of future threats. It considers infections that are caused by drug resistant pathogens and reviews available and new antibiotics that may meet unmet medical needs. Despite increasing political engagement, the global antimicrobial resistance landscape remains imbalanced. In high income hospital settings, diagnostics, antimicrobial stewardship, and infection prevention and control are improving and may be further enabled by artificial intelligence and information systems. The development and use of new antibiotics is a major focus. By contrast, in low and middle income countries, access to most of these advances is limited. In all settings, empirical prescribing of essential antibiotics remains the cornerstone of treatment and conserving their efficacy is critical to effective healthcare. Targeted prevention and optimal treatment strategies are needed to mitigate antimicrobial resistance across all settings.","author":[{"family":"Reza","given":"Nada"},{"family":"Dubey","given":"Vineet"},{"family":"Sharland","given":"Michael"},{"family":"Hope","given":"William"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/bmj-2024-082681","URL":"https://doi.org/10.1136/bmj-2024-082681","source":"openalex"},{"id":"oa:W7153217607","type":"article-journal","title":"AI Use for Medical Students: Impact on Clinical Skill Acquisition and Retention. A Systematic Review","abstract":"Purpose: Artificial Intelligence (AI) is increasingly used in undergraduate medical education but has a potentially negative impact on clinical reasoning development. Specifically, the use of AI in medical student education may lead to deskilling and upskilling inhibition - where automation reduces practice or limits skill development - potentially impairing clinical reasoning. This systematic review aimed to synthesise evidence regarding AI-supported learning effects on acquisition and retention of clinical skills in medical students to assess its potential negative impact in medical education. Methods: A systematic search was conducted on 21 October 2025 across PubMed, Scopus, and Embase using structured Boolean queries restricted to titles and abstracts. Inclusion criteria targeted published studies involving medical students exposed to AI tools in clinical learning, reporting outcomes related to skill acquisition, reasoning, or overreliance on AI. Exclusions included non-AI digital tools, administrative AI applications, and studies without clear educational outcomes. Screening followed PRISMA guidelines. Results: From 420 records, 255 were screened. Four studies met the screening criteria, incorporating a total of 408 medical students. Across included studies, AI exposure was associated with improved efficiency and improved basic knowledge acquisition. When higher-order clinical reasoning and complex decision-making were assessed, findings were mixed: one study reported no overall difference, while others suggested weaker performance or reduced engagement when AI-supported approaches were used. Conclusion: Current evidence suggests that AI-supported learning may be associated with improved efficiency and basic knowledge acquisition in undergraduate medical education. Findings were less consistently supportive of higher-order reasoning outcomes compared with traditional teaching approaches, although the evidence base was limited. Potential risks of deskilling and upskilling inhibition warrant attention as medical schools increasingly integrate AI tools into their curricula. A striking finding of our systematic review was the very low number of existing studies identified in this important field. Further research should explore the long-term impacts of AI on medical students' independent clinical judgement and consider strategies to mitigate overreliance on AI given the profound potential impact on future patient care.","author":[{"family":"Turney","given":"Jonathan"},{"family":"Young","given":"Timothy"},{"family":"Chauhan","given":"Dhyana"},{"family":"Beeharry","given":"Roshni"},{"family":"Mahmud","given":"Mohammad"}],"issued":{"date-parts":[[2026]]},"DOI":"10.2147/amep.s583763","URL":"https://doi.org/10.2147/amep.s583763","source":"openalex"},{"id":"oa:W7154774280","type":"article-journal","title":"A Dataset for Evaluating Large Language Models on Chinese National Medical Licensing Examinations","abstract":"Large language models (LLMs) are increasingly applied in medical education, question answering, and clinical reasoning, yet standardized datasets in non-English contexts remain limited. To address this gap, we present CNMLEQA, a benchmark dataset for evaluating LLMs on the Chinese National Medical Licensing Examination. The dataset integrates question-answer pairs from three sources, including PubMed, GitHub, and MedExamLLM. CNMLEQA comprises two subsets: CNMLEQA-10k (9,890 questions) and CNMLEQA-3k (2,949 questions), each consisting of multiple-choice questions with five options and one correct answer. Questions are annotated with key dimensions including: (1) question type (knowledge-based or case-based), (2) auxiliary metadata such as examination year, 3) clinical scenario information across five dimensions: disease or diagnosis, surgery, medication, laboratory examination, and symptom or sign. Annotation was conducted by clinical experts. To validate the dataset, we evaluated state-of-the-art LLMs including Gemini, DeepSeek, GPT, Qwen, and LLaMA, and conducted fine-tuning experiments specifically on Qwen models. Results show that Qwen2.5-32B achieved the accuracy of 90.88% on CNMLEQA-10k, while DeepSeek-R1 achieved the accuracy of 91.59% on CNMLEQA-3k. The fine-tuning experiments further demonstrated significant performance improvements. CNMLEQA provides a multidimensional, clinically grounded benchmark for advancing LLM evaluation in Chinese medical applications.","author":[{"family":"Zong","given":"Hui"},{"family":"Cha","given":"Jiaxue"},{"family":"Wang","given":"Yi"},{"family":"Song","given":"Yu"},{"family":"Zhao","given":"Yan"},{"family":"Shi","given":"Muyun"},{"family":"Shen","given":"Bairong"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1038/s41597-026-07261-9","URL":"https://doi.org/10.1038/s41597-026-07261-9","source":"openalex"},{"id":"oa:W7155410582","type":"article-journal","title":"The effect of medical explanations from large language models on diagnostic accuracy in radiology","abstract":"Large language models (LLMs) are increasingly used by physicians for diagnostic support. A key advantage of LLMs is the ability to generate explanations that can help physicians understand the reasoning behind a diagnosis. However, the best-suited format for LLM-generated explanations remains unclear. In this large-scale study, we examined the effect of different formats for LLM explanations on clinical decision-making. For this, we conducted a randomized experiment with radiologists reviewing patient cases with radiological images (N = 2020 assessments). Participants received either no LLM support (control group) or were supported by one of three LLM-generated explanations: (1) a standard output providing the diagnosis without explanation; (2) a differential diagnosis comparing multiple possible diagnoses; or (3) a chain-of-thought explanation offering a detailed reasoning process for the diagnosis. We find that the format of explanations significantly influences diagnostic accuracy. The chain-of-thought explanations yielded the best performance, improving the diagnostic accuracy by 12.2% compared to the control condition without LLM support (P = 0.001). The chain-of-thought explanations are also superior to the standard output without explanation ( + 7.2%; P = 0.040) and the differential diagnosis format ( + 9.7%; P = 0.004). We further assessed the robustness of these findings across case difficulty and different physician backgrounds, such as general vs. specialized radiologists. Evidently, in the controlled setting of our vignette study, explaining the reasoning for a diagnosis helps physicians to identify and correct potential errors in LLM predictions and thus improve overall decisions. Altogether, the results highlight the importance of explanations in medical LLMs to support the reasoning processes of physicians, so that medical LLMs can improve diagnostic performance and, ultimately, patient outcomes.","author":[{"family":"Spitzer","given":"Philipp"},{"family":"Hendriks","given":"Daniel"},{"family":"Rudolph","given":"Jan"},{"family":"Schlaeger","given":"Sarah"},{"family":"Ricke","given":"Jens"},{"family":"Kühl","given":"Niklas"},{"family":"Hoppe","given":"Boj"},{"family":"Feuerriegel","given":"Stefan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1038/s41746-026-02619-0","URL":"https://doi.org/10.1038/s41746-026-02619-0","source":"openalex"},{"id":"oa:W7119539028","type":"article-journal","title":"Artificial intelligence application in the prediction of spontaneous preterm birth by cervical length in the first trimester of pregnancy: Comparison of three measurement methods","abstract":"OBJECTIVES: The current study evaluates the efficacy of artificial intelligence (AI)-assisted measurement of cervical length (CL) in predicting spontaneous preterm birth (sPTB), comparing the traditional single-line and two-line methods with the innovative AI-line method in the first trimester of pregnancy. MATERIALS AND METHODS: This study is a retrospective secondary analysis of ultrasound images collected prospectively from women with a viable singleton pregnancy who were undergoing Down syndrome screening at Prince of Wales Hospital, Hong Kong SAR. CL was measured using transvaginal ultrasound, with a secondary analysis of archived 1664 images acquired during a prospective study and processed through a ResUNet-based model. This model, combining UNet and ResNet architectures, a modified ResUNet framework, aimed to overcome the limitations of current measurement techniques by providing a more accurate prediction of CL, particularly in cases where the cervix is curved. RESULTS: The AI-line method demonstrated superior accuracy in predicting sPTB at <37 and <32 weeks of gestation compared with conventional methods, with higher areas under the receiver operating characteristic curve (AUROC). The AUROC of CL measured by the AI-line method (0.676 [95% CI, 0.616-0.735], P < 0.05) in predicting sPTB at <37 weeks of gestation was significantly higher than the single-line (0.537 [95% CI, 0.474-0.6]) and two-line (0.54 [95% CI, 0.473-0.66]) methods. For the prediction of sPTB at <32 weeks of gestation, the AI-line method achieved an AUROC of 0.777 (95% CI, 0.703-0.850). CONCLUSION: The AI-line method offers a more accurate measurement of CL in the first trimester, showing potential as a tool for early screening of sPTB risk. The study's results could significantly influence clinical decision-making, providing a basis for the potential future clinical application of AI in prenatal care.","author":[{"family":"Tai","given":"Yi‐yun"},{"family":"Tseng","given":"Bor‐yann"},{"family":"Yang","given":"Zhu‐han"},{"family":"Ch","given":"Yu"},{"family":"Poon","given":"Liona"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1002/ijgo.70744","URL":"https://doi.org/10.1002/ijgo.70744","source":"openalex"},{"id":"oa:W4406382240","type":"article-journal","title":"Review of current progress on additive manufacturing of medical implants and natural/synthetic fibre reinforced composites","abstract":"Abstract Additive manufacturing, or 3D printing technique, is a technology that uses computerized information to generate three‐dimensional solid objects. These objects are produced by feed‐stocking and fusing materials layer by layer. Compared to conventional manufacturing, additive manufacturing can make geometrical shapes that are very complex within a short time with less material wastage. Remarkable applications of manufacturing technology are found in automobile, aerospace, medicine, and natural/synthetic fibre‐reinforced composites. Manufactured parts are fabricated using metals, ceramics, and mainly polymers or composites. Advancements in research have recently been implemented to optimize the process. This review focuses on the research progress on current methods applied to optimize 3D printed biopolymer medical implants and natural/synthetic fibre‐reinforced composites. The objective of this article is to review new opportunities to produce multifunctional materials and suggest solutions to solve persisting challenges in additive manufacturing of medical implants using natural/synthetic fiber reinforced composites. The influence of process parameters on output performance measures, as well as the modelling and simulation techniques applied, are critically established in this paper. Current 3D printing processes and technologies, including the status and future of additive manufacturing, are also critically presented. Finally, challenges and research opportunities for improved high‐performing and less costly printed parts are also illustrated.","author":[{"family":"Nsanzumuhire","given":"C"},{"family":"Daramola","given":"Oluyemi"},{"family":"Oladele","given":"Isiaka"},{"family":"Akinwekomi","given":"Akeem"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/mawe.202400070","URL":"https://doi.org/10.1002/mawe.202400070","source":"openalex"},{"id":"oa:W7127610098","type":"article-journal","title":"Advancing AI Competency in Graduate Medical Education: A Developmental Framework","abstract":"The integration of artificial intelligence (AI) into health care is no longer a scenario reserved for science fiction; it is rapidly becoming a critical component of clinical practice.1 A 2024 survey of 43 U.S. health systems found that 100% had implemented AI-based clinical documentation tools, whereas the Food and Drug Administration has authorized over 1,000 AI-enabled medical devices as of 2024—representing exponential growth from just 6 approvals in 2015 to 223 in 2023.2 As development and implementation accelerate, medical education confronts an urgent imperative to reimagine how to prepare current and future physicians for a landscape where AI tools increasingly influence healthcare policy, clinical decision making, documentation, and patient interaction.","author":[{"family":"Mahmud","given":"Tauhid"},{"family":"Jadotte","given":"Yuri"},{"family":"Lane","given":"Dorothy"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1016/j.focus.2026.100484","URL":"https://doi.org/10.1016/j.focus.2026.100484","source":"openalex"},{"id":"oa:W7134058249","type":"article-journal","title":"Policies and Guidelines for the Use of Artificial Intelligence in Latin American Journals Indexed in Scopus and Classified According to the Scimago Journal Rank (SJR)","abstract":"The emergence of artificial intelligence tools in scientific production is generating significant challenges for scientific integrity and editorial governance, prompting journals and publishers to develop normative guidelines for their use. This study analyzes the current state of guideline implementation among Latin American journals indexed in Scopus and classified according to the Scimago Journal Rank (SJR). A quantitative approach was adopted, complemented by a descriptive documentary analysis based on a detailed review of the websites of 1119 journals from 17 Latin American countries. The collected data were systematized using Excel and analyzed through descriptive and inferential statistical techniques. The results indicate that only 27.2% of journals have explicit guidelines on the use of artificial intelligence, with a predominantly regulatory rather than punitive orientation that prioritizes technical support while restricting practices that compromise human intellectual control. Additionally, statistically significant differences were identified according to quality indicators, showing that journals with higher quality levels are more likely to have such guidelines. Overall, the findings reveal an incipient and heterogeneous regulatory development, underscoring the need to strengthen and harmonize editorial guidelines on artificial intelligence in order to safeguard transparency, clarify the responsibilities of the actors involved in the production and publication process, and protect the integrity of scientific communication.","author":[{"family":"Zahn-Muñoz","given":"Cristian"},{"family":"Viancos","given":"Patricio"},{"family":"Alarcón-Henríquez","given":"Nancy"},{"family":"Aravena-Niño","given":"Bastián"},{"family":"Martínez-Rojas","given":"Ezequiel"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/publications14010017","URL":"https://doi.org/10.3390/publications14010017","source":"openalex"},{"id":"oa:W7164741966","type":"article-journal","title":"Artificial Intelligence-Assisted Quantification of Longitudinal HRCT Changes During Treatment of Pulmonary Tuberculosis: An Exploratory Proof-of-Concept Study","abstract":"Background: Treatment monitoring in pulmonary tuberculosis increasingly requires assessment of residual inflammatory burden and structural lung damage beyond microbiologic response alone. High-resolution computed tomography (HRCT) can provide this information, but interpretation of serial examinations is time-consuming and partly subjective. This study did not aim to evaluate AI for the diagnosis of pulmonary tuberculosis. Instead, it explored whether artificial intelligence (AI)-assisted quantitative HRCT analysis could support longitudinal assessment of treatment-related imaging changes in patients with microbiologically confirmed pulmonary tuberculosis. Methods: We conducted a retrospective, single-center, exploratory longitudinal study of patients receiving treatment for pulmonary tuberculosis. HRCT examinations acquired at diagnosis and during follow-up were anonymized, reviewed by an expert thoracic radiologist, and processed using AVIEW Lung Texture (Coreline Soft v2.0). The software quantified total lung volume and six predefined parenchymal categories: normal lung, ground-glass opacity, consolidation, reticulation, honeycombing, and emphysema. Results: Ninety-six patients contributed 256 HRCT examinations. The most frequent software-detected abnormalities were ground-glass opacity, consolidation, and emphysema-labeled low-attenuation areas. Ground-glass opacity and consolidation showed the clearest decline across serial examinations, consistent with regression of active inflammatory disease during treatment. Reticulation showed a heterogeneous course, likely reflecting both inflammatory resolution and residual structural remodeling. Honeycombing was infrequent and quantitatively limited. Lung volume changed variably and did not consistently parallel visual improvement. A key methodological limitation was the absence of a dedicated cavity class. As a result, emphysema-labeled low-attenuation areas should not be interpreted as conventional emphysema alone, because tuberculous cavities and post-destructive abnormalities were frequently included in this category. Conclusions: AI-assisted HRCT quantification may support longitudinal assessment of pulmonary tuberculosis by providing structured and reproducible measures of interval change. However, tuberculosis-specific interpretation remains dependent on expert radiologic oversight, particularly in cavitary disease.","author":[{"family":"Russo","given":"A"},{"family":"Patanè","given":"Vittorio"},{"family":"Ruotolo","given":"Francesco"},{"family":"Brunese","given":"Maria"},{"family":"Canto","given":"Maria"},{"family":"Alessio","given":"Loredana"},{"family":"Monari","given":"Caterina"},{"family":"Coppola","given":"Nicola"},{"family":"Reginelli","given":"Alfonso"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/diagnostics16121822","URL":"https://doi.org/10.3390/diagnostics16121822","source":"openalex"},{"id":"oa:W7160492411","type":"article-journal","title":"GenAI-Supported Virtual Patients in Health Care Education: Systematic Review","abstract":"Background: Generative artificial intelligence (GenAI) is enhancing virtual patient simulations in health care education by enabling dynamic, adaptive interactions, reshaping how clinical skills are taught. A synthesis of the current evidence is needed to guide implementation and future research, given the pace of technological advancement. Objective: This systematic review aims to synthesize empirical research on the design, implementation, and educational impact of GenAI-supported virtual patients in health care education. Methods: A systematic search was conducted across 5 databases (CINAHL, Medline, Embase, Scopus, and Web of Science) from their inception to March 19, 2026. Reference lists of included studies and relevant systematic reviews were also screened. Peer-reviewed studies in English that evaluated GenAI-supported virtual patients using quantitative or mixed methods were included. Two reviewers independently screened studies and extracted data. Study quality and risk of bias were assessed critically using JBI (Joanna Briggs Institute) checklists, with disagreements resolved by consensus. Results: A total of 15 studies met the inclusion criteria (total participants N=645), spanning health care disciplines, including nursing, medicine, pharmacy, radiography, and medical first-responder training. The virtual patients varied in design; input modalities included text (9 studies), voice (5 studies), or hybrid (1 study); output was text (9 studies), speech (5 studies), or both (1 study); 6 studies used 3D-embodied avatars, while 9 used nonembodied interfaces. A total of 13 studies used OpenAI GPT models (eg, ChatGPT), 1 used a fine-tuned model from a different provider, and 1 evaluated multiple model families (Claude, GPT, and open-source). Further, 6 studies used controlled experimental designs, including 3 randomized controlled trials (RCTs); the remainder were cross-sectional or prepost evaluations. Primary outcomes included user perceptions (14 studies), communication skills (4 studies), clinical reasoning (3 studies), and performance (7 studies). In controlled comparisons, GenAI-supported virtual patients consistently improved outcomes relative to control conditions: for example, enhanced clinical decision-making (RCT, n=21), ophthalmology history-taking skills (RCT, n=26), and medical history-taking performance (crossover RCT, n=20). The evidence base is characterized by brief intervention durations, a predominant reliance on single-session interactions, and a general lack of underpinning educational theory. No meta-analysis was performed due to the limited number of studies and significant heterogeneity in designs, interventions, and outcome measures. Conclusions: The evidence supports the feasibility and acceptability of GenAI-supported virtual patients, with positive learner perceptions and promising outcomes for skills development. However, critical limitations remain in emotional-behavioral complexity, simulation adaptability, and research design rigor (eg, limited use of control groups and validated instruments). The review offers educators, instructional designers, and policymakers actionable insights for integrating dynamic, artificial intelligence-driven simulations while identifying crucial gaps-such as the need for theoretical grounding, longitudinal studies, and standardized design protocols-that must be addressed for safe and effective implementation.","author":[{"family":"Jiang","given":"Juming"},{"family":"Ye","given":"Megan"},{"family":"Kwok","given":"T"},{"family":"Wong","given":"Janet"}],"issued":{"date-parts":[[2026]]},"DOI":"10.2196/82756","URL":"https://doi.org/10.2196/82756","source":"openalex"},{"id":"oa:W7133128916","type":"article-journal","title":"Artificial Intelligence for Pulmonary Abnormality Detection in Chest X-Ray Imaging: A Detailed Review of Methods, Datasets and Future Directions","abstract":"Chest X-ray (CXR) imaging remains the most widely used radiological modality for assessing pulmonary and cardiothoracic disease, yet its interpretation is inherently constrained by tissue superposition, subtle radiographic findings and marked inter-observer variability. Recent advances in artificial intelligence (AI) have driven significant progress in automated CXR analysis, supported by large public datasets, evolving annotation strategies and increasingly expressive deep learning architectures. This review presents a comprehensive synthesis of approaches for pulmonary abnormality detection, encompassing convolutional neural networks, transformers, multimodal and vision–language models and self-supervised representation learning. We critically discuss their strengths, limitations and vulnerability to label noise, domain shift and shortcut learning. In parallel, we examine dataset properties, annotation practices, robustness challenges, explainability methods and the heterogeneity of evaluation protocols that hinder fair comparison and clinical translation. Building on these observations, the review identifies key future directions, including foundation models, multimodal integration, federated and domain-generalized training, longitudinal modeling, synthetic data generation and standardized clinical evaluation frameworks. By integrating methodological and clinical perspectives, this work offers an up-to-date reference for researchers and clinicians and outlines a roadmap toward reliable, interpretable and clinically deployable AI systems for chest radiography.","author":[{"family":"Parra-Cabrera","given":"G"},{"family":"Jiménez-Delgado","given":"JJ"},{"family":"Pérez-Cano","given":"FD"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/technologies14030147","URL":"https://doi.org/10.3390/technologies14030147","source":"openalex"},{"id":"oa:W7165117924","type":"article-journal","title":"Andes virus outbreak linked to expedition cruise ship travel, multi-country investigation and response, April to June 2026","abstract":"As at 9 July 2026 †† , 13 cases (12 confirmed and one probable) of Andes orthohantavirus have been reported (case fatality: 23%), linked to the Dutch-flagged expedition cruise ship m/v Hondius . The event involved individuals from 23 nationalities and required medical evacuation, repatriation, coordinated international contact tracing, isolation, quarantine and clinical and laboratory testing follow-up. To date, all cases have been passengers (10/121; 8%) or crew members (3/61; 5%). Ongoing monitoring and investigations aim to clarify the source of the outbreak, identify risk factors and prevent further spread.","author":[{"family":"Berg","given":"O"},{"family":"Team","given":"Ukhsa"},{"family":"Severi","given":"Ettore"},{"family":"Mutoka-Banza","given":"Freddy"},{"family":"Vugt","given":"Michèle"},{"family":"Schadd","given":"Esther"},{"family":"Reyes","given":"María"},{"family":"Larrégola","given":"Laura"},{"family":"Prieto","given":"Pedro"},{"family":"Lescure","given":"François"},{"family":"Cachia","given":"Mark"},{"family":"Singal","given":"Mayank"},{"family":"Faber","given":"Mirko"},{"family":"Zingg","given":"W"},{"family":"Ismail","given":"Nazir"},{"family":"Wierik","given":"Margreet"},{"family":"Leenstra","given":"Tjalling"},{"family":"Reusken","given":"Chantal"},{"family":"Hof","given":"Susan"},{"family":"Team","given":"The"}],"issued":{"date-parts":[[2026]]},"DOI":"10.2807/1560-7917.es.2026.31.24.2600477","URL":"https://doi.org/10.2807/1560-7917.es.2026.31.24.2600477","source":"openalex"},{"id":"oa:W7155523385","type":"article-journal","title":"Underrepresentation of children in public medical imaging datasets","abstract":"Abstract Artificial intelligence (AI) has the potential to transform healthcare for all patients. Yet, there are disproportionately fewer paediatric AI studies and US Food and Drug Administration approvals relative to adults, indicating less effort focused on AI for children. Here, given that innovations in medical AI are accelerated by community-driven research on public datasets, we hypothesized that the disparity in AI for paediatrics is tied to the lack of public paediatric medical imaging data to support their development and evaluation. To that end, we systematically identified and reviewed 203 datasets, revealing that 33% of datasets lacked metadata on patient ages, and when available, children represented less than 2% of patients. To illustrate how a lack of paediatric data can lead to harmful algorithmic bias, we trained models to classify adult cardiomegaly and evaluated them on healthy children. We found a consistent pattern of age-related bias, reproducible across four large-scale public chest X-ray datasets. These findings suggest that the lack of public paediatric data hinders the development of safe AI for children, producing a landscape of adult-first and adult-only AI models with unknown patterns of bias in children.","author":[{"family":"Hua","given":"Stanley"},{"family":"Heller","given":"Nicholas"},{"family":"He","given":"Ping"},{"family":"Towbin","given":"Alexander"},{"family":"Chen","given":"Irene"},{"family":"Lu","given":"Alex"},{"family":"Erdman","given":"Lauren"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1038/s44360-026-00111-3","URL":"https://doi.org/10.1038/s44360-026-00111-3","source":"openalex"},{"id":"oa:W7134810623","type":"article-journal","title":"Clinical validation of artificial intelligence algorithms for the detection of different central-involved retinal pathologies and glaucoma from non-mydriatic images","abstract":"The use of Artificial intelligence (AI) algorithms for detecting different ophthalmic diseases, especially diabetic retinopathy (DR), has become increasingly popular. In this paper, we evaluate the screening performance of different AI algorithms based on convolutional neural networks (CNNs) in a real-world scenario. To that aim, we conducted an observational and cross-sectional study on patients aged ≥18 years with type-2 diabetes mellitus, who had undergone fundus examination for DR screening using a teleophthalmology program. We used the UPRETINA diagnostic system, which consists of 8 AI algorithms based on CNNs. A total of 1,652 eyes from 871 patients were analyzed. The AI algorithms had a sensitivity/specificity of 86.8%/95.6% for detecting DR; 94.9%/94.3% for detecting age-related macular degeneration (AMD); 82.7%/92.4% for detecting glaucomatous optic neuropathy (GON); 87.0%/87.5% for detecting epiretinal membrane; and 89.7%/98.0% for detecting nevus. Additionally, the sensitivity/specificity for correctly classifying images as right eye/left eye and to correctly classifying images gradeability (medium or high quality) were 100% /100 and 92.9%/90.5%, respectively. The AUROC of the AI algorithms ranged between 0.9777 (AMD) and 0.9122 (GON). UPRETINA system was capable of automatically and accurately classifying the screening retinographies, reducing workload and leading to a scenario of more efficient optimization of resources. Clinical trial registration: https://clinicaltrials.gov/study/NCT04132401 NCT04132401.","author":[{"family":"Vidal","given":"Josep"},{"family":"Bonache","given":"Alba"},{"family":"Solé-Casals","given":"Jordi"},{"family":"Fibla","given":"Dídac"},{"family":"Marín-Gomez","given":"Francesc"},{"family":"Distéfano","given":"Laura"},{"family":"Boixadera","given":"Anna"},{"family":"Casado-García","given":"Ángela"},{"family":"García-Domínguez","given":"Manuel"},{"family":"Inés","given":"Adrián"},{"family":"Heras","given":"Jonathan"},{"family":"Zapata","given":"Miguel"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3389/frai.2026.1754682","URL":"https://doi.org/10.3389/frai.2026.1754682","source":"openalex"},{"id":"doi:10.5281/zenodo.22059975","type":"article-journal","title":"ATLAS-A: ECG metadata controls","abstract":"This archive contains version 1.0.0 of ATLAS-A: ECG metadata controls, the reproducibility package for a study evaluating whether demographic and anthropometric metadata improve multi-label 12-lead ECG classification beyond ECG-only models. The archive includes analysis code, locked analysis plans, audit records, deterministic verification tests, aggregate numerical results, and sanitized run-level metrics for PTB-XL v1.0.3 and a patient-disjoint CODE-15% inter-dataset replication with retraining. The CODE-15% analysis is not presented as strict external validation of a fixed PTB-XL model because only three target conduction disorders could be harmonized across datasets. Raw ECG recordings, patient-level indices, model checkpoints, prediction arrays, execution logs, and the manuscript are intentionally excluded. The underlying datasets must be obtained from their official repositories. Complete environment, data-preparation, execution, and verification instructions are provided in the accompanying documentation.","author":[{"family":"Segnane","given":"Ezyn"},{"family":"Ommane","given":"Younes"},{"family":"Peluffo-Ordóñez","given":"Diego"},{"family":"Mouadili","given":"Maryam"},{"family":"El Waled","given":"Khalil"},{"family":"Cheikh Tourad","given":"Mohamedou"},{"family":"Beddi","given":"Mohamed"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22059975","URL":"https://doi.org/10.5281/zenodo.22059975","source":"datacite"},{"id":"doi:10.5281/zenodo.22059974","type":"article-journal","title":"ATLAS-A: ECG metadata controls","abstract":"This archive contains version 1.0.0 of ATLAS-A: ECG metadata controls, the reproducibility package for a study evaluating whether demographic and anthropometric metadata improve multi-label 12-lead ECG classification beyond ECG-only models. The archive includes analysis code, locked analysis plans, audit records, deterministic verification tests, aggregate numerical results, and sanitized run-level metrics for PTB-XL v1.0.3 and a patient-disjoint CODE-15% inter-dataset replication with retraining. The CODE-15% analysis is not presented as strict external validation of a fixed PTB-XL model because only three target conduction disorders could be harmonized across datasets. Raw ECG recordings, patient-level indices, model checkpoints, prediction arrays, execution logs, and the manuscript are intentionally excluded. The underlying datasets must be obtained from their official repositories. Complete environment, data-preparation, execution, and verification instructions are provided in the accompanying documentation.","author":[{"family":"Segnane","given":"Ezyn"},{"family":"Ommane","given":"Younes"},{"family":"Peluffo-Ordóñez","given":"Diego"},{"family":"Mouadili","given":"Maryam"},{"family":"El Waled","given":"Khalil"},{"family":"Cheikh Tourad","given":"Mohamedou"},{"family":"Beddi","given":"Mohamed"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22059974","URL":"https://doi.org/10.5281/zenodo.22059974","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33314797","type":"article-journal","title":"Left ventricular hypertrophy as a predictor of adverse maternal and neonatal outcomes in chronic hypertension in pregnancy: a multicenter study","abstract":"The impact of chronic hypertension (CHTN) combined with left ventricular hypertrophy (LVH) on adverse maternal and fetal pregnancy outcomes remains unclear. This multicenter retrospective cohort study included pregnant women with CHTN from four tertiary hospitals in China. Baseline characteristics were compared using the Kruskal–Wallis and Chi-squared tests with Bonferroni correction. Logistic regression (LR) identified risk factors for adverse outcomes. Four machine learning (ML) algorithms were validated, nomograms were developed for model visualization, and SHapley Additive exPlanations (SHAP) determined variable importance. Among 500 women, 117 (23.4%) had normal geometry, 118 (23.6%) concentric remodeling, 88 (17.6%) eccentric hypertrophy, and 177 (35.4%) concentric hypertrophy. Adverse maternal and fetal outcomes occurred in 126 (25.2%) and 359 (71.8%) women, respectively. LVH was independently associated with adverse maternal outcomes (OR 2.16, 95% CI 1.13–4.13) and fetal outcomes (OR 2.42, 95% CI 1.31–4.45). Additional maternal predictors included NYHA class III–IV, pre-eclampsia, oligohydramnios, elevated alanine aminotransferase and lactate dehydrogenase, hypoalbuminemia, greater bleeding loss, and blood transfusion. Fetal predictors included increased posterior wall thickness, oligohydramnios, abnormal umbilical artery flow, elevated alanine aminotransferase and blood urea nitrogen, low total protein, and proteinuria. ML models showed AUCs of 0.70–0.87 for maternal and 0.80–0.88 for fetal outcomes; SHAP identified LVH as an important contributor in both models. Early-pregnancy LVH was associated with higher risks of adverse maternal and fetal outcomes in women with CHTN. These models may support individualized risk stratification but require prospective external validation before clinical implementation.","author":[{"family":"Zhang","given":"Bei"},{"family":"Pan","given":"Guilong"},{"family":"Wang","given":"Wei"},{"family":"Zhang","given":"Kai"},{"family":"Song","given":"Xiao"},{"family":"Tian","given":"Wei"},{"family":"Chu","given":"Ran"},{"family":"Li","given":"Shuyi"},{"family":"An","given":"Hui"},{"family":"Zhang","given":"Shuo"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33314797","URL":"https://doi.org/10.6084/m9.figshare.33314797","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33314797.v1","type":"article-journal","title":"Left ventricular hypertrophy as a predictor of adverse maternal and neonatal outcomes in chronic hypertension in pregnancy: a multicenter study","abstract":"The impact of chronic hypertension (CHTN) combined with left ventricular hypertrophy (LVH) on adverse maternal and fetal pregnancy outcomes remains unclear. This multicenter retrospective cohort study included pregnant women with CHTN from four tertiary hospitals in China. Baseline characteristics were compared using the Kruskal–Wallis and Chi-squared tests with Bonferroni correction. Logistic regression (LR) identified risk factors for adverse outcomes. Four machine learning (ML) algorithms were validated, nomograms were developed for model visualization, and SHapley Additive exPlanations (SHAP) determined variable importance. Among 500 women, 117 (23.4%) had normal geometry, 118 (23.6%) concentric remodeling, 88 (17.6%) eccentric hypertrophy, and 177 (35.4%) concentric hypertrophy. Adverse maternal and fetal outcomes occurred in 126 (25.2%) and 359 (71.8%) women, respectively. LVH was independently associated with adverse maternal outcomes (OR 2.16, 95% CI 1.13–4.13) and fetal outcomes (OR 2.42, 95% CI 1.31–4.45). Additional maternal predictors included NYHA class III–IV, pre-eclampsia, oligohydramnios, elevated alanine aminotransferase and lactate dehydrogenase, hypoalbuminemia, greater bleeding loss, and blood transfusion. Fetal predictors included increased posterior wall thickness, oligohydramnios, abnormal umbilical artery flow, elevated alanine aminotransferase and blood urea nitrogen, low total protein, and proteinuria. ML models showed AUCs of 0.70–0.87 for maternal and 0.80–0.88 for fetal outcomes; SHAP identified LVH as an important contributor in both models. Early-pregnancy LVH was associated with higher risks of adverse maternal and fetal outcomes in women with CHTN. These models may support individualized risk stratification but require prospective external validation before clinical implementation.","author":[{"family":"Zhang","given":"Bei"},{"family":"Pan","given":"Guilong"},{"family":"Wang","given":"Wei"},{"family":"Zhang","given":"Kai"},{"family":"Song","given":"Xiao"},{"family":"Tian","given":"Wei"},{"family":"Chu","given":"Ran"},{"family":"Li","given":"Shuyi"},{"family":"An","given":"Hui"},{"family":"Zhang","given":"Shuo"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33314797.v1","URL":"https://doi.org/10.6084/m9.figshare.33314797.v1","source":"datacite"},{"id":"doi:10.5281/zenodo.19681252","type":"article-journal","title":"A Comprehensive Review of Machine Learning Techniques for Early Diagnosis of Cardiovascular Disease","abstract":"Heart diseases (CVDs) continue to be significant data has enabled the development of advanced ML models causes of morbidity in the world. mortality, and emphasizing the necessity of early, correct and prognostic and diagnostic systems that are ethically right. Recent improvements in artificial intelligence (AI) and machine. ML have facilitated the creation of data-driven learning (ML). clinical decision support models that can improve. Early disease diagnosis, risk prioritization and personalized treatment planning. The review summarizes current articles (2022-2025) that pay attention to ML- and AI- based. CVD risk prediction methods, CVD diagnosis methods. Assessment of myocardial ischemia, CVD early diagnosis, and. purposive treatment recommendation. The reviewed works use nonhomogeneous data, such as open repositories, large-scale real world, multi-institutional benchmarks. Hospital and ICU data. These studies are methodologically the same. supervised, ensemble, deep learning. learning architectures and hybrid models of ML and. DL and reinforcement learning of sequential clinical decision-making. Some of the studies highlight explainable AI. (XAI) methods and moral aspects to improve on. Clinical trust and safety. In comparison with other businesses, it implies that. Hybrid structures and ensemble structures frequently perform better. Stronger than predictive performance and strength. Separate models on benchmark data. However, problems connected with interpretability, extrinsic validation, fairness, and real-time clinical deployment are maintained. The review concludes with naming of important gaps in research and pointing out. Prospects at glorifiable, ethical and scalable AI. Effective real world healthcare systems.","author":[{"family":"Azeez","given":"Mohd"},{"family":"Ahmad","given":"Prof"},{"family":"Maurya","given":"Mr"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19681252","URL":"https://doi.org/10.5281/zenodo.19681252","source":"datacite"},{"id":"doi:10.5281/zenodo.19681253","type":"article-journal","title":"A Comprehensive Review of Machine Learning Techniques for Early Diagnosis of Cardiovascular Disease","abstract":"Heart diseases (CVDs) continue to be significant data has enabled the development of advanced ML models causes of morbidity in the world. mortality, and emphasizing the necessity of early, correct and prognostic and diagnostic systems that are ethically right. Recent improvements in artificial intelligence (AI) and machine. ML have facilitated the creation of data-driven learning (ML). clinical decision support models that can improve. Early disease diagnosis, risk prioritization and personalized treatment planning. The review summarizes current articles (2022-2025) that pay attention to ML- and AI- based. CVD risk prediction methods, CVD diagnosis methods. Assessment of myocardial ischemia, CVD early diagnosis, and. purposive treatment recommendation. The reviewed works use nonhomogeneous data, such as open repositories, large-scale real world, multi-institutional benchmarks. Hospital and ICU data. These studies are methodologically the same. supervised, ensemble, deep learning. learning architectures and hybrid models of ML and. DL and reinforcement learning of sequential clinical decision-making. Some of the studies highlight explainable AI. (XAI) methods and moral aspects to improve on. Clinical trust and safety. In comparison with other businesses, it implies that. Hybrid structures and ensemble structures frequently perform better. Stronger than predictive performance and strength. Separate models on benchmark data. However, problems connected with interpretability, extrinsic validation, fairness, and real-time clinical deployment are maintained. The review concludes with naming of important gaps in research and pointing out. Prospects at glorifiable, ethical and scalable AI. Effective real world healthcare systems.","author":[{"family":"Azeez","given":"Mohd"},{"family":"Ahmad","given":"Prof"},{"family":"Maurya","given":"Mr"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19681253","URL":"https://doi.org/10.5281/zenodo.19681253","source":"datacite"},{"id":"doi:10.24412/cl-34438-2026-785-139-149","type":"article-journal","title":"ИСКУССТВЕННЫЙ ИНТЕЛЛЕКТ В НЕЙРОХИРУРГИИ ОТ ДИАГНОСТИКИ ДО ПРОГНОЗИРОВАНИЯ ИСХОДОВ ОПЕРАЦИЙ","abstract":"Современная нейрохирургия требует прецизионной точности, минимизации интраоперационных рисков и объективного прогнозирования результатов лечения. Стремительное увеличение объемов диагностических данных и необходимость персонализации хирургических подходов обуславливают интеграцию методов искусственного интеллекта (ИИ) и машинного обучения (МО) в клиническую практику. Установлено, что применение ИИ в нейрорадиологии обеспечивает точность автоматической сегментации новообразований и сосудистых мальформаций на уровне 95-97%, а радиогеномика позволяет неинвазивно верифицировать молекулярный профиль опухолей. Доказана эффективность ИИ в компенсации феномена смещения мозга (brain shift) в режиме реального времени. В области прогностического моделирования алгоритмы машинного обучения демонстрируют высокую прогностическую ценность (AUC-ROC &gt; 0.88) при оценке рисков послеоперационных осложнений и выживаемости пациентов. Технологии ИИ трансформируют концепцию нейрохирургической помощи, выступая в роли когнитивного ассистента врача. Несмотря на барьеры в виде проблемы «черного ящика» и юридической неопределенности, дальнейшее развитие гибридного интеллекта (синергии хирурга и нейросети) является главным вектором эволюции специальности, способным значимо снизить уровень инвалидизации пациентов.","author":[{"family":"Қанағатқызы","given":"Құрбанова"},{"family":"Мухамедалиулы","given":"Каримбердиев"},{"family":"Бекзатұлы","given":"Жолдасов"},{"family":"Зарафулы","given":"Ниязов"},{"family":"Рустамович","given":"Абдуллаев"}],"issued":{"date-parts":[[2026]]},"DOI":"10.24412/cl-34438-2026-785-139-149","URL":"https://doi.org/10.24412/cl-34438-2026-785-139-149","source":"datacite"},{"id":"doi:10.5281/zenodo.20111083","type":"article-journal","title":"Mingzheng — Reproducibility Data Package","abstract":"This Zenodo record provides the reproducibility package for the Mingzheng study: a taxonomy-based multimodal artificial intelligence system for interpretable Traditional Chinese Medicine syndrome differentiation in cancer patients with comorbid sleep disorders. This package is independent of manuscript version. It supports reproduction of the main model benchmarks, leave-one-site-out (LOSO) ensemble inference, temporal validation, reader-study model outputs, zero-shot LLM comparisons, and supplementary subgroup analyses reported across manuscript versions. Related paper: Zheng Xiao, Xie Yong, Luo Song, et al. A taxonomy-based multimodal artificial intelligence system for interpretable syndrome differential diagnosis in cancer patients with comorbid sleep disorders. Manuscript, 2026. Zenodo DOI: https://doi.org/10.5281/zenodo.20111083 Code repository: https://github.com/Jayden-XL/mingzheng For now, this Code repository is accessible only to the editor and editor-invited reviewers. If you are accessing this record during peer review, please contact us through the editor. Contents checkpoints/ - Trained LOSO Ensemble Models, approximately 7.1 GB Trained model checkpoints for the fully deployable version of Mingzheng. The released version contains 15 checkpoints: checkpoints/ |-- loso_seed42/fold{1-5}_best_ckpt.pth |-- loso_seed43/fold{1-5}_best_ckpt.pth `-- loso_seed44/fold{1-5}_best_ckpt.pth This corresponds to 3 random seeds x 5 leave-one-site-out cross-validation folds. For site-matched LOSO inference, a patient from site s is scored only by the three checkpoints whose held-out site is s, one checkpoint per seed. The three predicted probability vectors are then averaged. This scheme prevents the patient's own clinical site from contributing to the training data and is used for deployment-style cross-site inference, temporal validation, and reader-study model outputs. The same checkpoint family can also be used to compute pooled internal LOSO ensemble metrics across the five development sites. Alternative ensemble summaries are provided only as sensitivity analyses where indicated in the analysis files. fold_splits/ - Data Partition Metadata, approximately 13 KB LOSO fold assignments and metadata: fold_splits/ `-- loso_seed{42,43,44}/loso{1-5}/split_info.json Each split_info.json contains: holdout_site test_patient_ids n_train n_val n_test Patient identifiers are de-identified study IDs. analysis/ - Aggregated Results and Manuscript Mapping, approximately 70 KB Aggregated analysis outputs used to reproduce model rankings, ablations, subgroup analyses, and manuscript display items. Key files include: ranking.csv - model ranking table with paper display names and experiment identifiers experiment_map.csv - mapping from display name to experiment name, model file, training script, base experiment, key CLI flags, seeds, and notes per_syndrome.csv - per-syndrome F1/AUC summaries ablation_v3_expert.csv - expert-annotation ablation results ablation_v4_llm.csv - LLM-component ablation results ablation_v4_bge_expert.csv - BGE-embedding ablation results fold_diagnostic.csv - per-fold diagnostic metrics *_cv_summary.json - per-seed or aggregated cross-validation summaries with fold-level F1/AUC breakdowns Because figure and table numbering may differ across journal submissions, use experiment_map.csv and the GitHub repository README to map these files to the corresponding manuscript display items. llm_responses/ - Zero-Shot LLM Outputs, Numeric-Only, approximately 61 KB Per-patient JSON-Lines outputs from the zero-shot LLM baselines evaluated in the study. Free-text fields have been removed before public release. Each record contains only the numeric and administrative fields required by the downstream reproduction scripts, such as: patient_id hashed name field where applicable cancer type parsed per-syndrome confidence scores parsed overall difficulty error status number of attempts Removed fields include raw model responses, supporting/","author":[{"family":"Zheng","given":"Xueer"},{"family":"Xie","given":"Ying"},{"family":"Luo","given":"Shijun"},{"family":"Yan","given":"Yici"},{"family":"Ruan","given":"Shanming"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20111083","URL":"https://doi.org/10.5281/zenodo.20111083","source":"datacite"},{"id":"doi:10.5281/zenodo.20111082","type":"article-journal","title":"Mingzheng — Reproducibility Data Package","abstract":"Zenodo v2 v2 (2026-05-25): Added a public-cohort transfer stress test on TCM-SD (Supplementary Part I). This release adds a subdirectory, tcmsd_external_validation/ (approximately 270 MB), containing reproducibility artifacts for the TCM-SD external-transfer stress test reported in Supplementary Part I: Qwen3-30B-A3B-Instruct-2507 16-slot structured extraction outputs for 5,316 TCM-SD test records, with an extraction success rate of 99.55%. BGE-m3 dense embeddings of the 16-slot sentences, with shape (5316, 16, 1024). Predictions from 15 frozen LOSO ensemble models and proxy labels for {phlegm-dampness, blood stasis, yin deficiency}. Raw logits for each ensemble member, with shape (15, 5316, 3). Five evaluation reports: A, Mingzheng zero-shot paper-locked thresholds; B, A plus half-retained threshold calibration; C, BGE-slot linear probe with 5-fold cross-validation; D, BGE-only zero-shot with a probe trained on the CRSD source cohort; and E, TF-IDF + SGD self-baseline reference. Reports include two-sided paired-bootstrap p values. All 10 pipeline scripts: L2 LLM extraction, BGE encoding, frozen Mingzheng forward propagation, the monitoring program, and five evaluation scripts. A subdirectory README documenting the workflow, results, per-file licenses, and reproduction commands. Main result (Group A vs Group D; n = 5,316; both true zero-shot, with no TCM-SD labels used): Macro-AUC = +0.020 (95% CI +0.002 to +0.038), two-sided paired-bootstrap p = 0.025, driven mainly by blood stasis (Delta AUC = +0.052, p < 0.001). This advantage did not extend to a statistically detectable macro-F1 difference (Delta F1 = +0.007, p = 0.250), and was not observed for yin deficiency. We position this analysis as a public-cohort transfer stress test rather than the primary external validation, because its operating-point performance is limited by label-transfer effects, especially the approximately 14.8-fold lower prevalence of yin deficiency relative to CRSD. The clinical-grade external-validation evidence for Mingzheng remains the prospective multicenter temporal cohort study (n = 47, collected from n = 105). License for the new subdirectory: Scripts are released under the MIT License, consistent with the Mingzheng GitHub repository. TCM-SD-derived outputs (tcmsd_l2_output.json, tcmsd_bge_4zhen.npz, tcmsd_predictions.json, tcmsd_labels.json, tcmsd_ensemble_logits.npz, tcmsd_test_l2_input.jsonl, tcmsd_eval_v*.json, and *.md) are released under CC BY-NC-SA 4.0, inheriting the ShareAlike requirement of the upstream TCM-SD dataset (Ren et al., CCL 2022; ModelScope OmniData/TCM-SD). Related materials: The corresponding code updates have been committed to the private Mingzheng GitHub repository under external_validation_tcmsd/ (commit 0046d59; https://github.com/Jayden-XL/mingzheng). For now, this repository is accessible only to the editor and editor-invited reviewers. If you are accessing this record during peer review, please contact us through the editor. Mingzheng - Reproducibility Data Package This Zenodo record provides the reproducibility package for the Mingzheng study: a taxonomy-based multimodal artificial intelligence system for interpretable Traditional Chinese Medicine syndrome differentiation in cancer patients with comorbid sleep disorders. This package is independent of manuscript version. It supports reproduction of the main model benchmarks, leave-one-site-out (LOSO) ensemble inference, temporal validation, reader-study model outputs, zero-shot LLM comparisons, and supplementary subgroup analyses reported across manuscript versions. Related paper: Zheng Xiao, Xie Yong, Luo Song, et al. A taxonomy-based multimodal artificial intelligence system for interpretable syndrome differential diagnosis in cancer patients with comorbid sleep disorders. Manuscript, 2026. Zenodo DOI: https://doi.org/10.5281/zenodo.20111083 Code repository: https://github.com/Jayden-XL/mingzheng For now, this repository is accessible only to the editor and edito","author":[{"family":"Zheng","given":"Xueer"},{"family":"Xie","given":"Ying"},{"family":"Luo","given":"Shijun"},{"family":"Yan","given":"Yici"},{"family":"Ruan","given":"Shanming"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20111082","URL":"https://doi.org/10.5281/zenodo.20111082","source":"datacite"},{"id":"doi:10.5281/zenodo.20382160","type":"article-journal","title":"Mingzheng — Reproducibility Data Package","abstract":"Zenodo v2 v2 (2026-05-25): Added a public-cohort transfer stress test on TCM-SD (Supplementary Part I). This release adds a subdirectory, tcmsd_external_validation/ (approximately 270 MB), containing reproducibility artifacts for the TCM-SD external-transfer stress test reported in Supplementary Part I: Qwen3-30B-A3B-Instruct-2507 16-slot structured extraction outputs for 5,316 TCM-SD test records, with an extraction success rate of 99.55%. BGE-m3 dense embeddings of the 16-slot sentences, with shape (5316, 16, 1024). Predictions from 15 frozen LOSO ensemble models and proxy labels for {phlegm-dampness, blood stasis, yin deficiency}. Raw logits for each ensemble member, with shape (15, 5316, 3). Five evaluation reports: A, Mingzheng zero-shot paper-locked thresholds; B, A plus half-retained threshold calibration; C, BGE-slot linear probe with 5-fold cross-validation; D, BGE-only zero-shot with a probe trained on the CRSD source cohort; and E, TF-IDF + SGD self-baseline reference. Reports include two-sided paired-bootstrap p values. All 10 pipeline scripts: L2 LLM extraction, BGE encoding, frozen Mingzheng forward propagation, the monitoring program, and five evaluation scripts. A subdirectory README documenting the workflow, results, per-file licenses, and reproduction commands. Main result (Group A vs Group D; n = 5,316; both true zero-shot, with no TCM-SD labels used): Macro-AUC = +0.020 (95% CI +0.002 to +0.038), two-sided paired-bootstrap p = 0.025, driven mainly by blood stasis (Delta AUC = +0.052, p < 0.001). This advantage did not extend to a statistically detectable macro-F1 difference (Delta F1 = +0.007, p = 0.250), and was not observed for yin deficiency. We position this analysis as a public-cohort transfer stress test rather than the primary external validation, because its operating-point performance is limited by label-transfer effects, especially the approximately 14.8-fold lower prevalence of yin deficiency relative to CRSD. The clinical-grade external-validation evidence for Mingzheng remains the prospective multicenter temporal cohort study (n = 47, collected from n = 105). License for the new subdirectory: Scripts are released under the MIT License, consistent with the Mingzheng GitHub repository. TCM-SD-derived outputs (tcmsd_l2_output.json, tcmsd_bge_4zhen.npz, tcmsd_predictions.json, tcmsd_labels.json, tcmsd_ensemble_logits.npz, tcmsd_test_l2_input.jsonl, tcmsd_eval_v*.json, and *.md) are released under CC BY-NC-SA 4.0, inheriting the ShareAlike requirement of the upstream TCM-SD dataset (Ren et al., CCL 2022; ModelScope OmniData/TCM-SD). Related materials: The corresponding code updates have been committed to the private Mingzheng GitHub repository under external_validation_tcmsd/ (commit 0046d59; https://github.com/Jayden-XL/mingzheng). For now, this repository is accessible only to the editor and editor-invited reviewers. If you are accessing this record during peer review, please contact us through the editor. Mingzheng - Reproducibility Data Package This Zenodo record provides the reproducibility package for the Mingzheng study: a taxonomy-based multimodal artificial intelligence system for interpretable Traditional Chinese Medicine syndrome differentiation in cancer patients with comorbid sleep disorders. This package is independent of manuscript version. It supports reproduction of the main model benchmarks, leave-one-site-out (LOSO) ensemble inference, temporal validation, reader-study model outputs, zero-shot LLM comparisons, and supplementary subgroup analyses reported across manuscript versions. Related paper: Zheng Xiao, Xie Yong, Luo Song, et al. A taxonomy-based multimodal artificial intelligence system for interpretable syndrome differential diagnosis in cancer patients with comorbid sleep disorders. Manuscript, 2026. Zenodo DOI: https://doi.org/10.5281/zenodo.20111083 Code repository: https://github.com/Jayden-XL/mingzheng For now, this repository is accessible only to the editor and edito","author":[{"family":"Zheng","given":"Xueer"},{"family":"Xie","given":"Ying"},{"family":"Luo","given":"Shijun"},{"family":"Yan","given":"Yici"},{"family":"Ruan","given":"Shanming"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20382160","URL":"https://doi.org/10.5281/zenodo.20382160","source":"datacite"},{"id":"oa:W4416566022","type":"article-journal","title":"The impact of artificial intelligence-driven simulation on the development of non-technical skills in medical education: a systematic review","abstract":"PURPOSE: Artificial intelligence (AI)-driven simulation is an emerging approach in healthcare education that enhances learning effectiveness. This review examined its impact on the development of non-technical skills among medical learners. METHODS: Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, a systematic review was conducted using the following databases: Web of Science, ScienceDirect, Scopus, and PubMed. The quality of the included studies was assessed using the Mixed. Methods: Appraisal Tool. The protocol was previously registered in PROSPERO (CRD420251038024). RESULTS: Of the 1,442 studies identified in the initial search, 20 met the inclusion criteria, involving 2,535 participants. The simulators varied considerably, ranging from platforms built on symbolic AI methods to social robots powered by computational AI. Among the 15 AI-driven simulators, 10 used ChatGPT or its variants as virtual patients. Several studies evaluated multiple non-technical skills simultaneously. Communication and clinical reasoning were the most frequently assessed skills, appearing in 12 and 6 studies, respectively, which generally reported positive outcomes. Improvements were also noted in decision-making, empathy, self-confidence, critical thinking, and problem-solving. In contrast, emotional regulation, assessed in a single study, showed no significant difference. Notably, none of the studies examined reflection, reflective practice, teamwork, or leadership. CONCLUSION: AI-driven simulation shows substantial potential for enhancing non-technical skills in medical education, particularly communication and clinical reasoning. However, its effects on several other non-technical skills remain unclear. Given heterogeneity in study designs and outcome measures, these findings should be interpreted cautiously. These considerations highlight the need for further research to support integrating this innovative approach into medical curricula.","author":[{"family":"Loubbairi","given":"Sana"},{"family":"Moussaoui","given":"Yassmine"},{"family":"Lahlou","given":"Laila"},{"family":"Chakri","given":"Imad"},{"family":"Nassik","given":"Hicham"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3352/jeehp.2025.22.37","URL":"https://doi.org/10.3352/jeehp.2025.22.37","source":"openalex"},{"id":"oa:W4408736524","type":"article-journal","title":"A systematic review and meta-analysis of diagnostic performance comparison between generative AI and physicians","abstract":"While generative artificial intelligence (AI) has shown potential in medical diagnostics, comprehensive evaluation of its diagnostic performance and comparison with physicians has not been extensively explored. We conducted a systematic review and meta-analysis of studies validating generative AI models for diagnostic tasks published between June 2018 and June 2024. Analysis of 83 studies revealed an overall diagnostic accuracy of 52.1%. No significant performance difference was found between AI models and physicians overall (p = 0.10) or non-expert physicians (p = 0.93). However, AI models performed significantly worse than expert physicians (p = 0.007). Several models demonstrated slightly higher performance compared to non-experts, although the differences were not significant. Generative AI demonstrates promising diagnostic capabilities with accuracy varying by model. Although it has not yet achieved expert-level reliability, these findings suggest potential for enhancing healthcare delivery and medical education when implemented with appropriate understanding of its limitations.","author":[{"family":"Takita","given":"Hirotaka"},{"family":"Kabata","given":"Daijiro"},{"family":"Walston","given":"Shannon"},{"family":"Tatekawa","given":"Hiroyuki"},{"family":"Saito","given":"Kenichi"},{"family":"Tsujimoto","given":"Yasushi"},{"family":"Miki","given":"Yukio"},{"family":"Ueda","given":"Daiju"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41746-025-01543-z","URL":"https://doi.org/10.1038/s41746-025-01543-z","source":"openalex"},{"id":"oa:W4407242328","type":"article-journal","title":"Recent Emerging Techniques in Explainable Artificial Intelligence to Enhance the Interpretable and Understanding of AI Models for Human","abstract":"Recent advancements in Explainable Artificial Intelligence (XAI) aim to bridge the gap between complex artificial intelligence (AI) models and human understanding, fostering trust and usability in AI systems. However, challenges persist in comprehensively interpreting these models, hindering their widespread adoption. This study addresses these challenges by exploring recently emerging techniques in XAI. The primary problem addressed is the lack of transparency and interpretability in AI models to humanity for institution-wide use, which undermines user trust and inhibits their integration into critical decision-making processes. Through an in-depth review, this study identifies the objectives of enhancing the interpretability of AI models and improving human understanding of their decision-making processes. Various methodological approaches, including post-hoc explanations, model transparency methods, and interactive visualization techniques, are investigated to elucidate AI model behaviours. We further present techniques and methods to make AI models more interpretable and understandable to humans including their strengths and weaknesses to demonstrate promising advancements in model interpretability, facilitating better comprehension of complex AI systems by humans. In addition, we provide the application of XAI in local use cases. Challenges, solutions, and open research directions were highlighted to clarify these compelling XAI utilization challenges. The implications of this research are profound, as enhanced interpretability fosters trust in AI systems across diverse applications, from healthcare to finance. By empowering users to understand and scrutinize AI decisions, these techniques pave the way for more responsible and accountable AI deployment.","author":[{"family":"Mathew","given":"Daniel"},{"family":"Ebem","given":"Deborah"},{"family":"Ikegwu","given":"Anayo"},{"family":"Ukeoma","given":"Pamela"},{"family":"Dibiaezue","given":"Ngozi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s11063-025-11732-2","URL":"https://doi.org/10.1007/s11063-025-11732-2","source":"openalex"},{"id":"oa:W7125589886","type":"article-journal","title":"Clinical research on artificial intelligence medical diagnostic devices: A scoping review","abstract":"Artificial intelligence medical diagnostic devices (AIMDDs) show strong potential but face barriers to clinical use, emphasizing the need for rigorous clinical research. We assessed current AIMDD research, key challenges, and future directions. A scoping review followed Arksey and O'Malley's methodological framework and the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews guidelines. PubMed, Web of Science Core Collection, and the Cochrane Database of Systematic Reviews (January 2020–December 2024) were searched on AIMDD design, implementation, and evaluation. Two independent researchers screened and extracted data from the literature using predefined criteria. Ninety-seven articles met the inclusion criteria. Machine learning and deep learning approaches dominated across diverse disease fields, with oncology being the most frequent (41 %). The key challenges identified include insufficient quantity, quality, representativeness, and diversity of data; research designs that do not adequately address clinical needs; poor patient selection; poorly defined gold standards; lack of external and prospective validation; and a disconnect between validation strategies and clinical practice. Additionally, issues such as the “black box” phenomenon, overfitting, and data privacy concerns hinder clinical translation. Completeness and standardization of reporting were also found to be lacking. Significant challenges remain in the development and clinical application of AIMDD. To facilitate their clinical translation, improvements are needed in dataset optimization, clinically driven research design, development of evaluation frameworks, enhanced interpretability, and standardized reporting and validation of algorithms. • This study used a scoping review to synthesize evidence in the complex, high-volume AI medical diagnosis field. • Oncology was the most studied discipline, comprising 41 % of all included publications. • Key barriers include poor data quality, limited external validation, and incomplete methodological reporting. • Future work should focus on dataset optimization, clinician-led design, explainable AI, and standardized validation.","author":[{"family":"Zhang","given":"Xiaowei"},{"family":"Zhang","given":"Xiaowei"},{"family":"Liu","given":"Changning"},{"family":"Sun","given":"Yifan"},{"family":"You","given":"Liangzhen"},{"family":"Zhang","given":"Xiaoyu"},{"family":"Zhang","given":"Xiaoyu"},{"family":"Shang","given":"Hongcai"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1016/j.engmed.2026.100120","URL":"https://doi.org/10.1016/j.engmed.2026.100120","source":"openalex"},{"id":"oa:W4411100445","type":"article-journal","title":"Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology","abstract":"Clinical decision-making in oncology is complex, requiring the integration of multimodal data and multidomain expertise. We developed and evaluated an autonomous clinical artificial intelligence (AI) agent leveraging GPT-4 with multimodal precision oncology tools to support personalized clinical decision-making. The system incorporates vision transformers for detecting microsatellite instability and KRAS and BRAF mutations from histopathology slides, MedSAM for radiological image segmentation and web-based search tools such as OncoKB, PubMed and Google. Evaluated on 20 realistic multimodal patient cases, the AI agent autonomously used appropriate tools with 87.5% accuracy, reached correct clinical conclusions in 91.0% of cases and accurately cited relevant oncology guidelines 75.5% of the time. Compared to GPT-4 alone, the integrated AI agent drastically improved decision-making accuracy from 30.3% to 87.2%. These findings demonstrate that integrating language models with precision oncology and search tools substantially enhances clinical accuracy, establishing a robust foundation for deploying AI-driven personalized oncology support systems.","author":[{"family":"Ferber","given":"Dyke"},{"family":"Nahhas","given":"Omar"},{"family":"Wölflein","given":"Georg"},{"family":"Wiest","given":"Isabella"},{"family":"Clusmann","given":"Jan"},{"family":"Leßmann","given":"Marie"},{"family":"Foersch","given":"Sebastian"},{"family":"Lammert","given":"Jacqueline"},{"family":"Tschochohei","given":"Maximilian"},{"family":"Jaeger","given":"Dirk"},{"family":"Saltotellez","given":"Manuel"},{"family":"Schultz","given":"Nikolaus"},{"family":"Truhn","given":"Daniel"},{"family":"Kather","given":"Jakob"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s43018-025-00991-6","URL":"https://doi.org/10.1038/s43018-025-00991-6","source":"openalex"},{"id":"doi:10.5281/zenodo.19356456","type":"article-journal","title":"Impact and acceptance of digital non-pharmacological treatments for insomnia among cancer patients: a scoping review protocol","abstract":"This document outlines a comprehensive protocol for a scoping review that examines how digital technologies can assist cancer patients suffering from insomnia. The researchers aim to investigate the effectiveness and user satisfaction of non-pharmacological interventions, including specialized software, wearable devices, and artificial intelligence. This review will analyze primary studies published between 2020 and 2025. The goal is to explore how medical treatments can be integrated with technological solutions. The methodology focuses on identifying research gaps related to how these tools address various comorbidities and the types of data they collect. Ultimately, the study aims to provide a structured overview of the current landscape of electronic sleep therapies within oncology care.","author":[{"family":"Velez Gutierrez","given":"Jose"},{"family":"Seepold","given":"Ralf"},{"family":"Ortega","given":"Juan"},{"family":"Martínez Madrid","given":"Natividad"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.19356456","URL":"https://doi.org/10.5281/zenodo.19356456","source":"datacite"},{"id":"doi:10.5281/zenodo.19356457","type":"article-journal","title":"Impact and acceptance of digital non-pharmacological treatments for insomnia among cancer patients: a scoping review protocol","abstract":"This document outlines a comprehensive protocol for a scoping review that examines how digital technologies can assist cancer patients suffering from insomnia. The researchers aim to investigate the effectiveness and user satisfaction of non-pharmacological interventions, including specialized software, wearable devices, and artificial intelligence. This review will analyze primary studies published between 2020 and 2025. The goal is to explore how medical treatments can be integrated with technological solutions. The methodology focuses on identifying research gaps related to how these tools address various comorbidities and the types of data they collect. Ultimately, the study aims to provide a structured overview of the current landscape of electronic sleep therapies within oncology care.","author":[{"family":"Velez Gutierrez","given":"Jose"},{"family":"Seepold","given":"Ralf"},{"family":"Ortega","given":"Juan"},{"family":"Martínez Madrid","given":"Natividad"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.19356457","URL":"https://doi.org/10.5281/zenodo.19356457","source":"datacite"},{"id":"doi:10.6084/m9.figshare.30113146","type":"article-journal","title":"<b>Machine Learning Approaches in Multimodal Analysis of Lung Cancer:</b><b>A Comprehensive Scoping Review</b>","abstract":"Protocol for Scoping Review Title Machine Learning Applications in Multimodal Analysis of Lung Cancer: A Scoping Review Introduction Lung cancer remains the leading cause of cancer-related mortality worldwide. Its biological heterogeneity presents significant challenges for diagnosis, prognosis, and treatment. Advances in artificial intelligence (AI) and machine learning (ML) have enabled the integration of multimodal datasets (e.g., imaging, histopathology, genomics, clinical records), creating new opportunities for precision oncology. Although individual modalities have been extensively studied, multimodal ML applications in lung cancer remain less systematically mapped. Understanding which methodologies are being employed across data fusion, model architectures, validation strategies, and explainability is essential to identify methodological strengths and gaps. This protocol describes the planned scoping review, following JBI methodology for scoping reviews (Aromataris et al., 2024) and the PRISMA Extension for Scoping Reviews (PRISMA-ScR) checklist (Tricco et al., 2018), with alignment to the updated PRISMA 2020 guidelines (Page et al., 2021a; Page et al., 2021b). Objectives • To map methodological approaches in ML applied to multimodal lung cancer datasets. • To classify ML studies by data modalities, fusion techniques, learning strategies, and validation methods. • To highlight methodological trends, challenges, and research gaps. Review Questions 1. What ML methodologies have been applied to multimodal datasets in lung cancer? 2. Which combinations of data modalities are most commonly studied? 3. Which fusion strategies, learning paradigms, and validation methods dominate the field? 4. What gaps and methodological challenges remain in this domain? Eligibility Criteria Population: Studies on lung cancer patients (all subtypes). Concept: Application of ML methodologies (supervised, unsupervised, semi-supervised, reinforcement learning, ensemble methods, deep learning, transfer learning, explainable AI). Context: Multimodal data integration (e.g., imaging + genomics, clinical + omics, pathology + imaging). Outcomes: Diagnostic, prognostic, predictive, and treatment-response applications. Study types: Peer-reviewed studies from 2013 onward; English language. Exclusion: Reviews, editorials, commentaries, non-lung cancer studies, single-modality ML studies. Methods Information Sources Databases: PubMed, Scopus, Embase, Web of Science, and IEEE Xplore. Supplementary hand-searching of reference lists and Google Scholar will also be conducted. Search Strategy The following database-specific strategies will be applied. Actual searches already run are included, with enhancements for sensitivity and specificity using subject headings and synonyms. • PubMed (Feb 6, 2025 – enhanced): ((\"NSCLC\"[Title/Abstract] OR \"non small cell lung cancer\"[Title/Abstract] OR \"lung cancer\"[Title/Abstract] OR \"Lung Neoplasms\"[MeSH])) AND ((\"machine learning\"[Title/Abstract] OR \"deep learning\"[Title/Abstract] OR \"artificial intelligence\"[Title/Abstract] OR \"Machine Learning\"[MeSH])) OR (\"image processing\"[Title/Abstract] OR \"Image Processing, Computer-Assisted\"[MeSH])) AND ((booksdocs[Filter] OR clinicaltrial[Filter] OR randomizedcontrolledtrial[Filter]) AND (2013:2025[pdat])) NOT (\"review\"[Publication Type] OR \"systematic review\"[tiab] OR \"meta-analysis\"[tiab]). • Scopus (Feb 8, 2025 – enhanced): TITLE-ABS-KEY (\"NSCLC\" OR \"non small cell lung cancer\" OR \"lung cancer\") AND TITLE-ABS-KEY (\"machine learning\" OR \"deep learning\" OR \"artificial intelligence\" OR \"image processing\") AND TITLE-ABS-KEY (\"clinical trial\" OR \"randomized controlled trial\" OR \"diagnostic accuracy\" OR \"prediction model\") AND NOT TITLE-ABS-KEY (\"review\" OR \"systematic review\" OR \"meta-analysis\" OR \"scoping review\" OR \"editorial\" OR \"commentary\" OR \"letter\") AND PUBYEAR &gt; 2012 AND PUBYEAR &lt; 2025 AND (LIMIT-TO (LANGUAGE , \"English\")) AND (LIMIT-TO (DOCTYPE , \"ar\") OR LIMIT-TO (D","author":[{"family":"Hashempour","given":"Sara"},{"family":"Han","given":"Lee"},{"family":"Galanti","given":"Mattia"},{"family":"Mayhue","given":"Sari"}],"issued":{"date-parts":[[2025]]},"DOI":"10.6084/m9.figshare.30113146","URL":"https://doi.org/10.6084/m9.figshare.30113146","source":"datacite"},{"id":"doi:10.17605/osf.io/497wc","type":"article-journal","title":"Inteligência artificial no planejamento e prognóstico do tratamento endodôntico: um protocolo de revisão de escopo.","abstract":"RESUMO O tratamento endodôntico requer pleno conhecimento da anatomia dos canais radiculares, sendo os exames de imagem fundamentais para o planejamento e a obtenção de um prognóstico favorável. Nesse contexto, a Inteligência Artificial (IA) tem emergido como uma ferramenta promissora na análise de imagens e no suporte à tomada de decisão clínica. Nesse viés, o presente estudo objetiva mapear as ferramentas de Inteligência Artificial disponíveis na literatura, a fim de auxiliar no planejamento e no prognóstico do tratamento endodôntico. O presente trabalho seguirá a metodologia de revisão de escopo, conforme os padrões e normas do Joanna Briggs Institute (JBI) e do checklist PRISMA-ScR. A pergunta norteadora foi elaborada com base na estratégia PCC: população de elementos dentários submetidos ao tratamento endodôntico; uso de ferramentas de inteligência artificial; e o planejamento e prognóstico do tratamento endodôntico. A busca será realizada nas plataformas PubMed via MEDLINE, Google Acadêmico, SciELO, Scopus e Dentistry &amp; Oral Sciences Source (DoSS), utilizando descritores DeCS/MeSH associados a operadores booleanos. Os resultados serão sintetizados em tabelas e gráficos informativos. Espera-se encontrar diversas ferramentas úteis no dia a dia endodôntico, além de identificar lacunas do conhecimento a serem desenvolvidas. Palavras-chave: Endodontia; Inteligência Artificial; Tratamento do Canal Radicular; Planejamento; Prognóstico. ABSTRACT Endodontic treatment requires thorough knowledge of root canal anatomy, and imaging exams are essential for planning and achieving a favorable prognosis. In this context, Artificial Intelligence (AI) has emerged as a promising tool for image analysis and clinical decision-making support. From this perspective, the present study aims to map the Artificial Intelligence tools available in the literature in order to assist in the planning and prognosis of endodontic treatment. This study will follow the scoping review methodology, in accordance with the standards and guidelines of the Joanna Briggs Institute (JBI) and the PRISMA-ScR checklist. The guiding research question was developed based on the PCC strategy: population of teeth undergoing endodontic treatment; use of Artificial Intelligence tools; and planning and prognosis of endodontic treatment. The search will be conducted in the databases PubMed via MEDLINE, Google Scholar, SciELO, Scopus, and Dentistry &amp; Oral Sciences Source (DoSS), using DeCS/MeSH descriptors combined with Boolean operators. The results will be synthesized in tables and informative charts. It is expected to identify several useful tools for daily endodontic practice, as well as to highlight knowledge gaps to be addressed. Keywords: Artificial Intelligence; Endodontics; Planning; Prognosis; Root Canal Therapy. INTRODUÇÃO O tratamento endodôntico tem como finalidade principal a descontaminação do sistema de canais radiculares infectados por microrganismos responsáveis pela formação das lesões apicais¹. Todavia, para a realização do preparo químico-mecânico do sistema de canais, é necessário um conhecimento prévio da anatomia dos canais radiculares do elemento dentário. Isso porque cada elemento dentário apresenta uma anatomia radicular característica e, apesar do conhecimento padrão já estabelecido, podem ocorrer variações anatômicas que dificultam o tratamento endodôntico, as quais podem ser identificadas radiograficamente². Para a realização de um tratamento endodôntico eficaz, exames de imagem são necessários, incluindo radiografia periapical, panorâmica e tomografia computadorizada de feixe cônico³. A partir da realização desses exames radiográficos, é possível identificar a anatomia dos canais radiculares de cada elemento dentário, bem como suas variações anatômicas, tais como canais laterais, ístimo e ramificações apicais, os quais podem dificultar o tratamento endodôntico. Além disso, por meio dos exames de imagem utilizados no tratamento de canais r","author":[{"family":"Dos Santos Sol","given":"Isabella"},{"family":"De Castro De Souza","given":"Maria"},{"family":"De Castro","given":"Antonio"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/497wc","URL":"https://doi.org/10.17605/osf.io/497wc","source":"datacite"},{"id":"doi:10.17605/osf.io/jxbms","type":"article-journal","title":"Artificial Intelligence and Radiomics for Differentiating Pseudoprogression from True Progression in High-Grade Gliomas: A Meta-Analysis","abstract":"Differentiating true progression (TP) from pseudoprogression (PsP) in high-grade gliomas (HGGs) on MRI is a critical clinical challenge. This meta-analysis evaluates the overall diagnostic accuracy of radiomics and artificial intelligence (AI) models to identify algorithmic determinants of optimal performance. A PRISMA-compliant search of PubMed, Ovid MEDLINE, and EMBASE (up to 2025) was conducted. Quality was assessed via QUADAS-2. Diagnostic metrics were pooled utilizing bivariate random-effects models and Summary Receiver Operating Characteristic (SROC) curves. Across 34 included studies, the overall pooled sensitivity and specificity were 82% and 79%, respectively. Shape-based and Deep Learning (DL) features achieved the highest Area Under the Curve (up to 0.96 and 0.95). First-order statistics and Gray-Level Co-occurrence Matrix (GLCM) yielded the highest consistency and accuracy peaks (up to 98%). Multiparametric MRI was the most robust and widely used imaging modality. Radiomics demonstrates high, reliable diagnostic potential for discriminating HGG PsP from TP. While advanced DL models and shape features maximize discriminative performance, classical handcrafted features remain the most extensively validated. Future studies must prioritize independent external validation, federated learning (FL), and Explainable AI (XAI) for clinical translation.","author":[{"family":"Facchinetti","given":"Giovanni"},{"family":"De Maria","given":"Lucio"},{"family":"Pagani","given":"Nicola"},{"family":"Ponzio","given":"Francesco"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/jxbms","URL":"https://doi.org/10.17605/osf.io/jxbms","source":"datacite"},{"id":"doi:10.17605/osf.io/nkbsf","type":"article-journal","title":"A Scoping Review of Actionable Recommendations for the Deployment of Artificial Intelligence in Healthcare: Protocol","abstract":"This pre-registered scoping review systematically maps actionable recommendations for AI deployment in healthcare across the full deployment lifecycle. Following JBI methodology and PRISMA-ScR standards, we search MEDLINE (via PubMed) and IEEE Xplore (January 2020 to search date), alongside a purposive systematic grey literature search covering regulatory authorities, policy bodies, and professional societies across Europe, the UK, and selected countries globally. Recommendations are mapped against the Compass AI deployment lifecycle framework, which organises deployment into six phases across three stages: Before going live (Needs Assessment and Procurement; Local Performance Testing); Going live (Clinical Workflow Integration; User Training); and Sustaining (Ongoing Monitoring and Maintenance; Decommissioning). Each extracted recommendation is assigned a deployment phase, specificity level (high / medium / low), recommendation type, target stakeholder, and evidence basis. The review will identify which phases have comprehensive versus sparse guidance, what proportion of recommendations are immediately implementable versus aspirational, which stakeholders are underrepresented, and what practical tools exist to support deployment. All data and materials will be made publicly available via this OSF repository.","author":[{"family":"Ossa","given":"Laura"},{"family":"Fehr","given":"Jana"},{"family":"Madai","given":"Vince"},{"family":"Topff","given":"Laurens"},{"family":"Zullino","given":"Sara"},{"family":"Beets-Tan","given":"Regina"},{"family":"Gordebeke","given":"Peter"},{"family":"Bruni","given":"Margherita"},{"family":"Rodriguez","given":"Pablo"},{"family":"Cerda","given":"Leonor"},{"family":"Van Diest","given":"Paul"},{"family":"Stockheim","given":"Jessica"},{"family":"Gumbs","given":"Andrew"},{"family":"Philippe","given":"Olivier"},{"family":"Polónia","given":"António"},{"family":"Patarnello","given":"Stefano"},{"family":"Bayarri","given":"Angel"},{"family":"Rosas","given":"Claudia"},{"family":"Lekadir","given":"Karim"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/nkbsf","URL":"https://doi.org/10.17605/osf.io/nkbsf","source":"datacite"},{"id":"doi:10.48550/arxiv.2603.19512","type":"manuscript","title":"FedAgain: A Trust-Based and Robust Federated Learning Strategy for an Automated Kidney Stone Identification in Ureteroscopy","abstract":"The reliability of artificial intelligence (AI) in medical imaging critically depends on its robustness to heterogeneous and corrupted images acquired with diverse devices across different hospitals which is highly challenging. Therefore, this paper introduces FedAgain, a trust-based Federated Learning (Federated Learning) strategy designed to enhance robustness and generalization for automated kidney stone identification from endoscopic images. FedAgain integrates a dual trust mechanism that combines benchmark reliability and model divergence to dynamically weight client contributions, mitigating the impact of noisy or adversarial updates during aggregation. The framework enables the training of collaborative models across multiple institutions while preserving data privacy and promoting stable convergence under real-world conditions. Extensive experiments across five datasets, including two canonical benchmarks (MNIST and CIFAR-10), two private multi-institutional kidney stone datasets, and one public dataset (MyStone), demonstrate that FedAgain consistently outperforms standard Federated Learning baselines under non-identically and independently distributed (non-IID) data and corrupted-client scenarios. By maintaining diagnostic accuracy and performance stability under varying conditions, FedAgain represents a practical advance toward reliable, privacy-preserving, and clinically deployable federated AI for medical imaging.","author":[{"family":"Reyes-Amezcua","given":"Ivan"},{"family":"Lopez-Tiro","given":"Francisco"},{"family":"Larose","given":"Clément"},{"family":"Daul","given":"Christian"},{"family":"Mendez-Vazquez","given":"Andres"},{"family":"Ochoa-Ruiz","given":"Gilberto"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2603.19512","URL":"https://doi.org/10.48550/arxiv.2603.19512","source":"datacite"},{"id":"doi:10.17605/osf.io/bg76j","type":"article-journal","title":"Ethical Governance and Leadership in the Integration of AI and Big Data within 'One Health', Veterinary Medicine, and Sustainable Agriculture Curricula: A Scoping Review of Multidisciplinary Educational Frameworks","abstract":"This scoping review investigates the intersection of Ethical Governance, Digital Leadership, and Multidisciplinary Education within the 'One Health' framework. As Artificial Intelligence (AI) and Big Data analytics become central to global health and food systems, there is a critical need for standardized pedagogical frameworks that ensure these technologies are integrated responsibly. The purpose of this research is to: Map the global landscape of educational models that combine AI/Big Data with Human Health, Veterinary Medicine, and Sustainable Agriculture. Identify the Ethical Governance structures required to lead multidisciplinary academic teams in high-velocity data environments. Synthesize leadership strategies that address algorithmic transparency, data privacy, and cross-disciplinary communication. Expected Outcomes: The review will produce a comprehensive taxonomy of leadership competencies and a conceptual framework for ethical AI integration. This will serve as a strategic roadmap for Higher Education leaders to prepare future professionals for the complex, data-driven challenges of zoonotic disease management, food security, and environmental sustainability.","author":[{"family":"Ismaile","given":"Samantha"},{"family":"Alhosban","given":"Fuad"},{"family":"Seboussi","given":"Rabiha"},{"family":"Abdallah","given":"Hanen"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/bg76j","URL":"https://doi.org/10.17605/osf.io/bg76j","source":"datacite"},{"id":"doi:10.48550/arxiv.2603.05884","type":"manuscript","title":"Computational Pathology in the Era of Emerging Foundation and Agentic AI -- International Expert Perspectives on Clinical Integration and Translational Readiness","abstract":"Recent breakthroughs in artificial intelligence through foundation models and agents have accelerated the evolution of computational pathology. Demonstrated performance gains reported across academia in benchmarking datasets in predictive tasks such as diagnosis, prognosis, and treatment response have ignited substantial enthusiasm for clinical application. Despite this development momentum, real world adoption has lagged, as implementation faces economic, technical, and administrative challenges. Beyond existing discussions of technical architectures and comparative performance, this review considers how these emerging AI systems can be responsibly integrated into medical practice by connecting deployable clinical relevance with downstream analytical capabilities and their technical maturity, operational readiness, and economic and regulatory context. Drawing on perspectives from an international group, we provide a practical assessment of current capabilities and barriers to adoption in patient care settings.","author":[{"family":"Da","given":"Qian"},{"family":"Chen","given":"Yijiang"},{"family":"Ju","given":"Min"},{"family":"Ji","given":"Zheyi"},{"family":"Zhou","given":"Albert"},{"family":"Wang","given":"Wenwen"},{"family":"Abikenari","given":"Matthew"},{"family":"Chikontwe","given":"Philip"},{"family":"Larghero","given":"Guillaume"},{"family":"Chen","given":"Bowen"},{"family":"Neidlinger","given":"Peter"},{"family":"Zhong","given":"Dingrong"},{"family":"Wang","given":"Shuhao"},{"family":"Xu","given":"Wei"},{"family":"Williamson","given":"Drew"},{"family":"Corredor","given":"German"},{"family":"Yang","given":"Sen"},{"family":"Lu","given":"Le"},{"family":"Han","given":"Xiao"},{"family":"Yu","given":"Kun"},{"family":"Huang","given":"Jun"},{"family":"Barisoni","given":"Laura"},{"family":"Litjens","given":"Geert"},{"family":"Madabhushi","given":"Anant"},{"family":"Zhu","given":"Lifeng"},{"family":"Wang","given":"Chaofu"},{"family":"Zhao","given":"Junhan"},{"family":"Hu","given":"Weiguo"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2603.05884","URL":"https://doi.org/10.48550/arxiv.2603.05884","source":"datacite"},{"id":"doi:10.17605/osf.io/f84rx","type":"article-journal","title":"Artificial Intelligence-Enhanced Electrocardiography for the Diagnosis of Arrhythmogenic Right Ventricular Cardiomyopathy: a Scoping Review","abstract":"Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a rare but potentially life-threatening cardiomyopathy and a leading cause of sudden cardiac death among young individuals and athletes. The diagnosis of ARVC encompasses electrocardiographic (ECG) and morphofunctional criteria, often requiring cardiac magnetic resonance (CMR) imaging, which is not routinely applicable or widely available, particularly in developing nations, and may be subtle during the early or concealed phases of the disease. ECG abnormalities may also be covert and nonspecific. In addition, conventional ECG interpretation is subject to interobserver variability. Recent advances in artificial intelligence (AI), particularly deep learning–based ECG analysis, have demonstrated improved diagnostic performance in detecting complex and subtle cardiac abnormalities. This scoping review aims to systematically map and summarize current evidence on the application, methodology, and diagnostic performance of AI-enhanced ECG for the detection of ARVC, identifying existing advances, limitations, and gaps in the literature.","author":[{"family":"Maharani","given":"Erika"},{"family":"Zahrani","given":"Sania"},{"family":"Prof Dr Dr Lucia Kris Dinarti","given":"Sppd"},{"family":"Prof Dr Dr Yoga Yuniadi","given":"Spjp"},{"family":"Dr Dyah Wulan Anggrahini","given":"Ph"},{"family":"Setiawan","given":"Noor"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/f84rx","URL":"https://doi.org/10.17605/osf.io/f84rx","source":"datacite"},{"id":"doi:10.17605/osf.io/z84xw","type":"article-journal","title":"The implementation of digital interventions to address alcohol use in primary care: a scoping review","abstract":"The advances in technology have significantly transformed healthcare, with digital health playing a pivotal role in improving patient outcomes. These innovations, such as mobile apps, wearables, and telemedicine, offer benefits like enhanced accessibility, real-time health monitoring, and early detection of medical issues. Digital health interventions empower patients to take an active role in their care, improving engagement and adherence to treatment. Such advances in the digital healthcare field call for further consideration using an implementation science approach to ensure this research can become integrated into standard medical care. An implementation science framework focuses on understanding the factors between clinical development and research and its eventual use as a medical tool. Given the increasing prevalence of alcohol-related harm and the potential for primary care providers to play a key role in identifying and addressing alcohol use, digital interventions offer an innovative and scalable solution for improving care delivery. Nair et al. 2015 provided valuable insights into the effectiveness and challenges of digital alcohol interventions in primary care settings. However, given the rapid advancements in technology over the past decade, the findings of this review have become outdated. Innovations such as the integration of artificial intelligence, machine learning, and wearable devices have significantly expanded the scope and capabilities of digital interventions. Additionally, the widespread adoption of telemedicine and mobile health applications has transformed patient care, improving accessibility and engagement. These technological developments necessitate a scoping review to capture the latest evidence and assess the implementation science of digital tools in healthcare. This review aims to: 1. Highlight implementation factors: Understanding the factors that influence the adoption and use of digital interventions in primary care—such as healthcare provider involvement, patient preferences, and technological barriers—will provide insights into the challenges and opportunities for integration into existing care models. 2. Identify and categorize digital interventions: By mapping the various types of digital tools available, the review will highlight the breadth of interventions used to address alcohol use in primary care, including apps, telehealth platforms, and online counseling. 3. Assess effectiveness and feasibility: The review will evaluate how well these digital interventions work in primary care settings, considering clinical outcomes, user engagement, and their potential for broad implementation within busy healthcare environments. 4. Identify gaps in the evidence: By synthesizing the current literature, the review will uncover areas where evidence is lacking, guiding future research efforts to strengthen the understanding of digital interventions' role in alcohol use management. This review seeks to inform healthcare providers, policymakers, and researchers about the current implementation landscape of digital interventions in alcohol use management, with the goal of improving the effectiveness, accessibility, and integration of these interventions into primary care practice.","author":[{"family":"Becker","given":"Sean"},{"family":"Edelman","given":"EJ"},{"family":"Kiluk","given":"Brian"},{"family":"Nair","given":"Shreyas"},{"family":"Bebarta","given":"Emma"},{"family":"Middya","given":"Ayesha"},{"family":"Grimshaw","given":"Alyssa"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/z84xw","URL":"https://doi.org/10.17605/osf.io/z84xw","source":"datacite"},{"id":"doi:10.5281/zenodo.18887718","type":"article-journal","title":"Medical Technologies: Pharmaceutical Cosmetology, Cosmeceuticals – Opportunities and Limitations","abstract":"Introduction. Pharmaceutical cosmetology is an interdisciplinary field at the junction of pharmacy, dermatology, and aesthetic medicine. The active development of cosmeceuticals is associated with the growing demand for products that combine a cosmetic effect with a biological effect on the skin. At the same time, the lack of a single regulatory status and mandatory randomized clinical trials creates several scientific and legal challenges. Purpose. To analyze the possibilities and limitations of cosmeceuticals in pharmaceutical cosmetology from the standpoint of modern medical technologies, evidence-based medicine, and pharmaceutical care. Methods. A systematic literature review was conducted using the PRISMA protocol using the PubMed, Scopus, Web of Science and Cochrane Library databases for 2015–2026. Out of 1243 publications, 62 sources with evidence levels I–II according to the Oxford Centre for Evidence-Based Medicine were selected after screening. Results. The article summarizes modern technological platforms of cosmeceuticals, including nanoemulsions, liposomes, multiple emulsions, and probiotic complexes, which increase the bioavailability of active ingredients by 3-5 times. The main groups of cosmeceutical active ingredients and their clinical efficacy in photoaging, acne, hyperpigmentation and alopecia are systematized. The importance of pharmaceutical care and consulting algorithms for increasing the safety of cosmeceutical use is shown. The prospects for personalized cosmeceuticals based on genetic tests and artificial intelligence are identified. Regulatory gaps in the USA, EU and Ukraine are outlined. Conclusions. Cosmeceuticals are a promising direction of integration of pharmacy, dermatology, and digital technologies, which demonstrates proven effectiveness in mild forms of dermatoses and prevention of photoaging. Further development requires standardization of the term, development of national regulatory approaches and implementation of pharmacovigilance systems.","author":[{"family":"Shapovalova","given":"Viktoriia"},{"family":"Osyntseva","given":"Alina"},{"family":"Shapovalov","given":"Valentyn"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18887718","URL":"https://doi.org/10.5281/zenodo.18887718","source":"datacite"},{"id":"doi:10.5281/zenodo.18887719","type":"article-journal","title":"Medical Technologies: Pharmaceutical Cosmetology, Cosmeceuticals – Opportunities and Limitations","abstract":"Introduction. Pharmaceutical cosmetology is an interdisciplinary field at the junction of pharmacy, dermatology, and aesthetic medicine. The active development of cosmeceuticals is associated with the growing demand for products that combine a cosmetic effect with a biological effect on the skin. At the same time, the lack of a single regulatory status and mandatory randomized clinical trials creates several scientific and legal challenges. Purpose. To analyze the possibilities and limitations of cosmeceuticals in pharmaceutical cosmetology from the standpoint of modern medical technologies, evidence-based medicine, and pharmaceutical care. Methods. A systematic literature review was conducted using the PRISMA protocol using the PubMed, Scopus, Web of Science and Cochrane Library databases for 2015–2026. Out of 1243 publications, 62 sources with evidence levels I–II according to the Oxford Centre for Evidence-Based Medicine were selected after screening. Results. The article summarizes modern technological platforms of cosmeceuticals, including nanoemulsions, liposomes, multiple emulsions, and probiotic complexes, which increase the bioavailability of active ingredients by 3-5 times. The main groups of cosmeceutical active ingredients and their clinical efficacy in photoaging, acne, hyperpigmentation and alopecia are systematized. The importance of pharmaceutical care and consulting algorithms for increasing the safety of cosmeceutical use is shown. The prospects for personalized cosmeceuticals based on genetic tests and artificial intelligence are identified. Regulatory gaps in the USA, EU and Ukraine are outlined. Conclusions. Cosmeceuticals are a promising direction of integration of pharmacy, dermatology, and digital technologies, which demonstrates proven effectiveness in mild forms of dermatoses and prevention of photoaging. Further development requires standardization of the term, development of national regulatory approaches and implementation of pharmacovigilance systems.","author":[{"family":"Shapovalova","given":"Viktoriia"},{"family":"Osyntseva","given":"Alina"},{"family":"Shapovalov","given":"Valentyn"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18887719","URL":"https://doi.org/10.5281/zenodo.18887719","source":"datacite"},{"id":"doi:10.17605/osf.io/ph2ad","type":"article-journal","title":"Application of Artificial Intelligence in Radiological Exams as a Medical Education Tool: A Scoping Review Protocol","abstract":"This study consists of a scoping review aimed at mapping and analyzing the use of Artificial Intelligence in radiology as a tool for medical education. The review seeks to identify the main strategies for using Artificial Intelligence in radiology teaching during medical training, as well as to recognize gaps related to its absence or limited integration into traditional educational models. The review will be conducted in accordance with the methodological guidelines of the Joanna Briggs Institute (JBI) for scoping reviews. The reporting of results will follow the recommendations of PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses – Scoping Review Extension). Studies published within the last five years, available in full text, in Portuguese, English, and Spanish, will be included. Searches will be performed in the following databases: PubMed, LILACS, Cochrane Library, Scopus, and Web of Science. The temporal scope was defined considering the rapid technological advancement of Artificial Intelligence in healthcare and medical education, particularly following recent developments in machine learning and generative AI from 2021 onward, aiming to capture the most current and relevant evidence. As this review follows JBI methodological guidance for scoping reviews, a formal risk of bias assessment will not be conducted, since the objective of this study is to map the available evidence rather than assess effectiveness.","author":[{"family":"Hikigi","given":"Juliana"},{"family":"Fernandes","given":"Isabela"},{"family":"Borges","given":"Maria"},{"family":"Monteiro","given":"Aline"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/ph2ad","URL":"https://doi.org/10.17605/osf.io/ph2ad","source":"datacite"},{"id":"doi:10.5281/zenodo.18742013","type":"article-journal","title":"mailcom: Pseudonymization Tool for Textual Data","abstract":"The rapid growth of data and its usage by Artificial Intelligence applications leads to heightened concerns about data privacy. Researchers often need to analyze datasets that contain personal information, sometimes paired with sensitive attributes such as medical records or political views. To support such analyses without exposing identifiable content, the Scientific Software Center (SSC) of Heidelberg University developed the mailcom package for pseudonymization. This capability is especially important when employing web-hosted Large Language Models for downstream analysis. As a use case, we applied mailcom to a multilingual email corpus in Spanish, French, and Portuguese contributed by multiple donors as part of a pilot study, in collaboration with the research group of Sybille Große (Department of Romance Studies, Heidelberg University). To protect donor privacy, sensitive information such as names, email addresses, and numbers is extracted and pseudonymized. The package processes text from email subjects and bodies in eml and html formats, as well as from csv rows, making it applicable to a wide range of textual data beyond email. mailcom is built entirely on open-source libraries and is designed for configurability and extensibility. Its core features are: (i) language identification, (ii) named-entity recognition, (iii) extraction of temporal expressions, and (iv) de-identifying sensitive data via pseudonyms. Three aforementioned languages are supported by default, with options to add further languages and change back-end libraries via configuration. We present these features in end-to-end processing pipelines using examples from our use case. The main parts include: (1) General workflow from raw text to pseudonymized output,(2) Default libraries and techniques (e.g. eml-parser, spaCy, langid, langdetect, transformers, and rule-based)(3) Mechanisms for adapting to new languages, transformer pipelines, and spaCy models with minimal effort. Since pseudonymized outputs still require human review to guarantee full anonymization, the package serves as a scalable pre-processing layer that reduces manual work while establishing a principled baseline of privacy protection. This reproducible, privacy-aware tool enables empirical research on digital text under current data-ethics and governance standards.","author":[{"family":"Le","given":"Kim"},{"family":"Gärtner","given":"Laura"},{"family":"Fleischle","given":"Felix"},{"family":"Schoeller","given":"Thore"},{"family":"Große","given":"Sybille"},{"family":"Ulusoy","given":"Inga"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18742013","URL":"https://doi.org/10.5281/zenodo.18742013","source":"datacite"},{"id":"doi:10.5281/zenodo.18742014","type":"article-journal","title":"mailcom: Pseudonymization Tool for Textual Data","abstract":"The rapid growth of data and its usage by Artificial Intelligence applications leads to heightened concerns about data privacy. Researchers often need to analyze datasets that contain personal information, sometimes paired with sensitive attributes such as medical records or political views. To support such analyses without exposing identifiable content, the Scientific Software Center (SSC) of Heidelberg University developed the mailcom package for pseudonymization. This capability is especially important when employing web-hosted Large Language Models for downstream analysis. As a use case, we applied mailcom to a multilingual email corpus in Spanish, French, and Portuguese contributed by multiple donors as part of a pilot study, in collaboration with the research group of Sybille Große (Department of Romance Studies, Heidelberg University). To protect donor privacy, sensitive information such as names, email addresses, and numbers is extracted and pseudonymized. The package processes text from email subjects and bodies in eml and html formats, as well as from csv rows, making it applicable to a wide range of textual data beyond email. mailcom is built entirely on open-source libraries and is designed for configurability and extensibility. Its core features are: (i) language identification, (ii) named-entity recognition, (iii) extraction of temporal expressions, and (iv) de-identifying sensitive data via pseudonyms. Three aforementioned languages are supported by default, with options to add further languages and change back-end libraries via configuration. We present these features in end-to-end processing pipelines using examples from our use case. The main parts include: (1) General workflow from raw text to pseudonymized output,(2) Default libraries and techniques (e.g. eml-parser, spaCy, langid, langdetect, transformers, and rule-based)(3) Mechanisms for adapting to new languages, transformer pipelines, and spaCy models with minimal effort. Since pseudonymized outputs still require human review to guarantee full anonymization, the package serves as a scalable pre-processing layer that reduces manual work while establishing a principled baseline of privacy protection. This reproducible, privacy-aware tool enables empirical research on digital text under current data-ethics and governance standards.","author":[{"family":"Le","given":"Kim"},{"family":"Gärtner","given":"Laura"},{"family":"Fleischle","given":"Felix"},{"family":"Schoeller","given":"Thore"},{"family":"Große","given":"Sybille"},{"family":"Ulusoy","given":"Inga"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18742014","URL":"https://doi.org/10.5281/zenodo.18742014","source":"datacite"},{"id":"doi:10.17605/osf.io/dj7ym","type":"article-journal","title":"Artificial intelligence for informed decision-making during clinical trial consent: A scoping review of applications, outcomes, and ethical considerations","abstract":"Clinical trials are essential for advancing medical treatments but face persistent challenges including slow recruitment, high costs, and complex regulatory demands. Artificial intelligence (AI), particularly large generative AI, has emerged as a potential tool to improve efficiency across multiple stages of trial design and conduct. One ethically sensitive domain is the informed consent process, where AI may support comprehension through document generation, conversational interfaces, and personalized explanations. Consent process in clinical trials is a longitudinal process, which includes information disclosure, decision supports, comprehension, dynamic conversations, and documentations. While early studies suggest potential benefits, concerns remain regarding safety, biases, and undue influence. The literature is rapidly expanding but fragmented across empirical studies and ethical analyses, warranting a scoping review to map the field.","author":[{"family":"Lee","given":"Seung"},{"family":"Yahya","given":"Ayesha"},{"family":"Serpico","given":"Kimberley"},{"family":"Yeh","given":"Natalie"},{"family":"Amin","given":"Amina"},{"family":"Bell","given":"Jennifer"},{"family":"Kieran","given":"Quinn"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/dj7ym","URL":"https://doi.org/10.17605/osf.io/dj7ym","source":"datacite"},{"id":"doi:10.17605/osf.io/mvtw9","type":"article-journal","title":"ORACLES-AI - ORal Annotated Clinical Lesion Evaluation dataset for Artificial Intelligence","abstract":"ORACLES-AI (ORal Annotated Clinical Lesion Evaluation dataset for Artificial Intelligence) is an annotated, image-based dataset developed to advance research, education, and computational modelling in the domain of oral mucosal health and disease. The dataset comprises high-quality intraoral clinical images collected from participants representing a broad spectrum of oral mucosal conditions, ranging from normal mucosa and variations from normal to Oral Potentially Malignant Disorders (OPMDs) and oral cancers. Each image is accompanied by expert-generated annotations and extensive, non-identifiable metadata, making ORACLES-AI a comprehensive and interpretable resource for both clinical and computational research applications. Data Collection and Methodology Participants were prospectively recruited, and intraoral images were acquired using standardized mobile phone–based intraoral photography protocols. Participants were enrolled from one of the spoke centers operating under a Hub-and-Spoke model established as part of an Indian Council of Medical Research (ICMR)–funded study. In this model, Ragas Dental College and Hospital, Chennai, functions as the central Hub, while private dental institutions and non-governmental organizations across different regions of India serve as Spokes responsible for data collection and community-level oral screening. The present dataset represents images and annotations obtained from one such Spoke center, collected in accordance with uniform methodological guidelines defined by the Hub institution. All intraoral images were captured following a copyrighted Standard Operating Procedure (SOP) for intraoral photography developed by the Hub to ensure consistency, reproducibility, and diagnostic usability across participating centers. Prior to imaging, camera lenses were cleaned, and photographs were taken at an approximate distance of 4–5 cm from the oral cavity, primarily under natural lighting conditions, with auxiliary lighting used when necessary. Retraction aids, including mouth mirrors or wooden retractors, were employed to optimize visualization. For each participant, eight standard intraoral sites were systematically photographed: dorsal tongue, ventral tongue, right buccal mucosa, left buccal mucosa, upper labial mucosa, lower labial mucosa, maxillary arch, and mandibular arch. Image quality was assessed based on predefined criteria, including centering, illumination, sharpness, and absence of motion blur or extraneous artifacts, and images not meeting these criteria were reacquired. The dataset is designed to be dynamic and will be periodically updated with additional participants and newly annotated images to enhance its coverage, diversity, and clinical representativeness. Annotation and Regional Attributes All images underwent expert review and annotation by Oral Pathologist and Public Health Dentistry specialists using the VGG Image Annotator (VIA), version 3.0.13. Regions of Interest (ROIs) were delineated using polygonal annotations to define lesion boundaries, mucosal sub-sites, and relevant diagnostic features such as surface texture, color variation, and border irregularity. Annotations were stored in JSON format to ensure compatibility with deep learning and computer vision frameworks. Each image file follows a structured naming convention in the format A_B_C.jpeg, where A represents a unique anonymized participant identifier, B denotes the site of data collection, and C specifies the intraoral region (e.g., DT for dorsal tongue, LB for left buccal mucosa). Data Structure The dataset is organized into four primary diagnostic categories: (1) normal mucosa, representing healthy oral tissues; (2) variations from normal, encompassing minor deviations from typical mucosal appearance; (3) Oral Potentially Malignant Disorders (OPMDs) and (4) oral cancer, comprising histopathologically confirmed squamous cell carcinoma cases. Each image is linked to its corresponding JSON annotation file ","author":[{"family":"Kumar","given":"Dr"},{"family":"Kannan","given":"Dr"},{"family":"Chattopadhyay","given":"Dr"},{"family":"Shetty","given":"Dr"},{"family":"Sen","given":"Dr"},{"family":"Bose","given":"Dr"},{"family":"Rajeshwari","given":"Dr"}],"issued":{"date-parts":[[2025]]},"DOI":"10.17605/osf.io/mvtw9","URL":"https://doi.org/10.17605/osf.io/mvtw9","source":"datacite"},{"id":"doi:10.17605/osf.io/4bzmh","type":"article-journal","title":"Pedagogical use of artificial intelligence in health sciences education: a systematic review of faculty perceptions and experiences","abstract":"This project is a systematic review on the pedagogical use of artificial intelligence (AI) in health sciences education, focusing on university faculty perceptions and experiences. The aim is to synthesize empirical evidence on how faculty conceptualize AI as an educational tool, how they describe its integration into teaching practices, and which benefits, challenges, and ethical concerns they report. The review targets health sciences programs within higher education settings and will use a structured search across relevant databases, applying predefined inclusion and exclusion criteria. The study will follow PRISMA reporting guidelines and will be organized according to the SALSA framework (Search, Appraisal, Synthesis, Analysis), conducting a narrative/thematic synthesis of the findings. This synthesis is intended to inform faculty development initiatives and institutional policies for a pedagogically robust and ethically responsible integration of AI in health sciences education.","author":[{"family":"Quidel","given":"Juan"},{"family":"Lindín","given":"Carles"},{"family":"Parcerisa","given":"Lluís"},{"family":"I Valero","given":"Joan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.17605/osf.io/4bzmh","URL":"https://doi.org/10.17605/osf.io/4bzmh","source":"datacite"},{"id":"doi:10.17605/osf.io/zxqcv","type":"article-journal","title":"Accuracy of Generative Artificial Intelligence in Risk-of-Bias Assessment using RoB 2.0 Tool and Extracting Data in Exercise Therapy for Chronic Low Back Pain RCTs","abstract":"The production of medical and scientific research has grown exponentially since the first medical articles were published. The current pyramid of evidence-based medicine represents the tip of the research iceberg, and the achievement of next-generation “deep” EBP requires synthesis and deep merging of all available data in the literature. For this reason, conducting systematic reviews and meta-analyses has become a necessity to more accurately and comprehensively present accumulated knowledge to a scientific, clinical, and general audience. First-level evidence can take a large amount of time and effort to complete, with an estimated mean time to accomplish a systematic review of 15 months, and the methodological standards for performing them have increased over time. Moreover, tools that meet these standards, such as RoB 2.0, are detailed and comprehensive, but at the same time can be challenging even for reviewers with considerable experience. For this reason, as mountains of unsynthesised research evidence accumulate, it is necessary to improve the tools for collecting, filtering, and synthesizing these evidences. In this way, the tasks required to develop a literature synthesis could be facilitated by the application of artificial intelligence (AI), which may be regarded as the development of algorithms that aim to simulate human intelligence, thereby allowing computers to perform tasks requiring human cognitive skills. ChatGPT can be described as a promising or even a revolutionary tool for scientific research in both academic writing and in the research process itself. It was listed in several sources as an efficient and promising tool for conducting comprehensive literature reviews and generating computer codes, thereby saving time for the research steps that require more effort from human intelligence. One of the most advanced recent models, GPT‑4o, released by OpenAI in May 2024, is a Large Multimodal Model (LMM), capable of processing diverse data modalities, such as text, images, audio, and video. The presented research project wants to evaluate the accuracy of ChatGPT in the assessment of Risk-of-Bias with RoB 2.0 tool and in the data extraction in musculoskeletal physiotherapy RCTs. For RoB assessment, a tailored prompt will be developed and applied to guide ChatGPT-4o in conducting an independent assessment using the RoB 2.0 tool. ChatGPT-4o’s evaluations will be compared item-by-item with baseline human assessments obtained in a previous study to determine agreement and reliability. For data extraction, human reviewers will independently perform a double-blinded extraction of the dataset. In parallel, two distinct prompts will be designed for ChatGPT-4o to extract the same information. AI-outputs extracted in this way will be subsequently compared with human extractions to evaluate their consistency and accuracy.","author":[{"family":"Carrer","given":"Luca"},{"family":"Pozzati","given":"Marco"},{"family":"Taborelli","given":"Dario"},{"family":"Maschi","given":"Niccolò"},{"family":"Innocenti","given":"Tiziano"},{"family":"Salvioli","given":"Stefano"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/zxqcv","URL":"https://doi.org/10.17605/osf.io/zxqcv","source":"datacite"},{"id":"doi:10.5281/zenodo.18802220","type":"article-journal","title":"PHARMACEUTICAL SCIENCE IN THE ERA OFARTIFICIAL INTELLIGENCE","abstract":"Purpose: This review discovers the transformative impression of Artificial Intelligence (AI) and Machine Learning (ML) on the pharmaceutical industry and healthcare delivery, focusing on drug discovery, clinical pharmacy practice, and operational productivity. Methods: The article inspects the historical development of AI, from early neural models to modern deep learning designs like GANs, RNNs, and Transformers, and assesses their specific requests transversely to the drug life span. Results: AI is publicised to significantly accelerate R&D by detecting drug leads quicker and adjusting clinical trials through patient-specific data analysis. In clinical settings, AI-driven choice support systems boost patient safety by reducing medication errors, predicting adverse reactions, and refining adherence—especially realising a 40% growth in adherence in community pharmacies. Still, technologies such as computer vision are restyling medicine supervision and analytical precision in medical imagination. Challenges: In spite of these benefits, the evolution characteristics sprints with \"black box\" interpretability, data privacy risks, algorithmic bias, and high implementation charges. Conclusion: This review highlights that despite the fact AI is redesigning pharmacy into an extra detailed and inventive field, its innocuous integration requires a specialised workforce exercise, strong governing agendas, and a constant emphasis on the vital social assembly in patient care.","author":[{"family":"Gambhir","given":"Dhairya"},{"family":"Tiwari","given":"Kanishek"},{"family":"Saini","given":"Govind"},{"family":"Dhakad","given":"Himanshu"},{"family":"Yadav","given":"Keshav"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18802220","URL":"https://doi.org/10.5281/zenodo.18802220","source":"datacite"},{"id":"doi:10.5281/zenodo.18802221","type":"article-journal","title":"PHARMACEUTICAL SCIENCE IN THE ERA OFARTIFICIAL INTELLIGENCE","abstract":"Purpose: This review discovers the transformative impression of Artificial Intelligence (AI) and Machine Learning (ML) on the pharmaceutical industry and healthcare delivery, focusing on drug discovery, clinical pharmacy practice, and operational productivity. Methods: The article inspects the historical development of AI, from early neural models to modern deep learning designs like GANs, RNNs, and Transformers, and assesses their specific requests transversely to the drug life span. Results: AI is publicised to significantly accelerate R&D by detecting drug leads quicker and adjusting clinical trials through patient-specific data analysis. In clinical settings, AI-driven choice support systems boost patient safety by reducing medication errors, predicting adverse reactions, and refining adherence—especially realising a 40% growth in adherence in community pharmacies. Still, technologies such as computer vision are restyling medicine supervision and analytical precision in medical imagination. Challenges: In spite of these benefits, the evolution characteristics sprints with \"black box\" interpretability, data privacy risks, algorithmic bias, and high implementation charges. Conclusion: This review highlights that despite the fact AI is redesigning pharmacy into an extra detailed and inventive field, its innocuous integration requires a specialised workforce exercise, strong governing agendas, and a constant emphasis on the vital social assembly in patient care.","author":[{"family":"Gambhir","given":"Dhairya"},{"family":"Tiwari","given":"Kanishek"},{"family":"Saini","given":"Govind"},{"family":"Dhakad","given":"Himanshu"},{"family":"Yadav","given":"Keshav"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18802221","URL":"https://doi.org/10.5281/zenodo.18802221","source":"datacite"},{"id":"doi:10.6084/m9.figshare.31429418","type":"article-journal","title":"Data augmentation strategies for GAN-based medical image analysis: an empirical review","abstract":"Medical image generation using artificial intelligence, especially Generative Adversarial Networks (GANs), has become a powerful solution to address challenges of limited data, class imbalance, and privacy restrictions in clinical imaging. While several GAN-based strategies have been proposed, earlier reviews often describe models in isolation without linking them clearly to medical datasets, evaluation measures, or diagnostic impact. This review focuses on GAN-based augmentation methods applied to medical imaging, systematically comparing prominent GAN families such as DCGAN, Pix2Pix, CycleGAN, StarGAN, and DualGAN across commonly used datasets, including BraTS, ISIC, DRIVE, and ADNI. Reported outcomes are synthesised using image-quality metrics (SSIM, PSNR, FID, LPIPS) and task-based measures (accuracy, sensitivity, Dice score). Findings suggest Pix2Pix frequently improves accuracy and SSIM on MRI and dermoscopy images, CycleGAN enhances sensitivity in retinal tasks, while newer models like StyleGAN and diffusion-based approaches achieve stronger perceptual fidelity. The review concludes that GAN-driven augmentation provides measurable benefits but is highly task- and dataset-dependent, emphasizing the future promise of hybrid GAN–diffusion pipelines and explainability tools for clinical adoption.","author":[{"family":"Dash","given":"Archana"},{"family":"Panigrahi","given":"Soumyarashmi"},{"family":"Adhikary","given":"Dibya"},{"family":"Swarnkar","given":"Tripti"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.31429418","URL":"https://doi.org/10.6084/m9.figshare.31429418","source":"datacite"},{"id":"doi:10.6084/m9.figshare.31429418.v1","type":"article-journal","title":"Data augmentation strategies for GAN-based medical image analysis: an empirical review","abstract":"Medical image generation using artificial intelligence, especially Generative Adversarial Networks (GANs), has become a powerful solution to address challenges of limited data, class imbalance, and privacy restrictions in clinical imaging. While several GAN-based strategies have been proposed, earlier reviews often describe models in isolation without linking them clearly to medical datasets, evaluation measures, or diagnostic impact. This review focuses on GAN-based augmentation methods applied to medical imaging, systematically comparing prominent GAN families such as DCGAN, Pix2Pix, CycleGAN, StarGAN, and DualGAN across commonly used datasets, including BraTS, ISIC, DRIVE, and ADNI. Reported outcomes are synthesised using image-quality metrics (SSIM, PSNR, FID, LPIPS) and task-based measures (accuracy, sensitivity, Dice score). Findings suggest Pix2Pix frequently improves accuracy and SSIM on MRI and dermoscopy images, CycleGAN enhances sensitivity in retinal tasks, while newer models like StyleGAN and diffusion-based approaches achieve stronger perceptual fidelity. The review concludes that GAN-driven augmentation provides measurable benefits but is highly task- and dataset-dependent, emphasizing the future promise of hybrid GAN–diffusion pipelines and explainability tools for clinical adoption.","author":[{"family":"Dash","given":"Archana"},{"family":"Panigrahi","given":"Soumyarashmi"},{"family":"Adhikary","given":"Dibya"},{"family":"Swarnkar","given":"Tripti"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.31429418.v1","URL":"https://doi.org/10.6084/m9.figshare.31429418.v1","source":"datacite"},{"id":"doi:10.17605/osf.io/cpv7j","type":"article-journal","title":"Analysis of technical nonconformities in health systems with LLMs: an interdisciplinary scoping review protocol","abstract":"This scoping review protocol outlines a methodological approach to investigate how Artificial Intelligence (AI), particularly Large Language Models (LLMs), is employed to support regulatory compliance and software quality within medical systems. The development of healthcare software requires strict adherence to complex global regulations (such as IEC 62304, FDA guidelines, and LGPD), making manual validation processes costly and time-consuming. Following the Joanna Briggs Institute (JBI) framework , this study will systematically search four major databases (IEEE Xplore, PubMed, Scopus, and Web of Science). The objective is to map current literature, identify existing AI-driven tools for source code and requirements analysis, and pinpoint knowledge gaps. Ultimately, this review aims to provide a scientific foundation for developing new solutions that mitigate regulatory bottlenecks and ensure patient safety during the medical software development lifecycle","author":[{"family":"Da Silva Júnior","given":"Klayton"},{"family":"Bublitz","given":"Frederico"},{"family":"De França Clemente E Rodrigues De Oliveira","given":"Rodolfo"},{"family":"De Barros Sales","given":"Débora"},{"family":"Vital","given":"José"},{"family":"Santos","given":"José"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/cpv7j","URL":"https://doi.org/10.17605/osf.io/cpv7j","source":"datacite"},{"id":"doi:10.17605/osf.io/8ysw6","type":"article-journal","title":"Healthcare Artificial Intelligence Policies in Australia and New Zealand","abstract":"Artificial intelligence (AI) technologies are rapidly transforming healthcare delivery globally. The pace of technological advancement is currently exceeding the speed of regulatory adaptation, creating a significant gap in published systematic evidence. Australia and New Zealand, as neighboring healthcare systems with shared professional bodies and similar regulatory approaches, provide an ideal comparative case study. Furthermore, the elevation of Indigenous data governance principles (honoring Te Tiriti o Waitangi and Aboriginal/Torres Strait Islander rights) is critical for ensuring cultural safety in AI systems.","author":[{"family":"Mordaunt","given":"Dylan"},{"family":"Palmer","given":"Lyle"},{"family":"Kirkpatrick","given":"Emily"},{"family":"Hosking","given":"Michael"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/8ysw6","URL":"https://doi.org/10.17605/osf.io/8ysw6","source":"datacite"},{"id":"doi:10.5281/zenodo.18736904","type":"article-journal","title":"Perceived Impact of Artificial Intelligence Usage on Academic Performance Among Undergraduate Health Science Students in Mirpurkhas, Sindh-Pakistan: A Multi-Center Study","abstract":"Background: Many medical and nursing students are increasingly utilizing artificial intelligence tools such as Telemedicine, Jeni.AI, Gemini, and ChatGPT to support both academic and clinical activities. These tools offer rapid responses and can save valuable time. However, there is growing concern that students in cities like Mirpurkhas may become overly reliant on AI, which could potentially compromise the development of critical thinking, problem-solving, and research skills. Methods: A descriptive cross-sectional study was conducted among undergraduate health science students at selected public and private centers in Mirpurkhas. Stratified random sampling was used to recruit a total of 197 participants, and data were collected via Google Forms from January to February 2026. An adapted questionnaire was employed, with minor modifications based on expert review to ensure validity. The collected data were analyzed using SPSS version 27.0, and descriptive statistics were generated for demographic subgroups. Findings: Among the 197 participants, the majority were BSN students (76.1%), aged 18–22 years (67.5%), and enrolled in private institutions (63%). Most students had a CGPA between 3.01–3.50. High perception scores were observed for General Use of AI (3.79 ± 0.40), Future of AI in Education (3.75 ± 0.46), and Impact on Student Skills (3.65 ± 0.48). Effectiveness of AI in Education was rated moderately high (3.49 ± 0.45), while Accuracy and Reliability received a moderate score (2.60 ± 0.50). These findings suggest that students generally find AI tools easy to use; however, they also express concerns about potential erosion of their natural skills and critical thinking. Conclusion: The healthcare undergraduate students demonstrated a high level of acceptance and utilization of artificial intelligence (AI) as a supportive academic tool. While AI use was generally associated with enhanced academic performance, concerns regarding accuracy and potential over-reliance remain. Implementation of structured training programs and clear ethical guidelines is recommended to ensure responsible and effective integration of AI in healthcare academia.","author":[{"family":"Ahmed","given":"Samia"},{"family":"Chandio","given":"Irfan"},{"family":"Pervaiz","given":"Anum"},{"family":"Fatima","given":"Aiman"},{"family":"Khadija"},{"family":"Malhi","given":"Sheela"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.18736904","URL":"https://doi.org/10.5281/zenodo.18736904","source":"datacite"},{"id":"doi:10.5281/zenodo.18736905","type":"article-journal","title":"Perceived Impact of Artificial Intelligence Usage on Academic Performance Among Undergraduate Health Science Students in Mirpurkhas, Sindh-Pakistan: A Multi-Center Study","abstract":"Background: Many medical and nursing students are increasingly utilizing artificial intelligence tools such as Telemedicine, Jeni.AI, Gemini, and ChatGPT to support both academic and clinical activities. These tools offer rapid responses and can save valuable time. However, there is growing concern that students in cities like Mirpurkhas may become overly reliant on AI, which could potentially compromise the development of critical thinking, problem-solving, and research skills. Methods: A descriptive cross-sectional study was conducted among undergraduate health science students at selected public and private centers in Mirpurkhas. Stratified random sampling was used to recruit a total of 197 participants, and data were collected via Google Forms from January to February 2026. An adapted questionnaire was employed, with minor modifications based on expert review to ensure validity. The collected data were analyzed using SPSS version 27.0, and descriptive statistics were generated for demographic subgroups. Findings: Among the 197 participants, the majority were BSN students (76.1%), aged 18–22 years (67.5%), and enrolled in private institutions (63%). Most students had a CGPA between 3.01–3.50. High perception scores were observed for General Use of AI (3.79 ± 0.40), Future of AI in Education (3.75 ± 0.46), and Impact on Student Skills (3.65 ± 0.48). Effectiveness of AI in Education was rated moderately high (3.49 ± 0.45), while Accuracy and Reliability received a moderate score (2.60 ± 0.50). These findings suggest that students generally find AI tools easy to use; however, they also express concerns about potential erosion of their natural skills and critical thinking. Conclusion: The healthcare undergraduate students demonstrated a high level of acceptance and utilization of artificial intelligence (AI) as a supportive academic tool. While AI use was generally associated with enhanced academic performance, concerns regarding accuracy and potential over-reliance remain. Implementation of structured training programs and clear ethical guidelines is recommended to ensure responsible and effective integration of AI in healthcare academia.","author":[{"family":"Ahmed","given":"Samia"},{"family":"Chandio","given":"Irfan"},{"family":"Pervaiz","given":"Anum"},{"family":"Fatima","given":"Aiman"},{"family":"Khadija"},{"family":"Malhi","given":"Sheela"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.18736905","URL":"https://doi.org/10.5281/zenodo.18736905","source":"datacite"},{"id":"doi:10.17605/osf.io/n7v2k","type":"article-journal","title":"Performance and Applicability of ChatGPT and Other Generative Artificial Intelligence on Hypertension: A Scoping Review","abstract":"This scoping review aims to map and synthesize the existing evidence on the performance and applicability of ChatGPT and other generative artificial intelligence (AI) tools in the context of hypertension. The review will examine how generative AI has been used across hypertension-related domains, including screening, diagnosis, risk stratification, patient education, clinical decision support, and health system applications. Additionally, it will explore reported accuracy, reliability, benefits, limitations, ethical considerations, and implementation challenges. By systematically identifying knowledge gaps and emerging trends, this scoping review seeks to inform clinicians, researchers, and policymakers about the current and potential roles of generative AI in hypertension care and research.","author":[{"family":"Gc","given":"Saroj"},{"family":"Maheshwari","given":"Dheeraj"},{"family":"Bhusal","given":"Pawan"},{"family":"Basnet","given":"Bibhusan"},{"family":"Basnet","given":"Roshan"},{"family":"Baruwal","given":"Ashma"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/n7v2k","URL":"https://doi.org/10.17605/osf.io/n7v2k","source":"datacite"},{"id":"doi:10.6084/m9.figshare.30287251.v1","type":"article-journal","title":"An audit of AI-related documents across U.S. medical schools: A framework-based qualitative content analysis","abstract":"Medical schools would benefit from systematic guidance for developing comprehensive artificial intelligence (AI) policies, given generative AI’s rapid integration into medical education. This study developed and applied an idealized AI policy framework to analyze AI-related documents at U.S. medical school institutions, providing reference points for the development and refinement of institutional policies. AI-related documents from institutions with U.S. allopathic and osteopathic medical schools were systematically collected (from August to October 2024) and analyzed using a comprehensive framework containing 24 subthemes across six themes: Background/Context, Governance, AI Literacy, Tools/Usage, Ethical/Legal Considerations, and Technology Support and Infrastructure. Publicly available online documents were systematically coded to generate framework subtheme scores indicating breadth of coverage across framework themes. AI-related documents retrieved from 73.7% (146/198) of U.S. medical school institutions covered an average of 8 of 24 subthemes, representing a mean framework coverage score of 32.3% ± 19.8 Rarely addressed subthemes included Audit and Compliance Mechanisms (6.8%, 10/146), Technical Infrastructure (6.2%, 9/146), and Environmental Stewardship (1.4%, 2/146). Academic Honesty and Plagiarism dominated AI-related documents (81.5%, 119/146), followed by Decision-Making Authority (54.1%, 79/146) and Critical Evaluation (52.1%, 76/146). Formal AI policies demonstrated significantly higher framework coverage than other AI document types (44.0% vs 30.4%, p = 0.003). Seven institutions with the highest coverage (≥13/24 subthemes) shared seven common distinguishing features, with six present universally. AI-related documents currently emphasize academic integrity over strategic planning, with substantial gaps in infrastructure and review mechanisms. Institutions can enhance their AI policies by incorporating common features identified in well-designed policies and following frameworks that strike a balance between immediate concerns and long-term adaptability.","author":[{"family":"Rush","given":"Emily"},{"family":"Byram","given":"Jessica"},{"family":"Garnett","given":"Colleen"},{"family":"Devaul","given":"Nicole"},{"family":"Smith","given":"Laura"},{"family":"Checchi","given":"Margaret"},{"family":"Martin","given":"Daniel"},{"family":"Hoffman","given":"Leslie"},{"family":"Brown","given":"Kirstin"},{"family":"Mumbower","given":"Daniel"},{"family":"Becker","given":"Robert"},{"family":"Roach","given":"Victoria"},{"family":"Doubleday","given":"Alison"},{"family":"Edwards","given":"Danielle"},{"family":"Lufler","given":"Rebecca"},{"family":"Wactor","given":"Alexandra"},{"family":"Boxerman","given":"Sophia"},{"family":"Smith","given":"Suzanne"},{"family":"Herriott","given":"Hannah"},{"family":"Wilson","given":"Adam"}],"issued":{"date-parts":[[2025]]},"DOI":"10.6084/m9.figshare.30287251.v1","URL":"https://doi.org/10.6084/m9.figshare.30287251.v1","source":"datacite"},{"id":"doi:10.6084/m9.figshare.30287251","type":"article-journal","title":"An audit of AI-related documents across U.S. medical schools: A framework-based qualitative content analysis","abstract":"Medical schools would benefit from systematic guidance for developing comprehensive artificial intelligence (AI) policies, given generative AI’s rapid integration into medical education. This study developed and applied an idealized AI policy framework to analyze AI-related documents at U.S. medical school institutions, providing reference points for the development and refinement of institutional policies. AI-related documents from institutions with U.S. allopathic and osteopathic medical schools were systematically collected (from August to October 2024) and analyzed using a comprehensive framework containing 24 subthemes across six themes: Background/Context, Governance, AI Literacy, Tools/Usage, Ethical/Legal Considerations, and Technology Support and Infrastructure. Publicly available online documents were systematically coded to generate framework subtheme scores indicating breadth of coverage across framework themes. AI-related documents retrieved from 73.7% (146/198) of U.S. medical school institutions covered an average of 8 of 24 subthemes, representing a mean framework coverage score of 32.3% ± 19.8 Rarely addressed subthemes included Audit and Compliance Mechanisms (6.8%, 10/146), Technical Infrastructure (6.2%, 9/146), and Environmental Stewardship (1.4%, 2/146). Academic Honesty and Plagiarism dominated AI-related documents (81.5%, 119/146), followed by Decision-Making Authority (54.1%, 79/146) and Critical Evaluation (52.1%, 76/146). Formal AI policies demonstrated significantly higher framework coverage than other AI document types (44.0% vs 30.4%, p = 0.003). Seven institutions with the highest coverage (≥13/24 subthemes) shared seven common distinguishing features, with six present universally. AI-related documents currently emphasize academic integrity over strategic planning, with substantial gaps in infrastructure and review mechanisms. Institutions can enhance their AI policies by incorporating common features identified in well-designed policies and following frameworks that strike a balance between immediate concerns and long-term adaptability.","author":[{"family":"Rush","given":"Emily"},{"family":"Byram","given":"Jessica"},{"family":"Garnett","given":"Colleen"},{"family":"Devaul","given":"Nicole"},{"family":"Smith","given":"Laura"},{"family":"Checchi","given":"Margaret"},{"family":"Martin","given":"Daniel"},{"family":"Hoffman","given":"Leslie"},{"family":"Brown","given":"Kirstin"},{"family":"Mumbower","given":"Daniel"},{"family":"Becker","given":"Robert"},{"family":"Roach","given":"Victoria"},{"family":"Doubleday","given":"Alison"},{"family":"Edwards","given":"Danielle"},{"family":"Lufler","given":"Rebecca"},{"family":"Wactor","given":"Alexandra"},{"family":"Boxerman","given":"Sophia"},{"family":"Smith","given":"Suzanne"},{"family":"Herriott","given":"Hannah"},{"family":"Wilson","given":"Adam"}],"issued":{"date-parts":[[2025]]},"DOI":"10.6084/m9.figshare.30287251","URL":"https://doi.org/10.6084/m9.figshare.30287251","source":"datacite"},{"id":"doi:10.17605/osf.io/zt6x4","type":"article-journal","title":"Exploring Artificial Intelligence Models for the Diagnosis of Sarcopenia: A Scoping Review","abstract":"Objetive: The objective of this scoping review is to systematically map the existing literature on AI and ML models developed for the detection or prediction of sarcopenia, to describe their methodological characteristics, input features, model performance, validation strategies, and adherence to reporting standards. By providing a structured overview of the current evidence landscape, this review aims to inform future research, promote methodological rigor, and facilitate the responsible integration of AI into sarcopenia assessment. METHODS We performed this scoping review in accordance with the PRISMA extension for scoping reviews (PRISMA-ScR, 2022). This review was reported and summarized using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations extension for scoping reviews (PRISMA-ScR, 2022)(8). The protocol for this scoping review is registered in OSF. The questions of the scoping review were: 1- What AI/ML algorithms have been used for binary diagnosis of sarcopenia, and how have these evolved over time? 2- What are the most frequently used input variables/features (e.g., DXA/CT/MRI/ultrasound, BIA, strength tests, gait/performance, labs, EHR variables)? 3- What are the reported sensitivity and specificity (and other metrics), and how are they estimated (internal/external validation)? 4- To what extent do studies adhere to TRIPOD+AI reporting items Search strategy and inclusion criteria The scoping review search was conducted across six databases (Medline, Scopus, Scielo, Web of Science and LILACS) up to February 2026. The search strategy, including all identified keywords, MeSH (Medical Subject Headings) and DeCS (Descriptores en Ciencias de la Salud) terms in English, Portuguese and Spanish according to the database requirement for search strategy is in Supplementary Table S1. The reference list of all included sources of evidence was screened for additional studies. This scoping review considered development and/or validation studies of AI/ML prediction/diagnostic models (including regression-based ML, classical ML, and deep learning) aimed at binary classification of sarcopenia (yes/no), that were published in Spanish, Portuguese and English. Studies that met the following criteria were included: 1) Population: Adults assessed sarcopenia that explicitly state that sarcopenia was diagnosed according to a recognized consensus or guideline (e.g., EWGSOP2, AWGS, FNIH, IWGS); 2) Concept: AI/ML models that output binary sarcopenia classification (present/absent) and 3) Context: Any setting (community, hospital, outpatient), any country. Studies will be excluded if sarcopenia is not defined according to an established consensus; if the AI model predicts continuous outcomes (e.g., muscle mass values) without a binary classification aligned to consensus criteria; if the study does not involve an AI-based model; or if it is a review, editorial, commentary, or animal study. Screening tools not embedded within an AI or machine learning framework will also be excluded. All citations were uploaded and duplicates removed using RAYYAN system wish is software for managing and streamlining a systematic review. Study selection All records were imported into Rayyan software for reference management and duplicate removal. Screening was conducted in two stages: 1) title and abstract screening and 2) full-text eligibility assessment. Both stages were performed independently by two reviewers (LAS and CBO). Discrepancies were resolved through discussion, and when necessary, by consultation with a third reviewer (WSL). The study selection process is presented using a PRISMA-ScR flow diagram. Reasons for exclusion at the full-text stage were documented. In accordance with JBI guidance, no formal risk of bias or methodological quality appraisal was performed, as the objective was to map the evidence rather than synthesize effectiveness or causal inference. Data extraction Data extraction was performe","author":[{"family":"Loyola","given":"Walter"},{"family":"Barros-Osorio","given":"Cristián"},{"family":"Gallegos","given":"Eduardo"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/zt6x4","URL":"https://doi.org/10.17605/osf.io/zt6x4","source":"datacite"},{"id":"doi:10.17605/osf.io/c3bea","type":"article-journal","title":"How socioeconomic status impacts the adoption and uptake of digitally-based health interventions: a scoping review protocol","abstract":"Over the past few decades, health service delivery has changed significantly with advancements in technology; this is reflected in the widespread integration of digital health within the health system. Digital health technologies or tools (DHTs) are becoming increasingly integral to best practice health care. Given the pace of innovation in digital health, DHTs are routinely being leveraged as interventions to aid individuals, the health workforce and health system users to address a variety of health and health system challenges. According to the World Health Organisation (WHO), these digital health interventions (DHIs) can be classified into four overarching groups based on the targeted primary user: (1) interventions for persons; (2) interventions for healthcare providers; (3) interventions for health management and support personnel; and (4) interventions for data services. Interventions for persons (i.e. person-centred DHIs), target members of the public who are potential or current users of health services and caregivers of individuals receiving health services. Technologies such as telemedicine, patient portals, mobile health applications and remote monitoring, are shown to improve patients’ quality of life and health outcomes. These technologies increase accessibility to health services, strengthen patient engagement and empowerment, and simultaneously reduce costs, thereby making healthcare more affordable. System-based DHIs (i.e. interventions for healthcare providers, health management and support personnel, and data services), such as electronic medical records, clinical decision support systems and artificial intelligence systems, are shown to improve overall health care delivery by optimising clinical practice and management, and reducing healthcare workers’ cognitive load. Despite their positive impacts on health outcomes, service delivery, and quality, the implementation of DHTs, particularly patient-centred DHTs, has faced a number of challenges. Although small-scale and proof-of-concept trials show promising results, DHTs are rarely sustained beyond the study period and often fail to translate to the same success in the real-world settings of busy public health clinics, large health systems, and various community settings. Additionally, their effectiveness is often hindered by a lack of uptake and usage. This disparity in access and use is known as the digital divide. Existing literature indicates that individuals belonging to more disadvantaged populations (i.e. those with lower levels of education, lower income, low literacy levels, or who are culturally diverse or live in rural areas) are less likely to start or continue using DHTs, despite the increased availability of the internet and digital technologies. However, there is a lack of evidence describing the differential impact of DHTs across different socioeconomic groups. Given the push towards the use of digital tools and interventions in health care15, underserved populations, especially those in lower socioeconomic groups, are at risk of being left behind in reaping the benefits of DHTs. Therefore, the objective of this scoping review is to explore the differential impact of patient-centred digitally-based interventions according to socioeconomic position. By understanding the equity impacts of DHTs across socioeconomic groups, this research can guide healthcare practices in implementing DHTs that are tailored across the population, concurrently ensuring that they are accessible and sustainable. Furthermore, policymakers could be more informed in developing equity-based criteria for the implementation of DHTs across health services to avoid perpetuating disparities in health outcomes. The research question is ‘What is the differential impact of patient-centred digital health interventions according to socioeconomic position?’ The objectives are as follows: Primary objective: 1. To describe the differential impact of patient-centred digital health int","author":[{"family":"Wang","given":"Sofia"},{"family":"Beauchamp","given":"Alison"},{"family":"Azar","given":"Denise"},{"family":"Shee","given":"Anna"},{"family":"Metcalf","given":"Olivia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.17605/osf.io/c3bea","URL":"https://doi.org/10.17605/osf.io/c3bea","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8315191.v1","type":"article-journal","title":"A scoping review of the use of generative artificial intelligence tools in health profession education","abstract":"Abstract Background Generative Artificial Intelligence (GenAI) is one of the leading innovations that is expected to reshape society for decades to come. Health professions education (HPE) programs are expected to prepare graduates with adequate knowledge and skills to provide high-quality patient-centered care. Although the use of GenAI in health professions is increasing, its optimal integration in HPE is still ambiguous. This scoping review aims to synthesize currently available literature regarding the use of GenAI in health professions education. Method This scoping review is conducted following JBI methodology for scoping reviews framework 2020 and aligned with PRISMA-ScR. A systematic and comprehensive search was conducted in PubMed, ERIC, CINAHL, Embase, Scopus, Cochrane Library, and ProQuest Central with no language restrictions. The identified evidence was screened and extracted using Covidence software. Publications on the integration of GenAI in undergraduate or graduate health profession education were considered. Data was analyzed and presented using graphs and charts. Followed by a narrative thematic mapping of the included studies. Results Out of 14,208 scanned records, 241 were considered eligible. The included studies discuss the application of GenAI in diverse education processes of different health professions, such as curriculum design, content creation, content delivery, personalized learning, assessment, evaluation, and feedback provision. Most studies focused on ChatGPT integration in medical and nursing education, with content creation emerging as the predominant area of integration, whereas curriculum design and GenAI literacy were underexplored. Perception studies reported a positive perspective regarding GenAI used in education among students and faculty. Conclusion This review provides an overview of the current integration of GenAI in HPE in the literature, highlighting the associated opportunities, challenges, facilitators, and barriers. Future education efforts should focus on enhancing GenAI literacy, developing policy, and adopting a balanced approach. In addition to conducting comparative studies and long-term assessment of GenAI impact.","author":[{"family":"Basil","given":"Mounyah"},{"family":"Ahmed","given":"Waad"},{"family":"Hajeomar","given":"Reem"},{"family":"Strawbridge","given":"Judith"},{"family":"Lynch","given":"Matthew"},{"family":"Mukhalalati","given":"Banan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8315191.v1","URL":"https://doi.org/10.6084/m9.figshare.c.8315191.v1","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8315191","type":"article-journal","title":"A scoping review of the use of generative artificial intelligence tools in health profession education","abstract":"Abstract Background Generative Artificial Intelligence (GenAI) is one of the leading innovations that is expected to reshape society for decades to come. Health professions education (HPE) programs are expected to prepare graduates with adequate knowledge and skills to provide high-quality patient-centered care. Although the use of GenAI in health professions is increasing, its optimal integration in HPE is still ambiguous. This scoping review aims to synthesize currently available literature regarding the use of GenAI in health professions education. Method This scoping review is conducted following JBI methodology for scoping reviews framework 2020 and aligned with PRISMA-ScR. A systematic and comprehensive search was conducted in PubMed, ERIC, CINAHL, Embase, Scopus, Cochrane Library, and ProQuest Central with no language restrictions. The identified evidence was screened and extracted using Covidence software. Publications on the integration of GenAI in undergraduate or graduate health profession education were considered. Data was analyzed and presented using graphs and charts. Followed by a narrative thematic mapping of the included studies. Results Out of 14,208 scanned records, 241 were considered eligible. The included studies discuss the application of GenAI in diverse education processes of different health professions, such as curriculum design, content creation, content delivery, personalized learning, assessment, evaluation, and feedback provision. Most studies focused on ChatGPT integration in medical and nursing education, with content creation emerging as the predominant area of integration, whereas curriculum design and GenAI literacy were underexplored. Perception studies reported a positive perspective regarding GenAI used in education among students and faculty. Conclusion This review provides an overview of the current integration of GenAI in HPE in the literature, highlighting the associated opportunities, challenges, facilitators, and barriers. Future education efforts should focus on enhancing GenAI literacy, developing policy, and adopting a balanced approach. In addition to conducting comparative studies and long-term assessment of GenAI impact.","author":[{"family":"Basil","given":"Mounyah"},{"family":"Ahmed","given":"Waad"},{"family":"Hajeomar","given":"Reem"},{"family":"Strawbridge","given":"Judith"},{"family":"Lynch","given":"Matthew"},{"family":"Mukhalalati","given":"Banan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8315191","URL":"https://doi.org/10.6084/m9.figshare.c.8315191","source":"datacite"},{"id":"doi:10.48550/arxiv.2510.09308","type":"manuscript","title":"A Model-Driven Engineering Approach to AI-Powered Healthcare Platforms","abstract":"Artificial intelligence (AI) has the potential to transform healthcare by supporting more accurate diagnoses and personalized treatments. However, its adoption in practice remains constrained by fragmented data sources, strict privacy rules, and the technical complexity of building reliable clinical systems. To address these challenges, we introduce a model driven engineering (MDE) framework designed specifically for healthcare AI. The framework relies on formal metamodels, domain-specific languages (DSLs), and automated transformations to move from high level specifications to running software. At its core is the Medical Interoperability Language (MILA), a graphical DSL that enables clinicians and data scientists to define queries and machine learning pipelines using shared ontologies. When combined with a federated learning architecture, MILA allows institutions to collaborate without exchanging raw patient data, ensuring semantic consistency across sites while preserving privacy. We evaluate this approach in a multi center cancer immunotherapy study. The generated pipelines delivered strong predictive performance, with support vector machines achieving up to 98.5 percent and 98.3 percent accuracy in key tasks, while substantially reducing manual coding effort. These findings suggest that MDE principles metamodeling, semantic integration, and automated code generation can provide a practical path toward interoperable, reproducible, and trustworthy digital health platforms.","author":[{"family":"Raheem","given":"Mira"},{"family":"Elgammal","given":"Amal"},{"family":"Papazoglou","given":"Michael"},{"family":"Krämer","given":"Bernd"},{"family":"El-Tazi","given":"Neamat"}],"issued":{"date-parts":[[2025]]},"DOI":"10.48550/arxiv.2510.09308","URL":"https://doi.org/10.48550/arxiv.2510.09308","source":"datacite"},{"id":"doi:10.6084/m9.figshare.30945099.v1","type":"article-journal","title":"Artificial intelligence in breast cancer: clinical applications in diagnosis, prognosis, and therapeutics","abstract":"Breast cancer (BC) presents a considerable global health challenge and is characterized by increasing mortality and morbidity rates. Prompt screening and accurate diagnosis are crucial for improving patient outcomes. For the assessment of BC, radiographic imaging modalities such as digital breast tomosynthesis (DBT), ultrasound, digital mammography (DM), magnetic resonance imaging (MRI), and nuclear medicine procedures are commonly used. The gold standard for confirming cancer is histopathology. To effectively support the segmentation, diagnosis, and prognosis of BC. Artificial intelligence (AI) technologies show great promise for the quantitative depiction of medical images. This review explores recent strides in AI applications for BC. The literature search from 2018 to 2025 was performed with the PubMed database. It includes rapid breast lesion detection, segmentation, cancer diagnosis and enhanced imaging quality through data augmentation. It also discusses the biological characterization of BC via AI-based classification tools, including subtyping and staging. Furthermore, this review also explores the use of multiomics data to predict clinical outcomes such as survival, treatment response, and metastasis in BC. Additionally, we recognized the challenges faced by AI in BC in real-world applications, including organizing data, model interpretability, and regulatory compliance. BC is one of the most common cancers in women worldwide. Early detection greatly increases the chances of successful treatment, but traditional screening and diagnosis methods rely on specialists who can review only many images and samples each day. AI offers new tools that learn from large collections of medical data to help doctors detect, diagnose, and treat BC more effectively. In screening, AI can quickly analyze mammograms, digital breast tomosynthesis, ultrasounds, and MRIs. It highlights suspicious areas, decreases false alarms, and can spot small lesions that humans might miss – especially in dense breast tissue. By prioritizing high-risk images for review, AI makes it easy for radiologists without compromising accuracy. When a biopsy is taken, AI systems examine digitized tissue slides to assess tumor features and key biomarkers, such as hormone receptors (ERs, PRs), HER2, and proliferation markers (Ki-67). These automated analyses are consistent, rapid, and can sometimes match or exceed human performance, helping pathologists make faster, more reliable diagnoses. AI models also combine imaging, genetic, and clinical information to predict an individual patient’s risk of developing BC, the likely course of the disease, and how it will respond to specific treatments. Such personalized predictions guide decisions about preventive measures, chemotherapy, surgery, and targeted therapies. Despite their great promise, AI tools require large, high-quality datasets for training, clear validation standards, and safeguards for patient privacy. Ongoing studies are working to ensure that these systems are robust across diverse populations. With continued research, careful testing, and thoughtful integration into clinical practice, AI has the potential to transform BC care to make screening more accurate, make diagnosis more precise, and make treatment truly personalized.","author":[{"family":"Singh","given":"Janhvi"},{"family":"Alsaidan","given":"Omar"},{"family":"Aodah","given":"Alhussain"},{"family":"Alrobaian","given":"Majed"},{"family":"Almalki","given":"Waleed"},{"family":"Almujri","given":"Salem"},{"family":"Sahoo","given":"Ankit"},{"family":"Alam","given":"Kainat"},{"family":"Lal","given":"Jonathan"},{"family":"Barkat","given":"Md"},{"family":"Rahman","given":"Mahfoozur"}],"issued":{"date-parts":[[2025]]},"DOI":"10.6084/m9.figshare.30945099.v1","URL":"https://doi.org/10.6084/m9.figshare.30945099.v1","source":"datacite"},{"id":"doi:10.6084/m9.figshare.30945099","type":"article-journal","title":"Artificial intelligence in breast cancer: clinical applications in diagnosis, prognosis, and therapeutics","abstract":"Breast cancer (BC) presents a considerable global health challenge and is characterized by increasing mortality and morbidity rates. Prompt screening and accurate diagnosis are crucial for improving patient outcomes. For the assessment of BC, radiographic imaging modalities such as digital breast tomosynthesis (DBT), ultrasound, digital mammography (DM), magnetic resonance imaging (MRI), and nuclear medicine procedures are commonly used. The gold standard for confirming cancer is histopathology. To effectively support the segmentation, diagnosis, and prognosis of BC. Artificial intelligence (AI) technologies show great promise for the quantitative depiction of medical images. This review explores recent strides in AI applications for BC. The literature search from 2018 to 2025 was performed with the PubMed database. It includes rapid breast lesion detection, segmentation, cancer diagnosis and enhanced imaging quality through data augmentation. It also discusses the biological characterization of BC via AI-based classification tools, including subtyping and staging. Furthermore, this review also explores the use of multiomics data to predict clinical outcomes such as survival, treatment response, and metastasis in BC. Additionally, we recognized the challenges faced by AI in BC in real-world applications, including organizing data, model interpretability, and regulatory compliance. BC is one of the most common cancers in women worldwide. Early detection greatly increases the chances of successful treatment, but traditional screening and diagnosis methods rely on specialists who can review only many images and samples each day. AI offers new tools that learn from large collections of medical data to help doctors detect, diagnose, and treat BC more effectively. In screening, AI can quickly analyze mammograms, digital breast tomosynthesis, ultrasounds, and MRIs. It highlights suspicious areas, decreases false alarms, and can spot small lesions that humans might miss – especially in dense breast tissue. By prioritizing high-risk images for review, AI makes it easy for radiologists without compromising accuracy. When a biopsy is taken, AI systems examine digitized tissue slides to assess tumor features and key biomarkers, such as hormone receptors (ERs, PRs), HER2, and proliferation markers (Ki-67). These automated analyses are consistent, rapid, and can sometimes match or exceed human performance, helping pathologists make faster, more reliable diagnoses. AI models also combine imaging, genetic, and clinical information to predict an individual patient’s risk of developing BC, the likely course of the disease, and how it will respond to specific treatments. Such personalized predictions guide decisions about preventive measures, chemotherapy, surgery, and targeted therapies. Despite their great promise, AI tools require large, high-quality datasets for training, clear validation standards, and safeguards for patient privacy. Ongoing studies are working to ensure that these systems are robust across diverse populations. With continued research, careful testing, and thoughtful integration into clinical practice, AI has the potential to transform BC care to make screening more accurate, make diagnosis more precise, and make treatment truly personalized.","author":[{"family":"Singh","given":"Janhvi"},{"family":"Alsaidan","given":"Omar"},{"family":"Aodah","given":"Alhussain"},{"family":"Alrobaian","given":"Majed"},{"family":"Almalki","given":"Waleed"},{"family":"Almujri","given":"Salem"},{"family":"Sahoo","given":"Ankit"},{"family":"Alam","given":"Kainat"},{"family":"Lal","given":"Jonathan"},{"family":"Barkat","given":"Md"},{"family":"Rahman","given":"Mahfoozur"}],"issued":{"date-parts":[[2025]]},"DOI":"10.6084/m9.figshare.30945099","URL":"https://doi.org/10.6084/m9.figshare.30945099","source":"datacite"},{"id":"doi:10.17605/osf.io/83rzt","type":"article-journal","title":"Implementation of artificial intelligence in prosthetics for amputees: a scoping review","abstract":"Amputation is the surgical removal of a limb secondary to injury or disease and represents a growing global health burden. Its incidence and prevalence have increased over recent years, driven primarily by trauma, peripheral arterial disease, diabetes mellitus, and hypoperfusion states. Individuals living with limb loss face significant physical, functional, and psychosocial challenges that substantially affect quality of life. Prosthetic devices have evolved significantly; however, achieving control and sensory feedback comparable to a biological limb remains challenging. Artificial Intelligence (AI) has emerged as a promising tool to enhance prosthetic functionality through adaptive motor control and personalization. Despite increasing research, evidence remains heterogeneous. This scoping review aims to systematically map existing evidence on AI implementation in prosthetics for amputee patients.","author":[{"family":"Rodriguez","given":"Maria"},{"family":"Rincon","given":"Erwin"},{"family":"Rodriguez","given":"Maria"},{"family":"Polo","given":"Maria"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/83rzt","URL":"https://doi.org/10.17605/osf.io/83rzt","source":"datacite"},{"id":"doi:10.17605/osf.io/4dn38","type":"article-journal","title":"Off-Label Combination Therapies for the Management of Atopic Dermatitis: A Protocol for a Scoping Review","abstract":"Introduction: Atopic dermatitis (AD) is a chronic, heterogeneous and relapsing inflammatory skin disease that presents a significant global health burden, affecting up to 20% of the paediatric population and 2-10% of adults worldwide.1–3 AD is considered the skin disease with the highest non-fatal health burden due to its profound negative impact on patient quality of life, sleep and mental health.1,4–6 AD is characterised by severe pruritus and xerosis, the pathogenesis of which is multifactorial, including an interplay between genetic predisposition, immune dysregulation with aberrant Th2 responses, skin barrier disruption and cutaneous dysbiosis; each of which may be modulated by environmental factors.2,7 These cytokines play a central role in B-cell maturation, IgE class switching and the down-regulation of essential skin barrier proteins such filaggrin, loricrin and involucrin.8 Current Therapeutic Landscape: The current management of AD follows a step-wise approach based on disease severity.9 For mild forms, the cornerstone of therapy includes trigger avoidance, regular use of emollients to combat xerosis and the application of topical corticosteroids (TCS) or topical calcineurin inhibitors (TCI).9 However, patients with moderate-to-severe disease that remains recalcitrant to topical interventions often require second-line treatments, including phototherapy or systemic immunomodulatory agents.5,9 Historically, systemic options were limited to conventional immunosuppressants such as ciclosporin A (CsA), methotrexate (MTX), azathioprine (AZA) and mycophenolate mofetil (MMF).10,11 Despite extensive clinical experience for use in management of moderate-severe AD, the above therapies are not licensed by the FDA/EMA/Japan for management of moderate to severe AD. Their long-term application is often curtailed by an unfavourable risk-benefit ratio, potential organ toxicity (e.g., nephrotoxicity with CsA or bone-marrow suppression with MTX), and a lack of robust long-term data.10,12 The therapeutic landscape was revolutionised in 2017 with the approval of dupilumab, the first biologic targeting the IL-4 receptor alpha chain to inhibit the signalling of IL-4 and IL-13.13 Since then the therapeutic landscape has expanded to include other biologics such as lebrikizumab, nemolizumab and tralokinumab, as well as oral small molecule drugs such as Janus kinase (JAK) inhibitors (e.g., abrocitinib, baricitinib, and upadacitinib).14 Despite these advancements, a subset of patients, demonstrates an inadequate response or resistance to biologic monotherapy.14,15 The Challenge of Recalcitrant Disease and Off-Label Combinations: In clinical practice, dermatologists often encounter patients whose disease is not adequately controlled by a single systemic agent.16 While phase III clinical trials typically evaluate these drugs as monotherapy (or with limited TCS use), daily practice sometimes involves an off-label combination of systemic therapies.1,3,17 There are two primary clinical scenarios where these combinations occur: 1. Transitioning and tapering: patients starting a new biologic like dupilumab may continue a conventional immunosuppressant (such as CsA) for several weeks to prevent disease flares during the transition period or to induce faster clinical remission.18 2. Rescue therapy: In patients experiencing a lack of response or loss of disease control on biologic monotherapy, a second systemic agent or phototherapy may be added as a \"rescue\" or ‘combination’ strategy to enhance effectiveness and maintain long-term control.3 Observational data indicate that up to 45% of patients in real-world settings may continue systemic immunosuppressants during the initiation of dupilumab.1 Common combinations reported in the literature include dupilumab paired with methylprednisolone, ciclosporin, methotrexate or azathioprine.3 Rationale for a Scoping Review: Despite the increasing use of combined systemic regimens, there is neither standardized gu","author":[{"family":"Sexton","given":"Fiona"},{"family":"Wang","given":"Leo"},{"family":"Stefanovic","given":"Nicholas"},{"family":"Irvine","given":"Alan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/4dn38","URL":"https://doi.org/10.17605/osf.io/4dn38","source":"datacite"},{"id":"doi:10.5281/zenodo.18476707","type":"article-journal","title":"Systematic Integration of Artificial Intelligence and Machine Learning in the Early Detection and Management of Goitre: A Global Epidemiological and Computational Framework","abstract":"Background: Goitre remains a high-signal global health indicator of thyroid dysfunction and population-level iodine status. Despite progress in salt iodization, early-stage thyroid enlargement is frequently under-detected in routine practice, especially when physical examination is confounded by body habitus and clinician subjectivity. Ultrasound is the preferred modality for early assessment, but interpretation is operator-dependent and increasingly burdened by rising thyroid nodule prevalence. Objective: This review synthesizes evidence on Machine Learning (ML) and Artificial Intelligence (AI) methods for predicting thyroid dysfunction and diagnosing early goitre (WHO Grade 1), with a practical emphasis on multi-modal “holistic AI” systems that combine tabular laboratory markers with imaging features. Methods: We summarize (i) supervised learning pipelines for structured clinical data (e.g., age, sex, TSH, T3, T4, T4U/FTI), (ii) deep learning architectures for ultrasound-based detection and segmentation (CNNs, U-Net variants, Vision Transformers), and (iii) deployment considerations including explainability, bias control, and reproducible benchmarking using open datasets. Following common clinical ML reporting practice, we emphasize confusion-matrixbased evaluation (precision/recall/F1/MCC) and strong ensemble baselines for tabular prediction. [30,31] Results: For tabular prediction tasks, stacked ensembles and gradient-boosted trees repeatedly rank among the best-performing approaches, particularly when combined with careful feature engineering and imbalance mitigation. For imaging, segmentation-first pipelines that estimate thyroid volume (e.g., U-Net family) and classification models leveraging multi-channel inputs or self-attention mechanisms (e.g., ViTs) report high diagnostic performance in differentiating benign enlargement from suspicious nodular patterns. Emerging smartphone-assisted workflows and LLM-based clinical summarization show promise for low-resource settings but require rigorous validation. Conclusion: AI can shift goitre management from late-stage detection to proactive screening by improving sensitivity for occult Grade 1 enlargement, standardizing ultrasound interpretation, and reducing unnecessary invasive procedures. Clinical adoption, however, depends on transparent explainability, external validation across diverse cohorts, and governance aligned with high-risk medical AI standards.","author":[{"family":"Akugri","given":"Kingdom"},{"family":"Agbenyo","given":"Prince"},{"family":"Akugri","given":"Marious"},{"family":"Keteku","given":"Lovelyn"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18476707","URL":"https://doi.org/10.5281/zenodo.18476707","source":"datacite"},{"id":"doi:10.5281/zenodo.18476708","type":"article-journal","title":"Systematic Integration of Artificial Intelligence and Machine Learning in the Early Detection and Management of Goitre: A Global Epidemiological and Computational Framework","abstract":"Background: Goitre remains a high-signal global health indicator of thyroid dysfunction and population-level iodine status. Despite progress in salt iodization, early-stage thyroid enlargement is frequently under-detected in routine practice, especially when physical examination is confounded by body habitus and clinician subjectivity. Ultrasound is the preferred modality for early assessment, but interpretation is operator-dependent and increasingly burdened by rising thyroid nodule prevalence. Objective: This review synthesizes evidence on Machine Learning (ML) and Artificial Intelligence (AI) methods for predicting thyroid dysfunction and diagnosing early goitre (WHO Grade 1), with a practical emphasis on multi-modal “holistic AI” systems that combine tabular laboratory markers with imaging features. Methods: We summarize (i) supervised learning pipelines for structured clinical data (e.g., age, sex, TSH, T3, T4, T4U/FTI), (ii) deep learning architectures for ultrasound-based detection and segmentation (CNNs, U-Net variants, Vision Transformers), and (iii) deployment considerations including explainability, bias control, and reproducible benchmarking using open datasets. Following common clinical ML reporting practice, we emphasize confusion-matrixbased evaluation (precision/recall/F1/MCC) and strong ensemble baselines for tabular prediction. [30,31] Results: For tabular prediction tasks, stacked ensembles and gradient-boosted trees repeatedly rank among the best-performing approaches, particularly when combined with careful feature engineering and imbalance mitigation. For imaging, segmentation-first pipelines that estimate thyroid volume (e.g., U-Net family) and classification models leveraging multi-channel inputs or self-attention mechanisms (e.g., ViTs) report high diagnostic performance in differentiating benign enlargement from suspicious nodular patterns. Emerging smartphone-assisted workflows and LLM-based clinical summarization show promise for low-resource settings but require rigorous validation. Conclusion: AI can shift goitre management from late-stage detection to proactive screening by improving sensitivity for occult Grade 1 enlargement, standardizing ultrasound interpretation, and reducing unnecessary invasive procedures. Clinical adoption, however, depends on transparent explainability, external validation across diverse cohorts, and governance aligned with high-risk medical AI standards.","author":[{"family":"Akugri","given":"Kingdom"},{"family":"Agbenyo","given":"Prince"},{"family":"Akugri","given":"Marious"},{"family":"Keteku","given":"Lovelyn"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18476708","URL":"https://doi.org/10.5281/zenodo.18476708","source":"datacite"},{"id":"doi:10.5281/zenodo.17367484","type":"article-journal","title":"VALIDATE D4.1 Study Initiation Package","abstract":"Artificial intelligence (AI)-powered prognostic tools and clinical decision support systems can predict the outcome of certain diseases based on a multitude of patient data at high speed, facilitating decisions by healthcare professionals. In acute ischemic stroke, the overall treatment effect and population-wide outcome benefit of treatments such as IV thrombolysis and mechanical thrombectomy are well established. However, in individual patients it is difficult to predict the prognosis in the acute phase of stroke: some patients are candidates for these treatments, but may have poor clinical outcomes (no improvement of stroke or even worsening). Our aim in this study is to validate an artificial intelligence (AI)-based prognostic tool to provide accurate real-time outcome prediction in patients with acute ischemic stroke. During the study, all patients admitted to the emergency room with an acute ischemic stroke will receive the usual treatment for acute stroke in accordance with the stroke neurologists in charge. A “shadow” clinical researcher, without interaction with treating physicians, will collect the data required by the AI model in vivo. These data will be obtained by filling in clinical data through an App on a hospital mobile/tablet, and by a connection with your electronic medical record. The AI models will estimate the outcome of an acute stroke patient, and this prediction will be compared with the real outcome of the patient after 3 months of follow-up. This document presents the first draft of the study initiation package of the study “Validation of a Trustworthy AI-based Clinical Decision Support System for Improving Patient Outcome in Acute Stroke Treatment””. The study aims to develop a demonstrator to provide accurate prognosis for acute ischemic stroke patients during the hyperacute phase. It has a preclinical phase with the development of the artificial intelligence (AI) models based on retrospective data, and a prospective phase to test these models in real time through a prospective multicenter shadowing observational non-interventional study. The predicted outcomes of the patients will be compared with real clinician and patient reported outcomes (PROs and CROs). Here we provide the protocol of the clinical study, the information form and informed consent form for patients/families and the interaction with the Institutional Review Board (IRB) in the different international centers. Now, an IRB approval from the pre-clinical retrospective study has been received, and we are waiting for the complete (pre-clinical and prospective data) IRB approval. The clinical study is also being evaluated for registration in clinicaltrials.gov.","author":[{"family":"Rubiera Del Fueyo","given":"Marta"},{"family":"Bonekamp","given":"Susanne"},{"family":"Leker","given":"Ronen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.17367484","URL":"https://doi.org/10.5281/zenodo.17367484","source":"datacite"},{"id":"doi:10.5281/zenodo.17367485","type":"article-journal","title":"VALIDATE D4.1 Study Initiation Package","abstract":"Artificial intelligence (AI)-powered prognostic tools and clinical decision support systems can predict the outcome of certain diseases based on a multitude of patient data at high speed, facilitating decisions by healthcare professionals. In acute ischemic stroke, the overall treatment effect and population-wide outcome benefit of treatments such as IV thrombolysis and mechanical thrombectomy are well established. However, in individual patients it is difficult to predict the prognosis in the acute phase of stroke: some patients are candidates for these treatments, but may have poor clinical outcomes (no improvement of stroke or even worsening). Our aim in this study is to validate an artificial intelligence (AI)-based prognostic tool to provide accurate real-time outcome prediction in patients with acute ischemic stroke. During the study, all patients admitted to the emergency room with an acute ischemic stroke will receive the usual treatment for acute stroke in accordance with the stroke neurologists in charge. A “shadow” clinical researcher, without interaction with treating physicians, will collect the data required by the AI model in vivo. These data will be obtained by filling in clinical data through an App on a hospital mobile/tablet, and by a connection with your electronic medical record. The AI models will estimate the outcome of an acute stroke patient, and this prediction will be compared with the real outcome of the patient after 3 months of follow-up. This document presents the first draft of the study initiation package of the study “Validation of a Trustworthy AI-based Clinical Decision Support System for Improving Patient Outcome in Acute Stroke Treatment””. The study aims to develop a demonstrator to provide accurate prognosis for acute ischemic stroke patients during the hyperacute phase. It has a preclinical phase with the development of the artificial intelligence (AI) models based on retrospective data, and a prospective phase to test these models in real time through a prospective multicenter shadowing observational non-interventional study. The predicted outcomes of the patients will be compared with real clinician and patient reported outcomes (PROs and CROs). Here we provide the protocol of the clinical study, the information form and informed consent form for patients/families and the interaction with the Institutional Review Board (IRB) in the different international centers. Now, an IRB approval from the pre-clinical retrospective study has been received, and we are waiting for the complete (pre-clinical and prospective data) IRB approval. The clinical study is also being evaluated for registration in clinicaltrials.gov.","author":[{"family":"Rubiera Del Fueyo","given":"Marta"},{"family":"Bonekamp","given":"Susanne"},{"family":"Leker","given":"Ronen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.17367485","URL":"https://doi.org/10.5281/zenodo.17367485","source":"datacite"},{"id":"doi:10.48550/arxiv.2506.17442","type":"manuscript","title":"Keeping Medical AI Healthy and Trustworthy: A Review of Detection and Correction Methods for System Degradation","abstract":"Artificial intelligence (AI) is increasingly integrated into modern healthcare, offering powerful support for clinical decision-making. However, in real-world settings, AI systems may experience performance degradation over time, due to factors such as shifting data distributions, changes in patient characteristics, evolving clinical protocols, and variations in data quality. These factors can compromise model reliability, posing safety concerns and increasing the likelihood of inaccurate predictions or adverse outcomes. This review presents a forward-looking perspective on monitoring and maintaining the \"health\" of AI systems in healthcare. We highlight the urgent need for continuous performance monitoring, early degradation detection, and effective self-correction mechanisms. The paper begins by reviewing common causes of performance degradation at both data and model levels. We then summarize key techniques for detecting data and model drift, followed by an in-depth look at root cause analysis. Correction strategies are further reviewed, ranging from model retraining to test-time adaptation. Our survey spans both traditional machine learning models and state-of-the-art large language models (LLMs), offering insights into their strengths and limitations. Finally, we discuss ongoing technical challenges and propose future research directions. This work aims to guide the development of reliable, robust medical AI systems capable of sustaining safe, long-term deployment in dynamic clinical settings.","author":[{"family":"Guan","given":"Hao"},{"family":"Bates","given":"David"},{"family":"Zhou","given":"Li"}],"issued":{"date-parts":[[2025]]},"DOI":"10.48550/arxiv.2506.17442","URL":"https://doi.org/10.48550/arxiv.2506.17442","source":"datacite"},{"id":"doi:10.17605/osf.io/3bv2p","type":"article-journal","title":"Artificial Intelligence for Stroke Risk Stratification and Surgical Decision-Making in Asymptomatic Carotid Stenosis: A Scoping Review","abstract":"This project is a scoping review examining how artificial intelligence (AI) and machine learning (ML) approaches have been applied to stroke risk stratification in patients with asymptomatic carotid stenosis, and how these models may inform decisions between surgical intervention and best medical therapy. The central premise is that, despite major advances in imaging, plaque characterisation, and computational modelling, the carotid field still lacks reliable tools to identify which asymptomatic patients are at sufficiently high risk of future stroke to justify carotid endarterectomy or stenting in the era of contemporary medical therapy. While numerous studies have developed AI models to analyse ultrasound, CT, and MRI plaque features, hemodynamics, and clinical variables, these efforts have not been synthesised in the context of the clinical question that matters most: patient selection for intervention. The review will systematically map the existing literature on AI-driven risk prediction in asymptomatic carotid disease, including models that predict ipsilateral stroke, transient ischaemic attack, plaque progression or instability, microembolisation, cognitive decline, and the need for intervention. It will characterise the data sources used by these models, such as radiomics, ultrasound texture analysis, MRI plaque imaging, CT angiography features, clinical and biochemical variables, and computational flow modelling, as well as the modelling techniques employed, including supervised machine learning, deep learning, clustering, and risk modelling approaches. In addition, the review will examine the stage of clinical translation of these models and how their outputs align, or fail to align, with current guideline-based risk stratification for carotid intervention. A key component of this review is conceptual: to evaluate whether existing AI models, even if not designed for this purpose, could theoretically function as decision-support tools to distinguish high-risk asymptomatic patients who may benefit from surgery from those better managed with medical therapy alone. By identifying gaps between AI risk prediction research and real-world surgical decision-making, this review aims to connect two parallel fields, computational plaque analysis and carotid management, that have largely evolved in isolation. The expected outcomes of this project include a comprehensive map of AI applications in asymptomatic carotid stenosis, identification of recurring methodological patterns and data modalities, clarification of how these models relate to clinically meaningful outcomes, and a clear delineation of the translational gap preventing their use in practice. Ultimately, the review aims to provide a foundation for future research and guideline development by outlining how AI-driven risk stratification could reshape patient selection for carotid endarterectomy, stenting, and best medical therapy in the modern era of stroke prevention.","author":[{"family":"Sharma","given":"Chinmay"},{"family":"Hilkiah","given":"Jeremiah"},{"family":"Saricilar","given":"Erin"},{"family":"Jesudason","given":"Daniel"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/3bv2p","URL":"https://doi.org/10.17605/osf.io/3bv2p","source":"datacite"},{"id":"doi:10.5281/zenodo.21313154","type":"article-journal","title":"Neuro-Oncology Benchmark: The Resource-Interpretability Tradeoff in Radiomics and Multiclass Brain Tumor Classification Based on Deep Transfer Learning vs Handcrafted Radiomics.","abstract":"The automatic multiclass brain tumor classification using MRI images plays an important role in a non-invasive clinical setting. However, the choice of a model demands the trade-off between accuracy, complexity, and interpretability of the classifier. In this study, we have established a benchmark for comparison of classical machine learning (ML) approaches to deep learning (DL) methods, using the same data distribution setup to describe the aforementioned complexity-interpretability trade-off frontier. We generate a 72-dimensional manually designed radiomic feature set, which comprises the first-order intensity features, GLCM features, LBP features, shape morphological measures, and FFT characteristics. These are compared against fine-tuned, ImageNet-pretrained ResNet50 and EfficientNet-B0 on a balanced 4-class dataset (Glioma, Meningioma, Pituitary, and No Tumor; $n=7,200$ images). To address the clinical \"black-box\" problem, a dual-axis Explainable AI (XAI) pipeline maps SHAP values to the radiomic feature space and Grad-CAM activations to the deep neural layers. Results show that, based on three independently seeded training cycles, ResNet50 (95.44% $\\pm$ 0.28%) and EfficientNet-B0 (95.44% $\\pm$ 0.22%) achieve statistically indistinguishable accuracy, while the radiomic-driven SVM reaches a robust 89.12% accuracy and 0.9613 Macro AUC, training in under 4 seconds versus over 17 minutes for EfficientNet-B0. Five-fold cross-validation confirms the stability of the SVM pipeline (92.43% $\\pm$ 0.89%). Despite matching ResNet50's accuracy, EfficientNet-B0 achieves this while reducing parameters by 83% (4.0M vs. 23.5M) and training in roughly half the time. We conclude that radiomic pipelines suit resource-constrained edge deployment, while lightweight deep networks integrated with visual XAI tools offer the ideal configuration for centralized diagnostic frameworks.","author":[{"family":"Bakkas","given":"Othmane"},{"family":"Ennagoura","given":"Drissia"},{"family":"El Kehal","given":"Kamal"},{"family":"Zbakh","given":"Abdelali"},{"family":"El Mahjouby","given":"Mohamed"},{"family":"Bossoufi","given":"Badre"},{"family":"El Fahssi","given":"Khalid"},{"family":"El Far","given":"Mohamed"},{"family":"Taj Bennani","given":"Mohamed"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21313154","URL":"https://doi.org/10.5281/zenodo.21313154","source":"datacite"},{"id":"doi:10.5281/zenodo.21313155","type":"article-journal","title":"Neuro-Oncology Benchmark: The Resource-Interpretability Tradeoff in Radiomics and Multiclass Brain Tumor Classification Based on Deep Transfer Learning vs Handcrafted Radiomics.","abstract":"The automatic multiclass brain tumor classification using MRI images plays an important role in a non-invasive clinical setting. However, the choice of a model demands the trade-off between accuracy, complexity, and interpretability of the classifier. In this study, we have established a benchmark for comparison of classical machine learning (ML) approaches to deep learning (DL) methods, using the same data distribution setup to describe the aforementioned complexity-interpretability trade-off frontier. We generate a 72-dimensional manually designed radiomic feature set, which comprises the first-order intensity features, GLCM features, LBP features, shape morphological measures, and FFT characteristics. These are compared against fine-tuned, ImageNet-pretrained ResNet50 and EfficientNet-B0 on a balanced 4-class dataset (Glioma, Meningioma, Pituitary, and No Tumor; $n=7,200$ images). To address the clinical \"black-box\" problem, a dual-axis Explainable AI (XAI) pipeline maps SHAP values to the radiomic feature space and Grad-CAM activations to the deep neural layers. Results show that, based on three independently seeded training cycles, ResNet50 (95.44% $\\pm$ 0.28%) and EfficientNet-B0 (95.44% $\\pm$ 0.22%) achieve statistically indistinguishable accuracy, while the radiomic-driven SVM reaches a robust 89.12% accuracy and 0.9613 Macro AUC, training in under 4 seconds versus over 17 minutes for EfficientNet-B0. Five-fold cross-validation confirms the stability of the SVM pipeline (92.43% $\\pm$ 0.89%). Despite matching ResNet50's accuracy, EfficientNet-B0 achieves this while reducing parameters by 83% (4.0M vs. 23.5M) and training in roughly half the time. We conclude that radiomic pipelines suit resource-constrained edge deployment, while lightweight deep networks integrated with visual XAI tools offer the ideal configuration for centralized diagnostic frameworks.","author":[{"family":"Bakkas","given":"Othmane"},{"family":"Ennagoura","given":"Drissia"},{"family":"El Kehal","given":"Kamal"},{"family":"Zbakh","given":"Abdelali"},{"family":"El Mahjouby","given":"Mohamed"},{"family":"Bossoufi","given":"Badre"},{"family":"El Fahssi","given":"Khalid"},{"family":"El Far","given":"Mohamed"},{"family":"Taj Bennani","given":"Mohamed"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21313155","URL":"https://doi.org/10.5281/zenodo.21313155","source":"datacite"},{"id":"doi:10.24406/publica-8676","type":"article-journal","title":"Distal radial epiphyseal fusion timing in Northwest Europeans and Middle Eastern asylum seekers: An automated ultrasound and machine learning approach","abstract":"Background: Age assessment in individuals without reliable identification is a recurrent medico-legal challenge in Europe, particularly in migration contexts. Distal radial epiphyseal fusion is among the last skeletal maturation events near the age of majority, yet population-specific reference data and radiation-free assessment strategies remain limited. Methods: In this multicentre study conducted across German clinical sites and initial reception facilities, Northwest European (NWE) and Middle Eastern (ME) individuals aged 13-25 years underwent distal forearm assessment using a fully automated ultrasound device. Reference ages were obtained from documented birthdates or Greulich-Pyle bone ages. Gradient-boosting machine-learning models (LightGBM) were evaluated using leave-one-out cross-validation across sex-specific age thresholds. Performance was quantified using precision, recall, F1-scores, ROC/AUC, and Clopper-Pearson confidence intervals for correctly detected fusions. Results: Of 1061 recordings, 830 high-quality ultrasound datasets were analysed across four sex-population clusters. Classification performance was biologically plausible and threshold-dependent, with optimal discrimination at 17.5 years in NWE females, 18.0-18.5 years in NWE males, 17.0-17.5 years in ME females, and 18.5 years in ME males. Females fused earlier than males in both populations, and sex-specific population shifts were observed. AUC values indicated good to excellent discrimination near the optimal thresholds (approximately 0.81-1.00). Conclusions: Machine-learning-assisted automated ultrasound provides a scalable, radiation-free approach for assessing distal radial epiphyseal fusion in late adolescence. Population- and sex-specific shifts underline the importance of population-appropriate reference data when interpreting fusion status around legally critical age thresholds. While an unfused distal radial plate strongly supported minority status, complete fusion alone could not be considered definitive proof of adulthood. Thus reinforcing the need for probabilistic interpretation, particularly around the legally critical 18-year threshold, within a broader forensic age-assessment framework.","author":[{"family":"Birken","given":"Charlotte"},{"family":"Hewener","given":"Holger"},{"family":"Lessmeister-Bastian","given":"Tina"},{"family":"Rohrer","given":"Tilman"},{"family":"Unav"}],"issued":{"date-parts":[[2026]]},"DOI":"10.24406/publica-8676","URL":"https://doi.org/10.24406/publica-8676","source":"datacite"},{"id":"doi:10.17605/osf.io/v8az5","type":"article-journal","title":"Integrating Machine Learning Across the Patient Journey: A Scoping Review and Patient-Oriented Framework","abstract":"This project reports a scoping review examining how machine learning (ML) and artificial intelligence (AI) have been applied across the full patient journey — defined as the longitudinal sequence of care stages spanning prevention and health promotion, diagnosis, prognosis and risk prediction, treatment and clinical decision support, hospital management and operational optimization, and ongoing monitoring and patient experience. The review was conducted following the PRISMA-ScR guidelines, the Arksey and O'Malley framework, and JBI guidance for scoping reviews. We searched PubMed, Scopus, and Web of Science in December 2025 using a predefined string combining ML/AI terminology with patient journey and care pathway concepts. Of 2,041 records retrieved, 99 peer-reviewed sources met all inclusion criteria after deduplication, title/abstract screening, and full-text review. Bibliographic coupling analysis via the Bibliometrix R package was used to organize the sample into thematic clusters for narrative synthesis. The primary purpose of this review is to map where ML has been applied across care stages, identify where cross-phase integration is absent, and characterize the methodological, structural, and ethical gaps that prevent individual tools from cohering into a system-level architecture. A secondary purpose is to use that map as the empirical basis for a proposed integrative framework. The expected outcome is twofold. First, a documented synthesis of the current state of ML deployment across the patient journey, with explicit identification of underrepresented stages — particularly care transitions, rehabilitation, and continuous monitoring. Second, the Integrated Patient Journey AI Framework (IPJ-AIF), a five-component architecture covering multi-source data integration via HL7 FHIR and OMOP-CDM standards, a modular ML pipeline with phase-specific architectures, secure cloud-edge infrastructure, model interpretability mechanisms including SHAP and LIME, and outcome- and equity-oriented evaluation incorporating QALY, ROI, and fairness indicators. The framework is intended as a practical roadmap for researchers and institutions seeking to move beyond siloed ML deployment toward genuinely longitudinal clinical AI. Note: This protocol was registered retrospectively following completion of the search and screening phases.","author":[{"family":"Matias","given":"Gustavo"},{"family":"Araújo","given":"Kauan"},{"family":"Moro","given":"Claudia"},{"family":"Bernardo","given":"Carlos"},{"family":"Lermen","given":"Fernando"},{"family":"Bertoni","given":"Vanessa"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/v8az5","URL":"https://doi.org/10.17605/osf.io/v8az5","source":"datacite"},{"id":"doi:10.6084/m9.figshare.32190618","type":"article-journal","title":"Predicting the risk of venous thromboembolism events in older adults with rheumatoid arthritis after initiating targeted disease-modifying antirheumatic drugs: a comparison of the random survival forest and regularized cox regression models","abstract":"To develop a random survival forest (RSF) machine learning (ML) model for predicting venous thromboembolism (VTE) risk in rheumatoid arthritis (RA) patients initiating biological (b) or targeted synthetic (ts) disease-modifying antirheumatic drugs (DMARDs) and compare its model performance with a regularized Cox regression (RegCox) model. This retrospective cohort study using the 5% Medicare data (2012-2020) identified older RA patients (≥ 65 years) initiating b/tsDMARDs (index date), including tumor necrosis factor inhibitors (TNFi) bDMARDs, non-TNFi bDMARDs, and tsDMARDs between January 1, 2013, through December 31, 2019. Study cohort was followed until an incident composite VTE event or censoring. Data were divided into training (75%) and testing (25%) sets. The RSF model was trained to predict VTE events during the follow-up period in the training set, with the RegCox model as the reference model. The performance of these models was evaluated in the testing data using the C-index. Variable importance of the predictors was assessed. Of 3,648 RA patients, 360 (9.87%) experienced any VTE event. The RSF model had better performance (C-index [95% CI] = 0.609[0.602-0.617]) than the RegCox model (C-index [95% CI] = 0.599[0.597-0.602], p = 0.0021). Variables commonly identified as the top influential variables were varicose veins, inpatient visits, Elixhauser score, emergency room visits, and outpatient visits. The RSF model performed slightly better in identifying VTE in RA patients after b/tsDMARDs initiation than RegCox. Incorporating additional clinical and contextual information beyond claims data may further enhance predictive accuracy in future studies. Biological (b) and targeted synthetic (ts) disease-modifying antirheumatic drugs (DMARDs) are newer medications that can treat rheumatoid arthritis (RA). However, recent randomized controlled clinical trials found the tsDMARDs may increase the risk of venous thromboembolism (VTE) among patients with RA. In this study, we developed a novel random survival forest (RSF) machine learning model to predict the VTE risk involving 3,648 older adults (≥65) with RA who were newly prescribed b/tsDMARD. Our results suggested that the RSF model achieved modest improved performance compared to the RegCox model in predicting the risk of VTE among these older adults with RA. Our findings from the RSF model may help rheumatologists better understand the high-risk RA patient profile who may be at risk of VTE. Incorporating additional variables from other datasets may improve the RSF’s model performance.","author":[{"family":"Huang","given":"Yinan"},{"family":"Bazzazzadehgan","given":"Shadi"},{"family":"Maharjan","given":"Shishir"},{"family":"Lin","given":"Ying"},{"family":"Bentley","given":"John"},{"family":"Agarwal","given":"Sandeep"},{"family":"Yang","given":"Yi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.32190618","URL":"https://doi.org/10.6084/m9.figshare.32190618","source":"datacite"},{"id":"doi:10.6084/m9.figshare.32190618.v1","type":"article-journal","title":"Predicting the risk of venous thromboembolism events in older adults with rheumatoid arthritis after initiating targeted disease-modifying antirheumatic drugs: a comparison of the random survival forest and regularized cox regression models","abstract":"To develop a random survival forest (RSF) machine learning (ML) model for predicting venous thromboembolism (VTE) risk in rheumatoid arthritis (RA) patients initiating biological (b) or targeted synthetic (ts) disease-modifying antirheumatic drugs (DMARDs) and compare its model performance with a regularized Cox regression (RegCox) model. This retrospective cohort study using the 5% Medicare data (2012-2020) identified older RA patients (≥ 65 years) initiating b/tsDMARDs (index date), including tumor necrosis factor inhibitors (TNFi) bDMARDs, non-TNFi bDMARDs, and tsDMARDs between January 1, 2013, through December 31, 2019. Study cohort was followed until an incident composite VTE event or censoring. Data were divided into training (75%) and testing (25%) sets. The RSF model was trained to predict VTE events during the follow-up period in the training set, with the RegCox model as the reference model. The performance of these models was evaluated in the testing data using the C-index. Variable importance of the predictors was assessed. Of 3,648 RA patients, 360 (9.87%) experienced any VTE event. The RSF model had better performance (C-index [95% CI] = 0.609[0.602-0.617]) than the RegCox model (C-index [95% CI] = 0.599[0.597-0.602], p = 0.0021). Variables commonly identified as the top influential variables were varicose veins, inpatient visits, Elixhauser score, emergency room visits, and outpatient visits. The RSF model performed slightly better in identifying VTE in RA patients after b/tsDMARDs initiation than RegCox. Incorporating additional clinical and contextual information beyond claims data may further enhance predictive accuracy in future studies. Biological (b) and targeted synthetic (ts) disease-modifying antirheumatic drugs (DMARDs) are newer medications that can treat rheumatoid arthritis (RA). However, recent randomized controlled clinical trials found the tsDMARDs may increase the risk of venous thromboembolism (VTE) among patients with RA. In this study, we developed a novel random survival forest (RSF) machine learning model to predict the VTE risk involving 3,648 older adults (≥65) with RA who were newly prescribed b/tsDMARD. Our results suggested that the RSF model achieved modest improved performance compared to the RegCox model in predicting the risk of VTE among these older adults with RA. Our findings from the RSF model may help rheumatologists better understand the high-risk RA patient profile who may be at risk of VTE. Incorporating additional variables from other datasets may improve the RSF’s model performance.","author":[{"family":"Huang","given":"Yinan"},{"family":"Bazzazzadehgan","given":"Shadi"},{"family":"Maharjan","given":"Shishir"},{"family":"Lin","given":"Ying"},{"family":"Bentley","given":"John"},{"family":"Agarwal","given":"Sandeep"},{"family":"Yang","given":"Yi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.32190618.v1","URL":"https://doi.org/10.6084/m9.figshare.32190618.v1","source":"datacite"},{"id":"doi:10.5167/uzh-435714","type":"article-journal","title":"A practical guide to the implementation of AI in orthopaedic research‐Part 4: Prerequisites for a successful orthopedics AI‐driven project in terms of interdisciplinary collaboration, data management, ethical approval and technology","abstract":"Translating artificial intelligence (AI) research in orthopedics from proof‐of‐concept studies into production‐grade clinical systems requires the systematic satisfaction of four prerequisite domains: interdisciplinary team architecture, technical data management, ethical and regulatory governance and production‐grade technology and deployment infrastructure. Despite a tenfold increase in orthopedic AI publications, fewer than 6% of studies reach routine clinical deployment, reflecting persistent gaps in each of these domains. This article provides a technically rigorous, evidence‐based framework organized around these four pillars. The interdisciplinary team may be structured using a product‐centric topology that decouples stream‐aligned clinical teams from platform infrastructure teams, following Huffman et al.'s six‐step AI project lifecycle: obtain/curate/label data; establish a reference standard; develop the model; evaluate performance; externally validate and iteratively reinforce until clinical implementation is viable. Data management requires data extraction protocols, integration for bulk exports and a multi‐component de‐identification pipeline. A multi‐stage Institutional Review Board framework governs ethical oversight, scaling from Exempt review for retrospective de‐identified studies to Full Board Review with prospective validation and mandatory human‐override mechanisms for interventional deployment. Responsible clinical deployment requires a multi‐layer Clinical Machine Learning Operations framework, implementing privacy‐preserving deployment, clinical observability, compliance audit trails and human‐in‐the‐loop governance. Model drift has to be monitored with a degradation threshold triggering mandatory human review. Level of Evidence Level V.","author":[{"family":"Longo","given":"Umile"},{"family":"Merone","given":"Mario"},{"family":"Schena","given":"Emiliano"},{"family":"Bandini","given":"Benedetta"},{"family":"Nicodemi","given":"Guido"},{"family":"Zsidai","given":"Bálint"},{"family":"Hilkert","given":"Ann‐sophie"},{"family":"Senorski","given":"Eric"},{"family":"Grassi","given":"Alberto"},{"family":"Ley","given":"Christophe"},{"family":"Herbst","given":"Elmar"},{"family":"Hirschmann","given":"Michael"},{"family":"Kopf","given":"Sebastian"},{"family":"Seil","given":"Romain"},{"family":"Tischer","given":"Thomas"},{"family":"Feldt","given":"Robert"},{"family":"Samuelsson","given":"Kristian"},{"family":"Oettl","given":"Felix"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5167/uzh-435714","URL":"https://doi.org/10.5167/uzh-435714","source":"datacite"},{"id":"doi:10.5061/dryad.x69p8czzx","type":"article-journal","title":"Data from: Comparing exercise with virtual reality gaming on gait and cognition in relapsing-remitting multiple sclerosis: a randomized controlled trial","abstract":"Background: Exercise and virtual reality gaming may mitigate gait and cognitive deficits in relapsing-remitting multiple sclerosis (RRMS). The main aim was to compare the efficacy of both interventions on gait and cognition and gait in RRMS. Secondary aims were to explore the efficacy of both interventions on serum biomarkers and to explore the predictors of treatment response. Methods: Forty-eight participants with RRMS were randomized to exercise (n=19), VR (n=19), or wait-list control (n=10) for eight weeks. Primary outcomes were the 10-meter walk test (10MWT) and the Symbol Digit Modalities Test (SDMT). Secondary outcomes included serum levels of neurofilament light chain (NfL), brain-derived neurotrophic factor (BDNF), and insulin-like growth factor-1 (IGF-1). Extreme Gradient Boosting (XGBoost), Random Forest, and logistic regression models were trained to predict treatment response. Results: The exercise group improved 10MWT performance by 2.41 seconds and increased IGF-1 levels by 100.25 ng/ml, significantly more than the VR and control groups (both p&lt;0.001). The VR group improved on the SDMT by 1.95 points (p=0.001 vs. control; p=0.05 vs. exercise). Both interventions reduced NfL concentrations compared to control (exercise: –2.07 pg/ml; VR: –0.60 pg/ml), with exercise showing a greater reduction than VR (p=0.02). XGBoost demonstrated highest predictive accuracy (10MWT: 87%; SDMT: 86%). SHapley Additive exPlanations (SHAP) analysis identified baseline IGF-1 and BDNF as top predictors of 10MWT, and baseline CognICA, BDNF, and age as predictors of SDMT performance. Conclusion: Exercise preferentially improves gait and IGF-1, whereas VR gaming yields modest cognitive gains. Serum biomarkers enhance machine learning prediction of treatment response, supporting a precision rehabilitation approach in RRMS. Keywords: Multiple Sclerosis, Virtual Reality, Exercise, Gait, Cognition, Biomarkers, Machine Learning","author":[{"family":"Sadeghi","given":"Maryam"},{"family":"Kordi","given":"Mohammadreza"},{"family":"Daemi","given":"Mehdi"},{"family":"Tabasi","given":"Seyed"},{"family":"Ebrahimnezhad Bashiri","given":"Mohammad"},{"family":"Nabavi","given":"Seyed"},{"family":"Khaligh-Razavi","given":"Seyed"},{"family":"Thompson","given":"Jeffrey"},{"family":"Sosnoff","given":"Jacob"},{"family":"Devos","given":"Hannes"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5061/dryad.x69p8czzx","URL":"https://doi.org/10.5061/dryad.x69p8czzx","source":"datacite"},{"id":"doi:10.6084/m9.figshare.31171437","type":"article-journal","title":"Evaluation of supervised machine learning models in predicting temporomandibular joint disc displacement on 3T magnetic resonance imaging","abstract":"To evaluate supervised machine learning (ML) models for classifying temporomandibular joint (TMJ) disc displacement on MRI using morphometric and signal intensity features. This retrospective study analyzed 324 TMJs from 162 individuals who underwent 3T MRI. Extracted features included condylar anteroposterior and mediolateral diameters, disc and condyle morphology, and lateral pterygoid muscle signal intensity ratios. Six ML algorithms (Random Forest, Gaussian Naïve Bayes, Logistic Regression, AdaBoost, Gradient Boost, and k-Nearest Neighbor) were evaluated using stratified 5-fold cross-validation. Performance was assessed with accuracy, macro-averaged recall, precision, F1-score, and ROC-AUC. All models demonstrated good classification performance (ROC-AUC &gt;0.80). AdaBoost achieved the highest ROC-AUC (0.88), while Gaussian Naïve Bayes showed the most balanced overall metrics. Mediolateral condylar diameter and disc morphology were key features associated with disc displacement categories. ML models can identify MRI-based morphometric patterns related to TMJ disc displacement and may support radiologic assessment, while clinical diagnosis should continue to rely on established standards of care.","author":[{"family":"Erol","given":"Seyit"},{"family":"Özer","given":"Halil"},{"family":"Gürhan","given":"Abdi"},{"family":"Koplay","given":"Mustafa"},{"family":"Erol","given":"Çağlagül"},{"family":"Seher","given":"Nusret"},{"family":"Öztürk","given":"Mehmet"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.31171437","URL":"https://doi.org/10.6084/m9.figshare.31171437","source":"datacite"},{"id":"doi:10.6084/m9.figshare.31171437.v1","type":"article-journal","title":"Evaluation of supervised machine learning models in predicting temporomandibular joint disc displacement on 3T magnetic resonance imaging","abstract":"To evaluate supervised machine learning (ML) models for classifying temporomandibular joint (TMJ) disc displacement on MRI using morphometric and signal intensity features. This retrospective study analyzed 324 TMJs from 162 individuals who underwent 3T MRI. Extracted features included condylar anteroposterior and mediolateral diameters, disc and condyle morphology, and lateral pterygoid muscle signal intensity ratios. Six ML algorithms (Random Forest, Gaussian Naïve Bayes, Logistic Regression, AdaBoost, Gradient Boost, and k-Nearest Neighbor) were evaluated using stratified 5-fold cross-validation. Performance was assessed with accuracy, macro-averaged recall, precision, F1-score, and ROC-AUC. All models demonstrated good classification performance (ROC-AUC &gt;0.80). AdaBoost achieved the highest ROC-AUC (0.88), while Gaussian Naïve Bayes showed the most balanced overall metrics. Mediolateral condylar diameter and disc morphology were key features associated with disc displacement categories. ML models can identify MRI-based morphometric patterns related to TMJ disc displacement and may support radiologic assessment, while clinical diagnosis should continue to rely on established standards of care.","author":[{"family":"Erol","given":"Seyit"},{"family":"Özer","given":"Halil"},{"family":"Gürhan","given":"Abdi"},{"family":"Koplay","given":"Mustafa"},{"family":"Erol","given":"Çağlagül"},{"family":"Seher","given":"Nusret"},{"family":"Öztürk","given":"Mehmet"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.31171437.v1","URL":"https://doi.org/10.6084/m9.figshare.31171437.v1","source":"datacite"},{"id":"doi:10.82992/intellectum/handle.10818.7197","type":"article-journal","title":"Evaluating AI Tools for COPD Management: NLP-Based Identification of Exacerbations in Local EHRs","abstract":"In this study we evaluate the performance of classical machine learning models, and transformer?based models (BioClinicalBERT and large language models (LLMs) for the automatic identification of COPD exacerbations using unstructured clinical texts from Colombian electronic health records (EHRs). The dataset included 5,924 outpatient notes written in Spanish and manually labeled by a pulmonologist, incorporating two key fields: “subjective” (patient-reported symptoms) and “analysis” (physician assessment). This work addresses a critical gap, as no prior models have been developed or validated for exacerbation detection in Spanish-language clinical texts. Results show that classical models—especially LightGBM and CatBoost—achieved the best balance between discrimination and calibration (ECE &lt; 1.3%), outperforming BioClinicalBERT and LLMs. These findings support the use of supervised, domain-adapted models for decision support in resource-limited clinical settings","author":[{"family":"Campos Lozano","given":"Jonathan"},{"family":"Ruiz Rodriguez","given":"Nini"},{"family":"Jaimes Fernandez","given":"Diego"},{"family":"Jaimes Fernandez","given":"Diego"},{"family":"Toledo Cortes","given":"Santiago"},{"family":"Toledo Cortes","given":"Santiago"}],"issued":{"date-parts":[[2025]]},"DOI":"10.82992/intellectum/handle.10818.7197","URL":"https://doi.org/10.82992/intellectum/handle.10818.7197","source":"datacite"},{"id":"doi:10.5281/zenodo.20081849","type":"article-journal","title":"Cognitive Retrain: A Web-Based Cognitive Skill Enhancement and Early Detection System for Children with ASD And Dyslexia","abstract":"Early detection and continuous therapeutic intervention are critical for children facing cognitive learning challenges, particularly Autism Spectrum Disorder (ASD) and Dyslexia. Traditional clinical diagnostic methodologies are resource-intensive, highly susceptible to observational bias, and lack continuous engagement mechanisms post-diagnosis. This research presents Cognitive Retrain, an integrated web-based platform engineered to facilitate early cognitive disorder screening and gamified therapeutic enhancement. Leveraging a decoupled microservices architecture, the system employs a Node.js and React infrastructure for low-latency user interaction, operating in tandem with a Python-based Machine Learning backend. Supervised learning classification models—specifically Random Forest ensembles and Support Vector Machines (SVM)—were trained on behavioral datasets to predict the likelihood of ASD and Dyslexia. Furthermore, the platform introduces a suite of adaptive therapeutic games targeting memory retention, visuospatial attention, linguistic processing, and emotional recognition, dynamically served via a cosine similarity recommendation engine. To bridge the gap in caregiver education, an advanced Retrieval-Augmented Generation (RAG) conversational agent is embedded, querying a FAISS-indexed ChromaDB vector database of verified medical literature. Empirical benchmarking demonstrates an exceptional 93.5% accuracy for the autism predictive model and an 89.2% accuracy for the dyslexia model, with an overall diagnostic inference latency under 150 milliseconds. The results highlight the efficacy of combining machine learning algorithms with customized gamified digital therapeutics for remote cognitive rehabilitation.","author":[{"family":"Narsimhulu","given":"Mr"},{"family":"Sai","given":"KG"},{"family":"Kashyap","given":"GG"},{"family":"Mangalarapu","given":"Gayathri"},{"family":"Deepthi","given":"K"},{"family":"Rohan","given":"SVSR"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20081849","URL":"https://doi.org/10.5281/zenodo.20081849","source":"datacite"},{"id":"doi:10.5281/zenodo.20081850","type":"article-journal","title":"Cognitive Retrain: A Web-Based Cognitive Skill Enhancement and Early Detection System for Children with ASD And Dyslexia","abstract":"Early detection and continuous therapeutic intervention are critical for children facing cognitive learning challenges, particularly Autism Spectrum Disorder (ASD) and Dyslexia. Traditional clinical diagnostic methodologies are resource-intensive, highly susceptible to observational bias, and lack continuous engagement mechanisms post-diagnosis. This research presents Cognitive Retrain, an integrated web-based platform engineered to facilitate early cognitive disorder screening and gamified therapeutic enhancement. Leveraging a decoupled microservices architecture, the system employs a Node.js and React infrastructure for low-latency user interaction, operating in tandem with a Python-based Machine Learning backend. Supervised learning classification models—specifically Random Forest ensembles and Support Vector Machines (SVM)—were trained on behavioral datasets to predict the likelihood of ASD and Dyslexia. Furthermore, the platform introduces a suite of adaptive therapeutic games targeting memory retention, visuospatial attention, linguistic processing, and emotional recognition, dynamically served via a cosine similarity recommendation engine. To bridge the gap in caregiver education, an advanced Retrieval-Augmented Generation (RAG) conversational agent is embedded, querying a FAISS-indexed ChromaDB vector database of verified medical literature. Empirical benchmarking demonstrates an exceptional 93.5% accuracy for the autism predictive model and an 89.2% accuracy for the dyslexia model, with an overall diagnostic inference latency under 150 milliseconds. The results highlight the efficacy of combining machine learning algorithms with customized gamified digital therapeutics for remote cognitive rehabilitation.","author":[{"family":"Narsimhulu","given":"Mr"},{"family":"Sai","given":"KG"},{"family":"Kashyap","given":"GG"},{"family":"Mangalarapu","given":"Gayathri"},{"family":"Deepthi","given":"K"},{"family":"Rohan","given":"SVSR"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20081850","URL":"https://doi.org/10.5281/zenodo.20081850","source":"datacite"},{"id":"doi:10.17605/osf.io/ey4sj","type":"article-journal","title":"External validation of prediction models for induction of labor outcomes: Individual Participant Data (IPD) Meta-Analysis – A study protocol","abstract":"This protocol is a systematic review to identify and select robust induction of labor (IOL) outcome prediction models to externally validate selected clinical prediction models. We will use individual participant data (IPD) from published randomised controlled trials (RCTs) to externally validate selected models. We will identify IOL prediction models from previously published systematic reviews on IOL outcome prediction and update the search after the recent review dated 24 April 2023. The search was conducted in the PubMed/MEDLINE, Embase, Scopus, and Web of Science databases. All identified IOL prediction models will be assessed for eligibility for external validation. Two reviewers will independently undertake study selection, data extraction, and risk-of-bias assessment using the PROBAST+AI tool. Only prediction models developed to predict IOL outcomes, most commonly caesarean section or vaginal delivery, will be considered for external validation. IOL prediction models will be excluded from external validation if they meet one or more of the following conditions: 1) inclusion of costly or non-routinely collected predictors (for example, predictors requiring expensive laboratory tests or specialised ultrasound measurements), 2) models included predictors collected after IOL initiation, 3) prediction model studies developed using data that could not establish the correct temporal relationship between predictors and outcome, such as a cross-sectional study design, 4) prediction model studies development using data from a case-control study without recalibration using external data on true prevalence, 5) prediction models developed using only a single predictor, such as the Bishop score, 6) prediction model studies with a high risk of bias, 7) prediction model studies that do not report parameters or specifications will be excluded unless the information can be obtained by contacting the authors, 8) inability to be validated using the validation data. For example, a model was developed in selected populations that are not sufficiently represented in the validation cohort.","author":[{"family":"Tiruneh","given":"Sofonyas"},{"family":"Hu","given":"Yanan"},{"family":"Au","given":"Ling"},{"family":"Mol","given":"Ben"},{"family":"Li","given":"Wentao"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/ey4sj","URL":"https://doi.org/10.17605/osf.io/ey4sj","source":"datacite"},{"id":"doi:10.17605/osf.io/cwzhu","type":"article-journal","title":"Artificial Intelligence-Enabled Wearable Devices for Cardiovascular Disease Screening, Monitoring, and Intervention: An Umbrella Review of Systematic Reviews","abstract":"Background Cardiovascular disease (CVD) remains the top cause of global mortality and imposes heavy chronic disease burden worldwide. In China, the prevalence and mortality of CVD have risen sharply over the past decades, accompanied by younger onset age and more asymptomatic hidden cardiac lesions. Traditional cardiac examination methods rely on intermittent hospital testing, which cannot capture transient silent cardiovascular events and lack long-term continuous physiological tracking. AI-integrated wearable devices break this limitation by real-time collecting heart rhythm, blood pressure and heart rate variability signals, enabling automated early warning and personalized cardiovascular management. Numerous published systematic reviews and meta‑analyses have explored the diagnostic and management value of AI‑enabled wearables in CVD. However, discrepancies in research conclusions, varied methodological quality, and overlapping primary study samples across existing reviews hinder the accurate interpretation and clinical translation of current evidence. Existing umbrella reviews in this field are mostly limited to single clinical scenarios, and there remains a lack of systematic synthesis and evidence grading targeting high‑quality systematic reviews and meta‑analyses covering the full spectrum of screening, monitoring and intervention. This preregistered umbrella review aims to systematically identify high‑quality systematic reviews and meta‑analyses in this field, conduct standardized methodological quality appraisal and evidence certainty grading, and provide a clearer evidence base for clinical cardiovascular practice. Primary Study Aims 1. Systematically identify and synthesize high-quality published systematic reviews and meta-analyses on AI-enabled wearables for CVD screening, monitoring and intervention, and summarize the evidence by cardiovascular disease subtypes and clinical application scenarios. 2. Appraise the methodological quality of included reviews using the AMSTAR 2 tool, and analyze the distribution of quality defects in current research. 3. Calculate the Corrected Covered Area (CCA) to quantify the overlap degree of primary studies among included reviews. 4. Grade the certainty of clinical evidence for core outcomes with the GRADE framework, so as to provide reference for clinical decision-making and follow-up original research design. The results will guide clinical cardiovascular management, health policy formulation and future original research design. Eligibility Criteria (PICOS) Population: Adults aged ≥18 years。 Eligible populations are classified into three categories: 1.CVD high-risk populations: individuals with at least one cardiovascular risk factor (hypertension, dyslipidemia, diabetes, smoking, family history of CVD, etc.) 2.Patients with suspected undiagnosed CVD: individuals with suspicious symptoms or abnormal screening indicators who have not received a definitive diagnosis 3.Patients with confirmed CVD: patients definitively diagnosed with hypertension, atrial fibrillation, heart failure, coronary heart disease, or ischemic/hemorrhagic stroke Excluded populations: minors under 18 years old, pregnant women, and patients with severe end‑stage comorbidities that independently affect cardiovascular indicators. Intervention: Wearable devices embedded with artificial intelligence algorithms (including machine learning and deep learning models) are applied in three major clinical scenarios. The eligible devices include smartwatches, wearable single‑lead ECG monitors, cuffless wearable blood pressure monitors, and wrist‑worn fitness trackers. The specific clinical scenarios are defined as follows: 1.Cardiovascular disease screening and risk prediction: Identify arrhythmias and abnormal blood pressure with artificial intelligence, and complete cardiovascular risk stratification. 2.Continuous physiological‑parameter monitoring: Perform real‑time dynamic monitoring of cardiovascular‑related indic","author":[{"family":"薛则佩"},{"family":"丁雄"},{"family":"Tian","given":"Maoyi"},{"family":"Tang","given":"Ning"},{"family":"Shui","given":"Dong"},{"family":"Zhang","given":"Ruolin"},{"family":"杜雪"},{"family":"陈萌萌"},{"family":"Zhang","given":"Jing"},{"family":"Tian","given":"Wei"},{"family":"张馨艺"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/cwzhu","URL":"https://doi.org/10.17605/osf.io/cwzhu","source":"datacite"},{"id":"doi:10.17605/osf.io/eqhjx","type":"article-journal","title":"Digital Intelligence technology empowers interprofessional collaboration and the well-being of medical social workers: a scoping review on health promotion for medical social workers","abstract":"Health and social workers (HSWs) and community health workers (CHWs) are critical to health equity in low- and middle-income countries (LMICs), yet they face persistent knowledge gaps, resource constraints, and limited institutional support. Digital technologies—from machine learning to large language models—offer new pathways for empowerment, but existing research has largely focused on clinical efficiency rather than on worker well-being and collaborative practice. This scoping review, following PRISMA-ScR guidelines and synthesizing 14 empirical studies published between 2016 and 2026, identifies three core application scenarios (prediction, intervention, evaluation) and four well-being promotion mechanisms (information relief, decision empowerment, supervisory care, and efficiency burden reduction). We argue that digital technologies are driving a fundamental shift from interpersonal to human-machine collaboration, reshaping trust structures, task divisions, and professional roles — yet this transformation is accompanied by three paradoxes: empowerment versus dependency, efficiency versus meaning, and care versus surveillance. We conclude that sustainable digital transformation in health social work requires long-term evaluation, standardized well-being metrics, attention to digital equity, and the formal integration of worker well-being into technology assessment frameworks.","author":[{"family":"Ma","given":"Jiyan"},{"family":"Dai","given":"Yuke"},{"family":"Kang","given":"Jie"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/eqhjx","URL":"https://doi.org/10.17605/osf.io/eqhjx","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8666056","type":"article-journal","title":"Validating objective and scalable speech markers of depression across two independent psychiatric cohorts","abstract":"Abstract Background Using speech as objective markers for major depressive disorder (MDD) has shown promise, yet their generalizability across clinical settings remains largely unvalidated. Objective This study aimed to validate previously identified speech markers of depressive symptoms in an independent clinical cohort, thereby assessing their reproducibility and robustness for cross-site application. Methods Speech data from two independent psychiatric cohorts (RWTH Aachen and University of Oldenburg, Germany) were analyzed, comprising 135 participants (71 healthy controls, 64 MDD patients). Participants completed a positive and a negative storytelling task, over 80 temporal, lexical, and spectral speech features were extracted from the acoustic signal. Statistical analyses assessed group differences and correlations with Beck Depression Inventory (BDI-II) scores. Machine learning models trained on the Aachen data were tested on the Oldenburg cohort. Results Several temporal and spectral speech features, including utterance duration, pause duration, and MFCCs, were consistently associated with MDD diagnosis and symptom severity across both cohorts. Machine learning models trained on Aachen data achieved a classification accuracy (ROC-AUC) of 0.63 on the Oldenburg sample, demonstrating above-chance but modest transfer performance. Voice quality features (shimmer, jitter) showed more variable associations: partial correlations indicated some significant effects (e.g., shimmer and jitter during positive storytelling), whereas moderation analyses revealed interaction effects, particularly for shimmer and jitter in negative storytelling, where MDD patients exhibited higher values in the Aachen cohort but lower values in the Oldenburg cohort compared to healthy controls. Conclusions The study indicates that temporal and spectral markers of speech are relatively robust across independent clinical samples, whereas voice quality markers (shimmer, jitter) show site-dependent inconsistencies, acting as technical artifacts of varying recording conditions rather than robust biomarkers. While current speech-based classifiers remain less accurate than established self-report measures, their integration with clinical scores offers a more balanced trade-off between sensitivity and specificity. Future work should prioritize systematic evaluation across elicitation tasks, languages, and longitudinal settings to delineate which speech features are transferable and which are task-specific.","author":[{"family":"Menne","given":"Felix"},{"family":"Dörr","given":"Felix"},{"family":"Tröger","given":"Johannes"},{"family":"König","given":"Alexandra"},{"family":"Schräder","given":"Julia"},{"family":"Immel","given":"Diana"},{"family":"Hurlemann","given":"René"},{"family":"Barton","given":"Simon"},{"family":"Wagels","given":"Lisa"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8666056","URL":"https://doi.org/10.6084/m9.figshare.c.8666056","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8666056.v1","type":"article-journal","title":"Validating objective and scalable speech markers of depression across two independent psychiatric cohorts","abstract":"Abstract Background Using speech as objective markers for major depressive disorder (MDD) has shown promise, yet their generalizability across clinical settings remains largely unvalidated. Objective This study aimed to validate previously identified speech markers of depressive symptoms in an independent clinical cohort, thereby assessing their reproducibility and robustness for cross-site application. Methods Speech data from two independent psychiatric cohorts (RWTH Aachen and University of Oldenburg, Germany) were analyzed, comprising 135 participants (71 healthy controls, 64 MDD patients). Participants completed a positive and a negative storytelling task, over 80 temporal, lexical, and spectral speech features were extracted from the acoustic signal. Statistical analyses assessed group differences and correlations with Beck Depression Inventory (BDI-II) scores. Machine learning models trained on the Aachen data were tested on the Oldenburg cohort. Results Several temporal and spectral speech features, including utterance duration, pause duration, and MFCCs, were consistently associated with MDD diagnosis and symptom severity across both cohorts. Machine learning models trained on Aachen data achieved a classification accuracy (ROC-AUC) of 0.63 on the Oldenburg sample, demonstrating above-chance but modest transfer performance. Voice quality features (shimmer, jitter) showed more variable associations: partial correlations indicated some significant effects (e.g., shimmer and jitter during positive storytelling), whereas moderation analyses revealed interaction effects, particularly for shimmer and jitter in negative storytelling, where MDD patients exhibited higher values in the Aachen cohort but lower values in the Oldenburg cohort compared to healthy controls. Conclusions The study indicates that temporal and spectral markers of speech are relatively robust across independent clinical samples, whereas voice quality markers (shimmer, jitter) show site-dependent inconsistencies, acting as technical artifacts of varying recording conditions rather than robust biomarkers. While current speech-based classifiers remain less accurate than established self-report measures, their integration with clinical scores offers a more balanced trade-off between sensitivity and specificity. Future work should prioritize systematic evaluation across elicitation tasks, languages, and longitudinal settings to delineate which speech features are transferable and which are task-specific.","author":[{"family":"Menne","given":"Felix"},{"family":"Dörr","given":"Felix"},{"family":"Tröger","given":"Johannes"},{"family":"König","given":"Alexandra"},{"family":"Schräder","given":"Julia"},{"family":"Immel","given":"Diana"},{"family":"Hurlemann","given":"René"},{"family":"Barton","given":"Simon"},{"family":"Wagels","given":"Lisa"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8666056.v1","URL":"https://doi.org/10.6084/m9.figshare.c.8666056.v1","source":"datacite"},{"id":"doi:10.48550/arxiv.2608.19965","type":"manuscript","title":"Flow Matching Meets 3D Curvilinear Structure Segmentation in Medical Imaging","abstract":"Segmentation of curvilinear anatomical structures in 3D medical images remains challenging due to complex topology, severe class imbalance, weak contrast, and large variations in structure morphology. While deep learning approaches for 3D curvilinear segmentation have been proposed, they are often tailored to specific anatomies or modalities, limiting generalization across clinical settings and leaving room for improvement. Recent generative models have shown the benefits of iterative prediction for structured segmentation tasks, yet diffusion-based methods suffer from computationally expensive sampling, hindering their use on high-resolution 3D volumes. We present 3D-CurvSegFlow, a flow matching-based model for 3D curvilinear structure segmentation. The model learns a continuous transformation from a simple source distribution to the target vascular representation, enabling progressive refinement of complex curvilinear geometries with efficient inference. We evaluate our method on Three public challenging datasets covering distinct anatomies and modalities: portal vein, cerebral vessel, and coronary arteries. Using a common architecture and training strategy across all tasks, our method outperforms general-purpose and vessel-specific approaches, with strong preservation of thin branches and vascular continuity. This work not only advances the state-of-the-art in 3D curvilinear segmentation but also opens new avenues for efficient, generalizable, and clinically applicable methods in medical image analysis.","author":[{"family":"Moctar","given":"Sidi"},{"family":"Vitry","given":"Nicolas"},{"family":"Bouvrais","given":"Hélène"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2608.19965","URL":"https://doi.org/10.48550/arxiv.2608.19965","source":"datacite"},{"id":"doi:10.5281/zenodo.22002707","type":"article-journal","title":"Data and Code for Machine Learning Based Reference Assisted Signal Recovery Under Anatomical Domain Shift in Simulated Radiofrequency Fracture Sensing","abstract":"Version 1.1.0 of the public data-and-code release supporting the article “Machine Learning Based Reference Assisted Signal Recovery Under Anatomical Domain Shift in Simulated Radiofrequency Fracture Sensing”. The release contains the canonical processed feature matrix comprising 16,740 records and 1,297 candidate descriptors, H2–H5 grouping manifests, the 1,235 modelling-descriptor definitions, held-out machine-learning predictions, calibration and operating-point outputs, hierarchical confidence intervals, false-negative burden analyses, five-seed LightGBM stability results, descriptor-stratified effects, MATLAB simulation and feature-extraction source code, Python analysis and reproduction scripts,software-environment specifications, and integrity checks. The release contains simulated data only. No human-participant, animal, personal, confidential, or patient-identifiable data are included. The exact matched-reference pathway is an engineering upper bound and should not be interpreted as ordinary screening or clinical-performance evidence. Version 1.1.0 supersedes Version 1.0.2. The canonical matrix, labels, primary held-out predictions, and reported numerical findings are unchanged. The public documentation, organisation, licensing, portability, and machine-readable metadata have been improved.","author":[{"family":"Aldelemy","given":"Ahmad"},{"family":"Al Dulaimi","given":"Ali"},{"family":"Abd-Alhameed","given":"Raed"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22002707","URL":"https://doi.org/10.5281/zenodo.22002707","source":"datacite"},{"id":"doi:10.5281/zenodo.22030635","type":"article-journal","title":"Data and Code for Machine Learning Based Reference Assisted Signal Recovery Under Anatomical Domain Shift in Simulated Radiofrequency Fracture Sensing","abstract":"Version 1.1.0 of the public data-and-code release supporting the article “Machine Learning Based Reference Assisted Signal Recovery Under Anatomical Domain Shift in Simulated Radiofrequency Fracture Sensing”. The release contains the canonical processed feature matrix comprising 16,740 records and 1,297 candidate descriptors, H2–H5 grouping manifests, the 1,235 modelling-descriptor definitions, held-out machine-learning predictions, calibration and operating-point outputs, hierarchical confidence intervals, false-negative burden analyses, five-seed LightGBM stability results, descriptor-stratified effects, MATLAB simulation and feature-extraction source code, Python analysis and reproduction scripts,software-environment specifications, and integrity checks. The release contains simulated data only. No human-participant, animal, personal, confidential, or patient-identifiable data are included. The exact matched-reference pathway is an engineering upper bound and should not be interpreted as ordinary screening or clinical-performance evidence. Version 1.1.0 supersedes Version 1.0.2. The canonical matrix, labels, primary held-out predictions, and reported numerical findings are unchanged. The public documentation, organisation, licensing, portability, and machine-readable metadata have been improved.","author":[{"family":"Aldelemy","given":"Ahmad"},{"family":"Al Dulaimi","given":"Ali"},{"family":"Abd-Alhameed","given":"Raed"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22030635","URL":"https://doi.org/10.5281/zenodo.22030635","source":"datacite"},{"id":"doi:10.5683/sp4/bl2yta","type":"article-journal","title":"A Labeled Accelerometry Dataset of Steps and Postures Across Diverse Mobility and Frailty Levels in Older Adults in a Hospital Setting","abstract":"The Mobility in Geriatrics (Mobi-G) dataset is an open-access dataset designed to support the development, validation, and benchmarking of machine learning algorithms for posture classification and step counting in older adults. The dataset includes ankle-worn Movesense accelerometry recordings (52 Hz), manual steps and postures annotations, and de-identified sociodemographic and clinical characteristics from 70 adults aged 65 years and older recruited at the Centre hospitalier de l'Université de Montréal (CHUM), Canada. Accelerometry data were collected during both standardized mobility assessments (lying, sitting, standing, transfers, regular walking, irregular walking and physiotherapy sessions) and free-living activities. Steps during walking activities were manually annotated using synchronized video recordings, while postures and transfers were annotated directly from the accelerometry signals. The repository includes raw and annotated accelerometry files and participant-level demographic and clinical variables. These data are intended to facilitate research in clinical mobility monitoring, digital biomarkers, wearable sensing, and artificial intelligence for geriatric care.","author":[{"family":"Lanoie","given":"Annik"},{"family":"Cheung","given":"Weng"},{"family":"Nguyen","given":"Quoc"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5683/sp4/bl2yta","URL":"https://doi.org/10.5683/sp4/bl2yta","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33296440","type":"article-journal","title":"Development and internal validation of a machine learning-based disease burden index for irritable bowel syndrome: a multicentre cross-sectional study","abstract":"Irritable bowel syndrome (IBS) is one of the most common disorders of gut-brain interaction in gastroenterology clinics worldwide, becoming a significant public health burden. This study aims to provide an exploratory approach for comprehensive assessment of IBS burden and may serve as a basis for future external validation and further evaluation of its potential clinical application. This multicentre cross-sectional study included IBS patients diagnosed according to the Rome III and/or Rome IV criteria. Rome criteria-based comparisons and latent class analysis (LCA) were performed to characterize clinical heterogeneity, followed by the development of the IBS burden index (IBS-BI) using machine learning. Compared to the Rome IV-ineligible group, the Rome IV-eligible group exhibited significantly more severe IBS symptoms, somatization symptoms, anxiety symptoms, depression symptoms, and poorer health-related quality of life (all p &lt;0.05). Linear regression analysis showed that the severity of these symptoms was inversely related to health-related quality of life (all p &lt;0.05). LCA identified four heterogeneous IBS burden phenotypes: high-symptom burden, low-symptom burden, somatization-dominant, and psychological comorbid phenotypes. Using machine learning, we developed the IBS-BI and an interactive clinical calculator. The validation was limited to an internal training-validation split, and no external validation was performed. Patients with IBS who fulfilled the Rome IV criteria showed a higher disease burden, which may be related to the more stringent diagnostic requirements of these criteria. The IBS-BI may have potential value for evaluating disease burden and patient stratification. Its clinical utility and broader applicability require further validation in independent external cohorts and prospective studies.","author":[{"family":"Wei","given":"Yanbin"},{"family":"Zhang","given":"Heyang"},{"family":"Wang","given":"Nan"},{"family":"Jin","given":"Haifeng"},{"family":"Feng","given":"Zitan"},{"family":"Guo","given":"Jiapeng"},{"family":"Peng","given":"Jiafei"},{"family":"Feng","given":"Jia"},{"family":"Zhi","given":"Jia"},{"family":"Zhang","given":"Shutian"},{"family":"Zhu","given":"Shengtao"},{"family":"Yao","given":"Xin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33296440","URL":"https://doi.org/10.6084/m9.figshare.33296440","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33296440.v1","type":"article-journal","title":"Development and internal validation of a machine learning-based disease burden index for irritable bowel syndrome: a multicentre cross-sectional study","abstract":"Irritable bowel syndrome (IBS) is one of the most common disorders of gut-brain interaction in gastroenterology clinics worldwide, becoming a significant public health burden. This study aims to provide an exploratory approach for comprehensive assessment of IBS burden and may serve as a basis for future external validation and further evaluation of its potential clinical application. This multicentre cross-sectional study included IBS patients diagnosed according to the Rome III and/or Rome IV criteria. Rome criteria-based comparisons and latent class analysis (LCA) were performed to characterize clinical heterogeneity, followed by the development of the IBS burden index (IBS-BI) using machine learning. Compared to the Rome IV-ineligible group, the Rome IV-eligible group exhibited significantly more severe IBS symptoms, somatization symptoms, anxiety symptoms, depression symptoms, and poorer health-related quality of life (all p &lt;0.05). Linear regression analysis showed that the severity of these symptoms was inversely related to health-related quality of life (all p &lt;0.05). LCA identified four heterogeneous IBS burden phenotypes: high-symptom burden, low-symptom burden, somatization-dominant, and psychological comorbid phenotypes. Using machine learning, we developed the IBS-BI and an interactive clinical calculator. The validation was limited to an internal training-validation split, and no external validation was performed. Patients with IBS who fulfilled the Rome IV criteria showed a higher disease burden, which may be related to the more stringent diagnostic requirements of these criteria. The IBS-BI may have potential value for evaluating disease burden and patient stratification. Its clinical utility and broader applicability require further validation in independent external cohorts and prospective studies.","author":[{"family":"Wei","given":"Yanbin"},{"family":"Zhang","given":"Heyang"},{"family":"Wang","given":"Nan"},{"family":"Jin","given":"Haifeng"},{"family":"Feng","given":"Zitan"},{"family":"Guo","given":"Jiapeng"},{"family":"Peng","given":"Jiafei"},{"family":"Feng","given":"Jia"},{"family":"Zhi","given":"Jia"},{"family":"Zhang","given":"Shutian"},{"family":"Zhu","given":"Shengtao"},{"family":"Yao","given":"Xin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33296440.v1","URL":"https://doi.org/10.6084/m9.figshare.33296440.v1","source":"datacite"},{"id":"doi:10.5281/zenodo.18887131","type":"article-journal","title":"Integrated AI-based Care Model Library I","abstract":"Deliverable D3.7, titled “Integrated AI-based Care Model Library I”, consolidates the state-of-the-art (SOTA) analysis, and the methodologies explored in tasks T3.5 – “Explainable Models and Algorithms for Trustworthy AI” and T3.6 – “Integrated Care Model Library: Implementation and Recalibration of Adaptive Models” of Work Package 3 (WP3 – “Data and AI Governance for Personalized Prediction, Monitoring and Recommendations”) of the project. Deliverable D3.7 is the first version of a two-part series of deliverables, namely D3.7 - “Integrated AI-based Care Model Library I” and D3.8 - “Integrated AI-based Care Model Library II” (to be submitted on M32). This deliverable D3.7 highlights the innovative research efforts being made in integrating sophisticated SOTA models and advanced Machine Learning (ML) technologies into the Virtual Health Platform (VHP). The deliverable further demonstrates how these advancements contribute to achieving the project’s overarching objectives. From the collective work conducted in WP3, this deliverable serves as both a practical guide and a repository of insights for stakeholders, transferring knowledge and fostering the adoption of advanced Integrated Care Model (ICM) technologies in healthcare. This document underscores the project’s dedication to driving impactful advancements and setting a benchmark for the future of predictive and preventive healthcare systems. There are several major challenges that should be addressed during the lifecycle of this project, which include: i) the integration of heterogeneous data sources into high-quality and bias-free datasets, ii) scalability and real-time processing of developed algorithms, iii) adaptability of AI/ML models to diverse and extremely complex healthcare settings, iv) explainability of AI models and their ethical compliance, and finally v) rigorous evaluation and validation. As this deliverable will demonstrate recent advancements in AI, and particularly multimodal processing can greatly assist in some of these challenges, given that they can process and integrate a wide range data sources to provide a holistic understanding of patient health with increased personalised accuracy and adaptability to new clinical settings.This report introduces key concepts in multimodal systems engineering as required for the development of the Integrated Care Model Library (ICML). This deliverable explores the advanced prediction analytical tools that will be developed to enhance the VHP and ultimately empower it with advanced AI methodologies. More specifically, the main objectives of this deliverable include the development of a novel methodology for ICML with an objective to promote accurate predictions and explainable outcomes across European populations irrespective of ethnicity or socioeconomic background. This deliverable also aims to tackle the ethical compliance aspects of the ICML and serve in the facilitation of a reliable framework that meaningfully assists clinicians in the decision-making process. Deliverable D3.7, titled “Integrated AI-based Care Model Library I,” establishes the foundation for achieving these objectives. More specifically, this document constitutes a detailed account of the state-of-the-art, methodologies, strategies, and outcomes associated with integrated care models, emphasising on its alignment with the broader goals of the COMFORTage project mission. The document bridges the gap between theoretical advancements and practical implementation, presenting solutions that are both scientifically robust and operationally feasible. A key focus of this deliverable is the alignment of project objectives with the complexity of the challenges identified. These challenges underscore the importance of a systematic and collaborative approach, as exemplified by the methodologies and frameworks described in this document.","author":[{"family":"Tsitiridis","given":"Aristeidis"},{"family":"Tsolis","given":"Dimitrios"},{"family":"Lykothanasi","given":"Kalliopi"},{"family":"Tsoukalos","given":"Dimitrios"},{"family":"Koutsomitropoulos","given":"Dimitrios"},{"family":"Giannaros","given":"Anastasios"},{"family":"Huang","given":"Shulei"},{"family":"Berenguer-Sánchez","given":"José"},{"family":"Van Den Heuvel","given":"Willem"},{"family":"Pilz","given":"Maximilian"},{"family":"Van Berlo","given":"Sander"},{"family":"Spathoulas","given":"Georgios"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18887131","URL":"https://doi.org/10.5281/zenodo.18887131","source":"datacite"},{"id":"doi:10.5281/zenodo.18887130","type":"article-journal","title":"Integrated AI-based Care Model Library I","abstract":"Deliverable D3.7, titled “Integrated AI-based Care Model Library I”, consolidates the state-of-the-art (SOTA) analysis, and the methodologies explored in tasks T3.5 – “Explainable Models and Algorithms for Trustworthy AI” and T3.6 – “Integrated Care Model Library: Implementation and Recalibration of Adaptive Models” of Work Package 3 (WP3 – “Data and AI Governance for Personalized Prediction, Monitoring and Recommendations”) of the project. Deliverable D3.7 is the first version of a two-part series of deliverables, namely D3.7 - “Integrated AI-based Care Model Library I” and D3.8 - “Integrated AI-based Care Model Library II” (to be submitted on M32). This deliverable D3.7 highlights the innovative research efforts being made in integrating sophisticated SOTA models and advanced Machine Learning (ML) technologies into the Virtual Health Platform (VHP). The deliverable further demonstrates how these advancements contribute to achieving the project’s overarching objectives. From the collective work conducted in WP3, this deliverable serves as both a practical guide and a repository of insights for stakeholders, transferring knowledge and fostering the adoption of advanced Integrated Care Model (ICM) technologies in healthcare. This document underscores the project’s dedication to driving impactful advancements and setting a benchmark for the future of predictive and preventive healthcare systems. There are several major challenges that should be addressed during the lifecycle of this project, which include: i) the integration of heterogeneous data sources into high-quality and bias-free datasets, ii) scalability and real-time processing of developed algorithms, iii) adaptability of AI/ML models to diverse and extremely complex healthcare settings, iv) explainability of AI models and their ethical compliance, and finally v) rigorous evaluation and validation. As this deliverable will demonstrate recent advancements in AI, and particularly multimodal processing can greatly assist in some of these challenges, given that they can process and integrate a wide range data sources to provide a holistic understanding of patient health with increased personalised accuracy and adaptability to new clinical settings.This report introduces key concepts in multimodal systems engineering as required for the development of the Integrated Care Model Library (ICML). This deliverable explores the advanced prediction analytical tools that will be developed to enhance the VHP and ultimately empower it with advanced AI methodologies. More specifically, the main objectives of this deliverable include the development of a novel methodology for ICML with an objective to promote accurate predictions and explainable outcomes across European populations irrespective of ethnicity or socioeconomic background. This deliverable also aims to tackle the ethical compliance aspects of the ICML and serve in the facilitation of a reliable framework that meaningfully assists clinicians in the decision-making process. Deliverable D3.7, titled “Integrated AI-based Care Model Library I,” establishes the foundation for achieving these objectives. More specifically, this document constitutes a detailed account of the state-of-the-art, methodologies, strategies, and outcomes associated with integrated care models, emphasising on its alignment with the broader goals of the COMFORTage project mission. The document bridges the gap between theoretical advancements and practical implementation, presenting solutions that are both scientifically robust and operationally feasible. A key focus of this deliverable is the alignment of project objectives with the complexity of the challenges identified. These challenges underscore the importance of a systematic and collaborative approach, as exemplified by the methodologies and frameworks described in this document.","author":[{"family":"Tsitiridis","given":"Aristeidis"},{"family":"Tsolis","given":"Dimitrios"},{"family":"Lykothanasi","given":"Kalliopi"},{"family":"Tsoukalos","given":"Dimitrios"},{"family":"Koutsomitropoulos","given":"Dimitrios"},{"family":"Giannaros","given":"Anastasios"},{"family":"Huang","given":"Shulei"},{"family":"Berenguer-Sánchez","given":"José"},{"family":"Van Den Heuvel","given":"Willem"},{"family":"Pilz","given":"Maximilian"},{"family":"Van Berlo","given":"Sander"},{"family":"Spathoulas","given":"Georgios"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18887130","URL":"https://doi.org/10.5281/zenodo.18887130","source":"datacite"},{"id":"oa:W4405283072","type":"article-journal","title":"Artificial intelligence–based rapid brain volumetry substantially improves differential diagnosis in dementia","abstract":"Abstract Introduction This study evaluates the clinical value of a deep learning–based artificial intelligence (AI) system that performs rapid brain volumetry with automatic lobe segmentation and age‐ and sex‐adjusted percentile comparisons. Methods Fifty‐five patients—17 with Alzheimer's disease (AD), 18 with frontotemporal dementia (FTD), and 20 healthy controls—underwent cranial magnetic resonance imaging scans. Two board‐certified neuroradiologists (BCNR), two board‐certified radiologists (BCR), and three radiology residents (RR) assessed the scans twice: first without AI support and then with AI assistance. Results AI significantly improved diagnostic accuracy for AD (area under the curve −AI: 0.800, +AI: 0.926, p < 0.05), with increased correct diagnoses ( p < 0.01) and reduced errors ( p < 0.03). BCR and RR showed notable performance gains (BCR: p < 0.04; RR: p < 0.02). For the diagnosis FTD, overall consensus ( p < 0.01), BCNR ( p < 0.02), and BCR ( p < 0.05) recorded significantly more correct diagnoses. Discussion AI‐assisted volumetry improves diagnostic performance in differentiating AD and FTD, benefiting all reader groups, including BCNR. Highlights Artificial intelligence (AI)‐supported brain volumetry significantly improved the diagnostic accuracy for Alzheimer's disease (AD) and frontotemporal dementia (FTD), with notable performance gains across radiologists of varying expertise levels. The presented AI tool is readily clinically available and reduces brain volumetry processing time from 12 to 24 hours to under 5 minutes, with full integration into picture archiving and communication systems, streamlining the workflow and facilitating real‐time clinical decision making. AI‐supported rapid brain volumetry has the potential to improve early diagnosis and to improve patient management.","author":[{"family":"Rudolph","given":"Jan"},{"family":"Rueckel","given":"Johannes"},{"family":"Döpfert","given":"Jörg"},{"family":"Ling","given":"Wen"},{"family":"Opalka","given":"J"},{"family":"Brem","given":"Christian"},{"family":"Hesse","given":"Nina"},{"family":"Ingenerf","given":"Maria"},{"family":"Koliogiannis","given":"Vanessa"},{"family":"Solyanik","given":"Olga"},{"family":"Hoppe","given":"Boj"},{"family":"Zimmermann","given":"Hanna"},{"family":"Flatz","given":"Wilhelm"},{"family":"Forbrig","given":"Robert"},{"family":"Patzig","given":"Maximilian"},{"family":"Rauchmann","given":"Boris‐stephan"},{"family":"Perneczky","given":"Robert"},{"family":"Peters","given":"Oliver"},{"family":"Priller","given":"Josef"},{"family":"Schneider","given":"Anja"},{"family":"Fließbach","given":"Klaus"},{"family":"Hermann","given":"Andreas"},{"family":"Wiltfang","given":"Jens"},{"family":"Jessen","given":"Frank"},{"family":"Düzel","given":"Emrah"},{"family":"Büerger","given":"Katharina"},{"family":"Teipel","given":"Stefan"},{"family":"Laske","given":"Christoph"},{"family":"Synofzik","given":"Matthis"},{"family":"Spottke","given":"Annika"},{"family":"Ewers","given":"Michael"},{"family":"Dechent","given":"Peter"},{"family":"Haynes","given":"John­–dylan"},{"family":"Levin","given":"Johannes"},{"family":"Liebig","given":"Thomas"},{"family":"Ricke","given":"Jens"},{"family":"Ingrisch","given":"Michael"},{"family":"Stoecklein","given":"Sophia"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/dad2.70037","URL":"https://doi.org/10.1002/dad2.70037","source":"openalex"},{"id":"oa:W4404852800","type":"article-journal","title":"Artificial intelligence and machine learning adoption in the financial sector: a holistic review","abstract":"The evolution of new technologies has spurred a growing body of literature exploring their application and impact on the financial sector, particularly the integration of artificial intelligence (AI). This paper delves into the rapid adoption of AI and machine learning within the financial sector, highlighting their potential to enhance financial stability and productivity. By reviewing research from 2018 to 2023, the study categorizes AI applications in finance into three main areas: cybersecurity, customer services, and financial management. Furthermore, the research identifies and classifies various threats posed to the integrity and stability of the financial system by AI, along with associated challenges for policy and regulatory frameworks. It also addresses the risks and obstacles inherent in deploying AI within financial markets and banking sectors, offering recommended strategies to mitigate these limitations. Despite the recognized advantages, the comprehensive understanding of AI's benefits and drawbacks remains incomplete due to its evolving nature and varied applications in banking. Clear policies governing AI usage are imperative to safeguard financial consumers and promote a fair and transparent financial market. These guidelines should prioritize human decision-making and foster an unbiased approach to policymaking, ultimately fostering innovation within the industry.","author":[{"family":"Sayari","given":"Karima"},{"family":"Firdouse","given":"MJ"},{"family":"Abri","given":"Fathiya"}],"issued":{"date-parts":[[2024]]},"DOI":"10.11591/ijai.v14.i1.pp19-31","URL":"https://doi.org/10.11591/ijai.v14.i1.pp19-31","source":"openalex"},{"id":"oa:W4387560915","type":"manuscript","title":"Predictable Artificial Intelligence","abstract":"We introduce the fundamental ideas and challenges of Predictable AI, a nascent research area that explores the ways in which we can anticipate key validity indicators (e.g., performance, safety) of present and future AI ecosystems. We argue that achieving predictability is crucial for fostering trust, liability, control, alignment and safety of AI ecosystems, and thus should be prioritised over performance. We formally characterise predictability, explore its most relevant components, illustrate what can be predicted, describe alternative candidates for predictors, as well as the trade-offs between maximising validity and predictability. To illustrate these concepts, we bring an array of illustrative examples covering diverse ecosystem configurations. Predictable AI is related to other areas of technical and non-technical AI research, but have distinctive questions, hypotheses, techniques and challenges. This paper aims to elucidate them, calls for identifying paths towards a landscape of predictably valid AI systems and outlines the potential impact of this emergent field.","author":[{"family":"Zhou","given":"Lexin"},{"family":"Moreno-Casares","given":"Pablo"},{"family":"Martínezplumed","given":"Fernando"},{"family":"Burden","given":"John"},{"family":"Burnell","given":"Ryan"},{"family":"Cheke","given":"Lucy"},{"family":"Ferri","given":"Cèsar"},{"family":"Marcoci","given":"Alexandru"},{"family":"Mehrbakhsh","given":"Behzad"},{"family":"Moros-Daval","given":"Yael"},{"family":"Héigeartaigh","given":"Seán"},{"family":"Rutar","given":"Danaja"},{"family":"Schellaert","given":"Wout"},{"family":"Voudouris","given":"Konstantinos"},{"family":"Hernándezorallo","given":"José"}],"issued":{"date-parts":[[2023]]},"DOI":"10.48550/arxiv.2310.06167","URL":"https://doi.org/10.48550/arxiv.2310.06167","source":"openalex"},{"id":"oa:W4411643584","type":"article-journal","title":"Opportunity or Threat: Investigating Faculty Readiness to Adopt Artificial Intelligence in Higher Education","abstract":"Artificial Intelligence (AI) is increasingly being integrated into higher education, providing opportunities to enhance teaching and learning through personalised instruction, automated assessments, and data–driven insights. However, faculty members’ readiness to integrate AI into their teaching practices plays a critical role in its effective implementation. While global studies have primarily examined AI adoption in higher education, research on faculty readiness within Caribbean institutions remains limited, presenting a significant gap in the literature. This study used the Unified Theory of Acceptance and Use of Technology (UTAUT) model to assess the influence of perceived benefits, facilitating conditions, and attitude towards AI on adoption readiness. A quantitative survey was administered to a sample of 78 faculty members from The University of the West Indies, Cave Hill campus, collecting data on their AI experience and perceptions. The findings indicate that perceived benefits of AI and institutional support are significant predictors of faculty readiness, while attitude towards AI does not significantly influence adoption. The study also identifies limited AI expertise, lack of training opportunities, and concerns about AI’s ethical implications as key barriers to readiness to adopt. These results highlight the need for structured AI training programs, enhanced institutional support, and clear policies to facilitate AI adoption among faculty. The findings have broader implications for universities in developing regions, emphasising the importance of targeted faculty development to ensure effective AI integration in higher education.","author":[{"family":"Jackman","given":"Grace"},{"family":"Marshall","given":"Ian"},{"family":"Carrington","given":"Troy"}],"issued":{"date-parts":[[2024]]},"DOI":"10.46425/cjed1201029055","URL":"https://doi.org/10.46425/cjed1201029055","source":"openalex"},{"id":"oa:W4404813021","type":"article-journal","title":"Focal therapy of prostate cancer: Use of artificial intelligence to define tumour volume and predict treatment outcomes","abstract":"Objectives: The aim of this study is to evaluate new software (Unfold AI) in the estimation of prostate tumour volume (TV) and prediction of focal therapy outcomes. Subjects/patients and methods: Subjects were 204 men with prostate cancer (PCa) of grade groups 2-4 (GG ≥ 2), who were enrolled in a trial of partial gland cryoablation (PGA) at UCLA from 2017 to 2022. Magnetic resonance imaging (MRI)-guided biopsy (MRGB) was performed at diagnosis and at 6 and 18 months following PGA. Utilising Unfold AI (FDA-cleared 2022), which generates a 3D map of GG ≥ 2 PCa margins, we retrospectively estimated TV for each patient. TV was compared against conventional baseline variables as a correlate of a successful primary outcome-defined here as the absence of GG ≥ 2 on follow-up MRGB at 6 months. Secondary outcomes were MRGB at 18 months and failure-free survival, that is, lack of metastasis or salvage whole gland therapy. Receiver operating curves and multivariate analysis were used to determine significance. Results: A successful primary outcome was observed in 77.7% of patients. Significant correlates of a successful ablation were percent pattern 4 and TV; areas under the curve (AUCs) were 0.60 and 0.73, respectively. GG was not a correlate of success (AUC = 0.51). A TV of 1.5 cc provided the optimal combination of sensitivity (55.8%) and specificity (85.7%) at 6 months. TV was also significantly associated with secondary outcomes. In multivariate analysis, TV was the variable most associated with 6- and 18-month biopsy success (adjusted odds ratios [aORs] were 6.1 and 4.2). Utilising TV ≤ 1.5 cc as a PGA criterion would have prevented 72% of failures at the cost of 42% of successes. Conclusion: The AI-based software Unfold AI estimates TV, which is significantly associated with biopsy outcomes after focal cryoablation. The rate of treatment success is inversely related to TV.","author":[{"family":"Brisbane","given":"Wayne"},{"family":"Priester","given":"Alan"},{"family":"Nguyen","given":"Anissa"},{"family":"Topoozian","given":"Mark"},{"family":"Mota","given":"Sakina"},{"family":"Delfin","given":"Merdie"},{"family":"Gonzalez","given":"Samantha"},{"family":"Grunden","given":"Kyla"},{"family":"Richardson","given":"Shannon"},{"family":"Natarajan","given":"Shyam"},{"family":"Marks","given":"Leonard"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/bco2.456","URL":"https://doi.org/10.1002/bco2.456","source":"openalex"},{"id":"oa:W4404957801","type":"article-journal","title":"Preclinical Cognitive Markers of Alzheimer Disease and Early Diagnosis Using Virtual Reality and Artificial Intelligence: Literature Review","abstract":"Background: This review explores the potential of virtual reality (VR) and artificial intelligence (AI) to identify preclinical cognitive markers of Alzheimer disease (AD). By synthesizing recent studies, it aims to advance early diagnostic methods to detect AD before significant symptoms occur. Objective: Research emphasizes the significance of early detection in AD during the preclinical phase, which does not involve cognitive impairment but nevertheless requires reliable biomarkers. Current biomarkers face challenges, prompting the exploration of cognitive behavior indicators beyond episodic memory. Methods: Using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we searched Scopus, PubMed, and Google Scholar for studies on neuropsychiatric disorders utilizing conversational data. Results: Following an analysis of 38 selected articles, we highlight verbal episodic memory as a sensitive preclinical AD marker, with supporting evidence from neuroimaging and genetic profiling. Executive functions precede memory decline, while processing speed is a significant correlate. The potential of VR remains underexplored, and AI algorithms offer a multidimensional approach to early neurocognitive disorder diagnosis. Conclusions: Emerging technologies like VR and AI show promise for preclinical diagnostics, but thorough validation and regulation for clinical safety and efficacy are necessary. Continued technological advancements are expected to enhance early detection and management of AD.","author":[{"family":"Parada","given":"María"},{"family":"González-Palau","given":"Fátima"},{"family":"Valladaresrodríguez","given":"Sonia"},{"family":"Rincón","given":"M"},{"family":"Barroeta","given":"Maria"},{"family":"Rodriguez","given":"Marta"},{"family":"Aguado","given":"Yolanda"},{"family":"Blanco","given":"A"},{"family":"Díaz-López","given":"Estela"},{"family":"Bachiller","given":"Margarita"},{"family":"Losada","given":"Raquel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/62914","URL":"https://doi.org/10.2196/62914","source":"openalex"},{"id":"oa:W4391186832","type":"article-journal","title":"Artificial-intelligence-powered customer service management in the logistics industry","abstract":"Objective: The article aims to show how people perceive future implications for logistics customer service resulting from the implementation of new technologies in the form of game-changer artificial intelligence (AI) solutions in the spirit of economy 4.0 and society 5.0. Research Design & Methods: The research process used a nomothetic approach based on the methodology of mixed research. The qualitative approach included a research study of monographs, publications, reports, and netographic sources. We used the technique of critical content analysis based on the co-occurrence of terms. In turn, we based the quantitative approach on the diagnostic survey method with the computer-assisted web interviewing (CAWI) technique. The sample size was 233. For further analysis, we used the statistical package for the social sciences (SPSS). Findings: The research shows that customer service in logistics already uses different forms of AI-based solutions (like Chabtbots, Voicebots, and voice assistants). Even customers positively evaluate those solutions, among others, for efficiency, competence, and service quality. Moreover, customers are aware of AI-based solutions and know that their usage will deepen in the future, as it is a game changer for the competitiveness of customer service in logistics. Implications & Recommendations: The conducted research indicates the need to constantly improve the digital competences of the users of last-mile logistics services in the context of technologization of transaction processes. Different areas of business will widely use AI-based solutions, because there is a need to develop systems which will help with the human-machine communication. This technology should be constructed as safe for people and easy to use; both with regard to users and customers. As a result of these processes, there is a greater need to educate people about AI-based solutions to develop awareness and improve future outcomes. Contribution & Value Added: The article’s main advantage is determining new possibilities in the area of logistics customer service as a result of the dissemination of solutions in the AI field, which may be a helpful instrument for enterprises in managing the last-mile scenario in the future.","author":[{"family":"Brzozowska","given":"Marta"},{"family":"Kolasińska-Morawska","given":"Katarzyna"},{"family":"Sułkowski","given":"Łukasz"},{"family":"Morawski","given":"Paweł"}],"issued":{"date-parts":[[2023]]},"DOI":"10.15678/eber.2023.110407","URL":"https://doi.org/10.15678/eber.2023.110407","source":"openalex"},{"id":"oa:W4401628611","type":"article-journal","title":"Preoperative Patient Guidance and Education in Aesthetic Breast Plastic Surgery: A Novel Proposed Application of Artificial Intelligence Large Language Models","abstract":"Background: At a time when Internet and social media use is omnipresent among patients in their self-directed research about their medical or surgical needs, artificial intelligence (AI) large language models (LLMs) are on track to represent hallmark resources in this context. Objectives: The authors aim to explore and assess the performance of a novel AI LLM in answering questions posed by simulated patients interested in aesthetic breast plastic surgery procedures. Methods: A publicly available AI LLM was queried using simulated interactions from the perspective of patients interested in breast augmentation, mastopexy, and breast reduction. Questions posed were standardized and categorized under aesthetic needs inquiries and awareness of appropriate procedures; patient candidacy and indications; procedure safety and risks; procedure information, steps, and techniques; patient assessment; preparation for surgery; postprocedure instructions and recovery; and procedure cost and surgeon recommendations. Using standardized Likert scales ranging from 1 to 10, 4 expert breast plastic surgeons evaluated responses provided by AI. A postparticipation survey assessed expert evaluators' experience with LLM technology, perceived utility, and limitations. Results: The overall performance across all question categories, assessment criteria, and procedures examined was 7.3/10 ± 0.5. Overall accuracy of information shared was scored at 7.1/10 ± 0.5; comprehensiveness at 7.0/10 ± 0.6; objectivity at 7.5/10 ± 0.4; safety at 7.5/10 ± 0.4; communication clarity at 7.3/10 ± 0.2; and acknowledgment of limitations at 7.7/10 ± 0.2. With regards to performance on procedures examined, the model's overall score was 7.0/10 ± 0.8 for breast augmentation; 7.6/10 ± 0.5 for mastopexy; and 7.4/10 ± 0.5 for breast reduction. The score on breast implant-specific knowledge was 6.7/10 ± 0.6. Conclusions: Albeit not without limitations, AI LLMs represent promising resources for patient guidance and patient education. The technology's machine learning capabilities may explain its improved performance efficiency.","author":[{"family":"Abirafeh","given":"Jad"},{"family":"Bassiri-Tehrani","given":"Brian"},{"family":"Kazan","given":"Roy"},{"family":"Furnas","given":"Heather"},{"family":"Hammond","given":"Dennis"},{"family":"Adams","given":"William"},{"family":"Nahai","given":"Foad"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/asjof/ojae062","URL":"https://doi.org/10.1093/asjof/ojae062","source":"openalex"},{"id":"oa:W4367857052","type":"article-journal","title":"Artificial intelligence to assist specialists in the detection of haematological diseases","abstract":"Artificial intelligence, particularly the growth of neural network research and development, has become an invaluable tool for data analysis, offering unrivalled solutions for image generation, natural language processing, and personalised suggestions. In the meantime, biomedicine has been presented as one of the pressing challenges of the 21st century. The inversion of the age pyramid, the increase in longevity, and the negative environment due to pollution and bad habits of the population have led to a necessity of research in the methodologies that can help to mitigate and fight against these changes. The combination of both fields has already achieved remarkable results in drug discovery, cancer prediction or gene activation. However, challenges such as data labelling, architecture improvements, interpretability of the models and translational implementation of the proposals still remain. In haematology, conventional protocols follow a stepwise approach that includes several tests and doctor-patient interactions to make a diagnosis. This procedure results in significant costs and workload for hospitals. In this paper, we present an artificial intelligence model based on neural networks to support practitioners in the identification of different haematological diseases using only rutinary and inexpensive blood count tests. In particular, we present both binary and multiclass classification of haematological diseases using a specialised neural network architecture where data is studied and combined along it, taking into account the clinical knowledge of the problem, obtaining results up to 96% accuracy for the binary classification experiment. Furthermore, we compare this method against traditional machine learning algorithms such as gradient boosting decision trees and transformers for tabular data. The use of these machine learning techniques could reduce the cost and decision time and improve the quality of life for both specialists and patients while producing more precise diagnoses.","author":[{"family":"Díaz-Del-Pino","given":"Sergio"},{"family":"Trellesmartínez","given":"Roberto"},{"family":"González-Fernández","given":"Fernando"},{"family":"Guil","given":"Nicolás"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.heliyon.2023.e15940","URL":"https://doi.org/10.1016/j.heliyon.2023.e15940","source":"openalex"},{"id":"oa:W4405844566","type":"article-journal","title":"ChatGPT-4 Performance on German Continuing Medical Education—Friend or Foe (Trick or Treat)? Protocol for a Randomized Controlled Trial","abstract":"BACKGROUND: The increasing development and spread of artificial and assistive intelligence is opening up new areas of application not only in applied medicine but also in related fields such as continuing medical education (CME), which is part of the mandatory training program for medical doctors in Germany. This study aimed to determine whether medical laypersons can successfully conduct training courses specifically for physicians with the help of a large language model (LLM) such as ChatGPT-4. This study aims to qualitatively and quantitatively investigate the impact of using artificial intelligence (AI; specifically ChatGPT) on the acquisition of credit points in German postgraduate medical education. OBJECTIVE: Using this approach, we wanted to test further possible applications of AI in the postgraduate medical education setting and obtain results for practical use. Depending on the results, the potential influence of LLMs such as ChatGPT-4 on CME will be discussed, for example, as part of a SWOT (strengths, weaknesses, opportunities, threats) analysis. METHODS: We designed a randomized controlled trial, in which adult high school students attempt to solve CME tests across six medical specialties in three study arms in total with 18 CME training courses per study arm under different interventional conditions with varying amounts of permitted use of ChatGPT-4. Sample size calculation was performed including guess probability (20% correct answers, SD=40%; confidence level of 1-α=.95/α=.05; test power of 1-β=.95; P<.05). The study was registered at open scientific framework. RESULTS: As of October 2024, the acquisition of data and students to participate in the trial is ongoing. Upon analysis of our acquired data, we predict our findings to be ready for publication as soon as early 2025. CONCLUSIONS: We aim to prove that the advances in AI, especially LLMs such as ChatGPT-4 have considerable effects on medical laypersons' ability to successfully pass CME tests. The implications that this holds on how the concept of continuous medical education requires reevaluation are yet to be contemplated. TRIAL REGISTRATION: OSF Registries 10.17605/OSF.IO/MZNUF; https://osf.io/mznuf. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/63887.","author":[{"family":"Burisch","given":"Christian"},{"family":"Bellary","given":"Abhav"},{"family":"Breuckmann","given":"Frank"},{"family":"Ehlers","given":"Jan"},{"family":"Thal","given":"Serge"},{"family":"Sellmann","given":"Timur"},{"family":"Gödde","given":"Daniel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/63887","URL":"https://doi.org/10.2196/63887","source":"openalex"},{"id":"oa:W4404048213","type":"article-journal","title":"Artificial Intelligence as a Tool for the Development of Soft Skills: A Bibliometric Review in the Context of Higher Education","abstract":"Skills such as communication, teamwork, adaptability and problem-solving are essential for professional and personal development. The integration of artificial intelligence (AI) in the process of building these skills in students represents a significant innovation in higher education. This is because AI offers new teaching methods that can be more effective and personalized than traditional approaches. Thus, it is relevant to investigate the bibliometric indicators that reflect the trends in which AI enables the development of soft skills in the context of higher education. The approach used is the mixed, exploratory-descriptive level. The total number of studies reviewed was 78, all of them extracted from the Scopus database. The results show that the most prevalent thematic areas are “Communication skills development”, “Teamwork and collaboration” and “Critical thinking and problem-solving”. This is a result of AI’s ability to create more personalized, interactive and adaptive learning environments. However, it is concluded that scientific production in this field of study is still developing and requires greater attention from researchers. It is important to reflect that the implementation of AI in higher education must be supported by policies that regulate its effective integration and maximize its impact. Future studies should employ systematic reviews to address the impact of AI on soft skills according to the area of knowledge, such as engineering, social sciences or health sciences, identifying the skills to which AI is contributing most significantly.","author":[{"family":"Alvarado-Bravo","given":"Nestor"},{"family":"Aldana-Trejo","given":"Florcita"},{"family":"Herrera","given":"Víctor"},{"family":"Rasilla-Rovegno","given":"José"},{"family":"Suarez-Bazalar","given":"Raul"},{"family":"Torres-Quiroz","given":"Almintor"},{"family":"Paredes-Soria","given":"Alejandro"},{"family":"Gonzales-Saldaña","given":"Susan"},{"family":"Quispe","given":"Gregorio"},{"family":"Olivares-Zegarra","given":"Soledad"}],"issued":{"date-parts":[[2024]]},"DOI":"10.26803/ijlter.23.10.18","URL":"https://doi.org/10.26803/ijlter.23.10.18","source":"openalex"},{"id":"oa:W4396935494","type":"article-journal","title":"Artificial intelligence based data curation: enabling a patient-centric European health data space","abstract":"The emerging European Health Data Space (EHDS) Regulation opens new prospects for large-scale sharing and re-use of health data. Yet, the proposed regulation suffers from two important limitations: it is designed to benefit the whole population with limited consideration for individuals, and the generation of secondary datasets from heterogeneous, unlinked patient data will remain burdensome. AIDAVA, a Horizon Europe project that started in September 2022, proposes to address both shortcomings by providing patients with an AI-based virtual assistant that maximises automation in the integration and transformation of their health data into an interoperable, longitudinal health record. This personal record can then be used to inform patient-related decisions at the point of care, whether this is the usual point of care or a possible cross-border point of care. The personal record can also be used to generate population datasets for research and policymaking. The proposed solution will enable a much-needed paradigm shift in health data management, implementing a 'curate once at patient level, use many times' approach, primarily for the benefit of patients and their care providers, but also for more efficient generation of high-quality secondary datasets. After 15 months, the project shows promising preliminary results in achieving automation in the integration and transformation of heterogeneous data of each individual patient, once the content of the data sources managed by the data holders has been formally described. Additionally, the conceptualization phase of the project identified a set of recommendations for the development of a patient-centric EHDS, significantly facilitating the generation of data for secondary use.","author":[{"family":"Zegher","given":"Isabelle"},{"family":"Norak","given":"Kerli"},{"family":"Steiger","given":"Dominik"},{"family":"Müller","given":"Heimo"},{"family":"Kalra","given":"Dipak"},{"family":"Scheenstra","given":"Bart"},{"family":"Cina","given":"Isabella"},{"family":"Schulz","given":"Stefan"},{"family":"Uma","given":"Kanimozhi"},{"family":"Kalendralis","given":"Petros"},{"family":"Lotman","given":"Eno"},{"family":"Benedikt","given":"Martin"},{"family":"Dumontier","given":"Michel"},{"family":"Çelebi","given":"Remzi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fmed.2024.1365501","URL":"https://doi.org/10.3389/fmed.2024.1365501","source":"openalex"},{"id":"oa:W4400123511","type":"article-journal","title":"Discriminative and exploitive stereotypes: Artificial intelligence generated images of aged care nurses and the impacts on recruitment and retention","abstract":"This article uses critical discourse analysis to investigate artificial intelligence (AI) generated images of aged care nurses and considers how perspectives and perceptions impact upon the recruitment and retention of nurses. The article demonstrates a recontextualization of aged care nursing, giving rise to hidden ideologies including harmful stereotypes which allow for discrimination and exploitation. It is argued that this may imply that nurses require fewer clinical skills in aged care, diminishing the value of working in this area. AI relies on existing data sets, and thus represent existing stereotypes and biases. The discourse analysis has highlighted key issues which may further impact upon nursing recruitment and retention, and advocates for stronger ethical consideration, including the use of experts in data validation, for the way that aged care services and nurses are depicted and thus valued.","author":[{"family":"Byrne","given":"Amy‐louise"},{"family":"Mulvogue","given":"Jennifer"},{"family":"Adhikari","given":"Siju"},{"family":"Cutmore","given":"Ellie"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/nin.12651","URL":"https://doi.org/10.1111/nin.12651","source":"openalex"},{"id":"oa:W4404961577","type":"article-journal","title":"Artificial Intelligence–Generated Emergency Department Summaries and Hospital Handoffs","abstract":"Artificial intelligence (AI) holds significant promise for enhancing quality, safety, efficiency, experience, and equity in health care delivery.Since the widespread release of large language models (LLMs) in late 2022, their capabilities to summarize complex text, extract structured data, and generate new content have fueled growing interest and hype around AI in health care.While applications of LLMs for clinical documentation, drafting in-basket messages, and summarizing patient encounters are emerging, comprehensive evaluations of their validity and impact remain limited, especially when compared with clinician-generated content, the current standard of care.Even less examined are discussions of AI use in acute care settings.","author":[{"family":"Landman","given":"Adam"},{"family":"Tilak","given":"Sharmila"},{"family":"Walker","given":"Graham"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1001/jamanetworkopen.2024.48729","URL":"https://doi.org/10.1001/jamanetworkopen.2024.48729","source":"openalex"},{"id":"oa:W4392378490","type":"article-journal","title":"Artificial Intelligence Techniques and Pedigree Charts in Oncogenetics: Towards an Experimental Multioutput Software System for Digitization and Risk Prediction","abstract":"Pedigree charts remain essential in oncological genetic counseling for identifying individuals with an increased risk of developing hereditary tumors. However, this valuable data source often remains confined to paper files, going unused. We propose a computer-aided detection/diagnosis system, based on machine learning and deep learning techniques, capable of the following: (1) assisting genetic oncologists in digitizing paper-based pedigree charts, and in generating new digital ones, and (2) automatically predicting the genetic predisposition risk directly from these digital pedigree charts. To the best of our knowledge, there are no similar studies in the current literature, and consequently, no utilization of software based on artificial intelligence on pedigree charts has been made public yet. By incorporating medical images and other data from omics sciences, there is also a fertile ground for training additional artificial intelligence systems, broadening the software predictive capabilities. We plan to bridge the gap between scientific advancements and practical implementation by modernizing and enhancing existing oncological genetic counseling services. This would mark the pioneering development of an AI-based application designed to enhance various aspects of genetic counseling, leading to improved patient care and advancements in the field of oncogenetics.","author":[{"family":"Conte","given":"Luana"},{"family":"Rizzo","given":"Emanuele"},{"family":"Grassi","given":"Tiziana"},{"family":"Bagordo","given":"Francesco"},{"family":"Matteis","given":"Elisabetta"},{"family":"Nunzio","given":"Giorgio"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/computation12030047","URL":"https://doi.org/10.3390/computation12030047","source":"openalex"},{"id":"oa:W4404042958","type":"article-journal","title":"Scaling equitable artificial intelligence in healthcare with machine learning operations","abstract":"Machine learning operations (MLOps), a discipline concerned with the production, monitoring and maintenance of artificial intelligence (AI) and machine learning (ML) models at scale, applied in healthcare can facilitate the transition of AI/ML-enabled healthcare tools from research to sustainable deployment.1–3 Adherence to MLOps best practices can address persistent challenges with AI/ML tools deployed into clinical workflows where models often struggle with generalisability, integration and robustness. As AI regulations continue to evolve such as the Department of Health and Human Services Office of Civil Rights final ruling that requires healthcare providers to ensure their AI/ML tools do not discriminate,4 it becomes increasingly essential for MLOps in healthcare to prioritise health equity.","author":[{"family":"Ng","given":"Madelena"},{"family":"Youssef","given":"Alexey"},{"family":"Pillai","given":"Malvika"},{"family":"Shah","given":"Vaibhavi"},{"family":"Hernandezboussard","given":"Tina"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1136/bmjhci-2024-101101","URL":"https://doi.org/10.1136/bmjhci-2024-101101","source":"openalex"},{"id":"oa:W4400742536","type":"article-journal","title":"Artificial Intelligence-Based Internet of Things for Industry 5.0","abstract":"In the Industry 5.0 paradigm, systems based on artificial intelligence are an important component of the Internet of Things. Industry 5.0 demonstrated the important link between intelligent systems and people in most applications through precision manufacturing automation and critical thinking. In addition, Industry 5.0 brings with it several valid tools that help organizations operate cheaply and change immediately without capital investment. In recent years, smart devices, wireless communication, and sensor nodes have advanced greatly, transforming Internet of Things (IoT) ecosystems. With IoT devices, users can receive information even in rural areas and generate unbounded reports. Also, as previously mentioned, they meticulously guide people with intelligent judgments through communication technology. Many connected devices collect significant amounts of detected raw data when they require pre-processing. Although, it hardly becomes valuable for IoT devices and sufficient resources require Edge computing. AI-based algorithms are essential tools for data inference in Edge computing. In addition, observed data collected by IoT applications is usually unstructured and needs further analysis, where AI-based models help extract relevant information. Furthermore, malicious files are possible when data are transferred from one device to another. Therefore, this chapter looks at Industry 5.0, IoT architecture, and AI-based IoT; we analyze the IoT network and its specifications; communication is possible thanks to technologies.","author":[{"family":"Goswami","given":"Shikha"},{"family":"Goswami","given":"Rohit"},{"family":"Verma","given":"Govind"}],"issued":{"date-parts":[[2024]]},"DOI":"10.5772/intechopen.115116","URL":"https://doi.org/10.5772/intechopen.115116","source":"openalex"},{"id":"oa:W4387326591","type":"article-journal","title":"Development and validation of an artificial intelligence assisted prenatal ultrasonography screening system for trainees","abstract":"OBJECTIVE: Fetal anomaly screening via ultrasonography, which involves capturing and interpreting standard views, is highly challenging for inexperienced operators. We aimed to develop and validate a prenatal-screening artificial intelligence system (PSAIS) for real-time evaluation of the quality of anatomical images, indicating existing and missing structures. METHODS: Still ultrasonographic images obtained from fetuses of 18-32 weeks of gestation between 2017 and 2018 were used to develop PSAIS based on YOLOv3 with global (anatomic site) and local (structures) feature extraction that could evaluate the image quality and indicate existing and missing structures in the fetal anatomical images. The performance of the PSAIS in recognizing 19 standard views was evaluated using retrospective real-world fetal scan video validation datasets from four hospitals. We stratified sampled frames (standard, similar-to-standard, and background views at approximately 1:1:1) for experts to blindly verify the results. RESULTS: The PSAIS was trained using 134 696 images and validated using 836 videos with 12 697 images. For internal and external validations, the multiclass macro-average areas under the receiver operating characteristic curve were 0.943 (95% confidence interval [CI], 0.815-1.000) and 0.958 (0.864-1.000); the micro-average areas were 0.974 (0.970-0.979) and 0.973 (0.965-0.981), respectively. For similar-to-standard views, the PSAIS accurately labeled 90.9% (90.0%-91.4%) with key structures and indicated missing structures. CONCLUSIONS: An artificial intelligence system developed to assist trainees in fetal anomaly screening demonstrated high agreement with experts in standard view identification.","author":[{"family":"Lei","given":"Ting"},{"family":"Feng","given":"Jie"},{"family":"Lin","given":"Mei"},{"family":"Xie","given":"Bai"},{"family":"Zhou","given":"Qian"},{"family":"Wang","given":"Nan"},{"family":"Zheng","given":"Qiao"},{"family":"Yang","given":"Yan"},{"family":"Guo","given":"Hong"},{"family":"Xie","given":"Hongning"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/ijgo.15167","URL":"https://doi.org/10.1002/ijgo.15167","source":"openalex"},{"id":"oa:W4389712719","type":"manuscript","title":"CLIP in Medical Imaging: A Survey","abstract":"Contrastive Language-Image Pre-training (CLIP), a simple yet effective pre-training paradigm, successfully introduces text supervision to vision models. It has shown promising results across various tasks due to its generalizability and interpretability. The use of CLIP has recently gained increasing interest in the medical imaging domain, serving as a pre-training paradigm for image-text alignment, or a critical component in diverse clinical tasks. With the aim of facilitating a deeper understanding of this promising direction, this survey offers an in-depth exploration of the CLIP within the domain of medical imaging, regarding both refined CLIP pre-training and CLIP-driven applications. In this paper, we (1) first start with a brief introduction to the fundamentals of CLIP methodology; (2) then investigate the adaptation of CLIP pre-training in the medical imaging domain, focusing on how to optimize CLIP given characteristics of medical images and reports; (3) further explore practical utilization of CLIP pre-trained models in various tasks, including classification, dense prediction, and cross-modal tasks; and (4) finally discuss existing limitations of CLIP in the context of medical imaging, and propose forward-looking directions to address the demands of medical imaging domain. Studies featuring technical and practical value are both investigated. We expect this survey will provide researchers with a holistic understanding of the CLIP paradigm and its potential implications. The project page of this survey can also be found on https://github.com/zhaozh10/Awesome-CLIP-in-Medical-Imaging.","author":[{"family":"Zhao","given":"Zihao"},{"family":"Liu","given":"Yu"},{"family":"Wu","given":"Han"},{"family":"Wang","given":"Mei"},{"family":"Li","given":"Yonghao"},{"family":"Wang","given":"Sheng"},{"family":"Teng","given":"Lin"},{"family":"Liu","given":"Disheng"},{"family":"Cui","given":"Zhiming"},{"family":"Wang","given":"Qian"},{"family":"Shen","given":"Dinggang"}],"issued":{"date-parts":[[2023]]},"DOI":"10.48550/arxiv.2312.07353","URL":"https://doi.org/10.48550/arxiv.2312.07353","source":"openalex"},{"id":"oa:W4404705278","type":"article-journal","title":"PICOT questions and search strategies formulation: A novel approach using artificial intelligence automation","abstract":"AIM: The aim of this study was to evaluate and compare artificial intelligence (AI)-based large language models (LLMs) (ChatGPT-3.5, Bing, and Bard) with human-based formulations in generating relevant clinical queries, using comprehensive methodological evaluations. METHODS: To interact with the major LLMs ChatGPT-3.5, Bing Chat, and Google Bard, scripts and prompts were designed to formulate PICOT (population, intervention, comparison, outcome, time) clinical questions and search strategies. Quality of the LLMs responses was assessed using a descriptive approach and independent assessment by two researchers. To determine the number of hits, PubMed, Web of Science, Cochrane Library, and CINAHL Ultimate search results were imported separately, without search restrictions, with the search strings generated by the three LLMs and an additional one by the expert. Hits from one of the scenarios were also exported for relevance evaluation. The use of a single scenario was chosen to provide a focused analysis. Cronbach's alpha and intraclass correlation coefficient (ICC) were also calculated. RESULTS: In five different scenarios, ChatGPT-3.5 generated 11,859 hits, Bing 1,376,854, Bard 16,583, and an expert 5919 hits. We then used the first scenario to assess the relevance of the obtained results. The human expert search approach resulted in 65.22% (56/105) relevant articles. Bing was the most accurate AI-based LLM with 70.79% (63/89), followed by ChatGPT-3.5 with 21.05% (12/45), and Bard with 13.29% (42/316) relevant hits. Based on the assessment of two evaluators, ChatGPT-3.5 received the highest score (M = 48.50; SD = 0.71). Results showed a high level of agreement between the two evaluators. Although ChatGPT-3.5 showed a lower percentage of relevant hits compared to Bing, this reflects the nuanced evaluation criteria, where the subjective evaluation prioritized contextual accuracy and quality over mere relevance. CONCLUSION: This study provides valuable insights into the ability of LLMs to formulate PICOT clinical questions and search strategies. AI-based LLMs, such as ChatGPT-3.5, demonstrate significant potential for augmenting clinical workflows, improving clinical query development, and supporting search strategies. However, the findings also highlight limitations that necessitate further refinement and continued human oversight. CLINICAL RELEVANCE: AI could assist nurses in formulating PICOT clinical questions and search strategies. AI-based LLMs offer valuable support to healthcare professionals by improving the structure of clinical questions and enhancing search strategies, thereby significantly increasing the efficiency of information retrieval.","author":[{"family":"Gosak","given":"Lucija"},{"family":"Štiglic","given":"Gregor"},{"family":"Pruinelli","given":"Lisiane"},{"family":"Vrbnjak","given":"Dominika"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/jnu.13036","URL":"https://doi.org/10.1111/jnu.13036","source":"openalex"},{"id":"oa:W4387955202","type":"article-journal","title":"BASELINE SPECTRAL DOMAIN OPTICAL COHERENCE TOMOGRAPHIC RETINAL LAYER FEATURES IDENTIFIED BY ARTIFICIAL INTELLIGENCE PREDICT THE COURSE OF CENTRAL SEROUS CHORIORETINOPATHY","abstract":"PURPOSE: To identify optical coherence tomography (OCT) features to predict the course of central serous chorioretinopathy (CSC) with an artificial intelligence-based program. METHODS: Multicenter, observational study with a retrospective design. Treatment-naïve patients with acute CSC and chronic CSC were enrolled. Baseline OCTs were examined by an artificial intelligence-developed platform (Discovery OCT Fluid and Biomarker Detector, RetinAI AG, Switzerland). Through this platform, automated retinal layer thicknesses and volumes, including intaretinal and subretinal fluid, and pigment epithelium detachment were measured. Baseline OCT features were compared between acute CSC and chronic CSC patients. RESULTS: One hundred and sixty eyes of 144 patients with CSC were enrolled, of which 100 had chronic CSC and 60 acute CSC. Retinal layer analysis of baseline OCT scans showed that the inner nuclear layer, the outer nuclear layer, and the photoreceptor-retinal pigmented epithelium complex were significantly thicker at baseline in eyes with acute CSC in comparison with those with chronic CSC ( P < 0.001). Similarly, choriocapillaris and choroidal stroma and retinal thickness (RT) were thicker in acute CSC than chronic CSC eyes ( P = 0.001). Volume analysis revealed average greater subretinal fluid volumes in the acute CSC group in comparison with chronic CSC ( P = 0.041). CONCLUSION: Optical coherence tomography features may be helpful to predict the clinical course of CSC. The baseline presence of an increased thickness in the outer retinal layers, choriocapillaris and choroidal stroma, and subretinal fluid volume seems to be associated with acute course of the disease.","author":[{"family":"Desideri","given":"Lorenzo"},{"family":"Anguita","given":"Rodrigo"},{"family":"Berger","given":"Lieselotte"},{"family":"Feenstra","given":"Helena"},{"family":"Scandella","given":"Davide"},{"family":"Sznitman","given":"Raphael"},{"family":"Boon","given":"Camiel"},{"family":"Dijk","given":"Elon"},{"family":"Zinkernagel","given":"Martin"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1097/iae.0000000000003965","URL":"https://doi.org/10.1097/iae.0000000000003965","source":"openalex"},{"id":"oa:W4403886078","type":"article-journal","title":"Integrating Big Data, Artificial Intelligence, and motion analysis for emerging precision medicine applications in Parkinson’s Disease","abstract":"One of the key challenges in Big Data for clinical research and healthcare is how to integrate new sources of data, whose relation to disease processes are often not well understood, with multiple classical clinical measurements that have been used by clinicians for years to describe disease processes and interpret therapeutic outcomes. Without such integration, even the most promising data from emerging technologies may have limited, if any, clinical utility. This paper presents an approach to address this challenge, illustrated through an example in Parkinson's Disease (PD) management. We show how data from various sensing sources can be integrated with traditional clinical measurements used in PD; furthermore, we show how leveraging Big Data frameworks, augmented by Artificial Intelligence (AI) algorithms, can distinctively enrich the data resources available to clinicians. We showcase the potential of this approach in a cohort of 50 PD patients who underwent both evaluations with an Integrated Motion Analysis Suite (IMAS) composed of a battery of multimodal, portable, and wearable sensors and traditional Unified Parkinson's Disease Rating Scale (UPDRS)-III evaluations. Through techniques including Principal Component Analysis (PCA), elastic net regression, and clustering analysis we demonstrate how this combined approach can be used to improve clinical motor assessments and to develop personalized treatments. The scalability of our approach enables systematic data generation and analysis on increasingly larger datasets, confirming the integration potential of IMAS, whose use in PD assessments is validated herein, within Big Data paradigms. Compared to existing approaches, our solution offers a more comprehensive, multi-dimensional view of patient data, enabling deeper clinical insights and greater potential for personalized treatment strategies. Additionally, we show how IMAS can be integrated into established clinical practices, facilitating its adoption in routine care and complementing emerging methods, for instance, non-invasive brain stimulation. Future work will aim to augment our data repositories with additional clinical data, such as imaging and biospecimen data, to further broaden and enhance these foundational methodologies, leveraging the full potential of Big Data and AI.","author":[{"family":"Dipietro","given":"Laura"},{"family":"Eden","given":"Uri"},{"family":"Elkin-Frankston","given":"Seth"},{"family":"El-Hagrassy","given":"Mirret"},{"family":"Camsari","given":"Deniz"},{"family":"Ramos-Estébanez","given":"Ciro"},{"family":"Fregni","given":"Felipe"},{"family":"Wagner","given":"Timothy"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s40537-024-01023-3","URL":"https://doi.org/10.1186/s40537-024-01023-3","source":"openalex"},{"id":"oa:W4404118195","type":"article-journal","title":"Artificial intelligence based assessment of minimally invasive surgical skills using standardised objective metrics – A narrative review","abstract":"INTRODUCTION: Many studies display significant heterogeneity in the reliability of artificial intelligence (AI) assessment of minimally invasive surgical (MIS) skills. Our objective is to investigate whether AI systems utilising standardised objective metrics (SOMs) as the basis of skill assessment can provide a clearer understanding of the current state of such technology. METHODS: We systematically searched Medline, Embase, Scopus, CENTRAL and Web of Science from March 2023 to September 2023. Results were compiled as a narrative review. RESULTS: Twenty-four citations were analysed. Overall accuracy of AI systems in predicting overall SOM score of a procedure ranged from 63 ​% to 100 ​%. The most frequently used SOM by AI algorithms were Objective Structured Assessment of Technical Skills (OSATS) (8/24) and Global Evaluative Assessment of Robotic Skills (GEARS) (8/24). CONCLUSIONS: Stratifying for AI studies which employed SOMs to assess surgical skill did not reduce heterogeneity of reported reliability. Our study identifies key issues within the current literature, which, once addressed, could allow more meaningful comparisons between studies.","author":[{"family":"Kankanamge","given":"Denuka"},{"family":"Wijeweera","given":"Chandana"},{"family":"Ong","given":"Z"},{"family":"Preda","given":"Tamara"},{"family":"Carney","given":"Terry"},{"family":"Wilson","given":"Michael"},{"family":"Preda","given":"Veronica"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.amjsurg.2024.116074","URL":"https://doi.org/10.1016/j.amjsurg.2024.116074","source":"openalex"},{"id":"oa:W4400045019","type":"article-journal","title":"Making sense of artificial intelligence and large language models—including ChatGPT—in pediatric hematology/oncology","abstract":"ChatGPT and other artificial intelligence (AI) systems have captivated the attention of healthcare providers and researchers for their potential to improve care processes and outcomes. While these technologies hold promise to automate processes, increase efficiency, and reduce cognitive burden, their use also carries risks. In this commentary, we review basic concepts of AI, outline some of the capabilities and limitations of currently available tools, discuss current and future applications in pediatric hematology/oncology, and provide an evaluation and implementation framework that can be used by pediatric hematologist/oncologists considering the use of AI in clinical practice.","author":[{"family":"Wyatt","given":"Kirk"},{"family":"Alexander","given":"Natasha"},{"family":"Hills","given":"Gerard"},{"family":"Liang","given":"Wayne"},{"family":"Kadauke","given":"Stephan"},{"family":"Volchenboum","given":"Samuel"},{"family":"Mian","given":"Amir"},{"family":"Phillips","given":"Charles"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/pbc.31143","URL":"https://doi.org/10.1002/pbc.31143","source":"openalex"},{"id":"doi:10.1080/0142159x.2024.2434101","type":"article-journal","title":"Using artificial intelligence to provide a ‘flipped assessment’ approach to medical education learning opportunities","abstract":"PURPOSE: Generative AI can potentially streamline the creation of practice exam questions. This study sought to evaluate medical students' confidence using generative AI for this purpose, and overall attitudes towards its use. MATERIALS AND METHODS: The study used a mixed-methods approach with a pre-post intervention design. 68 medical and physician associate students were recruited to attend a workshop where they were shown how to use Google Bard (now Gemini) to write exam questions before being encouraged to do this themselves with guidance. A survey was completed before and after. Seven students also participated in a follow-up focus group. RESULTS: < 0.001) after the workshop. Qualitative feedback highlighted pros and cons of using generative AI to write exam questions, alongside some concerns about its implementation. Students noted other positive uses in the curriculum and expressed a desire for institutional clarity on appropriate AI use. CONCLUSIONS: While increased confidence is positive, rigorous evaluation of AI-generated question quality is needed to confirm accuracy. Teaching students to use generative AI to create and critique practice questions represents a means of encouraging appropriate AI use.","author":[{"family":"Birks","given":"Samuel"},{"family":"Gray","given":"James"},{"family":"Darling-Pomranz","given":"Claire"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1080/0142159x.2024.2434101","URL":"https://doi.org/10.1080/0142159x.2024.2434101","source":"openalex"},{"id":"oa:W4402155151","type":"article-journal","title":"Artificial intelligence in the anterior segment of eye diseases","abstract":"Ophthalmology is a subject that highly depends on imaging examination. Artificial intelligence (AI) technology has great potential in medical imaging analysis, including image diagnosis, classification, grading, guiding treatment and evaluating prognosis. The combination of the two can realize mass screening of grass-roots eye health, making it possible to seek medical treatment in the mode of “first treatment at the grass-roots level, two-way referral, emergency and slow treatment, and linkage between the upper and lower levels”. On the basis of summarizing the AI technology carried out by scholars and their teams all over the world in the field of ophthalmology, quite a lot of studies have confirmed that machine learning can assist in diagnosis, grading, providing optimal treatment plans and evaluating prognosis in corneal and conjunctival diseases, ametropia, lens diseases, glaucoma, iris diseases, etc. This paper systematically shows the application and progress of AI technology in common anterior segment ocular diseases, the current limitations, and prospects for the future.","author":[{"family":"Liu","given":"Yao‐hong"},{"family":"Liu","given":"Sijia"},{"family":"Gao","given":"Lixiong"},{"family":"Tang","given":"Yong"},{"family":"Li","given":"Zhaohui"},{"family":"Ye","given":"Zi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.18240/ijo.2024.09.23","URL":"https://doi.org/10.18240/ijo.2024.09.23","source":"openalex"},{"id":"oa:W4392201266","type":"article-journal","title":"A 30-Year Review on Nanocomposites: Comprehensive Bibliometric Insights into Microstructural, Electrical, and Mechanical Properties Assisted by Artificial Intelligence","abstract":"From 1990 to 2024, this study presents a groundbreaking bibliometric and sentiment analysis of nanocomposite literature, distinguishing itself from existing reviews through its unique computational methodology. Developed by our research group, this novel approach systematically investigates the evolution of nanocomposites, focusing on microstructural characterization, electrical properties, and mechanical behaviors. By deploying advanced Boolean search strategies within the Scopus database, we achieve a meticulous extraction and in-depth exploration of thematic content, a methodological advancement in the field. Our analysis uniquely identifies critical trends and insights concerning nanocomposite microstructure, electrical attributes, and mechanical performance. The paper goes beyond traditional textual analytics and bibliometric evaluation, offering new interpretations of data and highlighting significant collaborative efforts and influential studies within the nanocomposite domain. Our findings uncover the evolution of research language, thematic shifts, and global contributions, providing a distinct and comprehensive view of the dynamic evolution of nanocomposite research. A critical component of this study is the \"State-of-the-Art and Gaps Extracted from Results and Discussions\" section, which delves into the latest advancements in nanocomposite research. This section details various nanocomposite types and their properties and introduces novel interpretations of their applications, especially in nanocomposite films. By tracing historical progress and identifying emerging trends, this analysis emphasizes the significance of collaboration and influential studies in molding the field. Moreover, the \"Literature Review Guided by Artificial Intelligence\" section showcases an innovative AI-guided approach to nanocomposite research, a first in this domain. Focusing on articles from 2023, selected based on citation frequency, this method offers a new perspective on the interplay between nanocomposites and their electrical properties. It highlights the composition, structure, and functionality of various systems, integrating recent findings for a comprehensive overview of current knowledge. The sentiment analysis, with an average score of 0.638771, reflects a positive trend in academic discourse and an increasing recognition of the potential of nanocomposites. Our bibliometric analysis, another methodological novelty, maps the intellectual domain, emphasizing pivotal research themes and the influence of crosslinking time on nanocomposite attributes. While acknowledging its limitations, this study exemplifies the indispensable role of our innovative computational tools in synthesizing and understanding the extensive body of nanocomposite literature. This work not only elucidates prevailing trends but also contributes a unique perspective and novel insights, enhancing our understanding of the nanocomposite research field.","author":[{"family":"Souza","given":"Fernando"},{"family":"Bhansali","given":"Shekhar"},{"family":"Pal","given":"Kaushik"},{"family":"Maranhão","given":"Fabíola"},{"family":"Oliveira","given":"Marcella"},{"family":"Valladão","given":"Viviane"},{"family":"Silva","given":"Daniele"},{"family":"Silva","given":"Gabriel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/ma17051088","URL":"https://doi.org/10.3390/ma17051088","source":"openalex"},{"id":"oa:W4402857715","type":"article-journal","title":"Artificial Intelligence Detection of Cervical Spine Fractures Using Convolutional Neural Network Models","abstract":"OBJECTIVE: To develop and evaluate a technique using convolutional neural networks (CNNs) for the computer-assisted diagnosis of cervical spine fractures from radiographic x-ray images. By leveraging deep learning techniques, the study might potentially lead to improved patient outcomes and clinical decision-making. METHODS: This study obtained 500 lateral radiographic cervical spine x-ray images from standard open-source dataset repositories to develop a classification model using CNNs. All the images contained diagnostic information, including normal cervical radiographic images (n=250) and fracture images of the cervical spine fracture (n=250). The model would classify whether the patient had a cervical spine fracture or not. Seventy percent of the images were training data sets used for model training, and 30% were for testing. Konstanz Information Miner (KNIME)'s graphic user interface-based programming enabled class label annotation, data preprocessing, CNNs model training, and performance evaluation. RESULTS: The performance evaluation of a model for detecting cervical spine fractures presents compelling results across various metrics. This model exhibits high sensitivity (recall) values of 0.886 for fractures and 0.957 for normal cases, indicating its proficiency in identifying true positives. Precision values of 0.954 for fractures and 0.893 for normal cases highlight the model's ability to minimize false positives. With specificity values of 0.957 for fractures and 0.886 for normal cases, the model effectively identifies true negatives. The overall accuracy of 92.14% highlights its reliability in correctly classifying cases by the area under the receiver operating characteristic curve. CONCLUSION: We successfully used deep learning models for computer-assisted diagnosis of cervical spine fractures from radiographic x-ray images. This approach can assist the radiologist in screening, detecting, and diagnosing cervical spine fractures.","author":[{"family":"Liawrungrueang","given":"Wongthawat"},{"family":"Han","given":"Inbo"},{"family":"Cholamjiak","given":"Watcharaporn"},{"family":"Sarasombath","given":"Peem"},{"family":"Riew","given":"KD"}],"issued":{"date-parts":[[2024]]},"DOI":"10.14245/ns.2448580.290","URL":"https://doi.org/10.14245/ns.2448580.290","source":"openalex"},{"id":"oa:W4401409653","type":"article-journal","title":"Revolutionizing Cardiac Imaging: A Scoping Review of Artificial Intelligence in Echocardiography, CTA, and Cardiac MRI","abstract":"BACKGROUND AND INTRODUCTION: Cardiac imaging is crucial for diagnosing heart disorders. Methods like X-rays, ultrasounds, CT scans, and MRIs provide detailed anatomical and functional heart images. AI can enhance these imaging techniques with its advanced learning capabilities. METHOD: In this scoping review, following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) Guidelines, we searched PubMed, Scopus, Web of Science, and Google Scholar using related keywords on 16 April 2024. From 3679 articles, we first screened titles and abstracts based on the initial inclusion criteria and then screened the full texts. The authors made the final selections collaboratively. RESULT: The PRISMA chart shows that 3516 articles were initially selected for evaluation after removing duplicates. Upon reviewing titles, abstracts, and quality, 24 articles were deemed eligible for the review. The findings indicate that AI enhances image quality, speeds up imaging processes, and reduces radiation exposure with sensitivity and specificity comparable to or exceeding those of qualified radiologists or cardiologists. Further research is needed to assess AI's applicability in various types of cardiac imaging, especially in rural hospitals where access to medical doctors is limited. CONCLUSIONS: AI improves image quality, reduces human errors and radiation exposure, and can predict cardiac events with acceptable sensitivity and specificity.","author":[{"family":"Moradi","given":"Ali"},{"family":"Olanisa","given":"Olawale"},{"family":"Nzeako","given":"Tochukwu"},{"family":"Shahrokhi","given":"Mehregan"},{"family":"Esfahani","given":"Eman"},{"family":"Fakher","given":"Nastaran"},{"family":"Tabari","given":"Mohammad"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/jimaging10080193","URL":"https://doi.org/10.3390/jimaging10080193","source":"openalex"},{"id":"oa:W4392244024","type":"article-journal","title":"Application of the AlphaFold2 Protein Prediction Algorithm Based on Artificial Intelligence","abstract":"As the expression products of genes and macromolecules in living organisms, proteins are the main material basis of life activities. They exist widely in various cells and have various functions such as catalysis, cell signaling and structural support, playing a key role in life activities and functional execution. At the same time, the study of protein can better grasp the life activities from the molecular level, and has important practical significance for disease management, new drug development and crop improvement. Due to advances in high-throughput sequencing technology, protein sequence data has grown exponentially. The protein function prediction problem can be seen as a multi-label binary classification problem by extracting the features of a given protein and mapping them to the protein function label space. A variety of data sources can be mined to obtain protein function prediction features, such as protein sequence, protein structure, protein family, protein interaction network, etc. The initial steps are classical sequence-based methods, such as BLAST, which calculate the similarity between protein sequences and transmit annotations between proteins whose similarity scores exceed a specific threshold. This method has great limitations for protein function prediction without sequence similarity. Therefore, this paper analyzes the development prospect of bioanalysis and artificial intelligence through the application status and realization path of AlphaFold2 protein prediction algorithm based on artificial intelligence.","author":[{"family":"Zhang","given":"Quan"},{"family":"Liu","given":"Beichang"},{"family":"Cai","given":"Guoqing"},{"family":"Qian","given":"Jili"},{"family":"Jin","given":"Zhengyu"}],"issued":{"date-parts":[[2024]]},"DOI":"10.53469/jtpes.2024.04(02).09","URL":"https://doi.org/10.53469/jtpes.2024.04(02).09","source":"openalex"},{"id":"oa:W4399118706","type":"article-journal","title":"Frameworks for procurement, integration, monitoring, and evaluation of artificial intelligence tools in clinical settings: A systematic review","abstract":"Research on the applications of artificial intelligence (AI) tools in medicine has increased exponentially over the last few years but its implementation in clinical practice has not seen a commensurate increase with a lack of consensus on implementing and maintaining such tools. This systematic review aims to summarize frameworks focusing on procuring, implementing, monitoring, and evaluating AI tools in clinical practice. A comprehensive literature search, following PRSIMA guidelines was performed on MEDLINE, Wiley Cochrane, Scopus, and EBSCO databases, to identify and include articles recommending practices, frameworks or guidelines for AI procurement, integration, monitoring, and evaluation. From the included articles, data regarding study aim, use of a framework, rationale of the framework, details regarding AI implementation involving procurement, integration, monitoring, and evaluation were extracted. The extracted details were then mapped on to the Donabedian Plan, Do, Study, Act cycle domains. The search yielded 17,537 unique articles, out of which 47 were evaluated for inclusion based on their full texts and 25 articles were included in the review. Common themes extracted included transparency, feasibility of operation within existing workflows, integrating into existing workflows, validation of the tool using predefined performance indicators and improving the algorithm and/or adjusting the tool to improve performance. Among the four domains (Plan, Do, Study, Act) the most common domain was Plan (84%, n = 21), followed by Study (60%, n = 15), Do (52%, n = 13), & Act (24%, n = 6). Among 172 authors, only 1 (0.6%) was from a low-income country (LIC) and 2 (1.2%) were from lower-middle-income countries (LMICs). Healthcare professionals cite the implementation of AI tools within clinical settings as challenging owing to low levels of evidence focusing on integration in the Do and Act domains. The current healthcare AI landscape calls for increased data sharing and knowledge translation to facilitate common goals and reap maximum clinical benefit.","author":[{"family":"Khan","given":"Sarim"},{"family":"Hoodbhoy","given":"Zahra"},{"family":"Raja","given":"Mohummad"},{"family":"Kim","given":"Jee"},{"family":"Hogg","given":"Henry"},{"family":"Manji","given":"Afshan"},{"family":"Gulamali","given":"Freya"},{"family":"Hasan","given":"Alifia"},{"family":"Shaikh","given":"Asim"},{"family":"Tajuddin","given":"Salma"},{"family":"Khan","given":"Nida"},{"family":"Patel","given":"Manesh"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1371/journal.pdig.0000514","URL":"https://doi.org/10.1371/journal.pdig.0000514","source":"openalex"},{"id":"oa:W4390541664","type":"article-journal","title":"The role of an artificial intelligence model in antiretroviral therapy counselling and advice for people living with HIV","abstract":"OBJECTIVES: People living with HIV may find personalized access to accurate information on antiretroviral therapy (ART) challenging given the stigma and costs potentially associated with attending physical consultations. Artificial intelligence (AI) chatbots such as ChatGPT may help to lower barriers to accessing information addressing concerns around ART initiation. However, the safety and accuracy of the information provided remains to be studied. METHODS: We instructed ChatGPT to answer questions that people living with HIV frequently ask about ART, covering i) knowledge of and access to ART; ii) ART initiation, side effects, and adherence, and iii) general sexual health practices while receiving ART. We checked the accuracy of the advice against international HIV clinical practice guidelines. RESULTS: ChatGPT answered all questions accurately and comprehensively. It recognized potentially life-threatening scenarios such as abacavir hypersensitivity reaction and gave appropriate advice. However, in certain contexts, such as specific geographic locations or for pregnant individuals, the advice lacked specificity to an individual's unique circumstances and may be inadequate. Nevertheless, ChatGPT consistently re-directed the individual to seek help from a healthcare professional to obtain targeted advice. CONCLUSIONS: ChatGPT may act as a useful adjunct in the process of ART counselling for people living with HIV. Improving access to information on and knowledge about ART may improve access and adherence to ART and outcomes for people living with HIV overall.","author":[{"family":"Koh","given":"Matthew"},{"family":"Ngiam","given":"Jinghao"},{"family":"Yong","given":"Joy"},{"family":"Tambyah","given":"Paul"},{"family":"Archuleta","given":"Sophia"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/hiv.13604","URL":"https://doi.org/10.1111/hiv.13604","source":"openalex"},{"id":"oa:W4391316610","type":"article-journal","title":"Artificial intelligence‐based prediction of the rheological properties of hydrocolloids for plant‐based meat analogues","abstract":"BACKGROUND: Methylcellulose has been applied as a primary binding agent to control the quality attributes of plant-based meat analogues. H owever, a great deal of effort has been made to search for hydrocolloids to replace methylcellulose because of increasing awareness of clean labels. In this study, a machine learning framework was proposed in order to describe and predict the flow behavior of six hydrocolloid solutions, and the predicted viscosities were correlated with the textural features of their corresponding plant-based meat analogues. RESULTS: Different shear-thinning and Newtonian behaviors were observed depending on the type of hydrocolloid and the shear rate. Methylcellulose exhibited an increasing viscosity pattern with increasing temperature, compared to the other hydrocolloids. The machine learning algorithms (random forest and multilayer perceptron models) showed a better viscosity fitting performance than the constitutive equations (power law and Cross models). In addition, three hyperparameters of the multilayer perceptron model (optimizer, learning rate, and the number of hidden layers) were tuned using the Bayesian optimization algorithm. CONCLUSION: = 0.9944-0.9961/RMSE = 0.0545-0.0708). Furthermore, the machine learning-predicted viscosities overall showed similar patterns to the textural parameters of the meat analogues. © 2024 Society of Chemical Industry.","author":[{"family":"Lee","given":"Da‐yeon"},{"family":"Jeong","given":"Sungmin"},{"family":"Yun","given":"Suin"},{"family":"Lee","given":"Suyong"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/jsfa.13334","URL":"https://doi.org/10.1002/jsfa.13334","source":"openalex"},{"id":"oa:W4403132582","type":"article-journal","title":"Development of an artificial intelligence curriculum design for children in Taiwan and its impact on learning outcomes","abstract":"In the digital age, the application of Artificial Intelligence (AI) has become an irreversible trend, with its potential in the field of education being particularly noteworthy. However, there are currently few AI education programs for children in Taiwan, and there is a lack of systematic teaching resources and methods, which poses a major challenge to the promotion of AI education. To address this challenge, this study designed a tailor-made AI curriculum for children in Taiwan, aimed at enhancing their foundational knowledge in the AI field and skills in using generative AI. This study was conducted in Taiwan, involving 30 elementary school students from grades 3 and 4, employing a single-group pre-test and post-test research design. Data were collected and analyzed through pre-and post-tests on AI knowledge quizzes and AI knowledge self-assessment, as well as the expert consensual assessment technique and the pupils’ attitude toward technology survey questionnaire. Research shows that students who participated in the AI course significantly improved their test scores on AI knowledge before and after the course. The students AI knowledge increased by 62.75%, demonstrating the course’s effectiveness and showing a positive attitude towards AI technology. Additionally, the student’s project outcomes demonstrated a high level of creativity. The students exhibited an enhanced interest and positive attitude towards learning AI, expressing a willingness to participate in more AI educational courses. This work provides valuable experience and guidance for the future integration of AI technology in children’s education, offering practical guidelines for teachers and researchers on how to effectively teach AI knowledge, as well as serving as a robust reference for educational policy makers in formulating strategies to promote AI education.","author":[{"family":"Zhao","given":"Hong"},{"family":"Li","given":"Xinzhu"},{"family":"Kang","given":"Xin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1057/s41599-024-03839-z","URL":"https://doi.org/10.1057/s41599-024-03839-z","source":"openalex"},{"id":"oa:W4391486627","type":"article-journal","title":"Revolutionizing nursing education and care: The role of artificial intelligence in nursing","abstract":"One of the primary applications of artificial intelligence (AI) in nursing is patient monitoring. AI-powered monitoring systems can continuously collect and analyze patient data, such as vital signs and patient behaviors.1, 2 These systems can detect subtle changes in a patient's condition, alerting nurses to potential issues before they become critical.2, 3 Remote monitoring allows nurses to keep a closer eye on patients, improving early intervention and reducing the risk of complications.4-6 The Internet of Things (IoT) is a new technology that has various applications in various fields, including medicine. It is possible to automatically connect sensors and devices to patients without human intervention through the IoT. A wireless body area network (WBAN) is an IoT subdomain which provides the possibility of monitoring vital signs remotely. In the medical field, WBAN consists of a small network of sensors, such as pulse oximeter, gyroscope, spirometer, global positioning system, and electrooculography. RPMs are used to continuously receive clinical data from patients through internal and external sensors and help physicians make appropriate decisions. The main stages of RPMs include the following: (1) Data collection: data such as (vital signs, EEG, ECG, blood pressure, heart rate, etc.) continuously using non-invasive techniques and Invasive techniques are reviewed. (2) Data transfer and storage: All data is collected and transferred to the cloud for analysis, sorting and processing. (3) Support systems: Data are analyzed and used to help clinicians make decisions. Implanted sensors: sensors that are implanted inside the patient's body (under the patient's skin). External sensors: sensors that are directly attached to the patient's skin. Cloud Computing and Fog Computing are the two systems used in storage server. RPMs have many advantages, which can be mentioned as follows: Provide patient assurance, Increase patient awareness and responsibility, Provision of low-cost solutions.5 The application of RPMs is mostly in the field of cardiac arrhythmias, hemodynamics and vital signs.6 Remote patient monitoring systems can be used in relation to various chronic diseases such as heart disease, fall detection, mental health, diabetes. In the field of heart diseases, it is possible to measure and collect heart rate, breathing rate, ECG, breathing rate through wearable sensors. Studies presented a fall detection system based on wearable and environmental sensors, which is of great importance in the elderly. In case of abnormal data, health care professionals and family would be notified. In the field of mental illnesses, systems that remind patients of the dosage of drugs at a certain time, and monitoring adherence to drugs were used. These systems have been used in Alzheimer's and bipolar patients. In the field of diabetes, these systems were used to evaluate the patient's blood glucose level, the amount of food consumed, and the patient's blood pressure.5 AI is also improving the management of electronic health records (EHRs). AI algorithms can extract valuable insights from these records, aiding nurses in making informed decisions about patient care. AI-driven predictive analytics can identify trends and risk factors, helping nurses anticipate patient needs and tailor interventions accordingly.5-7 During the last decade, the use of AI has grown in health data, which are mainly multimodal (EHR, medical imaging, multi-omics and environmental data). Artificial intelligence has led to transformation in various fields such as health education reaserch. Thanks to the advances in AI and machine learning (ML) models, multimodal data fusion with different features can be achieved. Integrating imaging data with specific laboratory test results and demographic data leads to improved outcomes.8 As EHR data is complex and contains diagnosis, scans, laboratory test results, administrative notes doctor's signature it is difficult in working wi","author":[{"family":"Ghane","given":"Golnar"},{"family":"Ghiyasvandian","given":"Shahrzad"},{"family":"Chekeni","given":"Amir"},{"family":"Karimi","given":"Raoofeh"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/nae2.12057","URL":"https://doi.org/10.1111/nae2.12057","source":"openalex"},{"id":"oa:W4405073941","type":"article-journal","title":"Multimodality Fusion Aspects of Medical Diagnosis: A Comprehensive Review","abstract":"Utilizing information from multiple sources is a preferred and more precise method for medical experts to confirm a diagnosis. Each source provides critical information about the disease that might otherwise be absent in other modalities. Combining information from various medical sources boosts confidence in the diagnosis process, enabling the creation of an effective treatment plan for the patient. The scarcity of medical experts to diagnose diseases motivates the development of automatic diagnoses relying on multimodal data. With the progress in artificial intelligence technology, automated diagnosis using multimodal fusion techniques is now possible. Nevertheless, the concept of multimodal medical diagnosis is still new and requires an understanding of the diverse aspects of multimodal data and its related challenges. This review article examines the various aspects of multimodal medical diagnosis to equip readers, academicians, and researchers with necessary knowledge to advance multimodal medical research. The chosen articles in the study underwent thorough screening from reputable journals and publishers to offer high-quality content to readers, who can then apply the knowledge to produce quality research. Besides, the need for multimodal information and the associated challenges are discussed with solutions. Additionally, ethical issues of using artificial intelligence in medical diagnosis is also discussed.","author":[{"family":"Kumar","given":"Sachin"},{"family":"Rani","given":"Sita"},{"family":"Sharma","given":"Shivani"},{"family":"Min","given":"Hong"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/bioengineering11121233","URL":"https://doi.org/10.3390/bioengineering11121233","source":"openalex"},{"id":"oa:W4402244112","type":"article-journal","title":"A quantitative analysis of artificial intelligence research in cervical cancer: a bibliometric approach utilizing CiteSpace and VOSviewer","abstract":"Background: Cervical cancer, a severe threat to women's health, is experiencing a global increase in incidence, notably among younger demographics. With artificial intelligence (AI) making strides, its integration into medical research is expanding, particularly in cervical cancer studies. This bibliometric study aims to evaluate AI's role, highlighting research trends and potential future directions in the field. Methods: This study systematically retrieved literature from the Web of Science Core Collection (WoSCC), employing VOSviewer and CiteSpace for analysis. This included examining collaborations and keyword co-occurrences, with a focus on the relationship between citing and cited journals and authors. A burst ranking analysis identified research hotspots based on citation frequency. Results: The study analyzed 927 articles from 2008 to 2024 by 5,299 authors across 81 regions. China, the U.S., and India were the top contributors, with key institutions like the Chinese Academy of Sciences and the NIH leading in publications. Schiffman, Mark, featured among the top authors, while Jemal, A, was the most cited. 'Diagnostics' and 'IEEE Access' stood out for publication volume and citation impact, respectively. Keywords such as 'cervical cancer,' 'deep learning,' 'classification,' and 'machine learning' were dominant. The most cited article was by Berner, ES; et al., published in 2008. Conclusions: AI's application in cervical cancer research is expanding, with a growing scholarly community. The study suggests that AI, especially deep learning and machine learning, will remain a key research area, focusing on improving diagnostics and treatment. There is a need for increased international collaboration to maximize AI's potential in advancing cervical cancer research and patient care.","author":[{"family":"Zhao","given":"Ziqi"},{"family":"Hu","given":"Boqian"},{"family":"Xu","given":"Kun"},{"family":"Jiang","given":"Yizhuo"},{"family":"Xu","given":"Xisheng"},{"family":"Liu","given":"Yuliang"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fonc.2024.1431142","URL":"https://doi.org/10.3389/fonc.2024.1431142","source":"openalex"},{"id":"oa:W4386535333","type":"article-journal","title":"Overview of trials on artificial intelligence algorithms in breast cancer screening – A roadmap for international evaluation and implementation","abstract":"Accumulating evidence from retrospective studies demonstrate at least non-inferior performance when using AI algorithms with different strategies versus double-reading in mammography screening. In addition, AI algorithms for mammography screening can reduce work load by moving to single human reading. Prospective trials are essential to avoid unintended adverse consequences before incorporation of AI algorithms into UK's National Health Service (NHS) Breast Screening Programme (BSP). A stakeholders' meeting was organized in Newnham College, Cambridge, UK to undertake a review of the current evidence to enable consensus discussion on next steps required before implementation into a screening programme. It was concluded that a multicentre multivendor testing platform study with opt-out consent is preferred. AI thresholds from different vendors should be determined while maintaining non-inferior screening performance results, particularly ensuring recall rates are not increased. Automatic recall of cases using an agreed high sensitivity AI score versus automatic rule out with a low AI score set at a high sensitivity could be used. A human reader should still be involved in decision making with AI-only recalls requiring human arbitration. Standalone AI algorithms used without prompting maintain unbiased screening reading performance, but reading with prompts should be tested prospectively and ideally provided for arbitration.","author":[{"family":"Nijnatten","given":"Thiemo"},{"family":"Payne","given":"Nicholas"},{"family":"Hickman","given":"Sarah"},{"family":"Ashrafian","given":"Hutan"},{"family":"Gilbert","given":"Fiona"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.ejrad.2023.111087","URL":"https://doi.org/10.1016/j.ejrad.2023.111087","source":"openalex"},{"id":"oa:W4313571717","type":"article-journal","title":"Risk Assessment and Pancreatic Cancer: Diagnostic Management and Artificial Intelligence","abstract":"Pancreatic cancer (PC) is one of the deadliest cancers, and it is responsible for a number of deaths almost equal to its incidence. The high mortality rate is correlated with several explanations; the main one is the late disease stage at which the majority of patients are diagnosed. Since surgical resection has been recognised as the only curative treatment, a PC diagnosis at the initial stage is believed the main tool to improve survival. Therefore, patient stratification according to familial and genetic risk and the creation of screening protocol by using minimally invasive diagnostic tools would be appropriate. Pancreatic cystic neoplasms (PCNs) are subsets of lesions which deserve special management to avoid overtreatment. The current PC screening programs are based on the annual employment of magnetic resonance imaging with cholangiopancreatography sequences (MR/MRCP) and/or endoscopic ultrasonography (EUS). For patients unfit for MRI, computed tomography (CT) could be proposed, although CT results in lower detection rates, compared to MRI, for small lesions. The actual major limit is the incapacity to detect and characterize the pancreatic intraepithelial neoplasia (PanIN) by EUS and MR/MRCP. The possibility of utilizing artificial intelligence models to evaluate higher-risk patients could favour the diagnosis of these entities, although more data are needed to support the real utility of these applications in the field of screening. For these motives, it would be appropriate to realize screening programs in research settings.","author":[{"family":"Granata","given":"Vincenza"},{"family":"Fusco","given":"Roberta"},{"family":"Setola","given":"Sergio"},{"family":"Galdiero","given":"Roberta"},{"family":"Maggialetti","given":"Nicola"},{"family":"Silvestro","given":"Lucrezia"},{"family":"Bellis","given":"Mario"},{"family":"Girolamo","given":"Elena"},{"family":"Grazzini","given":"Giulia"},{"family":"Chiti","given":"Giuditta"},{"family":"Brunese","given":"Maria"},{"family":"Belli","given":"Andrea"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/cancers15020351","URL":"https://doi.org/10.3390/cancers15020351","source":"openalex"},{"id":"oa:W4393308360","type":"article-journal","title":"Postgraduate Students’ Perceptions on the Benefits Associated with Artificial Intelligence Tools on Academic Success: In Case of ChatGPT AI tool","abstract":"Postgraduate students in developing nations, such as South Africa, are increasingly leveraging artificial intelligence tools like ChatGPT to elevate their academic success in the era of the fourth industrial revolution. This study aims to explore postgraduate students' perceptions of the benefits associated with the utilisation of artificial intelligence tools, with a specific focus on ChatGPT, in their academic success in South Africa’s historically disadvantaged universities. Employing a qualitative approach, the study aims to gain a deeper understanding of postgraduate views on this subject. The sample size comprised 10 postgraduate students pursuing master's degrees within the two selected South Africa’s historically disadvantaged universities, selected through purposive sampling. Semi-structured interviews were conducted to gather insights from the postgraduate students. Thematic analysis was employed to analyse the collected data. The study's findings shed light on the significant advantages of incorporating ChatGPT in students' academic journey with special focus on research success. The study found that ChatGPT proves beneficial for postgraduate students, with some utilising the AI tool to refine their research topics before submission to their supervisors. Moreover, ChatGPT assists postgraduate students in identifying grammatical errors and paraphrasing their academic writing, contributing to the enhancement of their writing skills. In light of these findings, the study recommends the immediate development of an innovative AI ethical use policy in South Africa’s historically disadvantaged universities. This policy should emphasise ethical guidelines for postgraduate students when utilising AI tools, such as ChatGPT to ensure responsible and effective integration into their academic success.","author":[{"family":"Chauke","given":"Thulani"},{"family":"Mkhize","given":"Themba"},{"family":"Methi","given":"Lina"},{"family":"Dlamini","given":"Ntandokamenzi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.46303/jcsr.2024.4","URL":"https://doi.org/10.46303/jcsr.2024.4","source":"openalex"},{"id":"oa:W4393226201","type":"article-journal","title":"A prediction model based on artificial intelligence techniques for disintegration time and hardness of fast disintegrating tablets in pre-formulation tests","abstract":"BACKGROUND: The pharmaceutical industry is continually striving to innovate drug development and formulation processes. Orally disintegrating tablets (ODTs) have gained popularity due to their quick release and patient-friendly characteristics. The choice of excipients in tablet formulations plays a critical role in ensuring product quality, highlighting its importance in tablet creation. The traditional trial-and-error approach to this process is both expensive and time-intensive. To tackle these obstacles, we introduce a fresh approach leveraging machine learning and deep learning methods to automate and enhance pre-formulation drug design. METHODS: We collected a comprehensive dataset of 1983 formulations, including excipient names, quantities, active ingredient details, and various physicochemical attributes. Our study focused on predicting two critical control test parameters: tablet disintegration time and hardness. We compared a range of models like deep learning, artificial neural networks, support vector machines, decision trees, multiple linear regression, and random forests. RESULTS: A 12-layer deep neural network, as a form of deep learning, surpassed alternative techniques by achieving 73% accuracy for disintegration time and 99% for tablet hardness. This success underscores its efficacy in predicting complex pharmaceutical factors. Such an approach streamlines the drug formulation process, reducing iterations and material consumption. CONCLUSIONS: Our findings highlight the deep learning potential in pharmaceutical formulations, particularly for tablet hardness prediction. Future work should focus on enlarging the dataset to improve model effectiveness and extend its application in pharmaceutical product development and assessment.","author":[{"family":"Momeni","given":"Mehri"},{"family":"Afkanpour","given":"Marziyeh"},{"family":"Rakhshani","given":"Saleh"},{"family":"Mehrabian","given":"Amin"},{"family":"Tabesh","given":"Hamed"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s12911-024-02485-4","URL":"https://doi.org/10.1186/s12911-024-02485-4","source":"openalex"},{"id":"oa:W4405215661","type":"article-journal","title":"Transforming Medical Libraries: Opportunities, Challenges, and Strategies for Integrating Artificial Intelligence","abstract":"This study examines the role of Artificial Intelligence (AI) in transforming medical libraries, focusing on identifying opportunities, challenges, and strategies for effective integration. Medical libraries, essential resources for healthcare professionals, are increasingly leveraging AI to improve information retrieval, data management, and user services. The study aims to analyze the benefits AI brings to medical libraries, the obstacles to its adoption, and best practices for implementation. To achieve this, three specific objectives guided the study. A qualitative approach was applied, using a systematic review of literature from databases such as Scopus, Web of Science, and Google Scholar. Articles published between 2020 and 2024 were included, with non-English publications excluded. Findings indicate that AI offers substantial benefits, including enhanced information retrieval, automated data management, and improved user engagement through personalized services and research support. However, challenges such as ethical issues, data privacy concerns, infrastructure needs, and staff training remain. The study concludes that, while AI holds great potential for advancing medical libraries, overcoming these challenges will require strategic planning, investment in infrastructure, and the use of transparent, explainable AI solutions.","author":[{"family":"Orubebe","given":"Ebiere"},{"family":"Ijaja","given":"Esther"},{"family":"Ogwula","given":"John"},{"family":"Oladokun","given":"Bolaji"}],"issued":{"date-parts":[[2024]]},"DOI":"10.70112/ajist-2024.14.2.4298","URL":"https://doi.org/10.70112/ajist-2024.14.2.4298","source":"openalex"},{"id":"oa:W4405397757","type":"article-journal","title":"Artificial intelligence-driven predictive maintenance in IoT systems","abstract":"The study looks at the application of AI-driven predictive maintenance in IoT systems. Predictive device failure, efficient reduction in system downtime, reduced maintenance costs, and overall efficiency in connected devices will be enabled through machine learning and deep learning algorithms. The AI models developed within this research were able to provide a prediction accuracy of 92%, while the traditional methods of maintenance were far behind at 78%. It resulted in a 35% reduction in system downtime and a 28% decrease in maintenance costs while reducing the error rate to 8%. The above results bring out the potential of AI-based solutions for real-time predictive maintenance over complex IoT networks. It concludes by indicating some further research vectors, such as the refinement of the model and the extension of AI-driven predictive maintenance for broader applications in IoT, such as smart cities and healthcare systems.","author":[{"family":"Abdulrazzq","given":"Raghdah"},{"family":"Sajid","given":"Nisreen"},{"family":"Hasan","given":"Md"}],"issued":{"date-parts":[[2024]]},"DOI":"10.46932/sfjdv5n12-030","URL":"https://doi.org/10.46932/sfjdv5n12-030","source":"openalex"},{"id":"oa:W4399527350","type":"article-journal","title":"Detection of oral cancer and oral potentially malignant disorders using artificial intelligence‐based image analysis","abstract":"BACKGROUND: We aimed to construct an artificial intelligence-based model for detecting oral cancer and dysplastic leukoplakia using oral cavity images captured with a single-lens reflex camera. SUBJECTS AND METHODS: We used 1043 images of lesions from 424 patients with oral squamous cell carcinoma (OSCC), leukoplakia, and other oral mucosal diseases. An object detection model was constructed using a Single Shot Multibox Detector to detect oral diseases and their locations using images. The model was trained using 523 images of oral cancer, and its performance was evaluated using images of oral cancer (n = 66), leukoplakia (n = 49), and other oral diseases (n = 405). RESULTS: For the detection of only OSCC versus OSCC and leukoplakia, the model demonstrated a sensitivity of 93.9% versus 83.7%, a negative predictive value of 98.8% versus 94.5%, and a specificity of 81.2% versus 81.2%. CONCLUSIONS: Our proposed model is a potential diagnostic tool for oral diseases.","author":[{"family":"Kouketsu","given":"Atsumu"},{"family":"Doi","given":"Chiaki"},{"family":"Tanaka","given":"Hiroaki"},{"family":"Araki","given":"Takashi"},{"family":"Nakayama","given":"Rina"},{"family":"Toyooka","given":"Tsuguyoshi"},{"family":"Hiyama","given":"Satoshi"},{"family":"Iikubo","given":"Masahiro"},{"family":"Osaka","given":"Ken"},{"family":"Sasaki","given":"Keiichi"},{"family":"Nagai","given":"Hirokazu"},{"family":"Sugiura","given":"Tsuyoshi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/hed.27843","URL":"https://doi.org/10.1002/hed.27843","source":"openalex"},{"id":"oa:W4404413393","type":"article-journal","title":"Human-Computer Interaction: A Literature Review of Artificial Intelligence and Communication in Healthcare","abstract":"The integration of artificial intelligence (AI) into healthcare communication has rapidly evolved, driven by advancements in large language models (LLMs) such as Chat Generative Pre-trained Transformer (ChatGPT). This literature review explores AI's role in patient-physician interactions, particularly focusing on its capacity to enhance communication by bridging language barriers, summarizing complex medical data, and offering empathetic responses. AI's strengths lie in its ability to deliver comprehensible, concise, and medically accurate information. Studies indicate AI can outperform human physicians in certain communicative aspects, such as empathy and clarity, with models like ChatGPT and the Medical Pathways Language Model (Med-PaLM) demonstrating high effectiveness in these areas. However, significant challenges remain, including occasional inaccuracies and \"hallucinations,\" where AI-generated content is irrelevant or medically inaccurate. These limitations highlight the need for continued refinement in AI algorithms to ensure reliability and consistency in sensitive healthcare settings. The review underscores the potential of AI as a transformative tool in health communication while advocating for further research and policy development to mitigate risks and enhance AI's integration into clinical practice.","author":[{"family":"Clay","given":"Theo"},{"family":"Steel","given":"Zephy"},{"family":"Jacobs","given":"Chris"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.73763","URL":"https://doi.org/10.7759/cureus.73763","source":"openalex"},{"id":"oa:W4399608418","type":"article-journal","title":"The Use of Artificial Intelligence for Skin Disease Diagnosis in Primary Care Settings: A Systematic Review","abstract":"The prevalence of dermatological conditions in primary care, coupled with challenges such as dermatologist shortages and rising consultation costs, highlights the need for innovative solutions. Artificial intelligence (AI) holds promise for improving the diagnostic analysis of skin lesion images, potentially enhancing patient care in primary settings. This systematic review following PRISMA guidelines examined primary studies (2012-2022) assessing AI algorithms' diagnostic accuracy for skin diseases in primary care. Studies were screened for eligibility based on their availability in the English language and exclusion criteria, with risk of bias evaluated using QUADAS-2. PubMed, Scopus, and Web of Science were searched. Fifteen studies (2019-2022), primarily from Europe and the USA, focusing on diagnostic accuracy were included. Sensitivity ranged from 58% to 96.1%, with accuracies varying from 0.41 to 0.93. AI applications encompassed triage and diagnostic support across diverse skin conditions in primary care settings, involving both patients and primary care professionals. While AI demonstrates potential for enhancing the accuracy of skin disease diagnostics in primary care, further research is imperative to address study heterogeneity and ensure algorithm reliability across diverse populations. Future investigations should prioritise robust dataset development and consider representative patient samples. Overall, AI may improve dermatological diagnosis in primary care, but careful consideration of algorithm limitations and implementation strategies is required.","author":[{"family":"Escalé-Besa","given":"Anna"},{"family":"Vidalalaball","given":"Josep"},{"family":"Catalina","given":"Queralt"},{"family":"Gracia","given":"Victor"},{"family":"Marín-Gomez","given":"Francesc"},{"family":"Fustercasanovas","given":"Aïna"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/healthcare12121192","URL":"https://doi.org/10.3390/healthcare12121192","source":"openalex"},{"id":"oa:W4400117494","type":"article-journal","title":"Resilient Artificial Intelligence in Health: Synthesis and Research Agenda Toward Next-Generation Trustworthy Clinical Decision Support","abstract":"Artificial intelligence (AI)-based clinical decision support systems are gaining momentum by relying on a greater volume and variety of secondary use data. However, the uncertainty, variability, and biases in real-world data environments still pose significant challenges to the development of health AI, its routine clinical use, and its regulatory frameworks. Health AI should be resilient against real-world environments throughout its lifecycle, including the training and prediction phases and maintenance during production, and health AI regulations should evolve accordingly. Data quality issues, variability over time or across sites, information uncertainty, human-computer interaction, and fundamental rights assurance are among the most relevant challenges. If health AI is not designed resiliently with regard to these real-world data effects, potentially biased data-driven medical decisions can risk the safety and fundamental rights of millions of people. In this viewpoint, we review the challenges, requirements, and methods for resilient AI in health and provide a research framework to improve the trustworthiness of next-generation AI-based clinical decision support.","author":[{"family":"Sáez","given":"Carlos"},{"family":"Ferri","given":"Pablo"},{"family":"Garcíagómez","given":"Juan"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/50295","URL":"https://doi.org/10.2196/50295","source":"openalex"},{"id":"oa:W4400141746","type":"manuscript","title":"The Rise of Artificial Intelligence in Educational Measurement: Opportunities and Ethical Challenges","abstract":"The integration of artificial intelligence (AI) in educational measurement has revolutionized assessment methods, enabling automated scoring, rapid content analysis, and personalized feedback through machine learning and natural language processing. These advancements provide timely, consistent feedback and valuable insights into student performance, thereby enhancing the assessment experience. However, the deployment of AI in education also raises significant ethical concerns regarding validity, reliability, transparency, fairness, and equity. Issues such as algorithmic bias and the opacity of AI decision-making processes pose risks of perpetuating inequalities and affecting assessment outcomes. Responding to these concerns, various stakeholders, including educators, policymakers, and organizations, have developed guidelines to ensure ethical AI use in education. The National Council of Measurement in Education's Special Interest Group on AI in Measurement and Education (AIME) also focuses on establishing ethical standards and advancing research in this area. In this paper, a diverse group of AIME members examines the ethical implications of AI-powered tools in educational measurement, explores significant challenges such as automation bias and environmental impact, and proposes solutions to ensure AI's responsible and effective use in education.","author":[{"family":"Bulut","given":"Okan"},{"family":"Beiting-Parrish","given":"Maggie"},{"family":"Casabianca","given":"Jodi"},{"family":"Slater","given":"Sharon"},{"family":"Jiao","given":"Hong"},{"family":"Song","given":"Dan"},{"family":"Ormerod","given":"Christopher"},{"family":"Fabiyi","given":"Deborah"},{"family":"Ivan","given":"Rodica"},{"family":"Walsh","given":"Cole"},{"family":"Rios","given":"Oscar"},{"family":"Wilson","given":"Joshua"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2406.18900","URL":"https://doi.org/10.48550/arxiv.2406.18900","source":"openalex"},{"id":"oa:W4386781106","type":"article-journal","title":"Accuracy, Reliability, and Comprehensibility of ChatGPT-Generated Medical Responses for Patients With Nonalcoholic Fatty Liver Disease","abstract":"Nonalcoholic fatty liver disease (NAFLD) is an increasing global health problem and is expected to become the leading indication for liver transplantation.1 There are no approved NAFLD-specific pharmacotherapies, and lifestyle modification is the primary recommended therapy.2 Innovative approaches to facilitate the implementation and long-term maintenance of lifestyle changes are needed to address the challenging and complex nature of the management of NAFLD, which recently was renamed as metabolic dysfunction–associated steatotic liver disease, to overcome the limitations and stigma of the previous name.3,4 Artificial intelligence (AI)-powered chatbots have been shown to provide effective personalized support and education to patients, with the potential to complement health care resources. The OpenAI Foundation’s AI chatbot, Chat Generative Pretrained Transformer (ChatGPT), has attracted worldwide attention for its remarkable performance in question–answer tasks.5–7 This study evaluated the accuracy, completeness, and comprehensiveness of chatGPT’s responses to NAFLD-related questions, with the aim of assessing its performance in addressing patients’ queries about the disease and lifestyle behaviors.","author":[{"family":"Pugliese","given":"Nicola"},{"family":"Wong","given":"Vincent"},{"family":"Schattenberg","given":"Jörn"},{"family":"Romerogómez","given":"Manuel"},{"family":"Sebastiani","given":"Giada"},{"family":"Castéra","given":"Laurent"},{"family":"Hassan","given":"Cesare"},{"family":"Manousou","given":"Pinelopi"},{"family":"Miele","given":"Luca"},{"family":"Peck","given":"Raquel"},{"family":"Petta","given":"Salvatore"},{"family":"Valenti","given":"Luca"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.cgh.2023.08.033","URL":"https://doi.org/10.1016/j.cgh.2023.08.033","source":"openalex"},{"id":"oa:W4398244260","type":"article-journal","title":"Interpretation of Clinical Retinal Images Using an Artificial Intelligence Chatbot","abstract":"Purpose To assess the performance of ChatGPT-4 in providing accurate diagnoses to retina teaching cases from OCTCases. Design Cross-sectional study. Subjects Retina teaching cases from OCTCases. Methods We prompted a custom chatbot with 69 retina cases containing multimodal ophthalmic images, asking it to provide the most likely diagnosis. In a sensitivity analysis, we inputted increasing amounts of clinical information pertaining to each case until the chatbot achieved a correct diagnosis. We performed multivariable logistic regressions on Stata v17.0 (StataCorp LLC, College Station, Texas) to investigate associations between the amount of text-based information inputted per prompt and the odds of the chatbot achieving a correct diagnosis, adjusting for the laterality of cases, number of ophthalmic images inputted, and imaging modalities. Main Outcome Measures Our primary outcome was the proportion of cases for which the chatbot was able to provide a correct diagnosis. Our secondary outcome was the chatbot's performance in relation to the amount of text-based information accompanying ophthalmic images. Results Across 69 retina cases collectively containing 139 ophthalmic images, the chatbot was able to provide a definitive, correct diagnosis for 35 (50.7%) cases. The chatbot needed variable amounts of clinical information to achieve a correct diagnosis, where the entire patient description as presented by OCTCases was required for a majority of correctly diagnosed cases (23/35 cases, 65.7%). Relative to when the chatbot was only prompted with a patient's age and sex, the chatbot achieved a higher odds of a correct diagnosis when prompted with an entire patient description (OR=10.1, 95%CI=[3.3, 30.3], p<0.01). Despite providing an incorrect diagnosis for 34 (49.3%) cases, the chatbot listed the correct diagnosis within its differential diagnosis for 7 (20.6%) of these incorrectly answered cases. Conclusions This custom chatbot was able to accurately diagnose approximately half of the retina cases requiring multimodal input, albeit relying heavily on text-based contextual information that accompanied ophthalmic images. The diagnostic ability of the chatbot in interpretation of multimodal imaging without text-based information is currently limited. The appropriate use of the chatbot in this setting is of utmost importance, given bioethical concerns.","author":[{"family":"Mihalache","given":"Andrew"},{"family":"Huang","given":"Ryan"},{"family":"Mikhail","given":"David"},{"family":"Popovic","given":"Marko"},{"family":"Shor","given":"Reut"},{"family":"Pereira","given":"Austin"},{"family":"Kwok","given":"Jason"},{"family":"Yan","given":"Peng"},{"family":"Wong","given":"David"},{"family":"Kertes","given":"Peter"},{"family":"Kohly","given":"Radha"},{"family":"Muni","given":"Rajeev"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.xops.2024.100556","URL":"https://doi.org/10.1016/j.xops.2024.100556","source":"openalex"},{"id":"oa:W4394687015","type":"article-journal","title":"Performance of an Artificial Intelligence System for Breast Cancer Detection on Screening Mammograms from BreastScreen Norway","abstract":"“Just Accepted” papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To explore the standalone breast cancer detection performance at different risk score thresholds of a commercially available artificial intelligence (AI) system. Materials and Methods This retrospective study included information from 661,695 digital mammographic examinations performed among 242,629 female individuals screened as a part of x, 2004–2018. The study sample included 3807 screen-detected cancers (SDC) and 1110 interval breast cancers (IC). A continuous examination level risk score by the AI system was used to measure performance as the area under the receiver operating characteristic curve (AUC) with 95% CIs and cancer detection at different AI risk score thresholds. Results The AUC of the AI system was 0.93 (95% CI: 0.92–0.93) for SDC and IC combined and 0.97 (95% CI: 0.97–0.97) for SDC. In a setting where 10% of the examinations with the highest AI risk scores were defined as positive and 90% with the lowest scores as negative, 92.0% (3502/3807) of the SDC and 44.6% (495/1100) of the IC were identified by AI. In this scenario, 68.5% (10 987/16 029) of false positive screening results (negative recall assessment) were considered negative by AI. When 50% was used as the cut-off, 99.3% (3781/3807) of the SDC and 85.2% (946/1100) of the IC were identified as positive by AI, while 17.0% (2725/16 029) of the false positives were considered as negative. Conclusion The AI system showed high performance in detecting breast cancers within 2 years of screening mammography and a potential for triaging low-risk mammograms to reduce radiologist workload. ©RSNA, 2024","author":[{"family":"Larsen","given":"Marthe"},{"family":"Olstad","given":"Camilla"},{"family":"Lee","given":"Christoph"},{"family":"Hovda","given":"Tone"},{"family":"Hoff","given":"Solveig"},{"family":"Martiniussen","given":"Marit"},{"family":"Mikalsen","given":"Karl"},{"family":"Lund-Hanssen","given":"Håkon"},{"family":"Solli","given":"Helene"},{"family":"Silberhorn","given":"Marko"},{"family":"Sulheim","given":"Åse"},{"family":"Auensen","given":"Steinar"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1148/ryai.230375","URL":"https://doi.org/10.1148/ryai.230375","source":"openalex"},{"id":"oa:W4396886447","type":"article-journal","title":"Accuracy of artificial intelligence-assisted endoscopy in the diagnosis of gastric intestinal metaplasia: A systematic review and meta-analysis","abstract":"BACKGROUND AND AIMS: Gastric intestinal metaplasia is a precancerous disease, and a timely diagnosis is essential to delay or halt cancer progression. Artificial intelligence (AI) has found widespread application in the field of disease diagnosis. This study aimed to conduct a comprehensive evaluation of AI's diagnostic accuracy in detecting gastric intestinal metaplasia in endoscopy, compare it to endoscopists' ability, and explore the main factors affecting AI's performance. METHODS: The study followed the PRISMA-DTA guidelines, and the PubMed, Embase, Web of Science, Cochrane, and IEEE Xplore databases were searched to include relevant studies published by October 2023. We extracted the key features and experimental data of each study and combined the sensitivity and specificity metrics by meta-analysis. We then compared the diagnostic ability of the AI versus the endoscopists using the same test data. RESULTS: Twelve studies with 11,173 patients were included, demonstrating AI models' efficacy in diagnosing gastric intestinal metaplasia. The meta-analysis yielded a pooled sensitivity of 94% (95% confidence interval: 0.92-0.96) and specificity of 93% (95% confidence interval: 0.89-0.95). The combined area under the receiver operating characteristics curve was 0.97. The results of meta-regression and subgroup analysis showed that factors such as study design, endoscopy type, number of training images, and algorithm had a significant effect on the diagnostic performance of AI. The AI exhibited a higher diagnostic capacity than endoscopists (sensitivity: 95% vs. 79%). CONCLUSIONS: AI-aided diagnosis of gastric intestinal metaplasia using endoscopy showed high performance and clinical diagnostic value. However, further prospective studies are required to validate these findings.","author":[{"family":"Li","given":"Na"},{"family":"Yang","given":"Jian"},{"family":"Li","given":"Xiaodong"},{"family":"Shi","given":"Yanting"},{"family":"Wang","given":"Kunhong"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1371/journal.pone.0303421","URL":"https://doi.org/10.1371/journal.pone.0303421","source":"openalex"},{"id":"oa:W4388342253","type":"article-journal","title":"An Artificial Intelligence Generated Automated Algorithm to Measure Total Kidney Volume in ADPKD","abstract":"Introduction: Accurate tools to inform individual prognosis in patients with autosomal dominant polycystic kidney disease (ADPKD) are lacking. Here, we report an artificial intelligence (AI)-generated method for routinely measuring total kidney volume (TKV). Methods: An ensemble U-net algorithm was created using the nnUNet approach. The training and internal cross-validation cohort consisted of all 1.5T magnetic resonance imaging (MRI) data acquired using 5 different MRI scanners (454 kidneys, 227 scans) in the CYSTic consortium, which was first manually segmented by a single human operator. As an independent validation cohort, we utilized 48 sequential clinical MRI scans with reference results of manual segmentation acquired by 6 individual analysts at a single center. The tool was then implemented for clinical use and its performance analyzed. Results: mutations (79%) and typical disease (Mayo Imaging class 1, 86%). The median DICE score on the clinical validation data set between the algorithm and human analysts was 0.96 for left and right kidneys with a median TKV error of -1.8%. The time taken to manually segment kidneys in the CYSTic data set was 56 (±28) minutes, whereas manual corrections of the algorithm output took 8.5 (±9.2) minutes per scan. Conclusion: Our AI-based algorithm demonstrates performance comparable to manual segmentation. Its rapidity and precision in real-world clinical cases demonstrate its suitability for clinical application.","author":[{"family":"Taylor","given":"Jonathan"},{"family":"Thomas","given":"Richard"},{"family":"Metherall","given":"Peter"},{"family":"Gastel","given":"Marieke"},{"family":"Gall","given":"Émilie"},{"family":"Caroli","given":"Anna"},{"family":"Furlano","given":"Mónica"},{"family":"Demoulin","given":"Nathalie"},{"family":"Devuyst","given":"Olivier"},{"family":"Winterbottom","given":"Jean"},{"family":"Torrá","given":"Roser"},{"family":"Perico","given":"Norberto"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.ekir.2023.10.029","URL":"https://doi.org/10.1016/j.ekir.2023.10.029","source":"openalex"},{"id":"oa:W4394911989","type":"article-journal","title":"Predicting non-muscle invasive bladder cancer outcomes using artificial intelligence: a systematic review using APPRAISE-AI","abstract":"Accurate prediction of recurrence and progression in non-muscle invasive bladder cancer (NMIBC) is essential to inform management and eligibility for clinical trials. Despite substantial interest in developing artificial intelligence (AI) applications in NMIBC, their clinical readiness remains unclear. This systematic review aimed to critically appraise AI studies predicting NMIBC outcomes, and to identify common methodological and reporting pitfalls. MEDLINE, EMBASE, Web of Science, and Scopus were searched from inception to February 5th, 2024 for AI studies predicting NMIBC recurrence or progression. APPRAISE-AI was used to assess methodological and reporting quality of these studies. Performance between AI and non-AI approaches included within these studies were compared. A total of 15 studies (five on recurrence, four on progression, and six on both) were included. All studies were retrospective, with a median follow-up of 71 months (IQR 32-93) and median cohort size of 125 (IQR 93-309). Most studies were low quality, with only one classified as high quality. While AI models generally outperformed non-AI approaches with respect to accuracy, c-index, sensitivity, and specificity, this margin of benefit varied with study quality (median absolute performance difference was 10 for low, 22 for moderate, and 4 for high quality studies). Common pitfalls included dataset limitations, heterogeneous outcome definitions, methodological flaws, suboptimal model evaluation, and reproducibility issues. Recommendations to address these challenges are proposed. These findings emphasise the need for collaborative efforts between urological and AI communities paired with rigorous methodologies to develop higher quality models, enabling AI to reach its potential in enhancing NMIBC care.","author":[{"family":"Kwong","given":"Jethro"},{"family":"Wu","given":"Jeremy"},{"family":"Malik","given":"Shamir"},{"family":"Khondker","given":"Adree"},{"family":"Gupta","given":"Naveen"},{"family":"Bodnariuc","given":"Nicole"},{"family":"Narayana","given":"Krishnateja"},{"family":"Malik","given":"Mikail"},{"family":"Kwast","given":"Theodorus"},{"family":"Johnson","given":"Alistair"},{"family":"Zlotta","given":"Alexandre"},{"family":"Kulkarni","given":"Girish"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41746-024-01088-7","URL":"https://doi.org/10.1038/s41746-024-01088-7","source":"openalex"},{"id":"oa:W4391751159","type":"article-journal","title":"Research and application of artificial intelligence in dentistry from lower-middle income countries – a scoping review","abstract":"Artificial intelligence (AI) has been integrated into dentistry for improvement of current dental practice. While many studies have explored the utilization of AI in various fields, the potential of AI in dentistry, particularly in low-middle income countries (LMICs) remains understudied. This scoping review aimed to study the existing literature on the applications of artificial intelligence in dentistry in low-middle income countries. A comprehensive search strategy was applied utilizing three major databases: PubMed, Scopus, and EBSCO Dentistry & Oral Sciences Source. The search strategy included keywords related to AI, Dentistry, and LMICs. The initial search yielded a total of 1587, out of which 25 articles were included in this review. Our findings demonstrated that limited studies have been carried out in LMICs in terms of AI and dentistry. Most of the studies were related to Orthodontics. In addition gaps in literature were noted such as cost utility and patient experience were not mentioned in the included studies.","author":[{"family":"Umer","given":"Fahad"},{"family":"Adnan","given":"Samira"},{"family":"Lal","given":"Abhishek"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s12903-024-03970-y","URL":"https://doi.org/10.1186/s12903-024-03970-y","source":"openalex"},{"id":"oa:W4399047878","type":"article-journal","title":"Revolutionizing Cardiology through Artificial Intelligence—Big Data from Proactive Prevention to Precise Diagnostics and Cutting-Edge Treatment—A Comprehensive Review of the Past 5 Years","abstract":"BACKGROUND: Artificial intelligence (AI) can radically change almost every aspect of the human experience. In the medical field, there are numerous applications of AI and subsequently, in a relatively short time, significant progress has been made. Cardiology is not immune to this trend, this fact being supported by the exponential increase in the number of publications in which the algorithms play an important role in data analysis, pattern discovery, identification of anomalies, and therapeutic decision making. Furthermore, with technological development, there have appeared new models of machine learning (ML) and deep learning (DP) that are capable of exploring various applications of AI in cardiology, including areas such as prevention, cardiovascular imaging, electrophysiology, interventional cardiology, and many others. In this sense, the present article aims to provide a general vision of the current state of AI use in cardiology. RESULTS: We identified and included a subset of 200 papers directly relevant to the current research covering a wide range of applications. Thus, this paper presents AI applications in cardiovascular imaging, arithmology, clinical or emergency cardiology, cardiovascular prevention, and interventional procedures in a summarized manner. Recent studies from the highly scientific literature demonstrate the feasibility and advantages of using AI in different branches of cardiology. CONCLUSIONS: The integration of AI in cardiology offers promising perspectives for increasing accuracy by decreasing the error rate and increasing efficiency in cardiovascular practice. From predicting the risk of sudden death or the ability to respond to cardiac resynchronization therapy to the diagnosis of pulmonary embolism or the early detection of valvular diseases, AI algorithms have shown their potential to mitigate human error and provide feasible solutions. At the same time, limits imposed by the small samples studied are highlighted alongside the challenges presented by ethical implementation; these relate to legal implications regarding responsibility and decision making processes, ensuring patient confidentiality and data security. All these constitute future research directions that will allow the integration of AI in the progress of cardiology.","author":[{"family":"Stamate","given":"Elena"},{"family":"Piraianu","given":"Alin"},{"family":"Ciobotaru","given":"Oana"},{"family":"Crassas","given":"R"},{"family":"Duca","given":"Oana"},{"family":"Fulga","given":"Ana"},{"family":"Grigore","given":"Ionica"},{"family":"Vintila","given":"V"},{"family":"Fulga","given":"Iuliu"},{"family":"Ciobotaru","given":"Octavian"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/diagnostics14111103","URL":"https://doi.org/10.3390/diagnostics14111103","source":"openalex"},{"id":"oa:W4402172079","type":"article-journal","title":"Impact of Artificial Intelligence in Endodontics: Precision, Predictions, and Prospects","abstract":"Artificial intelligence (AI) has become increasingly prevalent and significant across many industries, including the dental field. AI has shown accuracy and precision in detecting, evaluating, and predicting diseases. It can imitate human intelligence to carry out sophisticated predictions and decision-making in the health-care industry, especially in endodontics. AI models have demonstrated a wide range of applications in the field of endodontics. These include examining the anatomy of the root canal system, predicting the survival of dental pulp stem cells, gauging working lengths, identifying per apical lesions and root fractures, and predicting the outcome of retreatment treatments. Future uses of this technology were discussed in terms of robotic endodontic surgery, drug-drug interactions, patient care, scheduling, and prognostic diagnosis.","author":[{"family":"Parinitha","given":"MS"},{"family":"Doddawad","given":"Vidya"},{"family":"Kalgeri","given":"Sowmya"},{"family":"Gowda","given":"Samyuka"},{"family":"Patil","given":"Sahana"}],"issued":{"date-parts":[[2024]]},"DOI":"10.4103/jmss.jmss_7_24","URL":"https://doi.org/10.4103/jmss.jmss_7_24","source":"openalex"},{"id":"oa:W4320492850","type":"article-journal","title":"Internet of Medical Things Privacy and Security: Challenges, Solutions, and Future Trends from a New Perspective","abstract":"The Internet of Medical Things (IoMT), an application of the Internet of Things (IoT) in the medical domain, allows data to be transmitted across communication networks. In particular, IoMT can help improve the quality of life of citizens and older people by monitoring and managing the body’s vital signs, including blood pressure, temperature, heart rate, and others. Since IoMT has become the main platform for information exchange and making high-level decisions, it is necessary to guarantee its reliability and security. The growth of IoMT in recent decades has attracted the interest of many experts. This study provides an in-depth analysis of IoT and IoMT by focusing on security concerns from different points of view, making this comprehensive survey unique compared to other existing studies. A total of 187 articles from 2010 to 2022 are collected and categorized according to the type of applications, year of publications, variety of applications, and other novel perspectives. We compare the current studies based on the above criteria and provide a comprehensive analysis to pave the way for researchers working in this area. In addition, we highlight the trends and future work. We have found that blockchain, as a key technology, has solved many problems of security, authentication, and maintenance of IoT systems due to the decentralized nature of the blockchain. In the current study, this technology is examined from the application fields’ points of view, especially in the health sector, due to its additional importance compared to other fields.","author":[{"family":"Kamalov","given":"Firuz"},{"family":"Pourghebleh","given":"Behrouz"},{"family":"Gheisari","given":"Mehdi"},{"family":"Liu","given":"Yang"},{"family":"Moussa","given":"Sherif"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/su15043317","URL":"https://doi.org/10.3390/su15043317","source":"openalex"},{"id":"oa:W4394728138","type":"article-journal","title":"Diagnosis of soil-transmitted helminth infections with digital mobile microscopy and artificial intelligence in a resource-limited setting","abstract":"BACKGROUND: Infections caused by soil-transmitted helminths (STHs) are the most prevalent neglected tropical diseases and result in a major disease burden in low- and middle-income countries, especially in school-aged children. Improved diagnostic methods, especially for light intensity infections, are needed for efficient, control and elimination of STHs as a public health problem, as well as STH management. Image-based artificial intelligence (AI) has shown promise for STH detection in digitized stool samples. However, the diagnostic accuracy of AI-based analysis of entire microscope slides, so called whole-slide images (WSI), has previously not been evaluated on a sample-level in primary healthcare settings in STH endemic countries. METHODOLOGY/PRINCIPAL FINDINGS: Stool samples (n = 1,335) were collected during 2020 from children attending primary schools in Kwale County, Kenya, prepared according to the Kato-Katz method at a local primary healthcare laboratory and digitized with a portable whole-slide microscopy scanner and uploaded via mobile networks to a cloud environment. The digital samples of adequate quality (n = 1,180) were split into a training (n = 388) and test set (n = 792) and a deep-learning system (DLS) developed for detection of STHs. The DLS findings were compared with expert manual microscopy and additional visual assessment of the digital samples in slides with discordant results between the methods. Manual microscopy detected 15 (1.9%) Ascaris lumbricoides, 172 (21.7%) Tricuris trichiura and 140 (17.7%) hookworm (Ancylostoma duodenale or Necator americanus) infections in the test set. Importantly, more than 90% of all STH positive cases represented light intensity infections. With manual microscopy as the reference standard, the sensitivity of the DLS as the index test for detection of A. lumbricoides, T. trichiura and hookworm was 80%, 92% and 76%, respectively. The corresponding specificity was 98%, 90% and 95%. Notably, in 79 samples (10%) classified as negative by manual microscopy for a specific species, STH eggs were detected by the DLS and confirmed correct by visual inspection of the digital samples. CONCLUSIONS/SIGNIFICANCE: Analysis of digitally scanned stool samples with the DLS provided high diagnostic accuracy for detection of STHs. Importantly, a substantial number of light intensity infections were missed by manual microscopy but detected by the DLS. Thus, analysis of WSIs with image-based AI may provide a future tool for improved detection of STHs in a primary healthcare setting, which in turn could facilitate monitoring and evaluation of control programs.","author":[{"family":"Lundin","given":"Johan"},{"family":"Suutala","given":"Antti"},{"family":"Holmström","given":"Oscar"},{"family":"Henriksson","given":"Samuel"},{"family":"Valkamo","given":"Severi"},{"family":"Kaingu","given":"Harrison"},{"family":"Kinyua","given":"Felix"},{"family":"Muinde","given":"Martin"},{"family":"Lundin","given":"Mikael"},{"family":"Diwan","given":"Vinod"},{"family":"Mårtensson","given":"Andreas"},{"family":"Linder","given":"Nina"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1371/journal.pntd.0012041","URL":"https://doi.org/10.1371/journal.pntd.0012041","source":"openalex"},{"id":"oa:W4384277227","type":"article-journal","title":"Artificial intelligence in a prediction model for postendoscopic retrograde cholangiopancreatography pancreatitis","abstract":"OBJECTIVES: In this study we aimed to develop an artificial intelligence-based model for predicting postendoscopic retrograde cholangiopancreatography (ERCP) pancreatitis (PEP). METHODS: We retrospectively reviewed ERCP patients at Nagoya University Hospital (NUH) and Toyota Memorial Hospital (TMH). We constructed two prediction models, a random forest (RF), one of the machine-learning algorithms, and a logistic regression (LR) model. First, we selected features of each model from 40 possible features. Then the models were trained and validated using three fold cross-validation in the NUH cohort and tested in the TMH cohort. The area under the receiver operating characteristic curve (AUROC) was used to assess model performance. Finally, using the output parameters of the RF model, we classified the patients into low-, medium-, and high-risk groups. RESULTS: A total of 615 patients at NUH and 544 patients at TMH were enrolled. Ten features were selected for the RF model, including albumin, creatinine, biliary tract cancer, pancreatic cancer, bile duct stone, total procedure time, pancreatic duct injection, pancreatic guidewire-assisted technique without a pancreatic stent, intraductal ultrasonography, and bile duct biopsy. In the three fold cross-validation, the RF model showed better predictive ability than the LR model (AUROC 0.821 vs. 0.660). In the test, the RF model also showed better performance (AUROC 0.770 vs. 0.663, P = 0.002). Based on the RF model, we classified the patients according to the incidence of PEP (2.9%, 10.0%, and 23.9%). CONCLUSION: We developed an RF model. Machine-learning algorithms could be powerful tools to develop accurate prediction models.","author":[{"family":"Takahashi","given":"Hidekazu"},{"family":"Ohno","given":"Eizaburo"},{"family":"Furukawa","given":"Taiki"},{"family":"Yamao","given":"Kentaro"},{"family":"Ishikawa","given":"Takuya"},{"family":"Mizutani","given":"Yasuyuki"},{"family":"Iida","given":"Tadashi"},{"family":"Shiratori","given":"Yoshimune"},{"family":"Oyama","given":"Shintaro"},{"family":"Koyama","given":"Junji"},{"family":"Mori","given":"Kensaku"},{"family":"Hayashi","given":"Yuichiro"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1111/den.14622","URL":"https://doi.org/10.1111/den.14622","source":"openalex"},{"id":"oa:W4395039716","type":"article-journal","title":"Artificial intelligence in chorioretinal pathology through fundoscopy: a comprehensive review","abstract":"BACKGROUND: Applications for artificial intelligence (AI) in ophthalmology are continually evolving. Fundoscopy is one of the oldest ocular imaging techniques but remains a mainstay in posterior segment imaging due to its prevalence, ease of use, and ongoing technological advancement. AI has been leveraged for fundoscopy to accomplish core tasks including segmentation, classification, and prediction. MAIN BODY: In this article we provide a review of AI in fundoscopy applied to representative chorioretinal pathologies, including diabetic retinopathy and age-related macular degeneration, among others. We conclude with a discussion of future directions and current limitations. SHORT CONCLUSION: As AI evolves, it will become increasingly essential for the modern ophthalmologist to understand its applications and limitations to improve patient outcomes and continue to innovate.","author":[{"family":"Driban","given":"Matthew"},{"family":"Yan","given":"Audrey"},{"family":"Selvam","given":"Amrish"},{"family":"Ong","given":"Joshua"},{"family":"Vupparaboina","given":"Kiran"},{"family":"Chhablani","given":"Jay"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s40942-024-00554-4","URL":"https://doi.org/10.1186/s40942-024-00554-4","source":"openalex"},{"id":"oa:W4402380931","type":"article-journal","title":"Explainable artificial intelligence for genotype-to-phenotype prediction in plant breeding: a case study with a dataset from an almond germplasm collection","abstract":"Background Advances in DNA sequencing revolutionized plant genomics and significantly contributed to the study of genetic diversity. However, predicting phenotypes from genomic data remains a challenge, particularly in the context of plant breeding. Despite significant progress, accurately predicting phenotypes from high-dimensional genomic data remains a challenge, particularly in identifying the key genetic factors influencing these predictions. This study aims to bridge this gap by integrating explainable artificial intelligence (XAI) techniques with advanced machine learning models. This approach is intended to enhance both the predictive accuracy and interpretability of genotype-to-phenotype models, thereby improving their reliability and supporting more informed breeding decisions. Results This study compares several ML methods for genotype-to-phenotype prediction, using data available from an almond germplasm collection. After preprocessing and feature selection, regression models are employed to predict almond shelling fraction. Best predictions were obtained by the Random Forest method (correlation = 0.727 ± 0.020, an R 2 = 0.511 ± 0.025, and an RMSE = 7.746 ± 0.199). Notably, the application of the SHAP (SHapley Additive exPlanations) values algorithm to explain the results highlighted several genomic regions associated with the trait, including one, having the highest feature importance, located in a gene potentially involved in seed development. Conclusions Employing explainable artificial intelligence algorithms enhances model interpretability, identifying genetic polymorphisms associated with the shelling percentage. These findings underscore XAI’s efficacy in predicting phenotypic traits from genomic data, highlighting its significance in optimizing crop production for sustainable agriculture.","author":[{"family":"Novielli","given":"Pierfrancesco"},{"family":"Romano","given":"Donato"},{"family":"Pavan","given":"Stefano"},{"family":"Losciale","given":"Pasquale"},{"family":"Stellacci","given":"Anna"},{"family":"Diacono","given":"Domenico"},{"family":"Bellotti","given":"R"},{"family":"Tangaro","given":"Sabina"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fpls.2024.1434229","URL":"https://doi.org/10.3389/fpls.2024.1434229","source":"openalex"},{"id":"oa:W4393949164","type":"article-journal","title":"Artificial Intelligence-Based Left Ventricular Ejection Fraction by Medical Students for Mortality and Readmission Prediction","abstract":"Introduction: Point-of-care ultrasound has become a universal practice, employed by physicians across various disciplines, contributing to diagnostic processes and decision-making. Aim: To assess the association of reduced (<50%) left-ventricular ejection fraction (LVEF) based on prospective point-of-care ultrasound operated by medical students using an artificial intelligence (AI) tool and 1-year primary composite outcome, including mortality and readmission for cardiovascular-related causes. Methods: Eight trained medical students used a hand-held ultrasound device (HUD) equipped with an AI-based tool for automatic evaluation of the LVEF of non-selected patients hospitalized in a cardiology department from March 2019 through March 2020. Results: The study included 82 patients (72 males aged 58.5 ± 16.8 years), of whom 34 (41.5%) were diagnosed with AI-based reduced LVEF. The rates of the composite outcome were higher among patients with reduced systolic function compared to those with preserved LVEF (41.2% vs. 16.7%, p = 0.014). Adjusting for pertinent variables, reduced LVEF independently predicted the composite outcome (HR 2.717, 95% CI 1.083–6.817, p = 0.033). As compared to those with LVEF ≥ 50%, patients with reduced LVEF had a longer length of stay and higher rates of the secondary composite outcome, including in-hospital death, advanced ventilatory support, shock, and acute decompensated heart failure. Conclusion: AI-based assessment of reduced systolic function in the hands of medical students, independently predicted 1-year mortality and cardiovascular-related readmission and was associated with unfavorable in-hospital outcomes. AI utilization by novice users may be an important tool for risk stratification for hospitalized patients.","author":[{"family":"Dadon","given":"Ziv"},{"family":"Rav","given":"Moshe"},{"family":"Orlev","given":"Amir"},{"family":"Carasso","given":"Shemy"},{"family":"Glikson","given":"Michael"},{"family":"Gottlieb","given":"Shmuel"},{"family":"Alpert","given":"Evan"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/diagnostics14070767","URL":"https://doi.org/10.3390/diagnostics14070767","source":"openalex"},{"id":"oa:W4399295665","type":"article-journal","title":"Diagnosis of ADHD using virtual reality and artificial intelligence: an exploratory study of clinical applications","abstract":"Introduction: Diagnosis of Attention Deficit/Hyperactivity Disorder (ADHD) is based on clinical evaluation of symptoms by a psychiatrist, referencing results of psychological tests. When diagnosing ADHD, the child's behavior and functionality in real-life situations are critical components. However, direct observation by a clinician is often not feasible in practice. Therefore, such information is typically gathered from primary caregivers or teachers, which can introduce subjective elements. To overcome these limitations, we developed AttnKare-D, an innovative digital diagnostic tool that could analyze children's behavioral data in Virtual Reality using Artificial Intelligence. The purpose of this study was to explore the utility and safety of AttnKare-D for clinical application. Method: A total of 21 children aged between 6 and 12 years were recruited for this study. Among them, 15 were children diagnosed with ADHD, 5 were part of a normal control group, and 1 child was excluded due to withdrawal of consent. Psychological assessments, including K-WISC, Conners CPT, K-ARS, and K-CBCL, were conducted for participants and their primary caregivers. Diagnoses of ADHD were confirmed by child and adolescent psychiatrists based on comprehensive face-to-face evaluations and results of psychological assessments. Participants underwent VR diagnostic assessment by performing various cognitive and behavioral tasks in a VR environment. Collected data were analyzed using an AI model to assess ADHD diagnosis and the severity of symptoms. Results: AttnKare-D demonstrated diagnostic performance with an AUC of 0.893 when compared to diagnoses made by child and adolescent psychiatrist, showing a sensitivity of 0.8 and a specificity of 1.0 at a cut-off score of 18.44. AttnKare-D scores showed a high correlation with K-ARS scores rated by parents and experts, although the correlation was relatively low for inattention scores. Conclusion: Results of this study suggest that AttnKare-D can be a useful tool for diagnosing ADHD in children. This approach has potential to overcome limitations of current diagnostic methods, enhancing the accuracy and objectivity of ADHD diagnoses. This study lays the groundwork for further improvement and research on diagnostic tools integrating VR and AI technologies. For future clinical applications, it is necessary to conduct clinical trials involving a sufficient number of participants to ensure reliable use.","author":[{"family":"Oh","given":"Soohwan"},{"family":"Joung","given":"Yoo‐sook"},{"family":"Chung","given":"Tai‐myoung"},{"family":"Lee","given":"Junho"},{"family":"Seok","given":"Bum"},{"family":"Kim","given":"Nam"},{"family":"Son","given":"Ha"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fpsyt.2024.1383547","URL":"https://doi.org/10.3389/fpsyt.2024.1383547","source":"openalex"},{"id":"oa:W4401332440","type":"article-journal","title":"Applications of Multimodal Artificial Intelligence in Non-Hodgkin Lymphoma B Cells","abstract":"Given advancements in large-scale data and AI, integrating multimodal artificial intelligence into cancer research can enhance our understanding of tumor behavior by simultaneously processing diverse biomedical data types. In this review, we explore the potential of multimodal AI in comprehending B-cell non-Hodgkin lymphomas (B-NHLs). B-cell non-Hodgkin lymphomas (B-NHLs) represent a particular challenge in oncology due to tumor heterogeneity and the intricate ecosystem in which tumors develop. These complexities complicate diagnosis, prognosis, and therapy response, emphasizing the need to use sophisticated approaches to enhance personalized treatment strategies for better patient outcomes. Therefore, multimodal AI can be leveraged to synthesize critical information from available biomedical data such as clinical record, imaging, pathology and omics data, to picture the whole tumor. In this review, we first define various types of modalities, multimodal AI frameworks, and several applications in precision medicine. Then, we provide several examples of its usage in B-NHLs, for analyzing the complexity of the ecosystem, identifying immune biomarkers, optimizing therapy strategy, and its clinical applications. Lastly, we address the limitations and future directions of multimodal AI, highlighting the need to overcome these challenges for better clinical practice and application in healthcare.","author":[{"family":"Isavand","given":"Pouria"},{"family":"Aghamiri","given":"Sara"},{"family":"Amin","given":"Rada"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/biomedicines12081753","URL":"https://doi.org/10.3390/biomedicines12081753","source":"openalex"},{"id":"oa:W4401880940","type":"article-journal","title":"Assessment of Generative Artificial Intelligence (AI) Models in Creating Medical Illustrations for Various Corneal Transplant Procedures","abstract":"PURPOSE: This study aimed to task and assess generative artificial intelligence (AI) models in creating medical illustrations for corneal transplant procedures such as Descemet's stripping automated endothelial keratoplasty (DSAEK), Descemet's membrane endothelial keratoplasty (DMEK), deep anterior lamellar keratoplasty (DALK), and penetrating keratoplasty (PKP). Methods: Six engineered prompts were provided to Decoder-Only Autoregressive Language and Image Synthesis 3 (DALL-E 3) and Medical Illustration Manager (MIM) to guide these generative AI models in creating a final medical illustration for each of the four corneal transplant procedures. Control illustrations were created by the authors for each transplant technique for comparison. A grading system with five categories with a maximum score of 3 points each (15 points total) was designed to objectively assess AI's performance. Four independent reviewers analyzed and scored the final images produced by DALL-E 3 and MIM as well as the control illustrations. All AI-generated images and control illustrations were then provided to Chat Generative Pre-Trained Transformer-4o (ChatGPT-4o), which was tasked with grading each image with the grading system described above. All results were then tabulated and graphically depicted. RESULTS: The control illustration images received significantly higher scores than produced images from DALL-E 3 and MIM in legibility, anatomical realism and accuracy, procedural step accuracy, and lack of fictitious anatomy (p<0.001). For detail and clarity, the control illustrations and images produced by DALL-E 3 and MIM received statistically similar scores of 2.75±0.29, 2.19±0.24, and 2.50±0.29, respectively (p=0.0504). With regard to mean cumulative scores for each transplant procedure image, the control illustrations received a significantly higher score than DALL-E 3 and MIM (p<0.001). Additionally, the overall mean cumulative score for the control illustrations was significantly higher than DALL-E 3 and MIM (14.56±0.51 (97.1%), 4.38±1.2 (29.2%), and 5.63±1.82 (37.5%), respectively (p<0.001)). When assessing AI's grading performance, ChatGPT-4o scored the images produced by DALL-E 3 and MIM significantly higher than the average scores of the independent reviewers (DALL-E 3: 10.0±0.0 (66.6%) vs. 4.38±1.20 (29.2%), p<0.001; MIM: 10.0±0.0 (66.6%) vs. 5.63±1.82 (37.5%), p<0.001). However, mean scores for the control illustrations between ChatGPT-4o and the independent reviewers were comparable (15.0±0.0 (100%) vs. 14.56±0.13 (97.1%); p>0.05). CONCLUSION: AI is an extremely powerful and efficient tool for many tasks, but it is currently limited in producing accurate medical illustrations for corneal transplant procedures. Further development is required for generative AI models to create medically sound and accurate illustrations for use in ophthalmology.","author":[{"family":"Moin","given":"Kayvon"},{"family":"Nasir","given":"Ayesha"},{"family":"Petroff","given":"Dallas"},{"family":"Loveless","given":"Bosten"},{"family":"Moshirfar","given":"Omeed"},{"family":"Hoopes","given":"Phillip"},{"family":"Moshirfar","given":"Majid"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.67833","URL":"https://doi.org/10.7759/cureus.67833","source":"openalex"},{"id":"oa:W4391317825","type":"article-journal","title":"Artificial Intelligence Model Predicts Sudden Cardiac Arrest Manifesting With Pulseless Electric Activity Versus Ventricular Fibrillation","abstract":"BACKGROUND: There is no specific treatment for sudden cardiac arrest (SCA) manifesting as pulseless electric activity (PEA) and survival rates are low; unlike ventricular fibrillation (VF), which is treatable by defibrillation. Development of novel treatments requires fundamental clinical studies, but access to the true initial rhythm has been a limiting factor. METHODS: Using demographics and detailed clinical variables, we trained and tested an AI model (extreme gradient boosting) to differentiate PEA-SCA versus VF-SCA in a novel setting that provided the true initial rhythm. A subgroup of SCAs are witnessed by emergency medical services personnel, and because the response time is zero, the true SCA initial rhythm is recorded. The internal cohort consisted of 421 emergency medical services-witnessed out-of-hospital SCAs with PEA or VF as the initial rhythm in the Portland, Oregon metropolitan area. External validation was performed in 220 emergency medical services-witnessed SCAs from Ventura, CA. RESULTS: In the internal cohort, the artificial intelligence model achieved an area under the receiver operating characteristic curve of 0.68 (95% CI, 0.61-0.76). Model performance was similar in the external cohort, achieving an area under the receiver operating characteristic curve of 0.72 (95% CI, 0.59-0.84). Anemia, older age, increased weight, and dyspnea as a warning symptom were the most important features of PEA-SCA; younger age, chest pain as a warning symptom and established coronary artery disease were important features associated with VF. CONCLUSIONS: The artificial intelligence model identified novel features of PEA-SCA, differentiated from VF-SCA and was successfully replicated in an external cohort. These findings enhance the mechanistic understanding of PEA-SCA with potential implications for developing novel management strategies.","author":[{"family":"Holmström","given":"Lauri"},{"family":"Bednarski","given":"Bryan"},{"family":"Chugh","given":"Harpriya"},{"family":"Aziz","given":"Habiba"},{"family":"Pham","given":"Hoang"},{"family":"Sargsyan","given":"Arayik"},{"family":"Uyevanado","given":"Audrey"},{"family":"Dey","given":"Damini"},{"family":"Salvucci","given":"Angelo"},{"family":"Jui","given":"Jonathan"},{"family":"Reinier","given":"Kyndaron"},{"family":"Slomka","given":"Piotr"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1161/circep.123.012338","URL":"https://doi.org/10.1161/circep.123.012338","source":"openalex"},{"id":"oa:W4401989560","type":"article-journal","title":"Artificial Intelligence-Powered Surgical Consent: Patient Insights","abstract":"Introduction The integration of artificial intelligence (AI) in healthcare has revolutionized patient interactions and service delivery. AI's role extends from supporting clinical diagnostics and enhancing operational efficiencies to potentially improving informed consent processes in surgical settings. This study investigates the application of AI, particularly large language models like OpenAI's ChatGPT, in facilitating surgical consent, focusing on patient understanding, satisfaction, and trust. Methods We employed a mixed-methods approach involving 86 participants, including laypeople and medical staff, who engaged in a simulated AI-driven consent process for a tonsillectomy. Participants interacted with ChatGPT-4, which provided detailed procedure explanations, risks, and benefits. Post-interaction, participants completed a survey assessing their experience through quantitative and qualitative measures. Results Participants had a cautiously optimistic response to AI in the surgical consent process. Notably, 71% felt adequately informed, 86% found the information clear, and 71% felt they could make informed decisions. Overall, 71% were satisfied, 57% felt respected and confident, and 57% would recommend it, indicating areas needing refinement. However, concerns about data privacy and the lack of personal interaction were significant, with only 42% reassured about the security of their data. The standardization of information provided by AI was appreciated for potentially reducing human error, but the absence of empathetic human interaction was noted as a drawback. Discussion While AI shows promise in enhancing the consistency and comprehensiveness of information delivered during the consent process, significant challenges remain. These include addressing data privacy concerns and bridging the gap in personal interaction. The potential for AI to misinform due to system \"hallucinations\" or inherent biases also needs consideration. Future research should focus on refining AI interactions to support more nuanced and empathetic engagements, ensuring that AI supplements rather than replacing human elements in healthcare. Conclusion The integration of AI into surgical consent processes could standardize and potentially improve the delivery of information but must be balanced with efforts to maintain the critical human elements of care. Collaborative efforts between developers, clinicians, and ethicists are essential to optimize AI use, ensuring it complements the traditional consent process while enhancing patient satisfaction and trust.","author":[{"family":"Teasdale","given":"Alex"},{"family":"Mills","given":"Laura"},{"family":"Costello","given":"Rhodri"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.68134","URL":"https://doi.org/10.7759/cureus.68134","source":"openalex"},{"id":"oa:W4390863783","type":"article-journal","title":"A pedagogical design for self-regulated learning in academic writing using text-based generative artificial intelligence tools: 6-P pedagogy of plan, prompt, preview, produce, peer-review, portfolio-tracking","abstract":"The emergence and popularity of generative artificial intelligence (AI) tools, particularly text-based ones known as large language models, pose both opportunities and challenges to education. The ability of these tools to generate human-like texts based on minimal instructions causes concerns among educators about students’ use of these tools for academic writing, which may constitute a breach of academic integrity. We propose a pedagogical design that models on self-regulated learning and the authoring cycle and develops students’ critical thinking and self-regulation when composing academic writing using text-based generative AI tools. It contains six iterative and interactive phases. Students first plan the content and structure of the writing, then generate prompts for text-based generative AI tools. Next, students preview and verify the tools’ output, followed by the fourth phase of producing the writing using the corrected output. Fifthly, peer review by fellow students may be required to polish and proofread the writing. Lastly, through portfolio-tracking, students reflect on the writing process, and formulate strategies for future usage of text-based generative AI tools for writing. This pedagogical design helps students and teachers embrace text-based generative AI while addressing the perils these tools present, and guides the development of education interventions and instruments.","author":[{"family":"Kong","given":"Siu"},{"family":"Lee","given":"John"},{"family":"Tsang","given":"Olson"}],"issued":{"date-parts":[[2024]]},"DOI":"10.58459/rptel.2024.19030","URL":"https://doi.org/10.58459/rptel.2024.19030","source":"openalex"},{"id":"oa:W4395075544","type":"article-journal","title":"Artificial intelligence and smile design: An e‐Delphi consensus statement of ethical challenges","abstract":"PURPOSE: Smile design software increasingly relies on artificial intelligence (AI). However, using AI for smile design raises numerous technical and ethical concerns. This study aimed to evaluate these ethical issues. METHODS: An international consortium of experts specialized in AI, dentistry, and smile design was engaged to emulate and assess the ethical challenges raised by the use of AI for smile design. An e-Delphi protocol was used to seek the agreement of the ITU-WHO group on well-established ethical principles regarding the use of AI (wellness, respect for autonomy, privacy protection, solidarity, governance, equity, diversity, expertise/prudence, accountability/responsibility, sustainability, and transparency). Each principle included examples of ethical challenges that users might encounter when using AI for smile design. RESULTS: On the first round of the e-Delphi exercise, participants agreed that seven items should be considered in smile design (diversity, transparency, wellness, privacy protection, prudence, law and governance, and sustainable development), but the remaining four items (equity, accountability and responsibility, solidarity, and respect of autonomy) were rejected and had to be reformulated. After a second round, participants agreed to all items that should be considered while using AI for smile design. CONCLUSIONS: AI development and deployment for smile design should abide by the ethical principles of wellness, respect for autonomy, privacy protection, solidarity, governance, equity, diversity, expertise/prudence, accountability/responsibility, sustainability, and transparency.","author":[{"family":"Rokhshad","given":"Rata"},{"family":"Karteva","given":"Teodora"},{"family":"Chaurasia","given":"Akhilanand"},{"family":"Richert","given":"Raphaël"},{"family":"Mörch","given":"Carl‐maria"},{"family":"Tamimi","given":"Faleh"},{"family":"Ducret","given":"Maxime"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/jopr.13858","URL":"https://doi.org/10.1111/jopr.13858","source":"openalex"},{"id":"oa:W4404794268","type":"article-journal","title":"The Barriers and Solution to Artificial Intelligence Adoption in Medical Education: A Qualitative Study","abstract":"Background: Nowadays, AI adoption, in medical education is growing rapidly. Studies have been conducted on various aspects of AI in medical education, however, in the context of its challenges and solutions faced with AI adoption is rarely explored through qualitative approaches. It is necessary to know the barriers and remedies for AI utilization in medical education.Objectives: This study explores potential barriers and solutions to AI adoption in medical education. Materials and Methods: Based on Giorgi's phenomenological approach, this qualitative study explored an in-depth understanding of AI adoption in medical education through in-depth interviews from March to April 2024. By purposive sampling, sixteen participants across famous medical institutions of Pakistan were interviewed, and data collection was stopped upon saturation. Results: A total of 16 participants belonging to different cadres of undergraduate medical colleges of Pakistan were interviewed for the current study. A total of 219 quotations from the transcripts resulted in three main themes with subsequent subthemes (total 13), i.e. perceptions of AI in medical education with three subthemes, barriers to AI adoption with seven subthemes and conjoint solutions to barriers having three subthemes.Conclusion: AI adoption in medical education is evolving healthcare globally. Nevertheless, there are barriers like the complexity of AI, lack of technical skills for using AI, and scarcity of resources that hinder its utilization in medical education. Moreover, ethical concerns and the specific guidelines that attract investors are among other challenges. Countermeasures like improving technical infrastructure and faculty development initiatives through collaboration among policymakers, administrators, and teaching faculty can overcome the barriers. Keywords: Artificial Intelligence, Adoption, Challenges, Medical Education.","author":[{"family":"Khan","given":"Muhammad"},{"family":"Lajber","given":"Mehreen"},{"family":"Bilal","given":"Nazish"},{"family":"Khan","given":"Sana"},{"family":"Siddiqi","given":"Zaibunnisa"},{"family":"Ahmad","given":"Aziz"}],"issued":{"date-parts":[[2024]]},"DOI":"10.52206/jsmc.2024.14.4.957","URL":"https://doi.org/10.52206/jsmc.2024.14.4.957","source":"openalex"},{"id":"oa:W4391170104","type":"article-journal","title":"A Systematic Review on Artificial Intelligence Evaluating Metastatic Prostatic Cancer and Lymph Nodes on PSMA PET Scans","abstract":"Early detection of metastatic prostate cancer (mPCa) is crucial. Whilst the prostate-specific membrane antigen (PSMA) PET scan has high diagnostic accuracy, it suffers from inter-reader variability, and the time-consuming reporting process. This systematic review was registered on PROSPERO (ID CRD42023456044) and aims to evaluate AI's ability to enhance reporting, diagnostics, and predictive capabilities for mPCa on PSMA PET scans. Inclusion criteria covered studies using AI to evaluate mPCa on PSMA PET, excluding non-PSMA tracers. A search was conducted on Medline, Embase, and Scopus from inception to July 2023. After screening 249 studies, 11 remained eligible for inclusion. Due to the heterogeneity of studies, meta-analysis was precluded. The prediction model risk of bias assessment tool (PROBAST) indicated a low overall risk of bias in ten studies, though only one incorporated clinical parameters (such as age, and Gleason score). AI demonstrated a high accuracy (98%) in identifying lymph node involvement and metastatic disease, albeit with sensitivity variation (62-97%). Advantages included distinguishing bone lesions, estimating tumour burden, predicting treatment response, and automating tasks accurately. In conclusion, AI showcases promising capabilities in enhancing the diagnostic potential of PSMA PET scans for mPCa, addressing current limitations in efficiency and variability.","author":[{"family":"Liu","given":"Jianliang"},{"family":"Cundy","given":"Thomas"},{"family":"Woon","given":"Dixon"},{"family":"Lawrentschuk","given":"Nathan"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/cancers16030486","URL":"https://doi.org/10.3390/cancers16030486","source":"openalex"},{"id":"oa:W4394870730","type":"article-journal","title":"Enhanced control of periodontitis by an artificial intelligence‐enabled multimodal‐sensing toothbrush and targeted mHealth micromessages: A randomized trial","abstract":"AIM: Treatment of periodontitis, a chronic inflammatory disease driven by biofilm dysbiosis, remains challenging due to patients' poor performance and adherence to the necessary oral hygiene procedures. Novel, artificial intelligence-enabled multimodal-sensing toothbrushes (AI-MST) can guide patients' oral hygiene practices in real-time and transmit valuable data to clinicians, thus enabling effective remote monitoring and guidance. The aim of this trial was to assess the effect of such a system as an adjunct to clinical practice guideline-conform treatment. MATERIALS AND METHODS: This was a single-centre, double-blind, standard-of-care controlled, randomized, parallel-group, superiority trial. Male and female adults with generalized Stage II/III periodontitis were recruited at the Shanghai Ninth People's Hospital, China. Subjects received a standard-of-care oral hygiene regimen or a technology-enabled, theory-based digital intervention consisting of an AI-MST and targeted doctor's guidance by remote micromessaging. Additionally, both groups received guideline-conform periodontal treatment. The primary outcome was the resolution of inflamed periodontal pockets (≥4 mm with bleeding on probing) at 6 months. The intention-to-treat (ITT) analysis included all subjects who received the allocated treatment and at least one follow-up. RESULTS: One hundred patients were randomized and treated (50 tests/controls) between 1 February and 30 November 2022. Forty-eight tests (19 females) and 47 controls (16 females) were analysed in the ITT population. At 6 months, the proportion of inflamed periodontal pockets decreased from 80.7% (95% confidence interval [CI] 76.5-84.8) to 52.3% (47.7-57.0) in the control group, and from 81.4% (77.1-85.6) to 44.4% (39.9-48.9) in the test group. The inter-group difference was 7.9% (1.6-14.6, p < .05). Test subjects achieved better levels of oral hygiene (p < .001). No significant adverse events were observed. CONCLUSIONS: The tested digital health intervention significantly improved the outcome of periodontal therapy by enhancing the adherence and performance of self-performed oral hygiene. The model breaks the traditional model of oral health care and has the potential to improve efficiency and reduce costs (NCT05137392).","author":[{"family":"Li","given":"Yuan"},{"family":"Wu","given":"Xinyu"},{"family":"Liu","given":"MQ"},{"family":"Deng","given":"Ke"},{"family":"Tullini","given":"Annamaria"},{"family":"Zhang","given":"Xiao"},{"family":"Shi","given":"Junyu"},{"family":"Lai","given":"Hongchang"},{"family":"Tonetti","given":"Maurizio"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/jcpe.13987","URL":"https://doi.org/10.1111/jcpe.13987","source":"openalex"},{"id":"oa:W4399323408","type":"article-journal","title":"Buzzing with Intelligence: Current Issues in Apiculture and the Role of Artificial Intelligence (AI) to Tackle It","abstract":"L.) are important for agriculture and ecosystems; however, they are threatened by the changing climate. In order to adapt and respond to emerging difficulties, beekeepers require the ability to continuously monitor their beehives. To carry out this, the utilization of advanced machine learning techniques proves to be an exceptional tool. This review provides a comprehensive analysis of the available research on the different applications of artificial intelligence (AI) in beekeeping that are relevant to climate change. Presented studies have shown that AI can be used in various scientific aspects of beekeeping and can work with several data types (e.g., sound, sensor readings, images) to investigate, model, predict, and help make decisions in apiaries. Research articles related to various aspects of apiculture, e.g., managing hives, maintaining their health, detecting pests and diseases, and climate and habitat management, were analyzed. It was found that several environmental, behavioral, and physical attributes needed to be monitored in real-time to be able to understand and fully predict the state of the hives. Finally, it could be concluded that even if there is not yet a full-scale monitoring method for apiculture, the already available approaches (even with their identified shortcomings) can help maintain sustainability in the changing apiculture.","author":[{"family":"Astuti","given":"Putri"},{"family":"Hegedűs","given":"Bettina"},{"family":"Oleksa","given":"Andrzej"},{"family":"Bagi","given":"Zoltán"},{"family":"Kusza","given":"Szilvia"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/insects15060418","URL":"https://doi.org/10.3390/insects15060418","source":"openalex"},{"id":"oa:W4404701448","type":"article-journal","title":"Generative AI in Improving Personalized Patient Care Plans: Opportunities and Barriers Towards Its Wider Adoption","abstract":"The main aim of this study is to investigate the opportunities, challenges, and barriers in implementing generative artificial intelligence (Gen AI) in personalized patient care plans (PPCPs). This systematic review paper provides a comprehensive analysis of the current state, potential applications, and opportunities of Gen AI in patient care settings. This review aims to serve as a key resource for various stakeholders such as researchers, medical professionals, and data governance. We adopted the PRISMA review methodology and screened a total of 247 articles. After considering the eligibility and selection criteria, we selected 13 articles published between 2021 and 2024 (inclusive). The selection criteria were based on the inclusion of studies that report on the opportunities and challenges in improving PPCPs using Gen AI. We found that a holistic approach is required involving strategy, communications, integrations, and collaboration between AI developers, healthcare professionals, regulatory bodies, and patients. Developing frameworks that prioritize ethical considerations, patient privacy, and model transparency is crucial for the responsible deployment of Gen AI in healthcare. Balancing these opportunities and challenges requires collaboration between wider stakeholders to create a robust framework that maximizes the benefits of Gen AI in healthcare while addressing the key challenges and barriers such as explainability of the models, validation, regulation, and privacy integration with the existing clinical workflows.","author":[{"family":"Baig","given":"Mirza"},{"family":"Hobson","given":"Chris"},{"family":"Gholamhosseini","given":"Hamid"},{"family":"Ullah","given":"Ehsan"},{"family":"Afifi","given":"Shereen"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/app142310899","URL":"https://doi.org/10.3390/app142310899","source":"openalex"},{"id":"oa:W4402446266","type":"article-journal","title":"Artificial intelligence for automatic and objective assessment of competencies in flexible bronchoscopy","abstract":"Background: Bronchoscopy is a challenging technical procedure, and assessment of competence currently relies on expert raters. Human rating is time consuming and prone to rater bias. The aim of this study was to evaluate if a bronchial segment identification system based on artificial intelligence (AI) could automatically, instantly, and objectively assess competencies in flexible bronchoscopy in a valid way. Methods: September 2023. The participants performed one full diagnostic bronchoscopy in a simulated setting and were rated immediately by the AI according to its four outcome measures: diagnostic completeness (DC), structured progress (SP), procedure time (PT), and mean intersegmental time (MIT). The procedures were video-recorded and rated after the conference by two blinded, expert raters using a previously validated assessment tool with nine items regarding anatomy and dexterity. Results: Fifty-two participants from six different continents were included. All four outcome measures of the AI correlated significantly with the experts' anatomy-ratings (Pearson's correlation coefficient, P value): DC (r=0.47, P<0.001), SP (r=0.57, P<0.001), PT (r=-0.32, P=0.02), and MIT (r=-0.55, P<0.001) and also with the experts' dexterity-ratings: DC (r=0.38, P=0.006), SP (r=0.53, P<0.001), PT (r=-0.34, P=0.014), and MIT (r=-0.47, P<0.001). Conclusions: The study provides initial validity evidence for AI-based immediate and automatic assessment of anatomical and navigational competencies in flexible bronchoscopy. SP provided stronger correlations with human experts' ratings than the traditional DC.","author":[{"family":"Cold","given":"Kristoffer"},{"family":"Agbontaen","given":"Kaladerhan"},{"family":"Nielsen","given":"Anne"},{"family":"Andersen","given":"Christian"},{"family":"Singh","given":"Suveer"},{"family":"Konge","given":"Lars"}],"issued":{"date-parts":[[2024]]},"DOI":"10.21037/jtd-24-841","URL":"https://doi.org/10.21037/jtd-24-841","source":"openalex"},{"id":"oa:W4403637939","type":"article-journal","title":"Gender and Ethnicity Bias of Text-to-Image Generative Artificial Intelligence in Medical Imaging, Part 1: Preliminary Evaluation","abstract":"Generative artificial intelligence (AI) text-to-image production could reinforce or amplify gender and ethnicity biases. Several text-to-image generative AI tools are used for producing images that represent the medical imaging professions. White male stereotyping and masculine cultures can dissuade women and ethnically divergent people from being drawn into a profession. Methods: In March 2024, DALL-E 3, Firefly 2, Stable Diffusion 2.1, and Midjourney 5.2 were utilized to generate a series of individual and group images of medical imaging professionals: radiologist, nuclear medicine physician, radiographer, and nuclear medicine technologist. Multiple iterations of images were generated using a variety of prompts. Collectively, 184 images were produced for evaluation of 391 characters. All images were independently analyzed by 3 reviewers for apparent gender and skin tone. Results: Collectively (individual and group characters) (n = 391), 60.6% were male and 87.7% were of a light skin tone. DALL-E 3 (65.6%), Midjourney 5.2 (76.7%), and Stable Diffusion 2.1 (56.2%) had a statistically higher representation of men than Firefly 2 (42.9%) (P < 0.0001). With Firefly 2, 70.3% of characters had light skin tones, which was statistically lower (P < 0.0001) than for Stable Diffusion 2.1 (84.8%), Midjourney 5.2 (100%), and DALL-E 3 (94.8%). Overall, image quality metrics were average or better in 87.2% for DALL-E 3 and 86.2% for Midjourney 5.2, whereas 50.9% were inadequate or poor for Firefly 2 and 86.0% for Stable Diffusion 2.1. Conclusion: Generative AI text-to-image generation using DALL-E 3 via GPT-4 has the best overall quality compared with Firefly 2, Midjourney 5.2, and Stable Diffusion 2.1. Nonetheless, DALL-E 3 includes inherent biases associated with gender and ethnicity that demand more critical evaluation.","author":[{"family":"Currie","given":"Geoffrey"},{"family":"Hewis","given":"Johnathan"},{"family":"Hawk","given":"Elizabeth"},{"family":"Rohren","given":"Eric"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2967/jnmt.124.268332","URL":"https://doi.org/10.2967/jnmt.124.268332","source":"openalex"},{"id":"oa:W4399075444","type":"article-journal","title":"Liability of Health Professionals Using Sensors, Telemedicine and Artificial Intelligence for Remote Healthcare","abstract":"In the last few decades, there has been an ongoing transformation of our healthcare system with larger use of sensors for remote care and artificial intelligence (AI) tools. In particular, sensors improved by new algorithms with learning capabilities have proven their value for better patient care. Sensors and AI systems are no longer only non-autonomous devices such as the ones used in radiology or surgical robots; there are novel tools with a certain degree of autonomy aiming to largely modulate the medical decision. Thus, there will be situations in which the doctor is the one making the decision and has the final say and other cases in which the doctor might only apply the decision presented by the autonomous device. As those are two hugely different situations, they should not be treated the same way, and different liability rules should apply. Despite a real interest in the promise of sensors and AI in medicine, doctors and patients are reluctant to use it. One important reason is a lack clear definition of liability. Nobody wants to be at fault, or even prosecuted, because they followed the advice from an AI system, notably when it has not been perfectly adapted to a specific patient. Fears are present even with simple sensors and AI use, such as during telemedicine visits based on very useful, clinically pertinent sensors; with the risk of missing an important parameter; and, of course, when AI appears \"intelligent\", potentially replacing the doctors' judgment. This paper aims to provide an overview of the liability of the health professional in the context of the use of sensors and AI tools in remote healthcare, analyzing four regimes: the contract-based approach, the approach based on breach of duty to inform, the fault-based approach, and the approach related to the good itself. We will also discuss future challenges and opportunities in the promising domain of sensors and AI use in medicine.","author":[{"family":"Geny","given":"Marie"},{"family":"Andrès","given":"Emmanuel"},{"family":"Talha","given":"Samy"},{"family":"Gény","given":"Bernard"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/s24113491","URL":"https://doi.org/10.3390/s24113491","source":"openalex"},{"id":"oa:W4395447233","type":"article-journal","title":"Improving traumatic fracture detection on radiographs with artificial intelligence support: a multi-reader study","abstract":"Objectives: The aim of this study was to evaluate the diagnostic performance of nonspecialist readers with and without the use of an artificial intelligence (AI) support tool to detect traumatic fractures on radiographs of the appendicular skeleton. Methods: The design was a retrospective, fully crossed multi-reader, multi-case study on a balanced dataset of patients (≥2 years of age) with an AI tool as a diagnostic intervention. Fifteen readers assessed 340 radiographic exams, with and without the AI tool in 2 different sessions and the time spent was automatically recorded. Reference standard was established by 3 consultant radiologists. Sensitivity, specificity, and false positives per patient were calculated. Results: < .05) in exams aided by the AI tool compared to the unaided exams. The increase in sensitivity resulted in a relative reduction of missed fractures of 29%. The average rate of false positives per patient decreased from 0.16 to 0.14, corresponding to a relative reduction of 21%. There was no significant difference in average reading time spent per exam. The largest gain in fracture detection performance, with AI support, across all readers, was on nonobvious fractures with a significant increase in sensitivity of 11 percentage points (pp) (60%-71%). Conclusions: The diagnostic performance for detection of traumatic fractures on radiographs of the appendicular skeleton improved among nonspecialist readers tested AI fracture detection support tool showed an overall reader improvement in sensitivity and specificity when supported by an AI tool. Improvement was seen in both sensitivity and specificity without negatively affecting the interpretation time. Advances in knowledge: The division and analysis of obvious and nonobvious fractures are novel in AI reader comparison studies like this.","author":[{"family":"Bachmann","given":"Rikke"},{"family":"Gunes","given":"Gozde"},{"family":"Hangaard","given":"Stine"},{"family":"Nexmann","given":"Andreas"},{"family":"Lisouski","given":"Pavel"},{"family":"Boesen","given":"Mikael"},{"family":"Lundemann","given":"Michael"},{"family":"Baginski","given":"Scott"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1093/bjro/tzae011","URL":"https://doi.org/10.1093/bjro/tzae011","source":"openalex"},{"id":"oa:W4398742105","type":"article-journal","title":"Artificial Intelligence in Emergency Trauma Care: A Preliminary Scoping Review","abstract":"Abstract: This study aimed to analyze the use of generative artificial intelligence in the emergency trauma care setting through a brief scoping review of literature published between 2014 and 2024. An exploration of the NCBI repository was performed using a search string of selected keywords that returned N=87 results; articles that met the inclusion criteria (n=28) were reviewed and analyzed. Heterogeneity sources were explored and identified by a significance threshold of P < 0.10 or an I 2 value exceeding 50%. If applicable, articles were categorized within three primary domains: triage, diagnostics, or treatment. Findings suggest that CNNs demonstrate strong diagnostic performance for diverse traumatic injuries, but generalized integration requires expanded prospective multi-center validation. Injury scoring models currently experience calibration gaps in mortality quantification and lesion localization that can undermine clinical utility by permitting false negatives. Triage predictive models now confront transparency, explainability, and healthcare ecosystem integration barriers limiting real-world translation. The most significant literature gap centers on treatment-oriented generative AI applications that provide real-time guidance for urgent trauma interventions rather than just analytical support. Keywords: artificial intelligence, machine-learning, emergency medicine, traumatology","author":[{"family":"Ventura","given":"Christian"},{"family":"Denton","given":"Edward"},{"family":"David","given":"Jessica"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2147/mder.s467146","URL":"https://doi.org/10.2147/mder.s467146","source":"openalex"},{"id":"oa:W4403037144","type":"article-journal","title":"Educational innovation: Exploring the Potential of Generative Artificial Intelligence in cognitive schema building","abstract":"This study explores the use of generative artificial intelligence to enhance teaching and learning experience, focusing on strengthening and consolidating cognitive schemas. Research reveals that schemas can profoundly influence the improvement of the learning experience and promote the assimilation of new types of information and retention in students' memory. To improve the teaching and learning experience, the advantages, obstacles, and potential future trajectories of utilizing these technologies were examined by conducting a thorough literature review and analyzing relevant studies. Findings indicate that generative artificial intelligence has the potential to personalize learning, diversify educational content, and improve teaching efficiency and scalability. However, it also poses challenges related to content quality, data privacy, and equity in access to personalized learning. Future research should focus on the effectiveness of educational tools based on generative AI that promote equity and inclusion, ethical approaches, and interdisciplinary collaboration. Overall, this study provides a solid foundation for understanding and harnessing the potential of generative artificial intelligence in enhancing cognitive schemas, thereby promoting more effective, inclusive, and personalized education.","author":[{"family":"Granda","given":"Bernarda"},{"family":"Inzhivotkina","given":"Yana"},{"family":"Apolo","given":"María"},{"family":"Fajardo","given":"Jorge"}],"issued":{"date-parts":[[2024]]},"DOI":"10.21556/edutec.2024.89.3251","URL":"https://doi.org/10.21556/edutec.2024.89.3251","source":"openalex"},{"id":"oa:W4400289499","type":"article-journal","title":"Artificial intelligence innovations in neurosurgical oncology: a narrative review","abstract":"PURPOSE: Artificial Intelligence (AI) has become increasingly integrated clinically within neurosurgical oncology. This report reviews the cutting-edge technologies impacting tumor treatment and outcomes. METHODS: A rigorous literature search was performed with the aid of a research librarian to identify key articles referencing AI and related topics (machine learning (ML), computer vision (CV), augmented reality (AR), virtual reality (VR), etc.) for neurosurgical care of brain or spinal tumors. RESULTS: Treatment of central nervous system (CNS) tumors is being improved through advances across AI-such as AL, CV, and AR/VR. AI aided diagnostic and prognostication tools can influence pre-operative patient experience, while automated tumor segmentation and total resection predictions aid surgical planning. Novel intra-operative tools can rapidly provide histopathologic tumor classification to streamline treatment strategies. Post-operative video analysis, paired with rich surgical simulations, can enhance training feedback and regimens. CONCLUSION: While limited generalizability, bias, and patient data security are current concerns, the advent of federated learning, along with growing data consortiums, provides an avenue for increasingly safe, powerful, and effective AI platforms in the future.","author":[{"family":"Baker","given":"Clayton"},{"family":"Pease","given":"Matthew"},{"family":"Sexton","given":"Daniel"},{"family":"Abumoussa","given":"Andrew"},{"family":"Chambless","given":"Lola"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s11060-024-04757-5","URL":"https://doi.org/10.1007/s11060-024-04757-5","source":"openalex"},{"id":"oa:W4392632345","type":"article-journal","title":"Using ChatGPT-4 to Create Structured Medical Notes From Audio Recordings of Physician-Patient Encounters: Comparative Study","abstract":"BACKGROUND: Medical documentation plays a crucial role in clinical practice, facilitating accurate patient management and communication among health care professionals. However, inaccuracies in medical notes can lead to miscommunication and diagnostic errors. Additionally, the demands of documentation contribute to physician burnout. Although intermediaries like medical scribes and speech recognition software have been used to ease this burden, they have limitations in terms of accuracy and addressing provider-specific metrics. The integration of ambient artificial intelligence (AI)-powered solutions offers a promising way to improve documentation while fitting seamlessly into existing workflows. OBJECTIVE: This study aims to assess the accuracy and quality of Subjective, Objective, Assessment, and Plan (SOAP) notes generated by ChatGPT-4, an AI model, using established transcripts of History and Physical Examination as the gold standard. We seek to identify potential errors and evaluate the model's performance across different categories. METHODS: We conducted simulated patient-provider encounters representing various ambulatory specialties and transcribed the audio files. Key reportable elements were identified, and ChatGPT-4 was used to generate SOAP notes based on these transcripts. Three versions of each note were created and compared to the gold standard via chart review; errors generated from the comparison were categorized as omissions, incorrect information, or additions. We compared the accuracy of data elements across versions, transcript length, and data categories. Additionally, we assessed note quality using the Physician Documentation Quality Instrument (PDQI) scoring system. RESULTS: Although ChatGPT-4 consistently generated SOAP-style notes, there were, on average, 23.6 errors per clinical case, with errors of omission (86%) being the most common, followed by addition errors (10.5%) and inclusion of incorrect facts (3.2%). There was significant variance between replicates of the same case, with only 52.9% of data elements reported correctly across all 3 replicates. The accuracy of data elements varied across cases, with the highest accuracy observed in the \"Objective\" section. Consequently, the measure of note quality, assessed by PDQI, demonstrated intra- and intercase variance. Finally, the accuracy of ChatGPT-4 was inversely correlated to both the transcript length (P=.05) and the number of scorable data elements (P=.05). CONCLUSIONS: Our study reveals substantial variability in errors, accuracy, and note quality generated by ChatGPT-4. Errors were not limited to specific sections, and the inconsistency in error types across replicates complicated predictability. Transcript length and data complexity were inversely correlated with note accuracy, raising concerns about the model's effectiveness in handling complex medical cases. The quality and reliability of clinical notes produced by ChatGPT-4 do not meet the standards required for clinical use. Although AI holds promise in health care, caution should be exercised before widespread adoption. Further research is needed to address accuracy, variability, and potential errors. ChatGPT-4, while valuable in various applications, should not be considered a safe alternative to human-generated clinical documentation at this time.","author":[{"family":"Kernberg","given":"Annessa"},{"family":"Gold","given":"Jeffrey"},{"family":"Mohan","given":"Vishnu"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/54419","URL":"https://doi.org/10.2196/54419","source":"openalex"},{"id":"oa:W4388656590","type":"article-journal","title":"Assessing the advancement of artificial intelligence and drones’ integration in agriculture through a bibliometric study","abstract":"Integrating artificial intelligence (AI) with drones has emerged as a promising paradigm for advancing agriculture. This bibliometric analysis investigates the current state of research in this transformative domain by comprehensively reviewing 234 pertinent articles from Scopus and Web of Science databases. The problem involves harnessing AI-driven drones' potential to address agricultural challenges effectively. To address this, we conducted a bibliometric review, looking at critical components, such as prominent journals, co-authorship patterns across countries, highly cited articles, and the co-citation network of keywords. Our findings underscore a growing interest in using AI-integrated drones to revolutionize various agricultural practices. Noteworthy applications include crop monitoring, precision agriculture, and environmental sensing, indicative of the field’s transformative capacity. This pioneering bibliometric study presents a comprehensive synthesis of the dynamic research landscape, signifying the first extensive exploration of AI and drones in agriculture. The identified knowledge gaps point to future research opportunities, fostering the adoption and implementation of these technologies for sustainable farming practices and resource optimization. Our analysis provides essential insights for researchers and practitioners, laying the groundwork for steering agricultural advancements toward an enhanced efficiency and innovation era.","author":[{"family":"Slimani","given":"Hicham"},{"family":"Mhamdi","given":"Jamal"},{"family":"Jilbab","given":"Abdelilah"}],"issued":{"date-parts":[[2023]]},"DOI":"10.11591/ijece.v14i1.pp878-890","URL":"https://doi.org/10.11591/ijece.v14i1.pp878-890","source":"openalex"},{"id":"oa:W4403483525","type":"article-journal","title":"Digital Technologies Impact on Healthcare Delivery: A Systematic Review of Artificial Intelligence (AI) and Machine-Learning (ML) Adoption, Challenges, and Opportunities","abstract":"Recent significant advances in the healthcare industry due to artificial intelligence (AI) and machine learning (ML) have been shown to revolutionize healthcare delivery by improving efficiency, accuracy, and patient outcomes. However, these technologies can face significant challenges and ethical considerations. This systematic review aimed to gather and synthesize the current knowledge on the impact of AI and ML adoption in healthcare delivery, with its associated challenges and opportunities. This study adhered to the PRISMA guidelines. Articles from 2014 to 2024 were selected from various databases using specific keywords. Eligible studies were included after rigorous screening and quality assessment using checklist tools. Themes were identified through data analysis and thematic analysis. From 4981 articles screened, a data synthesis of nine eligible studies revealed themes, including productivity enhancement, improved patient care through decision support and precision medicine, legal and policy challenges, technological considerations, organizational and managerial aspects, ethical concerns, data challenges, and socioeconomic implications. There exist significant opportunities, as well as substantial challenges and ethical concerns, associated with integrating AI and ML into healthcare delivery. Implementation strategies must be carefully designed, considering technical, ethical, and social factors.","author":[{"family":"Okwor","given":"Ifeanyi"},{"family":"Hitch","given":"Geeta"},{"family":"Hakkim","given":"Saira"},{"family":"Akbar","given":"SKFA"},{"family":"Sookhoo","given":"Dave"},{"family":"Kainesie","given":"John"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/ai5040095","URL":"https://doi.org/10.3390/ai5040095","source":"openalex"},{"id":"oa:W4396938053","type":"article-journal","title":"Explainable Artificial Intelligence in Quantifying Breast Cancer Factors: Saudi Arabia Context","abstract":"Breast cancer represents a significant health concern, particularly in Saudi Arabia, where it ranks as the most prevalent cancer type among women. This study focuses on leveraging eXplainable Artificial Intelligence (XAI) techniques to predict benign and malignant breast cancer cases using various clinical and pathological features specific to Saudi Arabian patients. Six distinct models were trained and evaluated based on common performance metrics such as accuracy, precision, recall, F1 score, and AUC-ROC score. To enhance interpretability, Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) were applied. The analysis identified the Random Forest model as the top performer, achieving an accuracy of 0.72, along with robust precision, recall, F1 score, and AUC-ROC score values. Conversely, the Support Vector Machine model exhibited the poorest performance metrics, indicating its limited predictive capability. Notably, the XAI approaches unveiled variations in the feature importance rankings across models, underscoring the need for further investigation. These findings offer valuable insights into breast cancer diagnosis and machine learning interpretation, aiding healthcare providers in understanding and potentially integrating such technologies into clinical practices.","author":[{"family":"Alelyani","given":"Turki"},{"family":"Alshammari","given":"Maha"},{"family":"Almuhanna","given":"Afnan"},{"family":"Asan","given":"Onur"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/healthcare12101025","URL":"https://doi.org/10.3390/healthcare12101025","source":"openalex"},{"id":"oa:W4400823983","type":"article-journal","title":"Exploring the Role of Artificial Intelligence and Machine Learning in Pharmaceutical Formulation Design","abstract":"The integration of Artificial Intelligence (AI) and Machine Learning (ML) into pharmaceutical formulation design has brought about a significant transformation, opening up new avenues for innovation and operational efficiency. This review paper aims to extensively examine the utilization of AI and ML in pharmaceutical formulation development, consolidating recent empirical findings and emerging patterns. Meta-analyses examining AI-driven drug discovery and formulation design efforts have revealed promising outcomes, including the acceleration of drug development timelines and enhancements in success rates across preclinical and clinical trials. Notably, a meta-analysis featured in Nature Reviews Drug Discovery sheds light on the pivotal role of AI in rational drug design, resulting in the identification of novel therapeutic candidates boasting improved efficacy and diminished side effects. Furthermore, AI and ML techniques are increasingly being deployed to optimize drug delivery systems, with studies showcasing their effectiveness in devising controlled-release formulations and nano-scale delivery platforms. For instance, the research highlighted in Advanced Drug Delivery Reviews demonstrates the application of ML algorithms in predicting the physicochemical attributes of nanoparticles, thereby aiding in the development of more durable and efficient drug carriers. Despite these advancements, challenges persist, including data scarcity, regulatory complexities, and ethical considerations. Nevertheless, ongoing endeavors to tackle these obstacles coupled with the continual evolution of AI and ML technologies offer promising prospects for the future of pharmaceutical formulation design. In conclusion, this review underscores the transformative influence of AI and ML on pharmaceutical formulation development, underscoring the necessity for sustained research and collaboration to fully leverage these technologies in enhancing healthcare outcomes.","author":[{"family":"Dey","given":"Hrithik"},{"family":"Arya","given":"Nisha"},{"family":"Mathur","given":"Harshita"},{"family":"Chatterjee","given":"Neel"},{"family":"Jadon","given":"Ruchi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.61554/ijnrph.v2i1.2024.67","URL":"https://doi.org/10.61554/ijnrph.v2i1.2024.67","source":"openalex"},{"id":"oa:W4395005519","type":"article-journal","title":"The Potential of Artificial Intelligence in Prosthodontics: A Comprehensive Review","abstract":"Prosthodontics is a dental subspecialty that includes the preparation of dental prosthetics for missing or damaged teeth. It increasingly uses computer-assisted technologies for planning and preparing dental prosthetics. This study aims to present the findings from a systematic review of publications on artificial intelligence (AI) in prosthodontics to identify current trends and future opportunities. The review question was \"What are the applications of AI in prosthodontics and how good is their performance in prosthodontics?\" Electronic searching in the Web of Science, ScienceDirect, PubMed, and Cochrane Library was conducted. The search was limited to full text from January 2012 to January 2024. Quadas-2 was used for assessing quality and potential risk of bias for the selected studies. A total of 1925 studies were identified in the initial search. After removing the duplicates and applying exclusion criteria, a total of 30 studies were selected for this review. Results of the Quadas-2 assessment of included studies found that a total of 18.3% of studies were identified as low risk of bias studies, whereas 52.6% and 28.9% of included studies were identified as studies with high and unclear risk of bias, respectively. Although they are still developing, AI models have already shown promise in the areas of dental charting, tooth shade selection, automated restoration design, mapping the preparation finishing line, manufacturing casting optimization, predicting facial changes in patients wearing removable prostheses, and designing removable partial dentures.","author":[{"family":"Aljulayfi","given":"Ibrahim"},{"family":"Almatrafi","given":"Ali"},{"family":"Althubaitiy","given":"Ramzi"},{"family":"Alnafisah","given":"Fahad"},{"family":"Alshehri","given":"Khalid"},{"family":"Alzahrani","given":"Bandar"},{"family":"Gufran","given":"Khalid"}],"issued":{"date-parts":[[2024]]},"DOI":"10.12659/msm.944310","URL":"https://doi.org/10.12659/msm.944310","source":"openalex"},{"id":"oa:W4404127725","type":"article-journal","title":"Robustness in deep learning models for medical diagnostics: security and adversarial challenges towards robust AI applications","abstract":"The current study investigates the robustness of deep learning models for accurate medical diagnosis systems with a specific focus on their ability to maintain performance in the presence of adversarial or noisy inputs. We examine factors that may influence model reliability, including model complexity, training data quality, and hyperparameters; we also examine security concerns related to adversarial attacks that aim to deceive models along with privacy attacks that seek to extract sensitive information. Researchers have discussed various defenses to these attacks to enhance model robustness, such as adversarial training and input preprocessing, along with mechanisms like data augmentation and uncertainty estimation. Tools and packages that extend the reliability features of deep learning frameworks such as TensorFlow and PyTorch are also being explored and evaluated. Existing evaluation metrics for robustness are additionally being discussed and evaluated. This paper concludes by discussing limitations in the existing literature and possible future research directions to continue enhancing the status of this research topic, particularly in the medical domain, with the aim of ensuring that AI systems are trustworthy, reliable, and stable.","author":[{"family":"Javed","given":"Haseeb"},{"family":"Elsappagh","given":"Shaker"},{"family":"Abuhmed","given":"Tamer"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s10462-024-11005-9","URL":"https://doi.org/10.1007/s10462-024-11005-9","source":"openalex"},{"id":"oa:W4404485831","type":"article-journal","title":"Artificial Intelligence Diagnosing of Oral Lichen Planus: A Comparative Study","abstract":"Early diagnosis of oral lichen planus (OLP) is challenging, which traditionally is dependent on clinical experience and subjective interpretation. Artificial intelligence (AI) technology has been widely applied in objective and rapid diagnoses. In this study, we aim to investigate the potential of AI diagnosis in OLP and evaluate its effectiveness in improving diagnostic accuracy and accelerating clinical decision making. A total of 128 confirmed OLP patients were included, and lesion images from various anatomical sites were collected. The diagnosis was performed using AI platforms, including ChatGPT-4O, ChatGPT (Diagram-Date extension), and Claude Opus, for AI directly identification and AI pre-training identification. After OLP feature training, the diagnostic accuracy of the AI platforms significantly improved, with the overall recognition rates of ChatGPT-4O, ChatGPT (Diagram-Date extension), and Claude Opus increasing from 59%, 68%, and 15% to 77%, 80%, and 50%, respectively. Additionally, the pre-training recognition rates for buccal mucosa reached 94%, 93%, and 56%, respectively. However, the AI platforms performed less effectively when recognizing lesions in less common sites and complex cases; for instance, the pre-training recognition rates for the gums were only 60%, 60%, and 20%, demonstrating significant limitations. The study highlights the strengths and limitations of different AI technologies and provides a reference for future AI applications in oral medicine.","author":[{"family":"Yu","given":"Sensen"},{"family":"Sun","given":"Wansu"},{"family":"Mi","given":"Dawei"},{"family":"Jin","given":"Siyu"},{"family":"Wu","given":"Xing"},{"family":"Xin","given":"B"},{"family":"Zhang","given":"Hengguo"},{"family":"Wang","given":"Yuanyin"},{"family":"Sun","given":"Xiaoyu"},{"family":"He","given":"Xin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/bioengineering11111159","URL":"https://doi.org/10.3390/bioengineering11111159","source":"openalex"},{"id":"oa:W4396605792","type":"article-journal","title":"How artificial intelligence can provide information about subdural hematoma: Assessment of readability, reliability, and quality of ChatGPT, BARD, and perplexity responses","abstract":"Subdural hematoma is defined as blood collection in the subdural space between the dura mater and arachnoid. Subdural hematoma is a condition that neurosurgeons frequently encounter and has acute, subacute and chronic forms. The incidence in adults is reported to be 1.72-20.60/100.000 people annually. Our study aimed to evaluate the quality, reliability and readability of the answers to questions asked to ChatGPT, Bard, and perplexity about \"Subdural Hematoma.\" In this observational and cross-sectional study, we asked ChatGPT, Bard, and perplexity to provide the 100 most frequently asked questions about \"Subdural Hematoma\" separately. Responses from both chatbots were analyzed separately for readability, quality, reliability and adequacy. When the median readability scores of ChatGPT, Bard, and perplexity answers were compared with the sixth-grade reading level, a statistically significant difference was observed in all formulas (P < .001). All 3 chatbot responses were found to be difficult to read. Bard responses were more readable than ChatGPT's (P < .001) and perplexity's (P < .001) responses for all scores evaluated. Although there were differences between the results of the evaluated calculators, perplexity's answers were determined to be more readable than ChatGPT's answers (P < .05). Bard answers were determined to have the best GQS scores (P < .001). Perplexity responses had the best Journal of American Medical Association and modified DISCERN scores (P < .001). ChatGPT, Bard, and perplexity's current capabilities are inadequate in terms of quality and readability of \"Subdural Hematoma\" related text content. The readability standard for patient education materials as determined by the American Medical Association, National Institutes of Health, and the United States Department of Health and Human Services is at or below grade 6. The readability levels of the responses of artificial intelligence applications such as ChatGPT, Bard, and perplexity are significantly higher than the recommended 6th grade level.","author":[{"family":"Gül","given":"Şanser"},{"family":"Erdemir","given":"İsmail"},{"family":"Hancı","given":"Volkan"},{"family":"Aydoğmuş","given":"Evren"},{"family":"Erkoç","given":"Yavuz"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1097/md.0000000000038009","URL":"https://doi.org/10.1097/md.0000000000038009","source":"openalex"},{"id":"oa:W4397049662","type":"article-journal","title":"Using artificial intelligence to study atherosclerosis from computed tomography imaging: A state-of-the-art review of the current literature","abstract":"With the enormous progress in the field of cardiovascular imaging in recent years, computed tomography (CT) has become readily available to phenotype atherosclerotic coronary artery disease. New analytical methods using artificial intelligence (AI) enable the analysis of complex phenotypic information of atherosclerotic plaques. In particular, deep learning-based approaches using convolutional neural networks (CNNs) facilitate tasks such as lesion detection, segmentation, and classification. New radiotranscriptomic techniques even capture underlying bio-histochemical processes through higher-order structural analysis of voxels on CT images. In the near future, the international large-scale Oxford Risk Factors And Non-invasive Imaging (ORFAN) study will provide a powerful platform for testing and validating prognostic AI-based models. The goal is the transition of these new approaches from research settings into a clinical workflow. In this review, we present an overview of existing AI-based techniques with focus on imaging biomarkers to determine the degree of coronary inflammation, coronary plaques, and the associated risk. Further, current limitations using AI-based approaches as well as the priorities to address these challenges will be discussed. This will pave the way for an AI-enabled risk assessment tool to detect vulnerable atherosclerotic plaques and to guide treatment strategies for patients.","author":[{"family":"Klüner","given":"Laura"},{"family":"Chan","given":"Kenneth"},{"family":"Antoniades","given":"Charalambos"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.atherosclerosis.2024.117580","URL":"https://doi.org/10.1016/j.atherosclerosis.2024.117580","source":"openalex"},{"id":"doi:10.1201/9781003450153-17","type":"article-journal","title":"Effective Use of Computational Biology and Artificial Intelligence in the Domain of Medical Oncology","abstract":"Computational biology and artificial intelligence (AI) have become indispensable assets in the specialized field of medicine, especially in oncology, fundamentally altering our understanding and management of cancer. This chapter aims to provide insight into how computational biology and AI are reshaping oncological practices. It emphasizes their transformative potential in augmenting diagnostic precision, customizing therapeutic interventions, and fast-tracking the AI process. In terms of diagnostics, these technologies have proven invaluable in interpreting intricate medical images like X-rays, MRIs, and CT scans, thereby facilitating the early identification of cancerous abnormalities. Machine learning models have excelled in pinpointing oncological markers, thereby aiding clinicians in making more informed decisions. Additionally, computational simulations are increasingly employed to predict cancer progression and treatment responses, allowing for highly individualized patient care. In the drug discovery arena, AI algorithms examine extensive biological datasets to identify new oncological drug candidates, predict potential drug interactions, and streamline drug formulation, thereby accelerating the timeline from lab to clinic. Utilizing a blend of machine learning, deep learning, and predictive analytics, these computational tools equip oncologists with the resources necessary for precise diagnostics, tailored treatments, and efficient drug development. As the fields of computational biology and AI continue to evolve, they hold immense promise for revolutionizing oncological healthcare practices and enhancing patient outcomes.","author":[{"family":"Saraf","given":"Sameeksha"},{"family":"De","given":"Arka"},{"family":"Tripathy","given":"BK"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1201/9781003450153-17","URL":"https://doi.org/10.1201/9781003450153-17","source":"openalex"},{"id":"doi:10.1016/j.zemedi.2024.02.001","type":"article-journal","title":"Towards quality management of artificial intelligence systems for medical applications","abstract":"The use of artificial intelligence systems in clinical routine is still hampered by the necessity of a medical device certification and/or by the difficulty of implementing these systems in a clinic's quality management system. In this context, the key questions for a user are how to ensure robust model predictions and how to appraise the quality of a model's results on a regular basis. In this paper we discuss some conceptual foundation for a clinical implementation of a machine learning system and argue that both vendors and users should take certain responsibilities, as is already common practice for high-risk medical equipment. We propose the methodology from AAPM Task Group 100 report No. 283 as a conceptual framework for developing risk-driven a quality management program for a clinical process that encompasses a machine learning system. This is illustrated with an example of a clinical workflow. Our analysis shows how the risk evaluation in this framework can accommodate artificial intelligence based systems independently of their robustness evaluation or the user's in-house expertise. In particular, we highlight how the degree of interpretability of a machine learning system can be systematically accounted for within the risk evaluation and in the development of a quality management system.","author":[{"family":"Mercolli","given":"Lorenzo"},{"family":"Rominger","given":"Axel"},{"family":"Shi","given":"Kuangyu"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.zemedi.2024.02.001","URL":"https://doi.org/10.1016/j.zemedi.2024.02.001","source":"openalex"},{"id":"doi:10.20944/preprints202408.1215.v1","type":"manuscript","title":"Anxiety among Medical Students Regarding Generative Artificial Intelligence Models: A Pilot Descriptive Study","abstract":"Despite the potential benefits of generative Artificial Intelligence (genAI), concerns about its psy-chological impact on medical students, especially with regard to job displacement, are apparent. This pilot study, conducted in Jordan during July–August 2024, aimed to examine the specific fears, anxieties, mistrust, and ethical concerns medical students could harbor towards genAI. Using a cross-sectional survey design, data were collected from 164 medical students studying in Jordan across various academic years, employing a structured self-administered questionnaire with an internally consistent FAME scale—representing Fear, Anxiety, Mistrust, and Ethics and comprising 12 items, with three items for each construct. The results indicated variable levels of anxiety towards genAI among the participating medical students: 34.1% reported no anxiety about genAI role in their future careers (n = 56), while 41.5% were slightly anxious (n = 61), 22.0% somewhat anxious (n = 36), and 2.4% extremely anxious (n = 4). Among the FAME constructs, Mistrust was the most agreed upon (mean: 12.35±2.78), followed by Ethics construct (mean: 10.86±2.90), Fear (mean: 9.49±3.53), and Anxiety (mean: 8.91±3.68). Sex, academic level, and Grade Point Average (GPA) did not significantly affect the students’ perceptions of genAI. However, there was a notable direct association between the students’ general anxiety about genAI and elevated scores in the Fear, Anxiety, and Ethics constructs of the FAME scale. Prior exposure to genAI and its previous use did not significantly modify the scores of the FAME scale. These findings highlighted the critical need for refined educational strategies to address the integration of genAI in medical training. The results demonstrated a pervasive anxiety, fear, mistrust, and ethical concerns among medical students regarding the deployment of genAI in healthcare, indicating the necessity for curriculum modifi-cations that focus specifically on these areas. Interventions should be tailored to increase genAI familiarity and competency, which would alleviate apprehension and equip future physicians to engage with this inevitable technology effectively. The study also highlighted the importance of incorporating ethical discussions into medical courses to address mistrust and concerns about the human-centered aspects of genAI. Conclusively, the study calls for a proactive evolution of medical education to prepare students for AI-driven healthcare practices shortly to ensure that physicians are well-prepared, confident, and ethically informed in their professional interactions with genAI technologies.","author":[{"family":"Sallam","given":"Malik"},{"family":"Al-Mahzoum","given":"Kholoud"},{"family":"Almutairi","given":"Yousef"},{"family":"Alaqeel","given":"Omar"},{"family":"Abu-Salami","given":"Anan"},{"family":"Almutairi","given":"Zaid"},{"family":"Alsarraf","given":"Alhur"},{"family":"Barakat","given":"Muna"}],"issued":{"date-parts":[[2024]]},"DOI":"10.20944/preprints202408.1215.v1","URL":"https://doi.org/10.20944/preprints202408.1215.v1","source":"preprints"},{"id":"oa:W4319027038","type":"article-journal","title":"At the Confluence of Artificial Intelligence and Edge Computing in IoT-Based Applications: A Review and New Perspectives","abstract":"Given its advantages in low latency, fast response, context-aware services, mobility, and privacy preservation, edge computing has emerged as the key support for intelligent applications and 5G/6G Internet of things (IoT) networks. This technology extends the cloud by providing intermediate services at the edge of the network and improving the quality of service for latency-sensitive applications. Many AI-based solutions with machine learning, deep learning, and swarm intelligence have exhibited the high potential to perform intelligent cognitive sensing, intelligent network management, big data analytics, and security enhancement for edge-based smart applications. Despite its many benefits, there are still concerns about the required capabilities of intelligent edge computing to deal with the computational complexity of machine learning techniques for big IoT data analytics. Resource constraints of edge computing, distributed computing, efficient orchestration, and synchronization of resources are all factors that require attention for quality of service improvement and cost-effective development of edge-based smart applications. In this context, this paper aims to explore the confluence of AI and edge in many application domains in order to leverage the potential of the existing research around these factors and identify new perspectives. The confluence of edge computing and AI improves the quality of user experience in emergency situations, such as in the Internet of vehicles, where critical inaccuracies or delays can lead to damage and accidents. These are the same factors that most studies have used to evaluate the success of an edge-based application. In this review, we first provide an in-depth analysis of the state of the art of AI in edge-based applications with a focus on eight application areas: smart agriculture, smart environment, smart grid, smart healthcare, smart industry, smart education, smart transportation, and security and privacy. Then, we present a qualitative comparison that emphasizes the main objective of the confluence, the roles and the use of artificial intelligence at the network edge, and the key enabling technologies for edge analytics. Then, open challenges, future research directions, and perspectives are identified and discussed. Finally, some conclusions are drawn.","author":[{"family":"Bourechak","given":"Amira"},{"family":"Zedadra","given":"Ouarda"},{"family":"Kouahla","given":"Mohamed"},{"family":"Guerrieri","given":"Antonio"},{"family":"Séridi","given":"Hamid"},{"family":"Fortino","given":"Giancarlo"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/s23031639","URL":"https://doi.org/10.3390/s23031639","source":"openalex"},{"id":"oa:W4379473621","type":"article-journal","title":"Rise of Artificial Intelligence in Business and Industry","abstract":"The ongoing development of business and the most recent advances in artificial intelligence (AI) allow for the many business practices to be improved by the capacity to establish new forms of collaboration, which is a significant competitive advantage. This rapidly developing technology enables to offer brand services and even some new forms of business interactions with consumers and personnel. The digitalization of AI concurrently emphasized for businesses that they need concentrate on their present strategies while also routinely and early pursuing new chances in the market. Not only in business but also in different industry sectors, Al techniques are being used and revolutionized different industry sectors. This review focuses on the application of AI techniques in business and different industries.","author":[{"family":"Bharadiya","given":"Jasmin"},{"family":"Thomas","given":"Reji"},{"family":"Ahmed","given":"Farhan"}],"issued":{"date-parts":[[2023]]},"DOI":"10.9734/jerr/2023/v25i3893","URL":"https://doi.org/10.9734/jerr/2023/v25i3893","source":"openalex"},{"id":"oa:W4387966880","type":"article-journal","title":"Generative artificial intelligence empowers digital twins in drug discovery and clinical trials","abstract":"INTRODUCTION: The concept of Digital Twins (DTs) translated to drug development and clinical trials describes virtual representations of systems of various complexities, ranging from individual cells to entire humans, and enables in silico simulations and experiments. DTs increase the efficiency of drug discovery and development by digitalizing processes associated with high economic, ethical, or social burden. The impact is multifaceted: DT models sharpen disease understanding, support biomarker discovery and accelerate drug development, thus advancing precision medicine. One way to realize DTs is by generative artificial intelligence (AI), a cutting-edge technology that enables the creation of novel, realistic and complex data with desired properties. AREAS COVERED: The authors provide a brief introduction to generative AI and describe how it facilitates the modeling of DTs. In addition, they compare existing implementations of generative AI for DTs in drug discovery and clinical trials. Finally, they discuss technical and regulatory challenges that should be addressed before DTs can transform drug discovery and clinical trials. EXPERT OPINION: The current state of DTs in drug discovery and clinical trials does not exploit the entire power of generative AI yet and is limited to simulation of a small number of characteristics. Nonetheless, generative AI has the potential to transform the field by leveraging recent developments in deep learning and customizing models for the needs of scientists, physicians and patients.","author":[{"family":"Bordukova","given":"Maria"},{"family":"Makarov","given":"Nikita"},{"family":"Rodriguezesteban","given":"Raul"},{"family":"Schmich","given":"Fabian"},{"family":"Menden","given":"Michael"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1080/17460441.2023.2273839","URL":"https://doi.org/10.1080/17460441.2023.2273839","source":"openalex"},{"id":"oa:W4390906507","type":"article-journal","title":"Artificial Intelligence Alone Will Not Democratise Education: On Educational Inequality, Techno-Solutionism and Inclusive Tools","abstract":"Artificial Intelligence (AI) in Education claims to have the potential for building personalised curricula, as well as bringing opportunities for democratising education and creating a renaissance of new ways of teaching and learning. Millions of students are starting to benefit from the use of these technologies, but millions more around the world are not, due to the digital divide and deep pre-existing social and educational inequalities. If this trend continues, the first large-scale delivery of AI in Education could lead to greater educational inequality, along with a global misallocation of educational resources motivated by the current techno-solutionist narrative, which proposes technological solutions as a quick and flawless way to solve complex real-world problems. This work focuses on posing questions about the future of AI in Education, intending to initiate the pressing conversation that could set the right foundations (e.g., inclusion and diversity) for a new generation of education that is permeated with AI technology. The main goal of our opinion piece is to conceptualise a sustainable, large-scale and inclusive AI for the education ecosystem that facilitates equitable, high-quality lifelong learning opportunities for all. The contribution starts by synthesising how AI might change how we learn and teach, focusing on the case of personalised learning companions and assistive technology for disability. Then, we move on to discuss some socio-technical features that will be crucial to avoiding the perils of these AI systems worldwide (and perhaps ensuring their success by leveraging more inclusive education). This work also discusses the potential of using AI together with free, participatory and democratic resources, such as Wikipedia, Open Educational Resources and open-source tools. We emphasise the need for collectively designing human-centred, transparent, interactive and collaborative AI-based algorithms that empower and give complete agency to stakeholders, as well as supporting new emerging pedagogies. Finally, we ask what it would take for this educational revolution to provide egalitarian and empowering access to education that transcends any political, cultural, language, geographical and learning-ability barriers, so that educational systems can be responsive to all learners’ needs.","author":[{"family":"Bulathwela","given":"Sahan"},{"family":"Pérezortiz","given":"María"},{"family":"Holloway","given":"Catherine"},{"family":"Cukurova","given":"Mutlu"},{"family":"Shawetaylor","given":"John"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/su16020781","URL":"https://doi.org/10.3390/su16020781","source":"openalex"},{"id":"oa:W4386780386","type":"article-journal","title":"Artificial intelligence for medicine: Progress, challenges, and perspectives","abstract":"Artificial Intelligence (AI) has transformed how we live and how we think, and it will change how we practice medicine. With multimodal big data, we can develop large medical models that enables what used to unimaginable, such as early cancer detection several years in advance and effective control of virus outbreaks without imposing social burdens. The future is promising, and we are witnessing the advancement. That said, there are challenges that cannot be overlooked. For example, data generated is often isolated and difficult to integrate from both perspectives of data ownership and fusion algorithms. Additionally, existing AI models are often treated as black boxes, resulting in vague interpretation of the results. Patients also exhibit a lack of trust to AI applications, and there are insufficient regulations to protect patients’ privacy and rights. However, with the advancement of AI technologies, such as more sophisticated multimodal algorithms and federated learning, we may overcome the barriers posed by data silos. Deeper understanding of human brain and network structures can also help to unravel the mysteries of neural networks and construct more transparent yet more powerful AI models. It has become something of a trend that an increasing number of clinicians and patients will implement AI in their life and medical practice, which in turn can generate more data and improve the performance of models and networks. Last but not the least, it is crucial to monitor the practice of AI in medicine and ensure its equity, security, and responsibility.","author":[{"family":"Huang","given":"Tao"},{"family":"Xu","given":"Huiyu"},{"family":"Wang","given":"Haitao"},{"family":"Huang","given":"Haofan"},{"family":"Xu","given":"Yongjun"},{"family":"Li","given":"Baohua"},{"family":"Hong","given":"Shenda"},{"family":"Feng","given":"Guoshuang"},{"family":"Kui","given":"Shuyi"},{"family":"Liu","given":"Guangjian"},{"family":"Jiang","given":"Dehua"},{"family":"Li","given":"Zhicheng"},{"family":"Li","given":"Ye"},{"family":"Ma","given":"Congcong"},{"family":"Su","given":"Chunyan"},{"family":"Wang","given":"Wei"},{"family":"Li","given":"Rong"},{"family":"Lai","given":"Puxiang"},{"family":"Qiao","given":"Jie"}],"issued":{"date-parts":[[2023]]},"DOI":"10.59717/j.xinn-med.2023.100030","URL":"https://doi.org/10.59717/j.xinn-med.2023.100030","source":"openalex"},{"id":"oa:W4386416103","type":"article-journal","title":"Reimagining Healthcare: Unleashing the Power of Artificial Intelligence in Medicine","abstract":"Artificial intelligence (AI) has opened new medical avenues and revolutionized diagnostic and therapeutic practices, allowing healthcare providers to overcome significant challenges associated with cost, disease management, accessibility, and treatment optimization. Prominent AI technologies such as machine learning (ML) and deep learning (DL) have immensely influenced diagnostics, patient monitoring, novel pharmaceutical discoveries, drug development, and telemedicine. Significant innovations and improvements in disease identification and early intervention have been made using AI-generated algorithms for clinical decision support systems and disease prediction models. AI has remarkably impacted clinical drug trials by amplifying research into drug efficacy, adverse events, and candidate molecular design. AI's precision and analysis regarding patients' genetic, environmental, and lifestyle factors have led to individualized treatment strategies. During the COVID-19 pandemic, AI-assisted telemedicine set a precedent for remote healthcare delivery and patient follow-up. Moreover, AI-generated applications and wearable devices have allowed ambulatory monitoring of vital signs. However, apart from being immensely transformative, AI's contribution to healthcare is subject to ethical and regulatory concerns. AI-backed data protection and algorithm transparency should be strictly adherent to ethical principles. Vigorous governance frameworks should be in place before incorporating AI in mental health interventions through AI-operated chatbots, medical education enhancements, and virtual reality-based training. The role of AI in medical decision-making has certain limitations, necessitating the importance of hands-on experience. Therefore, reaching an optimal balance between AI's capabilities and ethical considerations to ensure impartial and neutral performance in healthcare applications is crucial. This narrative review focuses on AI's impact on healthcare and the importance of ethical and balanced incorporation to make use of its full potential.","author":[{"family":"Iqbal","given":"Javed"},{"family":"Jaimes","given":"Diana"},{"family":"Makineni","given":"Pallavi"},{"family":"Subramani","given":"Sachin"},{"family":"Hemaida","given":"Sarah"},{"family":"Thugu","given":"Thanmai"},{"family":"Butt","given":"Amna"},{"family":"Sikto","given":"Jarin"},{"family":"Kaur","given":"Pareena"},{"family":"Lak","given":"Muhammad"},{"family":"Augustine","given":"Monisha"},{"family":"Shahzad","given":"Roheen"},{"family":"Arain","given":"Mustafa"}],"issued":{"date-parts":[[2023]]},"DOI":"10.7759/cureus.44658","URL":"https://doi.org/10.7759/cureus.44658","source":"openalex"},{"id":"oa:W4319986951","type":"article-journal","title":"Metaverse for Healthcare: A Survey on Potential Applications, Challenges and Future Directions","abstract":"The rapid progress in digitalization and automation have led to an accelerated growth in healthcare, generating novel models that are creating new channels for rendering treatment at reduced cost. The Metaverse is an emerging technology in the digital space which has huge potential in healthcare, enabling realistic experiences to the patients as well as the medical practitioners. The Metaverse is a confluence of multiple enabling technologies such as artificial intelligence, virtual reality, augmented reality, internet of medical devices, robotics, quantum computing, etc. through which new directions for providing quality healthcare treatment and services can be explored. The amalgamation of these technologies ensures immersive, intimate and personalized patient care. It also provides adaptive intelligent solutions that eliminates the barriers between healthcare providers and receivers. This article provides a comprehensive review of the Metaverse for healthcare, emphasizing on the state of the art, the enabling technologies to adopt the Metaverse for healthcare, the potential applications, and the related projects. The issues in the adaptation of the Metaverse for healthcare applications are also identified and the plausible solutions are highlighted as part of future research directions.","author":[{"family":"Chengoden","given":"Rajeswari"},{"family":"Victor","given":"Nancy"},{"family":"Huynhthe","given":"Thien"},{"family":"Yenduri","given":"Gokul"},{"family":"Jhaveri","given":"Rutvij"},{"family":"Alazab","given":"Mamoun"},{"family":"Bhattacharya","given":"Sweta"},{"family":"Hegde","given":"Pawan"},{"family":"Maddikunta","given":"Praveen"},{"family":"Gadekallu","given":"Thippa"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/access.2023.3241628","URL":"https://doi.org/10.1109/access.2023.3241628","source":"openalex"},{"id":"oa:W4402037523","type":"article-journal","title":"Artificial Intelligence and the Dehumanization of Patient Care","abstract":"The integration of artificial intelligence (AI) into healthcare is rapidly transforming patient care, offering numerous advantages in diagnostics, efficiency, and clinical decision-making. However, this technological shift raises significant concerns about the potential erosion of the doctor-patient relationship, a cornerstone of effective medical practice. AI’s increasing role risks depersonalizing healthcare, as the emphasis on data-driven decisions may overshadow the empathy, trust, and personalized care traditionally provided by human clinicians. The \"black-box\" nature of AI algorithms further exacerbates this issue, as the lack of transparency in AI decision-making processes can undermine patient trust. Additionally, AI systems trained on biased datasets may inadvertently widen health disparities, particularly for underrepresented populations. While AI has the potential to streamline routine tasks and reduce the burden on healthcare providers, it is essential to ensure that these advancements do not come at the cost of the human connection vital to patient care. To address these challenges, future research and development should focus on creating AI systems that enhance, rather than replace, the compassionate aspects of healthcare. This balanced approach is crucial to preserving the integrity of the doctor-patient relationship while harnessing the benefits of AI, ultimately ensuring that technological progress aligns with the core values of medical practice.","author":[{"family":"Akingbola","given":"Adewunmi"},{"family":"Adeleke","given":"Oluwatimilehin"},{"family":"Idris","given":"Ayotomiwa"},{"family":"Adewole","given":"Olajumoke"},{"family":"Adegbesan","given":"Abiodun"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.glmedi.2024.100138","URL":"https://doi.org/10.1016/j.glmedi.2024.100138","source":"openalex"},{"id":"oa:W4394598267","type":"article-journal","title":"The potential for artificial intelligence to transform healthcare: perspectives from international health leaders","abstract":"Artificial intelligence (AI) has the potential to transform care delivery by improving health outcomes, patient safety, and the affordability and accessibility of high-quality care. AI will be critical to building an infrastructure capable of caring for an increasingly aging population, utilizing an ever-increasing knowledge of disease and options for precision treatments, and combatting workforce shortages and burnout of medical professionals. However, we are not currently on track to create this future. This is in part because the health data needed to train, test, use, and surveil these tools are generally neither standardized nor accessible. There is also universal concern about the ability to monitor health AI tools for changes in performance as they are implemented in new places, used with diverse populations, and over time as health data may change. The Future of Health (FOH), an international community of senior health care leaders, collaborated with the Duke-Margolis Institute for Health Policy to conduct a literature review, expert convening, and consensus-building exercise around this topic. This commentary summarizes the four priority action areas and recommendations for health care organizations and policymakers across the globe that FOH members identified as important for fully realizing AI's potential in health care: improving data quality to power AI, building infrastructure to encourage efficient and trustworthy development and evaluations, sharing data for better AI, and providing incentives to accelerate the progress and impact of AI.","author":[{"family":"Silcox","given":"Christina"},{"family":"Zimlichmann","given":"Eyal"},{"family":"Huber","given":"Katie"},{"family":"Rowen","given":"Neil"},{"family":"Saunders","given":"RS"},{"family":"Mcclellan","given":"Mark"},{"family":"Kahn","given":"Charles"},{"family":"Salzberg","given":"Claudia"},{"family":"Bates","given":"David"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41746-024-01097-6","URL":"https://doi.org/10.1038/s41746-024-01097-6","source":"openalex"},{"id":"oa:W4387966725","type":"article-journal","title":"Artificial Intelligence for Surface‐Enhanced Raman Spectroscopy","abstract":"Surface-enhanced Raman spectroscopy (SERS), well acknowledged as a fingerprinting and sensitive analytical technique, has exerted high applicational value in a broad range of fields including biomedicine, environmental protection, food safety among the others. In the endless pursuit of ever-sensitive, robust, and comprehensive sensing and imaging, advancements keep emerging in the whole pipeline of SERS, from the design of SERS substrates and reporter molecules, synthetic route planning, instrument refinement, to data preprocessing and analysis methods. Artificial intelligence (AI), which is created to imitate and eventually exceed human behaviors, has exhibited its power in learning high-level representations and recognizing complicated patterns with exceptional automaticity. Therefore, facing up with the intertwining influential factors and explosive data size, AI has been increasingly leveraged in all the above-mentioned aspects in SERS, presenting elite efficiency in accelerating systematic optimization and deepening understanding about the fundamental physics and spectral data, which far transcends human labors and conventional computations. In this review, the recent progresses in SERS are summarized through the integration of AI, and new insights of the challenges and perspectives are provided in aim to better gear SERS toward the fast track.","author":[{"family":"Bi","given":"Xinyuan"},{"family":"Lin","given":"Li"},{"family":"Chen","given":"Zhou"},{"family":"Ye","given":"Jian"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/smtd.202301243","URL":"https://doi.org/10.1002/smtd.202301243","source":"openalex"},{"id":"oa:W4401219516","type":"article-journal","title":"Artificial intelligence: revolutionizing robotic surgery: review","abstract":"Robotic surgery, known for its minimally invasive techniques and computer-controlled robotic arms, has revolutionized modern medicine by providing improved dexterity, visualization, and tremor reduction compared to traditional methods. The integration of artificial intelligence (AI) into robotic surgery has further advanced surgical precision, efficiency, and accessibility. This paper examines the current landscape of AI-driven robotic surgical systems, detailing their benefits, limitations, and future prospects. Initially, AI applications in robotic surgery focused on automating tasks like suturing and tissue dissection to enhance consistency and reduce surgeon workload. Present AI-driven systems incorporate functionalities such as image recognition, motion control, and haptic feedback, allowing real-time analysis of surgical field images and optimizing instrument movements for surgeons. The advantages of AI integration include enhanced precision, reduced surgeon fatigue, and improved safety. However, challenges such as high development costs, reliance on data quality, and ethical concerns about autonomy and liability hinder widespread adoption. Regulatory hurdles and workflow integration also present obstacles. Future directions for AI integration in robotic surgery include enhancing autonomy, personalizing surgical approaches, and refining surgical training through AI-powered simulations and virtual reality. Overall, AI integration holds promise for advancing surgical care, with potential benefits including improved patient outcomes and increased access to specialized expertise. Addressing challenges and promoting responsible adoption are essential for realizing the full potential of AI-driven robotic surgery.","author":[{"family":"Iftikhar","given":"Muhammad"},{"family":"Saqib","given":"Muhammad"},{"family":"Zareen","given":"Muhammad"},{"family":"Mumtaz","given":"Hassan"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1097/ms9.0000000000002426","URL":"https://doi.org/10.1097/ms9.0000000000002426","source":"openalex"},{"id":"oa:W4386855886","type":"article-journal","title":"ChatGPT in action: Harnessing artificial intelligence potential and addressing ethical challenges in medicine, education, and scientific research","abstract":"Artificial intelligence (AI) tools, like OpenAI's Chat Generative Pre-trained Transformer (ChatGPT), hold considerable potential in healthcare, academia, and diverse industries. Evidence demonstrates its capability at a medical student level in standardized tests, suggesting utility in medical education, radiology reporting, genetics research, data optimization, and drafting repetitive texts such as discharge summaries. Nevertheless, these tools should augment, not supplant, human expertise. Despite promising applications, ChatGPT confronts limitations, including critical thinking tasks and generating false references, necessitating stringent cross-verification. Ensuing concerns, such as potential misuse, bias, blind trust, and privacy, underscore the need for transparency, accountability, and clear policies. Evaluations of AI-generated content and preservation of academic integrity are critical. With responsible use, AI can significantly improve healthcare, academia, and industry without compromising integrity and research quality. For effective and ethical AI deployment, collaboration amongst AI developers, researchers, educators, and policymakers is vital. The development of domain-specific tools, guidelines, regulations, and the facilitation of public dialogue must underpin these endeavors to responsibly harness AI's potential.","author":[{"family":"Jeyaraman","given":"Madhan"},{"family":"Ramasubramanian","given":"Swaminathan"},{"family":"Balaji","given":"Sangeetha"},{"family":"Jeyaraman","given":"Naveen"},{"family":"Nallakumarasamy","given":"Arulkumar"},{"family":"Sharma","given":"Shilpa"}],"issued":{"date-parts":[[2023]]},"DOI":"10.5662/wjm.v13.i4.170","URL":"https://doi.org/10.5662/wjm.v13.i4.170","source":"openalex"},{"id":"oa:W4387949693","type":"article-journal","title":"A scoping review of artificial intelligence-based methods for diabetes risk prediction","abstract":"The increasing prevalence of type 2 diabetes mellitus (T2DM) and its associated health complications highlight the need to develop predictive models for early diagnosis and intervention. While many artificial intelligence (AI) models for T2DM risk prediction have emerged, a comprehensive review of their advancements and challenges is currently lacking. This scoping review maps out the existing literature on AI-based models for T2DM prediction, adhering to the PRISMA extension for Scoping Reviews guidelines. A systematic search of longitudinal studies was conducted across four databases, including PubMed, Scopus, IEEE-Xplore, and Google Scholar. Forty studies that met our inclusion criteria were reviewed. Classical machine learning (ML) models dominated these studies, with electronic health records (EHR) being the predominant data modality, followed by multi-omics, while medical imaging was the least utilized. Most studies employed unimodal AI models, with only ten adopting multimodal approaches. Both unimodal and multimodal models showed promising results, with the latter being superior. Almost all studies performed internal validation, but only five conducted external validation. Most studies utilized the area under the curve (AUC) for discrimination measures. Notably, only five studies provided insights into the calibration of their models. Half of the studies used interpretability methods to identify key risk predictors revealed by their models. Although a minority highlighted novel risk predictors, the majority reported commonly known ones. Our review provides valuable insights into the current state and limitations of AI-based models for T2DM prediction and highlights the challenges associated with their development and clinical integration.","author":[{"family":"Mohsen","given":"Farida"},{"family":"Al-Absi","given":"Hamada"},{"family":"Yousri","given":"Noha"},{"family":"Hajj","given":"Nady"},{"family":"Shah","given":"Zubair"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1038/s41746-023-00933-5","URL":"https://doi.org/10.1038/s41746-023-00933-5","source":"openalex"},{"id":"oa:W4384068429","type":"article-journal","title":"Artificial intelligence in supply chain and operations management: a multiple case study research","abstract":"Artificial intelligence (AI) is increasingly considered a source of competitive advantage in operations and supply chain management (OSCM). However, many organisations still struggle to adopt it successfully and empirical studies providing clear indications are scarce in the literature. This research aims to shed light on how AI applications can support OSCM processes and to identify benefits and barriers to their implementation. To this end, it conducts a multiple case study with semi-structured interviews in six companies, totalling 17 implementation cases. The Supply Chain Operations Reference (SCOR) model guided the entire study and the analysis of the results by targeting specific processes. The results highlighted how AI methods in OSCM can increase the companies’ competitiveness by reducing costs and lead times and improving service levels, quality, safety, and sustainability. However, they also identify barriers in the implementation of AI, such as ensuring data quality, lack of specific skills, need for high investments, lack of clarity on economic benefits and lack of experience in cost analysis for AI projects. Although the nature of the study is not suitable for wide generalisation, it offers clear guidance for practitioners facing AI dilemmas in specific SCOR processes and provides the basis for further future research.","author":[{"family":"Cannas","given":"Violetta"},{"family":"Ciano","given":"Maria"},{"family":"Saltalamacchia","given":"Mattia"},{"family":"Secchi","given":"Raffaele"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1080/00207543.2023.2232050","URL":"https://doi.org/10.1080/00207543.2023.2232050","source":"openalex"},{"id":"oa:W4389131563","type":"article-journal","title":"Empowering Medical Students: Harnessing Artificial Intelligence for Precision Point-of-Care Echocardiography Assessment of Left Ventricular Ejection Fraction","abstract":"Introduction: Point-of-care ultrasound (POCUS) use is now universal among nonexperts. Artificial intelligence (AI) is currently employed by nonexperts in various imaging modalities to assist in diagnosis and decision making. Aim: To evaluate the diagnostic accuracy of POCUS, operated by medical students with the assistance of an AI-based tool for assessing the left ventricular ejection fraction (LVEF) of patients admitted to a cardiology department. Methods: Eight students underwent a 6-hour didactic and hands-on training session. Participants used a hand-held ultrasound device (HUD) equipped with an AI-based tool for the automatic evaluation of LVEF. The clips were assessed for LVEF by three methods: visually by the students, by students + the AI-based tool, and by the cardiologists. All LVEF measurements were compared to formal echocardiography completed within 24 hours and were evaluated for LVEF using the Simpson method and eyeballing assessment by expert echocardiographers. Results: The study included 88 patients (aged 58.3 ± 16.3 years). The AI-based tool measurement was unsuccessful in 6 cases. Comparing LVEF reported by students' visual evaluation and students + AI vs. cardiologists revealed a correlation of 0.51 and 0.83, respectively. Comparing these three evaluation methods with the echocardiographers revealed a moderate/substantial agreement for the students + AI and cardiologists but only a fair agreement for the students' visual evaluation. Conclusion: Medical students' utilization of an AI-based tool with a HUD for LVEF assessment achieved a level of accuracy similar to that of cardiologists. Furthermore, the use of AI by the students achieved moderate to substantial inter-rater reliability with expert echocardiographers' evaluation.","author":[{"family":"Dadon","given":"Ziv"},{"family":"Orlev","given":"Amir"},{"family":"Butnaru","given":"Adi"},{"family":"Rosenmann","given":"David"},{"family":"Glikson","given":"Michael"},{"family":"Gottlieb","given":"Shmuel"},{"family":"Alpert","given":"Evan"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1155/2023/5225872","URL":"https://doi.org/10.1155/2023/5225872","source":"openalex"},{"id":"oa:W4403640721","type":"article-journal","title":"Attitudes and perceptions of Thai medical students regarding artificial intelligence in radiology and medicine","abstract":"INTRODUCTION: Artificial Intelligence (AI) has made a profound impact on the medical sector, particularly in radiology. The integration of AI knowledge into medical education is essential to equip future healthcare professionals with the skills needed to effectively leverage these advancements in their practices. Despite its significance, many medical schools have yet to incorporate AI into their curricula. This study aims to assess the attitudes of medical students in Thailand toward AI and its application in radiology, with the objective of better planning for its inclusion. METHODS: Between February and June 2022, we conducted a survey in two Thai medical schools: Chiang Mai University in Northern Thailand and Prince of Songkla University in Southern Thailand. We employed 5-point Likert scale questions (ranging from strongly agree to strongly disagree) to evaluate students' opinions on three main aspects: (1) their understanding of AI, (2) the inclusion of AI in their medical education, and (3) the potential impact of AI on medicine and radiology. RESULTS: Our findings revealed that merely 31% of medical students perceived to have a basic understanding of AI. Nevertheless, nearly all students (93.6%) recognized the value of AI training for their careers and strongly advocated for its inclusion in the medical school curriculum. Furthermore, those students who had a better understanding of AI were more likely to believe that AI would revolutionize the field of radiology (p = 0.02), making it more captivating and impactful (p = 0.04). CONCLUSION: Our study highlights a noticeable gap in the understanding of AI among medical students in Thailand and its practical applications in healthcare. However, the overwhelming consensus among these students is their readiness to embrace the incorporation of AI training into their medical education. This enthusiasm holds the promise of enhancing AI adoption, ultimately leading to an improvement in the standard of healthcare services in Thailand, aligning with the country's healthcare vision.","author":[{"family":"Angkurawaranon","given":"Salita"},{"family":"Inmutto","given":"Nakarin"},{"family":"Bannangkoon","given":"Kittipitch"},{"family":"Wonghan","given":"Surapat"},{"family":"Kham-Ai","given":"Thanawat"},{"family":"Khumma","given":"Porched"},{"family":"Daengpisut","given":"Kanvijit"},{"family":"Thabarsa","given":"Phattanun"},{"family":"Angkurawaranon","given":"Chaisiri"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s12909-024-06150-2","URL":"https://doi.org/10.1186/s12909-024-06150-2","source":"openalex"},{"id":"oa:W4385650719","type":"article-journal","title":"Artificial Intelligence Ethics and Challenges in Healthcare Applications: A Comprehensive Review in the Context of the European GDPR Mandate","abstract":"This study examines the ethical issues surrounding the use of Artificial Intelligence (AI) in healthcare, specifically nursing, under the European General Data Protection Regulation (GDPR). The analysis delves into how GDPR applies to healthcare AI projects, encompassing data collection and decision-making stages, to reveal the ethical implications at each step. A comprehensive review of the literature categorizes research investigations into three main categories: Ethical Considerations in AI; Practical Challenges and Solutions in AI Integration; and Legal and Policy Implications in AI. The analysis uncovers a significant research deficit in this field, with a particular focus on data owner rights and AI ethics within GDPR compliance. To address this gap, the study proposes new case studies that emphasize the importance of comprehending data owner rights and establishing ethical norms for AI use in medical applications, especially in nursing. This review makes a valuable contribution to the AI ethics debate and assists nursing and healthcare professionals in developing ethical AI practices. The insights provided help stakeholders navigate the intricate terrain of data protection, ethical considerations, and regulatory compliance in AI-driven healthcare. Lastly, the study introduces a case study of a real AI health-tech project named SENSOMATT, spotlighting GDPR and privacy issues.","author":[{"family":"Amini","given":"Mohammad"},{"family":"Jesus","given":"Marcia"},{"family":"Sheikholeslami","given":"Davood"},{"family":"Alves","given":"Paulo"},{"family":"Benam","given":"Aliakbar"},{"family":"Hariri","given":"Fatemeh"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/make5030053","URL":"https://doi.org/10.3390/make5030053","source":"openalex"},{"id":"oa:W4399564385","type":"article-journal","title":"Artificial intelligence and radiologists in prostate cancer detection on MRI (PI-CAI): an international, paired, non-inferiority, confirmatory study","abstract":"BACKGROUND: Artificial intelligence (AI) systems can potentially aid the diagnostic pathway of prostate cancer by alleviating the increasing workload, preventing overdiagnosis, and reducing the dependence on experienced radiologists. We aimed to investigate the performance of AI systems at detecting clinically significant prostate cancer on MRI in comparison with radiologists using the Prostate Imaging-Reporting and Data System version 2.1 (PI-RADS 2.1) and the standard of care in multidisciplinary routine practice at scale. METHODS: In this international, paired, non-inferiority, confirmatory study, we trained and externally validated an AI system (developed within an international consortium) for detecting Gleason grade group 2 or greater cancers using a retrospective cohort of 10 207 MRI examinations from 9129 patients. Of these examinations, 9207 cases from three centres (11 sites) based in the Netherlands were used for training and tuning, and 1000 cases from four centres (12 sites) based in the Netherlands and Norway were used for testing. In parallel, we facilitated a multireader, multicase observer study with 62 radiologists (45 centres in 20 countries; median 7 [IQR 5-10] years of experience in reading prostate MRI) using PI-RADS (2.1) on 400 paired MRI examinations from the testing cohort. Primary endpoints were the sensitivity, specificity, and the area under the receiver operating characteristic curve (AUROC) of the AI system in comparison with that of all readers using PI-RADS (2.1) and in comparison with that of the historical radiology readings made during multidisciplinary routine practice (ie, the standard of care with the aid of patient history and peer consultation). Histopathology and at least 3 years (median 5 [IQR 4-6] years) of follow-up were used to establish the reference standard. The statistical analysis plan was prespecified with a primary hypothesis of non-inferiority (considering a margin of 0·05) and a secondary hypothesis of superiority towards the AI system, if non-inferiority was confirmed. This study was registered at ClinicalTrials.gov, NCT05489341. FINDINGS: Of the 10 207 examinations included from Jan 1, 2012, through Dec 31, 2021, 2440 cases had histologically confirmed Gleason grade group 2 or greater prostate cancer. In the subset of 400 testing cases in which the AI system was compared with the radiologists participating in the reader study, the AI system showed a statistically superior and non-inferior AUROC of 0·91 (95% CI 0·87-0·94; p<0·0001), in comparison to the pool of 62 radiologists with an AUROC of 0·86 (0·83-0·89), with a lower boundary of the two-sided 95% Wald CI for the difference in AUROC of 0·02. At the mean PI-RADS 3 or greater operating point of all readers, the AI system detected 6·8% more cases with Gleason grade group 2 or greater cancers at the same specificity (57·7%, 95% CI 51·6-63·3), or 50·4% fewer false-positive results and 20·0% fewer cases with Gleason grade group 1 cancers at the same sensitivity (89·4%, 95% CI 85·3-92·9). In all 1000 testing cases where the AI system was compared with the radiology readings made during multidisciplinary practice, non-inferiority was not confirmed, as the AI system showed lower specificity (68·9% [95% CI 65·3-72·4] vs 69·0% [65·5-72·5]) at the same sensitivity (96·1%, 94·0-98·2) as the PI-RADS 3 or greater operating point. The lower boundary of the two-sided 95% Wald CI for the difference in specificity (-0·04) was greater than the non-inferiority margin (-0·05) and a p value below the significance threshold was reached (p<0·001). INTERPRETATION: An AI system was superior to radiologists using PI-RADS (2.1), on average, at detecting clinically significant prostate cancer and comparable to the standard of care. Such a system shows the potential to be a supportive tool within a primary diagnostic setting, with several associated benefits for patients and radiologists. Prospective validation is needed to test clinical applicab","author":[{"family":"Saha","given":"Anindo"},{"family":"Saha","given":"Anindo"},{"family":"Bosma","given":"Joeran"},{"family":"Twilt","given":"Jasper"},{"family":"Ginneken","given":"Bram"},{"family":"Bjartell","given":"Anders"},{"family":"Padhani","given":"Anwar"},{"family":"Bonekamp","given":"David"},{"family":"Villeirs","given":"Geert"},{"family":"Salomon","given":"Georg"},{"family":"Giannarini","given":"Gianluca"},{"family":"Kalpathycramer","given":"Jayashree"},{"family":"Barentsz","given":"Jelle"},{"family":"Maierhein","given":"Klaus"},{"family":"Rusu","given":"Mirabela"},{"family":"Rouvière","given":"Olivier"},{"family":"Bergh","given":"Roderick"},{"family":"Panebianco","given":"Valeria"},{"family":"Kasivisvanathan","given":"Veeru"},{"family":"Obuchowski","given":"Nancy"},{"family":"Yakar","given":"Derya"},{"family":"Elschot","given":"Mattijs"},{"family":"Veltman","given":"Jeroen"},{"family":"Fütterer","given":"Jurgen"},{"family":"Rooij","given":"Maarten"},{"family":"Huisman","given":"Henkjan"},{"family":"Saha","given":"Anindo"},{"family":"Saha","given":"Anindo"},{"family":"Bosma","given":"Joeran"},{"family":"Twilt","given":"Jasper"},{"family":"Ginneken","given":"Bram"},{"family":"Noordman","given":"Constant"},{"family":"Slootweg","given":"Ivan"},{"family":"Roest","given":"Christian"},{"family":"Fransen","given":"Stefan"},{"family":"Sunoqrot","given":"Mohammed"},{"family":"Bathen","given":"Tone"},{"family":"Rouw","given":"Dennis"},{"family":"Immerzeel","given":"Jos"},{"family":"Geerdink","given":"Jeroen"},{"family":"Run","given":"Chris"},{"family":"Groeneveld","given":"Miriam"},{"family":"Meakin","given":"James"},{"family":"Karagöz","given":"Ahmet"},{"family":"Bône","given":"Alexandre"},{"family":"Routier","given":"Alexandre"},{"family":"Marcoux","given":"Arnaud"},{"family":"Abi-Nader","given":"Clément"},{"family":"Li","given":"Cynthia"},{"family":"Feng","given":"Dagan"},{"family":"Alis","given":"Deniz"},{"family":"Karaarslan","given":"Ercan"},{"family":"Ahn","given":"Euijoon"},{"family":"Nicolas","given":"François"},{"family":"Sonn","given":"Geoffrey"},{"family":"Bhattacharya","given":"Indrani"},{"family":"Kim","given":"Jinman"},{"family":"Shi","given":"Jun"},{"family":"Jahanandish","given":"Hassan"},{"family":"An","given":"Hong"},{"family":"Kan","given":"Hongyu"},{"family":"Oksuz","given":"Ilkay"},{"family":"Qiao","given":"Liang"},{"family":"Rohé","given":"Marc"},{"family":"Yergin","given":"Mert"},{"family":"Khadra","given":"Mohamed"},{"family":"Şeker","given":"Mustafa"},{"family":"Kartal","given":"Mustafa"},{"family":"Debs","given":"Noëlie"},{"family":"Fan","given":"Richard"},{"family":"Saunders","given":"Sara"},{"family":"Soerensen","given":"Simon"},{"family":"Moroianu","given":"Stefania"},{"family":"Vesal","given":"Sulaiman"},{"family":"Yuan","given":"Yuan"},{"family":"Malakoti-Fard","given":"Afsoun"},{"family":"Mačiūnien","given":"Agnė"},{"family":"Kawashima","given":"Akira"},{"family":"Machadov","given":"Ana"},{"family":"Moreira","given":"Ana"},{"family":"Ponsiglione","given":"Andrea"},{"family":"Rappaport","given":"Annelies"},{"family":"Stanzione","given":"Arnaldo"},{"family":"Ciuvasovas","given":"Arturas"},{"family":"Turkbey","given":"Baris"},{"family":"Keyzer","given":"Bart"},{"family":"Pedersen","given":"Bodil"},{"family":"Eijlers","given":"Bram"},{"family":"Chen","given":"Christine"},{"family":"Riccardo","given":"Ciabattoni"},{"family":"Alis","given":"Deniz"},{"family":"Staal","given":"Ewout"},{"family":"Jäderling","given":"Fredrik"},{"family":"Langkilde","given":"Fredrik"},{"family":"Aringhieri","given":"Giacomo"},{"family":"Brembilla","given":"Giorgio"},{"family":"Son","given":"Hannah"},{"family":"Vanderlelij","given":"Hans"},{"family":"Raat","given":"Henricus"},{"family":"Pikūnienė","given":"Ingrida"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/s1470-2045(24)00220-1","URL":"https://doi.org/10.1016/s1470-2045(24)00220-1","source":"openalex"},{"id":"oa:W4389521069","type":"article-journal","title":"Medical students’ perceptions towards artificial intelligence in education and practice: A multinational, multicenter cross-sectional study","abstract":"Abstract Background Artificial intelligence (AI) is anticipated to fundamentally change the educational and professional landscape for the next generation of physicians, but its successful integration depends on the global perspectives of all stakeholders. Previous medical student surveys were limited by small sample sizes or geographic constraints, hindering a global comparison of perceptions. This study aims to explore current medical students’ attitudes towards AI in medical education and the profession on a broad, international scale and to examine regional differences in perspectives. Methods and Findings This international multicenter cross-sectional study developed and validated an anonymous online survey of 15 multiple-choice items to assess medical, dentistry, and veterinary students’ AI knowledge and attitudes toward the utilization of AI in healthcare, the current state of AI education, and regional differences in perspectives. Between April and October 2023, 4,313 medical, 205 dentistry, and 78 veterinary students from 192 faculties in 48 countries responded to the survey (average response rate: 0.2%, standard deviation: 0.4%). Most participants studied in European countries (N=2,350), followed by North/South America (N=1,070) and Asia (N=944). Students expressed predominantly positive attitudes towards the use of AI in healthcare (67.6%, N=3,091) and the desire for more AI teaching in their curricula (76.1%, N=3,474). However, they reported limited general knowledge of AI (75.3%, N=3,451), the absence of AI-related courses (76.3%, N=3,497), and felt inadequately prepared to use AI in their future careers (57.9%, N=2,652). The subgroup analyses revealed regional differences in perceptions, although predominantly with small effect sizes. The main limitations include the low response rate per institution, which was calculated on total enrollment across all degree programs, and the risk of selection bias. Conclusions This study highlights the favorable perceptions of international medical students towards incorporating AI in healthcare practice while emphasizing the importance of integrating AI teaching into medical education. Graphical abstract","author":[{"family":"Busch","given":"Felix"},{"family":"Hoffmann","given":"Lena"},{"family":"Truhn","given":"Daniel"},{"family":"Ortizprado","given":"Esteban"},{"family":"Makowski","given":"Marcus"},{"family":"Bressem","given":"Keno"},{"family":"Adams","given":"Lisa"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1101/2023.12.09.23299744","URL":"https://doi.org/10.1101/2023.12.09.23299744","source":"openalex"},{"id":"oa:W4381149654","type":"article-journal","title":"Artificial Intelligence in Andrology: From Semen Analysis to Image Diagnostics","abstract":"Artificial intelligence (AI) in medicine has gained a lot of momentum in the last decades and has been applied to various fields of medicine. Advances in computer science, medical informatics, robotics, and the need for personalized medicine have facilitated the role of AI in modern healthcare. Similarly, as in other fields, AI applications, such as machine learning, artificial neural networks, and deep learning, have shown great potential in andrology and reproductive medicine. AI-based tools are poised to become valuable assets with abilities to support and aid in diagnosing and treating male infertility, and in improving the accuracy of patient care. These automated, AI-based predictions may offer consistency and efficiency in terms of time and cost in infertility research and clinical management. In andrology and reproductive medicine, AI has been used for objective sperm, oocyte, and embryo selection, prediction of surgical outcomes, cost-effective assessment, development of robotic surgery, and clinical decision-making systems. In the future, better integration and implementation of AI into medicine will undoubtedly lead to pioneering evidence-based breakthroughs and the reshaping of andrology and reproductive medicine.","author":[{"family":"Ghayda","given":"Ramy"},{"family":"Cannarella","given":"Rossella"},{"family":"Calogero","given":"Aldo"},{"family":"Shah","given":"Rupin"},{"family":"Rambhatla","given":"Amarnath"},{"family":"Zohdy","given":"Wael"},{"family":"Kavoussi","given":"Parviz"},{"family":"Avidorreiss","given":"Tomer"},{"family":"Boitrelle","given":"Florence"},{"family":"Mostafa","given":"Taymour"},{"family":"Saleh","given":"Ramadan"},{"family":"Toprak","given":"Tuncay"},{"family":"Birowo","given":"Ponco"},{"family":"Salvio","given":"Gianmaria"},{"family":"Çalık","given":"Gökhan"},{"family":"Kuroda","given":"Shinnosuke"},{"family":"Kaiyal","given":"Raneen"},{"family":"Ziouziou","given":"Imad"},{"family":"Crafa","given":"Andrea"},{"family":"Phuoc","given":"Nguyen"},{"family":"Russo","given":"Giorgio"},{"family":"Durairajanayagam","given":"Damayanthi"},{"family":"Hashimi","given":"Manaf"},{"family":"Hamoda","given":"Taha"},{"family":"Pinggera","given":"Germar‐michael"},{"family":"Adriansjah","given":"Ricky"},{"family":"Rosas","given":"Israel"},{"family":"Arafa","given":"Mohamed"},{"family":"Chung","given":"Eric"},{"family":"Atmoko","given":"Widi"},{"family":"Rocco","given":"Lucia"},{"family":"Lin","given":"Haocheng"},{"family":"Huyghe","given":"É"},{"family":"Kothari","given":"Priyank"},{"family":"Vazquez","given":"Jesus"},{"family":"Dimitriadis","given":"Fotios"},{"family":"Garrido","given":"Nicolás"},{"family":"Homa","given":"Sheryl"},{"family":"Falcone","given":"Marco"},{"family":"Sabbaghian","given":"Marjan"},{"family":"Kandil","given":"Hussein"},{"family":"Ko","given":"Edmund"},{"family":"Martínez","given":"Marlon"},{"family":"Nguyen","given":"Quang"},{"family":"Harraz","given":"M"},{"family":"Şerefoğlu","given":"Ege"},{"family":"Karthikeyan","given":"Vilvapathy"},{"family":"Tien","given":"Dung"},{"family":"Jindal","given":"Sunil"},{"family":"Mičić","given":"S"},{"family":"Bellavia","given":"Marina"},{"family":"Alali","given":"Hamed"},{"family":"Gherabi","given":"Nazim"},{"family":"Lewis","given":"Sheena"},{"family":"Park","given":"Hyun"},{"family":"Simopoulou","given":"Mara"},{"family":"Sallam","given":"Hassan"},{"family":"Dominguez","given":"Liliana"},{"family":"Colpi","given":"Giovanni"},{"family":"Agarwal","given":"Ashok"},{"family":"Forum","given":"Global"}],"issued":{"date-parts":[[2023]]},"DOI":"10.5534/wjmh.230050","URL":"https://doi.org/10.5534/wjmh.230050","source":"openalex"},{"id":"oa:W4378715289","type":"article-journal","title":"Medical operational AI: artificial intelligence in routine medical operations","abstract":"Abstract Despite substantial gains facilitated by Artificial Intelligence (AI) in recent years, it has to be applied very cautiously in sensitive domains like medicine due to the lack of explainability of many methods in this field. We aim to provide a system to overcome these issues of medical AI applications by means of our concept of medical operational AI detailed in this paper. We make use of various methods of AI and utilize knowledge graphs in particular. The latter is continuously updated by medical experts based on medical literature such as peer-reviewed papers and standard online sources such as UpToDate. We thoroughly derive a multi-level system tackling the corresponding challenges. In particular, its design encompasses (i) holistic diagnostic assistance on a macro level, (ii) predicitions and detailed suggestions for specific medical domains on a micro level, as well as (iii) AI-based optimizations of the overall system on a meta level. We detail practical merits of medical operational AI and discuss the state of the art beyond our solution.","author":[{"family":"Berns","given":"Fabian"},{"family":"Heilig","given":"Niclas"},{"family":"Stumpe","given":"Florian"},{"family":"Kirchhoff","given":"Jan"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1515/labmed-2023-0011","URL":"https://doi.org/10.1515/labmed-2023-0011","source":"openalex"},{"id":"oa:W4361285006","type":"article-journal","title":"Explainable, Domain-Adaptive, and Federated Artificial Intelligence in Medicine","abstract":"Artificial intelligence (AI) continues to transform data analysis in many domains. Progress in each domain is driven by a growing body of annotated data, increased computational resources, and technological innovations. In medicine, the sensitivity of the data, the complexity of the tasks, the potentially high stakes, and a requirement of accountability give rise to a particular set of challenges. In this review, we focus on three key methodological approaches that address some of the particular challenges in AI-driven medical decision making. 1) Explainable AI aims to produce a human-interpretable justification for each output. Such models increase confidence if the results appear plausible and match the clinicians expectations. However, the absence of a plausible explanation does not imply an inaccurate model. Especially in highly non-linear, complex models that are tuned to maximize accuracy, such interpretable representations only reflect a small portion of the justification. 2) Domain adaptation and transfer learning enable AI models to be trained and applied across multiple domains. For example, a classification task based on images acquired on different acquisition hardware. 3) Federated learning enables learning large-scale models without exposing sensitive personal health information. Unlike centralized AI learning, where the centralized learning machine has access to the entire training data, the federated learning process iteratively updates models across multiple sites by exchanging only parameter updates, not personal health data. This narrative review covers the basic concepts, highlights relevant corner-stone and state-of-the-art research in the field, and discusses perspectives.","author":[{"family":"Chaddad","given":"Ahmad"},{"family":"Lu","given":"Qizong"},{"family":"Li","given":"Jiali"},{"family":"Katib","given":"Yousef"},{"family":"Kateb","given":"Reem"},{"family":"Tanougast","given":"Camel"},{"family":"Bouridane","given":"Ahmed"},{"family":"Abdulkadir","given":"Ahmed"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/jas.2023.123123","URL":"https://doi.org/10.1109/jas.2023.123123","source":"openalex"},{"id":"oa:W4321372215","type":"article-journal","title":"The Role of Artificial Intelligence in Echocardiography","abstract":"Echocardiography is an integral part of the diagnosis and management of cardiovascular disease. The use and application of artificial intelligence (AI) is a rapidly expanding field in medicine to improve consistency and reduce interobserver variability. AI can be successfully applied to echocardiography in addressing variance during image acquisition and interpretation. Furthermore, AI and machine learning can aid in the diagnosis and management of cardiovascular disease. In the realm of echocardiography, accurate interpretation is largely dependent on the subjective knowledge of the operator. Echocardiography is burdened by the high dependence on the level of experience of the operator, to a greater extent than other imaging modalities like computed tomography, nuclear imaging, and magnetic resonance imaging. AI technologies offer new opportunities for echocardiography to produce accurate, automated, and more consistent interpretations. This review discusses machine learning as a subfield within AI in relation to image interpretation and how machine learning can improve the diagnostic performance of echocardiography. This review also explores the published literature outlining the value of AI and its potential to improve patient care.","author":[{"family":"Barry","given":"Timothy"},{"family":"Farina","given":"Juan"},{"family":"Chao","given":"Chieh‐ju"},{"family":"Ayoub","given":"Chadi"},{"family":"Jeong","given":"Jiwoong"},{"family":"Patel","given":"Bhavik"},{"family":"Banerjee","given":"Imon"},{"family":"Arsanjani","given":"Reza"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/jimaging9020050","URL":"https://doi.org/10.3390/jimaging9020050","source":"openalex"},{"id":"oa:W4387348149","type":"article-journal","title":"A Review of Artificial Intelligence in Medical Prescription Analysis","abstract":"Accurate reading and comprehension of medical prescriptions are crucial for healthcare providers to ensure appropriate treatment for the patients. However, with the growing volume of prescriptions and increasingly complex medication regimens, errors can occur, which can result in severe consequences. To mitigate this issue, Artificial Intelligence (AI) can automate tasks such as medication identification, dosage calculation, and drug interaction checks, potentially improving the accuracy and efficiency of prescription analysis. This research study intends to examine the latest AI-based approaches for analyzing medical prescriptions in the healthcare industry, which can aid in the development of diagnostic systems capable of extracting patients' medical history, identifying incorrect medications and conducting statistical analyses on prescribed drugs. By analyzing the current state-of-the-art techniques in this field, one can identify areas of improvement to enhance prescription analysis accuracy and efficiency, ultimately benefiting patient care and medication management.","author":[{"family":"Agrawal","given":"Mayur"},{"family":"Mishra","given":"Vishwajeet"},{"family":"Jogani","given":"Hanesh"},{"family":"Kashyap","given":"Prerna"},{"family":"Mishra","given":"Raj"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/icpcsn58827.2023.00014","URL":"https://doi.org/10.1109/icpcsn58827.2023.00014","source":"openalex"},{"id":"oa:W4400126525","type":"article-journal","title":"The limits of fair medical imaging AI in real-world generalization","abstract":"As artificial intelligence (AI) rapidly approaches human-level performance in medical imaging, it is crucial that it does not exacerbate or propagate healthcare disparities. Previous research established AI's capacity to infer demographic data from chest X-rays, leading to a key concern: do models using demographic shortcuts have unfair predictions across subpopulations? In this study, we conducted a thorough investigation into the extent to which medical AI uses demographic encodings, focusing on potential fairness discrepancies within both in-distribution training sets and external test sets. Our analysis covers three key medical imaging disciplines-radiology, dermatology and ophthalmology-and incorporates data from six global chest X-ray datasets. We confirm that medical imaging AI leverages demographic shortcuts in disease classification. Although correcting shortcuts algorithmically effectively addresses fairness gaps to create 'locally optimal' models within the original data distribution, this optimality is not true in new test settings. Surprisingly, we found that models with less encoding of demographic attributes are often most 'globally optimal', exhibiting better fairness during model evaluation in new test environments. Our work establishes best practices for medical imaging models that maintain their performance and fairness in deployments beyond their initial training contexts, underscoring critical considerations for AI clinical deployments across populations and sites.","author":[{"family":"Yang","given":"Yuzhe"},{"family":"Zhang","given":"Haoran"},{"family":"Gichoya","given":"Judy"},{"family":"Katabi","given":"Dina"},{"family":"Ghassemi","given":"Marzyeh"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41591-024-03113-4","URL":"https://doi.org/10.1038/s41591-024-03113-4","source":"openalex"},{"id":"oa:W4315651387","type":"article-journal","title":"Artificial intelligence and inflammatory bowel disease: Where are we going?","abstract":"Inflammatory bowel diseases, namely ulcerative colitis and Crohn's disease, are chronic and relapsing conditions that pose a growing burden on healthcare systems worldwide. Because of their complex and partly unknown etiology and pathogenesis, the management of ulcerative colitis and Crohn's disease can prove challenging not only from a clinical point of view but also for resource optimization. Artificial intelligence, an umbrella term that encompasses any cognitive function developed by machines for learning or problem solving, and its subsets machine learning and deep learning are becoming ever more essential tools with a plethora of applications in most medical specialties. In this regard gastroenterology is no exception, and due to the importance of endoscopy and imaging numerous clinical studies have been gradually highlighting the relevant role that artificial intelligence has in inflammatory bowel diseases as well. The aim of this review was to summarize the most recent evidence on the use of artificial intelligence in inflammatory bowel diseases in various contexts such as diagnosis, follow-up, treatment, prognosis, cancer surveillance, data collection, and analysis. Moreover, insights into the potential further developments in this field and their effects on future clinical practice were discussed.","author":[{"family":"Rio","given":"Leonardo"},{"family":"Spadaccini","given":"Marco"},{"family":"Parigi","given":"Tommaso"},{"family":"Gabbiadini","given":"Roberto"},{"family":"Buono","given":"Arianna"},{"family":"Busacca","given":"Anita"},{"family":"Maselli","given":"Roberta"},{"family":"Fugazza","given":"Alessandro"},{"family":"Colombo","given":"Matteo"},{"family":"Carrara","given":"Silvia"},{"family":"Franchellucci","given":"Gianluca"},{"family":"Alfarone","given":"Ludovico"},{"family":"Facciorusso","given":"Antonio"},{"family":"Hassan","given":"Cesare"},{"family":"Repici","given":"Alessandro"},{"family":"Armuzzi","given":"Alessandro"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3748/wjg.v29.i3.508","URL":"https://doi.org/10.3748/wjg.v29.i3.508","source":"openalex"},{"id":"oa:W4392394652","type":"article-journal","title":"Generative Artificial Intelligence in Education: From Deceptive to Disruptive.","abstract":"Generative Artificial Intelligence (GenAI) has emerged as a promising technology that can create original content, such as text, images, and sound. The use of GenAI in educational settings is becoming increasingly popular and offers a range of opportunities and challenges. This special issue explores the management and integration of GenAI in educational settings, including the ethical considerations, best practices, and opportunities. The potential of GenAI in education is vast. By using algorithms and data, GenAI can create original content that can be used to augment traditional teaching methods, creating a more interactive and personalized learning experience. In addition, GenAI can be utilized as an assessment tool and for providing feedback to students using generated content. For instance, it can be used to create custom quizzes, generate essay prompts, or even grade essays. The use of GenAI as an assessment tool can reduce the workload of teachers and help students receive prompt feedback on their work. Incorporating GenAI in educational settings also poses challenges related to academic integrity. With availability of GenAI models, students can use them to study or complete their homework assignments, which can raise concerns about the authenticity and authorship of the delivered work. Therefore, it is important to ensure that academic standards are maintained, and the originality of the student's work is preserved. This issue highlights the need for implementing ethical practices in the use of GenAI models and ensuring that the technology is used to support and not replace the student's learning experience.","author":[{"family":"Alier","given":"Marc"},{"family":"Garcíapeñalvo","given":"Francisco"},{"family":"Camba","given":"Jorge"}],"issued":{"date-parts":[[2024]]},"DOI":"10.9781/ijimai.2024.02.011","URL":"https://doi.org/10.9781/ijimai.2024.02.011","source":"openalex"},{"id":"oa:W4404719075","type":"article-journal","title":"Artificial intelligence and pain medicine education: Benefits and pitfalls for the medical trainee","abstract":"OBJECTIVES: Artificial intelligence (AI) represents an exciting and evolving technology that is increasingly being utilized across pain medicine. Large language models (LLMs) are one type of AI that has become particularly popular. Currently, there is a paucity of literature analyzing the impact that AI may have on trainee education. As such, we sought to assess the benefits and pitfalls that AI may have on pain medicine trainee education. Given the rapidly increasing popularity of LLMs, we particularly assessed how these LLMs may promote and hinder trainee education through a pilot quality improvement project. MATERIALS AND METHODS: A comprehensive search of the existing literature regarding AI within medicine was performed to identify its potential benefits and pitfalls within pain medicine. The pilot project was approved by UPMC Quality Improvement Review Committee (#4547). Three of the most commonly utilized LLMs at the initiation of this pilot study - ChatGPT Plus, Google Bard, and Bing AI - were asked a series of multiple choice questions to evaluate their ability to assist in learner education within pain medicine. RESULTS: Potential benefits of AI within pain medicine trainee education include ease of use, imaging interpretation, procedural/surgical skills training, learner assessment, personalized learning experiences, ability to summarize vast amounts of knowledge, and preparation for the future of pain medicine. Potential pitfalls include discrepancies between AI devices and associated cost-differences, correlating radiographic findings to clinical significance, interpersonal/communication skills, educational disparities, bias/plagiarism/cheating concerns, lack of incorporation of private domain literature, and absence of training specifically for pain medicine education. Regarding the quality improvement project, ChatGPT Plus answered the highest percentage of all questions correctly (16/17). Lowest correctness scores by LLMs were in answering first-order questions, with Google Bard and Bing AI answering 4/9 and 3/9 first-order questions correctly, respectively. Qualitative evaluation of these LLM-provided explanations in answering second- and third-order questions revealed some reasoning inconsistencies (e.g., providing flawed information in selecting the correct answer). CONCLUSIONS: AI represents a continually evolving and promising modality to assist trainees pursuing a career in pain medicine. Still, limitations currently exist that may hinder their independent use in this setting. Future research exploring how AI may overcome these challenges is thus required. Until then, AI should be utilized as supplementary tool within pain medicine trainee education and with caution.","author":[{"family":"Glicksman","given":"Michael"},{"family":"Wang","given":"Sheri"},{"family":"Yellapragada","given":"Samir"},{"family":"Robinson","given":"Christopher"},{"family":"Orhurhu","given":"Vwaire"},{"family":"Emerick","given":"Trent"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/papr.13428","URL":"https://doi.org/10.1111/papr.13428","source":"openalex"},{"id":"oa:W4384563806","type":"article-journal","title":"Artificial intelligence, nutrition, and ethical issues: A mini-review","abstract":"Background and aims Artificial intelligence (AI) has expanded applications in both medicine and biomedical sciences, focusing on medical diagnosis, risk prediction of disease onset, support of therapeutic techniques, and other subjects. In parallel, several applications in nutrition have been developed, such as microbiota/genes-diet interactions, investigation of diet-disease relationships, chatbots for lifestyle intervention, dietary assessment with food photographs, and food composition applications. Methods The positive aspects and ethical concerns were analyzed regarding the use of AI in nutrition. Results In general, AI should do no harm and contribute to human well-being, while AI professionals should be honest, trustful, and fair. Privacy and confidentiality should be protected. Other concerns include the \"dehumanization\" of care, social disparities, responsibility assignment in case of errors or malfunctions, and bias in training models and delivering care. Moreover, the prediction of disease onset in high-risk individuals may be connected to stigma, over-medicalization, and stress. There is also a serious concern that AI systems in the field of nutrition and dietetics may cause a partial replacement of dietitians; however, health professionals can use such technologies as part of their work. The use of AI applications for persons with mental diseases or eating disorders is also crucial. Conclusion In conclusion, AI-assisted personalized nutrition needs further justification, while the regulatory framework requires a constant update to keep up with scientific advancements and address ethical issues. Coordinated global focus on these domains could effectively lead to a more embraced AI implementation in individuals and populations.","author":[{"family":"Detopoulou","given":"Paraskevi"},{"family":"Voulgaridou","given":"Gavriela"},{"family":"Moschos","given":"Panagiotis"},{"family":"Levidi","given":"Despoina"},{"family":"Anastasiou","given":"Thelma"},{"family":"Dedes","given":"Vasilios"},{"family":"Diplari","given":"Eirini"},{"family":"Fourfouri","given":"Nikoleta"},{"family":"Giaginis","given":"Constantinos"},{"family":"Panoutsopoulos","given":"Georgios"},{"family":"Papadopoulou","given":"Sousana"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.nutos.2023.07.001","URL":"https://doi.org/10.1016/j.nutos.2023.07.001","source":"openalex"},{"id":"oa:W4392555240","type":"article-journal","title":"The path from task-specific to general purpose artificial intelligence for medical diagnostics: A bibliometric analysis","abstract":"Artificial intelligence (AI) has revolutionized many fields, and its potential in healthcare has been increasingly recognized. Based on diverse data sources such as imaging, laboratory tests, medical records, and electrophysiological data, diagnostic AI has witnessed rapid development in recent years. A comprehensive understanding of the development status, contributing factors, and their relationships in the application of AI to medical diagnostics is essential to further promote its use in clinical practice. In this study, we conducted a bibliometric analysis to explore the evolution of task-specific to general-purpose AI for medical diagnostics. We used the Web of Science database to search for relevant articles published between 2010 and 2023, and applied VOSviewer, the R package Bibliometrix, and CiteSpace to analyze collaborative networks and keywords. Our analysis revealed that the field of AI in medical diagnostics has experienced rapid growth in recent years, with a focus on tasks such as image analysis, disease prediction, and decision support. Collaborative networks were observed among researchers and institutions, indicating a trend of global cooperation in this field. Additionally, we identified several key factors contributing to the development of AI in medical diagnostics, including data quality, algorithm design, and computational power. Challenges to progress in the field include model explainability, robustness, and equality, which will require multi-stakeholder, interdisciplinary collaboration to tackle. Our study provides a holistic understanding of the path from task-specific, mono-modal AI toward general-purpose, multimodal AI for medical diagnostics. With the continuous improvement of AI technology and the accumulation of medical data, we believe that AI will play a greater role in medical diagnostics in the future.","author":[{"family":"Chang","given":"Chuheng"},{"family":"Shi","given":"Wen"},{"family":"Wang","given":"Youyang"},{"family":"Zhang","given":"Zhan"},{"family":"Huang","given":"Xiaoming"},{"family":"Jiao","given":"Yang"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.compbiomed.2024.108258","URL":"https://doi.org/10.1016/j.compbiomed.2024.108258","source":"openalex"},{"id":"oa:W4321436564","type":"article-journal","title":"Artificial intelligence chatbots will revolutionize how cancer patients access information: ChatGPT represents a paradigm-shift","abstract":"On November 30, 2022, OpenAI enabled public access to ChatGPT, a next-generation artificial intelligence with a highly sophisticated ability to write, solve coding issues, and answer questions. This communication draws attention to the prospect that ChatGPT and its successors will become important virtual assistants to patients and health-care providers. In our assessments, ranging from answering basic fact-based questions to responding to complex clinical questions, ChatGPT demonstrated a remarkable ability to formulate interpretable responses, which appeared to minimize the likelihood of alarm compared with Google's feature snippet. Arguably, the ChatGPT use case presents an urgent need for regulators and health-care professionals to be involved in developing standards for minimum quality and to raise patient awareness of current limitations of emerging artificial intelligence assistants. This commentary aims to raise awareness at the tipping point of a paradigm shift.","author":[{"family":"Hopkins","given":"Ashley"},{"family":"Logan","given":"Jessica"},{"family":"Kichenadasse","given":"Ganessan"},{"family":"Sorich","given":"Michael"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1093/jncics/pkad010","URL":"https://doi.org/10.1093/jncics/pkad010","source":"openalex"},{"id":"oa:W4323567402","type":"article-journal","title":"Application of artificial intelligence in diagnosis and treatment of colorectal cancer: A novel Prospect","abstract":"In the past few decades, according to the rapid development of information technology, artificial intelligence (AI) has also made significant progress in the medical field. Colorectal cancer (CRC) is the third most diagnosed cancer worldwide, and its incidence and mortality rates are increasing yearly, especially in developing countries. This article reviews the latest progress in AI in diagnosing and treating CRC based on a systematic collection of previous literature. Most CRCs transform from polyp mutations. The computer-aided detection systems can significantly improve the polyp and adenoma detection rate by early colonoscopy screening, thereby lowering the possibility of mutating into CRC. Machine learning and bioinformatics analysis can help screen and identify more CRC biomarkers to provide the basis for non-invasive screening. The Convolutional neural networks can assist in reading histopathologic tissue images, reducing the experience difference among doctors. Various studies have shown that AI-based high-level auxiliary diagnostic systems can significantly improve the readability of medical images and help clinicians make more accurate diagnostic and therapeutic decisions. Moreover, Robotic surgery systems such as da Vinci have been more and more commonly used to treat CRC patients, according to their precise operating performance. The application of AI in neoadjuvant chemoradiotherapy has further improved the treatment and efficacy evaluation of CRC. In addition, AI represented by deep learning in gene sequencing research offers a new treatment option. All of these things have seen that AI has a promising prospect in the era of precision medicine.","author":[{"family":"Yin","given":"Zugang"},{"family":"Yao","given":"Chenhui"},{"family":"Zhang","given":"Limin"},{"family":"Qi","given":"Shaohua"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3389/fmed.2023.1128084","URL":"https://doi.org/10.3389/fmed.2023.1128084","source":"openalex"},{"id":"oa:W4385423139","type":"article-journal","title":"Artificial intelligence for Sustainable Development Goals : Bibliometric patterns and concept evolution trajectories","abstract":"Abstract The development of artificial intelligence (AI) as a field has impacted almost all aspects of human life. More recently it has found a role in addressing developmental challenges, specifically the Sustainable Development Goals (SDGs). However, there are not enough systematic studies on analysis of the role of AI research towards the SDGs. Therefore, this article attempts to bridge this gap by identifying the major bibliometric trends and concept‐evolution trajectories in the area of AI applications for sustainable‐development goals. The research publication data for the last 20 years in the areas of artificial intelligence, machine learning, deep learning, and so forth, is obtained and computationally analysed using a framework comprising bibliometrics, path analysis and content analysis. The findings show an incremental trend in overall publications on the application of AI for SDGs across the different regions of the world. SDGs 3 (good health & well‐being) and 7 (affordable and clean energy) are found as the areas with the most applications of AI. In SDG3, the literature reflects application of AI techniques such as deep learning for precision and personalised medicine while in SDG7, a number of studies have employed AI techniques for the integration of systems for efficient generation of solar power and improving the energy efficiency of a building. Furthermore, SDG 4 (quality education), SDG 13 (climate action), SDG 11 (sustainable cities and communities) and SDG 16 (peace, justice and strong institutions) are the other SDGs where AI approaches and techniques are applied. The analytical results present a detailed insight of application of AI for achieving the SDGs.","author":[{"family":"Singh","given":"Aakash"},{"family":"Kanaujia","given":"Anurag"},{"family":"Singh","given":"Vivek"},{"family":"Vinuesa","given":"Ricardo"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/sd.2706","URL":"https://doi.org/10.1002/sd.2706","source":"openalex"},{"id":"oa:W4380047614","type":"article-journal","title":"Exploring the Intersection of Artificial Intelligence and Clinical Healthcare: A Multidisciplinary Review","abstract":"Artificial intelligence (AI) plays a more and more important role in our everyday life due to the advantages that it brings when used, such as 24/7 availability, a very low percentage of errors, ability to provide real time insights, or performing a fast analysis. AI is increasingly being used in clinical medical and dental healthcare analyses, with valuable applications, which include disease diagnosis, risk assessment, treatment planning, and drug discovery. This paper presents a narrative literature review of AI use in healthcare from a multi-disciplinary perspective, specifically in the cardiology, allergology, endocrinology, and dental fields. The paper highlights data from recent research and development efforts in AI for healthcare, as well as challenges and limitations associated with AI implementation, such as data privacy and security considerations, along with ethical and legal concerns. The regulation of responsible design, development, and use of AI in healthcare is still in early stages due to the rapid evolution of the field. However, it is our duty to carefully consider the ethical implications of implementing AI and to respond appropriately. With the potential to reshape healthcare delivery and enhance patient outcomes, AI systems continue to reveal their capabilities.","author":[{"family":"Stafie","given":"Celina"},{"family":"Șufaru","given":"Irina"},{"family":"Ghiciuc","given":"Cristina"},{"family":"Stafie","given":"Ingrid"},{"family":"Sufaru","given":"Eduard"},{"family":"Solomon","given":"Sorina"},{"family":"Hăncianu","given":"Monica"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/diagnostics13121995","URL":"https://doi.org/10.3390/diagnostics13121995","source":"openalex"},{"id":"oa:W4395026179","type":"article-journal","title":"Ethical and regulatory challenges of large language models in medicine","abstract":"With the rapid growth of interest in and use of large language models (LLMs) across various industries, we are facing some crucial and profound ethical concerns, especially in the medical field. The unique technical architecture and purported emergent abilities of LLMs differentiate them substantially from other artificial intelligence (AI) models and natural language processing techniques used, necessitating a nuanced understanding of LLM ethics. In this Viewpoint, we highlight ethical concerns stemming from the perspectives of users, developers, and regulators, notably focusing on data privacy and rights of use, data provenance, intellectual property contamination, and broad applications and plasticity of LLMs. A comprehensive framework and mitigating strategies will be imperative for the responsible integration of LLMs into medical practice, ensuring alignment with ethical principles and safeguarding against potential societal risks.","author":[{"family":"Ong","given":"Jasmine"},{"family":"Chang","given":"Yin‐hsi"},{"family":"Wasswa","given":"William"},{"family":"Butte","given":"Atul"},{"family":"Shah","given":"Nigam"},{"family":"Chew","given":"Lita"},{"family":"Liu","given":"Nan"},{"family":"Doshivelez","given":"Finale"},{"family":"Lü","given":"Wei"},{"family":"Savulescu","given":"Julian"},{"family":"Ting","given":"Daniel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/s2589-7500(24)00061-x","URL":"https://doi.org/10.1016/s2589-7500(24)00061-x","source":"openalex"},{"id":"oa:W4316253891","type":"article-journal","title":"Analysing the Impact of Artificial Intelligence and Computational Sciences on Student Performance: Systematic Review and Meta-analysis","abstract":"Abstract Artificial intelligence (AI) and computational sciences have aroused a growing interest in education. Despite its relatively recent history, AI is increasingly being introduced into the classroom through different modalities, with the aim of improving student achievement. Thus, the purpose of the research is to analyse, quantitatively and qualitatively, the impact of AI components and computational sciences on student performance. For this purpose, a systematic review and meta-analysis have been carried out in WOS and Scopus databases. After applying the inclusion and exclusion criteria, the sample was set at 25 articles. The results support the positive impact that AI and computational sciences have on student performance, finding a rise in their attitude towards learning and their motivation, especially in the STEM (Science, Technology, Engineering, and Mathematics) areas. Despite the multiple benefits provided, the implementation of these technologies in instructional processes involves a great educational and ethical challenge for teachers in relation to their design and implementation, which requires further analysis from the educational research. These findings are consistent at all educational stages.","author":[{"family":"Garcíamartínez","given":"Inmaculada"},{"family":"Fernándezbatanero","given":"José"},{"family":"Cerero","given":"José"},{"family":"León","given":"Samuel"}],"issued":{"date-parts":[[2023]]},"DOI":"10.7821/naer.2023.1.1240","URL":"https://doi.org/10.7821/naer.2023.1.1240","source":"openalex"},{"id":"oa:W4367367475","type":"manuscript","title":"Towards Medical Artificial General Intelligence via Knowledge-Enhanced Multimodal Pretraining","abstract":"Medical artificial general intelligence (MAGI) enables one foundation model to solve different medical tasks, which is very practical in the medical domain. It can significantly reduce the requirement of large amounts of task-specific data by sufficiently sharing medical knowledge among different tasks. However, due to the challenges of designing strongly generalizable models with limited and complex medical data, most existing approaches tend to develop task-specific models. To take a step towards MAGI, we propose a new paradigm called Medical-knOwledge-enhanced mulTimOdal pretRaining (MOTOR). In MOTOR, we combine two kinds of basic medical knowledge, i.e., general and specific knowledge, in a complementary manner to boost the general pretraining process. As a result, the foundation model with comprehensive basic knowledge can learn compact representations from pretraining radiographic data for better cross-modal alignment. MOTOR unifies the understanding and generation, which are two kinds of core intelligence of an AI system, into a single medical foundation model, to flexibly handle more diverse medical tasks. To enable a comprehensive evaluation and facilitate further research, we construct a medical multimodal benchmark including a wide range of downstream tasks, such as chest x-ray report generation and medical visual question answering. Extensive experiments on our benchmark show that MOTOR obtains promising results through simple task-oriented adaptation. The visualization shows that the injected knowledge successfully highlights key information in the medical data, demonstrating the excellent interpretability of MOTOR. Our MOTOR successfully mimics the human practice of fulfilling a \"medical student\" to accelerate the process of becoming a \"specialist\". We believe that our work makes a significant stride in realizing MAGI.","author":[{"family":"Lin","given":"Bingqian"},{"family":"Chen","given":"Zicong"},{"family":"Li","given":"Mingjie"},{"family":"Lin","given":"Haokun"},{"family":"Xu","given":"Hang"},{"family":"Zhu","given":"Yi"},{"family":"Liu","given":"Jianzhuang"},{"family":"Cai","given":"Wenjia"},{"family":"Yang","given":"Lei"},{"family":"Zhao","given":"Shen"},{"family":"Wu","given":"Chenfei"},{"family":"Chen","given":"Ling"},{"family":"Chang","given":"Xiaojun"},{"family":"Yang","given":"Yi"},{"family":"Xing","given":"Lei"},{"family":"Liang","given":"Xiaodan"}],"issued":{"date-parts":[[2023]]},"DOI":"10.48550/arxiv.2304.14204","URL":"https://doi.org/10.48550/arxiv.2304.14204","source":"openalex"},{"id":"oa:W4390719454","type":"article-journal","title":"Warning: Artificial intelligence chatbots can generate inaccurate medical and scientific information and references","abstract":"The use of generative artificial intelligence (AI) chatbots, such as ChatGPT and YouChat, has increased enormously since their release in late 2022. Concerns have been raised over the potential of chatbots to facilitate cheating in education settings, including essay writing and exams. In addition, multiple publishers have updated their editorial policies to prohibit chatbot authorship on publications. This article highlights another potentially concerning issue; the strong propensity of chatbots in response to queries requesting medical and scientific information and its underlying references, to generate plausible looking but inaccurate responses, with the chatbots also generating nonexistent citations. As an example, a number of queries were generated and, using two popular chatbots, demonstrated that both generated inaccurate outputs. The authors thus urge extreme caution, because unwitting application of inconsistent and potentially inaccurate medical information could have adverse outcomes.","author":[{"family":"Clelland","given":"Catherine"},{"family":"Moss","given":"Stuart"},{"family":"Clelland","given":"James"}],"issued":{"date-parts":[[2024]]},"DOI":"10.37349/edht.2024.00006","URL":"https://doi.org/10.37349/edht.2024.00006","source":"openalex"},{"id":"oa:W4389286730","type":"article-journal","title":"ARTIFICIAL INTELLIGENCE IN DEVELOPING COUNTRIES: BRIDGING THE GAP BETWEEN POTENTIAL AND IMPLEMENTATION","abstract":"This paper examines the role of Artificial Intelligence (AI) in developing countries, focusing on bridging the gap between its vast potential and effective implementation. As AI technologies advance globally, their impact on socio-economic development becomes increasingly critical, particularly in regions with diverse challenges and opportunities. The study investigates the current landscape of AI adoption in developing countries, analyzing the potential benefits, challenges, and ethical considerations. Through a comprehensive review of literature and case studies, the paper explores strategies and solutions for harnessing AI's transformative power in diverse sectors such as healthcare, agriculture, and education. The findings emphasize the importance of capacity building, public-private partnerships, and tailored policy frameworks to address infrastructure limitations and skill gaps. The research contributes to a nuanced understanding of the opportunities and complexities surrounding AI implementation in developing countries, providing insights for policymakers, practitioners, and scholars seeking to navigate this evolving technological landscape Keywords: Artificial Intelligence; Global Connectivity; Emerging Technologies; Organizational Resilience; Sustainable Growth; Developing Country.","author":[{"family":"Aderibigbe","given":"Adebayo"},{"family":"Ohenhen","given":"Peter"},{"family":"Nwaobia","given":"Nwabueze"},{"family":"Gidiagba","given":"Joachim"},{"family":"Ani","given":"Emmanuel"}],"issued":{"date-parts":[[2023]]},"DOI":"10.51594/csitrj.v4i3.629","URL":"https://doi.org/10.51594/csitrj.v4i3.629","source":"openalex"},{"id":"oa:W4396833063","type":"article-journal","title":"Explainable Notes: Examining How to Unlock Meaning in Medical Notes with Interactivity and Artificial Intelligence","abstract":"Medical progress notes have recently become available to patients at an unprecedented scale. Progress notes offer patients insight into their care that they cannot find elsewhere. That said, reading a note requires patients to contend with the language, unspoken assumptions, and clutter common to clinical documentation. As the health system reinvents many of its interfaces to incorporate AI assistance, this paper examines what intelligent interfaces could do to help patients read their progress notes. In a qualitative study, we examine the needs of patients as they read a progress note. We then formulate a vision for the explainable note, an augmented progress note that provides support for directing attention, phrase-level understanding, and tracing lines of reasoning. This vision manifests in a set of patient-inspired opportunities for advancing intelligent interfaces for writing and reading progress notes.","author":[{"family":"Kambhamettu","given":"Hita"},{"family":"Metaxa","given":"Danaë"},{"family":"Johnson","given":"Kevin"},{"family":"Head","given":"Andrew"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1145/3613904.3642573","URL":"https://doi.org/10.1145/3613904.3642573","source":"openalex"},{"id":"oa:W4394832225","type":"article-journal","title":"Transparent medical image AI via an image–text foundation model grounded in medical literature","abstract":"Building trustworthy and transparent image-based medical artificial intelligence (AI) systems requires the ability to interrogate data and models at all stages of the development pipeline, from training models to post-deployment monitoring. Ideally, the data and associated AI systems could be described using terms already familiar to physicians, but this requires medical datasets densely annotated with semantically meaningful concepts. In the present study, we present a foundation model approach, named MONET (medical concept retriever), which learns how to connect medical images with text and densely scores images on concept presence to enable important tasks in medical AI development and deployment such as data auditing, model auditing and model interpretation. Dermatology provides a demanding use case for the versatility of MONET, due to the heterogeneity in diseases, skin tones and imaging modalities. We trained MONET based on 105,550 dermatological images paired with natural language descriptions from a large collection of medical literature. MONET can accurately annotate concepts across dermatology images as verified by board-certified dermatologists, competitively with supervised models built on previously concept-annotated dermatology datasets of clinical images. We demonstrate how MONET enables AI transparency across the entire AI system development pipeline, from building inherently interpretable models to dataset and model auditing, including a case study dissecting the results of an AI clinical trial. By learning to pair dermatological images and related concepts in a self-supervised manner, a visual-language foundation model is shown to have comparable performance to supervised models for concept annotation and is used to scrutinize model decisions for enhanced interpretability and accountability of medical imaging applications.","author":[{"family":"Kim","given":"Chanwoo"},{"family":"Gadgil","given":"Soham"},{"family":"Degrave","given":"Alex"},{"family":"Omiye","given":"Jesutofunmi"},{"family":"Cai","given":"Zhuo"},{"family":"Daneshjou","given":"Roxana"},{"family":"Lee","given":"Su‐in"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41591-024-02887-x","URL":"https://doi.org/10.1038/s41591-024-02887-x","source":"openalex"},{"id":"oa:W4388145633","type":"article-journal","title":"Federated Learning for Medical Applications: A Taxonomy, Current Trends, Challenges, and Future Research Directions","abstract":"With the advent of the Internet of Things (IoT), artificial intelligence (AI), machine learning (ML), and deep learning (DL) algorithms, the landscape of data-driven medical applications has emerged as a promising avenue for designing robust and scalable diagnostic and prognostic models from medical data. This has gained a lot of attention from both academia and industry, leading to significant improvements in healthcare quality. However, the adoption of AI-driven medical applications still faces tough challenges, including meeting security, privacy, and Quality-of-Service (QoS) standards. Recent developments in federated learning (FL) have made it possible to train complex machine-learned models in a distributed manner and have become an active research domain, particularly processing the medical data at the edge of the network in a decentralized way to preserve privacy and address security concerns. To this end, in this article, we explore the present and future of FL technology in medical applications where data sharing is a significant challenge. We delve into the current research trends and their outcomes, unraveling the complexities of designing reliable and scalable FL models. This article outlines the fundamental statistical issues in FL, tackles device-related problems, addresses security challenges, and navigates the complexity of privacy concerns, all while highlighting its transformative potential in the medical field. Our study primarily focuses on medical applications of FL, particularly in the context of global cancer diagnosis. We highlight the potential of FL to enable computer-aided diagnosis tools that address this challenge with greater effectiveness than traditional data-driven methods. Recent literature has shown that FL models are robust and generalize well to new data, which is essential for medical applications. We hope that this comprehensive review will serve as a checkpoint for the field, summarizing the current state of the art and identifying open problems and future research directions.","author":[{"family":"Rauniyar","given":"Ashish"},{"family":"Hagos","given":"Desta"},{"family":"Jha","given":"Debesh"},{"family":"Håkegård","given":"Jan"},{"family":"Bağcı","given":"Ulaş"},{"family":"Rawat","given":"Danda"},{"family":"Vlassov","given":"Vladimir"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/jiot.2023.3329061","URL":"https://doi.org/10.1109/jiot.2023.3329061","source":"openalex"},{"id":"oa:W4389253133","type":"article-journal","title":"Comparison of artificial intelligence-assisted informed consent obtained before coronary angiography with the conventional method: Medical competence and ethical assessment","abstract":"Objective: At the time of informed consent (IC) for coronary angiography (CAG), patients' knowledge of the process is inadequate. Time constraints and a lack of personalization of consent are the primary causes of inadequate information. This procedure can be enhanced by obtaining IC using a chatbot powered by artificial intelligence (AI). Methods: In the study, patients who will undergo CAG for the first time were randomly divided into two groups, and IC was given to one group using the conventional method and the other group using an AI-supported chatbot, chatGPT3. They were then evaluated with two distinct questionnaires measuring their satisfaction and capacity to understand CAG risks. Results: = 0.581), the correct understanding of CAG risk questionnaire was found to be significantly higher in the AI group (<0.001). Conclusions: AI can be trained to support clinicians in giving IC before CAG. In this way, the workload of healthcare professionals can be reduced while providing a better IC.","author":[{"family":"Aydın","given":"Fatih"},{"family":"Yıldırım","given":"Özge"},{"family":"Aydın","given":"Ayşe"},{"family":"Murat","given":"Bektas"},{"family":"Başaran","given":"Cem"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1177/20552076231218141","URL":"https://doi.org/10.1177/20552076231218141","source":"openalex"},{"id":"oa:W4405849905","type":"article-journal","title":"Race to the Moon or the Bottom? Applications, Performance, and Ethical Considerations of Artificial Intelligence in Prosthodontics and Implant Dentistry","abstract":"Objectives: This review aims to explore the applications of artificial intelligence (AI) in prosthodontics and implant dentistry, focusing on its performance outcomes and associated ethical concerns. Materials and Methods: Following the PRISMA guidelines, a search was conducted across databases such as PubMed, Medline, Web of Science, and Scopus. Studies published between January 2022 and May 2024, in English, were considered. The Population (P) included patients or extracted teeth with AI applications in prosthodontics and implant dentistry; the Intervention (I) was AI-based tools; the Comparison (C) was traditional methods, and the Outcome (O) involved AI performance outcomes and ethical considerations. The Newcastle–Ottawa Scale was used to assess the quality and risk of bias in the studies. Results: Out of 3420 initially identified articles, 18 met the inclusion criteria for AI applications in prosthodontics and implant dentistry. The review highlighted AI’s significant role in improving diagnostic accuracy, treatment planning, and prosthesis design. AI models demonstrated high accuracy in classifying dental implants and predicting implant outcomes, although limitations were noted in data diversity and model generalizability. Regarding ethical issues, five studies identified concerns such as data privacy, system bias, and the potential replacement of human roles by AI. While patients generally viewed AI positively, dental professionals expressed hesitancy due to a lack of familiarity and regulatory guidelines, highlighting the need for better education and ethical frameworks. Conclusions: AI has the potential to revolutionize prosthodontics and implant dentistry by enhancing treatment accuracy and efficiency. However, there is a pressing need to address ethical issues through comprehensive training and the development of regulatory frameworks. Future research should focus on broadening AI applications and addressing the identified ethical concerns.","author":[{"family":"Alfaraj","given":"Amal"},{"family":"Nagai","given":"Toshiki"},{"family":"Alqallaf","given":"Hawra"},{"family":"Lin","given":"Wei‐shao"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/dj13010013","URL":"https://doi.org/10.3390/dj13010013","source":"openalex"},{"id":"oa:W4401961962","type":"article-journal","title":"Accuracy and Readability of Artificial Intelligence Chatbot Responses to Vasectomy-Related Questions: Public Beware","abstract":"Purpose Artificial intelligence (AI) has rapidly gained popularity with the growth of ChatGPT (OpenAI, San Francisco, USA) and other large-language model chatbots, and these programs have tremendous potential to impact medicine. One important area of consequence in medicine and public health is that patients may use these programs in search of answers to medical questions. Despite the increased utilization of AI chatbots by the public, there is little research to assess the reliability of ChatGPT and alternative programs when queried for medical information. This study seeks to elucidate the accuracy and readability of AI chatbots in answering patient questions regarding urology. As vasectomy is one of the most common urologic procedures, this study investigates AI-generated responses to frequently asked vasectomy-related questions. For this study, five popular and free-to-access AI platforms were utilized to undertake this investigation. Methods Fifteen vasectomy-related questions were individually queried to five AI chatbots from November-December 2023: ChatGPT (OpenAI, San Francisco, USA), Bard (Google Inc., Mountainview, USA) Bing (Microsoft, Redmond, USA) Perplexity (Perplexity AI Inc., San Francisco, USA), and Claude (Anthropic, San Francisco, USA). Responses from each platform were graded by two attending urologists, two urology research faculty, and one urological resident physician using a Likert (1-6) scale: (1-completely inaccurate, 6-completely accurate) based on comparison to existing American Urological Association guidelines. Flesch-Kincaid Grade levels (FKGL) and Flesch Reading Ease scores (FRES) (1-100) were calculated for each response. To assess differences in Likert, FRES, and FKGL, Kruskal-Wallis tests were performed using GraphPad Prism V10.1.0 (GraphPad, San Diego, USA) with Alpha set at 0.05. Results Analysis shows that ChatGPT provided the most accurate responses across the five AI chatbots with an average score of 5.04 on the Likert scale. Subsequently, Microsoft Bing (4.91), Anthropic Claude (4.65), Google Bard (4.43), and Perplexity (4.41) followed. All five chatbots were found to score, on average, higher than 4.41 corresponding to a score of at least \"somewhat accurate.\" Google Bard received the highest Flesch Reading Ease score (49.67) and lowest Grade level (10.1) when compared to the other chatbots. Anthropic Claude scored 46.7 on the FRES and 10.55 on the FKGL. Microsoft Bing scored 45.57 on the FRES and 11.56 on the FKGL. Perplexity scored 36.4 on the FRES and 13.29 on the FKGL. ChatGPT had the lowest FRES of 30.4 and highest FKGL of 14.2. Conclusion This study investigates the use of AI in medicine, specifically urology, and it helps to determine whether large-language model chatbots can be reliable sources of freely available medical information. All five AI chatbots on average were able to achieve at least \"somewhat accurate\" on a 6-point Likert scale. In terms of readability, all five AI chatbots on average had Flesch Reading Ease scores of less than 50 and were higher than a 10th-grade level. In this small-scale study, there were several significant differences identified between the readability scores of each AI chatbot. However, there were no significant differences found among their accuracies. Thus, our study suggests that major AI chatbots may perform similarly in their ability to be correct but differ in their ease of being comprehended by the general public.","author":[{"family":"Carlson","given":"Jonathan"},{"family":"Cheng","given":"Robin"},{"family":"Lange","given":"Alyssa"},{"family":"Nagalakshmi","given":"Nadiminty"},{"family":"Rabets","given":"John"},{"family":"Shah","given":"Tariq"},{"family":"Sindhwani","given":"Puneet"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.67996","URL":"https://doi.org/10.7759/cureus.67996","source":"openalex"},{"id":"oa:W4400823820","type":"article-journal","title":"Random Forest Algorithm Overview","abstract":"A random forest is a machine learning model utilized in classification and forecasting. To train machine learning algorithms and artificial intelligence models, it is crucial to have a substantial amount of high-quality data for effective data collecting. System performance data is essential for refining algorithms, enhancing the efficiency of software and hardware, evaluating user be-havior, enabling pattern identification, decision-making, predictive modeling, and problem-solving, ultimately resulting in improved effectiveness and accuracy. The integration of diverse data collecting and processing methods enhances precision and innovation in problem-solving. Utilizing diverse methodologies in interdisciplinary research streamlines the research process, fosters innovation, and enables the application of data analysis findings to pattern recognition, decision-making, predictive modeling, and problem-solving. This approach also encourages in-novation in interdisciplinary research. This technique utilizes the concept of decision trees, con-structing a collection of decision trees and aggregating their outcomes to generate the ultimate prediction. Every decision tree inside a random forest is constructed using random subsets of data, and each individual tree is trained on a portion of the whole dataset. Subsequently, the outcomes of all decision trees are amalgamated to derive the ultimate forecast. One of the bene-fits of random forests is their capacity to handle unbalanced data and variables with missing values. Additionally, it mitigates the issue of arbitrary variable selection seen by certain alterna-tive models. Furthermore, random forests mitigate the issue of overfitting by training several de-cision trees on random subsets of data, hence enhancing their ability to generalize to novel data. Random forests are highly regarded as one of the most efficient and potent techniques in the domain of machine learning. They find extensive use in various applications such as automatic categorization, data forecasting, and supervisory learning.","author":[{"family":"Salman","given":"Hasan"},{"family":"Kalakech","given":"Ali"},{"family":"Steiti","given":"Amani"}],"issued":{"date-parts":[[2024]]},"DOI":"10.58496/bjml/2024/007","URL":"https://doi.org/10.58496/bjml/2024/007","source":"openalex"},{"id":"oa:W4400321274","type":"article-journal","title":"Medical Device-Associated Infections Caused by Biofilm-Forming Microbial Pathogens and Controlling Strategies","abstract":"Hospital-acquired infections, also known as nosocomial infections, include bloodstream infections, surgical site infections, skin and soft tissue infections, respiratory tract infections, and urinary tract infections. According to reports, Gram-positive and Gram-negative pathogenic bacteria account for up to 70% of nosocomial infections in intensive care unit (ICU) patients. Biofilm production is a main virulence mechanism and a distinguishing feature of bacterial pathogens. Most bacterial pathogens develop biofilms at the solid-liquid and air-liquid interfaces. An essential requirement for biofilm production is the presence of a conditioning film. A conditioning film provides the first surface on which bacteria can adhere and fosters the growth of biofilms by creating a favorable environment. The conditioning film improves microbial adherence by delivering chemical signals or generating microenvironments. Microorganisms use this coating as a nutrient source. The film gathers both inorganic and organic substances from its surroundings, or these substances are generated by microbes in the film. These nutrients boost the initial growth of the adhering bacteria and facilitate biofilm formation by acting as a food source. Coatings with combined antibacterial efficacy and antifouling properties provide further benefits by preventing dead cells and debris from adhering to the surfaces. In the present review, we address numerous pathogenic microbes that form biofilms on the surfaces of biomedical devices. In addition, we explore several efficient smart antiadhesive coatings on the surfaces of biomedical device-relevant materials that manage nosocomial infections caused by biofilm-forming microbial pathogens.","author":[{"family":"Mishra","given":"Akanksha"},{"family":"Aggarwal","given":"Ashish"},{"family":"Khan","given":"Fazlurrahman"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/antibiotics13070623","URL":"https://doi.org/10.3390/antibiotics13070623","source":"openalex"},{"id":"oa:W4393194929","type":"article-journal","title":"Use of artificial intelligence in breast surgery: a narrative review","abstract":"Background and Objective: We have witnessed tremendous advances in artificial intelligence (AI) technologies. Breast surgery, a subspecialty of general surgery, has notably benefited from AI technologies. This review aims to evaluate how AI has been integrated into breast surgery practices, to assess its effectiveness in improving surgical outcomes and operational efficiency, and to identify potential areas for future research and application. Methods: Two authors independently conducted a comprehensive search of PubMed, Google Scholar, EMBASE, and Cochrane CENTRAL databases from January 1, 1950, to September 4, 2023, employing keywords pertinent to AI in conjunction with breast surgery or cancer. The search focused on English language publications, where relevance was determined through meticulous screening of titles, abstracts, and full-texts, followed by an additional review of references within these articles. The review covered a range of studies illustrating the applications of AI in breast surgery encompassing lesion diagnosis to postoperative follow-up. Publications focusing specifically on breast reconstruction were excluded. Key Content and Findings: AI models have preoperative, intraoperative, and postoperative applications in the field of breast surgery. Using breast imaging scans and patient data, AI models have been designed to predict the risk of breast cancer and determine the need for breast cancer surgery. In addition, using breast imaging scans and histopathological slides, models were used for detecting, classifying, segmenting, grading, and staging breast tumors. Preoperative applications included patient education and the display of expected aesthetic outcomes. Models were also designed to provide intraoperative assistance for precise tumor resection and margin status assessment. As well, AI was used to predict postoperative complications, survival, and cancer recurrence. Conclusions: Extra research is required to move AI models from the experimental stage to actual implementation in healthcare. With the rapid evolution of AI, further applications are expected in the coming years including direct performance of breast surgery. Breast surgeons should be updated with the advances in AI applications in breast surgery to provide the best care for their patients.","author":[{"family":"Seth","given":"Ishith"},{"family":"Lim","given":"Bryan"},{"family":"Joseph","given":"Konrad"},{"family":"Gracias","given":"Dylan"},{"family":"Xie","given":"Yi"},{"family":"Ross","given":"Richard"},{"family":"Rozen","given":"Warren"}],"issued":{"date-parts":[[2024]]},"DOI":"10.21037/gs-23-414","URL":"https://doi.org/10.21037/gs-23-414","source":"openalex"},{"id":"oa:W4392599487","type":"article-journal","title":"ChatGPT’s Response Consistency: A Study on Repeated Queries of Medical Examination Questions","abstract":"(1) Background: As the field of artificial intelligence (AI) evolves, tools like ChatGPT are increasingly integrated into various domains of medicine, including medical education and research. Given the critical nature of medicine, it is of paramount importance that AI tools offer a high degree of reliability in the information they provide. (2) Methods: A total of n = 450 medical examination questions were manually entered into ChatGPT thrice, each for ChatGPT 3.5 and ChatGPT 4. The responses were collected, and their accuracy and consistency were statistically analyzed throughout the series of entries. (3) Results: ChatGPT 4 displayed a statistically significantly improved accuracy with 85.7% compared to that of 57.7% of ChatGPT 3.5 (p < 0.001). Furthermore, ChatGPT 4 was more consistent, correctly answering 77.8% across all rounds, a significant increase from the 44.9% observed from ChatGPT 3.5 (p < 0.001). (4) Conclusions: The findings underscore the increased accuracy and dependability of ChatGPT 4 in the context of medical education and potential clinical decision making. Nonetheless, the research emphasizes the indispensable nature of human-delivered healthcare and the vital role of continuous assessment in leveraging AI in medicine.","author":[{"family":"Funk","given":"Paul"},{"family":"Hoch","given":"Cosima"},{"family":"Knoedler","given":"Samuel"},{"family":"Knoedler","given":"Leonard"},{"family":"Cotofana","given":"Sebastian"},{"family":"Sofo","given":"Giuseppe"},{"family":"Dezfouli","given":"Ali"},{"family":"Wollenberg","given":"Barbara"},{"family":"Guntinaslichius","given":"Orlando"},{"family":"Alfertshofer","given":"Michael"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/ejihpe14030043","URL":"https://doi.org/10.3390/ejihpe14030043","source":"openalex"},{"id":"oa:W4390231752","type":"article-journal","title":"ROLE OF ARTIFICIAL INTELLIGENCE IN ELECTRIFICATION OF AFRICA","abstract":"The research explores the relationship between Artificial Intelligence (AI) and electrification in Africa, focusing on the challenges and emerging trends. The electrification deficit in Africa poses a significant impediment to economic development and social progress. This paper explores the pivotal role that Artificial Intelligence (AI) plays in addressing the challenges associated with electrification initiatives across the African continent. With its capacity for innovation and optimization, AI emerges as a transformative force capable of revolutionizing the planning, deployment, and management of electrification projects in a region characterized by diverse geographical landscapes and economic constraints. The paper investigates the potential of AI in addressing financial barriers associated with electrification projects. By facilitating innovative financing models, reducing operational costs, and attracting investments, AI contributes to creating sustainable and economically viable electrification solutions. The focus extends to decentralized energy systems and microgrids, exploring how AI can empower remote and underserved communities with reliable access to electricity. The socio-economic impact of AI-driven electrification initiatives is also scrutinized, emphasizing the potential for job creation, economic growth, and improved living standards. The paper discusses the importance of capacity building and local empowerment to ensure that AI technologies are effectively integrated into electrification projects while fostering inclusive and sustainable development. It highlights the role of AI in revolutionizing electrification, predicting electricity consumption and enabling decentralized solutions. AI-driven electrification has shown economic and social benefits, including enhanced productivity, improved quality of life, and increased market access. However, it also raises ethical concerns and privacy implications. The research emphasizes the need for proactive mitigation strategies and collaborations across sectors to drive regulatory frameworks, technological innovations, and global impact. The research envisions a future where AI plays a central role in fostering inclusive growth, connecting communities, and illuminating a digitally transformed and sustainable continent. Keywords: Artificial Intelligence, Electrification, Africa, Electricity, Energy","author":[{"family":"Cm","given":"Ibegbulam"},{"family":"Aigbovbiosa","given":"OJ"},{"family":"Olowonubi","given":"JA"},{"family":"Fatounde","given":"SA"}],"issued":{"date-parts":[[2023]]},"DOI":"10.51594/estj.v4i6.667","URL":"https://doi.org/10.51594/estj.v4i6.667","source":"openalex"},{"id":"oa:W4401167708","type":"article-journal","title":"Will Artificial Intelligence Be “Better” Than Humans in the Management of Syncope?","abstract":"Clinical decision-making regarding syncope poses challenges, with risk of physician error due to the elusive nature of syncope pathophysiology, diverse presentations, heterogeneity of risk factors, and limited therapeutic options. Artificial intelligence (AI)-based techniques, including machine learning (ML), deep learning (DL), and natural language processing (NLP), can uncover hidden and nonlinear connections among syncope risk factors, disease features, and clinical outcomes. ML, DL, and NLP models can analyze vast amounts of data effectively and assist physicians to help distinguish true syncope from other types of transient loss of consciousness. Additionally, short-term adverse events and length of hospital stay can be predicted by these models. In syncope research, AI-based models shift the focus from causality to correlation analysis between entities. This prompts the search for patterns rather than defining a hypothesis to be tested a priori. Furthermore, education of students, doctors, and health care providers engaged in continuing medical education may benefit from clinical cases of syncope interacting with NLP-based virtual patient simulators. Education may be of benefit to patients. This article explores potential strengths, weaknesses, and proposed solutions associated with utilization of ML and DL in syncope diagnosis and management. Three main topics regarding syncope are addressed: 1) clinical decision-making; 2) clinical research; and 3) education. Within each domain, we question whether \"AI will be better than humans,\" seeking evidence to support our objective inquiry.","author":[{"family":"Dipaola","given":"Franca"},{"family":"Gebska","given":"Milena"},{"family":"Gatti","given":"Mauro"},{"family":"Levra","given":"Alessandro"},{"family":"Parker","given":"William"},{"family":"Menè","given":"Roberto"},{"family":"Lee","given":"Sangil"},{"family":"Costantino","given":"Giorgio"},{"family":"Barsotti","given":"Ercole"},{"family":"Shiffer","given":"Dana"},{"family":"Johnston","given":"Samuel"},{"family":"Sutton","given":"Richard"},{"family":"Olshansky","given":"Brian"},{"family":"Furlan","given":"Raffaello"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.jacadv.2024.101072","URL":"https://doi.org/10.1016/j.jacadv.2024.101072","source":"openalex"},{"id":"oa:W4399250444","type":"article-journal","title":"Characterizing the Increase in Artificial Intelligence Content Detection in Oncology Scientific Abstracts From 2021 to 2023","abstract":"PURPOSE: Artificial intelligence (AI) models can generate scientific abstracts that are difficult to distinguish from the work of human authors. The use of AI in scientific writing and performance of AI detection tools are poorly characterized. METHODS: We extracted text from published scientific abstracts from the ASCO 2021-2023 Annual Meetings. Likelihood of AI content was evaluated by three detectors: GPTZero, Originality.ai, and Sapling. Optimal thresholds for AI content detection were selected using 100 abstracts from before 2020 as negative controls, and 100 produced by OpenAI's GPT-3 and GPT-4 models as positive controls. Logistic regression was used to evaluate the association of predicted AI content with submission year and abstract characteristics, and adjusted odds ratios (aORs) were computed. RESULTS: Fifteen thousand five hundred and fifty-three abstracts met inclusion criteria. Across detectors, abstracts submitted in 2023 were significantly more likely to contain AI content than those in 2021 (aOR range from 1.79 with Originality to 2.37 with Sapling). Online-only publication and lack of clinical trial number were consistently associated with AI content. With optimal thresholds, 99.5%, 96%, and 97% of GPT-3/4-generated abstracts were identified by GPTZero, Originality, and Sapling respectively, and no sampled abstracts from before 2020 were classified as AI generated by the GPTZero and Originality detectors. Correlation between detectors was low to moderate, with Spearman correlation coefficient ranging from 0.14 for Originality and Sapling to 0.47 for Sapling and GPTZero. CONCLUSION: There is an increasing signal of AI content in ASCO abstracts, coinciding with the growing popularity of generative AI models.","author":[{"family":"Howard","given":"Frederick"},{"family":"Li","given":"Anran"},{"family":"Riffon","given":"Mark"},{"family":"Garrettmayer","given":"Elizabeth"},{"family":"Pearson","given":"Alexander"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1200/cci.24.00077","URL":"https://doi.org/10.1200/cci.24.00077","source":"openalex"},{"id":"oa:W4400578172","type":"article-journal","title":"AI in Radiology: Navigating Medical Responsibility","abstract":"The application of Artificial Intelligence (AI) facilitates medical activities by automating routine tasks for healthcare professionals. AI augments but does not replace human decision-making, thus complicating the process of addressing legal responsibility. This study investigates the legal challenges associated with the medical use of AI in radiology, analyzing relevant case law and literature, with a specific focus on professional liability attribution. In the case of an error, the primary responsibility remains with the physician, with possible shared liability with developers according to the framework of medical device liability. If there is disagreement with the AI's findings, the physician must not only pursue but also justify their choices according to prevailing professional standards. Regulations must balance the autonomy of AI systems with the need for responsible clinical practice. Effective use of AI-generated evaluations requires knowledge of data dynamics and metrics like sensitivity and specificity, even without a clear understanding of the underlying algorithms: the opacity (referred to as the \"black box phenomenon\") of certain systems raises concerns about the interpretation and actual usability of results for both physicians and patients. AI is redefining healthcare, underscoring the imperative for robust liability frameworks, meticulous updates of systems, and transparent patient communication regarding AI involvement.","author":[{"family":"Contaldo","given":"Maria"},{"family":"Pasceri","given":"Giovanni"},{"family":"Vignati","given":"G"},{"family":"Bracchi","given":"Laura"},{"family":"Triggiani","given":"Sonia"},{"family":"Carrafiello","given":"Gianpaolo"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/diagnostics14141506","URL":"https://doi.org/10.3390/diagnostics14141506","source":"openalex"},{"id":"oa:W4403649741","type":"article-journal","title":"Artificial general intelligence in industry 4.0, 5.0, and society 5.0: Applications, opportunities, challenges, and future direction","abstract":"Artificial General Intelligence (AGI) is a giant step ahead of narrow Artificial Intelligence (AI), and in a wonder-form, it can really revolutionize Industry 4.0, 5.0, and Society 5.0. For instance, in Industry 4.0, AGI will permit absolutely autonomous fabrication procedures, optimizing supply chains and personalizing products with efficiency like never before. While bringing about Industry 5.0, the integration of AGI with human intelligence foresees fostering a collaborative, sustainable industrial environment more oriented to human-centric innovation and well-being. Society 5.0 has a vision of harmoniously blending cyberspace and physical space but leaves out the role AGI could play in dealing with the complex societal challenges of healthcare, education, and urban management with its intelligent solution tailoring to the needs of individuals and communities. However, there are big challenges to the use of AGI in these domains, and there are ethical concerns. Key issues to be tackled include risks of autonomy without accountability, possible biases of decision-making, and socio-economic effects from large-scale automation. Above all, safety, transparency, and fairness in using AGI systems are important to avoid undesirable consequences. The ethical concerns of AGI, like privacy issues, possible misuse, and the requirement for solid regulatory frameworks, are getting urgent considering major steps we are taking toward such a highly evolved paradigm in society and industries.","author":[{"family":"Rane","given":"Jayesh"},{"family":"Kaya","given":"Ömer"},{"family":"Mallick","given":"Suraj"},{"family":"Rane","given":"Nitin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.70593/978-81-981271-0-5_6","URL":"https://doi.org/10.70593/978-81-981271-0-5_6","source":"openalex"},{"id":"oa:W4394716306","type":"article-journal","title":"Extracting value from total-body PET/CT image data - the emerging role of artificial intelligence","abstract":"The evolution of Positron Emission Tomography (PET), culminating in the Total-Body PET (TB-PET) system, represents a paradigm shift in medical imaging. This paper explores the transformative role of Artificial Intelligence (AI) in enhancing clinical and research applications of TB-PET imaging. Clinically, TB-PET's superior sensitivity facilitates rapid imaging, low-dose imaging protocols, improved diagnostic capabilities and higher patient comfort. In research, TB-PET shows promise in studying systemic interactions and enhancing our understanding of human physiology and pathophysiology. In parallel, AI's integration into PET imaging workflows-spanning from image acquisition to data analysis-marks a significant development in nuclear medicine. This review delves into the current and potential roles of AI in augmenting TB-PET/CT's functionality and utility. We explore how AI can streamline current PET imaging processes and pioneer new applications, thereby maximising the technology's capabilities. The discussion also addresses necessary steps and considerations for effectively integrating AI into TB-PET/CT research and clinical practice. The paper highlights AI's role in enhancing TB-PET's efficiency and addresses the challenges posed by TB-PET's increased complexity. In conclusion, this exploration emphasises the need for a collaborative approach in the field of medical imaging. We advocate for shared resources and open-source initiatives as crucial steps towards harnessing the full potential of the AI/TB-PET synergy. This collaborative effort is essential for revolutionising medical imaging, ultimately leading to significant advancements in patient care and medical research.","author":[{"family":"Sundar","given":"Lalith"},{"family":"Gutschmayer","given":"Sebastian"},{"family":"Maenle","given":"Marcel"},{"family":"Beyer","given":"Thomas"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s40644-024-00684-w","URL":"https://doi.org/10.1186/s40644-024-00684-w","source":"openalex"},{"id":"oa:W4401701623","type":"article-journal","title":"Comparative Analysis of Accuracy, Readability, Sentiment, and Actionability: Artificial Intelligence Chatbots (ChatGPT and Google Gemini) versus Traditional Patient Information Leaflets for Local Anesthesia in Eye Surgery","abstract":"Background and Aim: Eye surgeries often evoke strong negative emotions in patients, including fear and anxiety. Patient education material plays a crucial role in informing and empowering individuals. Traditional sources of medical information may not effectively address individual patient concerns or cater to varying levels of understanding. This study aims to conduct a comparative analysis of the accuracy, completeness, readability, tone, and understandability of patient education material generated by AI chatbots versus traditional Patient Information Leaflets (PILs), focusing on local anesthesia in eye surgery. Methods: Expert reviewers evaluated responses generated by AI chatbots (ChatGPT and Google Gemini) and a traditional PIL (Royal College of Anaesthetists' PIL) based on accuracy, completeness, readability, sentiment, and understandability. Statistical analyses, including ANOVA and Tukey HSD tests, were conducted to compare the performance of the sources. Results: Readability analysis showed variations in complexity among the sources, with AI chatbots offering simplified language and PILs maintaining better overall readability and accessibility. Sentiment analysis revealed differences in emotional tone, with Google Gemini exhibiting the most positive sentiment. AI chatbots demonstrated superior understandability and actionability, while PILs excelled in completeness. Overall, ChatGPT showed slightly higher accuracy (scores expressed as mean ± standard deviation) (4.71 ± 0.5 vs 4.61 ± 0.62) and completeness (4.55 ± 0.58 vs 4.47 ± 0.58) compared to Google Gemini, but PILs performed best (4.84 ± 0.37 vs 4.88 ± 0.33) in terms of both accuracy and completeness (p-value for completeness <0.05). Conclusion: AI chatbots show promise as innovative tools for patient education, complementing traditional PILs. By leveraging the strengths of both AI-driven technologies and human expertise, healthcare providers can enhance patient education and empower individuals to make informed decisions about their health and medical care.","author":[{"family":"Gondode","given":"Prakash"},{"family":"Duggal","given":"Sakshi"},{"family":"Garg","given":"Neha"},{"family":"Lohakare","given":"Pooja"},{"family":"Jakhar","given":"Jubin"},{"family":"Bharti","given":"Swati"},{"family":"Dewangan","given":"Shraddha"}],"issued":{"date-parts":[[2024]]},"DOI":"10.22599/bioj.377","URL":"https://doi.org/10.22599/bioj.377","source":"openalex"},{"id":"oa:W4402882403","type":"article-journal","title":"Artificial intelligence-enabled multipurpose smart detection in active-matrix electrowetting-on-dielectric digital microfluidics","abstract":"An active-matrix electrowetting-on-dielectric (AM-EWOD) system integrates hundreds of thousands of active electrodes for sample droplet manipulation, which can enable simultaneous, automatic, and parallel on-chip biochemical reactions. A smart detection system is essential for ensuring a fully automatic workflow and online programming for the subsequent experimental steps. In this work, we demonstrated an artificial intelligence (AI)-enabled multipurpose smart detection method in an AM-EWOD system for different tasks. We employed the U-Net model to quantitatively evaluate the uniformity of the applied droplet-splitting methods. We used the YOLOv8 model to monitor the droplet-splitting process online. A 97.76% splitting success rate was observed with 18 different AM-EWOD chips. A 99.982% model precision rate and a 99.980% model recall rate were manually verified. We employed an improved YOLOv8 model to detect single-cell samples in nanolitre droplets. Compared with manual verification, the model achieved 99.260% and 99.193% precision and recall rates, respectively. In addition, single-cell droplet sorting and routing experiments were demonstrated. With an AI-based smart detection system, AM-EWOD has shown great potential for use as a ubiquitous platform for implementing true lab-on-a-chip applications.","author":[{"family":"Jia","given":"Zhongjie"},{"family":"Chang","given":"Chunyu"},{"family":"Hu","given":"Siyi"},{"family":"Li","given":"Jiahao"},{"family":"Ge","given":"Mingfeng"},{"family":"Dong","given":"Wen‐fei"},{"family":"Ma","given":"Hanbin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41378-024-00765-7","URL":"https://doi.org/10.1038/s41378-024-00765-7","source":"openalex"},{"id":"oa:W4390593459","type":"article-journal","title":"Search Engines and Generative Artificial Intelligence Integration: Public Health Risks and Recommendations to Safeguard Consumers Online","abstract":"BACKGROUND: The online pharmacy market is growing, with legitimate online pharmacies offering advantages such as convenience and accessibility. However, this increased demand has attracted malicious actors into this space, leading to the proliferation of illegal vendors that use deceptive techniques to rank higher in search results and pose serious public health risks by dispensing substandard or falsified medicines. Search engine providers have started integrating generative artificial intelligence (AI) into search engine interfaces, which could revolutionize search by delivering more personalized results through a user-friendly experience. However, improper integration of these new technologies carries potential risks and could further exacerbate the risks posed by illicit online pharmacies by inadvertently directing users to illegal vendors. OBJECTIVE: The role of generative AI integration in reshaping search engine results, particularly related to online pharmacies, has not yet been studied. Our objective was to identify, determine the prevalence of, and characterize illegal online pharmacy recommendations within the AI-generated search results and recommendations. METHODS: We conducted a comparative assessment of AI-generated recommendations from Google's Search Generative Experience (SGE) and Microsoft Bing's Chat, focusing on popular and well-known medicines representing multiple therapeutic categories including controlled substances. Websites were individually examined to determine legitimacy, and known illegal vendors were identified by cross-referencing with the National Association of Boards of Pharmacy and LegitScript databases. RESULTS: Of the 262 websites recommended in the AI-generated search results, 47.33% (124/262) belonged to active online pharmacies, with 31.29% (82/262) leading to legitimate ones. However, 19.04% (24/126) of Bing Chat's and 13.23% (18/136) of Google SGE's recommendations directed users to illegal vendors, including for controlled substances. The proportion of illegal pharmacies varied by drug and search engine. A significant difference was observed in the distribution of illegal websites between search engines. The prevalence of links leading to illegal online pharmacies selling prescription medications was significantly higher (P=.001) in Bing Chat (21/86, 24%) compared to Google SGE (6/92, 6%). Regarding the suggestions for controlled substances, suggestions generated by Google led to a significantly higher number of rogue sellers (12/44, 27%; P=.02) compared to Bing (3/40, 7%). CONCLUSIONS: While the integration of generative AI into search engines offers promising potential, it also poses significant risks. This is the first study to shed light on the vulnerabilities within these platforms while highlighting the potential public health implications associated with their inadvertent promotion of illegal pharmacies. We found a concerning proportion of AI-generated recommendations that led to illegal online pharmacies, which could not only potentially increase their traffic but also further exacerbate existing public health risks. Rigorous oversight and proper safeguards are urgently needed in generative search to mitigate consumer risks, making sure to actively guide users to verified pharmacies and prioritize legitimate sources while excluding illegal vendors from recommendations.","author":[{"family":"Ashraf","given":"Amir"},{"family":"Mackey","given":"Tim"},{"family":"Fittler","given":"András"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/53086","URL":"https://doi.org/10.2196/53086","source":"openalex"},{"id":"oa:W4390611806","type":"article-journal","title":"Brain organoids and organoid intelligence from ethical, legal, and social points of view","abstract":"Human brain organoids, aka cerebral organoids or earlier “mini-brains”, are 3D cellular models that recapitulate aspects of the developing human brain. They show tremendous promise for advancing our understanding of neurodevelopment and neurological disorders. However, the unprecedented ability to model human brain development and function in vitro also raises complex ethical, legal, and social challenges. Organoid Intelligence (OI) describes the ongoing movement to combine such organoids with Artificial Intelligence to establish basic forms of memory and learning. This article discusses key issues regarding the scientific status and prospects of brain organoids and OI, conceptualizations of consciousness and the mind–brain relationship, ethical and legal dimensions, including moral status, human–animal chimeras, informed consent, and governance matters, such as oversight and regulation. A balanced framework is needed to allow vital research while addressing public perceptions and ethical concerns. Interdisciplinary perspectives and proactive engagement among scientists, ethicists, policymakers, and the public can enable responsible translational pathways for organoid technology. A thoughtful, proactive governance framework might be needed to ensure ethically responsible progress in this promising field.","author":[{"family":"Härtung","given":"Thomas"},{"family":"Pantoja","given":"Itzy"},{"family":"Smirnova","given":"Lena"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/frai.2023.1307613","URL":"https://doi.org/10.3389/frai.2023.1307613","source":"openalex"},{"id":"oa:W4402047181","type":"article-journal","title":"Artificial Intelligence in Multilingual Interpretation and Radiology Assessment for Clinical Language Evaluation (AI-MIRACLE)","abstract":"The AI-MIRACLE Study investigates the efficacy of using ChatGPT 4.0, a large language model (LLM), for translating and simplifying radiology reports into multiple languages, aimed at enhancing patient comprehension. The study assesses the model's performance across the most spoken languages in the U.S., emphasizing the accuracy and clarity of translated and simplified radiology reports for non-medical readers. This study employed ChatGPT 4.0 to translate and simplify selected radiology reports into Vietnamese, Tagalog, Spanish, Mandarin, and Arabic. Hindi was used as a preliminary test language for validation of the questionnaire. Performance was assessed via Google form surveys distributed to bilingual physicians, which assessed the translation accuracy and clarity of simplified texts provided by ChatGPT 4. Responses from 24 participants showed mixed results. The study underscores the model's varying success across different languages, emphasizing both potential applications and limitations. ChatGPT 4.0 shows promise in breaking down language barriers in healthcare settings, enhancing patient comprehension of complex medical information. However, the performance is inconsistent across languages, indicating a need for further refinement and more inclusive training of AI models to handle diverse medical contexts and languages. The study highlights the role of LLMs in improving healthcare communication and patient comprehension, while indicating the need for continued advancements in AI technology, particularly in the translation of low-resource languages.","author":[{"family":"Khanna","given":"Praneet"},{"family":"Dhillon","given":"Gagandeep"},{"family":"Buddhavarapu","given":"Venkata"},{"family":"Verma","given":"Ram"},{"family":"Kashyap","given":"Rahul"},{"family":"Grewal","given":"Harpreet"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/jpm14090923","URL":"https://doi.org/10.3390/jpm14090923","source":"openalex"},{"id":"oa:W4392237966","type":"article-journal","title":"A framework for evaluating clinical artificial intelligence systems without ground-truth annotations","abstract":"A clinical artificial intelligence (AI) system is often validated on data withheld during its development. This provides an estimate of its performance upon future deployment on data in the wild; those currently unseen but are expected to be encountered in a clinical setting. However, estimating performance on data in the wild is complicated by distribution shift between data in the wild and withheld data and the absence of ground-truth annotations. Here, we introduce SUDO, a framework for evaluating AI systems on data in the wild. Through experiments on AI systems developed for dermatology images, histopathology patches, and clinical notes, we show that SUDO can identify unreliable predictions, inform the selection of models, and allow for the previously out-of-reach assessment of algorithmic bias for data in the wild without ground-truth annotations. These capabilities can contribute to the deployment of trustworthy and ethical AI systems in medicine.","author":[{"family":"Kiyasseh","given":"Dani"},{"family":"Cohen","given":"Aaron"},{"family":"Jiang","given":"Chengsheng"},{"family":"Altieri","given":"Nicholas"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41467-024-46000-9","URL":"https://doi.org/10.1038/s41467-024-46000-9","source":"openalex"},{"id":"oa:W4399770032","type":"article-journal","title":"Quality of science journalism in the age of Artificial Intelligence explored with a mixed methodology","abstract":"Science journalists, traditionally, play a key role in delivering science information to a wider audience. However, changes in the media ecosystem and the science-media relationship are posing challenges to reliable news production. Additionally, recent developments such as ChatGPT and Artificial Intelligence (AI) more generally, may have further consequences for the work of (science) journalists. Through a mixed-methodology, the quality of news reporting was studied within the context of AI. A content analysis of media output about AI (news articles published within the time frame 1 September 2022-28 February 2023) explored the adherence to quality indicators, while interviews shed light on journalism practices regarding quality reporting on and with AI. Perspectives from understudied areas in four European countries (Belgium, Italy, Portugal, and Spain) were included and compared. The findings show that AI received continuous media attention in the four countries. Furthermore, despite four different media landscapes, the reporting in the news articles adhered to the same quality criteria such as applying rigour, including sources of information, accessibility, and relevance. Thematic analysis of the interview findings revealed that impact of AI and ChatGPT on the journalism profession is still in its infancy. Expected benefits of AI related to helping with repetitive tasks (e.g. translations), and positively influencing journalistic principles of accessibility, engagement, and impact, while concerns showed fear for lower adherence to principles of rigour, integrity and transparency of sources of information. More generally, the interviewees expressed concerns about the state of science journalism, including a lack of funding influencing the quality of reporting. Journalists who were employed as staff as well as those who worked as freelancers put efforts in ensuring quality output, for example, via editorial oversight, discussions, or memberships of associations. Further research into the science-media relationship is recommended.","author":[{"family":"Dijkstra","given":"Anne"},{"family":"Jong","given":"Anouk"},{"family":"Boscolo","given":"Marco"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1371/journal.pone.0303367","URL":"https://doi.org/10.1371/journal.pone.0303367","source":"openalex"},{"id":"oa:W4405366153","type":"article-journal","title":"Lecturers’ Perceptions on the Integration of Artificial Intelligence Tools into Teaching Practice","abstract":"Higher education has witnessed a massive transformation due to the advent of generative artificial intelligence (AI) technologies such as ChatGPT. In essence, AI has transformed various aspects of society as a whole. Despite the growing interest in applying AI tools such as ChatGPT in higher education, there is limited understanding of lecturers’ perceptions regarding their use in teaching and learning contexts. Studies reported in the literature have not comprehensively explored lecturers’ attitudes towards AI adoption, particularly in terms of its impact on classroom activities, assessment, and feedback. This study aims to fill this gap by qualitatively studying lecturers’ perspectives. The findings reveal that the advent of AI was met with mixed feelings among lecturers. Some lecturers embraced AI technologies and developed mechanisms for utilizing them in the classroom, while others resisted the change. This research is significant, as it can inform best practices and guide future implementation strategies of technologies in education.","author":[{"family":"Mutanga","given":"Murimo"},{"family":"Jugoo","given":"Vikash"},{"family":"Adefemi","given":"Kuburat"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/higheredu3040066","URL":"https://doi.org/10.3390/higheredu3040066","source":"openalex"},{"id":"oa:W4390608381","type":"article-journal","title":"Lung Imaging and Artificial Intelligence in ARDS","abstract":"Artificial intelligence (AI) can make intelligent decisions in a manner akin to that of the human mind. AI has the potential to improve clinical workflow, diagnosis, and prognosis, especially in radiology. Acute respiratory distress syndrome (ARDS) is a very diverse illness that is characterized by interstitial opacities, mostly in the dependent areas, decreased lung aeration with alveolar collapse, and inflammatory lung edema resulting in elevated lung weight. As a result, lung imaging is a crucial tool for evaluating the mechanical and morphological traits of ARDS patients. Compared to traditional chest radiography, sensitivity and specificity of lung computed tomography (CT) and ultrasound are higher. The state of the art in the application of AI is summarized in this narrative review which focuses on CT and ultrasound techniques in patients with ARDS. A total of eighteen items were retrieved. The primary goals of using AI for lung imaging were to evaluate the risk of developing ARDS, the measurement of alveolar recruitment, potential alternative diagnoses, and outcome. While the physician must still be present to guarantee a high standard of examination, AI could help the clinical team provide the best care possible.","author":[{"family":"Chiumello","given":"Davide"},{"family":"Coppola","given":"Silvia"},{"family":"Catozzi","given":"Giulia"},{"family":"Danzo","given":"Fiammetta"},{"family":"Santus","given":"Pierachille"},{"family":"Radovanovic","given":"Dejan"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/jcm13020305","URL":"https://doi.org/10.3390/jcm13020305","source":"openalex"},{"id":"oa:W4390508424","type":"article-journal","title":"Artificial Intelligence: The Future","abstract":"Artificial intelligence is the intelligence of machines or software, as opposed to the intelligence of humans or animals. It is also the field of study in computer science that develops and studies intelligent machines. \"AI\" may also refer to the machines themselves. AI is not a new for the scientist, it was introduce in 1943 with artificial neurons model and get popular in 1950 due to “Turting test” the test was done to get answer that machine can think?, purposed by Alan Turing. Basically AI is categorized into three types, Artificial Narrow Intelligence, Artificial General Intelligence and Artificial Super Intelligence. Deep learning and machine learning is major subfield of Artificial intelligence. DL as a subset of ML, which is also another subset of AI. Therefore, AI is the all-encompassing concept that initially erupted. Application of AI fields are Healthcare, Business, Education, Agriculture, Finance, Law, Entertainment and media, Software coding and IT processes, Security, Manufacturing, Banking and Transportation. In reference to Job creation or distraction, Artificial Intelligence is not job killer but a job category killer”. In the latest report (May 2023) on The Future of Jobs, the World Economic Forum (WEF) predicts the creation of 69 million jobs by 2027 thanks to AI, but also the destruction of 89 million jobs. In context to intelligence level, Jan. 2022, Age of IQ level of AI was as 7 year Children and in Dec. 2022, Age of IQ level of AI was as 9 year Children. India is emerging market in global and it has around 12 % of work could be automated by AI. In India more than 2000 startup are related to AI and 90000 plus AI Professional work in India.The economic impact of AI, for select G20 countries and estimates AI to boost India’s annual growth rate by 1.3 percentage points by 2035. AI has potential to add 1 trillion to India’s economy in 2035. We are going to enter into new technological world, it’s may be our fortune or misfortune. Key Words: - Artificial Intelligence, Machine learning, Job, India","author":[{"family":"Yadav","given":"Narendra"},{"family":"Sharma","given":"Latika"},{"family":"Dhake","given":"Urmila"}],"issued":{"date-parts":[[2023]]},"DOI":"10.55041/ijsrem27796","URL":"https://doi.org/10.55041/ijsrem27796","source":"openalex"},{"id":"oa:W4404793685","type":"article-journal","title":"Artificial Intelligence in IVF Laboratories: Elevating Outcomes Through Precision and Efficiency","abstract":"Incorporating artificial intelligence (AI) into in vitro fertilization (IVF) laboratories signifies a significant advancement in reproductive medicine. AI technologies, such as neural networks, deep learning, and machine learning, promise to enhance quality control (QC) and quality assurance (QA) through increased accuracy, consistency, and operational efficiency. This comprehensive review examines the effects of AI on IVF laboratories, focusing on its role in automating processes such as embryo and sperm selection, optimizing clinical outcomes, and reducing human error. AI's data analysis and pattern recognition capabilities offer valuable predictive insights, enhancing personalized treatment plans and increasing success rates in fertility treatments. However, integrating AI also brings ethical, regulatory, and societal challenges, including concerns about data security, algorithmic bias, and the human-machine interface in clinical decision-making. Through an in-depth examination of current case studies, advancements, and future directions, this manuscript highlights how AI can revolutionize IVF by standardizing processes, improving patient outcomes, and advancing the precision of reproductive medicine. It underscores the necessity of ongoing research and ethical oversight to ensure fair and transparent applications in this sensitive field, assuring the responsible use of AI in reproductive medicine.","author":[{"family":"Hew","given":"Yaling"},{"family":"Kütük","given":"Duygu"},{"family":"Düzcü","given":"Tuba"},{"family":"Ergun","given":"Yagmur"},{"family":"Başar","given":"Murat"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/biology13120988","URL":"https://doi.org/10.3390/biology13120988","source":"openalex"},{"id":"oa:W4405244711","type":"article-journal","title":"Artificial intelligence in respiratory care: perspectives on critical opportunities and challenges","abstract":"Artificial intelligence (AI) is transforming respiratory healthcare through a wide range of deep learning and generative tools, and is increasingly integrated into both patients' lives and routine respiratory care. The implications of AI in respiratory care are vast and multifaceted, presenting both promises and uncertainties from the perspectives of clinicians, patients and society. Clinicians contemplate whether AI will streamline or complicate their daily tasks, while patients weigh the potential benefits of personalised self-management support against risks such as data privacy concerns and misinformation. The impact of AI on the clinician-patient relationship remains a pivotal consideration, with the potential to either enhance collaborative care or create depersonalised interactions. Societally, there is an imperative to leverage AI in respiratory care to bridge healthcare disparities, while safeguarding against the widening of inequalities. Strategic efforts to promote transparency and prioritise inclusivity and ease of understanding in algorithm co-design will be crucial in shaping future AI to maximise benefits and minimise risks for all stakeholders.","author":[{"family":"Drummond","given":"David"},{"family":"Adejumo","given":"Ireti"},{"family":"Hansen","given":"Kjeld"},{"family":"Poberezhets","given":"Vitalii"},{"family":"Slabaugh","given":"Greg"},{"family":"Hui","given":"Chi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1183/20734735.0189-2023","URL":"https://doi.org/10.1183/20734735.0189-2023","source":"openalex"},{"id":"oa:W4393098213","type":"article-journal","title":"Integration of cognitive tasks into artificial general intelligence test for large models","abstract":"During the evolution of large models, performance evaluation is necessary for assessing their capabilities. However, current model evaluations mainly rely on specific tasks and datasets, lacking a united framework for assessing the multidimensional intelligence of large models. In this perspective, we advocate for a comprehensive framework of cognitive science-inspired artificial general intelligence (AGI) tests, including crystallized, fluid, social, and embodied intelligence. The AGI tests consist of well-designed cognitive tests adopted from human intelligence tests, and then naturally encapsulates into an immersive virtual community. We propose increasing the complexity of AGI testing tasks commensurate with advancements in large models and emphasizing the necessity for the interpretation of test results to avoid false negatives and false positives. We believe that cognitive science-inspired AGI tests will effectively guide the targeted improvement of large models in specific dimensions of intelligence and accelerate the integration of large models into human society.","author":[{"family":"Qu","given":"Youzhi"},{"family":"Wei","given":"Chen"},{"family":"Du","given":"Penghui"},{"family":"Che","given":"WQ"},{"family":"Zhang","given":"Chi"},{"family":"Ouyang","given":"Wanli"},{"family":"Bian","given":"Yatao"},{"family":"Xu","given":"Feiyang"},{"family":"Hu","given":"Bin"},{"family":"Du","given":"Kai"},{"family":"Wu","given":"Haiyan"},{"family":"Liu","given":"Jia"},{"family":"Liu","given":"Quanying"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.isci.2024.109550","URL":"https://doi.org/10.1016/j.isci.2024.109550","source":"openalex"},{"id":"oa:W4403032893","type":"article-journal","title":"Ensuring Appropriate Representation in Artificial Intelligence–Generated Medical Imagery: Protocol for a Methodological Approach to Address Skin Tone Bias","abstract":"BACKGROUND: In medical education, particularly in anatomy and dermatology, generative artificial intelligence (AI) can be used to create customized illustrations. However, the underrepresentation of darker skin tones in medical textbooks and elsewhere, which serve as training data for AI, poses a significant challenge in ensuring diverse and inclusive educational materials. OBJECTIVE: This study aims to evaluate the extent of skin tone diversity in AI-generated medical images and to test whether the representation of skin tones can be improved by modifying AI prompts to better reflect the demographic makeup of the US population. METHODS: In total, 2 standard AI models (Dall-E [OpenAI] and Midjourney [Midjourney Inc]) each generated 100 images of people with psoriasis. In addition, a custom model was developed that incorporated a prompt injection aimed at \"forcing\" the AI (Dall-E 3) to reflect the skin tone distribution of the US population according to the 2012 American National Election Survey. This custom model generated another set of 100 images. The skin tones in these images were assessed by 3 researchers using the New Immigrant Survey skin tone scale, with the median value representing each image. A chi-square goodness of fit analysis compared the skin tone distributions from each set of images to that of the US population. RESULTS: The standard AI models (Dalle-3 and Midjourney) demonstrated a significant difference between the expected skin tones of the US population and the observed tones in the generated images (P<.001). Both standard AI models overrepresented lighter skin. Conversely, the custom model with the modified prompt yielded a distribution of skin tones that closely matched the expected demographic representation, showing no significant difference (P=.04). CONCLUSIONS: This study reveals a notable bias in AI-generated medical images, predominantly underrepresenting darker skin tones. This bias can be effectively addressed by modifying AI prompts to incorporate real-life demographic distributions. The findings emphasize the need for conscious efforts in AI development to ensure diverse and representative outputs, particularly in educational and medical contexts. Users of generative AI tools should be aware that these biases exist, and that similar tendencies may also exist in other types of generative AI (eg, large language models) and in other characteristics (eg, sex, gender, culture, and ethnicity). Injecting demographic data into AI prompts may effectively counteract these biases, ensuring a more accurate representation of the general population.","author":[{"family":"Omalley","given":"Andrew"},{"family":"Veenhuizen","given":"Miriam"},{"family":"Ahmed","given":"Ayla"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/58275","URL":"https://doi.org/10.2196/58275","source":"openalex"},{"id":"oa:W4388758788","type":"article-journal","title":"Artificial intelligence (AI) for neurologists: do digital neurones dream of electric sheep?","abstract":"Artificial intelligence (AI) is routinely mentioned in journals and newspapers, and non-technical outsiders may have difficulty in distinguishing hyperbole from reality. We present a practical guide to help non-technical neurologists to understand healthcare AI. AI is being used to support clinical decisions in treating neurological disorders. We introduce basic concepts of AI, such as machine learning and natural language processing, and explain how AI is being used in healthcare, giving examples its benefits and challenges. We also cover how AI performance is measured, and its regulatory aspects in healthcare. An important theme is that AI is a general-purpose technology like medical statistics, with broad utility applicable in various scenarios, such that niche approaches are outpaced by approaches that are broadly applicable in many disease areas and specialties. By understanding AI basics and its potential applications, neurologists can make informed decisions when evaluating AI used in their clinical practice. This article was written by four humans, with generative AI helping with formatting and image generation.","author":[{"family":"Yeung","given":"Joshua"},{"family":"Wang","given":"Yang"},{"family":"Kraljević","given":"Željko"},{"family":"Teo","given":"James"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1136/pn-2023-003757","URL":"https://doi.org/10.1136/pn-2023-003757","source":"openalex"},{"id":"oa:W4403288415","type":"article-journal","title":"The role of artificial intelligence in coronary CT angiography","abstract":"Coronary CT angiography (CCTA) offers an efficient and reliable tool for the non-invasive assessment of suspected coronary artery disease through the analysis of coronary artery plaque and stenosis. However, the detailed manual analysis of CCTA is a burdensome task requiring highly skilled experts. Recent advances in artificial intelligence (AI) have made significant progress toward a more comprehensive automated analysis of CCTA images, offering potential improvements in terms of speed, performance and scalability. This work offers an overview of the recent developments of AI in CCTA. We cover methodological advances for coronary artery tree and whole heart analysis, and provide an overview of AI techniques that have shown to be valuable for the analysis of cardiac anatomy and pathology in CCTA. Finally, we provide a general discussion regarding current challenges and limitations, and discuss prospects for future research.","author":[{"family":"Herten","given":"Rudolf"},{"family":"Lagogiannis","given":"Ioannis"},{"family":"Leiner","given":"Tim"},{"family":"Išgum","given":"Ivana"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s12471-024-01901-8","URL":"https://doi.org/10.1007/s12471-024-01901-8","source":"openalex"},{"id":"oa:W4387964872","type":"article-journal","title":"A Survey of Publicly Available MRI Datasets for Potential Use in Artificial Intelligence Research","abstract":"Artificial intelligence (AI) has the potential to bring transformative improvements to the field of radiology; yet, there are barriers to widespread clinical adoption. One of the most important barriers has been access to large, well-annotated, widely representative medical image datasets, which can be used to accurately train AI programs. Creating such datasets requires time and expertise and runs into constraints around data security and interoperability, patient privacy, and appropriate data use. Recognizing these challenges, several institutions have started curating and providing publicly available, high-quality datasets that can be accessed by researchers to advance AI models. The purpose of this work was to review the publicly available MRI datasets that can be used for AI research in radiology. Despite being an emerging field, a simple internet search for open MRI datasets presents an overwhelming number of results. Therefore, we decided to create a survey of the major publicly accessible MRI datasets in different subfields of radiology (brain, body, and musculoskeletal), and list the most important features of value to the AI researcher. To complete this review, we searched for publicly available MRI datasets and assessed them based on several parameters (number of subjects, demographics, area of interest, technical features, and annotations). We reviewed 110 datasets across sub-fields with 1,686,245 subjects in 12 different areas of interest ranging from spine to cardiac. This review is meant to serve as a reference for researchers to help spur advancements in the field of AI for radiology. LEVEL OF EVIDENCE: Level 4 TECHNICAL EFFICACY: Stage 6.","author":[{"family":"Dishner","given":"Katharine"},{"family":"Mcraeposani","given":"Bala"},{"family":"Bhowmik","given":"Arka"},{"family":"Jochelson","given":"Maxine"},{"family":"Holodny","given":"Andrei"},{"family":"Pinker","given":"Katja"},{"family":"Eskreiswinkler","given":"Sarah"},{"family":"Stember","given":"Joseph"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/jmri.29101","URL":"https://doi.org/10.1002/jmri.29101","source":"openalex"},{"id":"oa:W4392626370","type":"article-journal","title":"Nordic radiographers’ and students’ perspectives on artificial intelligence – A cross-sectional online survey","abstract":"INTRODUCTION: The integration of artificial intelligence (AI) into the domain of radiography holds substantial potential in various aspects including workflow efficiency, image processing, patient positioning, and quality assurance. The successful implementation of AI within a Radiology department necessitates the participation of key stakeholders, particularly radiographers. The study aimed to provide a comprehensive investigation about Nordic radiographers' perspectives and attitudes towards AI in radiography. METHODS: An online 29-item survey was distributed via social media platforms to Nordic students and radiographers working in Denmark, Norway, Sweden, Iceland, Greenland, and the Faroe Islands including items on demographics, specialization, educational background, place of work and perspectives and knowledge on AI. The items were a mix of closed-type and scaled questions, with the option for free-text responses when relevant. RESULTS: The survey received responses from all Nordic countries with 586 respondents, 26.8% males, 72.1% females, and 1.1% non-binary/self-defined or preferred not to say. The mean age was 37.2 with a standard deviation (SD) of ±12.1 years, and the mean number of years since qualification was 14.2 SD ± 10.3 years. A total of 43% (n = 254) of the respondents had not received any AI training in clinical practice. Whereas 13% (n = 76) had received AI during radiography undergrad training. A total of 77.9% (n = 412) expressed interest in pursuing AI education. The majority of respondents were aware of the potential use of AI (n = 485, 82.8%) and 39.1% (n = 204) had no reservations about AI. CONCLUSION: Overall, this study found that Nordic radiographers have a positive attitude toward AI. Very limited training or education has been provided to the radiographers. Especially since 82.8% reports on plans to implement AI in clinical practice. In general, awareness of AI applications is high, but the educational level is low for Nordic radiographers. IMPLICATION FOR PRACTICE: This study emphasises the favourable view of AI held by students and Nordic radiographers. However, there is a need for continuous professional development to facilitate the implementation and effective utilization of AI tools within the field of radiography.","author":[{"family":"Pedersen","given":"Malene"},{"family":"Kusk","given":"Martin"},{"family":"Lysdahlgaard","given":"Simon"},{"family":"Mork-Knudsen","given":"H"},{"family":"Malamateniou","given":"Christina"},{"family":"Jensen","given":"Janni"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.radi.2024.02.020","URL":"https://doi.org/10.1016/j.radi.2024.02.020","source":"openalex"},{"id":"oa:W4404613446","type":"article-journal","title":"Application of artificial intelligence in laryngeal lesions: a systematic review and meta-analysis","abstract":"OBJECTIVE: The objective of this systematic review and meta-analysis was to evaluate the diagnostic accuracy of AI-assisted technologies, including endoscopy, voice analysis, and histopathology, for detecting and classifying laryngeal lesions. METHODS: A systematic search was conducted in PubMed, Embase, etc. for studies utilizing voice analysis, histopathology for laryngeal lesions, or AI-assisted endoscopy. The results of diagnostic accuracy, sensitivity and specificity were synthesized by a meta-analysis. RESULTS: 12 studies employing AI-assisted endoscopy, 2 studies for voice analysis, and 4 studies for histopathology were included in the meta-analysis. The combined sensitivity of AI-assisted endoscopy was 91% (95% CI 87-94%) for the classification of benign from malignant lesions and 91% (95% CI 90-93%) for lesion detection. The highest accuracy pooled in detecting lesions versus healthy tissue was the AI-aided endoscopy was 94% (95% CI 92-97%). CONCLUSIONS: For laryngeal lesions, AI-assisted endoscopy shows excellent diagnosis accuracy. But more sizable prospective trials are needed to confirm the practical clinical value.","author":[{"family":"Marrerogonzalez","given":"Alejandro"},{"family":"Diemer","given":"Tanner"},{"family":"Nguyen","given":"Shaun"},{"family":"Camilon","given":"Terence"},{"family":"Meenan","given":"Kirsten"},{"family":"Orourke","given":"Ashli"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s00405-024-09075-0","URL":"https://doi.org/10.1007/s00405-024-09075-0","source":"openalex"},{"id":"oa:W4401499434","type":"article-journal","title":"Application of artificial intelligence in the diagnosis and treatment of urinary tumors","abstract":"Diagnosis and treatment of urological tumors, relying on auxiliary data such as medical imaging, while incorporating individual patient characteristics into treatment selection, has long been a key challenge in clinical medicine. Traditionally, clinicians used extensive experience for decision-making, but recent artificial intelligence (AI) advancements offer new solutions. Machine learning (ML) and deep learning (DL), notably convolutional neural networks (CNNs) in medical image recognition, enable precise tumor diagnosis and treatment. These technologies analyze complex medical image patterns, improving accuracy and efficiency. AI systems, by learning from vast datasets, reveal hidden features, offering reliable diagnostics and personalized treatment plans. Early detection is crucial for tumors like renal cell carcinoma (RCC), bladder cancer (BC), and Prostate Cancer (PCa). AI, coupled with data analysis, improves early detection and reduces misdiagnosis rates, enhancing treatment precision. AI's application in urological tumors is a research focus, promising a vital role in urological surgery with improved patient outcomes. This paper examines ML, DL in urological tumors, and AI's role in clinical decisions, providing insights for future AI applications in urological surgery.","author":[{"family":"Zhu","given":"Mengying"},{"family":"Gu","given":"Zhichao"},{"family":"Chen","given":"Fang"},{"family":"Chen","given":"Xi"},{"family":"Wang","given":"Yue"},{"family":"Zhao","given":"Guohua"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fonc.2024.1440626","URL":"https://doi.org/10.3389/fonc.2024.1440626","source":"openalex"},{"id":"oa:W4402577099","type":"article-journal","title":"Artificial Intelligence Tools in Pediatric Urology: A Comprehensive Review of Recent Advances","abstract":"Artificial intelligence (AI) is providing novel answers to long-standing clinical problems, and it is quickly changing pediatric urology. This thorough analysis focuses on current developments in AI technologies that improve pediatric urology diagnosis, treatment planning, and surgery results. Deep learning algorithms help detect problems with previously unheard-of precision in disorders including hydronephrosis, pyeloplasty, and vesicoureteral reflux, where AI-powered prediction models have demonstrated promising outcomes in boosting diagnostic accuracy. AI-enhanced image processing methods have significantly improved the quality and interpretation of medical images. Examples of these methods are deep-learning-based segmentation and contrast limited adaptive histogram equalization (CLAHE). These methods guarantee higher precision in the identification and classification of pediatric urological disorders, and AI-driven ground truth construction approaches aid in the standardization of and improvement in training data, resulting in more resilient and consistent segmentation models. AI is being used for surgical support as well. AI-assisted navigation devices help with difficult operations like pyeloplasty by decreasing complications and increasing surgical accuracy. AI also helps with long-term patient monitoring, predictive analytics, and customized treatment strategies, all of which improve results for younger patients. However, there are practical, ethical, and legal issues with AI integration in pediatric urology that need to be carefully navigated. To close knowledge gaps, more investigation is required, especially in the areas of AI-driven surgical methods and standardized ground truth datasets for pediatric radiologic image segmentation. In the end, AI has the potential to completely transform pediatric urology by enhancing patient care, increasing the effectiveness of treatments, and spurring more advancements in this exciting area.","author":[{"family":"Chowdhury","given":"Adiba"},{"family":"Salam","given":"Abdus"},{"family":"Naznine","given":"Mansura"},{"family":"Abdalla","given":"Da’ad"},{"family":"Erdman","given":"Lauren"},{"family":"Chowdhury","given":"Muhammad"},{"family":"Abbas","given":"Tariq"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/diagnostics14182059","URL":"https://doi.org/10.3390/diagnostics14182059","source":"openalex"},{"id":"oa:W4403454705","type":"article-journal","title":"Artificial intelligence and predictive models for early detection of acute kidney injury: transforming clinical practice","abstract":"Acute kidney injury (AKI) presents a significant clinical challenge due to its rapid progression to kidney failure, resulting in serious complications such as electrolyte imbalances, fluid overload, and the potential need for renal replacement therapy. Early detection and prediction of AKI can improve patient outcomes through timely interventions. This review was conducted as a narrative literature review, aiming to explore state-of-the-art models for early detection and prediction of AKI. We conducted a comprehensive review of findings from various studies, highlighting their strengths, limitations, and practical considerations for implementation in healthcare settings. We highlight the potential benefits and challenges of their integration into routine clinical care and emphasize the importance of establishing robust early-detection systems before the introduction of artificial intelligence (AI)-assisted prediction models. Advances in AI for AKI detection and prediction are examined, addressing their clinical applicability, challenges, and opportunities for routine implementation.","author":[{"family":"Tran","given":"Tu"},{"family":"Yun","given":"Giae"},{"family":"Kim","given":"Sejoong"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s12882-024-03793-7","URL":"https://doi.org/10.1186/s12882-024-03793-7","source":"openalex"},{"id":"oa:W4401456487","type":"article-journal","title":"Knowledge, Attitude, and Practices toward Artificial Intelligence among University Students in Lebanon","abstract":"Background: The expansion of artificial intelligence (AI) across diverse sectors worldwide demands an understanding of its impact on future generations. The studies of its influence on university students’ behavior and application in Lebanon are still limited. The present study aimed to explore the knowledge, attitudes, and practices (KAPs) of university students regarding AI and to identify factors affecting these dimensions. Methods: An online questionnaire (n = 457) was distributed to university students who were at least 18 years of age across Lebanon. Results: The results revealed that a significant majority (97.2%) of the participants were familiar with AI, from which 43% demonstrated a high level of knowledge. Furthermore, attitude toward AI role and integration in academic and professional paths was moderately satisfactory (43%), although it was reportedly used by 75% of students throughout their university years. There was a significant association between knowledge levels and sociodemographic factors such as age, sex, source of AI-related information, and knowledge rating (p < 0.05), whereas the academic major and knowledge rating affected attitudes toward AI (p < 0.05). Conclusion: These findings support the incorporation of AI education within the curriculum to increase acceptance of AI as a modern tool enhancing various sectors and serving as a facilitator for teaching and learning processes.","author":[{"family":"Kharroubi","given":"Samer"},{"family":"Tannir","given":"Iman"},{"family":"Hassan","given":"Rasha"},{"family":"Ballout","given":"Rouba"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/educsci14080863","URL":"https://doi.org/10.3390/educsci14080863","source":"openalex"},{"id":"oa:W4401498863","type":"article-journal","title":"Artificial intelligence reveals the predictions of hematological indexes in children with acute leukemia","abstract":"Childhood leukemia is a prevalent form of pediatric cancer, with acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) being the primary manifestations. Timely treatment has significantly enhanced survival rates for children with acute leukemia. This study aimed to develop an early and comprehensive predictor for hematologic malignancies in children by analyzing nutritional biomarkers, key leukemia indicators, and granulocytes in their blood. Using a machine learning algorithm and ten indices, the blood samples of 826 children with ALL and 255 children with AML were compared to a control group of 200 healthy children. The study revealed notable differences, including higher indicators in boys compared to girls and significant variations in most biochemical indicators between leukemia patients and healthy children. Employing a random forest model resulted in an area under the curve (AUC) of 0.950 for predicting leukemia subtypes and an AUC of 0.909 for forecasting AML. This research introduces an efficient diagnostic tool for early screening of childhood blood cancers and underscores the potential of artificial intelligence in modern healthcare.","author":[{"family":"Cheng","given":"Zhangkai"},{"family":"Li","given":"Haiyang"},{"family":"Liu","given":"Mingtao"},{"family":"Fu","given":"Xing"},{"family":"Liu","given":"Li"},{"family":"Liang","given":"Zhiman"},{"family":"Gan","given":"Hui"},{"family":"Sun","given":"Baoqing"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s12885-024-12646-3","URL":"https://doi.org/10.1186/s12885-024-12646-3","source":"openalex"},{"id":"oa:W4405306800","type":"article-journal","title":"American Academy of Otolaryngology–Head and Neck Surgery (AAO‐HNS) Report on Artificial Intelligence","abstract":"This report synthesizes the American Academy of Otolaryngology-Head and Neck Surgery (AAO-HNS) Task Force's guidance on the integration of artificial intelligence (AI) in otolaryngology-head and neck surgery (OHNS). A comprehensive literature review was conducted, focusing on the applications, benefits, and challenges of AI in OHNS, alongside ethical, legal, and social implications. The Task Force, formulated by otolaryngologist experts in AI, used an iterative approach, adapted from the Delphi method, to prioritize topics for inclusion and to reach a consensus on guiding principles. The Task Force's findings highlight AI's transformative potential for OHNS, offering potential advancements in precision medicine, clinical decision support, operational efficiency, research, and education. However, challenges such as data quality, health equity, privacy concerns, transparency, regulatory gaps, and ethical dilemmas necessitate careful navigation. Incorporating AI into otolaryngology practice in a safe, equitable, and patient-centered manner requires clinician judgment, transparent AI systems, and adherence to ethical and legal standards. The Task Force principles underscore the importance of otolaryngologists' involvement in AI's ethical development, implementation, and regulation to harness benefits while mitigating risks. The proposed principles inform the integration of AI in otolaryngology, aiming to enhance patient outcomes, clinician well-being, and efficiency of health care delivery.","author":[{"family":"Ayoub","given":"Noel"},{"family":"Rameau","given":"Anaïs"},{"family":"Brenner","given":"Michael"},{"family":"Bur","given":"Andrés"},{"family":"Ator","given":"Gregory"},{"family":"Briggs","given":"Selena"},{"family":"Takashima","given":"Masayoshi"},{"family":"Stanković","given":"Konstantina"},{"family":"Force","given":"Aao‐hns"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/ohn.1080","URL":"https://doi.org/10.1002/ohn.1080","source":"openalex"},{"id":"oa:W4315484276","type":"article-journal","title":"Reliability of Artificial Intelligence-Assisted Cephalometric Analysis. A Pilot Study","abstract":"Recently, Artificial Intelligence (AI) has spread in orthodontics, in particular within cephalometric analysis, where computerized digital software is able to provide linear-angular measurements upon manual landmark identification. A step forward is constituted by fully automated AI-assisted cephalometric analysis, where the landmarks are automatically detected by software. The aim of the study was to compare the reliability of a fully automated AI-assisted cephalometric analysis with the one obtained by a computerized digital software upon manual landmark identification. Fully automated AI-assisted cephalometric analysis of 13 lateral cephalograms were retrospectively compared to the cephalometric analysis performed twice by a blinded operator with a computerized software. Intra- and inter-operator (fully automated AI-assisted vs. computerized software with manual landmark identification) reliability in cephalometric parameters (maxillary convexity, facial conicity, facial axis angle, posterior and lower facial height) was tested with the Dahlberg equation and Bland–Altman plot. The results revealed no significant difference in intra- and inter-operator measurements. Although not significant, higher errors were observed within intra-operator measurements of posterior facial height and inter-operator measurements of facial axis angle. In conclusion, despite the small sample, the cephalometric measurements of a fully automated AI-assisted cephalometric software were reliable and accurate. Nevertheless, digital technological advances cannot substitute the critical role of the orthodontist toward a correct diagnosis.","author":[{"family":"Alessandribonetti","given":"Anna"},{"family":"Sangalli","given":"Linda"},{"family":"Salerno","given":"Martina"},{"family":"Gallenzi","given":"Patrizia"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/biomedinformatics3010003","URL":"https://doi.org/10.3390/biomedinformatics3010003","source":"openalex"},{"id":"oa:W4400084993","type":"article-journal","title":"Hybrid Explainable Artificial Intelligence Models for Targeted Metabolomics Analysis of Diabetic Retinopathy","abstract":"BACKGROUND: Diabetic retinopathy (DR) is a prevalent microvascular complication of diabetes mellitus, and early detection is crucial for effective management. Metabolomics profiling has emerged as a promising approach for identifying potential biomarkers associated with DR progression. This study aimed to develop a hybrid explainable artificial intelligence (XAI) model for targeted metabolomics analysis of patients with DR, utilizing a focused approach to identify specific metabolites exhibiting varying concentrations among individuals without DR (NDR), those with non-proliferative DR (NPDR), and individuals with proliferative DR (PDR) who have type 2 diabetes mellitus (T2DM). METHODS: A total of 317 T2DM patients, including 143 NDR, 123 NPDR, and 51 PDR cases, were included in the study. Serum samples underwent targeted metabolomics analysis using liquid chromatography and mass spectrometry. Several machine learning models, including Support Vector Machines (SVC), Random Forest (RF), Decision Tree (DT), Logistic Regression (LR), and Multilayer Perceptrons (MLP), were implemented as solo models and in a two-stage ensemble hybrid approach. The models were trained and validated using 10-fold cross-validation. SHapley Additive exPlanations (SHAP) were employed to interpret the contributions of each feature to the model predictions. Statistical analyses were conducted using the Shapiro-Wilk test for normality, the Kruskal-Wallis H test for group differences, and the Mann-Whitney U test with Bonferroni correction for post-hoc comparisons. RESULTS: The hybrid SVC + MLP model achieved the highest performance, with an accuracy of 89.58%, a precision of 87.18%, an F1-score of 88.20%, and an F-beta score of 87.55%. SHAP analysis revealed that glucose, glycine, and age were consistently important features across all DR classes, while creatinine and various phosphatidylcholines exhibited higher importance in the PDR class, suggesting their potential as biomarkers for severe DR. CONCLUSION: The hybrid XAI models, particularly the SVC + MLP ensemble, demonstrated superior performance in predicting DR progression compared to solo models. The application of SHAP facilitates the interpretation of feature importance, providing valuable insights into the metabolic and physiological markers associated with different stages of DR. These findings highlight the potential of hybrid XAI models combined with explainable techniques for early detection, targeted interventions, and personalized treatment strategies in DR management.","author":[{"family":"Yağın","given":"Fatma"},{"family":"Çolak","given":"Cemil"},{"family":"Algarni","given":"Abdulmohsen"},{"family":"Görmez","given":"Yasin"},{"family":"Güldoğan","given":"Emek"},{"family":"Ardigò","given":"Luca"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/diagnostics14131364","URL":"https://doi.org/10.3390/diagnostics14131364","source":"openalex"},{"id":"oa:W4388827078","type":"article-journal","title":"The Right to Transparency in Public Governance: Freedom of Information and the Use of Artificial Intelligence by Public Agencies","abstract":"What information should and can be transparent for artificial intelligence (AI) algorithms? This article examines the socio-technical and legal perspectives of transparency in relation to algorithmic decision-making in public administration. We show how transparency in AI can be understood in light of the various technologies and the challenges one may encounter. Despite some first steps in that direction, there exists so far no mature standard for documenting AI models. From a legal perspective, this article examined the applicable freedom of information (FOI) regimes across different jurisdictions, with a particular focus on Denmark and other Scandinavian countries. Despite notable differences, our findings show that the FOI regimes generally only grant access to existing documents, and that access can be denied on the basis of the wide proprietary interests and internal documents exemptions. This is why we ultimately conclude that the European data-protection framework and the proposed EU AI Act — with their far-reaching duties to document the functioning of AI systems — provide promising new avenues for research and insights into transparency in AI.","author":[{"family":"Olsen","given":"Henrik"},{"family":"Hildebrandt","given":"Thomas"},{"family":"Wiesener","given":"Cornelius"},{"family":"Larsen","given":"Matthias"},{"family":"Flügge","given":"Asbjørn"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1145/3632753","URL":"https://doi.org/10.1145/3632753","source":"openalex"},{"id":"oa:W4400037209","type":"article-journal","title":"Optimizing Nursing Productivity: Exploring the Role of Artificial Intelligence, Technology Integration, Competencies, and Leadership","abstract":"Background: In the rapidly evolving healthcare management landscape, technology integration and artificial intelligence utilization play pivotal roles in shaping employee productivity. This research investigates these dynamics within Riyadh Province, Kingdom of Saudi Arabia, focusing on the relationships between technology integration, the use of artificial intelligence in nursing profession, nursing workforce competencies, technological leadership, and employee productivity. Methods: A quantitative approach was employed, involving 329 nurses from five hospitals in Riyadh Province. Partial Least Squares Structural Equation Modeling facilitated comprehensive analysis of direct and indirect relationships among variables. Results: Findings reveal that technology integration significantly enhances nursing productivity, while the use of artificial intelligence initially presents disruptions before yielding productivity gains. Nursing workforce competencies mediate these relationships, emphasizing the critical role of workforce readiness in harnessing technology's benefits. Surprisingly, technological leadership did not significantly moderate these effects. Conclusions: This research offers vital insights for healthcare organizations, advocating strategic technology integration and workforce development. It underscores the significance of nursing competencies in navigating technological transformations and affirms the enduring importance of leadership in guiding these changes. As healthcare evolves, these findings provide guidance for optimizing technology and artificial intelligence to enhance employee productivity and patient care.","author":[{"family":"Alenezi","given":"Atallah"},{"family":"Alshammari","given":"Mohammed"},{"family":"Ibrahim","given":"Ibrahim"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1155/2024/8371068","URL":"https://doi.org/10.1155/2024/8371068","source":"openalex"},{"id":"oa:W4396581755","type":"article-journal","title":"Deep Learning Approaches for Medical Image Analysis and Diagnosis","abstract":"In addition to enhancing diagnostic accuracy, deep learning techniques offer the potential to streamline workflows, reduce interpretation time, and ultimately improve patient outcomes. The scalability and adaptability of deep learning algorithms enable their deployment across diverse clinical settings, ranging from radiology departments to point-of-care facilities. Furthermore, ongoing research efforts focus on addressing the challenges of data heterogeneity, model interpretability, and regulatory compliance, paving the way for seamless integration of deep learning solutions into routine clinical practice. As the field continues to evolve, collaborations between clinicians, data scientists, and industry stakeholders will be paramount in harnessing the full potential of deep learning for advancing medical image analysis and diagnosis. Furthermore, the integration of deep learning algorithms with other technologies, including natural language processing and computer vision, may foster multimodal medical data analysis and clinical decision support systems to improve patient care. The future of deep learning in medical image analysis and diagnosis is promising. With each success and advancement, this technology is getting closer to being leveraged for medical purposes. Beyond medical image analysis, patient care pathways like multimodal imaging, imaging genomics, and intelligent operating rooms or intensive care units can benefit from deep learning models.","author":[{"family":"Thakur","given":"Gopal"},{"family":"Thakur","given":"Abhishek"},{"family":"Kulkarni","given":"Shridhar"},{"family":"Khan","given":"Naseebia"},{"family":"Khan","given":"Shahnawaz"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.59507","URL":"https://doi.org/10.7759/cureus.59507","source":"openalex"},{"id":"oa:W4403078377","type":"article-journal","title":"Applications of artificial intelligence in interventional oncology: An up-to-date review of the literature","abstract":"Interventional oncology provides image-guided therapies, including transarterial tumor embolization and percutaneous tumor ablation, for malignant tumors in a minimally invasive manner. As in other medical fields, the application of artificial intelligence (AI) in interventional oncology has garnered significant attention. This narrative review describes the current state of AI applications in interventional oncology based on recent literature. A literature search revealed a rapid increase in the number of studies relevant to this topic recently. Investigators have attempted to use AI for various tasks, including automatic segmentation of organs, tumors, and treatment areas; treatment simulation; improvement of intraprocedural image quality; prediction of treatment outcomes; and detection of post-treatment recurrence. Among these, the AI-based prediction of treatment outcomes has been the most studied. Various deep and conventional machine learning algorithms have been proposed for these tasks. Radiomics has often been incorporated into prediction and detection models. Current literature suggests that AI is potentially useful in various aspects of interventional oncology, from treatment planning to post-treatment follow-up. However, most AI-based methods discussed in this review are still at the research stage, and few have been implemented in clinical practice. To achieve widespread adoption of AI technologies in interventional oncology procedures, further research on their reliability and clinical utility is necessary. Nevertheless, considering the rapid research progress in this field, various AI technologies will be integrated into interventional oncology practices in the near future.","author":[{"family":"Matsui","given":"Yusuke"},{"family":"Ueda","given":"Daiju"},{"family":"Fujita","given":"Shohei"},{"family":"Fushimi","given":"Yasutaka"},{"family":"Tsuboyama","given":"Takahiro"},{"family":"Kamagata","given":"Koji"},{"family":"Ito","given":"Rintaro"},{"family":"Yanagawa","given":"Masahiro"},{"family":"Yamada","given":"Akira"},{"family":"Kawamura","given":"Mariko"},{"family":"Nakaura","given":"Takeshi"},{"family":"Fujima","given":"Noriyuki"},{"family":"Nozaki","given":"Taiki"},{"family":"Tatsugami","given":"Fuminari"},{"family":"Fujioka","given":"Tomoyuki"},{"family":"Hirata","given":"Kenji"},{"family":"Naganawa","given":"Shinji"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s11604-024-01668-3","URL":"https://doi.org/10.1007/s11604-024-01668-3","source":"openalex"},{"id":"oa:W4404228387","type":"article-journal","title":"Representation of intensivists’ race/ethnicity, sex, and age by artificial intelligence: a cross-sectional study of two text-to-image models","abstract":"BACKGROUND: Integrating artificial intelligence (AI) into intensive care practices can enhance patient care by providing real-time predictions and aiding clinical decisions. However, biases in AI models can undermine diversity, equity, and inclusion (DEI) efforts, particularly in visual representations of healthcare professionals. This work aims to examine the demographic representation of two AI text-to-image models, Midjourney and ChatGPT DALL-E 2, and assess their accuracy in depicting the demographic characteristics of intensivists. METHODS: This cross-sectional study, conducted from May to July 2024, used demographic data from the USA workforce report (2022) and intensive care trainees (2021) to compare real-world intensivist demographics with images generated by two AI models, Midjourney v6.0 and ChatGPT 4.0 DALL-E 2. A total of 1,400 images were generated across ICU subspecialties, with outcomes being the comparison of sex, race/ethnicity, and age representation in AI-generated images to the actual workforce demographics. RESULTS: The AI models demonstrated noticeable biases when compared to the actual U.S. intensive care workforce data, notably overrepresenting White and young doctors. ChatGPT-DALL-E2 produced less female (17.3% vs 32.2%, p < 0.0001), more White (61% vs 55.1%, p = 0.002) and younger (53.3% vs 23.9%, p < 0.001) individuals. While Midjourney depicted more female (47.6% vs 32.2%, p < 0.001), more White (60.9% vs 55.1%, p = 0.003) and younger intensivist (49.3% vs 23.9%, p < 0.001). Substantial differences between the specialties within both models were observed. Finally when compared together, both models showed significant differences in the Portrayal of intensivists. CONCLUSIONS: Significant biases in AI images of intensivists generated by ChatGPT DALL-E 2 and Midjourney reflect broader cultural issues, potentially perpetuating stereotypes of healthcare worker within the society. This study highlights the need for an approach that ensures fairness, accountability, transparency, and ethics in AI applications for healthcare.","author":[{"family":"Gisselbaek","given":"Mia"},{"family":"Suppan","given":"Mélanie"},{"family":"Minsart","given":"Laurens"},{"family":"Köselerli","given":"Ekin"},{"family":"Myatra","given":"Sheila"},{"family":"Matot","given":"Idit"},{"family":"Chang","given":"Odmara"},{"family":"Saxena","given":"Sarah"},{"family":"Bergerestilita","given":"Joana"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s13054-024-05134-4","URL":"https://doi.org/10.1186/s13054-024-05134-4","source":"openalex"},{"id":"oa:W4402500282","type":"article-journal","title":"Challenges and opportunities to integrate artificial intelligence in radiation oncology: a narrative review","abstract":"Artificial intelligence (AI) is rapidly transforming various medical fields, including radiation oncology. This review explores the integration of AI into radiation oncology, highlighting both challenges and opportunities. AI can improve the precision, efficiency, and outcomes of radiation therapy by optimizing treatment planning, enhancing image analysis, facilitating adaptive radiation therapy, and enabling predictive analytics. Through the analysis of large datasets to identify optimal treatment parameters, AI can automate complex tasks, reduce planning time, and improve accuracy. In image analysis, AI-driven techniques enhance tumor detection and segmentation by processing data from CT, MRI, and PET scans to enable precise tumor delineation. In adaptive radiation therapy, AI is beneficial because it allows real-time adjustments to treatment plans based on changes in patient anatomy and tumor size, thereby improving treatment accuracy and effectiveness. Predictive analytics using historical patient data can predict treatment outcomes and potential complications, guiding clinical decision-making and enabling more personalized treatment strategies. Challenges to AI adoption in radiation oncology include ensuring data quality and quantity, achieving interoperability and standardization, addressing regulatory and ethical considerations, and overcoming resistance to clinical implementation. Collaboration among researchers, clinicians, data scientists, and industry stakeholders is crucial to overcoming these obstacles. By addressing these challenges, AI can drive advancements in radiation therapy, improving patient care and operational efficiencies. This review presents an overview of the current state of AI integration in radiation oncology and insights into future directions for research and clinical practice.","author":[{"family":"Jeong","given":"C"},{"family":"Goh","given":"YM"},{"family":"Kwak","given":"Jungwon"}],"issued":{"date-parts":[[2024]]},"DOI":"10.12771/emj.2024.e49","URL":"https://doi.org/10.12771/emj.2024.e49","source":"openalex"},{"id":"oa:W4391075605","type":"article-journal","title":"Artificial Intelligence (AI) in academic research. A multi-group analysis of students’ awareness and perceptions using gender and programme type","abstract":"The era of AI has brought tremendous impact in academic research, and this has provided the impetus for students to leverage on novel tools in carrying out a lot of quality research works. Previous studies have relied so much on AI for instruction, classroom management and assessment and utilisation of AI tools for research has scarcely been examined. This study covered the gap by examining students’ utilisation of AI tools based on their level of awareness and perception and finding out the difference based on gender and programme type in such prediction. A total of 5554 university students were used for the study. Exploratory factor analysis was first carried for dimensionality and other validity checks (convergent and discriminant) using Average Variance Extracted (AVE) and Fornel-Larcker criterion and methods. Population t-tests and multi-group analyses were performed using SPSS and Smart PLS 3. The study found that students have high level of awareness and positive perception of AI tools in research. Similarly, the level of utilisation of AI tools in research is high. Male and postgraduate students have a higher level of awareness and positive perception of AI tools in research, with female students stronger than male students in terms. Perception and awareness directly impacted on utilisation but perception mediates positively and significantly in the nexus between awareness and utilisation. The study findings provide useful insights into using AI tools among university students and also identify the rationale to consider variables like gender and programme type when developing curriculum that will meet the current technology needs in our higher institutions.","author":[{"family":"Joseph","given":"Usani"},{"family":"Arikpo","given":"Iyam"},{"family":"Victor","given":"Ovat"},{"family":"Chidirim","given":"Nwogwugwu"},{"family":"Mbua","given":"Anake"},{"family":"Ify","given":"Udeh"},{"family":"Diwa","given":"Out"}],"issued":{"date-parts":[[2024]]},"DOI":"10.37074/jalt.2024.7.1.9","URL":"https://doi.org/10.37074/jalt.2024.7.1.9","source":"openalex"},{"id":"oa:W4404196273","type":"article-journal","title":"THE role of Artificial Intelligence in industry 5.0: Enhancing human-machine collaboration","abstract":"The emergence of Industry 5.0 marks a transformative shift in the manufacturing landscape, emphasizing a synergistic relationship between humans and machines. This paper explores the pivotal role of Artificial Intelligence (AI) in enhancing human-machine collaboration within this paradigm. By leveraging AI technologies, industries can foster a more personalized, efficient, and innovative work environment. AI systems facilitate seamless communication and decision-making processes, thereby augmenting human capabilities rather than replacing them. This collaboration allows for the optimization of workflows, reduction of operational risks, and enhancement of product quality through advanced predictive analytics and real-time data processing. Furthermore, the integration of AI in Industry 5.0 supports sustainability initiatives by minimizing waste and energy consumption, aligning with the global push for greener manufacturing practices. Case studies demonstrate the successful implementation of AI-driven solutions across various sectors, showcasing improvements in productivity and employee satisfaction. As Industry 5.0 continues to evolve, the interplay between AI and human labour will redefine traditional roles, empowering workers with augmented intelligence tools. The findings indicate that embracing AI not only enhances operational efficiency but also contributes to a more resilient and adaptive workforce. Ultimately, this paper posits that the future of industry lies in the harmonious collaboration between human intellect and artificial intelligence, which together will drive innovation, productivity, and sustainable practices in the manufacturing sector.","author":[{"family":"Anang","given":"Andrew"},{"family":"Obidi","given":"Peter"},{"family":"Mesogboriwon","given":"Adeleye"},{"family":"Obidi","given":"James"},{"family":"Kuubata","given":"Maurice"},{"family":"Ogunbiyi","given":"Dabira"}],"issued":{"date-parts":[[2024]]},"DOI":"10.30574/wjarr.2024.24.2.3369","URL":"https://doi.org/10.30574/wjarr.2024.24.2.3369","source":"openalex"},{"id":"oa:W4400805135","type":"article-journal","title":"Validation of a novel, low-fidelity virtual reality simulator and an artificial intelligence assessment approach for peg transfer laparoscopic training","abstract":"Simulators are widely used in medical education, but objective and automatic assessment is not feasible with low-fidelity simulators, which can be solved with artificial intelligence (AI) and virtual reality (VR) solutions. The effectiveness of a custom-made VR simulator and an AI-based evaluator of a laparoscopic peg transfer exercise was investigated. Sixty medical students were involved in a single-blinded randomised controlled study to compare the VR simulator with the traditional box trainer. A total of 240 peg transfer exercises from the Fundamentals of Laparoscopic Surgery programme were analysed. The experts and AI-based software used the same criteria for evaluation. The algorithm detected pitfalls and measured exercise duration. Skill improvement showed no significant difference between the VR and control groups. The AI-based evaluator exhibited 95% agreement with the manual assessment. The average difference between the exercise durations measured by the two evaluation methods was 2.61 s. The duration of the algorithmic assessment was 59.47 s faster than the manual assessment. The VR simulator was an effective alternative practice compared with the training box simulator. The AI-based evaluation produced similar results compared with the manual assessment, and it could significantly reduce the evaluation time. AI and VR could improve the effectiveness of basic laparoscopic training.","author":[{"family":"Bogar","given":"Peter"},{"family":"Virag","given":"Mark"},{"family":"Bene","given":"Mátyás"},{"family":"Hardi","given":"Péter"},{"family":"Matuz","given":"András"},{"family":"Schlégl","given":"Ádám"},{"family":"Tóth","given":"Luca"},{"family":"Molnár","given":"Ferenc"},{"family":"Nagy","given":"Bálint"},{"family":"Rendeki","given":"Szilárd"},{"family":"Berner-Juhos","given":"Krisztina"},{"family":"Ferencz","given":"Andrea"},{"family":"Fischer","given":"Krisztina"},{"family":"Maróti","given":"Péter"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41598-024-67435-6","URL":"https://doi.org/10.1038/s41598-024-67435-6","source":"openalex"},{"id":"oa:W4392981548","type":"article-journal","title":"Ecotoxicological impacts of landfill sites: Towards risk assessment, mitigation policies and the role of artificial intelligence","abstract":"Waste disposal in landfills remains a global concern. Despite technological developments, landfill leachate poses a hazard to ecosystems and human health since it acts as a secondary reservoir for legacy and emerging pollutants. This study provides a systematic and scientometric review of the nature and toxicity of pollutants generated by landfills and means of assessing their potential risks. Regarding human health, unregulated waste disposal and pathogens in leachate are the leading causes of diseases reported in local populations. Both in vitro and in vivo approaches have been employed in the ecotoxicological risk assessment of landfill leachate, with model organisms ranging from bacteria to birds. These studies demonstrate a wide range of toxic effects that reflect the complex composition of leachate and geographical variations in climate, resource availability and management practices. Based on bioassay (and other) evidence, categories of persistent chemicals of most concern include brominated flame retardants, per- and polyfluorinated chemicals, pharmaceuticals and alkyl phenol ethoxylates. However, the emerging and more general literature on microplastic toxicity suggests that these particles might also be problematic in leachate. Various mitigation strategies have been identified, with most focussing on improving landfill design or leachate treatment, developing alternative disposal methods and reducing waste volume through recycling or using more sustainable materials. The success of these efforts will rely on policies and practices and their enforcement, which is seen as a particular challenge in developing nations and at the international (and transboundary) level. Artificial intelligence and machine learning afford a wide range of options for evaluating and reducing the risks associated with leachates and gaseous emissions from landfills, and various approaches tested or having potential are discussed. However, addressing the limitations in data collection, model accuracy, real-time monitoring and our understanding of environmental impacts will be critical for realising this potential.","author":[{"family":"Gautam","given":"Krishna"},{"family":"Pandey","given":"Namrata"},{"family":"Yadav","given":"Dhvani"},{"family":"Parthasarathi","given":"Ramakrishnan"},{"family":"Turner","given":"Andrew"},{"family":"Anbumani","given":"Sadasivam"},{"family":"Jha","given":"Awadhesh"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.scitotenv.2024.171804","URL":"https://doi.org/10.1016/j.scitotenv.2024.171804","source":"openalex"},{"id":"oa:W4395070146","type":"article-journal","title":"eXplainable Artificial Intelligence (XAI) for improving organisational regility","abstract":"Since the pandemic started, organisations have been actively seeking ways to improve their organisational agility and resilience (regility) and turn to Artificial Intelligence (AI) to gain a deeper understanding and further enhance their agility and regility. Organisations are turning to AI as a critical enabler to achieve these goals. AI empowers organisations by analysing large data sets quickly and accurately, enabling faster decision-making and building agility and resilience. This strategic use of AI gives businesses a competitive advantage and allows them to adapt to rapidly changing environments. Failure to prioritise agility and responsiveness can result in increased costs, missed opportunities, competition and reputational damage, and ultimately, loss of customers, revenue, profitability, and market share. Prioritising can be achieved by utilising eXplainable Artificial Intelligence (XAI) techniques, illuminating how AI models make decisions and making them transparent, interpretable, and understandable. Based on previous research on using AI to predict organisational agility, this study focuses on integrating XAI techniques, such as Shapley Additive Explanations (SHAP), in organisational agility and resilience. By identifying the importance of different features that affect organisational agility prediction, this study aims to demystify the decision-making processes of the prediction model using XAI. This is essential for the ethical deployment of AI, fostering trust and transparency in these systems. Recognising key features in organisational agility prediction can guide companies in determining which areas to concentrate on in order to improve their agility and resilience.","author":[{"family":"Shafiabady","given":"Niusha"},{"family":"Hadjinicolaou","given":"Nick"},{"family":"Hettikankanamage","given":"Nadeesha"},{"family":"Mohammadisavadkoohi","given":"Ehsan"},{"family":"Wu","given":"Mingxuan"},{"family":"Vakilian","given":"James"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1371/journal.pone.0301429","URL":"https://doi.org/10.1371/journal.pone.0301429","source":"openalex"},{"id":"oa:W4403002502","type":"article-journal","title":"Perspectives on Intelligence in Soft Robotics","abstract":"Engineers frequently aim to streamline environmental factors to facilitate the effective operation of robots. However, in nature, environmental considerations play a crucial role in shaping the embodiment of organisms. To comply robots with the complexity of real‐world environments, embedding similar intelligence is key. In the field of soft robotics, various approaches offer insight into how intelligence can be integrated into artificial agents. A discussed topic is the intricate relationship between the brain and the body at the core of intelligence in robots. The goal of this article is, therefore, to unravel the strategies to implement different types of intelligence currently adopted in soft robots. A classification is made by making a distinction between agents that adapt to their environment by 1) their adaptive shape, 2) their adaptive functionality, and 3) their adaptive mechanics. Additionally, the perspectives on intelligence based on their computational approach are distinguished: centralized computation, decentralized computation, or embedded computation. It is concluded that a tailored robotic design approach attuned to specific environmental demands is needed. To unlock the full potential of soft robots, a fresh perspective on embodied intelligence is described, so‐called mechanical intelligence, emphasizing the robot's responsiveness to changing external conditions of a real‐world environment.","author":[{"family":"Kortman","given":"Vera"},{"family":"Mazzolai","given":"Barbara"},{"family":"Sakes","given":"Aimée"},{"family":"Jovanova","given":"Jovana"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/aisy.202400294","URL":"https://doi.org/10.1002/aisy.202400294","source":"openalex"},{"id":"oa:W4404780331","type":"article-journal","title":"Integrating Artificial Intelligence and Microfluidics Technology for Psoriasis Therapy: A Comprehensive Review for Research and Clinical Applications","abstract":"Microfluidics has evolved into a transformative technology with far‐reaching applications in biomedical research. However, designing and optimizing custom microfluidic systems remains challenging because of their inherent complexities. Integrating artificial intelligence (AI) with microfluidics promises to overcome these barriers by leveraging AI algorithms to automate device design, streamline experimentation, and enhance diagnostic and therapeutic outcomes. Psoriasis is an incurable dermatological condition that is difficult to diagnose and treat owing to its complex pathogenesis. Traditional diagnostic and therapeutic approaches are often ineffective and fail to address individual variabilities in disease progression and treatment responses. However, AI‐coupled microfluidic platforms have the potential to revolutionize psoriasis research and clinical applications with expansive dermatological applications. AI‐driven microfluidic chips with embedded biosensors have the potential to precisely detect biomarkers (BMs), manipulate biological samples, and mimic psoriasis‐like in vivo and in vitro models, thereby allowing real‐time monitoring and optimized therapeutic testing. This review examines the transformative potential of AI and AI‐powered microfluidic platforms for advancing psoriasis research. It examines the design and mechanisms of AI‐coupled microfluidic platforms for cell screening, disease diagnosis, and drug delivery. It highlights recent advances, clinical applications, challenges, future perspectives, and ethical considerations to enhance personalized care and patient outcomes.","author":[{"family":"Shaw","given":"Ibrahim"},{"family":"Ali","given":"Yimer"},{"family":"Changhong","given":"Nie"},{"family":"Zhang","given":"Kexin"},{"family":"Chen","given":"Chuanpin"},{"family":"Xiao","given":"Yin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/aisy.202400558","URL":"https://doi.org/10.1002/aisy.202400558","source":"openalex"},{"id":"oa:W4390778841","type":"article-journal","title":"Identifying Individuals at High Risk for HIV and Sexually Transmitted Infections With an Artificial Intelligence–Based Risk Assessment Tool","abstract":"Background: We have previously developed an artificial intelligence-based risk assessment tool to identify the individual risk of HIV and sexually transmitted infections (STIs) in a sexual health clinical setting. Based on this tool, this study aims to determine the optimal risk score thresholds to identify individuals at high risk for HIV/STIs. Methods: machine learning models to estimate infection risk scores. Optimal cutoffs for determining high-risk individuals were determined using Youden's index. Results: The HIV risk score cutoff for high risk was 0.56, with 86.0% sensitivity (95% CI, 82.9%-88.7%) and 65.6% specificity (95% CI, 65.4%-65.8%). Thirty-five percent of participants were classified as high risk, which accounted for 86% of HIV cases. The corresponding cutoffs were 0.49 for syphilis (sensitivity, 77.6%; 95% CI, 76.2%-78.9%; specificity, 78.1%; 95% CI, 77.9%-78.3%), 0.52 for gonorrhea (sensitivity, 78.3%; 95% CI, 77.6%-78.9%; specificity, 71.9%; 95% CI, 71.7%-72.0%), and 0.47 for chlamydia (sensitivity, 68.8%; 95% CI, 68.3%-69.4%; specificity, 63.7%; 95% CI, 63.5%-63.8%). High-risk groups identified using these thresholds accounted for 78% of syphilis, 78% of gonorrhea, and 69% of chlamydia cases. The odds of positivity were significantly higher in the high-risk group than otherwise across all infections: 11.4 (95% CI, 9.3-14.8) times for HIV, 12.3 (95% CI, 11.4-13.3) for syphilis, 9.2 (95% CI, 8.8-9.6) for gonorrhea, and 3.9 (95% CI, 3.8-4.0) for chlamydia. Conclusions: , together with Youden's index, are effective in determining high-risk subgroups for HIV/STIs. The thresholds can aid targeted HIV/STI screening and prevention.","author":[{"family":"Latt","given":"Phyu"},{"family":"Soe","given":"Nyi"},{"family":"Xu","given":"Xianglong"},{"family":"Ong","given":"Jason"},{"family":"Chow","given":"Eric"},{"family":"Fairley","given":"Christopher"},{"family":"Zhang","given":"Lei"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/ofid/ofae011","URL":"https://doi.org/10.1093/ofid/ofae011","source":"openalex"},{"id":"oa:W4400371216","type":"article-journal","title":"Assessing the Readability of Patient Education Materials on Cardiac Catheterization From Artificial Intelligence Chatbots: An Observational Cross-Sectional Study","abstract":"BACKGROUND: Artificial intelligence (AI) is a burgeoning new field that has increased in popularity over the past couple of years, coinciding with the public release of large language model (LLM)-driven chatbots. These chatbots, such as ChatGPT, can be engaged directly in conversation, allowing users to ask them questions or issue other commands. Since LLMs are trained on large amounts of text data, they can also answer questions reliably and factually, an ability that has allowed them to serve as a source for medical inquiries. This study seeks to assess the readability of patient education materials on cardiac catheterization across four of the most common chatbots: ChatGPT, Microsoft Copilot, Google Gemini, and Meta AI. METHODOLOGY: A set of 10 questions regarding cardiac catheterization was developed using website-based patient education materials on the topic. We then asked these questions in consecutive order to four of the most common chatbots: ChatGPT, Microsoft Copilot, Google Gemini, and Meta AI. The Flesch Reading Ease Score (FRES) was used to assess the readability score. Readability grade levels were assessed using six tools: Flesch-Kincaid Grade Level (FKGL), Gunning Fog Index (GFI), Coleman-Liau Index (CLI), Simple Measure of Gobbledygook (SMOG) Index, Automated Readability Index (ARI), and FORCAST Grade Level. RESULTS: The mean FRES across all four chatbots was 40.2, while overall mean grade levels for the four chatbots were 11.2, 13.7, 13.7, 13.3, 11.2, and 11.6 across the FKGL, GFI, CLI, SMOG, ARI, and FORCAST indices, respectively. Mean reading grade levels across the six tools were 14.8 for ChatGPT, 12.3 for Microsoft Copilot, 13.1 for Google Gemini, and 9.6 for Meta AI. Further, FRES values for the four chatbots were 31, 35.8, 36.4, and 57.7, respectively. CONCLUSIONS: -grade level, depending on the tool used. This means that the materials were at the high school and even college reading level, which far exceeds the recommended sixth-grade level for patient education materials. Further, there is significant variability in the readability levels provided by different chatbots as, across all six grade-level assessments, Meta AI had the lowest scores and ChatGPT generally had the highest.","author":[{"family":"Behers","given":"Benjamin"},{"family":"Vargas","given":"Ian"},{"family":"Behers","given":"Brett"},{"family":"Rosario","given":"Manuel"},{"family":"Wojtas","given":"Caroline"},{"family":"Deevers","given":"Alexander"},{"family":"Hamad","given":"Karen"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.63865","URL":"https://doi.org/10.7759/cureus.63865","source":"openalex"},{"id":"oa:W4404435192","type":"article-journal","title":"The contagion effect of artificial intelligence across innovative industries: From blockchain and metaverse to cleantech and beyond","abstract":"Artificial Intelligence (AI) stands as a transformative force across business, technology, and science, yet its comprehensive impact on innovative industries remains relatively unexplored. This study delves into the interconnectedness between AI and pivotal sectors such as cryptocurrency , blockchain, metaverse, democratized banking, and Cleantech, among others. Employing the conditional autoregressive value-at-risk (CAViaR) and time-varying parameters vector autoregressions (TVP-VAR) methods, we scrutinize daily data spanning from June 1, 2018, to October 11, 2023, encompassing 12 stock indices representing each industry. Our findings unveil a strong contagion effect from AI to other innovative sectors, with the exception of Cleantech, which appears to have decoupled from the AI surge. Notably, democratized banking and the metaverse emerge as key recipients of this contagion. Examination of tail-risk spillovers highlights AI as one of the most influential risk transmitters during market tumult, while cryptocurrency and blockchain consistently function as net risk receivers throughout the sample period. The implications of these findings are multifaceted, offering substantive insights into the risk profiles of these critical innovative sectors. Investors and regulatory bodies stand to benefit significantly from this analysis, as it illuminates potential avenues for portfolio diversification and deepens understanding of contagion mechanisms within these evolving industries.","author":[{"family":"Naeem","given":"Muhammad"},{"family":"Arfaoui","given":"Nadia"},{"family":"Yarovaya","given":"Larisa"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.techfore.2024.123822","URL":"https://doi.org/10.1016/j.techfore.2024.123822","source":"openalex"},{"id":"oa:W4386892967","type":"article-journal","title":"Evaluating the performance of ChatGPT-4 on the United Kingdom Medical Licensing Assessment","abstract":"Introduction: Recent developments in artificial intelligence large language models (LLMs), such as ChatGPT, have allowed for the understanding and generation of human-like text. Studies have found LLMs abilities to perform well in various examinations including law, business and medicine. This study aims to evaluate the performance of ChatGPT in the United Kingdom Medical Licensing Assessment (UKMLA). Methods: Two publicly available UKMLA papers consisting of 200 single-best-answer (SBA) questions were screened. Nine SBAs were omitted as they contained images that were not suitable for input. Each question was assigned a specialty based on the UKMLA content map published by the General Medical Council. A total of 191 SBAs were inputted in ChatGPT-4 through three attempts over the course of 3 weeks (once per week). Results: ChatGPT scored 74.9% (143/191), 78.0% (149/191) and 75.6% (145/191) on three attempts, respectively. The average of all three attempts was 76.3% (437/573) with a 95% confidence interval of (74.46% and 78.08%). ChatGPT answered 129 SBAs correctly and 32 SBAs incorrectly on all three attempts. On three attempts, ChatGPT performed well in mental health (8/9 SBAs), cancer (11/14 SBAs) and cardiovascular (10/13 SBAs). On three attempts, ChatGPT did not perform well in clinical haematology (3/7 SBAs), endocrine and metabolic (2/5 SBAs) and gastrointestinal including liver (3/10 SBAs). Regarding to response consistency, ChatGPT provided correct answers consistently in 67.5% (129/191) of SBAs but provided incorrect answers consistently in 12.6% (24/191) and inconsistent response in 19.9% (38/191) of SBAs, respectively. Discussion and conclusion: This study suggests ChatGPT performs well in the UKMLA. There may be a potential correlation between specialty performance. LLMs ability to correctly answer SBAs suggests that it could be utilised as a supplementary learning tool in medical education with appropriate medical educator supervision.","author":[{"family":"Lai","given":"UH"},{"family":"Wu","given":"Keng"},{"family":"Hsu","given":"Ting"},{"family":"Kan","given":"Jessie"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3389/fmed.2023.1240915","URL":"https://doi.org/10.3389/fmed.2023.1240915","source":"openalex"},{"id":"oa:W4402221781","type":"article-journal","title":"Understanding the skills gap between higher education and industry in the UK in artificial intelligence sector","abstract":"As Artificial Intelligence (AI) changes how businesses work, there’s a growing need for people who can work in this sector. This paper investigates how well universities in United Kingdom offering courses in AI, prepare students for jobs in the real world. To gain insight into the differences between university curricula and industry demands we review the contents of taught courses and job advertisement portals. By using custom data scraping tools to gather information from job advertisements and university curricula, and frequency and Naive Bayes classifier analysis, this study will show exactly what skills industry is looking for. In this study we identified 12 skill categories that were used for mapping. The study showed that the university curriculum in the AI domain is well balanced in most technical skills, including Programming and Machine learning subjects, but have a gap in Data Science and Maths and Statistics sk\\ill categories.","author":[{"family":"Jaiswal","given":"Khushi"},{"family":"Kuzminykh","given":"Ievgeniia"},{"family":"Modgil","given":"Sanjay"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1177/09504222241280441","URL":"https://doi.org/10.1177/09504222241280441","source":"openalex"},{"id":"oa:W4404159957","type":"article-journal","title":"Optimization of diagnosis and treatment of hematological diseases via artificial intelligence","abstract":"Background: Optimizing the diagnosis and treatment of hematological diseases is a challenging yet crucial research area. Effective treatment plans typically require the comprehensive integration of cell morphology, immunology, cytogenetics, and molecular biology. These plans also consider patient-specific factors such as disease stage, age, and genetic mutation status. With the advancement of artificial intelligence (AI), more \"AI + medical\" application models are emerging. In clinical practice, many AI-assisted systems have been successfully applied to the diagnosis and treatment of hematological diseases, enhancing precision and efficiency and offering valuable solutions for clinical practice. Objective: This study summarizes the research progress of various AI-assisted systems applied in the clinical diagnosis and treatment of hematological diseases, with a focus on their application in morphology, immunology, cytogenetics, and molecular biology diagnosis, as well as prognosis prediction and treatment. Methods: Using PubMed, Web of Science, and other network search engines, we conducted a literature search on studies from the past 5 years using the main keywords \"artificial intelligence\" and \"hematological diseases.\" We classified the clinical applications of AI systems according to the diagnosis and treatment. We outline and summarize the current advancements in AI for optimizing the diagnosis and treatment of hematological diseases, as well as the difficulties and challenges in promoting the standardization of clinical diagnosis and treatment in this field. Results: AI can significantly shorten turnaround times, reduce diagnostic costs, and accurately predict disease outcomes through applications in image-recognition technology, genomic data analysis, data mining, pattern recognition, and personalized medicine. However, several challenges remain, including the lack of AI product standards, standardized data, medical-industrial collaboration, and the complexity and non-interpretability of AI systems. In addition, regulatory gaps can lead to data privacy issues. Therefore, more research and improvements are needed to fully leverage the potential of AI to promote standardization of the clinical diagnosis and treatment of hematological diseases. Conclusion: Our results serve as a reference point for the clinical diagnosis and treatment of hematological diseases and the development of AI-assisted clinical diagnosis and treatment systems. We offer suggestions for further development of AI in hematology and standardization of clinical diagnosis and treatment.","author":[{"family":"Wang","given":"Shixuan"},{"family":"Huang","given":"Zoufang"},{"family":"Li","given":"Jing"},{"family":"Wu","given":"Yin"},{"family":"Du","given":"Jun"},{"family":"Li","given":"Ting"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fmed.2024.1487234","URL":"https://doi.org/10.3389/fmed.2024.1487234","source":"openalex"},{"id":"oa:W4396870875","type":"article-journal","title":"Artificial intelligence applied to MRI data to tackle key challenges in multiple sclerosis","abstract":"Artificial intelligence (AI) is the branch of science aiming at creating algorithms able to carry out tasks that typically require human intelligence. In medicine, there has been a tremendous increase in AI applications thanks to increasingly powerful computers and the emergence of big data repositories. Multiple sclerosis (MS) is a chronic autoimmune condition affecting the central nervous system with a complex pathogenesis, a challenging diagnostic process strongly relying on magnetic resonance imaging (MRI) and a high and largely unexplained variability across patients. Therefore, AI applications in MS have the great potential of helping us better support the diagnosis, find markers for prognosis to eventually design more powerful randomised clinical trials and improve patient management in clinical practice and eventually understand the mechanisms of the disease. This topical review aims to summarise the recent advances in AI applied to MRI data in MS to illustrate its achievements, limitations and future directions.","author":[{"family":"Collorone","given":"Sara"},{"family":"Coll","given":"Llucia"},{"family":"Lorenzi","given":"Marco"},{"family":"Lladó","given":"Xavier"},{"family":"Sastregarriga","given":"Jaume"},{"family":"Tintoré","given":"Mar"},{"family":"Montalbán","given":"Xavier"},{"family":"Rovira","given":"Àlex"},{"family":"Pareto","given":"Deborah"},{"family":"Tur","given":"Carmen"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1177/13524585241249422","URL":"https://doi.org/10.1177/13524585241249422","source":"openalex"},{"id":"oa:W4390695989","type":"article-journal","title":"First clinical data on artificial intelligence‐guided catheter ablation in long‐standing persistent atrial fibrillation","abstract":"INTRODUCTION: Despite advanced ablation strategies and major technological improvements, treatment of persistent atrial fibrillation (AF) remains challenging and the underlying pathophysiology is not fully understood. This study analyzed the multiple procedure outcome and safety of catheter ablation of spatiotemporal dispersions (DISPERS) detected by artificial intelligence (AI)-guided software in patients with long-standing persistent AF. METHODS AND RESULTS: The Volta VX1 software was used for 50 consecutive patients undergoing catheter ablation for persistent AF. First, high-density mapping (78% biatrial) with a multipolar mapping catheter was performed. In addition to pulmonary vein isolation (PVI), ablation of DISPERS was performed aiming at homogenizing, dissecting, isolating, or connecting DISPERS areas to nonconducting anatomical structures. Follow-up contained regular visits at our outpatient clinic at 1, 3, 6, and 12 months including 7-day Holter electrocardiograms. Patients were mainly suffering from long-standing persistent AF (mean AF duration 50.30 ± 54.28 months). Following PVI, ablation of left atrial and right atrial DISPERS areas led to AF cycle length prolongation (mean of 162.0 ± 16.6 to 202.2 ± 21.6 ms after) and AF termination to atrial tachycardia (AT) or sinus rhythm (SR) in 12 patients (24%). No stroke or pericardial effusion occurred; major groin complications (pseudoaneurysm n = 1, atrioventricular fistula n = 1) were detected in two patients. After a blanking period of 6 weeks, recurrence of any atrial arrhythmia was documented in 26 patients (52%). The majority of patients presented with organized AT (n = 15) while AF was present in n = 9 patients and AT/AF was observed in n = 2 patients. Twenty-two patients underwent reablation. During a mean follow-up of 363.14 ± 187.42 days and after an average of 1.46 ± 0.68 procedures, 82% of patients remained in stable SR. CONCLUSION: DISPERS-guided ablation using machine learning software (the Volta VX1 software) in addition to PVI in long-standing persistent AF ablation resulted in high long-term success rates regarding AF and AT elimination. Most arrhythmia recurrences were reentrant AT. After a total of 1.46 ± 0.68 procedures, freedom from AF/AT was 82%. Despite prolonged procedure times complication rates were low. Randomized studies are necessary to evaluate long-term efficacy of dispersion-guided ablation using AI.","author":[{"family":"Bahlke","given":"Fabian"},{"family":"Englert","given":"Florian"},{"family":"Popa","given":"Miruna"},{"family":"Bourier","given":"Félix"},{"family":"Reents","given":"Tilko"},{"family":"Lennerz","given":"Carsten"},{"family":"Kraft","given":"Hannah"},{"family":"Martinez","given":"Alex"},{"family":"Kottmaier","given":"Marc"},{"family":"Syväri","given":"Jan"},{"family":"Tydecks","given":"Madeleine"},{"family":"Telishevska","given":"Marta"},{"family":"Lengauer","given":"Sarah"},{"family":"Hessling","given":"Gabriele"},{"family":"Deisenhofer","given":"Isabel"},{"family":"Erhard","given":"Nico"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/jce.16184","URL":"https://doi.org/10.1111/jce.16184","source":"openalex"},{"id":"oa:W4398780684","type":"article-journal","title":"Using artificial intelligence chatbots to improve patient history taking in dental education (Pilot study)","abstract":"In all health professions, taking a patient history is critical to accurate diagnosis and the formulation of an appropriate patient-focused management plan.1 Misdiagnosis and inappropriate treatment are common sequelae when clinicians fail to work through history in a logical and thorough manner. Traditionally, history taking in dentistry is taught in a clinical setting with some opportunities to practice in a simulated environment prior to students entering treatment clinics. Pre-clinical teaching is currently done in a tutorial setting where a clinical educator acts as a patient and the students act as the dentist. Such simulations allow repetitive practice in different clinical situations, within a risk-free environment.2 However, interviews with dental school staff members and observation of these tutorials revealed that in most classes, only one or two students actively participated in this role-playing. ChatGPT is a publicly available artificial intelligence (AI) large language model from OpenAI.3 It receives natural language inputs and responds realistically based on the conversation context. AI can be used to develop educational chatbots that are affordable and can provide a new medium of simulation with more scalability, personalization, and access.4 Research studies have demonstrated the effectiveness of ChatGPT in improving diagnostic accuracy, clinical judgment, and knowledge retention among healthcare learners.4 The use of an AI-generated chatbot is proposed as a solution to these problems by requiring individual interaction with the chatbot prior to clinical training. An educational chatbot was developed and trialed in a group of third-year Doctor of Dental Medicine students. A history-taking chatbot was developed with the student user/“clinician” taking a history from the chatbot/“patient” Students could interact either verbally or by typing. The chatbot was hosted on Vercel, using the Next.js framework for the front end and utilizing Vercel's native serverless functions for access to OpenAI's Application Programming Interface at the backend (Figure 1). The chatbot employs the GPT-3.5 Instruct model with structured chat prompts based on an ideal script to guide patient-practitioner dialogue. A small illustrative section is shown in Figure 2. The use of generative AI allows the chatbot to understand variably expressed questions and respond as realistically as possible. Observation of the tutorial with the clinical educator acting as the patient found only two of 13 students actively participated by asking questions compared with 100% involvement with the chatbot. Students generally found the chatbot useful and perceived competence was improved following chatbot use (Figure 3). Most students also agreed that they participated more with the chatbot and that the chatbot would provide more opportunities for them to practice. Staff also recognized the versatility of the tool and AI's potential to generate more cases, although cautious of the current fallibility of generative accuracy. There is clear potential for educational chatbots in dental education which were well-received by staff and senior students. Future directions include development with GPT-4 and updated AI models, trialing with early-year dental students, and iterations of more cases. This study was supported by the FMH Media lab at The University of Sydney. Open access publishing facilitated by The University of Sydney, as part of the Wiley - The University of Sydney agreement via the Council of Australian University Librarians.","author":[{"family":"Or","given":"Aidan"},{"family":"Sukumar","given":"Smitha"},{"family":"Ritchie","given":"Helen"},{"family":"Sarrafpour","given":"Babak"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/jdd.13591","URL":"https://doi.org/10.1002/jdd.13591","source":"openalex"},{"id":"oa:W4401715493","type":"article-journal","title":"Exploring the role of artificial intelligence, large language models: Comparing patient‐focused information and clinical decision support capabilities to the gynecologic oncology guidelines","abstract":"Gynecologic cancer requires personalized care to improve outcomes. Large language models (LLMs) hold the potential to provide intelligent question-answering with reliable information about medical queries in clear and plain English, which can be understood by both healthcare providers and patients. We aimed to evaluate two freely available LLMs (ChatGPT and Google's Bard) in answering questions regarding the management of gynecologic cancer. The LLMs' performances were evaluated by developing a set questions that addressed common gynecologic oncologic findings from a patient's perspective and more complex questions to elicit recommendations from a clinician's perspective. Each question was presented to the LLM interface, and the responses generated by the artificial intelligence (AI) model were recorded. The responses were assessed based on the adherence to the National Comprehensive Cancer Network and European Society of Gynecological Oncology guidelines. This evaluation aimed to determine the accuracy and appropriateness of the information provided by LLMs. We showed that the models provided largely appropriate responses to questions regarding common cervical cancer screening tests and BRCA-related questions. Less useful answers were received to complex and controversial gynecologic oncology cases, as assessed by reviewing the common guidelines. ChatGPT and Bard lacked knowledge of regional guideline variations, However, it provided practical and multifaceted advice to patients and caregivers regarding the next steps of management and follow up. We conclude that LLMs may have a role as an adjunct informational tool to improve outcomes.","author":[{"family":"Reicher","given":"Lee"},{"family":"Lutsker","given":"Guy"},{"family":"Michaan","given":"Nadav"},{"family":"Grisaru","given":"Dan"},{"family":"Laskov","given":"Ido"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/ijgo.15869","URL":"https://doi.org/10.1002/ijgo.15869","source":"openalex"},{"id":"oa:W4400297840","type":"article-journal","title":"Application of Photoactive Compounds in Cancer Theranostics: Review on Recent Trends from Photoactive Chemistry to Artificial Intelligence","abstract":"According to the World Health Organization (WHO) and the International Agency for Research on Cancer (IARC), the number of cancer cases and deaths worldwide is predicted to nearly double by 2030, reaching 21.7 million cases and 13 million fatalities. The increase in cancer mortality is due to limitations in the diagnosis and treatment options that are currently available. The close relationship between diagnostics and medicine has made it possible for cancer patients to receive precise diagnoses and individualized care. This article discusses newly developed compounds with potential for photodynamic therapy and diagnostic applications, as well as those already in use. In addition, it discusses the use of artificial intelligence in the analysis of diagnostic images obtained using, among other things, theranostic agents.","author":[{"family":"Szymaszek","given":"Patryk"},{"family":"Tyszkaczochara","given":"Małgorzata"},{"family":"Ortyl","given":"Joanna"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/molecules29133164","URL":"https://doi.org/10.3390/molecules29133164","source":"openalex"},{"id":"oa:W4400123292","type":"article-journal","title":"Medical-informed machine learning: integrating prior knowledge into medical decision systems","abstract":"BACKGROUND: Clinical medicine offers a promising arena for applying Machine Learning (ML) models. However, despite numerous studies employing ML in medical data analysis, only a fraction have impacted clinical care. This article underscores the importance of utilising ML in medical data analysis, recognising that ML alone may not adequately capture the full complexity of clinical data, thereby advocating for the integration of medical domain knowledge in ML. METHODS: The study conducts a comprehensive review of prior efforts in integrating medical knowledge into ML and maps these integration strategies onto the phases of the ML pipeline, encompassing data pre-processing, feature engineering, model training, and output evaluation. The study further explores the significance and impact of such integration through a case study on diabetes prediction. Here, clinical knowledge, encompassing rules, causal networks, intervals, and formulas, is integrated at each stage of the ML pipeline, resulting in a spectrum of integrated models. RESULTS: The findings highlight the benefits of integration in terms of accuracy, interpretability, data efficiency, and adherence to clinical guidelines. In several cases, integrated models outperformed purely data-driven approaches, underscoring the potential for domain knowledge to enhance ML models through improved generalisation. In other cases, the integration was instrumental in enhancing model interpretability and ensuring conformity with established clinical guidelines. Notably, knowledge integration also proved effective in maintaining performance under limited data scenarios. CONCLUSIONS: By illustrating various integration strategies through a clinical case study, this work provides guidance to inspire and facilitate future integration efforts. Furthermore, the study identifies the need to refine domain knowledge representation and fine-tune its contribution to the ML model as the two main challenges to integration and aims to stimulate further research in this direction.","author":[{"family":"Sirocchi","given":"Christel"},{"family":"Bogliolo","given":"Alessandro"},{"family":"Montagna","given":"Sara"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s12911-024-02582-4","URL":"https://doi.org/10.1186/s12911-024-02582-4","source":"openalex"},{"id":"oa:W4404942414","type":"article-journal","title":"Artificial intelligence and digital tools for design and execution of cardiovascular clinical trials","abstract":"Recent advances have given rise to a spectrum of digital health technologies that have the potential to revolutionize the design and conduct of cardiovascular clinical trials. Advances in domain tasks such as automated diagnosis and classification, synthesis of high-volume data and latent data from adjacent modalities, patient discovery, telemedicine, remote monitoring, augmented reality, and in silico modelling have the potential to enhance the efficiency, accuracy, and cost-effectiveness of cardiovascular clinical trials. However, early experience with these tools has also exposed important issues, including regulatory barriers, clinical validation and acceptance, technological literacy, integration with care models, and health equity concerns. This narrative review summarizes the landscape of digital tools at each stage of clinical trial planning and execution and outlines roadblocks and opportunities for successful implementation of digital tools in cardiovascular clinical trials.","author":[{"family":"Hu","given":"Jiun‐ruey"},{"family":"Power","given":"John"},{"family":"Zannad","given":"Faı̈ez"},{"family":"Lam","given":"Carolyn"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/eurheartj/ehae794","URL":"https://doi.org/10.1093/eurheartj/ehae794","source":"openalex"},{"id":"oa:W4404291277","type":"article-journal","title":"Artificial Intelligence for Mechanical Ventilation: A Transformative Shift in Critical Care","abstract":"With the large volume of data coming from implemented technologies and monitoring systems, intensive care units (ICUs) represent a key area for artificial intelligence (AI) application. Despite the last decade has been marked by studies focused on the use of AI in medicine, its application in mechanical ventilation management is still limited. Optimizing mechanical ventilation is a complex and high-stake intervention, which requires a deep understanding of respiratory pathophysiology. Therefore, this complex task might be supported by AI and machine learning. Most of the studies already published involve the use of AI to predict outcomes for mechanically ventilated patients, including the need for intubation, the respiratory complications, and the weaning readiness and success. In conclusion, the application of AI for the management of mechanical ventilation is still at an early stage and requires a cautious and much less enthusiastic approach. Future research should be focused on AI progressive introduction in the everyday management of mechanically ventilated patients, with the aim to explore the great potentiality of this tool.","author":[{"family":"Misseri","given":"Giovanni"},{"family":"Piattoli","given":"Matteo"},{"family":"Cuttone","given":"Giuseppe"},{"family":"Gregoretti","given":"Cesare"},{"family":"Bignami","given":"Elena"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1177/29768675241298918","URL":"https://doi.org/10.1177/29768675241298918","source":"openalex"},{"id":"oa:W4404812876","type":"article-journal","title":"Challenges and Opportunities for Data Sharing Related to Artificial Intelligence Tools in Health Care in Low- and Middle-Income Countries: Systematic Review and Case Study From Thailand","abstract":"BACKGROUND: Health care systems in low- and middle-income countries (LMICs) can greatly benefit from artificial intelligence (AI) interventions in various use cases such as diagnostics, treatment, and public health monitoring but face significant challenges in sharing data for developing and deploying AI in health care. OBJECTIVE: This study aimed to identify barriers and enablers to data sharing for AI in health care in LMICs and to test the relevance of these in a local context. METHODS: First, we conducted a systematic literature search using PubMed, SCOPUS, Embase, Web of Science, and ACM using controlled vocabulary. Primary research studies, perspectives, policy landscape analyses, and commentaries performed in or involving an LMIC context were included. Studies that lacked a clear connection to health information exchange systems or were not reported in English were excluded from the review. Two reviewers independently screened titles and abstracts of the included articles and critically appraised each study. All identified barriers and enablers were classified according to 7 categories as per the predefined framework-technical, motivational, economic, political, legal and policy, ethical, social, organisational, and managerial. Second, we tested the local relevance of barriers and enablers in Thailand through stakeholder interviews with 15 academic experts, technology developers, regulators, policy makers, and health care providers. The interviewers took notes and analyzed data using framework analysis. Coding procedures were standardized to enhance the reliability of our approach. Coded data were reverified and themes were readjusted where necessary to avoid researcher bias. RESULTS: We identified 22 studies, the majority of which were conducted across Africa (n=12, 55%) and Asia (n=6, 27%). The most important data-sharing challenges were unreliable internet connectivity, lack of equipment, poor staff and management motivation, uneven resource distribution, and ethical concerns. Possible solutions included improving IT infrastructure, enhancing funding, introducing user-friendly software, and incentivizing health care organizations and personnel to share data for AI-related tools. In Thailand, inconsistent data systems, limited staff time, low health data literacy, complex and unclear policies, and cybersecurity issues were important data-sharing challenges. Key solutions included building a conducive digital ecosystem-having shared data input platforms for health facilities to ensure data uniformity and to develop easy-to-understand consent forms, having standardized guidelines for data sharing, and having compensation policies for data breach victims. CONCLUSIONS: Although AI in LMICs has the potential to overcome health inequalities, these countries face technical, political, legal, policy, and organizational barriers to sharing data, which impede effective AI development and deployment. When tested in a local context, most of these barriers were relevant. Although our findings might not be generalizable to other contexts, this study can be used by LMICs as a framework to identify barriers and strengths within their health care systems and devise localized solutions for enhanced data sharing. TRIAL REGISTRATION: PROSPERO CRD42022360644; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=360644.","author":[{"family":"Kaushik","given":"Aprajita"},{"family":"Barcellona","given":"Capucine"},{"family":"Mandyam","given":"Nikita"},{"family":"Tan","given":"Si"},{"family":"Tromp","given":"Jasper"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/58338","URL":"https://doi.org/10.2196/58338","source":"openalex"},{"id":"oa:W4404642446","type":"article-journal","title":"Unlocking artificial intelligence for strategic market development and business growth: innovations, opportunities, and future directions","abstract":"This study explores the role of Artificial Intelligence (AI) in strategic market development and business growth. It aims to examine how AI innovations create new opportunities, enhance operational efficiencies, and drive competitive advantages for businesses across various sectors. This study relies significantly on previously published literature and secondary data sourced from esteemed academic databases, including Web of Science, SCOPUS, Google Scholar, and Research gate. Initially, we collected a dataset of 300 papers spanning the period from January 2020 to 2024. After a rigorous screening process, we narrowed it down to 65 papers to derive the desired insights and produce robust results. AI is revolutionizing market strategies by enabling personalized marketing, optimizing supply chains, and improving decision-making through predictive analytics. Companies are leveraging AI for automating customer service, enhancing product development, and discovering new business models. The study also highlights AI’s potential in facilitating global market expansion and localized product offerings. The findings suggest that AI is central to enhancing business efficiency, innovation, and customer satisfaction. However, ethical concerns such as data privacy, algorithmic bias, and regulatory challenges remain significant barriers to AI adoption. Companies must balance innovation with responsible AI use to ensure long-term success and sustainability. Businesses adopting AI technologies can gain a competitive edge by improving operational efficiency and creating more tailored customer experiences. However, they must invest in AI transparency, employee reskilling, and regulatory compliance to mitigate risks associated with AI deployment. Further research is needed to explore the ethical and societal impacts of AI, including its role in job displacement and its potential for environmental sustainability. The integration of AI with emerging technologies like blockchain and IoT also offers exciting opportunities for future growth. Organizations should invest in AI solutions aligned with business objectives, ensuring ethical practices and continuous innovation to maintain a competitive edge in the evolving market landscape.","author":[{"family":"Hossain","given":"Fahim"},{"family":"Ahmed","given":"GR"},{"family":"Shuvo","given":"Shree"},{"family":"Kona","given":"Abeda"},{"family":"Raina","given":"Mostarifa"},{"family":"Shikder","given":"Fisan"}],"issued":{"date-parts":[[2024]]},"DOI":"10.55214/25768484.v8i6.3263","URL":"https://doi.org/10.55214/25768484.v8i6.3263","source":"openalex"},{"id":"oa:W4404402066","type":"article-journal","title":"Examining the Role of Large Language Models in Orthopedics: Systematic Review","abstract":"BACKGROUND: Large language models (LLMs) can understand natural language and generate corresponding text, images, and even videos based on prompts, which holds great potential in medical scenarios. Orthopedics is a significant branch of medicine, and orthopedic diseases contribute to a significant socioeconomic burden, which could be alleviated by the application of LLMs. Several pioneers in orthopedics have conducted research on LLMs across various subspecialties to explore their performance in addressing different issues. However, there are currently few reviews and summaries of these studies, and a systematic summary of existing research is absent. OBJECTIVE: The objective of this review was to comprehensively summarize research findings on the application of LLMs in the field of orthopedics and explore the potential opportunities and challenges. METHODS: PubMed, Embase, and Cochrane Library databases were searched from January 1, 2014, to February 22, 2024, with the language limited to English. The terms, which included variants of \"large language model,\" \"generative artificial intelligence,\" \"ChatGPT,\" and \"orthopaedics,\" were divided into 2 categories: large language model and orthopedics. After completing the search, the study selection process was conducted according to the inclusion and exclusion criteria. The quality of the included studies was assessed using the revised Cochrane risk-of-bias tool for randomized trials and CONSORT-AI (Consolidated Standards of Reporting Trials-Artificial Intelligence) guidance. Data extraction and synthesis were conducted after the quality assessment. RESULTS: A total of 68 studies were selected. The application of LLMs in orthopedics involved the fields of clinical practice, education, research, and management. Of these 68 studies, 47 (69%) focused on clinical practice, 12 (18%) addressed orthopedic education, 8 (12%) were related to scientific research, and 1 (1%) pertained to the field of management. Of the 68 studies, only 8 (12%) recruited patients, and only 1 (1%) was a high-quality randomized controlled trial. ChatGPT was the most commonly mentioned LLM tool. There was considerable heterogeneity in the definition, measurement, and evaluation of the LLMs' performance across the different studies. For diagnostic tasks alone, the accuracy ranged from 55% to 93%. When performing disease classification tasks, ChatGPT with GPT-4's accuracy ranged from 2% to 100%. With regard to answering questions in orthopedic examinations, the scores ranged from 45% to 73.6% due to differences in models and test selections. CONCLUSIONS: LLMs cannot replace orthopedic professionals in the short term. However, using LLMs as copilots could be a potential approach to effectively enhance work efficiency at present. More high-quality clinical trials are needed in the future, aiming to identify optimal applications of LLMs and advance orthopedics toward higher efficiency and precision.","author":[{"family":"Zhang","given":"Cheng"},{"family":"Liu","given":"Shanshan"},{"family":"Zhou","given":"Xingyu"},{"family":"Zhou","given":"Siyu"},{"family":"Tian","given":"Yinglun"},{"family":"Wang","given":"Shenglin"},{"family":"Xu","given":"Nanfang"},{"family":"Li","given":"Weishi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/59607","URL":"https://doi.org/10.2196/59607","source":"openalex"},{"id":"oa:W4399187438","type":"article-journal","title":"Cloud-magnetic resonance imaging system: In the era of 6G and artificial intelligence","abstract":"Magnetic resonance imaging (MRI) plays an important role in medical diagnosis, generating petabytes of image data annually in large hospitals. This voluminous data stream requires a significant amount of network bandwidth and extensive storage infrastructure. Additionally, local data processing demands substantial manpower and hardware investments. Data isolation across different healthcare institutions hinders cross-institutional collaboration in clinics and research. In this work, we anticipate an innovative MRI system and its four generations that integrate emerging distributed cloud computing, 6G bandwidth, edge computing, federated learning, and blockchain technology. This system is called Cloud-MRI, aiming at solving the problems of MRI data storage security, transmission speed, artificial intelligence (AI) algorithm maintenance, hardware upgrading, and collaborative work. The workflow commences with the transformation of k-space raw data into the standardized Imaging Society for Magnetic Resonance in Medicine Raw Data (ISMRMRD) format. Then, the data are uploaded to the cloud or edge nodes for fast image reconstruction, neural network training, and automatic analysis. Then, the outcomes are seamlessly transmitted to clinics or research institutes for diagnosis and other services. The Cloud-MRI system will save the raw imaging data, reduce the risk of data loss, facilitate inter-institutional medical collaboration, and finally improve diagnostic accuracy and work efficiency.","author":[{"family":"Zhou","given":"Yirong"},{"family":"Wu","given":"Yanhuang"},{"family":"Su","given":"Yuhan"},{"family":"Li","given":"Jing"},{"family":"Cai","given":"Jianyu"},{"family":"You","given":"Yongfu"},{"family":"Zhou","given":"Jianjun"},{"family":"Guo","given":"Di"},{"family":"Qu","given":"Xiaobo"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.mrl.2024.200138","URL":"https://doi.org/10.1016/j.mrl.2024.200138","source":"openalex"},{"id":"oa:W4404202452","type":"article-journal","title":"Testing process for artificial intelligence applications in radiology practice","abstract":"Artificial intelligence (AI) applications are becoming increasingly common in radiology. However, ensuring reliable operation and expected clinical benefits remains a challenge. A systematic testing process aims to facilitate clinical deployment by confirming software applicability to local patient populations, practises, adherence to regulatory and safety requirements, and compatibility with existing systems. In this work, we present our testing process developed based on practical experience. First, a survey and pre-evaluation is conducted, where information requests are sent for potential products, and the specifications are evaluated against predetermined requirements. In the second phase, data collection, testing, and analysis are conducted. In the retrospective stage, the application undergoes testing with a pre selected dataset and is evaluated against specified key performance indicators (KPIs). In the prospective stage, the application is integrated into the clinical workflow and evaluated with additional process-specific KPIs. In the final phase, the results are evaluated in terms of safety, effectiveness, productivity, and integration. The final report summarises the results and includes a procurement/deployment or rejection recommendation. The process allows termination at any phase if the application fails to meet essential criteria. In addition, we present practical remarks from our experiences in AI testing and provide forms to guide and document the testing process. The established AI testing process facilitates a systematic evaluation and documentation of new technologies ensuring that each application undergoes equal and sufficient validation. Testing with local data is crucial for identifying biases and pitfalls of AI algorithms to improve the quality and safety, ultimately benefiting patient care.","author":[{"family":"Ketola","given":"Juuso"},{"family":"Inkinen","given":"Satu"},{"family":"Mäkelä","given":"Teemu"},{"family":"Syväranta","given":"Suvi"},{"family":"Peltonen","given":"Juha"},{"family":"Kaasalainen","given":"Touko"},{"family":"Kortesniemi","given":"Mika"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.ejmp.2024.104842","URL":"https://doi.org/10.1016/j.ejmp.2024.104842","source":"openalex"},{"id":"oa:W4395000105","type":"article-journal","title":"Exploring Embodied Intelligence in Soft Robotics: A Review","abstract":"Soft robotics is closely related to embodied intelligence in the joint exploration of the means to achieve more natural and effective robotic behaviors via physical forms and intelligent interactions. Embodied intelligence emphasizes that intelligence is affected by the synergy of the brain, body, and environment, focusing on the interaction between agents and the environment. Under this framework, the design and control strategies of soft robotics depend on their physical forms and material properties, as well as algorithms and data processing, which enable them to interact with the environment in a natural and adaptable manner. At present, embodied intelligence has comprehensively integrated related research results on the evolution, learning, perception, decision making in the field of intelligent algorithms, as well as on the behaviors and controls in the field of robotics. From this perspective, the relevant branches of the embodied intelligence in the context of soft robotics were studied, covering the computation of embodied morphology; the evolution of embodied AI; and the perception, control, and decision making of soft robotics. Moreover, on this basis, important research progress was summarized, and related scientific problems were discussed. This study can provide a reference for the research of embodied intelligence in the context of soft robotics.","author":[{"family":"Zhao","given":"Zikai"},{"family":"Wu","given":"Qiuxuan"},{"family":"Wang","given":"Jian"},{"family":"Zhang","given":"Botao"},{"family":"Zhong","given":"Chaoliang"},{"family":"Zhilenkov","given":"Anton"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/biomimetics9040248","URL":"https://doi.org/10.3390/biomimetics9040248","source":"openalex"},{"id":"oa:W4401844219","type":"article-journal","title":"Super AI, Generative AI, Narrow AI and Chatbots: An Assessment of Artificial Intelligence Technologies for The Public Sector and Public Administration","abstract":"Artificial intelligence encompasses a wide range of approaches, methodologies, and techniques aimed at mimicking human intelligence in machines. In recent times, the concepts of Generative Artificial Intelligence (AI), Super AI, and Narrow AI have attracted considerable attention. Undoubtedly, the success of ChatGPT in capturing all attention has played a significant role in this. Artificial intelligence technology has a profound impact on all sectors, and sector representatives are striving to adapt to this technology more quickly. It is projected that artificial intelligence could generate an economic size of 13 trillion American dollars by 2030. Developments in artificial intelligence technologies undoubtedly lead to significant improvements in the functioning of public institutions and access for citizens. Artificial intelligence has the potential to be used in many public services, including security and defense, healthcare services, education, transportation and infrastructure, environmental and natural resource management, law and justice systems, among others. Therefore, evaluating the types of artificial intelligence, Narrow AI applications, and chatbots for public use is seen as highly beneficial from the perspective of public administration and the public sector. In our study, the topics of super artificial intelligence, generative artificial intelligence, narrow artificial intelligence, and chatbots have been extensively evaluated within the context of the public sector and public administration. Utilizing findings from both Turkish and English literature reviews, the importance and potential impacts of artificial intelligence within the public sector, along with current trends, have been comprehensively assessed. This research delves into the concepts of artificial intelligence and its subsets—super AI, generative AI, narrow AI, and chatbots—within the general framework of the public sector. China and the United States are pioneering and leading countries in terms of investment. Although the U.S. stands out in many areas regarding investment, China's integration of artificial intelligence with national strategies and its policies indicate that it may play a more dominant role in the future. There are four main implementation areas of artificial intelligence in the public sector: efficiency and automation, service delivery, data-driven governance, and ethical and regulatory challenges. A review of the literature reveals that the ethical, legal, and social implications of implementing artificial intelligence in the public sector require more careful consideration. The study makes a significant contribution to the field of artificial intelligence discussions in public administration and the public sector, providing a comprehensive assessment of current discussions on artificial intelligence in the literature.","author":[{"family":"Damar","given":"Muhammet"},{"family":"Özen","given":"Ahmet"},{"family":"Çakmak","given":"Ülkü"},{"family":"Özoğuz","given":"Eren"},{"family":"Erenay","given":"Fatih"}],"issued":{"date-parts":[[2024]]},"DOI":"10.61969/jai.1512906","URL":"https://doi.org/10.61969/jai.1512906","source":"openalex"},{"id":"oa:W4401851847","type":"article-journal","title":"Trusted artificial intelligence for environmental assessments: An explainable high-precision model with multi-source big data","abstract":"Environmental assessments are critical for ensuring the sustainable development of human civilization. The integration of artificial intelligence (AI) in these assessments has shown great promise, yet the \"black box\" nature of AI models often undermines trust due to the lack of transparency in their decision-making processes, even when these models demonstrate high accuracy. To address this challenge, we evaluated the performance of a transformer model against other AI approaches, utilizing extensive multivariate and spatiotemporal environmental datasets encompassing both natural and anthropogenic indicators. We further explored the application of saliency maps as a novel explainability tool in multi-source AI-driven environmental assessments, enabling the identification of individual indicators' contributions to the model's predictions. We find that the transformer model outperforms others, achieving an accuracy of about 98% and an area under the receiver operating characteristic curve (AUC) of 0.891. Regionally, the environmental assessment values are predominantly classified as level II or III in the central and southwestern study areas, level IV in the northern region, and level V in the western region. Through explainability analysis, we identify that water hardness, total dissolved solids, and arsenic concentrations are the most influential indicators in the model. Our AI-driven environmental assessment model is accurate and explainable, offering actionable insights for targeted environmental management. Furthermore, this study advances the application of AI in environmental science by presenting a robust, explainable model that bridges the gap between machine learning and environmental governance, enhancing both understanding and trust in AI-assisted environmental assessments.","author":[{"family":"Xu","given":"Haoli"},{"family":"Yang","given":"Xing"},{"family":"Hu","given":"Yihua"},{"family":"Wang","given":"Daqing"},{"family":"Liang","given":"Zhenyu"},{"family":"Mu","given":"Hua"},{"family":"Wang","given":"Yangyang"},{"family":"Shi","given":"Liang"},{"family":"Gao","given":"Haoqi"},{"family":"Song","given":"Daoqing"},{"family":"Cheng","given":"Zijian"},{"family":"Lu","given":"Zhao"},{"family":"Zhao","given":"Xiaoning"},{"family":"Lu","given":"Jun"},{"family":"Wang","given":"Bingwen"},{"family":"Hu","given":"Zhiyang"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.ese.2024.100479","URL":"https://doi.org/10.1016/j.ese.2024.100479","source":"openalex"},{"id":"oa:W4403987666","type":"article-journal","title":"Artificial Intelligence for English Language Learning and Teaching: Advancing Sustainable Development Goals","abstract":"This study explores the affordance of Artificial Intelligence (AI) to English language learning and teaching, focusing on its alignment with the United Nations' Sustainable Development Goals (SDGs). It aims to investigate the role of AI in enhancing language education and fostering student-centered learning. Data for this study were collected through semi-structured interviews with 18 English teachers to gather qualitative insights into their experiences with AI-powered language learning tools. The findings reveal that the teachers have positive appraisals of AI that its use has six major impacts: i) enhancing the personalization of learning; ii) contributing to improved learning outcomes by advancing students' speaking, listening, reading, and writing skills; iii) playing a fundamental role in bridging educational gaps; iv) enhancing students’ engagement and motivation; v) empowering educators with professional development opportunities; vi) and encouraging self-directed learning. This study argues that, if implemented thoughtfully, AI can enhance language learning outcomes and create an environment conducive to student engagement and success.","author":[{"family":"Alsmadi","given":"Omar"},{"family":"Rashid","given":"Radzuwan"},{"family":"Saad","given":"Hadeel"},{"family":"Zrekat","given":"Yousef"},{"family":"Kamal","given":"Siti"},{"family":"Uktamovich","given":"Gaforov"}],"issued":{"date-parts":[[2024]]},"DOI":"10.17507/jltr.1506.09","URL":"https://doi.org/10.17507/jltr.1506.09","source":"openalex"},{"id":"oa:W4392415015","type":"article-journal","title":"Evaluating Artificial Intelligence in Clinical Settings—Let Us Not Reinvent the Wheel","abstract":"Given the requirement to minimize the risks and maximize the benefits of technology applications in health care provision, there is an urgent need to incorporate theory-informed health IT (HIT) evaluation frameworks into existing and emerging guidelines for the evaluation of artificial intelligence (AI). Such frameworks can help developers, implementers, and strategic decision makers to build on experience and the existing empirical evidence base. We provide a pragmatic conceptual overview of selected concrete examples of how existing theory-informed HIT evaluation frameworks may be used to inform the safe development and implementation of AI in health care settings. The list is not exhaustive and is intended to illustrate applications in line with various stakeholder requirements. Existing HIT evaluation frameworks can help to inform AI-based development and implementation by supporting developers and strategic decision makers in considering relevant technology, user, and organizational dimensions. This can facilitate the design of technologies, their implementation in user and organizational settings, and the sustainability and scalability of technologies.","author":[{"family":"Cresswell","given":"Kathrin"},{"family":"Keizer","given":"Nicolette"},{"family":"Magrabi","given":"Farah"},{"family":"Williams","given":"Robin"},{"family":"Rigby","given":"Michael"},{"family":"Prgomet","given":"Mirela"},{"family":"Kukhareva","given":"Polina"},{"family":"Wong","given":"Zoie"},{"family":"Scott","given":"Philip"},{"family":"Craven","given":"Catherine"},{"family":"Georgiou","given":"Andrew"},{"family":"Medlock","given":"Stephanie"},{"family":"Mcnair","given":"Jytte"},{"family":"Ammenwerth","given":"Elske"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/46407","URL":"https://doi.org/10.2196/46407","source":"openalex"},{"id":"oa:W4403463361","type":"article-journal","title":"Smart Vision Transparency: Efficient Ocular Disease Prediction Model Using Explainable Artificial Intelligence","abstract":"The early prediction of ocular disease is certainly an obligatory concern in the domain of ophthalmic medicine. Although modern scientific discoveries have shown the potential to treat eye diseases by using artificial intelligence (AI) and machine learning, explainable AI remains a crucial challenge confronting this area of research. Although some traditional methods put in significant effort, they cannot accurately predict the proper ocular diseases. However, incorporating AI into diagnosing eye diseases in healthcare complicates the situation as the decision-making process of AI demonstrates complexity, which is a significant concern, especially in major sectors like ocular disease prediction. The lack of transparency in the AI models may hinder the confidence and trust of the doctors and the patients, as well as their perception of the AI and its abilities. Accordingly, explainable AI is significant in ensuring trust in the technology, enhancing clinical decision-making ability, and deploying ocular disease detection. This research proposed an efficient transfer learning model for eye disease prediction to transform smart vision potential in the healthcare sector and meet conventional approaches' challenges while integrating explainable artificial intelligence (XAI). The integration of XAI in the proposed model ensures the transparency of the decision-making process through the comprehensive provision of rationale. This proposed model provides promising results with 95.74% accuracy and explains the transformative potential of XAI in advancing ocular healthcare. This significant milestone underscores the effectiveness of the proposed model in accurately determining various types of ocular disease. It is clearly shown that the proposed model is performing better than the previously published methods.","author":[{"family":"Abbas","given":"Sagheer"},{"family":"Qaisar","given":"Adnan"},{"family":"Farooq","given":"Muhammad"},{"family":"Saleem","given":"Muhammad"},{"family":"Ahmad","given":"Munir"},{"family":"Khan","given":"Muhammad"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/s24206618","URL":"https://doi.org/10.3390/s24206618","source":"openalex"},{"id":"oa:W4404933079","type":"article-journal","title":"Artificial intelligence for collective intelligence: a national-scale research strategy","abstract":"Abstract Advances in artificial intelligence (AI) have great potential to help address societal challenges that are both collective in nature and present at national or transnational scale. Pressing challenges in healthcare, finance, infrastructure and sustainability, for instance, might all be productively addressed by leveraging and amplifying AI for national-scale collective intelligence . The development and deployment of this kind of AI faces distinctive challenges, both technical and socio-technical. Here, a research strategy for mobilising inter-disciplinary research to address these challenges is detailed and some of the key issues that must be faced are outlined.","author":[{"family":"Bullock","given":"Seth"},{"family":"Ajmeri","given":"Nirav"},{"family":"Batty","given":"Mike"},{"family":"Black","given":"Michaela"},{"family":"Cartlidge","given":"John"},{"family":"Challen","given":"Robert"},{"family":"Chen","given":"Cangxiong"},{"family":"Jing","given":"Chen"},{"family":"Condell","given":"Joan"},{"family":"Danon","given":"León"},{"family":"Dennett","given":"Adam"},{"family":"Heppenstall","given":"Alison"},{"family":"Marshall","given":"Paul"},{"family":"Morgan","given":"Phil"},{"family":"Okane","given":"Aisling"},{"family":"Smith","given":"Laura"},{"family":"Smith","given":"Theresa"},{"family":"Williams","given":"Hywel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1017/s0269888924000110","URL":"https://doi.org/10.1017/s0269888924000110","source":"openalex"},{"id":"oa:W4400695423","type":"article-journal","title":"A systematic review and research recommendations on artificial intelligence for automated cervical cancer detection","abstract":"Abstract Early diagnosis of abnormal cervical cells enhances the chance of prompt treatment for cervical cancer (CrC). Artificial intelligence (AI)‐assisted decision support systems for detecting abnormal cervical cells are developed because manual identification needs trained healthcare professionals, and can be difficult, time‐consuming, and error‐prone. The purpose of this study is to present a comprehensive review of AI technologies used for detecting cervical pre‐cancerous lesions and cancer. The review study includes studies where AI was applied to Pap Smear test (cytological test), colposcopy, sociodemographic data and other risk factors, histopathological analyses, magnetic resonance imaging‐, computed tomography‐, and positron emission tomography‐scan‐based imaging modalities. We performed searches on Web of Science, Medline, Scopus, and Inspec. The preferred reporting items for systematic reviews and meta‐analysis guidelines were used to search, screen, and analyze the articles. The primary search resulted in identifying 9745 articles. We followed strict inclusion and exclusion criteria, which include search windows of the last decade, journal articles, and machine/deep learning‐based methods. A total of 58 studies have been included in the review for further analysis after identification, screening, and eligibility evaluation. Our review analysis shows that deep learning models are preferred for imaging techniques, whereas machine learning‐based models are preferred for sociodemographic data. The analysis shows that convolutional neural network‐based features yielded representative characteristics for detecting pre‐cancerous lesions and CrC. The review analysis also highlights the need for generating new and easily accessible diverse datasets to develop versatile models for CrC detection. Our review study shows the need for model explainability and uncertainty quantification to increase the trust of clinicians and stakeholders in the decision‐making of automated CrC detection models. Our review suggests that data privacy concerns and adaptability are crucial for deployment hence, federated learning and meta‐learning should also be explored. This article is categorized under: Fundamental Concepts of Data and Knowledge > Explainable AI Technologies > Machine Learning Technologies > Classification","author":[{"family":"Khare","given":"Smith"},{"family":"Blanesvidal","given":"Victoria"},{"family":"Booth","given":"Berit"},{"family":"Petersen","given":"Lone"},{"family":"Nadimi","given":"Esmaeil"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/widm.1550","URL":"https://doi.org/10.1002/widm.1550","source":"openalex"},{"id":"oa:W4405225674","type":"article-journal","title":"Ambient artificial intelligence scribes: physician burnout and perspectives on usability and documentation burden","abstract":"OBJECTIVE: This study evaluates the pilot implementation of ambient AI scribe technology to assess physician perspectives on usability and the impact on physician burden and burnout. MATERIALS AND METHODS: This prospective quality improvement study was conducted at Stanford Health Care with 48 physicians over a 3-month period. Outcome measures included burden, burnout, usability, and perceived time savings. RESULTS: Paired survey analysis (n = 38) revealed large statistically significant reductions in task load (-24.42, p <.001) and burnout (-1.94, p <.001), and moderate statistically significant improvements in usability scores (+10.9, p <.001). Post-survey responses (n = 46) indicated favorable utility with improved perceptions of efficiency, documentation quality, and ease of use. DISCUSSION: In one of the first pilot implementations of ambient AI scribe technology, improvements in physician task load, burnout, and usability were demonstrated. CONCLUSION: Ambient AI scribes like DAX Copilot may enhance clinical workflows. Further research is needed to optimize widespread implementation and evaluate long-term impacts.","author":[{"family":"Shah","given":"Shreya"},{"family":"Devon-Sand","given":"Anna"},{"family":"Stephen","given":"P"},{"family":"Jeong","given":"Yejin"},{"family":"Crowell","given":"Trevor"},{"family":"Smith","given":"Margaret"},{"family":"Liang","given":"April"},{"family":"Delahaie","given":"Clarissa"},{"family":"Hsia","given":"Caroline"},{"family":"Shanafelt","given":"Tait"},{"family":"Pfeffer","given":"Michael"},{"family":"Sharp","given":"Christopher"},{"family":"Lin","given":"Steven"},{"family":"García","given":"Patricia"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/jamia/ocae295","URL":"https://doi.org/10.1093/jamia/ocae295","source":"openalex"},{"id":"oa:W4391362070","type":"article-journal","title":"The dark side of artificial intelligence in services","abstract":"Artificial intelligence (AI) initiatives, including Generative AI, are being increasingly implemented in service industries, and are having a great impact on service operations and on customers’ reactions and behaviors. Previous literature is overoptimistic about AI implementation, and there is still a need to explore the dark side of this technology; that is, its potential negative impacts on consumers, businesses, and society, as well as the moral concerns associated with AI use in services. To establish some fundamental insights related to this research domain, this paper contributes to previous AI based-services literature by proposing a three-part conceptual model inspired by Belanche et al. (2020a), comprised of AI design, customers, and the service encounter. Specifically, we identify key factors and research gaps within each category that need to be addressed. The final research questions provide a research agenda to guide scholars and help practitioners implement AI-based services while avoiding their potential negative outcomes.","author":[{"family":"Belanche","given":"Daniel"},{"family":"Belk","given":"Russell"},{"family":"Casaló","given":"Luis"},{"family":"Flavián","given":"Carlos"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1080/02642069.2024.2305451","URL":"https://doi.org/10.1080/02642069.2024.2305451","source":"openalex"},{"id":"oa:W4378746207","type":"article-journal","title":"Explainable artificial intelligence (XAI) in radiology and nuclear medicine: a literature review","abstract":"Rational Deep learning (DL) has demonstrated a remarkable performance in diagnostic imaging for various diseases and modalities and therefore has a high potential to be used as a clinical tool. However, current practice shows low deployment of these algorithms in clinical practice, because DL algorithms lack transparency and trust due to their underlying black-box mechanism. For successful employment, explainable artificial intelligence (XAI) could be introduced to close the gap between the medical professionals and the DL algorithms. In this literature review, XAI methods available for magnetic resonance (MR), computed tomography (CT), and positron emission tomography (PET) imaging are discussed and future suggestions are made. Methods PubMed, Embase.com and Clarivate Analytics/Web of Science Core Collection were screened. Articles were considered eligible for inclusion if XAI was used (and well described) to describe the behavior of a DL model used in MR, CT and PET imaging. Results A total of 75 articles were included of which 54 and 17 articles described post and ad hoc XAI methods, respectively, and 4 articles described both XAI methods. Major variations in performance is seen between the methods. Overall, post hoc XAI lacks the ability to provide class-discriminative and target-specific explanation. Ad hoc XAI seems to tackle this because of its intrinsic ability to explain. However, quality control of the XAI methods is rarely applied and therefore systematic comparison between the methods is difficult. Conclusion There is currently no clear consensus on how XAI should be deployed in order to close the gap between medical professionals and DL algorithms for clinical implementation. We advocate for systematic technical and clinical quality assessment of XAI methods. Also, to ensure end-to-end unbiased and safe integration of XAI in clinical workflow, (anatomical) data minimization and quality control methods should be included.","author":[{"family":"Vries","given":"Bart"},{"family":"Zwezerijnen","given":"Gerben"},{"family":"Burchell","given":"George"},{"family":"Velden","given":"Floris"},{"family":"Oordt","given":"CWMD"},{"family":"Boellaard","given":"Ronald"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3389/fmed.2023.1180773","URL":"https://doi.org/10.3389/fmed.2023.1180773","source":"openalex"},{"id":"oa:W4402400123","type":"article-journal","title":"Revolutionizing Health Care: The Transformative Impact of Large Language Models in Medicine","abstract":"Large language models (LLMs) are rapidly advancing medical artificial intelligence, offering revolutionary changes in health care. These models excel in natural language processing (NLP), enhancing clinical support, diagnosis, treatment, and medical research. Breakthroughs, like GPT-4 and BERT (Bidirectional Encoder Representations from Transformer), demonstrate LLMs' evolution through improved computing power and data. However, their high hardware requirements are being addressed through technological advancements. LLMs are unique in processing multimodal data, thereby improving emergency, elder care, and digital medical procedures. Challenges include ensuring their empirical reliability, addressing ethical and societal implications, especially data privacy, and mitigating biases while maintaining privacy and accountability. The paper emphasizes the need for human-centric, bias-free LLMs for personalized medicine and advocates for equitable development and access. LLMs hold promise for transformative impacts in health care.","author":[{"family":"Zhang","given":"Kuo"},{"family":"Meng","given":"Xiangbin"},{"family":"Yan","given":"Xiangyu"},{"family":"Ji","given":"Jiaming"},{"family":"Liu","given":"Jingqian"},{"family":"Xu","given":"Hua"},{"family":"Zhang","given":"Heng"},{"family":"Liu","given":"Da"},{"family":"Wang","given":"Jingjia"},{"family":"Wang","given":"Xuliang"},{"family":"Gao","given":"Jun"},{"family":"Wang","given":"Yuan"},{"family":"Shao","given":"Chunli"},{"family":"Wang","given":"Wenyao"},{"family":"Li","given":"Jiarong"},{"family":"Zheng","given":"Ming"},{"family":"Yang","given":"Yaodong"},{"family":"Tang","given":"Yi‐da"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/59069","URL":"https://doi.org/10.2196/59069","source":"openalex"},{"id":"oa:W4377097363","type":"article-journal","title":"The Future of Artificial Intelligence in Special Education Technology","abstract":"This manuscript presents a preliminary discussion of Artificial Intelligence (AI) as a disruptive technology with the potential to significantly change special education practices. The article begins with a brief description of the development of AI. The authors recognize our assertions are subjective and require further research. Several references are not peer-reviewed because educational research takes years to conduct, analyze, and disseminate outcomes. In this manuscript, we discuss current software used for writing with students in special education and discuss similarities and differences with AI software. This discussion is followed by questions and examples related to implementation, ethical and policy considerations, and preservice special education teacher preparation. The article concludes with future considerations for how AI will impact the special education technology field.","author":[{"family":"Marino","given":"Matthew"},{"family":"Vasquez","given":"Eleazar"},{"family":"Dieker","given":"Lisa"},{"family":"Basham","given":"James"},{"family":"Blackorby","given":"José"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1177/01626434231165977","URL":"https://doi.org/10.1177/01626434231165977","source":"openalex"},{"id":"oa:W4387710210","type":"article-journal","title":"Redefining biomaterial biocompatibility: challenges for artificial intelligence and text mining","abstract":"The surge in 'Big data' has significantly influenced biomaterials research and development, with vast data volumes emerging from clinical trials, scientific literature, electronic health records, and other sources. Biocompatibility is essential in developing safe medical devices and biomaterials to perform as intended without provoking adverse reactions. Therefore, establishing an artificial intelligence (AI)-driven biocompatibility definition has become decisive for automating data extraction and profiling safety effectiveness. This definition should both reflect the attributes related to biocompatibility and be compatible with computational data-mining methods. Here, we discuss the need for a comprehensive and contemporary definition of biocompatibility and the challenges in developing one. We also identify the key elements that comprise biocompatibility, and propose an integrated biocompatibility definition that enables data-mining approaches.","author":[{"family":"Mateusanz","given":"Miguel"},{"family":"Fuenteslópez","given":"Carla"},{"family":"Uribe","given":"Juan"},{"family":"Haugen","given":"Håvard"},{"family":"Pandit","given":"Abhay"},{"family":"Ginebra","given":"Maria‐pau"},{"family":"Hakimi","given":"Osnat"},{"family":"Krallinger","given":"Martin"},{"family":"Samara","given":"Athina"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.tibtech.2023.09.015","URL":"https://doi.org/10.1016/j.tibtech.2023.09.015","source":"openalex"},{"id":"oa:W4386390931","type":"article-journal","title":"Future of education in the era of generative artificial intelligence: Consensus among Chinese scholars on applications of ChatGPT in schools","abstract":"Abstract ChatGPT is an artificial intelligence chatbot that utilizes advanced natural language processing technologies, including large language models, to produce human‐like responses to user queries spanning a wide range of topics from programming to mathematics. As an emerging generative artificial intelligence (GAI) tool, it presents novel opportunities and challenges to the ongoing digital transformation of education. This article employs a systematic review approach to summarize the viewpoints of Chinese scholars and experts regarding the implementation of GAI in education. The research findings indicate that a majority of Chinese scholars support the cautious integration of GAI into education as it serves as a learning tool that offers personalized educational experiences for students. However, it also raises concerns related to academic integrity and the potential hindrance to students' critical thinking skills. Consequently, a framework called DATS, which outlines an optimization path for future GAI applications in schools, is proposed. The framework takes into account the perspectives of four key stakeholders: developers, administrators, teachers, and students.","author":[{"family":"Liu","given":"Ming"},{"family":"Ren","given":"Yiling"},{"family":"Nyagoga","given":"Lucy"},{"family":"Stonier","given":"Francis"},{"family":"Wu","given":"Zhongming"},{"family":"Yu","given":"Liang"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/fer3.10","URL":"https://doi.org/10.1002/fer3.10","source":"openalex"},{"id":"oa:W4394859173","type":"article-journal","title":"Knowledge and Perception of Artificial Intelligence among Faculty Members and Students at Batterjee Medical College","abstract":"Background: Mounting research suggests that artificial intelligence (AI) is one of the innovations that aid in the patient's diagnosis and treatment, but unfortunately limited research has been conducted in this regard in the Kingdom of Saudi Arabia (KSA). Hence, this study aimed to assess the level of knowledge and awareness of AI among faculty members and medicine students in one of the premier medical colleges in KSA. Methods: A cross-sectional descriptive study was conducted at Batterjee Medical College (BMC), Jeddah (KSA), from November 2022 to April 2023. Result: A total of 131 participants contributed to our study, of which three were excluded due to incomplete responses, thereby giving a response rate of 98%. Conclusion: 85.4% of the respondents believe that AI has a positive impact on the healthcare system and physicians in general. Hence, there should be a mandatory course in medical schools that can prepare future doctors to diagnose patients more accurately, make predictions about patients' future health, and recommend better treatments.","author":[{"family":"Alshanberi","given":"Asim"},{"family":"Mousa","given":"Ahmed"},{"family":"Hashim","given":"Sama"},{"family":"Almutairi","given":"Reem"},{"family":"Alrehali","given":"Sara"},{"family":"Hamisu","given":"Aisha"},{"family":"Shaikhomer","given":"Mohammed"},{"family":"Ansari","given":"Shakeel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.4103/jpbs.jpbs_1162_23","URL":"https://doi.org/10.4103/jpbs.jpbs_1162_23","source":"openalex"},{"id":"oa:W4380995257","type":"article-journal","title":"Utility of ChatGPT in Clinical Practice","abstract":"ChatGPT is receiving increasing attention and has a variety of application scenarios in clinical practice. In clinical decision support, ChatGPT has been used to generate accurate differential diagnosis lists, support clinical decision-making, optimize clinical decision support, and provide insights for cancer screening decisions. In addition, ChatGPT has been used for intelligent question-answering to provide reliable information about diseases and medical queries. In terms of medical documentation, ChatGPT has proven effective in generating patient clinical letters, radiology reports, medical notes, and discharge summaries, improving efficiency and accuracy for health care providers. Future research directions include real-time monitoring and predictive analytics, precision medicine and personalized treatment, the role of ChatGPT in telemedicine and remote health care, and integration with existing health care systems. Overall, ChatGPT is a valuable tool that complements the expertise of health care providers and improves clinical decision-making and patient care. However, ChatGPT is a double-edged sword. We need to carefully consider and study the benefits and potential dangers of ChatGPT. In this viewpoint, we discuss recent advances in ChatGPT research in clinical practice and suggest possible risks and challenges of using ChatGPT in clinical practice. It will help guide and support future artificial intelligence research similar to ChatGPT in health.","author":[{"family":"Liu","given":"Jialin"},{"family":"Wang","given":"Changyu"},{"family":"Liu","given":"Siru"}],"issued":{"date-parts":[[2023]]},"DOI":"10.2196/48568","URL":"https://doi.org/10.2196/48568","source":"openalex"},{"id":"oa:W4366753220","type":"article-journal","title":"Artificial Intelligence Applications in Hepatology","abstract":"Over the past 2 decades, the field of hepatology has witnessed major developments in diagnostic tools, prognostic models, and treatment options making it one of the most complex medical subspecialties. Through artificial intelligence (AI) and machine learning, computers are now able to learn from complex and diverse clinical datasets to solve real-world medical problems with performance that surpasses that of physicians in certain areas. AI algorithms are currently being implemented in liver imaging, interpretation of liver histopathology, noninvasive tests, prediction models, and more. In this review, we provide a summary of the state of AI in hepatology and discuss current challenges for large-scale implementation including some ethical aspects. We emphasize to the readers that most AI-based algorithms that are discussed in this review are still considered in early development and their utility and impact on patient outcomes still need to be assessed in future large-scale and inclusive studies. Our vision is that the use of AI in hepatology will enhance physician performance, decrease the burden and time spent on documentation, and reestablish the personalized patient-physician relationship that is of utmost importance for obtaining good outcomes.","author":[{"family":"Schattenberg","given":"Jörn"},{"family":"Chalasani","given":"Naga"},{"family":"Alkhouri","given":"Naim"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.cgh.2023.04.007","URL":"https://doi.org/10.1016/j.cgh.2023.04.007","source":"openalex"},{"id":"oa:W4405335081","type":"article-journal","title":"Advancement of post-market surveillance of medical devices leveraging artificial intelligence: Patient monitors case study","abstract":"BackgroundHealthcare institutions throughout the world rely on medical devices to provide their services reliably and effectively. However, medical devices can, and do sometimes fail. These failures pose significant risk to patients.ObjectiveOne way to address these issues is through the use of artificial intelligence for the detection of medical device failure. This goal of this study was to develop automated systems utilising machine learning algorithms to predict patient monitor performance and potential failures based on data collected during regular safety and performance inspections.MethodsThe system developed in this study utilised machine learning techniques as its core. Throughout the study four algorithms were utilised. These algorithms include Decision Tree, Random Forest, Linear Regression and Support Vector Machines.ResultsFinal results showed that Random Forest algorithms had the best performance on various metrics among the four developed models. It achieved accuracy of 94% and precision and recall of 70% and 93% respectively.ConclusionThis study shows that use of systems like the one developed in this study have the potential to improve management and maintenance of medical devices.","author":[{"family":"Bećirović","given":"Faruk"},{"family":"Spahić","given":"Lemana"},{"family":"Merdović","given":"Nejra"},{"family":"Pokvić","given":"Lejla"},{"family":"Badnjević","given":"Almir"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1177/09287329241291424","URL":"https://doi.org/10.1177/09287329241291424","source":"openalex"},{"id":"oa:W4403553192","type":"article-journal","title":"Harnessing Artificial Intelligence for Advancing Medical Manuscript Composition: Applications and Ethical Considerations","abstract":"Scientific medical manuscripts are fundamental to advancing research and enhancing patient care. With the emergence of artificial intelligence (AI), the process of composing such manuscripts has witnessed profound transformations. This review delves into the multifaceted role of AI in medical manuscript composition, analyzing its applications, benefits, drawbacks, and ethical implications. Employing a comprehensive narrative review methodology, we explored databases such as PubMed, Google Scholar, and Science Direct. The review charts the evolution of AI in medical writing, from basic word processing to sophisticated neural network-based models like GPT-3 and GPT-4. Various AI-powered tools such as ChatGPT, Google Bard, Elicit, and Consensus AI are examined in terms of their functionalities and contributions to research and medical writing. While AI technologies offer notable advantages in automating content creation and boosting research productivity, concerns persist regarding overreliance, potential homogenization of writing styles, and ethical considerations such as originality and authorship. Because of this concern, some companies are restricting the use of AI in peer review processes, medical examinations, etc. It is crucial to strike a balance in integrating AI tools, ensuring human oversight, conducting thorough algorithm audits, addressing financial implications, and upholding academic integrity. The review underscores the transformative potential of AI in medical manuscript composition while emphasizing the ongoing significance of human expertise, creativity, and ethical responsibility in scientific communication. Recommendations are provided for the effective integration of AI tools into medical writing processes, emphasizing collaborative efforts between AI developers, researchers, and journal editors to navigate ethical dilemmas and maximize the benefits of AI-driven advancements in scientific publishing.","author":[{"family":"Singh","given":"Shruti"},{"family":"Kumar","given":"Rajesh"},{"family":"Maharshi","given":"Vikas"},{"family":"Singh","given":"Prashant"},{"family":"Kumari","given":"Veena"},{"family":"Tiwari","given":"Meenakshi"},{"family":"Harsha","given":"Divya"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.71744","URL":"https://doi.org/10.7759/cureus.71744","source":"openalex"},{"id":"oa:W4398141110","type":"article-journal","title":"Assessment of Artificial Intelligence Platforms With Regard to Medical Microbiology Knowledge: An Analysis of ChatGPT and Gemini","abstract":"The performance of two artificial intelligence (AI) platforms, ChatGPT 3.5 (OpenAI, California, United States) and Gemini (Google AI, California, United States) was assessed by answering 200 questions of microbiology drawn from validated sources. The questions were selected from topics such as General Microbiology, Immunology, and Microbiology Applied to Infectious Diseases. The study was conducted from December 2023 to March 2024, and the responses of the different AI platforms were compared with an answer key. Statistical analysis was performed to assess accuracy. ChatGPT 3.5 and Gemini had comparable accuracy with correct response scores of 71% and 70.5%, respectively. Their performance varied across different sections. Gemini performed better in General Microbiology and Immunology, and ChatGPT 3.5 had a better score in the Applied Microbiology section. The study's findings highlight that AI platforms such as ChatGPT and Gemini can be utilized in microbiology and medical education. The evolution and continuous updating of AI platforms are required to improve their performance.","author":[{"family":"Ranjan","given":"Jai"},{"family":"Ahmad","given":"Absar"},{"family":"Subudhi","given":"Monalisa"},{"family":"Kumar","given":"Ajay"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.60675","URL":"https://doi.org/10.7759/cureus.60675","source":"openalex"},{"id":"oa:W4385421241","type":"article-journal","title":"Advancing Patient Care: How Artificial Intelligence Is Transforming Healthcare","abstract":"Artificial Intelligence (AI) has emerged as a transformative technology with immense potential in the field of medicine. By leveraging machine learning and deep learning, AI can assist in diagnosis, treatment selection, and patient monitoring, enabling more accurate and efficient healthcare delivery. The widespread implementation of AI in healthcare has the role to revolutionize patients' outcomes and transform the way healthcare is practiced, leading to improved accessibility, affordability, and quality of care. This article explores the diverse applications and reviews the current state of AI adoption in healthcare. It concludes by emphasizing the need for collaboration between physicians and technology experts to harness the full potential of AI.","author":[{"family":"Poalelungi","given":"Diana"},{"family":"Mușat","given":"Carmina"},{"family":"Fulga","given":"Ana"},{"family":"Neagu","given":"Marius"},{"family":"Neagu","given":"Anca‐iulia"},{"family":"Piraianu","given":"Alin"},{"family":"Fulga","given":"Iuliu"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/jpm13081214","URL":"https://doi.org/10.3390/jpm13081214","source":"openalex"},{"id":"oa:W4385241201","type":"article-journal","title":"Use of an Artificial Intelligence-Driven Digital Platform for Reflective Learning to Support Continuing Medical and Professional Education and Opportunities for Interprofessional Education and Equitable Access","abstract":"Continuing medical education (CME) and continuing education (CE) provide frameworks for assimilating and disseminating new advancements and are mainstays of clinicians’ professional development and accreditation. However, traditional CME/CE approaches may be challenged in providing opportunities for integrated and interprofessional learning and helping clinicians effectively translate innovations into individual practice. This Commentary describes the reflective learning approach, including its integration into CME/CE and how it can support interprofessional education. Also identified are barriers to reflective and interprofessional learning implementation and CME/CE access. The Commentary provides insights based on point-of-care reflection data and outlines considerations in trialing the use of an artificial intelligence (AI)-driven digital platform for reflective learning. Further, the Commentary describes how the AI-driven digital platform may help overcome barriers to reflective learning and interprofessional education and support equitable CME/CE program access.","author":[{"family":"Cohen","given":"Brian"},{"family":"Dubois","given":"Sasha"},{"family":"Lynch","given":"Patricia"},{"family":"Swami","given":"Niraj"},{"family":"Noftle","given":"Kelli"},{"family":"Arensberg","given":"Mary"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/educsci13080760","URL":"https://doi.org/10.3390/educsci13080760","source":"openalex"},{"id":"oa:W4388540336","type":"article-journal","title":"Introduction of telemedicine technologies based on artificial intelligence into practice of providing outpatient care for medical examination","abstract":"To date, the main focus in the implementation of medical activities is focused on the organization of outpatient care for the population. It is at the level of primary health care, where it is possible to carry out primary prevention measures and the formation of a healthy lifestyle, these issues should be given special attention when providing medical care to the population. One of the main tasks of modern healthcare is to reduce the number of chronic non-communicable diseases of the adult population using modern tools of early preclinical diagnostics. To date, one of the most promising areas that have a significant impact on modern healthcare is digital telemedicine technologies based on artificial intelligence. They can be confidently attributed to the most popular and rapidly developing groups of services developed for primary health care for the purpose of primary diagnostics. Nevertheless, no matter how fast information technologies develop, the opinions of experts in the subject area remain relevant, which significantly enrich the results of the expert opinion.","author":[{"family":"Seliverstov","given":"P"},{"family":"Шаповалов","given":"ВВ"},{"family":"Aleshko","given":"OV"}],"issued":{"date-parts":[[2023]]},"DOI":"10.33667/2078-5631-2023-28-44-49","URL":"https://doi.org/10.33667/2078-5631-2023-28-44-49","source":"openalex"},{"id":"oa:W4389179445","type":"article-journal","title":"Radiologists’ and Radiographers’ Perspectives on Artificial Intelligence in Medical Imaging in Saudi Arabia","abstract":"INTRODUCTION: Artificial intelligence (AI) in medical imaging rapidly expands regarding image processing and interpretation. Therefore, the aim was to explore radiographers’ and radiologists’ perceptions and attitudes towards AI use in medical imaging technologies in Saudi Arabia. METHODS: The survey was distributed online, and responses were collected from 173 participants nationwide. Data analysis was performed using SPSS Statistics (version 27). RESULTS: The participants scored an average of 1.7, 1.6, and 1.8 on a scale of 1–3 for attitudinal perspectives on clinical application and the positive and negative impact of integrating AI technology in diagnostic radiology. Lack of knowledge (43.9%) and perceived cyber threats (37.7%) were the most cited factors hindering AI implementation in Saudi Arabia. CONCLUSION: The radiographers and radiologists in this study had a favorable attitude toward AI integration in diagnostic radiology; nonetheless, concerns were raised about data protection, cyber security, AI-related errors, and decision-making challenges.","author":[{"family":"Alyami","given":"Ali"},{"family":"Majrashi","given":"Naif"},{"family":"Shubayr","given":"Nasser"}],"issued":{"date-parts":[[2023]]},"DOI":"10.2174/0115734056250970231117111810","URL":"https://doi.org/10.2174/0115734056250970231117111810","source":"openalex"},{"id":"oa:W4388525108","type":"article-journal","title":"Characterizing the Clinical Adoption of Medical AI Devices through U.S. Insurance Claims","abstract":"There are now over 500 medical artificial intelligence (AI) devices that are approved by the U.S. Food and Drug Administration. However, little is known about where and how often these devices are actually used after regulatory approval. In this article, we systematically quantify the adoption and usage of medical AI devices in the United States by tracking Current Procedural Terminology (CPT) codes explicitly created for medical AI. CPT codes are widely used for documenting billing and payment for medical procedures, providing a measure of device utilization across different clinical settings. We examined a comprehensive nationwide claims database of 11 billion CPT claims between January 1, 2018, and June 1, 2023 to analyze the prevalence of medical AI devices based on submitted claims. Our results indicate that medical AI device adoption is still nascent, with most usage driven by a handful of leading devices. For example, only AI devices used for assessing coronary artery disease and for diagnosing diabetic retinopathy have accumulated more than 10,000 CPT claims. Furthermore, we found that zip codes that had a higher income level, were metropolitan, and had academic medical centers were much more likely to have medical AI usage. Our study sheds light on the current landscape of medical AI device adoption and usage in the United States, underscoring the need to further investigate barriers and incentives to promote equitable access and broader integration of AI technologies in health care.","author":[{"family":"Wu","given":"Kevin"},{"family":"Wu","given":"Eric"},{"family":"Theodorou","given":"Brandon"},{"family":"Liang","given":"Weixin"},{"family":"Mack","given":"Christina"},{"family":"Glass","given":"Lucas"},{"family":"Sun","given":"Jimeng"},{"family":"Zou","given":"James"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1056/aioa2300030","URL":"https://doi.org/10.1056/aioa2300030","source":"openalex"},{"id":"oa:W4403052161","type":"article-journal","title":"From Theory to Practice: Artificial Intelligence (AI) Literacy Course for First-Year Medical Students","abstract":"Artificial intelligence (AI) is rapidly transforming healthcare by enhancing diagnostics, personalized medicine, and clinical decision-making. In medical education, AI chatbots have the potential to be used as virtual tutors or learning assistants. Despite AI's growing impact, its integration into medical education remains limited. AI is not a standard component of medical curricula, which could leave many future healthcare professionals unprepared for an AI-driven workplace. To address this significant gap, this editorial describes the development of a mini-course to integrate AI training for first-year medical students. The course was focused on the fundamentals of AI, prompt engineering, practical applications of chatbots as learning assistants, and ethical use of generative AI.","author":[{"family":"Levingston","given":"Hunter"},{"family":"Anderson","given":"Max"},{"family":"Roni","given":"Monzurul"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.70706","URL":"https://doi.org/10.7759/cureus.70706","source":"openalex"},{"id":"oa:W4388570274","type":"article-journal","title":"Strategies of Artificial intelligence tools in the domain of nanomedicine","abstract":"Nanomedicine is a field of medicine that uses nanotechnology to develop new diagnostic tools and therapies for a wide range of medical conditions. It encompasses a variety of different approaches, including the use of nanoparticles , nano-robots, and nanodevices. Some examples of how nanotechnology is being used in medicine include drug delivery, diagnostics imaging, tissue engineering, and biomedical devices. One of the main advantages of using nanomedicine-based drug delivery is the ability to deliver drugs at fixed and controlled doses. This is achieved by engineering nanoparticles to release drugs at a specific rate, which can be attuned to match the desires of the patient. Determining the optimal combination of nanotherapy, dosages, and administration schedule can be challenging, as it requires a thorough understanding of the pharmacokinetics and pharmacodynamics of the drugs, as well as the patient's individual physiognomies and the stage of the disease. To overcome this challenge, researchers and practitioners are increasingly using computational models, Artificial intelligence and machine learning algorithms, and personalized medicine approaches to predict the optimal drug combination, dosage, and administration schedule for each patient. In this review, we were given the outline of the different Artificial intelligence tools for the prediction of nanomedicine in the field of various medical applications.","author":[{"family":"Habeeb","given":"Mohammad"},{"family":"You","given":"Huay"},{"family":"Umapathi","given":"Mutheeswaran"},{"family":"Kanna","given":"RK"},{"family":"Hariyadi"},{"family":"Mishra","given":"Shweta"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.jddst.2023.105157","URL":"https://doi.org/10.1016/j.jddst.2023.105157","source":"openalex"},{"id":"oa:W4405033145","type":"manuscript","title":"Explainable Artificial Intelligence for Medical Applications: A Review","abstract":"The continuous development of artificial intelligence (AI) theory has propelled this field to unprecedented heights, owing to the relentless efforts of scholars and researchers. In the medical realm, AI takes a pivotal role, leveraging robust machine learning (ML) algorithms. AI technology in medical imaging aids physicians in X-ray, computed tomography (CT) scans, and magnetic resonance imaging (MRI) diagnoses, conducts pattern recognition and disease prediction based on acoustic data, delivers prognoses on disease types and developmental trends for patients, and employs intelligent health management wearable devices with human-computer interaction technology to name but a few. While these well-established applications have significantly assisted in medical field diagnoses, clinical decision-making, and management, collaboration between the medical and AI sectors faces an urgent challenge: How to substantiate the reliability of decision-making? The underlying issue stems from the conflict between the demand for accountability and result transparency in medical scenarios and the black-box model traits of AI. This article reviews recent research grounded in explainable artificial intelligence (XAI), with an emphasis on medical practices within the visual, audio, and multimodal perspectives. We endeavour to categorise and synthesise these practices, aiming to provide support and guidance for future researchers and healthcare professionals.","author":[{"family":"Sun","given":"Qiyang"},{"family":"Akman","given":"Alican"},{"family":"Schuller","given":"Björn"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2412.01829","URL":"https://doi.org/10.48550/arxiv.2412.01829","source":"openalex"},{"id":"oa:W4387873179","type":"article-journal","title":"Toward Explainable Artificial Intelligence for Precision Pathology","abstract":"The rapid development of precision medicine in recent years has started to challenge diagnostic pathology with respect to its ability to analyze histological images and increasingly large molecular profiling data in a quantitative, integrative, and standardized way. Artificial intelligence (AI) and, more precisely, deep learning technologies have recently demonstrated the potential to facilitate complex data analysis tasks, including clinical, histological, and molecular data for disease classification; tissue biomarker quantification; and clinical outcome prediction. This review provides a general introduction to AI and describes recent developments with a focus on applications in diagnostic pathology and beyond. We explain limitations including the black-box character of conventional AI and describe solutions to make machine learning decisions more transparent with so-called explainable AI. The purpose of the review is to foster a mutual understanding of both the biomedical and the AI side. To that end, in addition to providing an overview of the relevant foundations in pathology and machine learning, we present worked-through examples for a better practical understanding of what AI can achieve and how it should be done.","author":[{"family":"Klauschen","given":"Frederick"},{"family":"Dippel","given":"Jonas"},{"family":"Keyl","given":"Philipp"},{"family":"Jurmeister","given":"Philipp"},{"family":"Bockmayr","given":"Michael"},{"family":"Möck","given":"Andreas"},{"family":"Buchstab","given":"Oliver"},{"family":"Alber","given":"Maximilian"},{"family":"Ruff","given":"Lukas"},{"family":"Montavon","given":"Grégoire"},{"family":"Müller","given":"Klaus‐robert"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1146/annurev-pathmechdis-051222-113147","URL":"https://doi.org/10.1146/annurev-pathmechdis-051222-113147","source":"openalex"},{"id":"oa:W4368275176","type":"article-journal","title":"An artificial intelligence-based chatbot for prostate cancer education: Design and patient evaluation study","abstract":"Introduction: Artificial intelligence (AI) is increasingly used in healthcare. AI-based chatbots can act as automated conversational agents, capable of promoting health and providing education at any time. The objective of this study was to develop and evaluate a user-friendly medical chatbot (prostate cancer communication assistant (PROSCA)) for provisioning patient information about early detection of prostate cancer (PC). Methods: The chatbot was developed to provide information on prostate diseases, diagnostic tests for PC detection, stages, and treatment options. Ten men aged 49 to 81 years with suspicion of PC were enrolled in this study. Nine of ten patients used the chatbot during the evaluation period and filled out the questionnaires on usage and usability, perceived benefits, and potential for improvement. Results: The chatbot was straightforward to use, with 78% of users not needing any assistance during usage. In total, 89% of the chatbot users in the study experienced a clear to moderate increase in knowledge about PC through the chatbot. All study participants who tested the chatbot would like to re-use a medical chatbot in the future and support the use of chatbots in the clinical routine. Conclusions: Through the introduction of the chatbot PROSCA, we created and evaluated an innovative evidence-based health information tool in the field of PC, allowing targeted support for doctor-patient communication and offering great potential in raising awareness, patient education, and support. Our study revealed that a medical chatbot in the field of early PC detection is readily accepted and benefits patients as an additional informative tool.","author":[{"family":"Görtz","given":"Magdalena"},{"family":"Baumgärtner","given":"Kilian"},{"family":"Schmid","given":"T"},{"family":"Muschko","given":"Marc"},{"family":"Woessner","given":"Philipp"},{"family":"Gerlach","given":"Axel"},{"family":"Byczkowski","given":"Michael"},{"family":"Sültmann","given":"Holger"},{"family":"Duensing","given":"Stefan"},{"family":"Hohenfellner","given":"Markus"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1177/20552076231173304","URL":"https://doi.org/10.1177/20552076231173304","source":"openalex"},{"id":"oa:W4324359891","type":"article-journal","title":"Artificial Intelligence in Food Safety: A Decade Review and Bibliometric Analysis","abstract":"Artificial Intelligence (AI) technologies have been powerful solutions used to improve food yield, quality, and nutrition, increase safety and traceability while decreasing resource consumption, and eliminate food waste. Compared with several qualitative reviews on AI in food safety, we conducted an in-depth quantitative and systematic review based on the Core Collection database of WoS (Web of Science). To discover the historical trajectory and identify future trends, we analysed the literature concerning AI technologies in food safety from 2012 to 2022 by CiteSpace. In this review, we used bibliometric methods to describe the development of AI in food safety, including performance analysis, science mapping, and network analysis by CiteSpace. Among the 1855 selected articles, China and the United States contributed the most literature, and the Chinese Academy of Sciences released the largest number of relevant articles. Among all the journals in this field, PLoS ONE and Computers and Electronics in Agriculture ranked first and second in terms of annual publications and co-citation frequency. The present character, hot spots, and future research trends of AI technologies in food safety research were determined. Furthermore, based on our analyses, we provide researchers, practitioners, and policymakers with the big picture of research on AI in food safety across the whole process, from precision agriculture to precision nutrition, through 28 enlightening articles.","author":[{"family":"Liu","given":"Zhe"},{"family":"Wang","given":"Shuzhe"},{"family":"Zhang","given":"Yudong"},{"family":"Feng","given":"Yichen"},{"family":"Liu","given":"Jiajia"},{"family":"Zhu","given":"Hengde"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/foods12061242","URL":"https://doi.org/10.3390/foods12061242","source":"openalex"},{"id":"oa:W4376280052","type":"article-journal","title":"Artificial intelligence in thyroid ultrasound","abstract":"Artificial intelligence (AI), particularly deep learning (DL) algorithms, has demonstrated remarkable progress in image-recognition tasks, enabling the automatic quantitative assessment of complex medical images with increased accuracy and efficiency. AI is widely used and is becoming increasingly popular in the field of ultrasound. The rising incidence of thyroid cancer and the workload of physicians have driven the need to utilize AI to efficiently process thyroid ultrasound images. Therefore, leveraging AI in thyroid cancer ultrasound screening and diagnosis cannot only help radiologists achieve more accurate and efficient imaging diagnosis but also reduce their workload. In this paper, we aim to present a comprehensive overview of the technical knowledge of AI with a focus on traditional machine learning (ML) algorithms and DL algorithms. We will also discuss their clinical applications in the ultrasound imaging of thyroid diseases, particularly in differentiating between benign and malignant nodules and predicting cervical lymph node metastasis in thyroid cancer. Finally, we will conclude that AI technology holds great promise for improving the accuracy of thyroid disease ultrasound diagnosis and discuss the potential prospects of AI in this field.","author":[{"family":"Cao","given":"Chun‐li"},{"family":"Li","given":"Qiaoli"},{"family":"Jin","given":"Tong"},{"family":"Shi","given":"Li‐nan"},{"family":"Li","given":"Wen‐xiao"},{"family":"Ya","given":"Xu"},{"family":"Cheng","given":"Jing"},{"family":"Du","given":"Tingting"},{"family":"Li","given":"Jun"},{"family":"Cui","given":"Xin‐wu"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3389/fonc.2023.1060702","URL":"https://doi.org/10.3389/fonc.2023.1060702","source":"openalex"},{"id":"oa:W4386607939","type":"article-journal","title":"A Review on Applications of Artificial Intelligence in Wastewater Treatment","abstract":"In recent years, artificial intelligence (AI), as a rapidly developing and powerful tool to solve practical problems, has attracted much attention and has been widely used in various areas. Owing to their strong learning and accurate prediction abilities, all sorts of AI models have also been applied in wastewater treatment (WWT) to optimize the process, predict the efficiency and evaluate the performance, so as to explore more cost-effective solutions to WWT. In this review, we summarize and analyze various AI models and their applications in WWT. Specifically, we briefly introduce the commonly used AI models and their purposes, advantages and disadvantages, and comprehensively review the inputs, outputs, objectives and major findings of particular AI applications in water quality monitoring, laboratory-scale research and process design. Although AI models have gained great success in WWT-related fields, there are some challenges and limitations that hinder the widespread applications of AI models in real WWT, such as low interpretability, poor model reproducibility and big data demand, as well as a lack of physical significance, mechanism explanation, academic transparency and fair comparison. To overcome these hurdles and successfully apply AI models in WWT, we make recommendations and discuss the future directions of AI applications.","author":[{"family":"Wáng","given":"Yì"},{"family":"Cheng","given":"Yuhan"},{"family":"Liu","given":"He"},{"family":"Guo","given":"Qing"},{"family":"Dai","given":"Chuanjun"},{"family":"Zhao","given":"Min"},{"family":"Liu","given":"Dezhao"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/su151813557","URL":"https://doi.org/10.3390/su151813557","source":"openalex"},{"id":"oa:W4386304030","type":"article-journal","title":"Artificial Intelligence in Lung Cancer Screening: The Future Is Now","abstract":"Lung cancer has one of the worst morbidity and fatality rates of any malignant tumour. Most lung cancers are discovered in the middle and late stages of the disease, when treatment choices are limited, and patients' survival rate is low. The aim of lung cancer screening is the identification of lung malignancies in the early stage of the disease, when more options for effective treatments are available, to improve the patients' outcomes. The desire to improve the efficacy and efficiency of clinical care continues to drive multiple innovations into practice for better patient management, and in this context, artificial intelligence (AI) plays a key role. AI may have a role in each process of the lung cancer screening workflow. First, in the acquisition of low-dose computed tomography for screening programs, AI-based reconstruction allows a further dose reduction, while still maintaining an optimal image quality. AI can help the personalization of screening programs through risk stratification based on the collection and analysis of a huge amount of imaging and clinical data. A computer-aided detection (CAD) system provides automatic detection of potential lung nodules with high sensitivity, working as a concurrent or second reader and reducing the time needed for image interpretation. Once a nodule has been detected, it should be characterized as benign or malignant. Two AI-based approaches are available to perform this task: the first one is represented by automatic segmentation with a consequent assessment of the lesion size, volume, and densitometric features; the second consists of segmentation first, followed by radiomic features extraction to characterize the whole abnormalities providing the so-called \"virtual biopsy\". This narrative review aims to provide an overview of all possible AI applications in lung cancer screening.","author":[{"family":"Cellina","given":"Michaela"},{"family":"Cacioppa","given":"Laura"},{"family":"Cè","given":"Maurizio"},{"family":"Chiarpenello","given":"Vittoria"},{"family":"Costa","given":"Marco"},{"family":"Vincenzo","given":"Zakaria"},{"family":"Pais","given":"Daniele"},{"family":"Bausano","given":"Maria"},{"family":"Rossini","given":"Nicolò"},{"family":"Bruno","given":"Alessandra"},{"family":"Floridi","given":"Chiara"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/cancers15174344","URL":"https://doi.org/10.3390/cancers15174344","source":"openalex"},{"id":"oa:W4366415335","type":"article-journal","title":"Analysis of IoT Security Challenges and Its Solutions Using Artificial Intelligence","abstract":"The Internet of Things (IoT) is a well-known technology that has a significant impact on many areas, including connections, work, healthcare, and the economy. IoT has the potential to improve life in a variety of contexts, from smart cities to classrooms, by automating tasks, increasing output, and decreasing anxiety. Cyberattacks and threats, on the other hand, have a significant impact on intelligent IoT applications. Many traditional techniques for protecting the IoT are now ineffective due to new dangers and vulnerabilities. To keep their security procedures, IoT systems of the future will need AI-efficient machine learning and deep learning. The capabilities of artificial intelligence, particularly machine and deep learning solutions, must be used if the next-generation IoT system is to have a continuously changing and up-to-date security system. IoT security intelligence is examined in this paper from every angle available. An innovative method for protecting IoT devices against a variety of cyberattacks is to use machine learning and deep learning to gain information from raw data. Finally, we discuss relevant research issues and potential next steps considering our findings. This article examines how machine learning and deep learning can be used to detect attack patterns in unstructured data and safeguard IoT devices. We discuss the challenges that researchers face, as well as potential future directions for this research area, considering these findings. Anyone with an interest in the IoT or cybersecurity can use this website's content as a technical resource and reference.","author":[{"family":"Mazhar","given":"Tehseen"},{"family":"Talpur","given":"Dhani"},{"family":"Shloul","given":"Tamara"},{"family":"Ghadi","given":"Yazeed"},{"family":"Haq","given":"Inayatul"},{"family":"Ullah","given":"Inam"},{"family":"Ouahada","given":"Khmaies"},{"family":"Hamam","given":"Habib"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/brainsci13040683","URL":"https://doi.org/10.3390/brainsci13040683","source":"openalex"},{"id":"oa:W4396768244","type":"article-journal","title":"Role of artificial intelligence in revolutionizing drug discovery","abstract":"The application of artificial intelligence (AI) in medicine, particularly through machine learning (ML), marked a significant progression in drug discovery. AI acts as a powerful catalyst in narrowing the gap between disease understanding and the identification of potential therapeutic agents. This review provides an inclusive summary of the latest advancements in AI and its application in drug discovery. We examine the various stages of the drug discovery process, starting from disease identification and encompassing diagnosis, target identification, screening, and lead discovery. AI's capability to analyze extensive datasets and discern patterns is essential in these stages, enhancing predictions and efficiencies in disease identification, drug discovery, and clinical trial management. The role of AI in expediting drug development is emphasized, highlighting its potential to analyze vast data volumes, thus reducing the time and costs associated with new drug market introduction. The importance of data quality, algorithm training, and ethical considerations, especially in patient data handling during clinical trials, is addressed. By considering these factors, AI promises to transform drug development, offering significant benefits to patients and society.","author":[{"family":"Rehman","given":"Ashfaq"},{"family":"Li","given":"Mingyu"},{"family":"Wu","given":"Binjian"},{"family":"Ali","given":"Yasir"},{"family":"Rasheed","given":"Salman"},{"family":"Shaheen","given":"Sana"},{"family":"Liu","given":"Xinyi"},{"family":"Luo","given":"Ray"},{"family":"Zhang","given":"Jian"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.fmre.2024.04.021","URL":"https://doi.org/10.1016/j.fmre.2024.04.021","source":"openalex"},{"id":"oa:W4394014418","type":"article-journal","title":"Application of Artificial Intelligence in Tissue Engineering","abstract":"Tissue engineering, a crucial approach in medical research and clinical applications, aims to regenerate damaged organs. By combining stem cells, biochemical factors, and biomaterials, it encounters challenges in designing complex 3D structures. Artificial intelligence (AI) enhances tissue engineering through computational modeling, biomaterial design, cell culture optimization, and personalized medicine. This review explores AI applications in organ tissue engineering (bone, heart, nerve, skin, cartilage), employing various machine learning (ML) algorithms for data analysis, prediction, and optimization. Each section discusses common ML algorithms and specific applications, emphasizing the potential and challenges in advancing regenerative therapies.","author":[{"family":"Bagherpour","given":"Reza"},{"family":"Bagherpour","given":"Ghasem"},{"family":"Mohammadi","given":"Parvin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1089/ten.teb.2024.0022","URL":"https://doi.org/10.1089/ten.teb.2024.0022","source":"openalex"},{"id":"oa:W4405292214","type":"article-journal","title":"Is artificial intelligence for everyone? Analyzing the role of ChatGPT as a writing assistant for medical students","abstract":"This study explores the potential impact of ChatGPT on the academic writing skills development of medical students enrolled in a compulsory 3-unit writing course at a medical university. The research focuses on two primary objectives, which are formulated as two research questions: Firstly, does the use of ChatGPT enhance medical students’ English academic writing skills compared to conventional writing training? Secondly, how does the use of ChatGPT impact on different components of academic writing? A longitudinal intervention design was employed with 83 participants from two writing classes in the experimental and control groups. The findings demonstrated ChatGPT’s significant impact on enhancing medical students’ English academic writing skills, with large effect sizes. ChatGPT enhanced students’ writing skills, especially content, organization, vocabulary, and mechanics in the experimental group, while its impact on language use is limited. AI tools like ChatGPT can be valuable in assisting with certain aspects of writing, but they should not be considered a one-size-fits-all solution for enhancing writing skills. The result of the study can be beneficial for educators, particularly those interested in teaching writing.","author":[{"family":"Shahsavar","given":"Zahra"},{"family":"Kafipour","given":"Reza"},{"family":"Khojasteh","given":"Laleh"},{"family":"Pakdel","given":"Farhad"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/feduc.2024.1457744","URL":"https://doi.org/10.3389/feduc.2024.1457744","source":"openalex"},{"id":"oa:W4394011823","type":"article-journal","title":"Artificial intelligence in lung cancer screening: Detection, classification, prediction, and prognosis","abstract":"BACKGROUND: The exceptional capabilities of artificial intelligence (AI) in extracting image information and processing complex models have led to its recognition across various medical fields. With the continuous evolution of AI technologies based on deep learning, particularly the advent of convolutional neural networks (CNNs), AI presents an expanded horizon of applications in lung cancer screening, including lung segmentation, nodule detection, false-positive reduction, nodule classification, and prognosis. METHODOLOGY: This review initially analyzes the current status of AI technologies. It then explores the applications of AI in lung cancer screening, including lung segmentation, nodule detection, and classification, and assesses the potential of AI in enhancing the sensitivity of nodule detection and reducing false-positive rates. Finally, it addresses the challenges and future directions of AI in lung cancer screening. RESULTS: AI holds substantial prospects in lung cancer screening. It demonstrates significant potential in improving nodule detection sensitivity, reducing false-positive rates, and classifying nodules, while also showing value in predicting nodule growth and pathological/genetic typing. CONCLUSIONS: AI offers a promising supportive approach to lung cancer screening, presenting considerable potential in enhancing nodule detection sensitivity, reducing false-positive rates, and classifying nodules. However, the universality and interpretability of AI results need further enhancement. Future research should focus on the large-scale validation of new deep learning-based algorithms and multi-center studies to improve the efficacy of AI in lung cancer screening.","author":[{"family":"Quanyang","given":"Wu"},{"family":"Yao","given":"Huang"},{"family":"Sicong","given":"Wang"},{"family":"Linlin","given":"Qi"},{"family":"Zewei","given":"Zhang"},{"family":"Donghui","given":"Hou"},{"family":"Li","given":"Hongjia"},{"family":"Zhao","given":"Shijun"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/cam4.7140","URL":"https://doi.org/10.1002/cam4.7140","source":"openalex"},{"id":"oa:W4393993063","type":"article-journal","title":"Artificial Intelligence in Medical Imaging: Analyzing the Performance of ChatGPT and Microsoft Bing in Scoliosis Detection and Cobb Angle Assessment","abstract":"Open-source artificial intelligence models (OSAIM) find free applications in various industries, including information technology and medicine. Their clinical potential, especially in supporting diagnosis and therapy, is the subject of increasingly intensive research. Due to the growing interest in artificial intelligence (AI) for diagnostic purposes, we conducted a study evaluating the capabilities of AI models, including ChatGPT and Microsoft Bing, in the diagnosis of single-curve scoliosis based on posturographic radiological images. Two independent neurosurgeons assessed the degree of spinal deformation, selecting 23 cases of severe single-curve scoliosis. Each posturographic image was separately implemented onto each of the mentioned platforms using a set of formulated questions, starting from 'What do you see in the image?' and ending with a request to determine the Cobb angle. In the responses, we focused on how these AI models identify and interpret spinal deformations and how accurately they recognize the direction and type of scoliosis as well as vertebral rotation. The Intraclass Correlation Coefficient (ICC) with a 'two-way' model was used to assess the consistency of Cobb angle measurements, and its confidence intervals were determined using the F test. Differences in Cobb angle measurements between human assessments and the AI ChatGPT model were analyzed using metrics such as RMSEA, MSE, MPE, MAE, RMSLE, and MAPE, allowing for a comprehensive assessment of AI model performance from various statistical perspectives. The ChatGPT model achieved 100% effectiveness in detecting scoliosis in X-ray images, while the Bing model did not detect any scoliosis. However, ChatGPT had limited effectiveness (43.5%) in assessing Cobb angles, showing significant inaccuracy and discrepancy compared to human assessments. This model also had limited accuracy in determining the direction of spinal curvature, classifying the type of scoliosis, and detecting vertebral rotation. Overall, although ChatGPT demonstrated potential in detecting scoliosis, its abilities in assessing Cobb angles and other parameters were limited and inconsistent with expert assessments. These results underscore the need for comprehensive improvement of AI algorithms, including broader training with diverse X-ray images and advanced image processing techniques, before they can be considered as auxiliary in diagnosing scoliosis by specialists.","author":[{"family":"Fabijan","given":"Artur"},{"family":"Zawadzka-Fabijan","given":"Agnieszka"},{"family":"Fabijan","given":"Robert"},{"family":"Zakrzewski","given":"Krzysztof"},{"family":"Nowosławska","given":"Emilia"},{"family":"Polis","given":"Bartosz"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/diagnostics14070773","URL":"https://doi.org/10.3390/diagnostics14070773","source":"openalex"},{"id":"oa:W4400170199","type":"article-journal","title":"Explainable Artificial Intelligence in Medical Imaging: A Case Study on Enhancing Lung Cancer Detection through CT Images","abstract":"This study tackles the pressing challenge of lung cancer detection, the foremost cause of cancer-related mortality worldwide, hindered by late detection and diagnostic limitations. Aiming to improve early detection rates and diagnostic reliability, we propose an approach integrating Deep Convolutional Neural Networks (DCNN) with Explainable Artificial Intelligence (XAI) techniques, specifically focusing on the Residual Network (ResNet) architecture and Gradient-weighted Class Activation Mapping (Grad-CAM). Utilizing a dataset of 1,000 CT scans, categorized into normal, non-cancerous, and three types of lung cancer images, we adapted the ResNet50 model through transfer learning and fine-tuning for enhanced specificity in lung cancer subtype detection. Our methodology demonstrated the modified ResNet50 model's effectiveness, significantly outperforming the original architecture in accuracy (91.11%), precision (91.66%), sensitivity (91.11%), specificity (96.63%), and F1-score (91.10%). The inclusion of Grad-CAM provided insightful visual explanations for the model's predictions, fostering transparency and trust in computer-assisted diagnostics. The study highlights the potential of combining DCNN with XAI to advance lung cancer detection, suggesting future research should expand dataset diversity and explore multimodal data integration for broader applicability and improved diagnostic capabilities.","author":[{"family":"Noviandy","given":"Teuku"},{"family":"Maulana","given":"Aga"},{"family":"Zulfikar","given":"Teuku"},{"family":"Rusyana","given":"Asep"},{"family":"Enitan","given":"Seyi"},{"family":"Idroes","given":"Rinaldi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.60084/ijcr.v2i1.150","URL":"https://doi.org/10.60084/ijcr.v2i1.150","source":"openalex"},{"id":"oa:W4372291572","type":"article-journal","title":"Implementing Artificial Intelligence in Higher Education: Pros and Cons from the Perspectives of Academics","abstract":"This article investigates the perspectives of Romanian academics on implementing Artificial Intelligence (AI) in Higher Education (HE). The article analyzes the pros and cons of AI in HE, based on the views of eighteen academics from five Romanian universities. There is a large and heated debate about the proliferation of AI in many domains, with strong supporters and determined deniers. Studies that research the implications of AI enrich the evidence-based literature on the advantages, disadvantages, threats, or opportunities that AI creates for us, for businesses, or for societies. Though many aspects are still less well known, attitudes toward AI are still under construction. HE is a domain where the implications of AI create passionate discussions. HE is, eventually, the sector that shapes the masterminds of societies’ leaders. There is a quest to find the perspectives of those who will apply AI, who will work with or for AI, and those who are opposed to or in favor of implementing AI in HE. The conclusions revealed by this study are in line with similar studies that exist in the literature. The positive aspects of AI implementation in HE are related, in the view of academics, to gains in the learning–teaching process, improvements in students skills and competences, better inclusion, and greater efficiency in administrative costs. Similarly, the negative aspects revealed by the research are linked to psychosocial effects, data security, ethical aspects, and unemployment threats. However, there are some aspects (mostly negative) related to implementing AI in HE that are less exposed by the interviewed academics, which are mostly related to the costs and efforts of implementing AI in HE. The possible explanation of this situation is related to the lack of strategic vision on what, in fact, the implementation of AI in HE means, what this process involves, and the fact that digitalization in Romanian universities (as well as in the Romanian economy) is in its infancy. The contribution of the results of this research is mainly empirical and practical. These opinions should be used as resources for managers of HE institutions to develop better policies concerning the implementation of AI in HE and for strategic vision toward AI, with the ultimate purpose of achieving progress and prosperity for the entire society.","author":[{"family":"Pisica","given":"Alina"},{"family":"Edu","given":"Tudor"},{"family":"Zaharia","given":"Rodica"},{"family":"Zaharia","given":"Rodica"},{"family":"Zaharia","given":"Razvan"},{"family":"Zaharia","given":"Razvan"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/soc13050118","URL":"https://doi.org/10.3390/soc13050118","source":"openalex"},{"id":"oa:W4386223230","type":"article-journal","title":"Internet of Medical Things and Healthcare 4.0: Trends, Requirements, Challenges, and Research Directions","abstract":"Healthcare 4.0 is a recent e-health paradigm associated with the concept of Industry 4.0. It provides approaches to achieving precision medicine that delivers healthcare services based on the patient's characteristics. Moreover, Healthcare 4.0 enables telemedicine, including telesurgery, early predictions, and diagnosis of diseases. This represents an important paradigm for modern societies, especially with the current situation of pandemics. The release of the fifth-generation cellular system (5G), the current advances in wearable device manufacturing, and the recent technologies, e.g., artificial intelligence (AI), edge computing, and the Internet of Things (IoT), are the main drivers of evolutions of Healthcare 4.0 systems. To this end, this work considers introducing recent advances, trends, and requirements of the Internet of Medical Things (IoMT) and Healthcare 4.0 systems. The ultimate requirements of such networks in the era of 5G and next-generation networks are discussed. Moreover, the design challenges and current research directions of these networks. The key enabling technologies of such systems, including AI and distributed edge computing, are discussed.","author":[{"family":"Osama","given":"Manar"},{"family":"Ateya","given":"Abdelhamied"},{"family":"Sayed","given":"Mohammed"},{"family":"Hammad","given":"Mohamed"},{"family":"Pławiak","given":"Paweł"},{"family":"Ellatif","given":"Ahmed"},{"family":"Elsayed","given":"Rania"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/s23177435","URL":"https://doi.org/10.3390/s23177435","source":"openalex"},{"id":"oa:W4392488972","type":"article-journal","title":"Data-Centric Artificial Intelligence","abstract":"Abstract Data-centric artificial intelligence (data-centric AI) represents an emerging paradigm that emphasizes the importance of enhancing data systematically and at scale to build effective and efficient AI-based systems. The novel paradigm complements recent model-centric AI, which focuses on improving the performance of AI-based systems based on changes in the model using a fixed set of data. The objective of this article is to introduce practitioners and researchers from the field of Business and Information Systems Engineering (BISE) to data-centric AI. The paper defines relevant terms, provides key characteristics to contrast the paradigm of data-centric AI with the model-centric one, and introduces a framework to illustrate the different dimensions of data-centric AI. In addition, an overview of available tools for data-centric AI is presented and this novel paradigm is differenciated from related concepts. Finally, the paper discusses the longer-term implications of data-centric AI for the BISE community.","author":[{"family":"Jakubik","given":"Johannes"},{"family":"Vössing","given":"Michael"},{"family":"Kühl","given":"Niklas"},{"family":"Walk","given":"Jannis"},{"family":"Satzger","given":"Gerhard"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s12599-024-00857-8","URL":"https://doi.org/10.1007/s12599-024-00857-8","source":"openalex"},{"id":"oa:W4385652559","type":"article-journal","title":"Utilizing Artificial Intelligence for Crafting Medical Examinations: A Medical Education Study with GPT-4","abstract":"Abstract Background. The task of writing multiple choice question examinations for medical students is complex, timely and requires significant efforts from clinical staff and faculty. Applying artificial intelligence algorithms in this field of medical education may be advisable. Methods. We utilized GPT-4, an OpenAI application, to write a 210 multi choice questions-MCQs examination based on an existing exam template and thoroughly investigated the output by specialist physicians who were blinded to the source of the questions. Algorithm mistakes and inaccuracies were categorized by their characteristics. Results. After inputting a detailed prompt, GPT-4 produced the test rapidly and effectively. Only 1 question (0.5%) was defined as false; 15% of questions necessitated revisions. Errors in the AI-generated questions included: the use of outdated or inaccurate terminology, age-sensitive inaccuracies, gender-sensitive inaccuracies, and geographically sensitive inaccuracies. Questions that were disqualified due to flawed methodology basis included elimination-based questions and questions that did not include elements of integrating knowledge with clinical reasoning. Conclusion. GPT can be used as an adjunctive tool in creating multi-choice question medical examinations yet rigorous inspection by specialist physicians remains pivotal.","author":[{"family":"Klang","given":"Eyal"},{"family":"Portugez","given":"Shir"},{"family":"Gross","given":"Raz"},{"family":"Lerner","given":"Reut"},{"family":"Brenner","given":"Alina"},{"family":"Gilboa","given":"Maayan"},{"family":"Ortal","given":"Tal"},{"family":"Ron","given":"Sophi"},{"family":"Robinzon","given":"Vered"},{"family":"Meiri","given":"Hila"},{"family":"Segal","given":"Gad"}],"issued":{"date-parts":[[2023]]},"DOI":"10.21203/rs.3.rs-3146947/v1","URL":"https://doi.org/10.21203/rs.3.rs-3146947/v1","source":"openalex"},{"id":"oa:W4393405308","type":"article-journal","title":"Sora for Computational Social Systems: From Counterfactual Experiments to Artificiofactual Experiments With Parallel Intelligence","abstract":"Welcome to the second issue of IEEE Transactions on Computational Social Systems (TCSS) of 2024. This issue showcases an impressive array of 104 regular papers alongside our Special Issue on Big Data and Computational Social Intelligence for Guaranteed Financial Security, highlighting cutting-edge research aimed at harnessing big data and computational techniques to fortify financial security amidst the digital finance evolution. With a focus on addressing the intricate challenges of financial big data, enhancing the efficacy of artificial intelligence, and covering critical topics from data mining to digital currencies, this issue underscores the vital role of cross-disciplinary efforts in mitigating financial security risks.","author":[{"family":"Qin","given":"Rui"},{"family":"Wang","given":"Fei–yue"},{"family":"Zheng","given":"Xiaolong"},{"family":"Ni","given":"Qinghua"},{"family":"Li","given":"Juanjuan"},{"family":"Xue","given":"Xiao"},{"family":"Hu","given":"Bin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1109/tcss.2024.3373928","URL":"https://doi.org/10.1109/tcss.2024.3373928","source":"openalex"},{"id":"oa:W4400668233","type":"article-journal","title":"From Pixels to Information: Artificial Intelligence in Fluorescence Microscopy","abstract":"This review explores how artificial intelligence (AI) is transforming fluorescence microscopy, providing an overview of its fundamental principles and recent advancements. The roles of AI in improving image quality and introducing new imaging modalities are discussed, offering a comprehensive perspective on these changes. Additionally, a unified framework is introduced for comprehending AI‐driven microscopy methodologies and categorizing them into linear inverse problem‐solving, denoising, and nonlinear prediction. Furthermore, the potential of self‐supervised learning techniques that address the challenges associated with training the networks are explored, utilizing unlabeled microscopy data to enhance data quality and expand imaging capabilities. It is worth noting that while the specific examples and advancements discussed in this review focus on fluorescence microscopy, the general approaches and theories are directly applicable to other optical microscopy methods.","author":[{"family":"Han","given":"Seungjae"},{"family":"You","given":"Joshua"},{"family":"Eom","given":"Minho"},{"family":"Ahn","given":"Sungjin"},{"family":"Cho","given":"Eun‐seo"},{"family":"Yoon","given":"Young‐gyu"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/adpr.202300308","URL":"https://doi.org/10.1002/adpr.202300308","source":"openalex"},{"id":"oa:W4404547649","type":"article-journal","title":"Temporomandibular joint assessment in MRI images using artificial intelligence tools: where are we now? A systematic review","abstract":"OBJECTIVES: To summarize the current evidence on the performance of artificial intelligence (AI) algorithms for the temporomandibular joint (TMJ) disc assessment and TMJ internal derangement diagnosis in magnetic resonance imaging (MRI) images. METHODS: Studies were gathered by searching 5 electronic databases and partial grey literature up to May 27, 2024. Studies in humans using AI algorithms to detect or diagnose internal derangements in MRI images were included. The methodological quality of the studies was evaluated using the Quality Assessment Tool for Diagnostic of Accuracy Studies-2 (QUADAS-2) and a proposed checklist for dental AI studies. RESULTS: Thirteen studies were included in this systematic review. Most of the studies assessed disc position. One study assessed disc perforation. A high heterogeneity related to the patient selection domain was found between the studies. The studies used a variety of AI approaches and performance metrics with CNN-based models being the most used. A high performance of AI models compared to humans was reported with accuracy ranging from 70% to 99%. CONCLUSIONS: The integration of AI, particularly deep learning, in TMJ MRI, shows promising results as a diagnostic-assistance tool to segment TMJ structures and classify disc position. Further studies exploring more diverse and multicentre data will improve the validity and generalizability of the models before being implemented in clinical practice.","author":[{"family":"Manek","given":"Mitul"},{"family":"Maita","given":"Ibraheem"},{"family":"Silva","given":"Diego"},{"family":"Melo","given":"Daniela"},{"family":"Major","given":"Paul"},{"family":"Jaremko","given":"Jacob"},{"family":"Almeida","given":"Fabiana"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/dmfr/twae055","URL":"https://doi.org/10.1093/dmfr/twae055","source":"openalex"},{"id":"oa:W4401558357","type":"article-journal","title":"Potential roles for artificial intelligence in clinical microbiology from improved diagnostic accuracy to solving the staffing crisis","abstract":"OBJECTIVES: This review summarizes the current and potential uses of artificial intelligence (AI) in the current state of clinical microbiology with a focus on replacement of labor-intensive tasks. METHODS: A search was conducted on PubMed using the key terms clinical microbiology and artificial intelligence. Studies were reviewed for relevance to clinical microbiology, current diagnostic techniques, and potential advantages of AI in routine microbiology workflows. RESULTS: Numerous studies highlight potential labor, as well as diagnostic accuracy, benefits to the implementation of AI for slide-based and macroscopic digital image analyses. These range from Gram stain interpretation to categorization and quantitation of culture growth. CONCLUSIONS: Artificial intelligence applications in clinical microbiology significantly enhance diagnostic accuracy and efficiency, offering promising solutions to labor-intensive tasks and staffing shortages. More research efforts and US Food and Drug Administration clearance are still required to fully incorporate these AI applications into routine clinical laboratory practices.","author":[{"family":"Graf","given":"Erin"},{"family":"Soliman","given":"Amr"},{"family":"Marouf","given":"Mohamed"},{"family":"Parwani","given":"Anil"},{"family":"Pancholi","given":"Preeti"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/ajcp/aqae107","URL":"https://doi.org/10.1093/ajcp/aqae107","source":"openalex"},{"id":"oa:W4402317328","type":"article-journal","title":"From Theory to Practice: A Holistic Study of the Application of Artificial Intelligence Methods and Techniques in Higher Education and Science","abstract":"This study endeavors to conduct an exhaustive analysis of the integration of artificial intelligence (AI) into educational and scientific practices, and to elucidate potential pathways for progress in this domain. It involves reflecting on the impact of AI across various education domains, the advancement of scientific methodologies and discoveries, and the broader societal development. With the help of a systematic review of the relevant literature, examples, and trends of the application of AI in education and science are studied, emphasizing their methodological and conceptual basis. The qualitative approach of this study is based on a systematic analytical review of academic publications, with an attempt to identify key topics and trends in the integration of AI in the fields of education and science. The critical analysis of relevant research assesses the reliability of the presented evidence and applied research methods, and examines the differences and convergence of the approaches of different authors. This methodological approach allows a more profound analysis of AI's impact, while also exploring AI as an advanced research methodology and analyzing various perspectives and contributions of authors within the realms of education and science. The results demonstrate that AI integration significantly contributes to improving educational processes, fostering student creativity, and enhancing scientific practices. Furthermore, the study identifies research gaps, emphasizing the need to explore ethical implications, long-term impacts, and inclusive models of AI. Based on these findings, further studies employing longitudinal/ experimental methods and large dataset analyses are recommended. The findings are expected to advance research by deepening our understanding of the complex interactions between artificial intelligence, education, and science.","author":[{"family":"Džogović","given":"Suada"},{"family":"Zdravkovska-Adamova","given":"Blagojka"},{"family":"Serpil","given":"Harun"}],"issued":{"date-parts":[[2024]]},"DOI":"10.21554/hrr.092406","URL":"https://doi.org/10.21554/hrr.092406","source":"openalex"},{"id":"oa:W4404583323","type":"article-journal","title":"Disruptive technologies in the university curriculum: use of artificial intelligence","abstract":"The so-called “digital era” is synonymous with the transformation of every aspect of human life. This transformation is given by the development of new technologies that modify the way humans communicate and cooperate. Now, it can be said that formal education, compared to other economic sectors, is lagging in the integration of novel technologies in higher education curricula, especially in terms of implementing artificial intelligence (AI). The objective of this research was to conduct a systematic review of the scientific production related to the incorporation of artificial intelligence as a disruptive technology in the university curriculum. It was carried out using a qualitative approach based on a systematic review. The review showed a greater scientific production between 2022 and 2023; it was also evidenced that, as a technology, artificial intelligence has become a disruptive element thanks to its ability to change the role and work performed by teachers, students, and educational institutions. Consequently, the university of the future urgently needs to plan, design, develop, and implement curricula that include artificial intelligence, with the purpose of training better professionals, capable of acting effectively in a technological and productive environment.","author":[{"family":"Huapaya","given":"Enma"},{"family":"Chucos","given":"Gilmer"},{"family":"Sosa","given":"Efrain"},{"family":"Meza","given":"Melva"}],"issued":{"date-parts":[[2024]]},"DOI":"10.11591/ijere.v14i1.30450","URL":"https://doi.org/10.11591/ijere.v14i1.30450","source":"openalex"},{"id":"oa:W4404866020","type":"article-journal","title":"Artificial intelligence literacy scale: A study of reliability and validity in Turkish university students","abstract":"This study aims to adapt to Turkish the \"Scale for the assessment of non-experts: AI literacy\" developed by Laupichler et al. (2023a). The scale consists of 31 items with three sub-dimensions: technical understanding, critical thinking, and practical applications. The data required for the validity and reliability study of the scale were collected from 642 undergraduate and graduate students studying in different departments of a state university in the fall semester of the 2023-2024 academic year. First, CFA was applied to the data according to the factor structure in the original scale, but as acceptable fit values could not be obtained because of the analysis, exploratory factor analysis was performed. In the reliability analysis of the factor structure determined by EFA, KMO was calculated as =0.948. It was determined that the scale items were collected in three factors and explained 61.1% of the total variance (\"critical thinking\" is 25.8%, \"technical knowledge\" is 25.2%, and \"practical applications\" explains 10.2% of the total variance). As a result of EFA, it was seen that the sub-dimensions of some of the items in the original scale had changed, and since the factor load values of the three items were very close to each other, they were removed from the scale. Because of CFA, which was conducted to evaluate whether the data supported the hypothesized relationships between the measured variables, Cronbach’s alpha value was found to be 0.90. As a result of the CFA analysis conducted with the 3 sub-dimensions and 28 items in the scale, the Chi-square value (X²=2.85; df=345, N=317, p< .001), which is the fit index of the model, has a good fit and is significant, SRMR = 0.0545and RMSEA = 0.077 values and fit indices and the model has an acceptable fit.","author":[{"family":"Topal","given":"Arzu"},{"family":"Gökçe","given":"Asiye"},{"family":"Eren","given":"Canan"},{"family":"Geçer","given":"Aynur"}],"issued":{"date-parts":[[2024]]},"DOI":"10.53850/joltida.1440845","URL":"https://doi.org/10.53850/joltida.1440845","source":"openalex"},{"id":"oa:W4387110043","type":"article-journal","title":"Actionable Artificial Intelligence for the Future of Production","abstract":"Abstract The Internet of Production (IoP) promises to be the answer to major challenges facing the Industrial Internet of Things (IIoT) and Industry 4.0. The lack of inter-company communication channels and standards, the need for heightened safety in Human Robot Collaboration (HRC) scenarios, and the opacity of data-driven decision support systems are only a few of the challenges we tackle in this chapter. We outline the communication and data exchange within the World Wide Lab (WWL) and autonomous agents that query the WWL which is built on the Digital Shadows (DS). We categorize our approaches into machine level, process level, and overarching principles. This chapter surveys the interdisciplinary work done in each category, presents different applications of the different approaches, and offers actionable items and guidelines for future work.The machine level handles the robots and machines used for production and their interactions with the human workers. It covers low-level robot control and optimization through gray-box models, task-specific motion planning, and optimization through reinforcement learning. In this level, we also examine quality assurance through nonintrusive real-time quality monitoring, defect recognition, and quality prediction. Work on this level also handles confidence, verification, and validation of re-configurable processes and reactive, modular, transparent process models. The process level handles the product life cycle, interoperability, and analysis and optimization of production processes, which is overall attained by analyzing process data and event logs to detect and eliminate bottlenecks and learn new process models. Moreover, this level presents a communication channel between human workers and processes by extracting and formalizing human knowledge into ontology and providing a decision support by reasoning over this information. Overarching principles present a toolbox of omnipresent approaches for data collection, analysis, augmentation, and management, as well as the visualization and explanation of black-box models.","author":[{"family":"Behery","given":"Mohamed"},{"family":"Brauner","given":"Philipp"},{"family":"Zhou","given":"Hans"},{"family":"Uysal","given":"Merih"},{"family":"Самсонов","given":"Владимир"},{"family":"Bellgardt","given":"Martin"},{"family":"Brillowski","given":"Florian"},{"family":"Brockhoff","given":"Tobias"},{"family":"Ghahfarokhi","given":"Anahita"},{"family":"Gleim","given":"Lars"},{"family":"Gorißen","given":"Leon"},{"family":"Grochowski","given":"Marco"},{"family":"Henn","given":"Thomas"},{"family":"Iacomini","given":"Elisa"},{"family":"Käster","given":"Thomas"},{"family":"Koren","given":"István"},{"family":"Liebenberg","given":"Martin"},{"family":"Reinsch","given":"Leon"},{"family":"Tirpitz","given":"Liam"},{"family":"Trinh","given":"Minh"},{"family":"Posada-Moreno","given":"Andrés"},{"family":"Liehner","given":"Luca"},{"family":"Schemmer","given":"Thomas"},{"family":"Vervier","given":"Luisa"},{"family":"Völker","given":"Marcus"},{"family":"Walderich","given":"Philipp"},{"family":"Zhang","given":"Song"},{"family":"Brecher","given":"Christian"},{"family":"Schmitt","given":"Robert"},{"family":"Decker","given":"Stefan"},{"family":"Gries","given":"Thomas"},{"family":"Häfner","given":"Constantin"},{"family":"Herty","given":"Michaël"},{"family":"Jarke","given":"Matthias"},{"family":"Kowalewski","given":"Stefan"},{"family":"Kuhlen","given":"Torsten"},{"family":"Schleifenbaum","given":"Johannes"},{"family":"Trimpe","given":"Sebastian"},{"family":"Aalst","given":"Wil"},{"family":"Ziefle","given":"Martina"},{"family":"Lakemeyer","given":"Gerhard"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1007/978-3-030-98062-7_4-2","URL":"https://doi.org/10.1007/978-3-030-98062-7_4-2","source":"openalex"},{"id":"oa:W4400158994","type":"article-journal","title":"Integrating routine blood biomarkers and artificial intelligence for supporting diagnosis of silicosis in engineered stone workers","abstract":"Abstract Engineered stone silicosis (ESS), primarily caused by inhaling respirable crystalline silica, poses a significant occupational health risk globally. ESS has no effective treatment and presents a rapid progression from simple silicosis (SS) to progressive massive fibrosis (PMF), with respiratory failure and death. Despite the use of diagnostic methods like chest x‐rays and high‐resolution computed tomography, early detection of silicosis remains challenging. Since routine blood tests have shown promise in detecting inflammatory markers associated with the disease, this study aims to assess whether routine blood biomarkers, coupled with machine learning techniques, can effectively differentiate between healthy individuals, subjects with SS, and PMF. To this end, 107 men diagnosed with silicosis, ex‐workers in the engineered stone (ES) sector, and 22 healthy male volunteers as controls not exposed to ES dust were recruited. Twenty‐one primary biochemical markers derived from peripheral blood extraction were obtained retrospectively from clinical hospital records. Relief‐F features selection technique was applied, and the resulting subset of 11 biomarkers was used to build five machine learning models, demonstrating high performance with sensitivities and specificities in the best case greater than 82% and 89%, respectively. The percentage of lymphocytes, the angiotensin‐converting enzyme, and lactate dehydrogenase indexes were revealed, among others, as blood biomarkers with significant cumulative importance for the machine learning models. Our study reveals that these biomarkers could detect a chronic inflammatory status and potentially serve as a supportive tool for the diagnosis, monitoring, and early detection of the progression of silicosis.","author":[{"family":"Morillo","given":"Daniel"},{"family":"Leónjiménez","given":"Antonio"},{"family":"Guerrerochanivet","given":"María"},{"family":"Jiménezgómez","given":"Gema"},{"family":"Hidalgomolina","given":"Antonio"},{"family":"Camposcaro","given":"Antonio"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/btm2.10694","URL":"https://doi.org/10.1002/btm2.10694","source":"openalex"},{"id":"oa:W4388030783","type":"article-journal","title":"Artificial intelligent based teaching and learning approaches: A comprehensive review","abstract":"The goal of this study is to investigate the potential effects that Artificial intelligence (AI) could have on education. The narrative and framework for investigating AI that emerged from the preliminary research served as the basis for the study’s emphasis, which was narrowed down to the use of AI and its effects on administration, instruction, and student learning. According to the findings, artificial intelligence has seen widespread adoption and use in education, particularly by educational institutions and in various contexts and applications. The development of AI began with computers and technologies related to computers; it then progressed to web-based and online intelligent education systems; and finally, it applied embedded computer systems in conjunction with other technologies, humanoid robots, and web-based chatbots to execute instructor tasks and functions either independently or in partnership with instructors. By utilizing these platforms, educators have been able to accomplish a variety of administrative tasks. In addition, because the systems rely on machine learning and flexibility, the curriculum and content have been modified to match the needs of students. This has led to improved learning outcomes in the form of higher uptake and retention rates.","author":[{"family":"Nguyen","given":"Thuong"},{"family":"Nguyễn","given":"Minh"},{"family":"Tran","given":"Hoang"}],"issued":{"date-parts":[[2023]]},"DOI":"10.11591/ijere.v12i4.26623","URL":"https://doi.org/10.11591/ijere.v12i4.26623","source":"openalex"},{"id":"oa:W4384823872","type":"article-journal","title":"Regulating Artificial Intelligence in the EU, United States and China - Implications for energy systems","abstract":"The growing prevalence and potential impact of artificial intelligence (AI) on society rises the need for regulation. In return, the shape of regulations will affect the application potential of AI across all economic sectors. This study compares the approaches to regulate AI in the European Union (EU), the United States (US) and China (CN). We then apply the findings of our comparative analysis on the energy sector, assessing the effects of each regulatory approach on the operation of a AI-based short-term electricity demand forecasting application. Our findings show that operationalizing AI applications will face very different challenges across geographies, with important implications for policy making and business development.","author":[{"family":"Heymann","given":"Fabian"},{"family":"Parginos","given":"Konstantinos"},{"family":"Hariri","given":"Ali"},{"family":"Franco","given":"Gabriele"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/isgteurope56780.2023.10407482","URL":"https://doi.org/10.1109/isgteurope56780.2023.10407482","source":"openalex"},{"id":"oa:W4405351251","type":"article-journal","title":"The 2024 revision of the Declaration of Helsinki: a modern ethical framework for medical research","abstract":"The Declaration of Helsinki, established in 1964, remains a foundational document in medical research ethics. This review examines the 2024 revision, endorsed by the 75th World Medical Association (WMA) Assembly, highlighting its impact on modern clinical research. Major updates include the shift from \"subjects\" to \"participants,\" promoting autonomy and active involvement, and the introduction of dual ethical review requirements for cross-border studies to strengthen accountability. New guidelines for data privacy address AI-related ethical concerns, while enhanced community engagement fosters transparency and shared decision-making. Additionally, standards for environmental sustainability encourage research practices that minimize ecological impacts. In response to global health crises such as COVID-19, the revised Declaration sets forth ethical protections to balance participant safety with research urgency during emergencies. Despite these advances, areas for improvement remain, especially in AI ethics, emergency research protocols, and the extension the Declaration's scope to include forensic and specimen research. The 2024 revision thus strengthens the Declaration's role as an adaptive, relevant framework for safeguarding participant rights and research integrity in a changing landscape.","author":[{"family":"Wen","given":"Boyuan"},{"family":"Zhang","given":"Guochao"},{"family":"Zhan","given":"Chang"},{"family":"Chen","given":"Chen"},{"family":"Yi","given":"Hang"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/postmj/qgae181","URL":"https://doi.org/10.1093/postmj/qgae181","source":"openalex"},{"id":"oa:W4405186881","type":"article-journal","title":"Translating ophthalmic medical jargon with artificial intelligence: a comparative comprehension study","abstract":"OBJECTIVE: Our goal was to evaluate the efficacy of OpenAI's ChatGPT-4.0 large language model (LLM) in translating technical ophthalmology terminology into more comprehensible language for allied health care professionals and compare it with other LLMs. DESIGN: Observational cross-sectional study. PARTICIPANTS: Five ophthalmologists each contributed three clinical encounter notes, totaling 15 reports for analysis. METHODS: Notes were translated into more comprehensible language using ChatGPT-4.0, ChatGPT-4o, Claude 3 Sonnet, and Google Gemini. Ten family physicians, masked to whether the note was original or translated by an LLM, independently evaluated both sets using Likert scales to assess comprehension and utility for clinical decision-making. Readability was evaluated using Flesch Reading Ease and Flesch-Kincaid Grade Level scores. Five ophthalmologist raters compared performance between LLMs and identified translation errors. RESULTS: LLM translations significantly outperformed the original notes in terms of comprehension (mean score of 4.7/5.0 vs 3.7/5.0; p < 0.001) and perceived usefulness (mean score of 4.6/5.0 vs 3.8/5.0; p < 0.005). Readability analysis demonstrated mildly increased linguistic complexity in the translated notes. ChatGPT-4.0 was preferred in 8 of 15 cases, ChatGPT-4o in 4, Gemini in 3, and Claude 3 Sonnet in 0 cases. All models exhibited some translation errors, but ChatGPT-4o and ChatGPT-4.0 had fewer inaccuracies. CONCLUSIONS: ChatGPT-4.0 can significantly enhance the comprehensibility of ophthalmic notes, facilitating better interprofessional communication and suggesting a promising role for LLMs in medical translation. However, the results also underscore the need for ongoing refinement and careful implementation of such technologies. Further research is needed to validate these findings across a broader range of specialties and languages.","author":[{"family":"Balas","given":"Michael"},{"family":"Kaplan","given":"Alexander"},{"family":"Esmail","given":"Kaisra"},{"family":"Saleh","given":"Solin"},{"family":"Sharma","given":"RC"},{"family":"Yan","given":"Peng"},{"family":"Arjmand","given":"Parnian"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.jcjo.2024.11.003","URL":"https://doi.org/10.1016/j.jcjo.2024.11.003","source":"openalex"},{"id":"oa:W4403750183","type":"article-journal","title":"Let's Have a Chat: How Well Does an Artificial Intelligence Chatbot Answer Clinical Infectious Diseases Pharmacotherapy Questions?","abstract":"Background: It is unknown whether ChatGPT provides quality responses to infectious diseases (ID) pharmacotherapy questions. This study surveyed ID pharmacist subject matter experts (SMEs) to assess the quality of ChatGPT version 3.5 (GPT-3.5) responses. Methods: The primary outcome was the percentage of GPT-3.5 responses considered useful by SME rating. Secondary outcomes were SMEs' ratings of correctness, completeness, and safety. Rating definitions were based on literature review. One hundred ID pharmacotherapy questions were entered into GPT-3.5 without custom instructions or additional prompts, and responses were recorded. A 0-10 rating scale for correctness, completeness, and safety was developed and validated for interrater reliability. Continuous and categorical variables were assessed for interrater reliability via average measures intraclass correlation coefficient and Fleiss multirater kappa, respectively. SMEs' responses were compared by the Kruskal-Wallis test and chi-square test for continuous and categorical variables. Results: SMEs considered 41.8% of responses useful. Median (IQR) ratings for correctness, completeness, and safety were 7 (4-9), 5 (3-8), and 8 (4-10), respectively. The Fleiss multirater kappa for usefulness was 0.379 (95% CI, .317-.441) indicating fair agreement, and intraclass correlation coefficients were 0.820 (95% CI, .758-.870), 0.745 (95% CI, .656-.816), and 0.833 (95% CI, .775-.880) for correctness, completeness, and safety, indicating at least substantial agreement. No significant difference was observed among SME responses for percentage of responses considered useful. Conclusions: Fewer than 50% of GPT-3.5 responses were considered useful by SMEs. Responses were mostly considered correct and safe but were often incomplete, suggesting that GPT-3.5 responses may not replace an ID pharmacist's responses.","author":[{"family":"Kufel","given":"Wesley"},{"family":"Hanrahan","given":"Kathleen"},{"family":"Seabury","given":"Robert"},{"family":"Parsels","given":"Katie"},{"family":"Gallagher","given":"Jason"},{"family":"Macdougall","given":"Conan"},{"family":"Covington","given":"Elizabeth"},{"family":"Chahine","given":"Elias"},{"family":"Britt","given":"Rachel"},{"family":"Steele","given":"Jeffrey"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/ofid/ofae641","URL":"https://doi.org/10.1093/ofid/ofae641","source":"openalex"},{"id":"oa:W4367556344","type":"article-journal","title":"Artificial Intelligence (AI) in Nursing Services: A Literature Review","abstract":"Research on artificial intelligence (AI) in health services is increased in recent decades, demonstrating great potential for improving the quality of care. However, the application of AI in nursing raises concerns regarding data bias and the potential impacts on patients. In addition, research on AI and the benefits in nursing is still limited. The aim of this study was to outline the results of research on AI technology in nursing services. The research method was literature review. Searches for relevant studies were carried out on several databases, such as PUBMED, Scopus, and Google Scholar by using keywords and terms related to nursing, artificial intelligence, and machine learning between 2012 and 2022. The inclusion criteria included research and development or AI-based technological validation used in nursing services and the research design, including experiment or observation with qualitative, quantitative, or mixed approached. Meanwhile, the exclusion criteria were articles which are irrelevant to nursing, non-experimental, non-observational, or literature review. The results showed total of 3,713 articles found only 10 articles fulfilled the criteria. The use of AI in nursing services can provide many benefits, but it requires careful consideration too. Therefore, more research is needed to address the challenges faced and to maximize the potential of AI in improving the quality of nursing services.","author":[{"family":"Kurniawan","given":"Moh"},{"family":"Handiyani","given":"Hanny"},{"family":"Nuraini","given":"Tuti"},{"family":"Hariyati","given":"Rr"}],"issued":{"date-parts":[[2023]]},"DOI":"10.33746/fhj.v10i01.556","URL":"https://doi.org/10.33746/fhj.v10i01.556","source":"openalex"},{"id":"oa:W4386889960","type":"article-journal","title":"Artificial Intelligence in Healthcare: Perception and Reality","abstract":"Artificial intelligence (AI) has birthed the new \"big thing\" in modern medicine. It promises to bring about safer and improved care that will be beneficial to patients and become a helpful tool in the hands of a skilled physician. Despite its anticipation, however, the implementation and usage of AI are still in their elementary phases, particularly due to legal and ethical considerations that border on \"data.\" These challenges should not be brushed aside but rather be recognized and resolved to enable acceptance by all relevant stakeholders without prejudice. Once these challenges can be overcome, AI will truly revolutionize the field of medicine with improved diagnostic accuracy, a reduction in physician burnout, and an enhanced treatment modality. It is therefore paramount that AI be embraced by physicians and integrated into medical education in order to be well-prepared for our role in the future of medicine.","author":[{"family":"Akinrinmade","given":"Abidemi"},{"family":"Adebile","given":"Temitayo"},{"family":"Ezuma-Ebong","given":"Chioma"},{"family":"Bolaji","given":"Kafayat"},{"family":"Ajufo","given":"Afomachukwu"},{"family":"Adigun","given":"Aisha"},{"family":"Mohammad","given":"Majed"},{"family":"Dike","given":"Juliet"},{"family":"Okobi","given":"Okelue"}],"issued":{"date-parts":[[2023]]},"DOI":"10.7759/cureus.45594","URL":"https://doi.org/10.7759/cureus.45594","source":"openalex"},{"id":"oa:W4382058325","type":"article-journal","title":"Recent Advances of Artificial Intelligence in Healthcare: A Systematic Literature Review","abstract":"The implementation of artificial intelligence (AI) is driving significant transformation inside the administrative and clinical workflows of healthcare organizations at an accelerated rate. This modification highlights the significant impact that AI has on a variety of tasks, especially in health procedures relating to early detection and diagnosis. Papers done in the past imply that AI has the potential to increase the overall quality of services provided in the healthcare industry. There have been reports that technology based on AI can improve the quality of human existence by making life simpler, safer, and more productive. A comprehensive analysis of previous scholarly research on the use of AI in the health area is provided in this research in the form of a literature review. In order to propose a classification framework, the review took into consideration 132 academic publications sourced from scholarly sources. The presentation covers both the benefits and the issues that AI capabilities provide for individuals, medical professionals, corporations, and the health industry. In addition, the social and ethical implications of AI are examined in the context of the output of value-added medical services for decision-making processes in healthcare, privacy and security measures for patient data, and health monitoring capabilities.","author":[{"family":"Kitsios","given":"Fotis"},{"family":"Kamariotou","given":"Maria"},{"family":"Syngelakis","given":"Aristomenis"},{"family":"Talias","given":"Μichael"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/app13137479","URL":"https://doi.org/10.3390/app13137479","source":"openalex"},{"id":"oa:W4403457107","type":"article-journal","title":"A survey of explainable artificial intelligence in healthcare: Concepts, applications, and challenges","abstract":"Explainable AI (XAI) has the potential to transform healthcare by making AI-driven medical decisions more transparent, reliable, and ethically compliant. Despite its promise, the healthcare sector faces several challenges, including the need to balance interpretability and accuracy, integrating XAI into clinical workflows, and ensuring adherence to rigorous regulatory standards. This paper provides a comprehensive review of XAI in healthcare, covering techniques, challenges, opportunities, and advancements, thereby enhancing the understanding and practical application of XAI in healthcare. The study also explores responsible AI in healthcare, discussing new perspectives and emerging trends, offering valuable insights for researchers and practitioners. The insights and recommendations presented aim to guide future research and policy-making, fostering the development of transparent, trustworthy, and effective AI-driven solutions.","author":[{"family":"Mienye","given":"Ibomoiye"},{"family":"Obaido","given":"George"},{"family":"Jere","given":"Nobert"},{"family":"Mienye","given":"Ebikella"},{"family":"Aruleba","given":"Kehinde"},{"family":"Emmanuel","given":"Ikiomoye"},{"family":"Ogbuokiri","given":"Blessing"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.imu.2024.101587","URL":"https://doi.org/10.1016/j.imu.2024.101587","source":"openalex"},{"id":"oa:W4403371652","type":"article-journal","title":"Artificial Intelligence Applications in Medical Mycology: Current and Future","abstract":"The application of artificial intelligence (AI) in the medical mycology field represents a new era in the diagnosis and management of fungal infections. AI technologies, particularly machine learning (ML) and deep learning (DL) methods, enhance diagnostic accuracy by leveraging large datasets and complex algorithms. This review examines current applications of AI in laboratory and clinical settings for fungal diagnostics. In the laboratory, AI models analyze microscopic images from potassium hydroxide (KOH) examinations, fungal culture tests, and histopathologic slides, which improves the detection rates of fungal pathogens significantly. In the clinical setting, AI assists the diagnosis of fungal infections using medical images, exhibiting high efficacy in binary classification tasks. However, challenges include small sample sizes, class imbalances, reliance on expert-labeled data, and the black box nature of AI models. Explainable AI offers potential solutions by providing human-comprehensible insights into AI decisionmaking processes. In addition, human-computer collaboration can enhance diagnostic accuracy, particularly for less experienced clinicians. The development of generative AI models, e.g., large language models and multimodal AI, promises to create extensive datasets and integrate various data sources for comprehensive diagnostics. Addressing these limitations through prospective clinical validation and continuous feedback will be essential for realizing the full potential of AI in medical mycology.","author":[{"family":"Kim","given":"Jemin"},{"family":"Boo","given":"Jihee"},{"family":"Park","given":"Chang"}],"issued":{"date-parts":[[2024]]},"DOI":"10.17966/jmi.2024.29.3.85","URL":"https://doi.org/10.17966/jmi.2024.29.3.85","source":"openalex"},{"id":"oa:W4386024862","type":"article-journal","title":"Radiographers’ Acceptance on the Integration of Artificial Intelligence into Medical Imaging Practice","abstract":"Artificial intelligence (AI) integration in medical imaging is a promising field for enhancing patient care, performance, and efficiency. Radiographers, on the other hand, are concerned about AI's acceptance and potential to replace them. This study assessed radiographers' acceptance of AI integration by considering their knowledge, attitudes, and job security. Based on demographic characteristics, there were no significant differences in knowledge, attitude, or job security level. Completing AI training, on the other hand, had a considerable influence. Overall, radiographers have a good level of knowledge and are enthusiastic about using AI tools into their regular activities.","author":[{"family":"Sharip","given":"Hairenanorashikin"},{"family":"Zakaria","given":"Wan"},{"family":"Leong","given":"Sook"},{"family":"Masoud","given":"Maida"},{"family":"Junaidi","given":"Mohamad"}],"issued":{"date-parts":[[2023]]},"DOI":"10.21834/e-bpj.v8i25.4872","URL":"https://doi.org/10.21834/e-bpj.v8i25.4872","source":"openalex"},{"id":"oa:W4391650202","type":"article-journal","title":"Artificial intelligence in future nursing care: Exploring perspectives of nursing professionals - A descriptive qualitative study","abstract":"Background: The healthcare landscape is rapidly evolving, with artificial intelligence (AI) emerging as a transformative force. In this context, understanding the viewpoints of nursing professionals regarding the integration of AI in future nursing care is crucial. Aims: This study aimed to provide insights into the perceptions of nursing professionals regarding the role of AI in shaping the future of healthcare. Methods: A cohort of 23 nursing professionals was recruited between April 7, 2023, and May 4, 2023, for this study. Employing a thematic analysis approach, qualitative data from interviews with nursing professionals were analyzed. Verbatim transcripts underwent rigorous coding, and these codes were organized into themes through constant comparative analysis. The themes were refined and developed through the grouping of related codes, ensuring an authentic representation of participants' viewpoints. Results: After careful data analysis, ten key themes emerged including: (I) Perceptions of AI readiness; (II) Benefits and concerns; (III) Enhanced patient outcomes; (IV) Collaboration and workflow; (V) Human-tech balance: (VI) Training and skill development; (VII) Ethical and legal considerations; (VIII) AI implementation barriers; (IX) Patient-nurse relationships; (X) Future vision and adaptation. Conclusion: This study provides valuable insights into nursing professionals' perspectives on the integration of AI in future nursing care. It highlights their enthusiasm for AI's potential benefits while emphasizing the importance of ethical and compassionate nursing practice. The findings underscore the need for comprehensive training programs to equip nursing professionals with the skills necessary for successful AI integration. Ultimately, this research contributes to the ongoing discourse on the role of AI in nursing, paving the way for a future where innovative technologies complement and enhance the delivery of patient-centered care.","author":[{"family":"Rony","given":"Moustaq"},{"family":"Kayesh","given":"Ibne"},{"family":"Bala","given":"Shuvashish"},{"family":"Akter","given":"Fazila"},{"family":"Parvin","given":"Mst"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.heliyon.2024.e25718","URL":"https://doi.org/10.1016/j.heliyon.2024.e25718","source":"openalex"},{"id":"oa:W4391820411","type":"article-journal","title":"Artificial Intelligence in Operating Room Management","abstract":"This systematic review examines the recent use of artificial intelligence, particularly machine learning, in the management of operating rooms. A total of 22 selected studies from February 2019 to September 2023 are analyzed. The review emphasizes the significant impact of AI on predicting surgical case durations, optimizing post-anesthesia care unit resource allocation, and detecting surgical case cancellations. Machine learning algorithms such as XGBoost, random forest, and neural networks have demonstrated their effectiveness in improving prediction accuracy and resource utilization. However, challenges such as data access and privacy concerns are acknowledged. The review highlights the evolving nature of artificial intelligence in perioperative medicine research and the need for continued innovation to harness artificial intelligence's transformative potential for healthcare administrators, practitioners, and patients. Ultimately, artificial intelligence integration in operative room management promises to enhance healthcare efficiency and patient outcomes.","author":[{"family":"Bellini","given":"Valentina"},{"family":"Russo","given":"Michele"},{"family":"Domenichetti","given":"Tania"},{"family":"Panizzi","given":"Matteo"},{"family":"Allai","given":"Simone"},{"family":"Bignami","given":"Elena"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s10916-024-02038-2","URL":"https://doi.org/10.1007/s10916-024-02038-2","source":"openalex"},{"id":"oa:W4400820534","type":"article-journal","title":"Performance of ChatGPT in Solving Questions From the Progress Test (Brazilian National Medical Exam): A Potential Artificial Intelligence Tool in Medical Practice","abstract":"Background The use of artificial intelligence (AI) is not a recent phenomenon, but the latest advancements in this technology are making a significant impact across various fields of human knowledge. In medicine, this trend is no different, although it has developed at a slower pace. ChatGPT is an example of an AI-based algorithm capable of answering questions, interpreting phrases, and synthesizing complex information, potentially aiding and even replacing humans in various areas of social interest. Some studies have compared its performance in solving medical knowledge exams with medical students and professionals to verify AI accuracy. This study aimed to measure the performance of ChatGPT in answering questions from the Progress Test from 2021 to 2023. Methodology An observational study was conducted in which questions from the 2021 Progress Test and the regional tests (Southern Institutional Pedagogical Support Center II) of 2022 and 2023 were presented to ChatGPT 3.5. The results obtained were compared with the scores of first- to sixth-year medical students from over 120 Brazilian universities. All questions were presented sequentially, without any modification to their structure. After each question was presented, the platform's history was cleared, and the site was restarted. Results The platform achieved an average accuracy rate in 2021, 2022, and 2023 of 69.7%, 68.3%, and 67.2%, respectively, surpassing students from all medical years in the three tests evaluated, reinforcing findings in the current literature. The subject with the best score for the AI was Public Health, with a mean grade of 77.8%. Conclusions ChatGPT demonstrated the ability to answer medical questions with higher accuracy than humans, including students from the last year of medical school.","author":[{"family":"Alessi","given":"Mateus"},{"family":"Gomes","given":"Heitor"},{"family":"Castro","given":"Matheus"},{"family":"Okamoto","given":"Cristina"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.64924","URL":"https://doi.org/10.7759/cureus.64924","source":"openalex"},{"id":"oa:W4379875293","type":"article-journal","title":"Smart Shoe Classification Using Artificial Intelligence on EfficientnetB3 Model","abstract":"In both athletics and daily life, shoes are crucial. Recently, smart sneakers have also taken on a certain trend among both sportsmen and the general public. Therefore, it has actually become crucial to classify shoes in a way that enables industry specialists to do so in accordance with customer demands keeping medical and health point of view. Each shoe product has four distinct perspectives, front, left, right, and rear, but only a few of these views have the product information that allows researchers to create innovative transfer learning models to understand its qualities with precision. Selection, prediction, and categorization of the specific perspective from the provided dataset from many viewpoints are all made possible via the development of machine learning models and the explanation of the artificial intelligence method. In order to create a sound and useful framework for shoe categorization, several efforts have been made over the past few decades. The suggested model in this research exhibits an accuracy of 88.8% on the Adam optimizer utilizing the EfficientNetB3 model and illustrates the accurate classification of the shoe class as utilized in the dataset.","author":[{"family":"Gill","given":"Kanwarpartap"},{"family":"Sharma","given":"Avinash"},{"family":"Anand","given":"Vatsala"},{"family":"Gupta","given":"Rupesh"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/incacct57535.2023.10141787","URL":"https://doi.org/10.1109/incacct57535.2023.10141787","source":"openalex"},{"id":"oa:W4327952037","type":"article-journal","title":"How to Bell the Cat? A Theoretical Review of Generative Artificial Intelligence towards Digital Disruption in All Walks of Life","abstract":"Generative Artificial Intelligence (GAI) has brought revolutionary changes to the world, enabling businesses to create new experiences by combining virtual and physical worlds. As the use of GAI grows along with the Metaverse, it is explored by academics, researchers, and industry communities for its endless possibilities. From ChatGPT by OpenAI to Bard AI by Google, GAI is a leading technology in physical and virtual business platforms. This paper focuses on GAI’s economic and societal impact and the challenges it poses. Businesses must rethink their operations and strategies to create hybrid physical and virtual experiences using GAI. This study proposes a framework that can help business managers develop effective strategies to enhance their operations. It analyzes the initial applications of GAI in multiple sectors to promote the development of future customer solutions and explores how GAI can help businesses create new value propositions and experiences for their customers, and the possibilities of digital communication and information technology. A research agenda is proposed for developing GAI for business management to enhance organizational efficiency. The results highlight a healthy conversation on the potential of GAI in various business sectors to improve customer experience.","author":[{"family":"Mondal","given":"Subhra"},{"family":"Das","given":"Subhankar"},{"family":"Vrana","given":"Vasiliki"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/technologies11020044","URL":"https://doi.org/10.3390/technologies11020044","source":"openalex"},{"id":"oa:W4398770695","type":"article-journal","title":"Generative artificial intelligence in academic medical writing","abstract":"Dear Editor, As dedicated readers of the Medical Journal of Babylon, we find it imperative to illuminate the captivating domain of generative artificial intelligence (GenAI) and its ever-growing impact on academic medical prose. As humanity progresses along the Kardashev scale, GenAI could stand as a milestone on the exponentially advancing path toward the much-debated realm of general AI and artificial superintelligence—a concept intertwined with the notion of technological singularity.[1] Text-based GenAI technology garnered remarkable attention across various fields and now impacts scholarly publications. By harnessing the capabilities of deep learning algorithms and neural networks, GenAI introduces innovative dimensions to academic medical writing.[2] Its diverse applications in assisting authors and biomedical researchers offer a range of advantages and disadvantages that warrant meticulous examination. In this discourse, we delve into the constraints and prospects inherent in text-based GenAI. Although GenAI models represent a relatively recent phenomenon, they produced unprecedented attention and fearful speculations among scholars, media, and the public; for example, concerning ChatGPT (based on GPT-3 model) and other models such as CTRL, GPT-4, GrammarlyGO, T5, and XLNet [Figure 1; Supplementary materials].[3,4] GenAI brings valuable advantages to biomedical research and academic writing. First, its speed and efficiency aid in drafting manuscripts, abstracts, and complex summaries, which expedites writing, allowing researchers to prioritize tasks. Second, these tools enhance language use, ensuring explicit, coherent content adhering to medical terminology, which is substantially advantageous, especially for non-native English speakers. Finally, GenAI transforms intricate data sets into accessible narratives, catering to policymakers and stakeholders, thus augmenting research impact.[2]Figure 1:: Features of different text-based GenAI models. GenAI: generative artificial intelligence; NLP: natural language processingIntegrating GenAI in academic writing brings forth some disadvantages. The foremost issue is the potential loss of authenticity, as it risks diluting the author’s distinct voice and perspective. Another downside is GenAI’s struggle to comprehend nuanced contextual understanding within biomedical research, resulting in inaccuracies, misinterpretations, or unsuitable recommendations. Ethical concerns also arise concerning proper attribution, plagiarism, and authorship transparency. GenAI might also offer misleading information with questionable bibliographies, known as “hallucinations,” thus intensifying bias.[2,5,6] Nevertheless, newer models (e.g., Scopus AI) promise to overcome such limitations.[6] Acknowledging the current limitations of GenAI is crucial. These AI models cannot often discern the subtleties of medical concepts, especially those rooted in deep domain expertise, including clinical experience. Additionally, although they can generate coherent sentences, their output might lack the critical analysis or logical reasoning that characterizes high-quality scholarly work.[3,5,6] Therefore, the application of GenAI should not be tempered without the understanding that it is a tool to support human creativity and insight rather than a replacement for the intellectual rigor that underpins academic excellence. Concerning the future potential, a careful approach is paramount. Authors and researchers must recognize GenAI as a tool to complement their expertise, not replace it. Rigorous revisions remain essential for producing high-quality academic content. The medical community in Iraq, aligned with the excellence and innovation principles worldwide, should cautiously embrace GenAI. Collaborative efforts among researchers, AI experts, and journal editors can pave the way for guidelines that ensure this technology’s responsible and effective use. Other regulations may emerge for GenAI’s involvement in mo","author":[{"family":"Al-Imam","given":"Ahmed"},{"family":"Al-Hadithi","given":"Nawfal"},{"family":"Alissa","given":"Faisel"},{"family":"Michalak","given":"Michał"}],"issued":{"date-parts":[[2023]]},"DOI":"10.4103/mjbl.mjbl_1176_23","URL":"https://doi.org/10.4103/mjbl.mjbl_1176_23","source":"openalex"},{"id":"oa:W4403426352","type":"article-journal","title":"Early Warning Scores With and Without Artificial Intelligence","abstract":"Importance: Early warning decision support tools to identify clinical deterioration in the hospital are widely used, but there is little information on their comparative performance. Objective: To compare 3 proprietary artificial intelligence (AI) early warning scores and 3 publicly available simple aggregated weighted scores. Design, Setting, and Participants: This retrospective cohort study was performed at 7 hospitals in the Yale New Haven Health System. All consecutive adult medical-surgical ward hospital encounters between March 9, 2019, and November 9, 2023, were included. Exposures: Simultaneous Epic Deterioration Index (EDI), Rothman Index (RI), eCARTv5 (eCART), Modified Early Warning Score (MEWS), National Early Warning Score (NEWS), and NEWS2 scores. Main Outcomes and Measures: Clinical deterioration, defined as a transfer from ward to intensive care unit or death within 24 hours of an observation. Results: Of the 362 926 patient encounters (median patient age, 64 [IQR, 47-77] years; 200 642 [55.3%] female), 16 693 (4.6%) experienced a clinical deterioration event. eCART had the highest area under the receiver operating characteristic curve at 0.895 (95% CI, 0.891-0.900), followed by NEWS2 at 0.831 (95% CI, 0.826-0.836), NEWS at 0.829 (95% CI, 0.824-0.835), RI at 0.828 (95% CI, 0.823-0.834), EDI at 0.808 (95% CI, 0.802-0.812), and MEWS at 0.757 (95% CI, 0.750-0.764). After matching scores at the moderate-risk sensitivity level for a NEWS score of 5, overall positive predictive values (PPVs) ranged from a low of 6.3% (95% CI, 6.1%-6.4%) for an EDI score of 41 to a high of 17.3% (95% CI, 16.9%-17.8%) for an eCART score of 94. Matching scores at the high-risk specificity of a NEWS score of 7 yielded overall PPVs ranging from a low of 14.5% (95% CI, 14.0%-15.2%) for an EDI score of 54 to a high of 23.3% (95% CI, 22.7%-24.2%) for an eCART score of 97. The moderate-risk thresholds provided a median of at least 20 hours of lead time for all the scores. Median lead time at the high-risk threshold was 11 (IQR, 0-69) hours for eCART, 8 (IQR, 0-63) hours for NEWS, 6 (IQR, 0-62) hours for NEWS2, 5 (IQR, 0-56) hours for MEWS, 1 (IQR, 0-39) hour for EDI, and 0 (IQR, 0-42) hours for RI. Conclusions and Relevance: In this cohort study of inpatient encounters, eCART outperformed the other AI and non-AI scores, identifying more deteriorating patients with fewer false alarms and sufficient time to intervene. NEWS, a non-AI, publicly available early warning score, significantly outperformed EDI. Given the wide variation in accuracy, additional transparency and oversight of early warning tools may be warranted.","author":[{"family":"Edelson","given":"Dana"},{"family":"Churpek","given":"Matthew"},{"family":"Carey","given":"Kyle"},{"family":"Lin","given":"Zhenqiu"},{"family":"Huang","given":"Chenxi"},{"family":"Siner","given":"Jonathan"},{"family":"Johnson","given":"Jennifer"},{"family":"Krumholz","given":"Harlan"},{"family":"Rhodes","given":"Deborah"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1001/jamanetworkopen.2024.38986","URL":"https://doi.org/10.1001/jamanetworkopen.2024.38986","source":"openalex"},{"id":"oa:W4379796020","type":"article-journal","title":"Perception of Pathologists in Poland of Artificial Intelligence and Machine Learning in Medical Diagnosis—A Cross-Sectional Study","abstract":"Background: In the past vicennium, several artificial intelligence (AI) and machine learning (ML) models have been developed to assist in medical diagnosis, decision making, and design of treatment protocols. The number of active pathologists in Poland is low, prolonging tumor patients’ diagnosis and treatment journey. Hence, applying AI and ML may aid in this process. Therefore, our study aims to investigate the knowledge of using AI and ML methods in the clinical field in pathologists in Poland. To our knowledge, no similar study has been conducted. Methods: We conducted a cross-sectional study targeting pathologists in Poland from June to July 2022. The questionnaire included self-reported information on AI or ML knowledge, experience, specialization, personal thoughts, and level of agreement with different aspects of AI and ML in medical diagnosis. Data were analyzed using IBM® SPSS® Statistics v.26, PQStat Software v.1.8.2.238, and RStudio Build 351. Results: Overall, 68 pathologists in Poland participated in our study. Their average age and years of experience were 38.92 ± 8.88 and 12.78 ± 9.48 years, respectively. Approximately 42% used AI or ML methods, which showed a significant difference in the knowledge gap between those who never used it (OR = 17.9, 95% CI = 3.57–89.79, p < 0.001). Additionally, users of AI had higher odds of reporting satisfaction with the speed of AI in the medical diagnosis process (OR = 4.66, 95% CI = 1.05–20.78, p = 0.043). Finally, significant differences (p = 0.003) were observed in determining the liability for legal issues used by AI and ML methods. Conclusion: Most pathologists in this study did not use AI or ML models, highlighting the importance of increasing awareness and educational programs regarding applying AI and ML in medical diagnosis.","author":[{"family":"Ahmed","given":"Alhassan"},{"family":"Brychcy","given":"Agnieszka"},{"family":"Abouzid","given":"Mohamed"},{"family":"Witt","given":"Martin"},{"family":"Kaczmarek","given":"Elżbieta"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/jpm13060962","URL":"https://doi.org/10.3390/jpm13060962","source":"openalex"},{"id":"oa:W4327919044","type":"article-journal","title":"Explainable AI in medical imaging: An overview for clinical practitioners – Beyond saliency-based XAI approaches","abstract":"Driven by recent advances in Artificial Intelligence (AI) and Computer Vision (CV), the implementation of AI systems in the medical domain increased correspondingly. This is especially true for the domain of medical imaging, in which the incorporation of AI aids several imaging-based tasks such as classification, segmentation, and registration. Moreover, AI reshapes medical research and contributes to the development of personalized clinical care. Consequently, alongside its extended implementation arises the need for an extensive understanding of AI systems and their inner workings, potentials, and limitations which the field of eXplainable AI (XAI) aims at. Because medical imaging is mainly associated with visual tasks, most explainability approaches incorporate saliency-based XAI methods. In contrast to that, in this article we would like to investigate the full potential of XAI methods in the field of medical imaging by specifically focusing on XAI techniques not relying on saliency, and providing diversified examples. We dedicate our investigation to a broad audience, but particularly healthcare professionals. Moreover, this work aims at establishing a common ground for cross-disciplinary understanding and exchange across disciplines between Deep Learning (DL) builders and healthcare professionals, which is why we aimed for a non-technical overview. Presented XAI methods are divided by a method's output representation into the following categories: Case-based explanations, textual explanations, and auxiliary explanations.","author":[{"family":"Borys","given":"Katarzyna"},{"family":"Schmitt","given":"Yasmin"},{"family":"Nauta","given":"Meike"},{"family":"Seifert","given":"Christin"},{"family":"Krämer","given":"Nicole"},{"family":"Friedrich","given":"Christoph"},{"family":"Nensa","given":"Felix"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.ejrad.2023.110786","URL":"https://doi.org/10.1016/j.ejrad.2023.110786","source":"openalex"},{"id":"oa:W4324046738","type":"article-journal","title":"Application of artificial intelligence for resilient and sustainable healthcare system: systematic literature review and future research directions","abstract":"Recent years have witnessed increased pressure across the global healthcare system during the COVID-19 pandemic. The COVID-19 pandemic shattered existing healthcare operations and taught us the importance of a resilient and sustainable healthcare system. Digitisation, specifically adoption of Artificial Intelligence (AI) has positively contributed to developing a resilient healthcare system in recent past. To understand how AI contributes to building a resilient and sustainable healthcare system, this study based on systematic literature review of 89 articles extracted from Scopus and Web of Science databases is conducted. The study is organised around several key themes such as applications, benefits, and challenges of using AI technology in healthcare sector. It is observed that AI has wide applications in radiology, surgery, medical, research, and development of healthcare sector. Based on the analysis, a research framework is proposed using an extended Antecedents, Practices, and Outcomes (APO) framework. This framework comprises AI applications’ antecedents, practices, and outcomes for building a resilient and sustainable healthcare system. Consequently, three propositions are drawn in this study. Furthermore, our study has adopted the theory, context and methodology (TCM) framework to provide future research directions, which can be used as a reference point for future studies.","author":[{"family":"Vishwakarma","given":"Laxmi"},{"family":"Singh","given":"Rajesh"},{"family":"Mishra","given":"Ruchi"},{"family":"Kumari","given":"Archana"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1080/00207543.2023.2188101","URL":"https://doi.org/10.1080/00207543.2023.2188101","source":"openalex"},{"id":"oa:W4313582012","type":"article-journal","title":"Review on the Evaluation and Development of Artificial Intelligence for COVID-19 Containment","abstract":"Artificial intelligence has significantly enhanced the research paradigm and spectrum with a substantiated promise of continuous applicability in the real world domain. Artificial intelligence, the driving force of the current technological revolution, has been used in many frontiers, including education, security, gaming, finance, robotics, autonomous systems, entertainment, and most importantly the healthcare sector. With the rise of the COVID-19 pandemic, several prediction and detection methods using artificial intelligence have been employed to understand, forecast, handle, and curtail the ensuing threats. In this study, the most recent related publications, methodologies and medical reports were investigated with the purpose of studying artificial intelligence's role in the pandemic. This study presents a comprehensive review of artificial intelligence with specific attention to machine learning, deep learning, image processing, object detection, image segmentation, and few-shot learning studies that were utilized in several tasks related to COVID-19. In particular, genetic analysis, medical image analysis, clinical data analysis, sound analysis, biomedical data classification, socio-demographic data analysis, anomaly detection, health monitoring, personal protective equipment (PPE) observation, social control, and COVID-19 patients' mortality risk approaches were used in this study to forecast the threatening factors of COVID-19. This study demonstrates that artificial-intelligence-based algorithms integrated into Internet of Things wearable devices were quite effective and efficient in COVID-19 detection and forecasting insights which were actionable through wide usage. The results produced by the study prove that artificial intelligence is a promising arena of research that can be applied for disease prognosis, disease forecasting, drug discovery, and to the development of the healthcare sector on a global scale. We prove that artificial intelligence indeed played a significantly important role in helping to fight against COVID-19, and the insightful knowledge provided here could be extremely beneficial for practitioners and research experts in the healthcare domain to implement the artificial-intelligence-based systems in curbing the next pandemic or healthcare disaster.","author":[{"family":"Hasan","given":"Md"},{"family":"Islam","given":"Muhammad"},{"family":"Sadeq","given":"Muhammad"},{"family":"Fung","given":"Wai"},{"family":"Uddin","given":"Jasim"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/s23010527","URL":"https://doi.org/10.3390/s23010527","source":"openalex"},{"id":"oa:W4388406898","type":"article-journal","title":"Artificial Intelligence in the Military: An Overview of the Capabilities, Applications, and Challenges","abstract":"Artificial intelligence (AI) has become a reality in today’s world with the rise of the 4th industrial revolution, especially in the armed forces. Military AI systems can process more data more effectively than traditional systems. Due to its intrinsic computing and decision‐making capabilities, AI also increases combat systems’ self‐control, self‐regulation, and self‐actuation. Artificial intelligence is used in almost every military application, and increased research and development support from military research agencies to develop new and advanced AI technologies is expected to drive the widespread demand for AI‐driven systems in the military. This essay will discuss several AI applications in the military, as well as their capabilities, opportunities, and potential harm and devastation when there is instability. The article looks at current and future potential for developing artificial intelligence algorithms, particularly in military applications. Most of the discussion focused on the seven patterns of AI, the usage and implementation of AI algorithms in the military, object detection, military logistics, and robots, the global instability induced by AI use, and nuclear risk. The article also looks at the current and future potential for developing artificial intelligence algorithms, particularly in military applications.","author":[{"family":"Rashid","given":"Adib"},{"family":"Kausik","given":"Ashfakul"},{"family":"Sunny","given":"Ahamed"},{"family":"Bappy","given":"Mehedy"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1155/2023/8676366","URL":"https://doi.org/10.1155/2023/8676366","source":"openalex"},{"id":"oa:W4401727214","type":"article-journal","title":"Generative Artificial Intelligence in Education: Advancing Adaptive and Personalized Learning","abstract":"The integration of generative artificial intelligence (AI) into adaptive and personalized learning represents a transformative shift in the educational landscape.This research paper investigates the impact of incorporating generative AI into adaptive and personalized learning environments, with a focus on tracing the evolution from conventional artificial intelligence methods to generative AI and identifying its diverse applications in education.The study begins with a comprehensive review of the evolution of generative AI models and frameworks.A framework of selection criteria is established to curate case studies showcasing the applications of generative AI in education.These case studies are analysed to elucidate the benefits and challenges associated with integrating generative AI into adaptive learning frameworks.Through an in-depth analysis of selected case studies, the study reveals tangible benefits derived from generative AI integration, including increased student engagement, improved test scores and accelerated skill development.Ethical, technical and pedagogical challenges related to generative AI integration are identified, emphasizing the need for careful consideration and collaborative efforts between educators and technologists.The findings underscore the transformative potential of generative AI in revolutionizing education.By addressing ethical concerns, navigating technical challenges and embracing human-centric approaches, educators and technologists can collaboratively harness the power of generative AI to create innovative and inclusive learning environments.Additionally, the study highlights the transition from Education 4.0 to Education 5.0, emphasizing the importance of social-emotional learning and human connection alongside personalization in shaping the future of education.","author":[{"family":"Guettala","given":"Manel"},{"family":"Bourekkache","given":"Samir"},{"family":"Kazar","given":"Okba"},{"family":"Harous","given":"Saad"}],"issued":{"date-parts":[[2024]]},"DOI":"10.18267/j.aip.235","URL":"https://doi.org/10.18267/j.aip.235","source":"openalex"},{"id":"oa:W4402564190","type":"article-journal","title":"Generative artificial intelligence and ethical considerations in health care: a scoping review and ethics checklist","abstract":"The widespread use of Chat Generative Pre-trained Transformer (known as ChatGPT) and other emerging technology that is powered by generative artificial intelligence (GenAI) has drawn attention to the potential ethical issues they can cause, especially in high-stakes applications such as health care, but ethical discussions have not yet been translated into operationalisable solutions. Furthermore, ongoing ethical discussions often neglect other types of GenAI that have been used to synthesise data (eg, images) for research and practical purposes, which resolve some ethical issues and expose others. We did a scoping review of the ethical discussions on GenAI in health care to comprehensively analyse gaps in the research. To reduce the gaps, we have developed a checklist for comprehensive assessment and evaluation of ethical discussions in GenAI research. The checklist can be integrated into peer review and publication systems to enhance GenAI research and might be useful for ethics-related disclosures for GenAI-powered products and health-care applications of such products and beyond.","author":[{"family":"Ning","given":"Yilin"},{"family":"Teixayavong","given":"Salinelat"},{"family":"Shang","given":"Yuqing"},{"family":"Savulescu","given":"Julian"},{"family":"Nagaraj","given":"Vaishaanth"},{"family":"Miao","given":"Di"},{"family":"Mertens","given":"Mayli"},{"family":"Ting","given":"Daniel"},{"family":"Ong","given":"Jasmine"},{"family":"Liu","given":"Mingxuan"},{"family":"Cao","given":"Jiuwen"},{"family":"Dunn","given":"Michael"},{"family":"Vaughan","given":"Roger"},{"family":"Ong","given":"Marcus"},{"family":"Sung","given":"Joseph"},{"family":"Topol","given":"Eric"},{"family":"Liu","given":"Nan"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/s2589-7500(24)00143-2","URL":"https://doi.org/10.1016/s2589-7500(24)00143-2","source":"openalex"},{"id":"oa:W4403245162","type":"article-journal","title":"Using artificial intelligence to document the hidden RNA virosphere","abstract":"Current metagenomic tools can fail to identify highly divergent RNA viruses. We developed a deep learning algorithm, termed LucaProt, to discover highly divergent RNA-dependent RNA polymerase (RdRP) sequences in 10,487 metatranscriptomes generated from diverse global ecosystems. LucaProt integrates both sequence and predicted structural information, enabling the accurate detection of RdRP sequences. Using this approach, we identified 161,979 potential RNA virus species and 180 RNA virus supergroups, including many previously poorly studied groups, as well as RNA virus genomes of exceptional length (up to 47,250 nucleotides) and genomic complexity. A subset of these novel RNA viruses was confirmed by RT-PCR and RNA/DNA sequencing. Newly discovered RNA viruses were present in diverse environments, including air, hot springs, and hydrothermal vents, with virus diversity and abundance varying substantially among ecosystems. This study advances virus discovery, highlights the scale of the virosphere, and provides computational tools to better document the global RNA virome.","author":[{"family":"Hou","given":"Xin"},{"family":"He","given":"Yong"},{"family":"Fang","given":"Pan"},{"family":"Mei","given":"Shi"},{"family":"Xu","given":"Zan"},{"family":"Wu","given":"Wei"},{"family":"Tian","given":"Jun"},{"family":"Zhang","given":"Shun"},{"family":"Zeng","given":"Zhenyu"},{"family":"Gou","given":"Qinyu"},{"family":"Xin","given":"Gen"},{"family":"Le","given":"Shi"},{"family":"Xia","given":"Yinyue"},{"family":"Zhou","given":"Yu"},{"family":"Hui","given":"Fengming"},{"family":"Pan","given":"Yuanfei"},{"family":"Eden","given":"John‐sebastian"},{"family":"Yang","given":"Zhaohui"},{"family":"Han","given":"Chong"},{"family":"Shu","given":"Yuelong"},{"family":"Guo","given":"Deyin"},{"family":"Li","given":"Jun"},{"family":"Holmes","given":"Edward"},{"family":"Li","given":"Zhao‐rong"},{"family":"Shī","given":"Mǎng"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.cell.2024.09.027","URL":"https://doi.org/10.1016/j.cell.2024.09.027","source":"openalex"},{"id":"oa:W3156012754","type":"article-journal","title":"Artificial Intelligence and Big Data in Sustainable Entrepreneurship","abstract":"Abstract There is an urgent need to transition our economy, society, and culture towards systems and actions that facilitate ecological sustainability. Such radical change requires equally radical transformation of approaches to decision making and resource use. Sustainable entrepreneurship (SE) is often presented as the answer to meeting the triple‐bottom‐line challenges that businesses face; however, there are very real limits to what it can achieve. SE is in the early stages of adopting tools at the technological frontier that offer empirical guidance at every point of an entrepreneurial decision‐making process. Big Data (BD) advances the potential for artificial intelligence (AI) to inform decision making, while also charting pathways to achieve desired outcomes. So far, the interactions between AI, BD, and SE have been generally under‐studied. In this primarily conceptual paper, we address the lack of work consolidating and synthesizing these literatures. We suggest that AI and BD readily contribute to further sustainable development of the weak form, but that it also holds great promise for achieving the strong sustainability ideal. We offer two propositions regarding how the integration of AI and BD can inform/support SE. We conclude by mapping out potential avenues for future research.","author":[{"family":"Bickley","given":"Steve"},{"family":"Macintyre","given":"Alison"},{"family":"Torgler","given":"Benno"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/joes.12611","URL":"https://doi.org/10.1111/joes.12611","source":"openalex"},{"id":"oa:W4385562476","type":"article-journal","title":"Artificial intelligence and human behavioral development: A perspective on new skills and competences acquisition for the educational context","abstract":"Despite the significant emphasis placed on incorporating 21st century skills into the educational framework, particularly at the primary level, recent scholarly works indicate considerable variation in the implementation of these skills across different countries and regions, suggesting a demand for further research specifically focusing on primary education. The indications of the Digicomp framework 1 and 21st-century skills in Europe have outlined the key competences for lifelong learning needed for all citizens, including teachers and students. In this perspective, Education plays a fundamental role in ensuring that citizens acquire the required skills. The objective in the common European framework is clear: to initiate a transition from the culture of knowledge to the culture of competence. Nowadays, technological advancement allows the researchers to create and combine different frameworks with the perspective of an even more tailored, and engaged education, some examples derived from the implementation of Virtual Reality (VR) and Augmented Reality (AR), in the combination of Gamification and AI, or the development of Intelligent Tutoring Systems (ITS) to foster and create an even more personalized learning and teaching. Following these premises, in this paper, we want to point out new research reflections and perspectives that could help researchers, teachers, educators (and consequently students) to reflect on the introduction of new technologies (e.g., artificial intelligence, robot tutors) and on how these can affect on human behavioral development and on the acquisition of new skills and competences (Specifically: Creativity, Critical Thinking, Problem Solving, and Computational Thinking) for the educational context. The analysis carried on, suggests a perspective on how creativity, critical thinking, and problem-solving can be effective in promoting computational thinking, and how Artificial Intelligence (AI) could be an aid instrument to teachers in the fostering of creativity, critical thinking, and problem-solving in schools and educational contexts.","author":[{"family":"Benvenuti","given":"Martina"},{"family":"Cangelosi","given":"Angelo"},{"family":"Weinberger","given":"Armin"},{"family":"Mazzoni","given":"Elvis"},{"family":"Benassi","given":"Mariagrazia"},{"family":"Barbaresi","given":"Mattia"},{"family":"Orsoni","given":"Matteo"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.chb.2023.107903","URL":"https://doi.org/10.1016/j.chb.2023.107903","source":"openalex"},{"id":"oa:W4362716434","type":"article-journal","title":"Overview of Early ChatGPT’s Presence in Medical Literature: Insights From a Hybrid Literature Review by ChatGPT and Human Experts","abstract":"ChatGPT, an artificial intelligence chatbot, has rapidly gained prominence in various domains, including medical education and healthcare literature. This hybrid narrative review, conducted collaboratively by human authors and ChatGPT, aims to summarize and synthesize the current knowledge of ChatGPT in the indexed medical literature during its initial four months. A search strategy was employed in PubMed and EuropePMC databases, yielding 65 and 110 papers, respectively. These papers focused on ChatGPT's impact on medical education, scientific research, medical writing, ethical considerations, diagnostic decision-making, automation potential, and criticisms. The findings indicate a growing body of literature on ChatGPT's applications and implications in healthcare, highlighting the need for further research to assess its effectiveness and ethical concerns.","author":[{"family":"Temsah","given":"Omar"},{"family":"Khan","given":"Samina"},{"family":"Chaiah","given":"Yazan"},{"family":"Senjab","given":"Abdulrahman"},{"family":"Alhasan","given":"Khalid"},{"family":"Jamal","given":"Amr"},{"family":"Aljamaan","given":"Fadi"},{"family":"Malki","given":"Khalid"},{"family":"Halwani","given":"Rabih"},{"family":"Altawfiq","given":"Jaffar"},{"family":"Temsah","given":"Mohamad‐hani"},{"family":"Aleyadhy","given":"Ayman"}],"issued":{"date-parts":[[2023]]},"DOI":"10.7759/cureus.37281","URL":"https://doi.org/10.7759/cureus.37281","source":"openalex"},{"id":"oa:W4320891340","type":"manuscript","title":"Acceptance of Medical Artificial Intelligence in Skin Cancer Screening: Choice-Based Conjoint Survey (Preprint)","abstract":"BACKGROUND There is great interest in using artificial intelligence (AI) to screen for skin cancer. This is fueled by a rising incidence of skin cancer and an increasing scarcity of trained dermatologists. AI systems capable of identifying melanoma could save lives, enable immediate access to screenings, and reduce unnecessary care and health care costs. While such AI-based systems are useful from a public health perspective, past research has shown that individual patients are very hesitant about being examined by an AI system. OBJECTIVE The aim of this study was two-fold: (1) to determine the relative importance of the provider (in-person physician, physician via teledermatology, AI, personalized AI), costs of screening (free, 10€, 25€, 40€; 1€=US $1.09), and waiting time (immediate, 1 day, 1 week, 4 weeks) as attributes contributing to patients’ choices of a particular mode of skin cancer screening; and (2) to investigate whether sociodemographic characteristics, especially age, were systematically related to participants’ individual choices. METHODS A choice-based conjoint analysis was used to examine the acceptance of medical AI for a skin cancer screening from the patient’s perspective. Participants responded to 12 choice sets, each containing three screening variants, where each variant was described through the attributes of provider, costs, and waiting time. Furthermore, the impacts of sociodemographic characteristics (age, gender, income, job status, and educational background) on the choices were assessed. RESULTS Among the 383 clicks on the survey link, a total of 126 (32.9%) respondents completed the online survey. The conjoint analysis showed that the three attributes had more or less equal importance in contributing to the participants’ choices, with provider being the most important attribute. Inspecting the individual part-worths of conjoint attributes showed that treatment by a physician was the most preferred modality, followed by electronic consultation with a physician and personalized AI; the lowest scores were found for the three AI levels. Concerning the relationship between sociodemographic characteristics and relative importance, only age showed a significant positive association to the importance of the attribute provider (r=0.21, P=.02), in which younger participants put less importance on the provider than older participants. All other correlations were not significant. CONCLUSIONS This study adds to the growing body of research using choice-based experiments to investigate the acceptance of AI in health contexts. Future studies are needed to explore the reasons why AI is accepted or rejected and whether sociodemographic characteristics are associated with this decision.","author":[{"family":"Jagemann","given":"Inga"},{"family":"Wensing","given":"Ole"},{"family":"Stegemann","given":"Manuel"},{"family":"Hirschfeld","given":"Gerrit"}],"issued":{"date-parts":[[2023]]},"DOI":"10.2196/preprints.46402","URL":"https://doi.org/10.2196/preprints.46402","source":"openalex"},{"id":"oa:W4396615793","type":"article-journal","title":"Bio‐Inspired Sensory Receptors for Artificial‐Intelligence Perception","abstract":"In the era of artificial intelligence (AI), there is a growing interest in replicating human sensory perception. Selective and sensitive bio-inspired sensory receptors with synaptic plasticity have recently gained significant attention in developing energy-efficient AI perception. Various bio-inspired sensory receptors and their applications in AI perception are reviewed here. The critical challenges for the future development of bio-inspired sensory receptors are outlined, emphasizing the need for innovative solutions to overcome hurdles in sensor design, integration, and scalability. AI perception can revolutionize various fields, including human-machine interaction, autonomous systems, medical diagnostics, environmental monitoring, industrial optimization, and assistive technologies. As advancements in bio-inspired sensing continue to accelerate, the promise of creating more intelligent and adaptive AI systems becomes increasingly attainable, marking a significant step forward in the evolution of human-like sensory perception.","author":[{"family":"Bag","given":"Atanu"},{"family":"Ghosh","given":"Gargi"},{"family":"Sultan","given":"MJ"},{"family":"Chouhdry","given":"Hamna"},{"family":"Hong","given":"Seok"},{"family":"Trung","given":"Tran"},{"family":"Kang","given":"Geun‐young"},{"family":"Lee","given":"Nae‐eung"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/adma.202403150","URL":"https://doi.org/10.1002/adma.202403150","source":"openalex"},{"id":"oa:W4391265101","type":"article-journal","title":"Fuzzy inference system with interpretable fuzzy rules: Advancing explainable artificial intelligence for disease diagnosis—A comprehensive review","abstract":"Interpretable artificial intelligence (AI), also known as explainable AI, is indispensable in establishing trustable AI for bench-to-bedside translation, with substantial implications for human well-being. However, the majority of existing research in this area has centered on designing complex and sophisticated methods, regardless of their interpretability. Consequently, the main prerequisite for implementing trustworthy AI in medical domains has not been met. Scientists have developed various explanation methods for interpretable AI. Among these methods, fuzzy rules embedded in a fuzzy inference system (FIS) have emerged as a novel and powerful tool to bridge the communication gap between humans and advanced AI machines. However, there have been few reviews of the use of FISs in medical diagnosis. In addition, the application of fuzzy rules to different kinds of multimodal medical data has received insufficient attention, despite the potential use of fuzzy rules in designing appropriate methodologies for available datasets. This review provides a fundamental understanding of interpretability and fuzzy rules, conducts comparative analyses of the use of fuzzy rules and other explanation methods in handling three major types of multimodal data (i.e., sequence signals, medical images, and tabular data), and offers insights into appropriate fuzzy rule application scenarios and recommendations for future research.","author":[{"family":"Cao","given":"Jin"},{"family":"Zhou","given":"Ta"},{"family":"Zhi","given":"Shaohua"},{"family":"Lam","given":"Saikit"},{"family":"Ren","given":"Ge"},{"family":"Zhang","given":"Yuanpeng"},{"family":"Wang","given":"Yongqiang"},{"family":"Dong","given":"Yanjing"},{"family":"Cai","given":"Jing"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.ins.2024.120212","URL":"https://doi.org/10.1016/j.ins.2024.120212","source":"openalex"},{"id":"oa:W4367311208","type":"article-journal","title":"How Chatbots and Large Language Model Artificial Intelligence Systems Will Reshape Modern Medicine","abstract":"In an era of clinicians being burned out by electronic medical records and documentation burdens, we might all dream of having a personal scribe to draft progress notes, translate patient instructions, summarize the literature, complete insurance authorization paperwork, and respond to unending in-basket messages, as described in the Perspective in this issue of JAMA Internal Medicine. 1 This would have sounded like a fantasy just a few years ago, but the release of rapidly developing chatbots now demonstrates the potential of large language model artificial intelligence (AI) systems with surprisingly adept language manipulation and knowledge processing capabilities.The underlying foundation model technology rides atop the peak of inflated expectations, 2 reflecting a disruptive technology likely to change the way we work and live, even as we must be aware of substantial limitations.Good or bad, ready or not, Pandora's box has already been opened.One such large language model, ChatGPT, is the fastest-growing internet application in history with more than 100 million users.3 This has shifted access to sophisticated AI capabilities away from concentrated pockets of technical experts to the masses, where all types of otherwise unimaginable (and unintended) use cases are being discovered.To ensure that the adoption of such tools into health care practice is done effectively and responsibly, physicians must lean in to understand and drive this conversation.Large language models represent the underlying class of machine learning models trained in autocomplete tasks.Given the words \"coronary artery,\" these models may predict the next word to be \"disease,\" \"bypass graft,\" or \"calcification\" based on statistical parameters learned from prior training data text on how often those words appear together.","author":[{"family":"Li","given":"Ron"},{"family":"Kumar","given":"Andre"},{"family":"Chen","given":"Jonathan"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1001/jamainternmed.2023.1835","URL":"https://doi.org/10.1001/jamainternmed.2023.1835","source":"openalex"},{"id":"oa:W4381743186","type":"article-journal","title":"Artificial Intelligence in Ophthalmology: A Comparative Analysis of GPT-3.5, GPT-4, and Human Expertise in Answering StatPearls Questions","abstract":"Importance Chat Generative Pre-Trained Transformer (ChatGPT) has shown promising performance in various fields, including medicine, business, and law, but its accuracy in specialty-specific medical questions, particularly in ophthalmology, is still uncertain. Purpose This study evaluates the performance of two ChatGPT models (GPT-3.5 and GPT-4) and human professionals in answering ophthalmology questions from the StatPearls question bank, assessing their outcomes, and providing insights into the integration of artificial intelligence (AI) technology in ophthalmology. Methods ChatGPT's performance was evaluated using 467 ophthalmology questions from the StatPearls question bank. These questions were stratified into 11 subcategories, four difficulty levels, and three generalized anatomical categories. The answer accuracy of GPT-3.5, GPT-4, and human participants was assessed. Statistical analysis was conducted via the Kolmogorov-Smirnov test for normality, one-way analysis of variance (ANOVA) for the statistical significance of GPT-3 versus GPT-4 versus human performance, and repeated unpaired two-sample t-tests to compare the means of two groups. Results GPT-4 outperformed both GPT-3.5 and human professionals on ophthalmology StatPearls questions, except in the \"Lens and Cataract\" category. The performance differences were statistically significant overall, with GPT-4 achieving higher accuracy (73.2%) compared to GPT-3.5 (55.5%, p-value < 0.001) and humans (58.3%, p-value < 0.001). There were variations in performance across difficulty levels (rated one to four), but GPT-4 consistently performed better than both GPT-3.5 and humans on level-two, -three, and -four questions. On questions of level-four difficulty, human performance significantly exceeded that of GPT-3.5 (p = 0.008). Conclusion The study's findings demonstrate GPT-4's significant performance improvements over GPT-3.5 and human professionals on StatPearls ophthalmology questions. Our results highlight the potential of advanced conversational AI systems to be utilized as important tools in the education and practice of medicine.","author":[{"family":"Moshirfar","given":"Majid"},{"family":"Altaf","given":"Amal"},{"family":"Stoakes","given":"Isabella"},{"family":"Tuttle","given":"Jared"},{"family":"Hoopes","given":"Phillip"}],"issued":{"date-parts":[[2023]]},"DOI":"10.7759/cureus.40822","URL":"https://doi.org/10.7759/cureus.40822","source":"openalex"},{"id":"oa:W4402298814","type":"article-journal","title":"Embedding Internal Accountability Into Health Care Institutions for Safe, Effective, and Ethical Implementation of Artificial Intelligence Into Medical Practice: A Mayo Clinic Case Study","abstract":"Health care organizations are building, deploying, and self-governing digital health technologies (DHTs), including artificial intelligence, at an increasing rate. This scope necessitates expertise and quality infrastructure to ensure that the technology impacting patient care is safe, effective, and ethical throughout its lifecycle. The objective of this article is to describe Mayo Clinic's approach for embedding internal accountability as a case study for other health care institutions seeking modalities for responsible implementation of artificial intelligence-enabled DHTs. Mayo Clinic aims to enable and empower innovators by (1) building internal skills and expertise, (2) establishing a centralized review board, and (3) aligning development and deployment processes with regulations, standards, and best practices. In 2022, Mayo Clinic established the Software as a Medical Device Review Board (The Board), an independent body of physicians and domain experts to represent the organization in providing innovators regulatory and risk mitigation recommendations for DHTs. Hundreds of digital health product teams have since benefited from this function, intended to enable responsible innovation in alignment with regulation and state-of-the-art quality management practices. Other health care institutions can adopt similar internal accountability bodies using this framework. Opportunity remains to iterate on Mayo Clinic's approach in alignment with advancing best practices and enhance representation on The Board as part of standard continuous improvement practices.","author":[{"family":"Loufek","given":"Brenna"},{"family":"Vidal","given":"David"},{"family":"Mcclintock","given":"David"},{"family":"Lifson","given":"Mark"},{"family":"Williamson","given":"Eric"},{"family":"Overgaard","given":"Shauna"},{"family":"Mcnaughton","given":"Kathleen"},{"family":"Lipford","given":"Melissa"},{"family":"Pardi","given":"Darrell"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.mcpdig.2024.08.008","URL":"https://doi.org/10.1016/j.mcpdig.2024.08.008","source":"openalex"},{"id":"oa:W4324311092","type":"article-journal","title":"Potential and Pitfalls of ChatGPT and Natural-Language Artificial Intelligence Models for Diabetes Education","abstract":"Diabetes self-management and education (DSME) is an integral part of diabetes care and has been shown to improve glycemic control, reduce complications, and increase quality of life (1).The traditional model in which clinicians and diabetes educators share responsibility for patient education faces challenges such as reduced access to care during the pandemic and a shortage of trained educators.Artificial intelligence (AI) solutions are increasingly recognized to have a strong use case in DSME (2).Smart conversational agents (\"chatbots\") have shown potential as tools for direct patient engagement and education (3).While previous generations of chatbots delivered structured output based on preset queries and responses, modern natural-language AI models are designed to accept unstructured or nonstandardized inputs and provide human-like responses.These models draw on a large repository of humangenerated textual content to produce responses statistically likely to match the query.While chatbots may be able to augment patient care by providing ondemand answers to patient questions, they are based on language patterns rather than objective databases and may provide patients with authoritativesounding information that is inaccurate.ChatGPT is a chatbot developed by OpenAI based on the GPT3 large language model.It is readily accessible by the general public and has gained popular traction.It has recently been shown to be able to pass the U.S.","author":[{"family":"Sng","given":"Gerald"},{"family":"Tung","given":"Joshua"},{"family":"Lim","given":"Daniel"},{"family":"Bee","given":"Yong"}],"issued":{"date-parts":[[2023]]},"DOI":"10.2337/dc23-0197","URL":"https://doi.org/10.2337/dc23-0197","source":"openalex"},{"id":"oa:W4323667995","type":"article-journal","title":"Artificial intelligence-based traffic flow prediction: a comprehensive review","abstract":"Abstract The expansion of the Internet of Things has resulted in new creative solutions, such as smart cities, that have made our lives more productive, convenient, and intelligent. The core of smart cities is the Intelligent Transportation System (ITS) which has been integrated into several smart city applications that improve transportation and mobility. ITS aims to resolve many traffic issues, such as traffic congestion issues. Recently, new traffic flow prediction models and frameworks have been rapidly developed in tandem with the introduction of artificial intelligence approaches to improve the accuracy of traffic flow prediction. Traffic forecasting is a crucial duty in the transportation industry. It can significantly affect the design of road constructions and projects in addition to its importance for route planning and traffic rules. Furthermore, traffic congestion is a critical issue in urban areas and overcrowded cities. Therefore, it must be accurately evaluated and forecasted. Hence, a reliable and efficient method for predicting traffic is essential. The main objectives of this study are: First, present a comprehensive review of the most popular machine learning and deep learning techniques applied in traffic prediction. Second, identifying inherent obstacles to applying machine learning and deep learning in the domain of traffic prediction.","author":[{"family":"Ahmed","given":"Sayed"},{"family":"Abdelhamid","given":"Yasser"},{"family":"Hefny","given":"Hesham"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1186/s43067-023-00081-6","URL":"https://doi.org/10.1186/s43067-023-00081-6","source":"openalex"},{"id":"oa:W4391196614","type":"article-journal","title":"Postoperative accurate pain assessment of children and artificial intelligence: A medical hypothesis and planned study","abstract":"Although the pediatric perioperative pain management has been improved in recent years, the valid and reliable pain assessment tool in perioperative period of children remains a challenging task. Pediatric perioperative pain management is intractable not only because children cannot express their emotions accurately and objectively due to their inability to describe physiological characteristics of feeling which are different from those of adults, but also because there is a lack of effective and specific assessment tool for children. In addition, exposure to repeated painful stimuli early in life is known to have short and long-term adverse sequelae. The short-term sequelae can induce a series of neurological, endocrine, cardiovascular system stress related to psychological trauma, while long-term sequelae may alter brain maturation process, which can lead to impair neurodevelopmental, behavioral, and cognitive function. Children's facial expressions largely reflect the degree of pain, which has led to the developing of a number of pain scoring tools that will help improve the quality of pain management in children if they are continually studied in depth. The artificial intelligence (AI) technology represented by machine learning has reached an unprecedented level in image processing of deep facial models through deep convolutional neural networks, which can effectively identify and systematically analyze various subtle features of children's facial expressions. Based on the construction of a large database of images of facial expressions in children with perioperative pain, this study proposes to develop and apply automatic facial pain expression recognition software using AI technology. The study aims to improve the postoperative pain management for pediatric population and the short-term and long-term quality of life for pediatric patients after operational event.","author":[{"family":"Yue","given":"Jian"},{"family":"Wang","given":"Qi"},{"family":"Liu","given":"Bin"},{"family":"Zhou","given":"Leng"}],"issued":{"date-parts":[[2024]]},"DOI":"10.12998/wjcc.v12.i4.681","URL":"https://doi.org/10.12998/wjcc.v12.i4.681","source":"openalex"},{"id":"oa:W4317727238","type":"article-journal","title":"A Survey on Optimization Techniques for Edge Artificial Intelligence (AI)","abstract":"Artificial Intelligence (Al) models are being produced and used to solve a variety of current and future business and technical problems. Therefore, AI model engineering processes, platforms, and products are acquiring special significance across industry verticals. For achieving deeper automation, the number of data features being used while generating highly promising and productive AI models is numerous, and hence the resulting AI models are bulky. Such heavyweight models consume a lot of computation, storage, networking, and energy resources. On the other side, increasingly, AI models are being deployed in IoT devices to ensure real-time knowledge discovery and dissemination. Real-time insights are of paramount importance in producing and releasing real-time and intelligent services and applications. Thus, edge intelligence through on-device data processing has laid down a stimulating foundation for real-time intelligent enterprises and environments. With these emerging requirements, the focus turned towards unearthing competent and cognitive techniques for maximally compressing huge AI models without sacrificing AI model performance. Therefore, AI researchers have come up with a number of powerful optimization techniques and tools to optimize AI models. This paper is to dig deep and describe all kinds of model optimization at different levels and layers. Having learned the optimization methods, this work has highlighted the importance of having an enabling AI model optimization framework.","author":[{"family":"Surianarayanan","given":"Chellammal"},{"family":"Lawrence","given":"John"},{"family":"Chelliah","given":"Pethuru"},{"family":"Prakash","given":"Edmond"},{"family":"Hewage","given":"Chaminda"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/s23031279","URL":"https://doi.org/10.3390/s23031279","source":"openalex"},{"id":"oa:W4401974153","type":"article-journal","title":"Bridging the Artificial Intelligence (AI) Divide: Do Postgraduate Medical Students Outshine Undergraduate Medical Students in AI Readiness?","abstract":"INTRODUCTION: As artificial intelligence (AI) transforms healthcare, medical education must adapt to equip future physicians with the necessary competencies. However, little is known about the differences in AI knowledge, attitudes, and practices between undergraduate and postgraduate medical students. This study aims to assess and compare AI knowledge, attitudes, and practices among undergraduate and postgraduate medical students, and to explore the associated factors and qualitative themes. METHODS: A mixed-methods study was conducted, involving 605 medical students (404 undergraduates, 201 postgraduates) from a tertiary care center. Participants completed a survey assessing AI knowledge, attitudes, and practices. Semi-structured interviews and focus group discussions were conducted to explore qualitative themes. Quantitative data were analyzed using descriptive statistics, t-tests, chi-square tests, and regression analyses. Qualitative data underwent thematic analysis. RESULTS: Postgraduate students demonstrated significantly higher AI knowledge scores than undergraduates (38.9±4.9 vs. 29.6±6.8, p<0.001). Both groups held positive attitudes, but postgraduates showed greater confidence in AI's potential (p<0.001). Postgraduates reported more extensive AI-related practices (p<0.001). Key qualitative themes included excitement about AI's potential, concerns about job security, and the need for AI education. AI knowledge, attitudes, and practices were positively correlated (p<0.01). CONCLUSIONS: This study reveals a significant AI knowledge gap between undergraduate and postgraduate medical students, highlighting the need for targeted AI education. The findings can inform curriculum development and policies to prepare medical students for the AI-driven future of healthcare. Further research should explore the long-term impact of AI education on clinical practice.","author":[{"family":"Gandhi","given":"Rohankumar"},{"family":"Parmar","given":"Alpesh"},{"family":"Kagathara","given":"Jimmy"},{"family":"Lakkad","given":"Dhruv"},{"family":"Kakadiya","given":"Jay"},{"family":"Murugan","given":"Yogesh"}],"issued":{"date-parts":[[2024]]},"DOI":"10.7759/cureus.67288","URL":"https://doi.org/10.7759/cureus.67288","source":"openalex"},{"id":"oa:W4387857082","type":"article-journal","title":"Artificial intelligence technology in Alzheimer's disease research","abstract":"Alzheimer's disease is a neurocognitive disorder and one of the contributing factors to dementia. According to the World Health Organization, this disease has a sig-nificant impact on the global population's health, with the number of affected individuals steadily increasing each year. Amidst rapid technological development, the use of artificial intelligence has significantly expanded into the field of medical diagnostics, encompassing areas such as the analysis of medical images, drug development, design of personalized treatment plans, and disease prediction and treatment. Deep learning, which is an important branch in the field of artificial intelligence, is playing a key role in solving several medical challenges by providing important technical support for the early detection, diagnosis, and treatment of Alzheimer's disease. Given this context, this review aims to explore the differences between conventional methods and artificial intelligence techniques in Alzheimer's disease research. Additionally, it aims to summarize current non-invasive and portable techniques for detection of Alzheimer's disease, offering support and guidance for the future prediction and management of the disease.","author":[{"family":"Zhang","given":"Wenli"},{"family":"Li","given":"Yifan"},{"family":"Ren","given":"Wentao"},{"family":"Liu","given":"Bo"}],"issued":{"date-parts":[[2023]]},"DOI":"10.5582/irdr.2023.01091","URL":"https://doi.org/10.5582/irdr.2023.01091","source":"openalex"},{"id":"oa:W4391541001","type":"article-journal","title":"Artificial intelligence powered Metaverse: analysis, challenges and future perspectives","abstract":"Abstract The Metaverse, a virtual reality (VR) space where users can interact with each other and digital objects, is rapidly becoming a reality. As this new world evolves, Artificial Intelligence (AI) is playing an increasingly important role in shaping its development. Integrating AI with emerging technologies in the Metaverse creates new possibilities for immersive experiences that were previously impossible. This paper explores how AI is integrated with technologies such as the Internet of Things, blockchain, Natural Language Processing, virtual reality, Augmented Reality, Mixed Reality, and Extended Reality. One potential benefit of using AI in the Metaverse is the ability to create personalized experiences for individual users, based on their behavior and preferences. Another potential benefit of using AI in the Metaverse is the ability to automate repetitive tasks, freeing up time and resources for more complex and creative endeavors. However, there are also challenges associated with using AI in the Metaverse, such as ensuring user privacy and addressing issues of bias and discrimination. By examining the potential benefits and challenges of using AI in the Metaverse, including ethical considerations, we can better prepare for this exciting new era of VR. This paper presents a comprehensive survey of AI and its integration with other emerging technologies in the Metaverse, as the Metaverse continues to evolve and grow, it will be important for developers and researchers to stay up to date with the latest developments in AI and emerging technologies to fully leverage their potential.","author":[{"family":"Soliman","given":"Mona"},{"family":"Ahmed","given":"Eman"},{"family":"Darwish","given":"Ashraf"},{"family":"Hassanien","given":"Aboul"},{"family":"Soliman","given":"Mona"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s10462-023-10641-x","URL":"https://doi.org/10.1007/s10462-023-10641-x","source":"openalex"},{"id":"oa:W4385553785","type":"article-journal","title":"Identifying fake conclusions of forensic medical examinations using an artificial intelligence technology based on the experience in the Republic of Kazakhstan: a Review","abstract":"This review discusses the legal aspects on the use of mathematical statistics and machine learning (hereinafter referred to as artificial intelligence) in forensic activities to identify both expert errors and fake expert opinions. An attempt has been made to establish the criteria for evaluating conclusions of forensic examinations by determining their relevance, admissibility, reliability, and objectivity, as well as the objective possibility of distinguishing expert errors from deliberately false and fake expert opinions. A SWOT analysis on the use of artificial intelligence was carried out to solve the issue of its application in the field under consideration, which revealed its advantages, and disadvantages. The use of mathematical statistics and machine learning methods is not a universal method to identify fakes in expert opinions. However, given that this method can give both false-positive and false-negative results, its outcomes should be verified by independent experts. In addition, to effectively prevent the facts of falsification, comprehensive measures should be taken, including not only the detection of manipulations but also the prevention of the possibility of their occurrence, as well as the punishment of the perpetrators. Thus, this review proposed several amendments and additions to the current legislation of the Republic of Kazakhstan.","author":[{"family":"Voyevodkin","given":"Denis"},{"family":"Rustemova","given":"Gauhar"},{"family":"Begaliyev","given":"Yernar"},{"family":"Igembayev","given":"Kussain"},{"family":"Ayupova","given":"Zauresh"}],"issued":{"date-parts":[[2023]]},"DOI":"10.17816/fm8270","URL":"https://doi.org/10.17816/fm8270","source":"openalex"},{"id":"oa:W4390571745","type":"article-journal","title":"Artificial intelligence: revolutionizing cardiology with large language models","abstract":"Natural language processing techniques are having an increasing impact on clinical care from patient, clinician, administrator, and research perspective. Among others are automated generation of clinical notes and discharge letters, medical term coding for billing, medical chatbots both for patients and clinicians, data enrichment in the identification of disease symptoms or diagnosis, cohort selection for clinical trial, and auditing purposes. In the review, an overview of the history in natural language processing techniques developed with brief technical background is presented. Subsequently, the review will discuss implementation strategies of natural language processing tools, thereby specifically focusing on large language models, and conclude with future opportunities in the application of such techniques in the field of cardiology.","author":[{"family":"Boonstra","given":"Machteld"},{"family":"Weissenbacher","given":"Davy"},{"family":"Moore","given":"Jason"},{"family":"Gonzalezhernandez","given":"Graciela"},{"family":"Asselbergs","given":"Folkert"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1093/eurheartj/ehad838","URL":"https://doi.org/10.1093/eurheartj/ehad838","source":"openalex"},{"id":"oa:W4402966535","type":"article-journal","title":"The impact of artificial intelligence on women’s empowerment, and work-life balance in Saudi educational institutions","abstract":"Gender prejudice and stereotypes are prevalent in the workplace, particularly for women in the Artificial Intelligence (AI) industry, where they can significantly hinder professional development and limit prospects for growth. These challenges contribute to the underrepresentation of executives in AI. However, with the right measures, these barriers can be overcome, leading to a more inclusive and diverse AI industry. Women in this demanding technological domain often face additional difficulties in achieving a work-life balance, further constraining their professional advancement and engagement in the industry. This research aims to examine the implications of AI capabilities on work-life balance and the empowerment of female faculty members in enhancing the efficiency of educational institutions. The research performs a structural equation modeling (SEM) approach, using a survey conducted on female faculty of Saudi Arabian universities. The study specifically considers moderating variables such as age, education level, experience, and marital status. The findings, which reveal that AI managerial capability, as well as AI infrastructure agility, impacts work-life balance and empowerment of women faculties in educational institution efficiency, underscore the significance of considering demographic factors when analyzing women's empowerment and work-life balance as outcomes. By exploring these factors, the research provides a comprehensive understanding of how AI capabilities impact women's empowerment and their ability to maintain a work-life balance, ultimately contributing to the efficiency and effectiveness of educational institutions. These results emphasize the value of increasing women's empowerment and raising the standard of performance evaluation systems in educational sectors.","author":[{"family":"Meharunisa","given":"Sayeda"},{"family":"Almugren","given":"Hawazen"},{"family":"Sarabdeen","given":"Masahina"},{"family":"Mabrouk","given":"Fatma"},{"family":"Kijas","given":"ACM"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fpsyg.2024.1432541","URL":"https://doi.org/10.3389/fpsyg.2024.1432541","source":"openalex"},{"id":"oa:W4392055393","type":"article-journal","title":"Artificial intelligence algorithms for predicting post-operative ileus after laparoscopic surgery","abstract":"Objective: By constructing a predictive model using machine learning and deep learning technologies, we aim to understand the risk factors for postoperative intestinal obstruction in laparoscopic colorectal cancer patients, and establish an effective artificial intelligence-based predictive model to guide individualized prevention and treatment, thus improving patient outcomes. Methods: We constructed a model of the artificial intelligence algorithm in Python. Subjects were randomly assigned to either a training set for variable identification and model construction, or a test set for testing model performance, at a ratio of 7:3. The model was trained with ten algorithms. We used the AUC values of the ROC curves, as well as accuracy, precision, recall rate and F1 scores. Results: The results of feature engineering composited with the GBDT algorithm showed that opioid use, anesthesia duration, and body weight were the top three factors in the development of POI. We used ten machine learning and deep learning algorithms to validate the model, and the results were as follows: the three algorithms with best accuracy were XGB (0.807), Decision Tree (0.807) and Neural DecisionTree (0.807); the two algorithms with best precision were XGB (0.500) and Decision Tree (0.500); the two algorithms with best recall rate were adab (0.243) and Decision Tree (0.135); the two algorithms with highest F1 score were adab (0.290) and Decision Tree (0.213); and the three algorithms with best AUC were Gradient Boosting (0.678), XGB (0.638) and LinearSVC (0.633). Conclusion: This study shows that XGB and Decision Tree are the two best algorithms for predicting the risk of developing ileus after laparoscopic colon cancer surgery. It provides new insight and approaches to the field of postoperative intestinal obstruction in colorectal cancer through the application of machine learning techniques, thereby improving our understanding of the disease and offering strong support for clinical decision-making.","author":[{"family":"Zhou","given":"Cheng"},{"family":"Li","given":"Huijuan"},{"family":"Xue","given":"Qiong"},{"family":"Yang","given":"Jianjun"},{"family":"Zhu","given":"Yu"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.heliyon.2024.e26580","URL":"https://doi.org/10.1016/j.heliyon.2024.e26580","source":"openalex"},{"id":"oa:W4400118784","type":"article-journal","title":"Artificial Intelligence-Driven Facial Image Analysis for the Early Detection of Rare Diseases: Legal, Ethical, Forensic, and Cybersecurity Considerations","abstract":"This narrative review explores the potential, complexities, and consequences of using artificial intelligence (AI) to screen large government-held facial image databases for the early detection of rare genetic diseases. Government-held facial image databases, combined with the power of artificial intelligence, offer the potential to revolutionize the early diagnosis of rare genetic diseases. AI-powered phenotyping, as exemplified by the Face2Gene app, enables highly accurate genetic assessments from simple photographs. This and similar breakthrough technologies raise significant privacy and ethical concerns about potential government overreach augmented with the power of AI. This paper explores the concept, methods, and legal complexities of AI-based phenotyping within the EU. It highlights the transformative potential of such tools for public health while emphasizing the critical need to balance innovation with the protection of individual privacy and ethical boundaries. This comprehensive overview underscores the urgent need to develop robust safeguards around individual rights while responsibly utilizing AI’s potential for improved healthcare outcomes, including within a forensic context. Furthermore, the intersection of AI and sensitive genetic data necessitates proactive cybersecurity measures. Current and future developments must focus on securing AI models against attacks, ensuring data integrity, and safeguarding the privacy of individuals within this technological landscape.","author":[{"family":"Kováč","given":"Peter"},{"family":"Jackuliak","given":"Peter"},{"family":"Bražinová","given":"Alexandra"},{"family":"Varga","given":"Ivan"},{"family":"Aláč","given":"Michal"},{"family":"Smatana","given":"Martin"},{"family":"Lovich","given":"Dušan"},{"family":"Thurzo","given":"Andrej"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/ai5030049","URL":"https://doi.org/10.3390/ai5030049","source":"openalex"},{"id":"oa:W4401328425","type":"article-journal","title":"Navigating artificial intelligence in care homes: Competing stakeholder views of trust and logics of care","abstract":"The COVID-19 pandemic shed light on systemic issues plaguing care (nursing) homes, from staff shortages to substandard healthcare. Artificial Intelligence (AI) technologies, including robots and chatbots, have been proposed as solutions to such issues. Yet, socio-ethical concerns about the implications of AI for health and care practices have also been growing among researchers and practitioners. At a time of AI promise and concern, it is critical to understand how those who develop and implement these technologies perceive their use and impact in care homes. Combining a sociological approach to trust with Annemarie Mol's logic of care and Jeanette Pol's concept of fitting, we draw on 18 semi-structured interviews with care staff, advocates, and AI developers to explore notions of human-AI care. Our findings show positive perceptions and experiences of AI in care homes, but also ambivalence. While integrative care incorporating humans and technology was salient across interviewees, we also identified experiential, contextual, and knowledge divides between AI developers and care staff. For example, developers lacked experiential knowledge of care homes' daily functioning and constraints, influencing how they designed AI. Care staff demonstrated limited experiential knowledge of AI or more critical views about contexts of use, affecting their trust in these technologies. Different understandings of 'good care' were evident, too: 'warm' care was sometimes linked to human care and 'cold' care to technology. In conclusion, understandings and experiences of AI are marked by different logics of sociotechnical care and related levels of trust in these sensitive settings.","author":[{"family":"Neves","given":"Bárbara"},{"family":"Omori","given":"Maho"},{"family":"Petersen","given":"Alan"},{"family":"Vered","given":"Mor"},{"family":"Carter","given":"Adrian"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.socscimed.2024.117187","URL":"https://doi.org/10.1016/j.socscimed.2024.117187","source":"openalex"},{"id":"oa:W4388973748","type":"article-journal","title":"Transformers in medical image segmentation: a narrative review","abstract":"Background and Objective: Transformers, which have been widely recognized as state-of-the-art tools in natural language processing (NLP), have also come to be recognized for their value in computer vision tasks. With this increasing popularity, they have also been extensively researched in the more complex medical imaging domain. The associated developments have resulted in transformers being on par with sought-after convolution neural networks, particularly for medical image segmentation. Methods combining both types of networks have proven to be especially successful in capturing local and global contexts, thereby significantly boosting their performances in various segmentation problems. Motivated by this success, we have attempted to survey the consequential research focused on innovative transformer networks, specifically those designed to cater to medical image segmentation in an efficient manner. Methods: Databases like Google Scholar, arxiv, ResearchGate, Microsoft Academic, and Semantic Scholar have been utilized to find recent developments in this field. Specifically, research in the English language from 2021 to 2023 was considered. Key Content and Findings: In this survey, we look into the different types of architectures and attention mechanisms that uniquely improve performance and the structures that are in place to handle complex medical data. Through this survey, we summarize the popular and unconventional transformer-based research as seen through different key angles and analyze quantitatively the strategies that have proven more advanced. Conclusions: We have also attempted to discern existing gaps and challenges within current research, notably highlighting the deficiency of annotated medical data for precise deep learning model training. Furthermore, potential future directions for enhancing transformers' utility in healthcare are outlined, encompassing strategies such as transfer learning and exploiting foundation models for specialized medical image segmentation.","author":[{"family":"Khan","given":"Rabeea"},{"family":"Lee","given":"Byoung"},{"family":"Lee","given":"Mu"}],"issued":{"date-parts":[[2023]]},"DOI":"10.21037/qims-23-542","URL":"https://doi.org/10.21037/qims-23-542","source":"openalex"},{"id":"oa:W4392392203","type":"article-journal","title":"Artificial Intelligence in the Diagnosis and Management of Appendicitis in Pediatric Departments: A Systematic Review","abstract":"INTRODUCTION: Artificial intelligence (AI) is a growing field in medical research that could potentially help in the challenging diagnosis of acute appendicitis (AA) in children. However, usefulness of AI in clinical settings remains unclear. Our aim was to assess the accuracy of AIs in the diagnosis of AA in the pediatric population through a systematic literature review. METHODS: PubMed, Embase, and Web of Science were searched using the following keywords: \"pediatric,\" \"artificial intelligence,\" \"standard practices,\" and \"appendicitis,\" up to September 2023. The risk of bias was assessed using PROBAST. RESULTS: A total of 302 articles were identified and nine articles were included in the final review. Two studies had prospective validation, seven were retrospective, and no randomized control trials were found. All studies developed their own algorithms and had an accuracy greater than 90% or area under the curve >0.9. All studies were rated as a \"high risk\" concerning their overall risk of bias. CONCLUSION: We analyzed the current status of AI in the diagnosis of appendicitis in children. The application of AI shows promising potential, but the need for more rigor in study design, reporting, and transparency is urgent to facilitate its clinical implementation.","author":[{"family":"Rey","given":"Robin"},{"family":"Gualtieri","given":"Renato"},{"family":"Scala","given":"Giorgio"},{"family":"Posfaybarbe","given":"Klara"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1055/a-2257-5122","URL":"https://doi.org/10.1055/a-2257-5122","source":"openalex"},{"id":"oa:W4400695814","type":"article-journal","title":"Management of drug supply chain information based on “artificial intelligence + vendor managed inventory” in China: perspective based on a case study","abstract":"Objectives: To employ a drug supply chain information system to optimize drug management practices, reducing costs and improving efficiency in financial and asset management. Methods: A digital artificial intelligence + vendor managed inventory (AI+VMI)-based system for drug supply chain information management in hospitals has been established. The system enables digitalization and intelligentization of purchasing plans, reconciliations, and consumption settlements while generating purchase, sales, inventory reports as well as various query reports. The indicators for evaluating the effectiveness before and after project implementation encompass drug loss reporting, inventory discrepancies, inter-hospital medication retrieval frequency, drug expenditure, and cloud pharmacy service utilization. Results: The successful implementation of this system has reduced the hospital inventory rate to approximately 20% and decreased the average annual inventory error rate from 0.425‰ to 0.025‰, significantly boosting drug supply chain efficiency by 42.4%. It has also minimized errors in drug application, allocation, and distribution while increasing adverse reaction reports. Drug management across multiple hospital districts has been standardized, leading to improved access to medicines and enhanced patient satisfaction. Conclusion: The AI+VMI system improves drug supply chain management by ensuring security, reducing costs, enhancing efficiency and safety of drug management, and elevating the professional competence and service level of pharmaceutical personnel.","author":[{"family":"Shen","given":"Jianwen"},{"family":"Bu","given":"Fengjiao"},{"family":"Ye","given":"Zhengqiang"},{"family":"Zhang","given":"Min"},{"family":"Ma","given":"Qin"},{"family":"Yan","given":"Jingchao"},{"family":"Huang","given":"Taomin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fphar.2024.1373642","URL":"https://doi.org/10.3389/fphar.2024.1373642","source":"openalex"},{"id":"oa:W4403238905","type":"article-journal","title":"Perceptions and attitudes toward artificial intelligence among frontline physicians and physicians’ assistants in Kansas: a cross-sectional survey","abstract":"Abstract Objective This survey aims to understand frontline healthcare professionals’ perceptions of artificial intelligence (AI) in healthcare and assess how AI familiarity influences these perceptions. Materials and Methods We conducted a survey from February to March 2023 of physicians and physician assistants registered with the Kansas State Board of Healing Arts. Participants rated their perceptions toward AI-related domains and constructs on a 5-point Likert scale, with higher scores indicating stronger agreement. Two sub-groups were created for analysis to assess the impact of participants’ familiarity and experience with AI on the survey results. Results From 532 respondents, key concerns were Perceived Communication Barriers (median = 4.0, IQR = 2.8-4.8), Unregulated Standards (median = 4.0, IQR = 3.6-4.8), and Liability Issues (median = 4.0, IQR = 3.5-4.8). Lower levels of agreement were noted for Trust in AI Mechanisms (median = 3.0, IQR = 2.2-3.4), Perceived Risks of AI (median = 3.2, IQR = 2.6-4.0), and Privacy Concerns (median = 3.3, IQR = 2.3-4.0). Positive correlations existed between Intention to use AI and Perceived Benefits (r = 0.825) and Trust in AI Mechanisms (r = 0.777). Perceived risk negatively correlated with Intention to Use AI (r = −0.718). There was no difference in perceptions between AI experienced and AI naïve subgroups. Discussion The findings suggest that perceptions of benefits, trust, risks, communication barriers, regulation, and liability issues influence healthcare professionals’ intention to use AI, regardless of their AI familiarity. Conclusion The study highlights key factors affecting AI adoption in healthcare from the frontline healthcare professionals’ perspective. These insights can guide strategies for successful AI implementation in healthcare.","author":[{"family":"Dean","given":"Tanner"},{"family":"Seecheran","given":"Rajeev"},{"family":"Badgett","given":"Robert"},{"family":"Zackula","given":"Rosey"},{"family":"Symons","given":"John"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/jamiaopen/ooae100","URL":"https://doi.org/10.1093/jamiaopen/ooae100","source":"openalex"},{"id":"oa:W4402581887","type":"article-journal","title":"Transparent RFID tag wall enabled by artificial intelligence for assisted living","abstract":"Abstract Current approaches to activity-assisted living (AAL) are complex, expensive, and intrusive, which reduces their practicality and end user acceptance. However, emerging technologies such as artificial intelligence and wireless communications offer new opportunities to enhance AAL systems. These improvements could potentially lower healthcare costs and reduce hospitalisations by enabling more effective identification, monitoring, and localisation of hazardous activities, ensuring rapid response to emergencies. In response to these challenges, this paper introduces theTransparentRFIDTag Wall (TRT-Wall), a novel system taht utilises a passive ultra-high frequency (UHF) radio-frequency identification (RFID) tag array combined with deep learning for contactless human activity monitoring. TheTRT-Wallis tested on five distinct activities: sitting, standing, walking (in both directions), and no-activity. Experimental results demonstrate that theTRT-Walldistinguishes these activities with an impressive average accuracy of $$95.6\\%$$ 95.6% under four distinct distances (2, 2.5, 3.5 and 4.5 m) by capturing the RSSI and phase information. This suggests that our proposed contactless AAL system possesses significant potential to enhance elderly patient-assisted living.","author":[{"family":"Khan","given":"Muhammad"},{"family":"Usman","given":"Muhammad"},{"family":"Tahir","given":"Ahsen"},{"family":"Farooq","given":"Muhammad"},{"family":"Qayyum","given":"Adnan"},{"family":"Ahmad","given":"Jawad"},{"family":"Abbas","given":"Hasan"},{"family":"Imran","given":"Muhammad"},{"family":"Abbasi","given":"Qammer"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41598-024-64411-y","URL":"https://doi.org/10.1038/s41598-024-64411-y","source":"openalex"},{"id":"oa:W4401323124","type":"article-journal","title":"Artificial Intelligence Classification for Detecting and Grading Lumbar Intervertebral Disc Degeneration","abstract":"Introduction: Intervertebral disc degeneration (IDD) is a primary cause of chronic back pain and disability, highlighting the need for precise detection and grading for effective treatment. This study focuses on developing and validating a convolutional neural network (CNN) with a You Only Look Once (YOLO) architecture model using the Pfirrmann grading system to classify and grade lumbar intervertebral disc degeneration based on magnetic resonance imaging (MRI) scans. Methods: We developed a deep learning model trained on a dataset of anonymized MRI studies of patients with symptomatic back pain. MRI images were segmented and annotated by radiologists according to the Pfirrmann grading for the datasets. The segmentation MRI-disc image dataset was prepared for three groups: a training set (1,000), a testing set (500), and an external validation set (500) to assess model generalizability without overlapping images. The model's performance was evaluated using accuracy, sensitivity, specificity, F1 score, prediction error, and ROC-AUC. Results: The AI model showed high performance across all metrics. For Grade I IDD, the model achieved an accuracy of 97%, 95%, and 92% in the training, testing, and external validation sets, respectively. For Grade II, the sensitivity was 100% in both training and testing sets and 98% in the validation set. For Grade III, the specificity was 95.4% in the training set and 94% in both testing and validation sets. For Grade IV, the F1 score was 97.77% in the training set and 95% in both testing and validation sets. For Grade V, the prediction error was 2.3%, 2%, and 2.5% in the training, testing, and validation sets, respectively. The overall ROC-AUC was 97%, 92%, and 95% in the training, testing, and validation sets, respectively. Conclusions: The AI-based classification model exhibits high accuracy, sensitivity, and specificity in detecting and grading lumbar IDD using the Pfirrmann grading. AI has significantly enhanced diagnostic precision and reliability, providing a powerful tool for clinicians in managing IDD. The potential impact is substantial, although further clinical validation is necessary before integrating this model into routine practice.","author":[{"family":"Liawrungrueang","given":"Wongthawat"},{"family":"Cholamjiak","given":"Watcharaporn"},{"family":"Sarasombath","given":"Peem"},{"family":"Jitpakdee","given":"Khanathip"},{"family":"Kotheeranurak","given":"Vit"}],"issued":{"date-parts":[[2024]]},"DOI":"10.22603/ssrr.2024-0154","URL":"https://doi.org/10.22603/ssrr.2024-0154","source":"openalex"},{"id":"oa:W4396613362","type":"article-journal","title":"Artificial Intelligence-Assisted Automated Heart Failure Detection and Classification from Electronic Health Records","abstract":"AIMS: Electronic health records (EHR) linked to Digital Imaging and Communications in Medicine (DICOM), biological specimens, and deep learning (DL) algorithms could potentially improve patient care through automated case detection and surveillance. We hypothesized that by applying keyword searches to routinely stored EHR, in conjunction with AI-powered automated reading of DICOM echocardiography images and analysing biomarkers from routinely stored plasma samples, we were able to identify heart failure (HF) patients. METHODS AND RESULTS: We used EHR data between 1993 and 2021 from Tayside and Fife (~20% of the Scottish population). We implemented a keyword search strategy complemented by filtering based on International Classification of Diseases (ICD) codes and prescription data to EHR data set. We then applied DL for the automated interpretation of echocardiographic DICOM images. These methods were then integrated with the analysis of routinely stored plasma samples to identify and categorize patients into HF with reduced ejection fraction (HFrEF), HF with preserved ejection fraction (HFpEF), and controls without HF. The final diagnosis was verified through a manual review of medical records, measured natriuretic peptides in stored blood samples, and by comparing clinical outcomes among groups. In our study, we selected the patient cohort through an algorithmic workflow. This process started with 60 850 EHR data and resulted in a final cohort of 578 patients, divided into 186 controls, 236 with HFpEF, and 156 with HFrEF, after excluding individuals with mismatched data or significant valvular heart disease. The analysis of baseline characteristics revealed that compared with controls, patients with HFrEF and HFpEF were generally older, had higher BMI, and showed a greater prevalence of co-morbidities such as diabetes, COPD, and CKD. Echocardiographic analysis, enhanced by DL, provided high coverage, and detailed insights into cardiac function, showing significant differences in parameters such as left ventricular diameter, ejection fraction, and myocardial strain among the groups. Clinical outcomes highlighted a higher risk of hospitalization and mortality for HF patients compared with controls, with particularly elevated risk ratios for both HFrEF and HFpEF groups. The concordance between the algorithmic selection of patients and manual validation demonstrated high accuracy, supporting the effectiveness of our approach in identifying and classifying HF subtypes, which could significantly impact future HF diagnosis and management strategies. CONCLUSIONS: Our study highlights the feasibility of combining keyword searches in EHR, DL automated echocardiographic interpretation, and biobank resources to identify HF subtypes.","author":[{"family":"Oo","given":"Mon"},{"family":"Gao","given":"Chuang"},{"family":"Cole","given":"Christian"},{"family":"Hummel","given":"Yoran"},{"family":"Guignardduff","given":"Magalie"},{"family":"Jefferson","given":"Emily"},{"family":"Hare","given":"James"},{"family":"Voors","given":"Adriaan"},{"family":"Boer","given":"Rudolf"},{"family":"Lam","given":"Carolyn"},{"family":"Mordi","given":"Ify"},{"family":"Tromp","given":"Jasper"},{"family":"Lang","given":"Chim"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/ehf2.14828","URL":"https://doi.org/10.1002/ehf2.14828","source":"openalex"},{"id":"oa:W4396808870","type":"article-journal","title":"Artificial intelligence challenges in the face of biological threats: emerging catastrophic risks for public health","abstract":"The threat landscape of biological hazards with the evolution of AI presents challenges. While AI promises innovative solutions, concerns arise about its misuse in the creation of biological weapons. The convergence of AI and genetic editing raises questions about biosecurity, potentially accelerating the development of dangerous pathogens. The mapping conducted highlights the critical intersection between AI and biological threats, underscoring emerging risks in the criminal manipulation of pathogens. Technological advancement in biology requires preventative and regulatory measures. Expert recommendations emphasize the need for solid regulations and responsibility of creators, demanding a proactive, ethical approach and governance to ensure global safety.","author":[{"family":"Lima","given":"Renan"},{"family":"Sinclair","given":"Lucas"},{"family":"Megger","given":"Ricardo"},{"family":"Maciel","given":"Magno"},{"family":"Vasconcelos","given":"Pedro"},{"family":"Quaresma","given":"Juarez"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/frai.2024.1382356","URL":"https://doi.org/10.3389/frai.2024.1382356","source":"openalex"},{"id":"oa:W4404414807","type":"article-journal","title":"Addressing ethical issues in healthcare artificial intelligence using a lifecycle-informed process","abstract":"Objectives: Artificial intelligence (AI) proceeds through an iterative and evaluative process of development, use, and refinement which may be characterized as a lifecycle. Within this context, stakeholders can vary in their interests and perceptions of the ethical issues associated with this rapidly evolving technology in ways that can fail to identify and avert adverse outcomes. Identifying issues throughout the AI lifecycle in a systematic manner can facilitate better-informed ethical deliberation. Materials and Methods: We analyzed existing lifecycles from within the current literature for ethical issues of AI in healthcare to identify themes, which we relied upon to create a lifecycle that consolidates these themes into a more comprehensive lifecycle. We then considered the potential benefits and harms of AI through this lifecycle to identify ethical questions that can arise at each step and to identify where conflicts and errors could arise in ethical analysis. We illustrated the approach in 3 case studies that highlight how different ethical dilemmas arise at different points in the lifecycle. Results Discussion and Conclusion: Through case studies, we show how a systematic lifecycle-informed approach to the ethical analysis of AI enables mapping of the effects of AI onto different steps to guide deliberations on benefits and harms. The lifecycle-informed approach has broad applicability to different stakeholders and can facilitate communication on ethical issues for patients, healthcare professionals, research participants, and other stakeholders.","author":[{"family":"Collins","given":"Benjamin"},{"family":"Bélislepipon","given":"Jean‐christophe"},{"family":"Evans","given":"Barbara"},{"family":"Ferryman","given":"Kadija"},{"family":"Jiang","given":"Xiaoqian"},{"family":"Nebeker","given":"Camille"},{"family":"Novak","given":"Laurie"},{"family":"Roberts","given":"Kirk"},{"family":"Were","given":"Martin"},{"family":"Yin","given":"Zhijun"},{"family":"Ravitsky","given":"Vardit"},{"family":"Coco","given":"Joseph"},{"family":"Hendrickssturrup","given":"Rachele"},{"family":"Williams","given":"Ishan"},{"family":"Clayton","given":"Ellen"},{"family":"Malin","given":"Bradley"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/jamiaopen/ooae108","URL":"https://doi.org/10.1093/jamiaopen/ooae108","source":"openalex"},{"id":"oa:W4404138521","type":"article-journal","title":"Leveraging Medical Knowledge Graphs Into Large Language Models for Diagnosis Prediction: Design and Application Study","abstract":"BACKGROUND: Electronic health records (EHRs) and routine documentation practices play a vital role in patients' daily care, providing a holistic record of health, diagnoses, and treatment. However, complex and verbose EHR narratives can overwhelm health care providers, increasing the risk of diagnostic inaccuracies. While large language models (LLMs) have showcased their potential in diverse language tasks, their application in health care must prioritize the minimization of diagnostic errors and the prevention of patient harm. Integrating knowledge graphs (KGs) into LLMs offers a promising approach because structured knowledge from KGs could enhance LLMs' diagnostic reasoning by providing contextually relevant medical information. OBJECTIVE: This study introduces DR.KNOWS (Diagnostic Reasoning Knowledge Graph System), a model that integrates Unified Medical Language System-based KGs with LLMs to improve diagnostic predictions from EHR data by retrieving contextually relevant paths aligned with patient-specific information. METHODS: DR.KNOWS combines a stack graph isomorphism network for node embedding with an attention-based path ranker to identify and rank knowledge paths relevant to a patient's clinical context. We evaluated DR.KNOWS on 2 real-world EHR datasets from different geographic locations, comparing its performance to baseline models, including QuickUMLS and standard LLMs (Text-to-Text Transfer Transformer and ChatGPT). To assess diagnostic reasoning quality, we designed and implemented a human evaluation framework grounded in clinical safety metrics. RESULTS: -scores, highlighting the benefits of KG integration. Human evaluators found the diagnostic rationales of DR.KNOWS to be aligned strongly with correct clinical reasoning, indicating improved abstraction and reasoning. Recognized limitations include potential biases within the KG data, which we addressed by emphasizing case-specific path selection and proposing future bias-mitigation strategies. CONCLUSIONS: DR.KNOWS offers a robust approach for enhancing diagnostic accuracy and reasoning by integrating structured KG knowledge into LLM-based clinical workflows. Although further work is required to address KG biases and extend generalizability, DR.KNOWS represents progress toward trustworthy artificial intelligence-driven clinical decision support, with a human evaluation framework focused on diagnostic safety and alignment with clinical standards.","author":[{"family":"Gao","given":"Yanjun"},{"family":"Li","given":"Ruizhe"},{"family":"Croxford","given":"Emma"},{"family":"Caskey","given":"John"},{"family":"Patterson","given":"Brian"},{"family":"Churpek","given":"Matthew"},{"family":"Miller","given":"Timothy"},{"family":"Dligach","given":"Dmitriy"},{"family":"Afshar","given":"Majid"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/58670","URL":"https://doi.org/10.2196/58670","source":"openalex"},{"id":"oa:W4401336373","type":"article-journal","title":"ARTIFICIAL INTELLIGENCE – POWERED VIDEO CONTENT GENERATION TOOLS","abstract":"This article discusses the considerations of artificial intelligence-powered video content generation tools, exploring their applications, ethical considerations, and evaluation criteria. Through discussions of various artificial intelligence (AI) tools, including features, limitations, and implications, the authors analyze the evolving landscape of video creation in the digital age. Key themes include the ethical implications of deep fake technology and the importance of responsible AI principles, exemplified by Microsoft's guidelines. This paper identifies five of the most promoted free social media tools. Evaluation criteria for these tools, such as visual quality, relevance, coherence, authenticity, and transparency, are examined to assess the suitability of AI-generated videos. While AI offers promising opportunities, the discussion underscores the continued need for human oversight and ethical considerations to ensure the responsible use of AI technologies in video content generation.","author":[{"family":"Roșca","given":"Cosmina"},{"family":"Gortoescu","given":"Ionuț"},{"family":"Tanase","given":"M"}],"issued":{"date-parts":[[2024]]},"DOI":"10.51865/jpgt.2024.01.10","URL":"https://doi.org/10.51865/jpgt.2024.01.10","source":"openalex"},{"id":"oa:W4401105105","type":"article-journal","title":"Artificial Intelligence in the Diagnosis of Onychomycosis—Literature Review","abstract":"Onychomycosis is a common fungal nail infection that is difficult to diagnose due to its similarity to other nail conditions. Accurate identification is essential for effective treatment. The current gold standard methods include microscopic examination with potassium hydroxide, fungal cultures, and Periodic acid-Schiff biopsy staining. These conventional techniques, however, suffer from high turnover times, variable sensitivity, reliance on human interpretation, and costs. This study examines the potential of integrating AI (artificial intelligence) with visualization tools like dermoscopy and microscopy to improve the accuracy and efficiency of onychomycosis diagnosis. AI algorithms can further improve the interpretation of these images. The review includes 14 studies from PubMed and IEEE databases published between 2010 and 2024, involving clinical and dermoscopic pictures, histopathology slides, and KOH microscopic images. Data extracted include study type, sample size, image assessment model, AI algorithms, test performance, and comparison with clinical diagnostics. Most studies show that AI models achieve an accuracy comparable to or better than clinicians, suggesting a promising role for AI in diagnosing onychomycosis. Nevertheless, the niche nature of the topic indicates a need for further research.","author":[{"family":"Bulińska","given":"Barbara"},{"family":"Mazur-Milecka","given":"Magdalena"},{"family":"Sławińska","given":"Martyna"},{"family":"Rumiński","given":"Jacek"},{"family":"Nowicki","given":"Roman"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/jof10080534","URL":"https://doi.org/10.3390/jof10080534","source":"openalex"},{"id":"oa:W4402884190","type":"article-journal","title":"Artificial intelligence detection of cognitive impairment in older adults during walking","abstract":"INTRODUCTION: To detect early cognitive impairment in community-dwelling older adults, this study explored the viability of artificial intelligence (AI)-assisted linear acceleration and angular velocity analysis during walking. METHODS: This cross-sectional study included 879 participants without dementia (female, 60.6%; mean age, 73.5 years) from the 2011 Comprehensive Gerontology Survey. Sensors attached to the pelvis and left ankle recorded the triaxial linear acceleration and angular velocity while the participants walked at a comfortable speed. Cognitive impairment was determined using Mini-Mental State Examination scores. Deep learning models were used to discern the linear acceleration and angular velocity data of 12,302 walking strides. RESULTS: The models' average sensitivity, specificity, and area under the curve were 0.961, 0.643, and 0.833, respectively, across 30 testing datasets. DISCUSSION: AI-enabled gait analysis can be used to detect signs of cognitive impairment. Integrating this AI model into smartphones may help detect dementia early, facilitating better prevention. Highlights: Artificial intelligence (AI)-enabled gait analysis can be used to detect the early signs of cognitive decline.This AI model was constructed using data from a community-dwelling cohort.AI-assisted linear acceleration and angular velocity analysis during gait was used.The model may help in early detection of dementia.","author":[{"family":"Obuchi","given":"Shuichi"},{"family":"Kojima","given":"Motonaga"},{"family":"Suzuki","given":"Hiroyuki"},{"family":"Garbalosa","given":"Juan"},{"family":"Imamura","given":"Keigo"},{"family":"Ihara","given":"Kazushige"},{"family":"Hirano","given":"Hirohiko"},{"family":"Sasai","given":"Hiroyuki"},{"family":"Fujiwara","given":"Yoshinori"},{"family":"Kawai","given":"Hisashi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/dad2.70012","URL":"https://doi.org/10.1002/dad2.70012","source":"openalex"},{"id":"oa:W4399572703","type":"article-journal","title":"Application of radiomics for preoperative prediction of lymph node metastasis in colorectal cancer: a systematic review and meta-analysis","abstract":"BACKGROUND: Colorectal cancer (CRC) stands as the third most prevalent cancer globally, projecting 3.2 million new cases and 1.6 million deaths by 2040. Accurate lymph node metastasis (LNM) detection is critical for determining optimal surgical approaches, including preoperative neoadjuvant chemoradiotherapy and surgery, which significantly influence CRC prognosis. However, conventional imaging lacks adequate precision, prompting exploration into radiomics, which addresses this shortfall by converting medical images into reproducible, quantitative data. METHODS: Following PRISMA, Supplemental Digital Content 1 (http://links.lww.com/JS9/C77) and Supplemental Digital Content 2 (http://links.lww.com/JS9/C78), and AMSTAR-2 guidelines, Supplemental Digital Content 3 (http://links.lww.com/JS9/C79), we systematically searched PubMed, Web of Science, Embase, Cochrane Library, and Google Scholar databases until 11 January 2024, to evaluate radiomics models' diagnostic precision in predicting preoperative LNM in CRC patients. The quality and bias risk of the included studies were assessed using the Radiomics Quality Score (RQS) and the modified Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. Subsequently, statistical analyses were conducted. RESULTS: Thirty-six studies encompassing 8039 patients were included, with a significant concentration in 2022-2023 (20/36). Radiomics models predicting LNM demonstrated a pooled area under the curve (AUC) of 0.814 (95% CI: 0.78-0.85), featuring sensitivity and specificity of 0.77 (95% CI: 0.69, 0.84) and 0.73 (95% CI: 0.67, 0.78), respectively. Subgroup analyses revealed similar AUCs for CT and MRI-based models, and rectal cancer models outperformed colon and colorectal cancers. Additionally, studies utilizing cross-validation, 2D segmentation, internal validation, manual segmentation, prospective design, and single-center populations tended to have higher AUCs. However, these differences were not statistically significant. Radiologists collectively achieved a pooled AUC of 0.659 (95% CI: 0.627, 0.691), significantly differing from the performance of radiomics models (P<0.001). CONCLUSION: Artificial intelligence-based radiomics shows promise in preoperative lymph node staging for CRC, exhibiting significant predictive performance. These findings support the integration of radiomics into clinical practice to enhance preoperative strategies in CRC management.","author":[{"family":"Abbaspour","given":"Elahe"},{"family":"Karimzadhagh","given":"Sahand"},{"family":"Monsef","given":"Abbas"},{"family":"Joukar","given":"Farahnaz"},{"family":"Mansourghanaei","given":"Fariborz"},{"family":"Hassanipour","given":"Soheil"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1097/js9.0000000000001239","URL":"https://doi.org/10.1097/js9.0000000000001239","source":"openalex"},{"id":"oa:W4399419828","type":"article-journal","title":"Are Artificial Intelligence Virtual Simulated Patients (AI-VSP) a Valid Teaching Modality for Health Professional Students?","abstract":"Introduction Simulation-based learning is robust, but the COVID pandemic created opportunities for novel modalities, including Artificial Intelligence Virtual Simulated Patient (AI-VSP) scenarios. Methods Between 2019 and 2022, the following health professional students experienced AI-VSP at one US university: (a) \"Headache\" scenario: 24 Family Nurse Practitioner (FNP) and 48 Physician Assistants (PA) students, (b) \"Insomnia\" scenario: 64 Bachelor of Science in Nursing (BSN) and 47 Accelerated Bachelor of Science in Nursing (ABSN) students. Each individually conducted a brief strongly encouraged encounter and subsequently optionally participated in our study. Results When asked about the scenario's realism, positive answers were 50% (FNP), 16% (PA), 63% (BSN), and 87% (ABSN). Also, 41% FNP, 52% PA, 65% BSN, and 82% ABSN felt capable of creating diagnoses and treatment plans as thoroughly as they would with human patients. Regarding improving diagnostic abilities, favorable responses were 73% (FNP), 74% (PA), 72% (BSN), and 90% (ABSN). When asked whether they would recommend AI-VSP encounters to others, 91% (FNP), 84% (PA), 93% (BSN), and 90% (ABSN) agreed. Conclusions AI-VSP scenarios were well accepted by students and demonstrated significant promise as a complementary simulation-based learning modality.","author":[{"family":"Mattei","given":"Leticia"},{"family":"Morato","given":"Marcelino"},{"family":"Sidhu","given":"Vineet"},{"family":"Gautam","given":"Nodana"},{"family":"Mendonca","given":"C"},{"family":"Tsai","given":"Albert"},{"family":"Hammer","given":"Marjorie"},{"family":"Creighton-Wong","given":"Lynda"},{"family":"Azzam","given":"Amin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.ecns.2024.101536","URL":"https://doi.org/10.1016/j.ecns.2024.101536","source":"openalex"},{"id":"oa:W4400796849","type":"article-journal","title":"Emergence of Artificial Intelligence Art Therapies ( AIATs ) in Mental Health Care: A Systematic Review","abstract":"The application of artificial intelligence art therapies (AIATs) in mental health care represents an innovative merger between digital technology and the therapeutic potential of creative arts. This systematic review aimed to assess the effectiveness and ethical considerations of AIATs, incorporating robots, AI painting and AI Chatbots to augment traditional art therapies. Aligning with the Preferred Reporting Items for systematic reviews (PRISMA) guidelines, we meticulously searched PubMed, Cochrane Library, Web of Science and CNKI, resulting in 15 selected articles for detailed analysis. To ensure methodological quality, we applied the Joanna Briggs Institute (JBI) criteria for quality assessment and extracted data using the PICO(S) format, specifically targeting randomised controlled trials (RCTs). Our findings suggest that AIATs can profoundly enhance the therapeutic experience by providing new creative outlets and reinforcing existing methods, despite possible drawbacks and ethical challenges. This examination underscores AIATs' potential to enrich mental health therapies, emphasising the critical importance of ethical considerations and the responsible application of AI as the field evolves. With a focus on expanding treatment efficacy and patient expressiveness, the promise of AIATs in mental health care necessitates a careful balance between innovation and ethical responsibility. Trial Registration: PROSPERO: CRD42024504472.","author":[{"family":"Luo","given":"Xuexing"},{"family":"Zhang","given":"Aijia"},{"family":"Li","given":"Yu"},{"family":"Zhang","given":"Zheyu"},{"family":"Ying","given":"Fangtian"},{"family":"Lin","given":"Runqing"},{"family":"Yang","given":"Qianxu"},{"family":"Wang","given":"Jue"},{"family":"Huang","given":"Guanghui"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/inm.13384","URL":"https://doi.org/10.1111/inm.13384","source":"openalex"},{"id":"oa:W4378450174","type":"article-journal","title":"Artificial Intelligence in Medicine: Legal, Ethical and Social Aspects","abstract":"In this article, the authors reflected the legal, ethical and social aspects of the introduction of artificial intelligence in the field of medicine. The authors used the dialectical method to understand the problematic aspects of qualitative changes in the healthcare system of Ukraine in connection with the quantitative increase in the use of artificial intelligence technology. The system method contributed to determining the nature of the impact of the introduction of artificial intelligence on the transformation of the structural elements of legislation in the healthcare sector. Analytical and formal-logical methods were useful in the process of identifying legal and ethical and social problems from the introduction of artificial intelligence and providing proposals for their solution. Emphasis was placed on the current state of the legal regulation of artificial intelligence in Ukraine and the problems of a legal, ethical and social nature that need to be addressed in the process of its implementation. The authors came to the conclusion that Ukraine is now at the initial stage of introducing artificial intelligence into public life. The problem of the lack of legislative work to streamline public relations associated with the use of artificial intelligence has been identified. Proposals are provided that can help mitigate the risks from the introduction of artificial intelligence.","author":[{"family":"Anishchenko","given":"МА"},{"family":"Gidenko","given":"Ievgen"},{"family":"Kaliman","given":"Maksym"},{"family":"Polyvaniuk","given":"Vasyl"},{"family":"Demianchuk","given":"Yurii"}],"issued":{"date-parts":[[2023]]},"DOI":"10.4067/s1726-569x2023000100063","URL":"https://doi.org/10.4067/s1726-569x2023000100063","source":"openalex"},{"id":"oa:W4402692922","type":"article-journal","title":"Human versus Artificial Intelligence: ChatGPT-4 Outperforming Bing, Bard, ChatGPT-3.5 and Humans in Clinical Chemistry Multiple-Choice Questions","abstract":"Introduction: Artificial intelligence (AI) chatbots excel in language understanding and generation. These models can transform healthcare education and practice. However, it is important to assess the performance of such AI models in various topics to highlight its strengths and possible limitations. This study aimed to evaluate the performance of ChatGPT (GPT-3.5 and GPT-4), Bing, and Bard compared to human students at a postgraduate master’s level in Medical Laboratory Sciences. Methods: The study design was based on the METRICS checklist for the design and reporting of AI-based studies in healthcare. The study utilized a dataset of 60 Clinical Chemistry multiple-choice questions (MCQs) initially conceived for assessing 20 MSc students. The revised Bloom’s taxonomy was used as the framework for classifying the MCQs into four cognitive categories: Remember, Understand, Analyze, and Apply. A modified version of the CLEAR tool was used for the assessment of the quality of AI-generated content, with Cohen’s κ for inter-rater agreement. Results: Compared to the mean students’ score which was 0.68± 0.23, GPT-4 scored 0.90 ± 0.30, followed by Bing (0.77 ± 0.43), GPT-3.5 (0.73 ± 0.45), and Bard (0.67 ± 0.48). Statistically significant better performance was noted in lower cognitive domains (Remember and Understand) in GPT-3.5 ( P =0.041), GPT-4 ( P =0.003), and Bard ( P =0.017) compared to the higher cognitive domains (Apply and Analyze). The CLEAR scores indicated that ChatGPT-4 performance was “Excellent” compared to the “Above average” performance of ChatGPT-3.5, Bing, and Bard. Discussion: The findings indicated that ChatGPT-4 excelled in the Clinical Chemistry exam, while ChatGPT-3.5, Bing, and Bard were above average. Given that the MCQs were directed to postgraduate students with a high degree of specialization, the performance of these AI chatbots was remarkable. Due to the risk of academic dishonesty and possible dependence on these AI models, the appropriateness of MCQs as an assessment tool in higher education should be re-evaluated. Keywords: AI in healthcare education, higher education, large language models, evaluation","author":[{"family":"Sallam","given":"Malik"},{"family":"Alsalahat","given":"Khaled"},{"family":"Eid","given":"Huda"},{"family":"Egger","given":"Jan"},{"family":"Puladi","given":"Behrus"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2147/amep.s479801","URL":"https://doi.org/10.2147/amep.s479801","source":"openalex"},{"id":"oa:W4402237623","type":"article-journal","title":"Artificial intelligence-assisted interventions for perioperative anesthetic management: a systematic review and meta-analysis","abstract":"Integration of artificial intelligence (AI) into medical practice has increased recently. Numerous AI models have been developed in the field of anesthesiology; however, their use in clinical settings remains limited. This study aimed to identify the gap between AI research and its implementation in anesthesiology via a systematic review of randomized controlled trials with meta-analysis (CRD42022353727). We searched the databases of Medical Literature Analysis and Retrieval System Online (MEDLINE), Excerpta Medica Database (Embase), Web of Science, Cochrane Central Register of Controlled Trials (CENTRAL), Institute of Electrical and Electronics Engineers Xplore (IEEE), and Google Scholar and retrieved randomized controlled trials comparing conventional and AI-assisted anesthetic management published between the date of inception of the database and August 31, 2023. Eight randomized controlled trials were included in this systematic review ( n = 568 patients), including 286 and 282 patients who underwent anesthetic management with and without AI-assisted interventions, respectively. AI-assisted interventions used in the studies included fuzzy logic control for gas concentrations (one study) and the Hypotension Prediction Index (seven studies; adding only one indicator). Seven studies had small sample sizes ( n = 30 to 68, except for the largest), and meta-analysis including the study with the largest sample size ( n = 213) showed no difference in a hypotension-related outcome (mean difference of the time-weighted average of the area under the threshold 0.22, 95% confidence interval -0.03 to 0.48, P = 0.215, I 2 93.8%). This systematic review and meta-analysis revealed that randomized controlled trials on AI-assisted interventions in anesthesiology are in their infancy, and approaches that take into account complex clinical practice should be investigated in the future. This study was registered with the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD42022353727).","author":[{"family":"Shimada","given":"Kensuke"},{"family":"Inokuchi","given":"Ryota"},{"family":"Ohigashi","given":"Tomohiro"},{"family":"Iwagami","given":"Masao"},{"family":"Tanaka","given":"Makoto"},{"family":"Gosho","given":"Masahiko"},{"family":"Tamiya","given":"Nanako"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s12871-024-02699-z","URL":"https://doi.org/10.1186/s12871-024-02699-z","source":"openalex"},{"id":"oa:W4391954037","type":"article-journal","title":"Establishment and validation of an interactive artificial intelligence platform to predict postoperative ambulatory status for patients with metastatic spinal disease: a multicenter analysis","abstract":"BACKGROUND: Identification of patients with high-risk of experiencing inability to walk after surgery is important for surgeons to make therapeutic strategies for patients with metastatic spinal disease. However, there is a lack of clinical tool to assess postoperative ambulatory status for those patients. The emergence of artificial intelligence (AI) brings a promising opportunity to develop accurate prediction models. METHODS: This study collected 455 patients with metastatic spinal disease who underwent posterior decompressive surgery at three tertiary medical institutions. Of these, 220 patients were collected from one medical institution to form the model derivation cohort, while 89 and 146 patients were collected from two other medical institutions to form the external validation cohorts 1 and 2, respectively. Patients in the model derivation cohort were used to develop and internally validate models. To establish the interactive AI platform, machine learning techniques were used to develop prediction models, including logistic regression (LR), decision tree (DT), random forest (RF), extreme gradient boosting machine (eXGBM), support vector machine (SVM), and neural network (NN). Furthermore, to enhance the resilience of the study's model, an ensemble machine learning approach was employed using a soft-voting method by combining the results of the above six algorithms. A scoring system incorporating 10 evaluation metrics was used to comprehensively assess the prediction performance of the developed models. The scoring system had a total score of 0 to 60, with higher scores denoting better prediction performance. An interactive AI platform was further deployed via Streamlit. The prediction performance was compared between medical experts and the AI platform in assessing the risk of experiencing postoperative inability to walk among patients with metastatic spinal disease. RESULTS: Among all developed models, the ensemble model outperformed the six other models with the highest score of 57, followed by the eXGBM model (54), SVM model (50), and NN model (50). The ensemble model had the best performance in accuracy and calibration slope, and the second-best performance in precise, recall, specificity, area under the curve (AUC), Brier score, and log loss. The scores of the LR model, RF model, and DT model were 39, 46, and 26, respectively. External validation demonstrated that the ensemble model had an AUC value of 0.873 (95% CI: 0.809-0.936) in the external validation cohort 1 and 0.924 (95% CI: 0.890-0.959) in the external validation cohort 2. In the new ensemble machine learning model excluding the feature of the number of comorbidities, the AUC value was still as high as 0.916 (95% CI: 0.863-0.969). In addition, the AUC values of the new model were 0.880 (95% CI: 0.819-0.940) in the external validation cohort 1 and 0.922 (95% CI: 0.887-0.958) in the external validation cohort 2, indicating favorable generalization of the model. The interactive AI platform was further deployed online based on the final machine learning model, and it was available at https://postoperativeambulatory-izpdr6gsxxwhitr8fubutd.streamlit.app/ . By using the AI platform, researchers were able to obtain the individual predicted risk of postoperative inability to walk, gain insights into the key factors influencing the outcome, and find the stratified therapeutic recommendations. The AUC value obtained from the AI platform was significantly higher than the average AUC value achieved by the medical experts ( P <0.001), denoting that the AI platform obviously outperformed the individual medical experts. CONCLUSIONS: The study successfully develops and validates an interactive AI platform for evaluating the risk of postoperative loss of ambulatory ability in patients with metastatic spinal disease. This AI platform has the potential to serve as a valuable model for guiding healthcare professionals in implementing surgical plans and ultimately enhancing","author":[{"family":"Cui","given":"Yunpeng"},{"family":"Shi","given":"Xuedong"},{"family":"Qin","given":"Yong"},{"family":"Wan","given":"Qiwei"},{"family":"Cao","given":"Xuyong"},{"family":"Che","given":"Xiaotong"},{"family":"Pan","given":"Yuanxing"},{"family":"Wang","given":"Bing"},{"family":"Lei","given":"Mingxing"},{"family":"Liu","given":"Yaosheng"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1097/js9.0000000000001169","URL":"https://doi.org/10.1097/js9.0000000000001169","source":"openalex"},{"id":"oa:W4392795414","type":"article-journal","title":"Artificial Intelligence–Based Radiotherapy Contouring and Planning to Improve Global Access to Cancer Care","abstract":"PURPOSE: Increased automation has been identified as one approach to improving global cancer care. The Radiation Planning Assistant (RPA) is a web-based tool offering automated radiotherapy (RT) contouring and planning to low-resource clinics. In this study, the RPA workflow and clinical acceptability were assessed by physicians around the world. METHODS: The RPA output for 75 cases was reviewed by at least three physicians; 31 radiation oncologists at 16 institutions in six countries on five continents reviewed RPA contours and plans for clinical acceptability using a 5-point Likert scale. RESULTS: For cervical cancer, RPA plans using bony landmarks were scored as usable as-is in 81% (with minor edits 93%); using soft tissue contours, plans were scored as usable as-is in 79% (with minor edits 96%). For postmastectomy breast cancer, RPA plans were scored as usable as-is in 44% (with minor edits 91%). For whole-brain treatment, RPA plans were scored as usable as-is in 67% (with minor edits 99%). For head/neck cancer, the normal tissue autocontours were acceptable as-is in 89% (with minor edits 97%). The clinical target volumes (CTVs) were acceptable as-is in 40% (with minor edits 93%). The volumetric-modulated arc therapy (VMAT) plans were acceptable as-is in 87% (with minor edits 96%). For cervical cancer, the normal tissue autocontours were acceptable as-is in 92% (with minor edits 99%). The CTVs for cervical cancer were scored as acceptable as-is in 83% (with minor edits 92%). The VMAT plans for cervical cancer were acceptable as-is in 99% (with minor edits 100%). CONCLUSION: The RPA, a web-based tool designed to improve access to high-quality RT in low-resource settings, has high rates of clinical acceptability by practicing clinicians around the world. It has significant potential for successful implementation in low-resource clinics.","author":[{"family":"Court","given":"Laurence"},{"family":"Aggarwal","given":"Ajay"},{"family":"Jhingran","given":"Anuja"},{"family":"Naidoo","given":"Komeela"},{"family":"Netherton","given":"Tucker"},{"family":"Olanrewaju","given":"Adenike"},{"family":"Peterson","given":"Christine"},{"family":"Parkes","given":"Jeannette"},{"family":"Simonds","given":"Hannah"},{"family":"Trauernicht","given":"Christoph"},{"family":"Zhang","given":"Lifei"},{"family":"Beadle","given":"Beth"},{"family":"Ahmad","given":"Shareen"},{"family":"Anderson","given":"David"},{"family":"Baghwala","given":"Arjig"},{"family":"Chan","given":"Karen"},{"family":"Das","given":"Prajnan"},{"family":"Edwards","given":"Albert"},{"family":"Elbanna","given":"May"},{"family":"Elhalawani","given":"Hesham"},{"family":"Elsayed","given":"Medhat"},{"family":"Ewongwo","given":"Agnes"},{"family":"Fakie","given":"Nazia"},{"family":"Fuller","given":"Clifton"},{"family":"Garden","given":"Adam"},{"family":"Gove","given":"Matt"},{"family":"Urbano","given":"Teresa"},{"family":"Kaittany","given":"Njeri"},{"family":"Khan","given":"Mishal"},{"family":"Langer","given":"Joshua"},{"family":"Leeig","given":"Percy"},{"family":"Lee","given":"Becky"},{"family":"Lee","given":"Anna"},{"family":"Lee","given":"Belinda"},{"family":"Leech","given":"Michelle"},{"family":"Li","given":"Ben"},{"family":"Lichter","given":"Katie"},{"family":"Lin","given":"Lilie"},{"family":"Lin","given":"Stacy"},{"family":"Lombe","given":"Dorothy"},{"family":"Mallick","given":"Indranil"},{"family":"Maroongroge","given":"Sean"},{"family":"Martin","given":"Rachael"},{"family":"Mcginnis","given":"Gwendolyn"},{"family":"Mezera","given":"Megan"},{"family":"Mohammedsaid","given":"Mustefa"},{"family":"Nguyen","given":"Son"},{"family":"Nuanjing","given":"Jenny"},{"family":"Phillips","given":"Tony"},{"family":"Prajapati","given":"Surendra"},{"family":"Punt","given":"Lydia"},{"family":"Reed","given":"Valerie"},{"family":"Roniger","given":"Dominique"},{"family":"Shaitelman","given":"Simona"},{"family":"Sherriff","given":"Alicia"},{"family":"Shiao","given":"Jay"},{"family":"Skinner","given":"Heath"},{"family":"Susan","given":"A"},{"family":"Sutton","given":"Jordan"},{"family":"Syed","given":"Hamza"},{"family":"Thang","given":"Sandy"},{"family":"Thompson","given":"JS"},{"family":"Walker","given":"Gary"},{"family":"Wetter","given":"Julie"},{"family":"White","given":"Ingrid"},{"family":"Xu","given":"Melody"},{"family":"Yousif","given":"Yousif"},{"family":"Zhu","given":"Simeng"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1200/go.23.00376","URL":"https://doi.org/10.1200/go.23.00376","source":"openalex"},{"id":"oa:W4399648420","type":"article-journal","title":"Advancing Psoriasis Care through Artificial Intelligence: A Comprehensive Review","abstract":"Purpose of Review: Machine learning (ML), a subset of artificial intelligence (AI), has been vital in advancing tasks such as image classification and speech recognition. Its integration into clinical medicine, particularly dermatology, offers a significant leap in healthcare delivery. Recent Findings: This review examines the impact of ML on psoriasis-a condition heavily reliant on visual assessments for diagnosis and treatment. The review highlights five areas where ML is reshaping psoriasis care: diagnosis of psoriasis through clinical and dermoscopic images, skin severity quantification, psoriasis biomarker identification, precision medicine enhancement, and AI-driven education strategies. These advancements promise to improve patient outcomes, especially in regions lacking specialist care. However, the success of AI in dermatology hinges on dermatologists' oversight to ensure that ML's potential is fully realized in patient care, preserving the essential human element in medicine. Summary: This collaboration between AI and human expertise could define the future of dermatological treatments, making personalized care more accessible and precise.","author":[{"family":"Smith","given":"Payton"},{"family":"Johnson","given":"Chandler"},{"family":"Haran","given":"Kathryn"},{"family":"Orcales","given":"Faye"},{"family":"Kranyak","given":"Allison"},{"family":"Bhutani","given":"Tina"},{"family":"Rieramonroig","given":"Josep"},{"family":"Liao","given":"Wilson"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s13671-024-00434-y","URL":"https://doi.org/10.1007/s13671-024-00434-y","source":"openalex"},{"id":"oa:W4404055367","type":"article-journal","title":"Validation of an artificial intelligence-based prognostic biomarker in patients with oligometastatic Castration-Sensitive prostate cancer","abstract":"BACKGROUND: There is a need for clinically actionable prognostic and predictive tools to guide the management of oligometastatic castration-sensitive prostate cancer (omCSPC). METHODS: This is a multicenter retrospective study to assess the prognostic and predictive performance of a multimodal artificial intelligence biomarker (MMAI; the ArteraAI Prostate Test) in men with omCSPC (n = 222). The cohort also included 51 patients from the STOMP and ORIOLE phase 2 clinical trials which randomized patients to observation versus metastasis-directed therapy (MDT). MMAI scores were computed from digitized histopathology slides and clinical variables. Overall survival (OS) and time to castration-resistant prostate cancer (TTCRPC) were assessed for the entire cohort from time of diagnosis. Metastasis free survival (MFS) was assessed for the trial cohort from time of randomization. RESULTS: = 0.04. CONCLUSION: The ArteraAI MMAI biomarker is prognostic for OS and TTCRPC among patients with omCSPC and may predict for response to MDT. Further work is needed to validate the MMAI biomarker in a broader mCSPC cohort.","author":[{"family":"Wang","given":"Jarey"},{"family":"Deek","given":"Matthew"},{"family":"Mendes","given":"Adrianna"},{"family":"Song","given":"Yang"},{"family":"Shetty","given":"Amol"},{"family":"Bazyar","given":"Soha"},{"family":"Eecken","given":"Kim"},{"family":"Chen","given":"Emmalyn"},{"family":"Showalter","given":"Timothy"},{"family":"Royce","given":"Trevor"},{"family":"Todorović","given":"Tamara"},{"family":"Huang","given":"Huei–chung"},{"family":"Houck","given":"Scott"},{"family":"Yamashita","given":"Rikiya"},{"family":"Kiess","given":"Ana"},{"family":"Song","given":"Daniel"},{"family":"Lotan","given":"Tamara"},{"family":"Deweese","given":"Theodore"},{"family":"Marchionni","given":"Luigi"},{"family":"Ren","given":"Lei"},{"family":"Sawant","given":"Amit"},{"family":"Simone","given":"Nicole"},{"family":"Berlín","given":"Alejandro"},{"family":"Önal","given":"Cem"},{"family":"Esteva","given":"Andre"},{"family":"Feng","given":"Felix"},{"family":"Tran","given":"Phuoc"},{"family":"Sutera","given":"Philip"},{"family":"Ost","given":"Piet"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.radonc.2024.110618","URL":"https://doi.org/10.1016/j.radonc.2024.110618","source":"openalex"},{"id":"oa:W4402886857","type":"article-journal","title":"Opportunities or Challenges? The Interplay between Artificial Intelligence and Corporate Social Responsibility Communication","abstract":"Abstract Background The rapid development of Artificial Intelligence (AI) offers both opportunities and challenges for its application in Corporate Social Responsibility (CSR) communication. While AI can enhance CSR initiatives, its impact on consumer relations and brand perception remains inconsistent. Objectives This study aims to explore the academic landscape of AI’s role in CSR communication, focusing on publication trends, key authors, research topics, and future directions. Methods/Approach A bibliometric analysis was conducted on 1,094 articles related to AI and CSR communication, retrieved from the Web of Science database from 2000 to February 2024. Using CiteSpace software, the study mapped research trends by analysing disciplines, countries, institutions, authors, references, and keywords. Results The United States and China lead in publication output, with key research themes including social media impact, management strategies, and consumer trust. Emerging trends point to the importance of privacy, service quality, and perceived value in AI-driven CSR initiatives. Conclusions The integration of AI in CSR communication is an evolving field, with significant contributions from social media research and consumer behaviour studies. Future research should address ethical concerns and long-term effects on consumer trust and engagement.","author":[{"family":"Hua","given":"Xiangzhou"},{"family":"Hasan","given":"Nurul"},{"family":"Costa","given":"Feroz"},{"family":"Qiao","given":"Weihua"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2478/bsrj-2024-0007","URL":"https://doi.org/10.2478/bsrj-2024-0007","source":"openalex"},{"id":"oa:W4392861152","type":"article-journal","title":"Automated analysis and detection of epileptic seizures in video recordings using artificial intelligence","abstract":"Introduction Automated seizure detection promises to aid in the prevention of SUDEP and improve the quality of care by assisting in epilepsy diagnosis and treatment adjustment. Methods In this phase 2 exploratory study, the performance of a contactless, marker-free, video-based motor seizure detection system is assessed, considering video recordings of patients (age 0–80 years), in terms of sensitivity, specificity, and Receiver Operating Characteristic (ROC) curves, with respect to video-electroencephalographic monitoring (VEM) as the medical gold standard. Detection performances of five categories of motor epileptic seizures (tonic–clonic, hyperkinetic, tonic, unclassified motor, automatisms) and psychogenic non-epileptic seizures (PNES) with a motor behavioral component lasting for >10 s were assessed independently at different detection thresholds (rather than as a categorical classification problem). A total of 230 patients were recruited in the study, of which 334 in-scope (>10 s) motor seizures (out of 1,114 total seizures) were identified by VEM reported from 81 patients. We analyzed both daytime and nocturnal recordings. The control threshold was evaluated at a range of values to compare the sensitivity ( n = 81 subjects with seizures) and false detection rate (FDR) ( n = all 230 subjects). Results At optimal thresholds, the performance of seizure groups in terms of sensitivity (CI) and FDR/h (CI): tonic–clonic- 95.2% (82.4, 100%); 0.09 (0.077, 0.103), hyperkinetic- 92.9% (68.5, 98.7%); 0.64 (0.59, 0.69), tonic- 78.3% (64.4, 87.7%); 5.87 (5.51, 6.23), automatism- 86.7% (73.5, 97.7%); 3.34 (3.12, 3.58), unclassified motor seizures- 78% (65.4, 90.4%); 4.81 (4.50, 5.14), and PNES- 97.7% (97.7, 100%); 1.73 (1.61, 1.86). A generic threshold recommended for all motor seizures under study asserted 88% sensitivity and 6.48 FDR/h. Discussion These results indicate an achievable performance for major motor seizure detection that is clinically applicable for use as a seizure screening solution in diagnostic workflows.","author":[{"family":"Rai","given":"Pragya"},{"family":"Knight","given":"Andrew"},{"family":"Hiillos","given":"Matias"},{"family":"Kertész","given":"Csaba"},{"family":"Morales","given":"Elizabeth"},{"family":"Terney","given":"Daniella"},{"family":"Larsen","given":"Sidsel"},{"family":"Østerkjerhuus","given":"Tim"},{"family":"Peltola","given":"Jukka"},{"family":"Beniczky","given":"Sándor"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fninf.2024.1324981","URL":"https://doi.org/10.3389/fninf.2024.1324981","source":"openalex"},{"id":"oa:W4404254218","type":"article-journal","title":"Shareable artificial intelligence to extract cancer outcomes from electronic health records for precision oncology research","abstract":"Databases that link molecular data to clinical outcomes can inform precision cancer research into novel prognostic and predictive biomarkers. However, outside of clinical trials, cancer outcomes are typically recorded only in text form within electronic health records (EHRs). Artificial intelligence (AI) models have been trained to extract outcomes from individual EHRs. However, patient privacy restrictions have historically precluded dissemination of these models beyond the centers at which they were trained. In this study, the vulnerability of text classification models trained directly on protected health information to membership inference attacks is confirmed. A teacher-student distillation approach is applied to develop shareable models for annotating outcomes from imaging reports and medical oncologist notes. 'Teacher' models trained on EHR data from Dana-Farber Cancer Institute (DFCI) are used to label imaging reports and discharge summaries from the Medical Information Mart for Intensive Care (MIMIC)-IV dataset. 'Student' models are trained to use these MIMIC documents to predict the labels assigned by teacher models and sent to Memorial Sloan Kettering (MSK) for evaluation. The student models exhibit high discrimination across outcomes in both the DFCI and MSK test sets. Leveraging private labeling of public datasets to distill publishable clinical AI models from academic centers could facilitate deployment of machine learning to accelerate precision oncology research.","author":[{"family":"Kehl","given":"Kenneth"},{"family":"Jee","given":"Justin"},{"family":"Pichotta","given":"Karl"},{"family":"Paul","given":"Megan"},{"family":"Trukhanov","given":"Pavel"},{"family":"Fong","given":"Christopher"},{"family":"Waters","given":"Michele"},{"family":"Bakouny","given":"Ziad"},{"family":"Xu","given":"Wenxin"},{"family":"Choueiri","given":"Toni"},{"family":"Nichols","given":"Chelsea"},{"family":"Schrag","given":"Deborah"},{"family":"Schultz","given":"Nikolaus"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41467-024-54071-x","URL":"https://doi.org/10.1038/s41467-024-54071-x","source":"openalex"},{"id":"oa:W4323659035","type":"article-journal","title":"Using Explainable Artificial Intelligence to Predict Potentially Preventable Hospitalizations","abstract":"BACKGROUND: The increasing aging population and limited health care resources have placed new demands on the healthcare sector. Reducing the number of hospitalizations has become a political priority in many countries, and special focus has been directed at potentially preventable hospitalizations. OBJECTIVES: We aimed to develop an artificial intelligence (AI) prediction model for potentially preventable hospitalizations in the coming year, and to apply explainable AI to identify predictors of hospitalization and their interaction. METHODS: We used the Danish CROSS-TRACKS cohort and included citizens in 2016-2017. We predicted potentially preventable hospitalizations within the following year using the citizens' sociodemographic characteristics, clinical characteristics, and health care utilization as predictors. Extreme gradient boosting was used to predict potentially preventable hospitalizations with Shapley additive explanations values serving to explain the impact of each predictor. We reported the area under the receiver operating characteristic curve, the area under the precision-recall curve, and 95% confidence intervals (CI) based on five-fold cross-validation. RESULTS: The best performing prediction model showed an area under the receiver operating characteristic curve of 0.789 (CI: 0.782-0.795) and an area under the precision-recall curve of 0.232 (CI: 0.219-0.246). The predictors with the highest impact on the prediction model were age, prescription drugs for obstructive airway diseases, antibiotics, and use of municipality services. We found an interaction between age and use of municipality services, suggesting that citizens aged 75+ years receiving municipality services had a lower risk of potentially preventable hospitalization. CONCLUSION: AI is suitable for predicting potentially preventable hospitalizations. The municipality-based health services seem to have a preventive effect on potentially preventable hospitalizations.","author":[{"family":"Riis","given":"Anders"},{"family":"Kristensen","given":"Pia"},{"family":"Lauritsen","given":"Simon"},{"family":"Thiesson","given":"Bo"},{"family":"Jørgensen","given":"Marianne"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1097/mlr.0000000000001830","URL":"https://doi.org/10.1097/mlr.0000000000001830","source":"openalex"},{"id":"oa:W4401610634","type":"article-journal","title":"Enhancing Medical Interview Skills Through AI-Simulated Patient Interactions: Nonrandomized Controlled Trial","abstract":"BACKGROUND: Medical interviewing is a critical skill in clinical practice, yet opportunities for practical training are limited in Japanese medical schools, necessitating urgent measures. Given advancements in artificial intelligence (AI) technology, its application in the medical field is expanding. However, reports on its application in medical interviews in medical education are scarce. OBJECTIVE: This study aimed to investigate whether medical students' interview skills could be improved by engaging with AI-simulated patients using large language models, including the provision of feedback. METHODS: This nonrandomized controlled trial was conducted with fourth-year medical students in Japan. A simulation program using large language models was provided to 35 students in the intervention group in 2023, while 110 students from 2022 who did not participate in the intervention were selected as the control group. The primary outcome was the score on the Pre-Clinical Clerkship Objective Structured Clinical Examination (pre-CC OSCE), a national standardized clinical skills examination, in medical interviewing. Secondary outcomes included surveys such as the Simulation-Based Training Quality Assurance Tool (SBT-QA10), administered at the start and end of the study. RESULTS: The AI intervention group showed significantly higher scores on medical interviews than the control group (AI group vs control group: mean 28.1, SD 1.6 vs 27.1, SD 2.2; P=.01). There was a trend of inverse correlation between the SBT-QA10 and pre-CC OSCE scores (regression coefficient -2.0 to -2.1). No significant safety concerns were observed. CONCLUSIONS: Education through medical interviews using AI-simulated patients has demonstrated safety and a certain level of educational effectiveness. However, at present, the educational effects of this platform on nonverbal communication skills are limited, suggesting that it should be used as a supplementary tool to traditional simulation education.","author":[{"family":"Yamamoto","given":"Akira"},{"family":"Koda","given":"Masahide"},{"family":"Ogawa","given":"Hiroko"},{"family":"Miyoshi","given":"Tomoko"},{"family":"Maeda","given":"Yoshinobu"},{"family":"Otsuka","given":"Fumio"},{"family":"Ino","given":"Hideo"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/58753","URL":"https://doi.org/10.2196/58753","source":"openalex"},{"id":"oa:W4392797668","type":"article-journal","title":"Using artificial intelligence to improve human performance: efficient retinal disease detection training with synthetic images","abstract":"BACKGROUND: Artificial intelligence (AI) in medical imaging diagnostics has huge potential, but human judgement is still indispensable. We propose an AI-aided teaching method that leverages generative AI to train students on many images while preserving patient privacy. METHODS: A web-based course was designed using 600 synthetic ultra-widefield (UWF) retinal images to teach students to detect disease in these images. The images were generated by stable diffusion, a large generative foundation model, which we fine-tuned with 6285 real UWF images from six categories: five retinal diseases (age-related macular degeneration, glaucoma, diabetic retinopathy, retinal detachment and retinal vein occlusion) and normal. 161 trainee orthoptists took the course. They were evaluated with two tests: one consisting of UWF images and another of standard field (SF) images, which the students had not encountered in the course. Both tests contained 120 real patient images, 20 per category. The students took both tests once before and after training, with a cool-off period in between. RESULTS: On average, students completed the course in 53 min, significantly improving their diagnostic accuracy. For UWF images, student accuracy increased from 43.6% to 74.1% (p<0.0001 by paired t-test), nearly matching the previously published state-of-the-art AI model's accuracy of 73.3%. For SF images, student accuracy rose from 42.7% to 68.7% (p<0.0001), surpassing the state-of-the-art AI model's 40%. CONCLUSION: Synthetic images can be used effectively in medical education. We also found that humans are more robust to novel situations than AI models, thus showcasing human judgement's essential role in medical diagnosis.","author":[{"family":"Tabuchi","given":"Hitoshi"},{"family":"Engelmann","given":"Justin"},{"family":"Maeda","given":"Fumiatsu"},{"family":"Nishikawa","given":"Ryo"},{"family":"Nagasawa","given":"Toshihiko"},{"family":"Yamauchi","given":"Tomofusa"},{"family":"Tanabe","given":"Mao"},{"family":"Akada","given":"Masahiro"},{"family":"Kihara","given":"Keita"},{"family":"Nakae","given":"Yasuyuki"},{"family":"Kiuchi","given":"Yoshiaki"},{"family":"Bernabéu","given":"Miguel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1136/bjo-2023-324923","URL":"https://doi.org/10.1136/bjo-2023-324923","source":"openalex"},{"id":"oa:W4403203070","type":"article-journal","title":"A sustainable artificial-intelligence-augmented digital care pathway for epilepsy: Automating seizure tracking based on electroencephalogram data using artificial intelligence","abstract":"Objective: Scalp electroencephalograms (EEGs) are critical for neurological evaluations, particularly in epilepsy, yet they demand specialized expertise that is often lacking in many regions. Artificial intelligence (AI) offers potential solutions to this gap. While existing AI models address certain aspects of EEG analysis, a fully automated system for routine EEG interpretation is required for effective epilepsy management and healthcare professionals' decision-making. This study aims to develop an AI-augmented model for automating EEG seizure tracking, thereby supporting a sustainable digital care pathway for epilepsy (DCPE). The goal is to improve patient monitoring, facilitate collaborative decision-making, ensure timely medication adherence, and promote patient compliance. Method: The study proposes an AI-augmented framework using machine learning, focusing on quantitative analysis of EEG data to automate DCPE. A focus group discussion was conducted with healthcare professionals to find the problem of the current digital care pathway and assess the feasibility, usability, and sustainability of the AI-augmented system in the digital care pathway. Results: The study found that a combination of random forest with principal component analysis and support vector machines with KBest feature selection achieved high accuracy rates of 96.52% and 95.28%, respectively. Additionally, the convolutional neural networks model outperformed other deep learning algorithms with an accuracy of 97.65%. The focus group discussion revealed that automating the diagnostic process in digital care pathway could reduce the time needed to diagnose epilepsy. However, the sustainability of the AI-integrated framework depends on factors such as technological infrastructure, skilled personnel, training programs, patient digital literacy, financial resources, and regulatory compliance. Conclusion: The proposed AI-augmented system could enhance epilepsy management by optimizing seizure tracking accuracy, improving monitoring and timely interventions, facilitating collaborative decision-making, and promoting patient-centered care, thereby making the digital care pathway more sustainable.","author":[{"family":"Keikhosrokiani","given":"Pantea"},{"family":"Isomursu","given":"Minna"},{"family":"Uusimaa","given":"Johanna"},{"family":"Kortelainen","given":"Jukka"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1177/20552076241287356","URL":"https://doi.org/10.1177/20552076241287356","source":"openalex"},{"id":"oa:W4391225179","type":"article-journal","title":"Harnessing the potential of large language models in medical education: promise and pitfalls","abstract":"OBJECTIVES: To provide balanced consideration of the opportunities and challenges associated with integrating Large Language Models (LLMs) throughout the medical school continuum. PROCESS: Narrative review of published literature contextualized by current reports of LLM application in medical education. CONCLUSIONS: LLMs like OpenAI's ChatGPT can potentially revolutionize traditional teaching methodologies. LLMs offer several potential advantages to students, including direct access to vast information, facilitation of personalized learning experiences, and enhancement of clinical skills development. For faculty and instructors, LLMs can facilitate innovative approaches to teaching complex medical concepts and fostering student engagement. Notable challenges of LLMs integration include the risk of fostering academic misconduct, inadvertent overreliance on AI, potential dilution of critical thinking skills, concerns regarding the accuracy and reliability of LLM-generated content, and the possible implications on teaching staff.","author":[{"family":"Benítez","given":"Trista"},{"family":"Xu","given":"Yueyuan"},{"family":"Boudreau","given":"JD"},{"family":"Kow","given":"Alfred"},{"family":"Bello","given":"Fernando"},{"family":"Phuoc","given":"Le"},{"family":"Wang","given":"Xiaofei"},{"family":"Sun","given":"Xiaodong"},{"family":"Leung","given":"Gkk"},{"family":"Lan","given":"Yanyan"},{"family":"Wang","given":"Ya"},{"family":"Cheng","given":"Davy"},{"family":"Tham","given":"Yih"},{"family":"Wong","given":"Tien"},{"family":"Chung","given":"Kevin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/jamia/ocad252","URL":"https://doi.org/10.1093/jamia/ocad252","source":"openalex"},{"id":"oa:W4392807024","type":"article-journal","title":"Generative Pre-Trained Transformer-Empowered Healthcare Conversations: Current Trends, Challenges, and Future Directions in Large Language Model-Enabled Medical Chatbots","abstract":"This review explores the transformative integration of artificial intelligence (AI) and healthcare through conversational AI leveraging Natural Language Processing (NLP). Focusing on Large Language Models (LLMs), this paper navigates through various sections, commencing with an overview of AI’s significance in healthcare and the role of conversational AI. It delves into fundamental NLP techniques, emphasizing their facilitation of seamless healthcare conversations. Examining the evolution of LLMs within NLP frameworks, the paper discusses key models used in healthcare, exploring their advantages and implementation challenges. Practical applications in healthcare conversations, from patient-centric utilities like diagnosis and treatment suggestions to healthcare provider support systems, are detailed. Ethical and legal considerations, including patient privacy, ethical implications, and regulatory compliance, are addressed. The review concludes by spotlighting current challenges, envisaging future trends, and highlighting the transformative potential of LLMs and NLP in reshaping healthcare interactions.","author":[{"family":"Chow","given":"James"},{"family":"Wong","given":"Valerie"},{"family":"Li","given":"Kay"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/biomedinformatics4010047","URL":"https://doi.org/10.3390/biomedinformatics4010047","source":"openalex"},{"id":"oa:W4402586479","type":"article-journal","title":"Interdisciplinary research in artificial intelligence: Lessons from COVID-19","abstract":"Abstract Artificial intelligence (AI) is widely regarded as one of the most promising technologies for advancing science, fostering innovation, and solving global challenges. Recent years have seen a push for teamwork between experts from different fields and AI specialists, but the outcomes of these collaborations have yet to be studied. We focus on approximately 15,000 papers at the intersection of AI and COVID-19—arguably one of the major challenges of recent decades—and show that interdisciplinary collaborations between medical professionals and AI specialists have largely resulted in publications with low visibility and impact. Our findings suggest that impactful research depends less on the overall interdisciplinary of author teams and more on the diversity of knowledge they actually harness in their research. We conclude that team composition significantly influences the successful integration of new computational technologies into science and that obstacles still exist to effective interdisciplinary collaborations in the realm of AI.","author":[{"family":"Abbonato","given":"Diletta"},{"family":"Bianchini","given":"Stefano"},{"family":"Gargiulo","given":"Floriana"},{"family":"Venturini","given":"Tommaso"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1162/qss_a_00329","URL":"https://doi.org/10.1162/qss_a_00329","source":"openalex"},{"id":"oa:W4405241694","type":"article-journal","title":"BDLT-IoMT—a novel architecture: SVM machine learning for robust and secure data processing in Internet of Medical Things with blockchain cybersecurity","abstract":"The integration of artificial intelligence (AI) has caused information and communication technology (ICT) to undergo a number of recent rapid fluctuations. These changes have primarily affected the areas of management, end-to-end device interconnectivity, resource organization, communication, networking, and application-related aspects of ICT. Owing to the complex structure of applicational connectedness, evaluating each of the aforementioned opportunities concurrently reflects the idea of heterogeneity. The association of multiple end devices, particularly in interoperable space, integrity, privacy protection, security, provenance, and the massive volume of everyday media data generated in the modern healthcare setting could also provide significant issues. To address these issues, decentralized, secure, economical resource optimization, and intelligent network activities and organization are necessary. Blockchain technology plays a crucial role in providing distributed storage data organization, sharing, and exchange for automated decision-making, privacy, and security in AI-enabled machine learning (ML) models. However, machine learning models—support vector machine, in particular—have a significant impact on the growth of distributed consortium networks and the exchange of information among connected nodes, resolving issues with resource management, scalability, and data processing. By resolving the three main problems of seamless data integrity, peer-to-peer communication between nodes, and infrastructure security, we provide a novel interoperable technique in this proposed architecture. The approach is unique, as demonstrated by the simulation-based results, which display huge differences of 1.37%, 1.56%, and 1.87%, respectively. The background for the evaluation consists of the following three areas: (i) infrastructure security to protect automated decision-making; (ii) integrity between smooth data sharing and exchange; and (iii) network resource optimization to enable smooth communication across heterogeneous devices.","author":[{"family":"Khan","given":"Abdullah"},{"family":"Laghari","given":"Asif"},{"family":"Baqasah","given":"Abdullah"},{"family":"Bacarra","given":"Rex"},{"family":"Alroobaea","given":"Roobaea"},{"family":"Alsafyani","given":"Majed"},{"family":"Alsayaydeh","given":"Jamil"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s11227-024-06782-7","URL":"https://doi.org/10.1007/s11227-024-06782-7","source":"openalex"},{"id":"oa:W4399104858","type":"article-journal","title":"Perspectives on Artificial Intelligence Adoption for European Union Elderly in the Context of Digital Skills Development","abstract":"In today’s digitalized era, embracing new and emerging technologies is a requirement to remain competitive. The present research investigates the adoption of artificial intelligence (AI) by the elderly in the European landscape, emphasizing the importance of individuals’ digital skills. As has already been globally recognized, the most imminent demographic challenge is no longer represented by the rapid growth of the population but by its aging. Thus, the paper initially analyzed European perspectives on AI adoption, also discussing the importance of focusing on seniors. A bibliometric analysis was required afterward, and the review of the resulting relevant scientific publications uncovered gaps in understanding the relationship between older individuals and AI, particularly in terms of digital competence. Further exploration considered the EU population’s digital literacy and cultural influences using Hofstede’s model, while also identifying potential ways to improve the elderly’s digital skills and promote the adoption of AI. Results indicate a growing interest in AI adoption among the elderly, underscoring the urgent need for digital skills development. The imperative of personalized approach implementations, such as specialized courses, personalized training sessions, or mentoring programs, was underscored. Moreover, the importance of targeted strategies and collaborative efforts to ensure equitable participation in the digital age was identified as a prerequisite for AI adoption by seniors. In terms of potential implications, the research can serve as a starting point for various stakeholders in promoting an effective and sustainable adoption of AI among older citizens in the EU.","author":[{"family":"Bogoslov","given":"Ioana"},{"family":"Corman","given":"Sorina"},{"family":"Lungu","given":"Anca"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/su16114579","URL":"https://doi.org/10.3390/su16114579","source":"openalex"},{"id":"oa:W4404691645","type":"article-journal","title":"Artificial Intelligence and Statistical Models for the Prediction of Radiotherapy Toxicity in Prostate Cancer: A Systematic Review","abstract":"Background: Prostate cancer (PCa) is the second most common cancer in men, and radiotherapy (RT) is one of the main treatment options. Although effective, RT can cause toxic side effects. The accurate prediction of dosimetric parameters, enhanced by advanced technologies and AI-based predictive models, is crucial to optimize treatments and reduce toxicity risks. This study aims to explore current methodologies for predictive dosimetric parameters associated with RT toxicity in PCa patients, analyzing both traditional techniques and recent innovations. Methods: A systematic review was conducted using the PubMed, Scopus, and Medline databases to identify dosimetric predictive parameters for RT in prostate cancer. Studies published from 1987 to April 2024 were included, focusing on predictive models, dosimetric data, and AI techniques. Data extraction covered study details, methodology, predictive models, and results, with an emphasis on identifying trends and gaps in the research. Results: After removing duplicate manuscripts, 354 articles were identified from three databases, with 49 shortlisted for in-depth analysis. Of these, 27 met the inclusion criteria. Most studies utilized logistic regression models to analyze correlations between dosimetric parameters and toxicity, with the accuracy assessed by the area under the curve (AUC). The dosimetric parameter studies included Vdose, Dmax, and Dmean for the rectum, anal canal, bowel, and bladder. The evaluated toxicities were genitourinary, hematological, and gastrointestinal. Conclusions: Understanding dosimetric parameters, such as DVH, Dmax, and Dmean, is crucial for optimizing RT and predicting toxicity. Enhanced predictive accuracy improves treatment effectiveness and reduces side effects, ultimately improving patients’ quality of life. Emerging artificial intelligence and machine learning technologies offer the potential to further refine RT in PCa by analyzing complex data, and enabling more personalized treatment approaches.","author":[{"family":"Piras","given":"Antonio"},{"family":"Corso","given":"Rosario"},{"family":"Benfante","given":"Viviana"},{"family":"Ali","given":"Muhammad"},{"family":"Laudicella","given":"Riccardo"},{"family":"Alongi","given":"Pierpaolo"},{"family":"Daviero","given":"Andrea"},{"family":"Cusumano","given":"Davide"},{"family":"Boldrini","given":"Luca"},{"family":"Salvaggio","given":"Giuseppe"},{"family":"Raimondo","given":"Domenico"},{"family":"Tuttolomondo","given":"Antonino"},{"family":"Comelli","given":"Albert"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/app142310947","URL":"https://doi.org/10.3390/app142310947","source":"openalex"},{"id":"oa:W4400307907","type":"article-journal","title":"Reasoning with large language models for medical question answering","abstract":"OBJECTIVES: To investigate approaches of reasoning with large language models (LLMs) and to propose a new prompting approach, ensemble reasoning, to improve medical question answering performance with refined reasoning and reduced inconsistency. MATERIALS AND METHODS: We used multiple choice questions from the USMLE Sample Exam question files on 2 closed-source commercial and 1 open-source clinical LLM to evaluate our proposed approach ensemble reasoning. RESULTS: On GPT-3.5 turbo and Med42-70B, our proposed ensemble reasoning approach outperformed zero-shot chain-of-thought with self-consistency on Steps 1, 2, and 3 questions (+3.44%, +4.00%, and +2.54%) and (2.3%, 5.00%, and 4.15%), respectively. With GPT-4 turbo, there were mixed results with ensemble reasoning again outperforming zero-shot chain-of-thought with self-consistency on Step 1 questions (+1.15%). In all cases, the results demonstrated improved consistency of responses with our approach. A qualitative analysis of the reasoning from the model demonstrated that the ensemble reasoning approach produces correct and helpful reasoning. CONCLUSION: The proposed iterative ensemble reasoning has the potential to improve the performance of LLMs in medical question answering tasks, particularly with the less powerful LLMs like GPT-3.5 turbo and Med42-70B, which may suggest that this is a promising approach for LLMs with lower capabilities. Additionally, the findings show that our approach helps to refine the reasoning generated by the LLM and thereby improve consistency even with the more powerful GPT-4 turbo. We also identify the potential and need for human-artificial intelligence teaming to improve the reasoning beyond the limits of the model.","author":[{"family":"Lucas","given":"Mary"},{"family":"Yang","given":"Justin"},{"family":"Pomeroy","given":"Jon"},{"family":"Yang","given":"Christopher"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/jamia/ocae131","URL":"https://doi.org/10.1093/jamia/ocae131","source":"openalex"},{"id":"oa:W4403479497","type":"article-journal","title":"Perceptions of Artificial Intelligence and Its Impact on Academic Integrity Among University Students in Peru and Chile: An Approach to Sustainable Education","abstract":"In a context where artificial intelligence (AI) is transforming higher education, this study analyzes how students’ perceptions of AI influence their academic integrity (INA), with a focus on sustainable education. Through a correlational-explanatory analysis based on Structural Equation Models (SEMs) applied to a sample of 659 students from 13 universities in Chile and Peru, it is observed that AI has a significant and direct impact on academic integrity in both countries (β = 0.44). In Peru, the most influential dimension is trust in education (λ = 0.86), followed by social, economic, security, and risk implications (λ = 0.78), while attitudes towards AI also have a direct impact on integrity factors (β = 0.15). In Chile, the dimensions of trust in education (λ = 0.83) and social and economic impact (λ = 0.79) are most relevant, and the relationships between the dimensions of academic integrity such as justice, respect, and responsibility (λ = 0.71) are stronger. The study highlights the importance of incorporating AI literacy into educational curricula and developing regulatory frameworks that promote its ethical use, linking these actions to sustainable education. The findings highlight the need for sustainable educational approaches that enhance understanding of AI and ensure that its use in academia is beneficial, ethical, and contributes to sustainable development.","author":[{"family":"Vidaurre","given":"Sam"},{"family":"Rodríguez","given":"Norma"},{"family":"Quelopana","given":"Renza"},{"family":"Valdivia","given":"Ana"},{"family":"Rossi","given":"Ernesto"},{"family":"Nolasco-Mamani","given":"Marco"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/su16209005","URL":"https://doi.org/10.3390/su16209005","source":"openalex"},{"id":"oa:W4405728415","type":"article-journal","title":"Spotlight on the 2024 ESC/EACTS management of atrial fibrillation guidelines: 10 novel key aspects","abstract":"Atrial fibrillation (AF) remains the most common cardiac arrhythmia worldwide and is associated with significant morbidity and mortality. The European Society of Cardiology (ESC)/European Association for Cardio-Thoracic Surgery (EACTS) have recently released the 2024 guidelines for the management of AF. This review highlights 10 novel aspects of the ESC/EACTS 2024 Guidelines. The AF-CARE framework is introduced, a structural approach that aims to improve patient care and outcomes, comprising of four pillars: [C] Comorbidity and risk factor management, [A] Avoid stroke and thromboembolism, [R] Reduce symptoms by rate and rhythm control, and [E] Evaluation and dynamic reassessment. Additionally, graphical patient pathways are provided to enhance clinical application. A significant shift is the new emphasis on comorbidity and risk factor control to reduce AF recurrence and progression. Individualized assessment of risk is suggested to guide the initiation of oral anticoagulation to prevent thromboembolism. New guidance is provided for anticoagulation in patients with trigger-induced and device-detected sub-clinical AF, ischaemic stroke despite anticoagulation, and the indications for percutaneous/surgical left atrial appendage exclusion. AF ablation is a first-line rhythm control option for suitable patients with paroxysmal AF, and in specific patients, rhythm control can improve prognosis. The AF duration threshold for early cardioversion was reduced from 48 to 24 h, and a wait-and-see approach for spontaneous conversion is advised to promote patient safety. Lastly, strong emphasis is given to optimize the implementation of AF guidelines in daily practice using a patient-centred, multidisciplinary and shared-care approach, with the simultaneous launch of a patient version of the guideline.","author":[{"family":"Rienstra","given":"Michiel"},{"family":"Tzeis","given":"Stylianos"},{"family":"Bunting","given":"Karina"},{"family":"Caso","given":"Valeria"},{"family":"Crijns","given":"Harry"},{"family":"Potter","given":"Tom"},{"family":"Sanders","given":"Prashanthan"},{"family":"Svennberg","given":"Emma"},{"family":"Casado-Arroyo","given":"Rubén"},{"family":"Dwight","given":"Jeremy"},{"family":"Guasti","given":"Luigina"},{"family":"Hanke","given":"Thorsten"},{"family":"Jaarsma","given":"Tiny"},{"family":"Lettino","given":"Maddalena"},{"family":"Løchen","given":"Maja‐lisa"},{"family":"Lumbers","given":"RT"},{"family":"Maesen","given":"Bart"},{"family":"Mølgaard","given":"Inge"},{"family":"Rosano","given":"Giuseppe"},{"family":"Schnabel","given":"Renate"},{"family":"Suwalski","given":"Piotr"},{"family":"Tamargo","given":"Juan"},{"family":"Țica","given":"Otilia"},{"family":"Traykov","given":"Vassil"},{"family":"Kotecha","given":"Dipak"},{"family":"Gelder","given":"Isabelle"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/europace/euae298","URL":"https://doi.org/10.1093/europace/euae298","source":"openalex"},{"id":"oa:W4401895798","type":"article-journal","title":"Artificial Intelligence in Chronic Obstructive Pulmonary Disease: Research Status, Trends, and Future Directions --A Bibliometric Analysis from 2009 to 2023.","abstract":"Objective: A bibliometric analysis was conducted using VOSviewer and CiteSpace to examine studies published between 2009 and 2023 on the utilization of artificial intelligence (AI) in chronic obstructive pulmonary disease (COPD). Methods: On March 24, 2024, a computer search was conducted on the Web of Science (WOS) core collection dataset published between January 1, 2009, and December 30, 2023, to identify literature related to the application of artificial intelligence in chronic obstructive pulmonary disease (COPD). VOSviewer was utilized for visual analysis of countries, institutions, authors, co-cited authors, and keywords. CiteSpace was employed to analyze the intermediary centrality of institutions, references, keyword outbreaks, and co-cited literature. Relevant descriptive analysis tables were created using Excel2021 software. Results: This study included a total of 646 papers from WOS. The number of papers remained small and stable from 2009 to 2017 but started increasing significantly annually since 2018. The United States had the highest number of publications among countries/regions while Silverman Edwin K and Harvard Medical School were the most prolific authors and institutions respectively. Lynch DA, Kirby M. and Vestbo J. were among the top three most cited authors overall. Scientific Reports had the largest number of publications while Radiology ranked as one of the top ten influential journals. The Genetic Epidemiology of COPD (COPDGene) Study Design was frequently cited. Through keyword clustering analysis, all keywords were categorized into four groups: epidemiological study of COPD; AI-assisted imaging diagnosis; AI-assisted diagnosis; and AI-assisted treatment and prognosis prediction in the COPD research field. Currently, hot research topics include explainable artificial intelligence framework, chest CT imaging, and lung radiomics. Conclusion: At present, AI is predominantly employed in genetic biology, early diagnosis, risk staging, efficacy evaluation, and prediction modeling of COPD. This study's results offer novel insights and directions for future research endeavors related to COPD.","author":[{"family":"Bian","given":"Hupo"},{"family":"Zhu","given":"Shaoqi"},{"family":"Zhang","given":"Yonghua"},{"family":"Fei","given":"Qiang"},{"family":"Peng","given":"Xiuhua"},{"family":"Jin","given":"Zanhui"},{"family":"Zhou","given":"Tianxiang"},{"family":"Zhao","given":"Hongxing"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2147/copd.s474402","URL":"https://doi.org/10.2147/copd.s474402","source":"openalex"},{"id":"oa:W4400002936","type":"article-journal","title":"Artificial intelligence-enhanced electrocardiography derived body mass index as a predictor of future cardiometabolic disease","abstract":"Abstract The electrocardiogram (ECG) can capture obesity-related cardiac changes. Artificial intelligence-enhanced ECG (AI-ECG) can identify subclinical disease. We trained an AI-ECG model to predict body mass index (BMI) from the ECG alone. Developed from 512,950 12-lead ECGs from the Beth Israel Deaconess Medical Center (BIDMC), a secondary care cohort, and validated on UK Biobank (UKB) (n = 42,386), the model achieved a Pearson correlation coefficient (r) of 0.65 and 0.62, and an R2 of 0.43 and 0.39 in the BIDMC cohort and UK Biobank, respectively for AI-ECG BMI vs. measured BMI. We found delta-BMI, the difference between measured BMI and AI-ECG-predicted BMI (AI-ECG-BMI), to be a biomarker of cardiometabolic health. The top tertile of delta-BMI showed increased risk of future cardiometabolic disease (BIDMC: HR 1.15, p < 0.001; UKB: HR 1.58, p < 0.001) and diabetes mellitus (BIDMC: HR 1.25, p < 0.001; UKB: HR 2.28, p < 0.001) after adjusting for covariates including measured BMI. Significant enhancements in model fit, reclassification and improvements in discriminatory power were observed with the inclusion of delta-BMI in both cohorts. Phenotypic profiling highlighted associations between delta-BMI and cardiometabolic diseases, anthropometric measures of truncal obesity, and pericardial fat mass. Metabolic and proteomic profiling associates delta-BMI positively with valine, lipids in small HDL, syntaxin-3, and carnosine dipeptidase 1, and inversely with glutamine, glycine, colipase, and adiponectin. A genome-wide association study revealed associations with regulators of cardiovascular/metabolic traits, including SCN10A, SCN5A, EXOG and RXRG. In summary, our AI-ECG-BMI model accurately predicts BMI and introduces delta-BMI as a non-invasive biomarker for cardiometabolic risk stratification.","author":[{"family":"Pastika","given":"Libor"},{"family":"Sau","given":"Arunashis"},{"family":"Patlatzoglou","given":"Konstantinos"},{"family":"Sieliwończyk","given":"Ewa"},{"family":"Ribeiro","given":"Antônio"},{"family":"Mcgurk","given":"Kathryn"},{"family":"Khan","given":"Sadia"},{"family":"Mandic","given":"Danilo"},{"family":"Scott","given":"William"},{"family":"Ware","given":"James"},{"family":"Peters","given":"Nicholas"},{"family":"Ribeiro","given":"Antônio"},{"family":"Kramer","given":"Daniel"},{"family":"Waks","given":"Jonathan"},{"family":"Ng","given":"Fu"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41746-024-01170-0","URL":"https://doi.org/10.1038/s41746-024-01170-0","source":"openalex"},{"id":"oa:W4404160912","type":"article-journal","title":"Orchestration logics for artificial intelligence platforms: From raw data to industry‐specific applications","abstract":"Abstract Artificial intelligence (AI) platforms face distinct orchestration challenges in industry‐specific settings, such as the need for specialised resources, data‐sharing concerns, heterogeneous users and context‐sensitive applications. This study investigates how these platforms can effectively orchestrate autonomous actors in developing and consuming AI applications despite these challenges. Through an analysis of five AI platforms for medical imaging, we identify four orchestration logics: platform resourcing, data‐centric collaboration, distributed refinement and application brokering. These logics illustrate how platform owners can verticalize the AI development process by orchestrating actors who co‐create, share and refine data and AI models, ultimately producing industry‐specific applications capable of generalisation. Our findings extend research on platform orchestration logics and change our perspective from boundary resources to a process of boundary processing. These insights provide a theoretical foundation and practical strategies to build effective industry‐specific AI platforms.","author":[{"family":"Weber","given":"Michael"},{"family":"Hein","given":"Andreas"},{"family":"Weking","given":"Jörg"},{"family":"Krcmar","given":"Helmut"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/isj.12567","URL":"https://doi.org/10.1111/isj.12567","source":"openalex"},{"id":"oa:W4401815753","type":"article-journal","title":"Enhancing cervical cancer cytology screening via artificial intelligence innovation","abstract":"A double-check process helps prevent errors and ensures quality control. However, it may lead to decreased personal accountability, reduced effort, and declining quality checks. Introducing an artificial intelligence (AI)-based system in such scenarios could effectively address the risk of oversights. This study introduces an innovative AI-integrated workflow for cervical cytology screening that substantially improves efficiency and reduces the burden on cytologists. The AI model prioritizes cases for review based on anomaly scores and streamlines the first screening process to approximately 10 s per case. The model enhances the identification of high-risk cases via detailed microscopic observation, high anomaly scores cases, and a targeted review of low-score cases. The workflow highlights its capability for rapid, accurate, and less labor-intensive evaluations, demonstrating the potential to transform cervical cancer screening. This study highlights the importance of AI in modern medical diagnostics, particularly in areas with a high demand for accuracy and efficiency.","author":[{"family":"Kurita","given":"Yuki"},{"family":"Meguro","given":"Shiori"},{"family":"Kosugi","given":"Isao"},{"family":"Enomoto","given":"Yasunori"},{"family":"Kawasaki","given":"Hideya"},{"family":"Kano","given":"Tomoaki"},{"family":"Saitoh","given":"Takeji"},{"family":"Shinmura","given":"Kazuya"},{"family":"Iwashita","given":"Toshihide"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41598-024-70670-6","URL":"https://doi.org/10.1038/s41598-024-70670-6","source":"openalex"},{"id":"oa:W4403650365","type":"article-journal","title":"Role of artificial intelligence in haematolymphoid diagnostics","abstract":"The advent of digital pathology and the deployment of high-throughput molecular techniques are generating an unprecedented mass of data. Thanks to advances in computational sciences, artificial intelligence (AI) approaches represent a promising avenue for extracting relevant information from complex data structures. From diagnostic assistance to powerful research tools, the potential fields of application of machine learning techniques in pathology are vast and constitute the subject of considerable research work. The aim of this article is to provide an overview of the potential applications of AI in the field of haematopathology and to define the role that these emerging technologies could play in our laboratories in the short to medium term.","author":[{"family":"Syrykh","given":"Charlotte"},{"family":"Brand","given":"Michiel"},{"family":"Kather","given":"Jakob"},{"family":"Laurent","given":"Camille"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/his.15327","URL":"https://doi.org/10.1111/his.15327","source":"openalex"},{"id":"oa:W4404568512","type":"article-journal","title":"Advanced Artificial Intelligence Techniques for Comprehensive Dermatological Image Analysis and Diagnosis","abstract":"With the growing complexity of skin disorders and the challenges of traditional diagnostic methods, AI offers exciting new solutions that can enhance the accuracy and efficiency of dermatological assessments. Reflectance Confocal Microscopy (RCM) stands out as a non-invasive imaging technique that delivers detailed views of the skin at the cellular level, proving its immense value in dermatology. The manual analysis of RCM images, however, tends to be slow and inconsistent. By combining artificial intelligence (AI) with RCM, this approach introduces a transformative shift toward precise, data-driven dermatopathology, supporting more accurate patient stratification, tailored treatments, and enhanced dermatological care. Advancements in AI are set to revolutionize this process. This paper explores how AI, particularly Convolutional Neural Networks (CNNs), can enhance RCM image analysis, emphasizing machine learning (ML) and deep learning (DL) methods that improve diagnostic accuracy and efficiency. The discussion highlights AI’s role in identifying and classifying skin conditions, offering benefits such as a greater consistency and a reduced strain on healthcare professionals. Furthermore, the paper explores AI integration into dermatological practices, addressing current challenges and future possibilities. The synergy between AI and RCM holds the potential to significantly advance skin disease diagnosis, ultimately leading to better therapeutic personalization and comprehensive dermatological care.","author":[{"family":"Aksoy","given":"Serra"},{"family":"Demircioğlu","given":"Pınar"},{"family":"Böğrekçi","given":"İsmail"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/dermato4040015","URL":"https://doi.org/10.3390/dermato4040015","source":"openalex"},{"id":"oa:W4321004114","type":"article-journal","title":"Technical characterisation of digital stethoscopes: towards scalable artificial intelligence-based auscultation","abstract":"Digital stethoscopes can enable the development of integrated artificial intelligence (AI) systems that can remove the subjectivity of manual auscultation, improve diagnostic accuracy, and compensate for diminishing auscultatory skills. Developing scalable AI systems can be challenging, especially when acquisition devices differ and thus introduce sensor bias. To address this issue, a precise knowledge of these differences, i.e., frequency responses of these devices, is needed, but the manufacturers often do not provide complete device specifications. In this study, we reported an effective methodology for determining the frequency response of a digital stethoscope and used it to characterise three common digital stethoscopes: Littmann 3200, Eko Core, and Thinklabs One. Our results show significant inter-device variability in that the frequency responses of the three studied stethoscopes were distinctly different. A moderate intra-device variability was seen when comparing two separate units of Littmann 3200. The study highlights the need for normalisation across devices for developing successful AI-assisted auscultation and provides a technical characterisation approach as a first step to accomplish it.","author":[{"family":"Arjoune","given":"Youness"},{"family":"Nguyen","given":"Trong"},{"family":"Doroshow","given":"Robin"},{"family":"Shekhar","given":"Raj"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1080/03091902.2023.2174198","URL":"https://doi.org/10.1080/03091902.2023.2174198","source":"openalex"},{"id":"oa:W4401638119","type":"article-journal","title":"Assessment Study of ChatGPT-3.5’s Performance on the Final Polish Medical Examination: Accuracy in Answering 980 Questions","abstract":"Background/Objectives: The use of artificial intelligence (AI) in education is dynamically growing, and models such as ChatGPT show potential in enhancing medical education. In Poland, to obtain a medical diploma, candidates must pass the Medical Final Examination, which consists of 200 questions with one correct answer per question, is administered in Polish, and assesses students’ comprehensive medical knowledge and readiness for clinical practice. The aim of this study was to determine how ChatGPT-3.5 handles questions included in this exam. Methods: This study considered 980 questions from five examination sessions of the Medical Final Examination conducted by the Medical Examination Center in the years 2022–2024. The analysis included the field of medicine, the difficulty index of the questions, and their type, namely theoretical versus case-study questions. Results: The average correct answer rate achieved by ChatGPT for the five examination sessions hovered around 60% and was lower (p < 0.001) than the average score achieved by the examinees. The lowest percentage of correct answers was in hematology (42.1%), while the highest was in endocrinology (78.6%). The difficulty index of the questions showed a statistically significant correlation with the correctness of the answers (p = 0.04). Questions for which ChatGPT-3.5 provided incorrect answers had a lower (p < 0.001) percentage of correct responses. The type of questions analyzed did not significantly affect the correctness of the answers (p = 0.46). Conclusions: This study indicates that ChatGPT-3.5 can be an effective tool for assisting in passing the final medical exam, but the results should be interpreted cautiously. It is recommended to further verify the correctness of the answers using various AI tools.","author":[{"family":"Siebielec","given":"Julia"},{"family":"Ordak","given":"Michał"},{"family":"Oskroba","given":"Agata"},{"family":"Dworakowska","given":"Anna"},{"family":"Bujalskazadrożny","given":"Magdalena"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/healthcare12161637","URL":"https://doi.org/10.3390/healthcare12161637","source":"openalex"},{"id":"oa:W4405335762","type":"article-journal","title":"Explainable Artificial Intelligence in Paediatric: Challenges for the Future","abstract":"Background: Explainable artificial intelligence (XAI) emerged to improve the transparency of machine learning models and increase understanding of how models make actions and decisions. It helps to present complex models in a more digestible form from a human perspective. However, XAI is still in the development stage and must be used carefully in sensitive domains including paediatrics, where misuse might have adverse consequences. Objective: This commentary paper discusses concerns and challenges related to implementation and interpretation of XAI methods, with the aim of rising awareness of the main concerns regarding their adoption in paediatrics. Methods: A comprehensive literature review was undertaken to explore the challenges of adopting XAI in paediatrics. Results: Although XAI has several favorable outcomes, its implementation in paediatrics is prone to challenges including generalizability, trustworthiness, causality and intervention, and XAI evaluation. Conclusion: Paediatrics is a very sensitive domain where consequences of misinterpreting AI outcomes might be very significant. XAI should be adopted carefully with focus on evaluating the outcomes primarily by including paediatricians in the loop, enriching the pipeline by injecting domain knowledge promoting a cross-fertilization perspective aiming at filling the gaps still preventing its adoption.","author":[{"family":"Salih","given":"Ahmed"},{"family":"Menegaz","given":"Gloria"},{"family":"Pillay","given":"Thillagavathie"},{"family":"Boyle","given":"Elaine"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/hsr2.70271","URL":"https://doi.org/10.1002/hsr2.70271","source":"openalex"},{"id":"oa:W4392813893","type":"article-journal","title":"Assessment of patient perceptions of artificial intelligence use in dermatology: A cross‐sectional survey","abstract":"Dear Editor, The use of artificial intelligence (AI) in medicine has grown in recent decades, with deep neural networks demonstrating accuracies comparable to dermatologists when classifying melanoma, keratinocyte carcinomas, and other common skin conditions.1-3 With the future possibility that AI will be integrated into dermatology practice, it is important to understand how patients view these possible changes. Although prior studies have shown that patients are open to the use of AI in the diagnosis of skin cancer, little is known about patients' trust and perception of AI accuracy in general dermatology.4, 5 This survey study aimed to gather opinions from a diverse dermatology patient population on AI use in dermatology and establish a specific accuracy at which patients would be comfortable receiving a diagnosis solely from an AI tool. We created a 20-question survey utilizing a five-point Likert scale to assess patient opinions on AI in dermatology. Patients were given a specific example of AI use in dermatology in which a program would analyze a patient-acquired photograph of a skin lesion or rash and provide a list of potential diagnoses to the patient based on the photograph. Patients were then asked to complete a survey on their opinion of this type of AI (Attachment 1). The survey was given randomly via convenience sampling to adult patients who visited the University of Texas Southwestern Medical Center Dermatology Clinic between June 2023 and September 2023. Standard deviation, frequency distribution, and multivariable logistic regression were used in statistical analysis. The UT Southwestern Institutional Review Board approved this study. Among 157 patients informed about the study, 141 (89.8%) consented to complete the survey. Seventy-three respondents (51.8%) were male, 79 respondents (56.0%) were non-Hispanic white, and the mean (SD) age was 55.3 (16.5) years (Table 1). Respondents had a household income of $50,000–$99,999 (55 [39.0%]) and 61 (43.2%) respondents attained a bachelor's degree). Most respondents did not work in healthcare (125 [88.7%]), and 33 (23.4%) respondents obtained a degree in or held a job in computer science. The majority of respondents believed a dermatologist's diagnosis was accurate (131 [92.9%]), whereas only a minority believed a diagnosis made by AI to be accurate (52 [36.9%]) (Table 2). If differing diagnoses were received from a dermatologist and an AI model, most respondents would trust a dermatologist over an AI model(119 [84.4%]). Even with equal diagnostic accuracy, most respondents preferred to see a dermatologist over an AI model alone (119 [84.4%]). Respondents required the AI model to be 12.9% (SD, ± 8.1%) more accurate on average than a dermatologist in order for respondents to be comfortable only receiving evaluation from an AI model and not a dermatologist (Table 3). Some respondents were completely unwilling to be evaluated by an AI model alone (21 [14.9%]). Nonetheless, a majority of respondents believed that a model that could provide diagnoses based on a photograph could help improve the accuracy of dermatologists (88 [62.4%]), and most would rather get a diagnosis from a dermatologist working with an AI model than solely a dermatologist (96 [68.1%]). After performing a multivariable logistic regression controlling for sociodemographic factors, age 40–59 was significantly associated with a decrease in familiarity with AI (odds ratio: 0.21, p < 0.01) (Table S1). Being familiar with AI was significantly associated with a positive view of AI (odds ratio: 17.8, p < 0.01), belief that AI can improve the accuracy of dermatologists (odds ratio: 4.73, p = 0.04), and preference to receive a diagnosis from a dermatologist working with an AI over a dermatologist alone (odds ratio: 39.58, p < 0.01). Interestingly, having a computer science degree or working in computer science was not significantly associated with a more positive of AI. Our results suggest that although patien","author":[{"family":"Wu","given":"Alexander"},{"family":"Ngo","given":"Madeline"},{"family":"Thomas","given":"Cristina"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/srt.13656","URL":"https://doi.org/10.1111/srt.13656","source":"openalex"},{"id":"oa:W4402964664","type":"article-journal","title":"Enhancing Oral Cancer Detection: A Systematic Review of the Diagnostic Accuracy and Future Integration of Optical Coherence Tomography with Artificial Intelligence","abstract":"Introduction: Optical Coherence Tomography (OCT) has emerged as an important imaging modality in non-invasive diagnosis for oral cancer and can provide real-time visualisation of tissue morphology with the required high resolution. This systematic review aims to assess the diagnostic accuracy of OCT in the detection of oral cancers, and to explore the potential integration of OCT with artificial intelligence (AI) and other imaging techniques to enhance diagnostic precision and clinical outcomes in oral healthcare. Methods: A systematic literature search was conducted across PubMed, Embase, Scopus, Google Scholar, Cochrane Central Register, and Web of Science from inception until August 2024. Studies were included if they employed OCT for oral cancer detection, reported diagnostic outcomes, such as sensitivity and specificity, and were conducted on human subjects. Data extraction and quality assessment were performed independently by two reviewers. The synthesis highlights advancements in OCT technology, including AI-enhanced interpretations. Results: A total of 9 studies met the inclusion criteria, encompassing a total of 860 events (cancer detections). The studies spanned from 2008 to 2022 and utilised various OCT techniques, including clinician-based, algorithm-based, and AI-driven interpretations. The findings indicate OCT’s high diagnostic accuracy, with sensitivity ranging from 75% to 100% and specificity from 71% to 100%. AI-augmented OCT interpretations demonstrated the highest accuracy, emphasising OCT’s potential in early cancer detection and precision in guiding surgical interventions. Conclusions: OCT could play a very prominent role as a new diagnostic tool for oral cancer, with very high sensitivity and specificity. Future research pointed towards integrating OCT with other imaging methods and AI systems in providing better accuracy of diagnoses, plus more clinical usability. Further development and validation with large-scale multicentre trials is imperative for the realisation of this potential in changing the way we practice oral healthcare.","author":[{"family":"Jerjes","given":"Waseem"},{"family":"Stevenson","given":"Harvey"},{"family":"Ramsay","given":"Daniele"},{"family":"Hamdoon","given":"Zaid"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/jcm13195822","URL":"https://doi.org/10.3390/jcm13195822","source":"openalex"},{"id":"oa:W4402464029","type":"article-journal","title":"Barriers and facilitators to implementing imaging-based diagnostic artificial intelligence-assisted decision-making software in hospitals in China: a qualitative study using the updated Consolidated Framework for Implementation Research","abstract":"OBJECTIVES: To identify the barriers and facilitators to the successful implementation of imaging-based diagnostic artificial intelligence (AI)-assisted decision-making software in China, using the updated Consolidated Framework for Implementation Research (CFIR) as a theoretical basis to develop strategies that promote effective implementation. DESIGN: This qualitative study involved semistructured interviews with key stakeholders from both clinical settings and industry. Interview guide development, coding, analysis and reporting of findings were thoroughly informed by the updated CFIR. SETTING: Four healthcare institutions in Beijing and Shanghai and two vendors of AI-assisted decision-making software for lung nodules detection and diabetic retinopathy screening were selected based on purposive sampling. PARTICIPANTS: A total of 23 healthcare practitioners, 6 hospital informatics specialists, 4 hospital administrators and 7 vendors of the selected AI-assisted decision-making software were included in the study. RESULTS: Within the 5 CFIR domains, 10 constructs were identified as barriers, 8 as facilitators and 3 as both barriers and facilitators. Major barriers included unsatisfactory clinical performance (Innovation); lack of collaborative network between primary and tertiary hospitals, lack of information security measures and certification (outer setting); suboptimal data quality, misalignment between software functions and goals of healthcare institutions (inner setting); unmet clinical needs (individuals). Key facilitators were strong empirical evidence of effectiveness, improved clinical efficiency (innovation); national guidelines related to AI, deployment of AI software in peer hospitals (outer setting); integration of AI software into existing hospital systems (inner setting) and involvement of clinicians (implementation process). CONCLUSIONS: The study findings contributed to the ongoing exploration of AI integration in healthcare from the perspective of China, emphasising the need for a comprehensive approach considering both innovation-specific factors and the broader organisational and contextual dynamics. As China and other developing countries continue to advance in adopting AI technologies, the derived insights could further inform healthcare practitioners, industry stakeholders and policy-makers, guiding policies and practices that promote the successful implementation of imaging-based diagnostic AI-assisted decision-making software in healthcare for optimal patient care.","author":[{"family":"Liao","given":"Xiwen"},{"family":"Yao","given":"Chen"},{"family":"Jin","given":"Feifei"},{"family":"Zhang","given":"Jun"},{"family":"Liu","given":"Larry"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1136/bmjopen-2024-084398","URL":"https://doi.org/10.1136/bmjopen-2024-084398","source":"openalex"},{"id":"oa:W4405623173","type":"article-journal","title":"Large Language Models in Gastroenterology: Systematic Review","abstract":"BACKGROUND: As health care continues to evolve with technological advancements, the integration of artificial intelligence into clinical practices has shown promising potential to enhance patient care and operational efficiency. Among the forefront of these innovations are large language models (LLMs), a subset of artificial intelligence designed to understand, generate, and interact with human language at an unprecedented scale. OBJECTIVE: This systematic review describes the role of LLMs in improving diagnostic accuracy, automating documentation, and advancing specialist education and patient engagement within the field of gastroenterology and gastrointestinal endoscopy. METHODS: Core databases including MEDLINE through PubMed, Embase, and Cochrane Central registry were searched using keywords related to LLMs (from inception to April 2024). Studies were included if they satisfied the following criteria: (1) any type of studies that investigated the potential role of LLMs in the field of gastrointestinal endoscopy or gastroenterology, (2) studies published in English, and (3) studies in full-text format. The exclusion criteria were as follows: (1) studies that did not report the potential role of LLMs in the field of gastrointestinal endoscopy or gastroenterology, (2) case reports and review papers, (3) ineligible research objects (eg, animals or basic research), and (4) insufficient data regarding the potential role of LLMs. Risk of Bias in Non-Randomized Studies-of Interventions was used to evaluate the quality of the identified studies. RESULTS: Overall, 21 studies on the potential role of LLMs in gastrointestinal disorders were included in the systematic review, and narrative synthesis was done because of heterogeneity in the specified aims and methodology in each included study. The overall risk of bias was low in 5 studies and moderate in 16 studies. The ability of LLMs to spread general medical information, offer advice for consultations, generate procedure reports automatically, or draw conclusions about the presumptive diagnosis of complex medical illnesses was demonstrated by the systematic review. Despite promising benefits, such as increased efficiency and improved patient outcomes, challenges related to data privacy, accuracy, and interdisciplinary collaboration remain. CONCLUSIONS: We highlight the importance of navigating these challenges to fully leverage LLMs in transforming gastrointestinal endoscopy practices. TRIAL REGISTRATION: PROSPERO 581772; https://www.crd.york.ac.uk/prospero/.","author":[{"family":"Gong","given":"Eun"},{"family":"Bang","given":"Chang"},{"family":"Lee","given":"Jae"},{"family":"Park","given":"Jonghyung"},{"family":"Kim","given":"Eunsil"},{"family":"Kim","given":"Subeen"},{"family":"Kimm","given":"Minjae"},{"family":"Choi","given":"Seoung"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2196/66648","URL":"https://doi.org/10.2196/66648","source":"openalex"},{"id":"oa:W4401417975","type":"article-journal","title":"Advancing Healthcare Accessibility: Fusing Artificial Intelligence with Flexible Sensing to Forge Digital Health Innovations","abstract":"In recent years, the rapid advancement of digital technologies has precipitated a paradigm shift in global healthcare, heralding a new era of digital health methodologies.This transition underscores a universal consensus on the imperative of digitalization and the application of sophisticated information and communication technologies to achieve universal health coverage, aimed at enhancing health outcomes and overall well-being.Central to this transformation is the integration of advanced technologies such as artificial intelligence (AI), big data analytics, wearable smart devices, and the Internet of Things (IoT), which have greatly enhanced data collection, analysis, storage, and transmission, laying the foundation for a comprehensive healthcare system [1,2].This system significantly improves diagnostic accuracy, enables data-driven therapeutic interventions, supports individual health self-management, and promotes patient-centric care, thus playing a crucial role in the ongoing development of healthcare infrastructures.As digital health continues to evolve, it prompts a focused examination toward eliminating healthcare disparities, aiming for a universal healthcare model that emphasizes accessibility, affordability, and sustainability [3].Flexible sensing technologies stand out in today's healthcare system because of their adaptability, comfort, and lightweight construction, which improve data accuracy and continuity for enhanced health monitoring.Their integration with AI represents a significant advancement in personal health management and supports a more inclusive digital health ecosystem, vastly enhancing health data analysis and advancing healthcare equity and accessibility (Figure ).Constructed from elastic materials designed to conform to the human body's various shapes, flexible sensors are increasingly utilized in healthcare for monitoring health indicators.These sensors are adept at measuring a comprehensive array of physiological and biochemical markers, such as heart rate, blood pressure, respiratory functions, electrophysiological signals, metabolites, proteins, electrolytes, and microbial presence, showcasing their utility through high compliance, seamless integration, noninvasiveness, and cost-effectiveness [4].Their application spans remote disease monitoring, therapeutic interventions, intelligent prosthetics, and rehabilitation technologies,","author":[{"family":"Huang","given":"Lingting"},{"family":"Chen","given":"Zhengjie"},{"family":"Yáng","given":"Zhèn"},{"family":"Huang","given":"Wei"}],"issued":{"date-parts":[[2024]]},"DOI":"10.34133/bmef.0062","URL":"https://doi.org/10.34133/bmef.0062","source":"openalex"},{"id":"oa:W4391841559","type":"article-journal","title":"An artificial intelligence-based bone age assessment model for Han and Tibetan children","abstract":"Background: Manual bone age assessment (BAA) is associated with longer interpretation time and higher cost and variability, thus posing challenges in areas with restricted medical facilities, such as the high-altitude Tibetan Plateau. The application of artificial intelligence (AI) for automating BAA could facilitate resolving this issue. This study aimed to develop an AI-based BAA model for Han and Tibetan children. Methods: A model named “EVG-BANet” was trained using three datasets, including the Radiology Society of North America (RSNA) dataset (training set n = 12611, validation set n = 1425, and test set n = 200), the Radiological Hand Pose Estimation (RHPE) dataset (training set n = 5491, validation set n = 713, and test set n = 79), and a self-established local dataset [training set n = 825 and test set n = 351 (Han n = 216 and Tibetan n = 135)]. An open-access state-of-the-art model BoNet was used for comparison. The accuracy and generalizability of the two models were evaluated using the abovementioned three test sets and an external test set (n = 256, all were Tibetan). Mean absolute difference (MAD) and accuracy within 1 year were used as indicators. Bias was evaluated by comparing the MAD between the demographic groups. Results: EVG-BANet outperformed BoNet in the MAD on the RHPE test set (0.52 vs. 0.63 years, p < 0.001), the local test set (0.47 vs. 0.62 years, p < 0.001), and the external test set (0.53 vs. 0.66 years, p < 0.001) and exhibited a comparable MAD on the RSNA test set (0.34 vs. 0.35 years, p = 0.934). EVG-BANet achieved accuracy within 1 year of 97.7% on the local test set (BoNet 90%, p < 0.001) and 89.5% on the external test set (BoNet 85.5%, p = 0.066). EVG-BANet showed no bias in the local test set but exhibited a bias related to chronological age in the external test set. Conclusion: EVG-BANet can accurately predict the bone age (BA) for both Han children and Tibetan children living in the Tibetan Plateau with limited healthcare facilities.","author":[{"family":"Liu","given":"Qixing"},{"family":"Wang","given":"Huogen"},{"family":"Wangjiu","given":"Cidan"},{"family":"Awang","given":"Tudan"},{"family":"Yang","given":"Meijie"},{"family":"Qiongda","given":"Puqiong"},{"family":"Yang","given":"Xiao"},{"family":"Pan","given":"Hui"},{"family":"Wang","given":"Fengdan"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fphys.2024.1329145","URL":"https://doi.org/10.3389/fphys.2024.1329145","source":"openalex"},{"id":"oa:W4403544935","type":"article-journal","title":"AI Rx: Revolutionizing Healthcare Through Intelligence, Innovation, and Ethics","abstract":"The integration of artificial intelligence (AI) in healthcare presents significant promise to enhance clinical procedures and patient outcomes. This research examines the setting, methodology, conclusions, and issues associated with AI in healthcare. The swift proliferation of digital health data, encompassing medical imaging and clinical records, has generated substantial prospects for AI applications. Artificial intelligence methodologies, including machine learning, natural language processing, and computer vision, facilitate the derivation of significant insights from intricate datasets, hence improving clinical decision-making. A thorough literature review examines the practical applications of AI, encompassing its roles in medical diagnostics, treatment planning, and patient outcome prediction. The report also examines ethical issues, data protection, and legal frameworks, which are crucial for the responsible application of AI in healthcare. The results illustrate AI's capacity to enhance diagnostic precision, facilitate administrative efficiency, and optimise resource distribution, resulting in tailored therapies and improved healthcare administration. Nonetheless, obstacles persist, such as data integrity, algorithm transparency, and ethical considerations, which must be resolved to guarantee the secure and efficient deployment of AI. Continuous research, cooperation between healthcare and AI experts, and the establishment of comprehensive regulatory frameworks are essential for optimising the advantages of AI while minimising hazards. This research highlights AI's capacity to transform healthcare, stressing the necessity for a multidisciplinary strategy to effectively harness its benefits and tackle the associated ethical and regulatory dilemmas.","author":[{"family":"Wahed","given":"Mutaz"},{"family":"Alqaraleh","given":"Muhyeeddin"},{"family":"Alzboon","given":"Mowafaq"},{"family":"Al-Batah","given":"Mohammad"}],"issued":{"date-parts":[[2024]]},"DOI":"10.56294/mw202535","URL":"https://doi.org/10.56294/mw202535","source":"openalex"},{"id":"oa:W4399052098","type":"article-journal","title":"Systematic review: The use of large language models as medical chatbots in digestive diseases","abstract":"BACKGROUND: Interest in large language models (LLMs), such as OpenAI's ChatGPT, across multiple specialties has grown as a source of patient-facing medical advice and provider-facing clinical decision support. The accuracy of LLM responses for gastroenterology and hepatology-related questions is unknown. AIMS: To evaluate the accuracy and potential safety implications for LLMs for the diagnosis, management and treatment of questions related to gastroenterology and hepatology. METHODS: We conducted a systematic literature search including Cochrane Library, Google Scholar, Ovid Embase, Ovid MEDLINE, PubMed, Scopus and the Web of Science Core Collection to identify relevant articles published from inception until January 28, 2024, using a combination of keywords and controlled vocabulary for LLMs and gastroenterology or hepatology. Accuracy was defined as the percentage of entirely correct answers. RESULTS: Among the 1671 reports screened, we identified 33 full-text articles on using LLMs in gastroenterology and hepatology and included 18 in the final analysis. The accuracy of question-responding varied across different model versions. For example, accuracy ranged from 6.4% to 45.5% with ChatGPT-3.5 and was between 40% and 91.4% with ChatGPT-4. In addition, the absence of standardised methodology and reporting metrics for studies involving LLMs places all the studies at a high risk of bias and does not allow for the generalisation of single-study results. CONCLUSIONS: Current general-purpose LLMs have unacceptably low accuracy on clinical gastroenterology and hepatology tasks, which may lead to adverse patient safety events through incorrect information or triage recommendations, which might overburden healthcare systems or delay necessary care.","author":[{"family":"Giuffrè","given":"Mauro"},{"family":"Kresevic","given":"Simone"},{"family":"You","given":"Kisung"},{"family":"Dupont","given":"Johannes"},{"family":"Huebner","given":"Jack"},{"family":"Grimshaw","given":"Alyssa"},{"family":"Shung","given":"Dennis"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/apt.18058","URL":"https://doi.org/10.1111/apt.18058","source":"openalex"},{"id":"oa:W4405854719","type":"article-journal","title":"Why Should Users Take the Risk of Sustainable Use of Generative Artificial Intelligence Chatbots","abstract":"Despite the risks associated with generative AI (GenAI) chatbots, people increasingly use these technologies, which may seem contradictory. This study identified and explored factors and risks related to trust, perceived values, satisfaction, and sustainable use of GenAI chatbots. Relying on IS theories to build a stimulus-organism-response model, the authors tested a model using PLS-SEM with data from 393 ChatGPT users. The results show that user competence and autonomy dramatically increase a user's trust in ChatGPT, and trust improves hedonic value (HV), utilitarian value (UV), value-in-use, perceived task-technology fit (TTF), information accuracy, knowledge acquisition, perceived informativeness, and user satisfaction. In addition to trust, user satisfaction depends on HV, UV, and TTF. The sustainability use of ChatGPT depends on HV and satisfaction. However, perceived privacy concerns, perceived privacy risks, and privacy awareness do not affect consumer trust. There is a complete mediation between trust and sustainability, as well as HV and sustainability.","author":[{"family":"Wamba-Taguimdje","given":"Serge"},{"family":"Wamba","given":"Samuel"},{"family":"Twinomurinzi","given":"Hossana"}],"issued":{"date-parts":[[2024]]},"DOI":"10.4018/jgim.365600","URL":"https://doi.org/10.4018/jgim.365600","source":"openalex"},{"id":"oa:W4396724067","type":"article-journal","title":"Artificial intelligence–aided steatosis assessment in donor livers according to the Banff consensus recommendations","abstract":"OBJECTIVES: Severe macrovesicular steatosis in donor livers is associated with primary graft dysfunction. The Banff Working Group on Liver Allograft Pathology has proposed recommendations for steatosis assessment of donor liver biopsy specimens with a consensus for defining \"large droplet fat\" (LDF) and a 3-step algorithmic approach. METHODS: We retrieved slides and initial pathology reports from potential liver donor biopsy specimens from 2010 to 2021. Following the Banff approach, we reevaluated LDF steatosis and employed a computer-assisted manual quantification protocol and artificial intelligence (AI) model for analysis. RESULTS: In a total of 113 slides from 88 donors, no to mild (<33%) macrovesicular steatosis was reported in 88.5% (100/113) of slides; 8.8% (10/113) was reported as at least moderate steatosis (≥33%) initially. Subsequent pathology evaluation, following the Banff recommendation, revealed that all slides had LDF below 33%, a finding confirmed through computer-assisted manual quantification and an AI model. Correlation coefficients between pathologist and computer-assisted manual quantification, between computer-assisted manual quantification and the AI model, and between the AI model and pathologist were 0.94, 0.88, and 0.81, respectively (P < .0001 for all). CONCLUSIONS: The 3-step approach proposed by the Banff Working Group on Liver Allograft Pathology may be followed when evaluating steatosis in donor livers. The AI model can provide a rapid and objective assessment of liver steatosis.","author":[{"family":"Jiao","given":"Jingjing"},{"family":"Tang","given":"Haiming"},{"family":"Sun","given":"Nanfei"},{"family":"Zhang","given":"Xuchen"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/ajcp/aqae053","URL":"https://doi.org/10.1093/ajcp/aqae053","source":"openalex"},{"id":"oa:W4391067767","type":"article-journal","title":"Application of Artificial Intelligence in Infant Movement Classification: A Reliability and Validity Study in Infants Who Were Full-Term and Preterm","abstract":"OBJECTIVE: Preterm infants are at high risk of neuromotor disorders. Recent advances in digital technology and machine learning algorithms have enabled the tracking and recognition of anatomical key points of the human body. It remains unclear whether the proposed pose estimation model and the skeleton-based action recognition model for adult movement classification are applicable and accurate for infant motor assessment. Therefore, this study aimed to develop and validate an artificial intelligence (AI) model framework for movement recognition in full-term and preterm infants. METHODS: This observational study prospectively assessed 30 full-term infants and 54 preterm infants using the Alberta Infant Motor Scale (58 movements) from 4 to 18 months of age with their movements recorded by 5 video cameras simultaneously in a standardized clinical setup. The movement videos were annotated for the start/end times and presence of movements by 3 pediatric physical therapists. The annotated videos were used for the development and testing of an AI algorithm that consisted of a 17-point human pose estimation model and a skeleton-based action recognition model. RESULTS: The infants contributed 153 sessions of Alberta Infant Motor Scale assessment that yielded 13,139 videos of movements for data processing. The intra and interrater reliabilities for movement annotation of videos by the therapists showed high agreements (88%-100%). Thirty-one of the 58 movements were selected for machine learning because of sufficient data samples and developmental significance. Using the annotated results as the standards, the AI algorithm showed satisfactory agreement in classifying the 31 movements (accuracy = 0.91, recall = 0.91, precision = 0.91, and F1 score = 0.91). CONCLUSION: The AI algorithm was accurate in classifying 31 movements in full-term and preterm infants from 4 to 18 months of age in a standardized clinical setup. IMPACT: The findings provide the basis for future refinement and validation of the algorithm on home videos to be a remote infant movement assessment.","author":[{"family":"Lin","given":"Shiang"},{"family":"Chandra","given":"Erick"},{"family":"Tsao","given":"Po‐nien"},{"family":"Liao","given":"Wei‐chih"},{"family":"Chen","given":"Wei"},{"family":"Yen","given":"Ting‐an"},{"family":"Hsu","given":"Jane"},{"family":"Jeng","given":"Suh‐fang"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/ptj/pzad176","URL":"https://doi.org/10.1093/ptj/pzad176","source":"openalex"},{"id":"oa:W4401698185","type":"article-journal","title":"Machine Learning Prediction of Autism Spectrum Disorder From a Minimal Set of Medical and Background Information","abstract":"Importance: Early identification of the likelihood of autism spectrum disorder (ASD) using minimal information is crucial for early diagnosis and intervention, which can affect developmental outcomes. Objective: To develop and validate a machine learning (ML) model for predicting ASD using a minimal set of features from background and medical information and to evaluate the predictors and the utility of the ML model. Design, Setting, and Participants: For this diagnostic study, a retrospective analysis of the Simons Foundation Powering Autism Research for Knowledge (SPARK) database, version 8 (released June 6, 2022), was conducted, including data from 30 660 participants after adjustments for missing values and class imbalances (15 330 with ASD and 15 330 without ASD). The SPARK database contains participants recruited from 31 university-affiliated research clinicals and online in 26 states in the US. All individuals with a professional ASD diagnosis and their families were eligible to participate. The model performance was validated on independent datasets from SPARK, version 10 (released July 21, 2023), and the Simons Simplex Collection (SSC), consisting of 14 790 participants, followed by phenotypic associations. Exposures: Twenty-eight basic medical screening and background history items present before 24 months of age. Main Outcomes and Measures: Generalizable ML prediction models were developed for detecting ASD using 4 algorithms (logistic regression, decision tree, random forest, and eXtreme Gradient Boosting [XGBoost]). Performance metrics included accuracy, area under the receiver operating characteristics curve (AUROC), sensitivity, specificity, positive predictive value (PPV), and F1 score, offering a comprehensive assessment of the predictive accuracy of the model. Explainable AI methods were applied to determine the effect of individual features in predicting ASD as secondary outcomes, enhancing the interpretability of the best-performing model. The secondary outcome analyses were further complemented by examining differences in various phenotypic measures using nonparametric statistical methods, providing insights into the ability of the model to differentiate between different presentations of ASD. Results: The study included 19 477 (63.5%) male and 11 183 (36.5%) female participants (mean [SD] age, 106 [62] months). The mean (SD) age was 113 (68) months for the ASD group and 100 (55) months for the non-ASD group. The XGBoost (termed AutMedAI) model demonstrated strong performance with an AUROC score of 0.895, sensitivity of 0.805, specificity of 0.829, and PPV of 0.897. Developmental milestones and eating behavior were the most important predictors. Validation on independent cohorts showed an AUROC of 0.790, indicating good generalizability. Conclusions and Relevance: In this diagnostic study of ML prediction of ASD, robust model performance was observed to identify autistic individuals with more symptoms and lower cognitive levels. The robustness and ML model generalizability results are promising for further validation and use in clinical and population settings.","author":[{"family":"Rajagopalan","given":"Shyam"},{"family":"Zhang","given":"Yali"},{"family":"Yahia","given":"Ashraf"},{"family":"Tammimies","given":"Kristiina"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1001/jamanetworkopen.2024.29229","URL":"https://doi.org/10.1001/jamanetworkopen.2024.29229","source":"openalex"},{"id":"oa:W4399970933","type":"article-journal","title":"Artificial Intelligence in Infectious Skin Disease","abstract":"Abstract Background Infective and infectious dermatological diseases, which range from minor diseases like impetigo to serious diseases like deep fungal infections, pose significant public health issues. Given the growth of drug‐resistant microorganisms, it is critical to investigate novel techniques in dermatology. Artificial Intelligence (AI) has shown promising results in improving the diagnosis, treatment, and management of infectious skin disorders. This has the potential to significantly improve dermatological treatment by combining physician experience with data‐driven insights. Objective This review will look into the existing uses and future possibilities of AI technologies in infectious dermatology, including machine learning and deep learning. Its goal is to highlight major advances, identify gaps in understanding and technical advancement, and recommend viable future research directions. Methods A comprehensive literature search of the scientific literature was performed using well‐known databases such as PubMed, Google Scholar, and Embase. A specific set of phrases relevant to AI and infectious dermatology was used to ensure a thorough search. Articles were picked based on their relevance, timeliness, and quality, with a particular emphasis on research demonstrating how AI is being utilized to prevent, detect, diagnose, or manage infectious skin disorders. Results AI has made significant contributions to the management of infection in dermatology. It has improved diagnostic accuracy, predictive modeling of drug resistance, and individualized care regimens. Deep learning is used to evaluate clinical images, predictive models are developed to forecast antibiotic resistance, and AI‐powered diagnostic tools for uncommon infections. The assessment also throws light on AI's role in pandemic preparedness and response.","author":[{"family":"Rokni","given":"Ghasem"},{"family":"Gholizadeh","given":"Nasim"},{"family":"Babaei","given":"Mahsa"},{"family":"Das","given":"Kinnor"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/der2.241","URL":"https://doi.org/10.1002/der2.241","source":"openalex"},{"id":"oa:W4391252914","type":"article-journal","title":"Exploring Computing Paradigms for Electric Vehicles: From Cloud to Edge Intelligence, Challenges and Future Directions","abstract":"Electric vehicles are widely adopted globally as a sustainable mode of transportation. With the increased availability of onboard computation and communication capabilities, vehicles are moving towards automated driving and intelligent transportation systems. The adaption of technologies such as IoT, edge intelligence, 5G, and blockchain in vehicle architecture has increased possibilities towards efficient and sustainable transportation systems. In this article, we present a comprehensive study and analysis of the edge computing paradigm, explaining elements of edge AI. Furthermore, we discussed the edge intelligence approach for deploying AI algorithms and models on edge devices, which are typically resource-constrained devices located at the edge of the network. It mentions the advantages of edge intelligence and its use cases in smart electric vehicles. It also discusses challenges and opportunities and provides in-depth analysis for optimizing computation for edge intelligence. Finally, it sheds some light on the research roadmap on AI for edge and AI on edge by dividing efforts into topology, content, service segments, model adaptation, framework design, and processor acceleration, all of which stand to gain advantages from AI technologies. Investigating the incorporation of important technologies, issues, opportunities, and Roadmap in this study will be a valuable resource for the community engaged in research on edge intelligence in electric vehicles.","author":[{"family":"Chougule","given":"Sachin"},{"family":"Chaudhari","given":"Bharat"},{"family":"Ghorpade","given":"Sheetal"},{"family":"Zennaro","given":"Marco"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/wevj15020039","URL":"https://doi.org/10.3390/wevj15020039","source":"openalex"},{"id":"oa:W4366602741","type":"article-journal","title":"Artificial Intelligence Applications in Glioma With 1p/19q Co‐Deletion: A Systematic Review","abstract":"As an important genomic marker for oligodendrogliomas, early determination of 1p/19q co-deletion status is critical for guiding therapy and predicting prognosis in patients with glioma. The purpose of this study is to systematically review the literature concerning the magnetic resonance imaging (MRI) with artificial intelligence (AI) methods for predicting 1p/19q co-deletion status in glioma. PubMed, Scopus, Embase, and IEEE Xplore were searched in accordance with the Preferred Reporting Items for systematic reviews and meta-analyses guidelines. Methodological quality of studies was assessed according to the Quality Assessment of Diagnostic Accuracy Studies-2. Finally, 28 studies were included in the quantitative analysis. Diagnostic test accuracy reached an area under the ROC curve of 0.71-0.98 were reported in 24 studies. The remaining four studies with no available AUC provided an accuracy of 0.75-0. 89. The included studies varied widely in terms of imaging sequences, input features, and modeling methods. The current review highlighted that integrating MRI with AI technology is a potential tool for determination 1p/19q status pre-operatively and noninvasively, which can possibly help clinical decision-making. However, the reliability and feasibility of this approach still need to be further validated and improved in a real clinical setting. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: 2.","author":[{"family":"Zhang","given":"Simin"},{"family":"Yin","given":"Lijuan"},{"family":"Ma","given":"Lu"},{"family":"Sun","given":"Huaiqiang"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/jmri.28737","URL":"https://doi.org/10.1002/jmri.28737","source":"openalex"},{"id":"oa:W4401178944","type":"article-journal","title":"Radiographical diagnostic competences of dental students using various feedback methods and integrating an artificial intelligence application—A randomized clinical trial","abstract":"INTRODUCTION: Radiographic diagnostic competences are a primary focus of dental education. This study assessed two feedback methods to enhance learning outcomes and explored the feasibility of artificial intelligence (AI) to support education. MATERIALS AND METHODS: Fourth-year dental students had access to 16 virtual radiological example cases for 8 weeks. They were randomly assigned to either elaborated feedback (eF) or knowledge of results feedback (KOR) based on expert consensus. Students´ diagnostic competences were tested on bitewing/periapical radiographs for detection of caries, apical periodontitis, accuracy for all radiological findings and image quality. We additionally assessed the accuracy of an AI system (dentalXrai Pro 3.0), where applicable. Data were analysed descriptively and using ROC analysis (accuracy, sensitivity, specificity, AUC). Groups were compared with Welch's t-test. RESULTS: Among 55 students, the eF group by large performed significantly better than the KOR group in detecting enamel caries (accuracy 0.840 ± 0.041, p = .196; sensitivity 0.638 ± 0.204, p = .037; specificity 0.859 ± 0.050, p = .410; ROC AUC 0.748 ± 0.094, p = .020), apical periodontitis (accuracy 0.813 ± 0.095, p = .011; sensitivity 0.476 ± 0.230, p = .003; specificity 0.914 ± 0.108, p = .292; ROC AUC 0.695 ± 0.123, p = .001) and in assessing the image quality of periapical images (p = .031). No significant differences were observed for the other outcomes. The AI showed almost perfect diagnostic performance (enamel caries: accuracy 0.964, sensitivity 0.857, specificity 0.074; dentin caries: accuracy 0.988, sensitivity 0.941, specificity 1.0; overall: accuracy 0.976, sensitivity 0.958, specificity 0.983). CONCLUSION: Elaborated feedback can improve student's radiographic diagnostic competences, particularly in detecting enamel caries and apical periodontitis. Using an AI may constitute an alternative to expert labelling of radiographs.","author":[{"family":"Rampf","given":"Sarah"},{"family":"Gehrig","given":"Holger"},{"family":"Möltner","given":"Andreas"},{"family":"Fischer","given":"Martin"},{"family":"Schwendicke","given":"Falk"},{"family":"Huth","given":"Karin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/eje.13028","URL":"https://doi.org/10.1111/eje.13028","source":"openalex"},{"id":"oa:W4404642710","type":"article-journal","title":"BindingDB in 2024: a FAIR knowledgebase of protein-small molecule binding data","abstract":"BindingDB (bindingdb.org) is a public, web-accessible database of experimentally measured binding affinities between small molecules and proteins, which supports diverse applications including medicinal chemistry, biochemical pathway annotation, training of artificial intelligence models and computational chemistry methods development. This update reports significant growth and enhancements since our last review in 2016. Of note, the database now contains 2.9 million binding measurements spanning 1.3 million compounds and thousands of protein targets. This growth is largely attributable to our unique focus on curating data from US patents, which has yielded a substantial influx of novel binding data. Recent improvements include a remake of the website following responsive web design principles, enhanced search and filtering capabilities, new data download options and webservices and establishment of a long-term data archive replicated across dispersed sites. We also discuss BindingDB's positioning relative to related resources, its open data sharing policies, insights gleaned from the dataset and plans for future growth and development.","author":[{"family":"Liu","given":"Tiqing"},{"family":"Hwang","given":"Linda"},{"family":"Burley","given":"SK"},{"family":"Nitsche","given":"Carmen"},{"family":"Southan","given":"Christopher"},{"family":"Walters","given":"WP"},{"family":"Gilson","given":"Michael"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/nar/gkae1075","URL":"https://doi.org/10.1093/nar/gkae1075","source":"openalex"},{"id":"oa:W4404318930","type":"article-journal","title":"Artificial intelligence research in radiation oncology: a practical guide for the clinician on concepts and methods","abstract":"The use of artificial intelligence (AI) holds great promise for radiation oncology, with many applications being reported in the literature, including some of which are already in clinical use. These are mainly in areas where AI provides benefits in efficiency (such as automatic segmentation and treatment planning). Prediction models that directly impact patient decision-making are far less mature in terms of their application in clinical practice. Part of the limited clinical uptake of these models may be explained by the need for broader knowledge, among practising clinicians within the medical community, about the processes of AI development. This lack of understanding could lead to low commitment to AI research, widespread scepticism, and low levels of trust. This attitude towards AI may be further negatively impacted by the perception that deep learning is a \"black box\" with inherently low transparency. Thus, there is an unmet need to train current and future clinicians in the development and application of AI in medicine. Improving clinicians' AI-related knowledge and skills is necessary to enhance multidisciplinary collaboration between data scientists and physicians, that is, involving a clinician in the loop during AI development. Increased knowledge may also positively affect the acceptance and trust of AI. This paper describes the necessary steps involved in AI research and development, and thus identifies the possibilities, limitations, challenges, and opportunities, as seen from the perspective of a practising radiation oncologist. It offers the clinician with limited knowledge and experience in AI valuable tools to evaluate research papers related to an AI model application.","author":[{"family":"Hoebers","given":"Frank"},{"family":"Wee","given":"Leonard"},{"family":"Likitlersuang","given":"Jirapat"},{"family":"Mak","given":"Raymond"},{"family":"Bitterman","given":"Danielle"},{"family":"Huang","given":"Yanqi"},{"family":"Dekker","given":"André"},{"family":"Aerts","given":"Hugo"},{"family":"Kann","given":"Benjamin"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1093/bjro/tzae039","URL":"https://doi.org/10.1093/bjro/tzae039","source":"openalex"},{"id":"oa:W4403546675","type":"article-journal","title":"Development and Validation of an Artificial Intelligence–Assisted Patient Education Material for Ostomy Patients: A Methodological Study","abstract":"AIM: To develop and test the validity of an artificial intelligence-assisted patient education material for ostomy patients. DESIGN: A methodological study. METHODS: The study was carried out in two main stages and five steps: (1) determining the information needs of ostomy patients, (2) creating educational content, (3) converting the educational content into patient education material, (4) validation of patient education material based on expert review and (5) measuring the readability of the patient education material. We used ChatGPT 4.0 to determine the information needs and create patient education material content, and Publuu Online Flipbook Maker was used to convert the educational content into patient education material. Understandability and applicability scores were assessed using the Patient Education Materials Assessment Tool submitted to 10 expert reviews. The tool inter-rater reliability was determined via the intraclass correlation coefficient. Readability was analysed using the Flesch-Kincaid Grade Level, Gunning Fog Index and Simple Measure of Gobbledygook formula. RESULTS: The mean Patient Education Materials Assessment Tool understandability score of the patient education material was 81.91%, and the mean Patient Education Materials Assessment Tool actionability score was 85.33%. The scores for the readability indicators were calculated to be Flesch-Kincaid Grade Level: 8.53, Gunning Fog: 10.9 and Simple Measure of Gobbledygook: 7.99. CONCLUSIONS: The AI-assisted patient education material for ostomy patients provided accurate information with understandable and actionable responses to patients, but is at a high reading level for patients. IMPLICATIONS FOR THE PROFESSION AND PATIENT CARE: Artificial intelligence-assisted patient education materials can significantly increase patient information rates in the health system regarding ease of practice. Artificial intelligence is currently not an option for creating patient education material, and their impact on the patient is not fully known. REPORTING METHOD: The study followed the STROBE checklist guidelines. PATIENT OR PUBLIC CONTRIBUTION: No patient or public contributions.","author":[{"family":"Kaçmaz","given":"Hatice"},{"family":"Kahraman","given":"Hilal"},{"family":"Akutay","given":"Seda"},{"family":"Dağdelen","given":"Derya"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/jan.16542","URL":"https://doi.org/10.1111/jan.16542","source":"openalex"},{"id":"oa:W4390233959","type":"article-journal","title":"The European Union’s Approach to Artificial Intelligence and the Challenge of Financial Systemic Risk","abstract":"Abstract This piece examines the EU’s ‘Proposal for a Regulation of the European Parliament and of the Council Laying Down Harmonised Rules on Artificial Intelligence’ (‘AI Act’) with a view to determining the extent to which it addresses the systemic risk created by AI FinTech. Ultimately, it is argued that the notion of ‘high risk’ at the centre of the AI Act leaves out financial systemic risk. This exclusion can neither be justified by reasons of technology neutrality, nor by reasons of proportionality: neither is AI-driven financial systemic risk already covered by existing (or proposed) macroprudential frameworks and tools, nor can its omission from the AI Act be justified by the prioritisation of other types of risk. Moving forward, it is suggested that the EU’s AI Act would have benefited from a broader definition of ‘high risk’. It is also hoped that EU policy makers will soon begin to strengthen existing macroprudential toolkits to address the financial systemic risk created by AI.","author":[{"family":"Keller","given":"Anat"},{"family":"Pereira","given":"Clara"},{"family":"Pires","given":"Martinho"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1007/978-3-031-41264-6_22","URL":"https://doi.org/10.1007/978-3-031-41264-6_22","source":"openalex"},{"id":"oa:W4392292312","type":"article-journal","title":"The Digital Divide in Action: How Experiences of Digital Technology Shape Future Relationships with Artificial Intelligence","abstract":"The digital divide remains an ongoing societal concern, with digital exclusion shown to have a significantly detrimental impact on people’s quality of life. Artificial intelligence (AI), the latest wave of digitalisation, is being integrated into the fabric of society at an accelerated rate, the speed of which has prompted ethical concerns. Without addressing the digital divide, the AI revolution risks exacerbating the existing consequences of digital exclusion and limiting the potential for all people to reap the benefits provided by AI. To understand the factors that might contribute to experiences of AI, and how these might be related to digital exclusion, we surveyed a diverse online community sample (N = 303). We created a novel measure of digital confidence capturing individual levels of awareness, familiarity, and sense of competence with digital technology. Results indicated that measures of digital confidence were predicted by structural, behavioural, and psychological differences, such that women, older people, those on lower salaries, people with less digital access, and those with lower digital well-being, reported significantly less digital confidence. Furthermore, digital confidence significantly moderated the relationship between people’s experiences with everyday AI technologies and their general attitudes towards AI. This understanding of the spill-over effects of digital exclusion onto experiences of AI is fundamental to the articulation and delivery of inclusive AI.","author":[{"family":"Bentley","given":"Sarah"},{"family":"Naughtin","given":"Claire"},{"family":"Mcgrath","given":"Melanie"},{"family":"Irons","given":"Jessica"},{"family":"Cooper","given":"Patrick"}],"issued":{"date-parts":[[2024]]},"DOI":"10.31234/osf.io/wd4tr","URL":"https://doi.org/10.31234/osf.io/wd4tr","source":"openalex"},{"id":"oa:W4386126083","type":"article-journal","title":"Application of digital-intelligence technology in the processing of Chinese materia medica","abstract":"Processing of Chinese Materia Medica (PCMM) is the concentrated embodiment, which is the core of Chinese unique traditional pharmaceutical technology. The processing includes the preparation steps such as cleansing, cutting and stir-frying, to make certain impacts on the quality and efficacy of Chinese botanical drugs. The rapid development of new computer digital technologies, such as big data analysis, Internet of Things (IoT), blockchain and cloud computing artificial intelligence, has promoted the rapid development of traditional pharmaceutical manufacturing industry with digitalization and intellectualization. In this review, the application of digital intelligence technology in the PCMM was analyzed and discussed, which hopefully promoted the standardization of the process and secured the quality of botanical drugs decoction pieces. Through the intellectualization and the digitization of production, safety and effectiveness of clinical use of traditional Chinese medicine (TCM) decoction pieces were ensured. This review also provided a theoretical basis for further technical upgrading and high-quality development of TCM industry.","author":[{"family":"Zhang","given":"Wanlong"},{"family":"Zhang","given":"Changhua"},{"family":"Cao","given":"Lan"},{"family":"Fang","given":"Liang"},{"family":"Xie","given":"Weihua"},{"family":"Tao","given":"Liang"},{"family":"Chen","given":"Chen"},{"family":"Yang","given":"Ming"},{"family":"Zhong","given":"Lingyun"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3389/fphar.2023.1208055","URL":"https://doi.org/10.3389/fphar.2023.1208055","source":"openalex"},{"id":"oa:W4401214532","type":"article-journal","title":"Enhancing Obstetric Ultrasonography With Artificial Intelligence in Resource-Limited Settings","abstract":"Establishing gestational age (GA) is critical for guiding obstetric care and decision-making (including the timing of prenatal visits, laboratory testing, administration of medications, vaccinations and delivery) as well as neonatal care (including need for newborn resuscitation and neonatal intensive care).Obstetric ultrasonography is the criterion-standard tool for establishing pregnancy viability and dating, as well as for assessing fetal anatomy, growth, and well-being. 1 However, standard ultrasonography examination requires expensive equipment and skilled operators that are often not available in resource-limited settings.In this issue of JAMA, Stringer et al present promising findings from a prospective diagnostic study conducted in Lusaka, Zambia, and Chapel Hill, North Carolina.The study demonstrates that novice users, with no prior obstetrics ultrasonography training, using a low-cost, point-of-care artificial intelligence (AI) tool could estimate GA as accurately as credentialed sonographers using high-specification machines.2 In the study, 400 pregnant patients 18 years or older with a median age of 29 years, median body mass index of 25.9, and singleton, nonanomalous fetuses underwent blind sweeps of the maternal abdomen by novice users employing an AI-enabled device, as well as standard fetal biometry by trained sonographers using standard ultrasonography equipment.The blind sweeps were free-hand sweeps with a 2-dimensional probe approximately 10 seconds in length, including craniocaudal sweeps starting at the pubis symphysis and ending at the xyphoid process and lateral sweeps starting above the pubis and ending between the left and right uterine borders.GA was established by a transvaginal crown-rump length early in pregnancy.Patients were randomized to varying windows of GA for follow-up ultrasonography examination during a primary (14 0/7 to 27 6/7 weeks' gestation), secondary (20 0/7 to 36 6/7 weeks' gestation), and tertiary (37 0/7 to 40 6/7 weeks' gestation) window.The primary outcome measure","author":[{"family":"Gimovsky","given":"Alexis"},{"family":"Eke","given":"Ahizechukwu"},{"family":"Tuuli","given":"Methodius"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1001/jama.2024.14794","URL":"https://doi.org/10.1001/jama.2024.14794","source":"openalex"},{"id":"oa:W4401910776","type":"article-journal","title":"Lung cancer screening: where do we stand?","abstract":"Lung cancer screening (LCS) programmes have emerged over recent years around the world. LCS programmes present differences in delivery, inclusion criteria and resource allocation. On a national scale, only a few LCS programmes have been fully established, but more are anticipated to follow. Evidence has shown that, in combination with a low-dose chest computed tomography scan, smoking cessation should be offered as part of a LCS programme for improved patient outcomes. Promising tools in LCS include further refined risk prediction models, the use of biomarkers, artificial intelligence and radiomics. However, these tools require further study and clinical validation is required prior to routine implementation.","author":[{"family":"Hardavella","given":"Georgia"},{"family":"Frille","given":"Armin"},{"family":"Sreter","given":"Katherina"},{"family":"Atrafi","given":"Florence"},{"family":"Yousaf-Khan","given":"Uraujh"},{"family":"Beyaz","given":"Ferhat"},{"family":"Kyriakou","given":"Fotis"},{"family":"Bellou","given":"Elena"},{"family":"Mullin","given":"Monica"},{"family":"Janes","given":"Sam"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1183/20734735.0190-2023","URL":"https://doi.org/10.1183/20734735.0190-2023","source":"openalex"},{"id":"oa:W4403945659","type":"article-journal","title":"Synthetic Breast Ultrasound Images: A Study to Overcome Medical Data Sharing Barriers","abstract":"The vast potential of medical big data to enhance healthcare outcomes remains underutilized due to privacy concerns, which restrict cross-center data sharing and the construction of diverse, large-scale datasets. To address this challenge, we developed a deep generative model aimed at synthesizing medical data to overcome data sharing barriers, with a focus on breast ultrasound (US) image synthesis. Specifically, we introduce CoLDiT, a conditional latent diffusion model with a transformer backbone, to generate US images of breast lesions across various Breast Imaging Reporting and Data System (BI-RADS) categories. Using a training dataset of 9,705 US images from 5,243 patients across 202 hospitals with diverse US systems, CoLDiT generated breast US images without duplicating private information, as confirmed through nearest-neighbor analysis. Blinded reader studies further validated the realism of these images, with area under the receiver operating characteristic curve (AUC) scores ranging from 0.53 to 0.77. Additionally, synthetic breast US images effectively augmented the training dataset for BI-RADS classification, achieving performance comparable to that using an equal-sized training set comprising solely real images ( P = 0.81 for AUC). Our findings suggest that synthetic data, such as CoLDiT-generated images, offer a viable, privacy-preserving solution to facilitate secure medical data sharing and advance the utilization of medical big data.","author":[{"family":"Xu","given":"Jiale"},{"family":"Hua","given":"Qing"},{"family":"Jia","given":"Xiaohong"},{"family":"Zheng","given":"Yuhang"},{"family":"Hu","given":"Qiao"},{"family":"Bai","given":"Baoyan"},{"family":"Miao","given":"Juan"},{"family":"Zhu","given":"Lisha"},{"family":"Zhang","given":"Meixiang"},{"family":"Tao","given":"Ren"},{"family":"Li","given":"Yuheng"},{"family":"Luo","given":"Ting"},{"family":"Xie","given":"Jun"},{"family":"Zheng","given":"Xuebin"},{"family":"Gu","given":"Pengfei"},{"family":"Xing","given":"Fengyuan"},{"family":"He","given":"Chuan"},{"family":"Song","given":"Yanyan"},{"family":"Dong","given":"Yijie"},{"family":"Xia","given":"Shujun"},{"family":"Zhou","given":"Jianqiao"}],"issued":{"date-parts":[[2024]]},"DOI":"10.34133/research.0532","URL":"https://doi.org/10.34133/research.0532","source":"openalex"},{"id":"oa:W4376642747","type":"article-journal","title":"Fluorescence-guided surgery: comprehensive review","abstract":"BACKGROUND: Despite significant improvements in preoperative workup and surgical planning, surgeons often rely on their eyes and hands during surgery. Although this can be sufficient in some patients, intraoperative guidance is highly desirable. Near-infrared fluorescence has been advocated as a potential technique to guide surgeons during surgery. METHODS: A literature search was conducted to identify relevant articles for fluorescence-guided surgery. The literature search was performed using Medical Subject Headings on PubMed for articles in English until November 2022 and a narrative review undertaken. RESULTS: The use of invisible light, enabling real-time imaging, superior penetration depth, and the possibility to use targeted imaging agents, makes this optical imaging technique increasingly popular. Four main indications are described in this review: tissue perfusion, lymph node assessment, anatomy of vital structures, and tumour tissue imaging. Furthermore, this review provides an overview of future opportunities in the field of fluorescence-guided surgery. CONCLUSION: Fluorescence-guided surgery has proven to be a widely innovative technique applicable in many fields of surgery. The potential indications for its use are diverse and can be combined. The big challenge for the future will be in bringing experimental fluorophores and conjugates through trials and into clinical practice, as well as validation of computer visualization with large data sets. This will require collaborative surgical groups focusing on utility, efficacy, and outcomes for these techniques.","author":[{"family":"Sutton","given":"Paul"},{"family":"Dam","given":"Martijn"},{"family":"Cahill","given":"Ronan"},{"family":"Mieog","given":"JSD"},{"family":"Połom","given":"Karol"},{"family":"Vahrmeijer","given":"Alexander"},{"family":"Vorst","given":"Joost"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1093/bjsopen/zrad049","URL":"https://doi.org/10.1093/bjsopen/zrad049","source":"openalex"},{"id":"oa:W4396955550","type":"article-journal","title":"FI‐Net: Rethinking Feature Interactions for Medical Image Segmentation","abstract":"To solve the problems of existing hybrid networks based on convolutional neural networks (CNN) and Transformers, we propose a new encoder–decoder network FI‐Net based on CNN‐Transformer for medical image segmentation. In the encoder part, a dual‐stream encoder is used to capture local details and long‐range dependencies. Moreover, the attentional feature fusion module is used to perform interactive feature fusion of dual‐branch features, maximizing the retention of local details and global semantic information in medical images. At the same time, the multi‐scale feature aggregation module is used to aggregate local information and capture multi‐scale context to mine more semantic details. The multi‐level feature bridging module is used in skip connections to bridge multi‐level features and mask information to assist multi‐scale feature interaction. Experimental results on seven public medical image datasets fully demonstrate the effectiveness and advancement of our method. In future work, we plan to extend FI‐Net to support 3D medical image segmentation tasks and combine self‐supervised learning and knowledge distillation to alleviate the overfitting problem of limited data training.","author":[{"family":"Ding","given":"Yuhan"},{"family":"Liu","given":"Jinhui"},{"family":"He","given":"Yunbo"},{"family":"Huang","given":"Jinliang"},{"family":"Liang","given":"Haisu"},{"family":"Yi","given":"Zhenglin"},{"family":"Wang","given":"Yongjie"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/aisy.202400201","URL":"https://doi.org/10.1002/aisy.202400201","source":"openalex"},{"id":"oa:W4394758452","type":"article-journal","title":"Segment Anything Is Not Always Perfect: An Investigation of SAM on Different Real-world Applications","abstract":"Abstract Recently, Meta AI Research approaches a general, promptable segment anything model (SAM) pre-trained on an unprecedentedly large segmentation dataset (SA-1B). Without a doubt, the emergence of SAM will yield significant benefits for a wide array of practical image segmentation applications. In this study, we conduct a series of intriguing investigations into the performance of SAM across various applications, particularly in the fields of natural images, agriculture, manufacturing, remote sensing and healthcare. We analyze and discuss the benefits and limitations of SAM, while also presenting an outlook on its future development in segmentation tasks. By doing so, we aim to give a comprehensive understanding of SAM’s practical applications. This work is expected to provide insights that facilitate future research activities toward generic segmentation. Source code is publicly available at https://github.com/LiuTingWed/SAM-Not-Perfect .","author":[{"family":"Ji","given":"Wei"},{"family":"Li","given":"Jingjing"},{"family":"Bi","given":"Qi"},{"family":"Liu","given":"Tingwei"},{"family":"Li","given":"Wenbo"},{"family":"Cheng","given":"Li"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s11633-023-1385-0","URL":"https://doi.org/10.1007/s11633-023-1385-0","source":"openalex"},{"id":"oa:W4384296340","type":"article-journal","title":"Computational Science Role in Medical and Healthcare‐Related Approach","abstract":"Medical imaging, human genome research, clinical diagnosis, and medical data management are all areas where computer technology plays a critical role in modern medicine, healthcare, and life sciences. Computer science solutions will undoubtedly become a vital element of modern medicine and healthcare as computer technology continues to advance. Computational research and applications in medicine and healthcare on modeling, creating, addressing, and assessing fundamental problems are not only vital but also indispensable! With an emphasis on medical imaging, we describe a collection of key computational challenges and methodologies in current medical research, clinical practice, and applications in this session. Computational issues that emerge in medical imaging, clinical diagnosis, therapy, and other medical applications are discussed in detail. These include immune cell identification and distribution analysis, optimal segmentation, and analysis of many medical objects in 3D images, cancer research and analysis, motion tracking of massive swarming bacteria, and so on. In this book chapter, we show how to formulate difficulties as computational problems using new models, as well as how to solve them effectively. Our methods rely on cutting-edge machine learning, data mining, and geometric optimization techniques, particularly novel deep learning models and algorithms. We also offer experimental data and findings to demonstrate how our ideas may be used in clinical settings. Finally, in the exciting new field of computational medicine and healthcare, we highlight several crucial future research themes and issues.","author":[{"family":"Whig","given":"Pawan"},{"family":"Velu","given":"Arun"},{"family":"Nadikattu","given":"Rahul"},{"family":"Alkali","given":"Yusuf"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/9781119763468.ch12","URL":"https://doi.org/10.1002/9781119763468.ch12","source":"openalex"},{"id":"oa:W4385754666","type":"article-journal","title":"Monitoring blood pressure and cardiac function without positioning via a deep learning–assisted strain sensor array","abstract":"Continuous and reliable monitoring of blood pressure and cardiac function is of great importance for diagnosing and preventing cardiovascular diseases. However, existing cardiovascular monitoring approaches are bulky and costly, limiting their wide applications for early diagnosis. Here, we developed an intelligent blood pressure and cardiac function monitoring system based on a conformal and flexible strain sensor array and deep learning neural networks. The sensor has a variety of advantages, including high sensitivity, high linearity, fast response and recovery, and high isotropy. Experiments and simulation synergistically verified that the sensor array can acquire high-precise and feature-rich pulse waves from the wrist without precise positioning. By combining high-quality pulse waves with a well-trained deep learning model, we can monitor blood pressure and cardiac function parameters. As a proof of concept, we further constructed an intelligent wearable system for real-time and long-term monitoring of blood pressure and cardiac function, which may contribute to personalized health management, precise and early diagnosis, and remote treatment.","author":[{"family":"Li","given":"Shuo"},{"family":"Wang","given":"Haomin"},{"family":"Ma","given":"Wei"},{"family":"Qiu","given":"Lin"},{"family":"Xia","given":"Kailun"},{"family":"Zhang","given":"Yong"},{"family":"Zhang","given":"Yong"},{"family":"Lü","given":"Haojie"},{"family":"Zhu","given":"Mengjia"},{"family":"Liang","given":"Xiaoping"},{"family":"Wu","given":"Xun‐en"},{"family":"Liang","given":"Huarun"},{"family":"Zhang","given":"Yingying"},{"family":"Zhang","given":"Yingying"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1126/sciadv.adh0615","URL":"https://doi.org/10.1126/sciadv.adh0615","source":"openalex"},{"id":"oa:W4400095485","type":"article-journal","title":"Comparative Performance of ChatGPT 3.5 and GPT4 on Rhinology Standardized Board Examination Questions","abstract":"Objective: Advances in deep learning and artificial intelligence (AI) have led to the emergence of large language models (LLM) like ChatGPT from OpenAI. The study aimed to evaluate the performance of ChatGPT 3.5 and GPT4 on Otolaryngology (Rhinology) Standardized Board Examination questions in comparison to Otolaryngology residents. Methods: This study selected all 127 rhinology standardized questions from www.boardvitals.com, a commonly used study tool by otolaryngology residents preparing for board exams. Ninety-three text-based questions were administered to ChatGPT 3.5 and GPT4, and their answers were compared with the average results of the question bank (used primarily by otolaryngology residents). Thirty-four image-based questions were provided to GPT4 and underwent the same analysis. Based on the findings of an earlier study, a pass-fail cutoff was set at the 10th percentile. Results: = .001). GPT4 answered image-based questions correctly 64.7% of the time. Projections suggest that ChatGPT 3.5 might not pass the American Board of Otolaryngology Written Question Exam (ABOto WQE), whereas GPT4 stands a strong chance of passing. Discussion: The older LLM, ChatGPT 3.5, is unlikely to pass the ABOto WQE. However, the advanced GPT4 model exhibits a much higher likelihood of success. This rapid progression in AI indicates its potential future role in otolaryngology education. Implications for Practice: As AI technology rapidly advances, it may be that AI-assisted medical education, diagnosis, and treatment planning become commonplace in the medical and surgical landscape. Level of Evidence: Level 5.","author":[{"family":"Patel","given":"Evan"},{"family":"Fleischer","given":"Lindsay"},{"family":"Filip","given":"Peter"},{"family":"Eggerstedt","given":"Michael"},{"family":"Hutz","given":"Michael"},{"family":"Michaelides","given":"Elias"},{"family":"Batra","given":"Pete"},{"family":"Tajudeen","given":"Bobby"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/oto2.164","URL":"https://doi.org/10.1002/oto2.164","source":"openalex"},{"id":"oa:W4403910234","type":"article-journal","title":"A critical analysis of the integration of life cycle methods and quantitative methods for sustainability assessment","abstract":"Abstract The utilization of Life Cycle Assessment (LCA) and its corresponding methodologies gained considerable attention within the realm of corporate social responsibility (CSR) and sustainability assessment. Nevertheless, a lack of extensive investigation on their integration along with quantitative procedures, for example, statistical techniques and artificial intelligence (AI), has emerged. The purpose of this literature review is to investigate the extent to which these methodologies have been connected so far in order to achieve objectives concerning the assessment of sustainability. The scope of the study was restricted to articles published in peer‐reviewed journals throughout the period from 1960 to 2022. The investigation was conducted by using a broad set of keywords, encompassing both life cycle methods, including Life Cycle Assessment, Environmental Life Cycle Costing and Emergy Accounting, and quantitative methods, such as mathematical methods, economic methods and building information modeling methods. A total of 144 articles addressing the combined use of life cycle methods and quantitative methodologies for the evaluation of sustainability were identified in the literature review. The greater part of these studies relied on the combined use of LCA with mathematical models, statistical methods and AI methods. As a result, the studies proven that the joint application of such methods can improve consistency of sustainability assessment and enhance CSR. Additionally, many articles suggested novel approaches, including a combined use of LCA and building information modeling as well as simulation methods. The combined use of life cycle methods alongside quantitative methods offers promise in enhancing sustainability assessment by offering more precise and consistent outcomes. However, it becomes imperative to carefully evaluate the hypotheses, accuracy of data, and uncertainty associated with each method within the integration process. Additional research needs to be conducted in order to establish standardized protocols for combining these methodologies as well as to identify the most suitable procedures for their integration with respect to specific objectives concerning sustainability assessment.","author":[{"family":"Cerchione","given":"Roberto"},{"family":"Morelli","given":"Mariarosaria"},{"family":"Passaro","given":"Renato"},{"family":"Quinto","given":"Ivana"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/csr.3010","URL":"https://doi.org/10.1002/csr.3010","source":"openalex"},{"id":"oa:W4388616875","type":"article-journal","title":"Revolutionizing peptide‐based drug discovery: Advances in the post‐AlphaFold era","abstract":"Peptide-based drugs offer high specificity, potency, and selectivity. However, their inherent flexibility and differences in conformational preferences between their free and bound states create unique challenges that have hindered progress in effective drug discovery pipelines. The emergence of AlphaFold (AF) and Artificial Intelligence (AI) presents new opportunities for enhancing peptide-based drug discovery. We explore recent advancements that facilitate a successful peptide drug discovery pipeline, considering peptides' attractive therapeutic properties and strategies to enhance their stability and bioavailability. AF enables efficient and accurate prediction of peptide-protein structures, addressing a critical requirement in computational drug discovery pipelines. In the post-AF era, we are witnessing rapid progress with the potential to revolutionize peptide-based drug discovery such as the ability to rank peptide binders or classify them as binders/non-binders and the ability to design novel peptide sequences. However, AI-based methods are struggling due to the lack of well-curated datasets, for example to accommodate modified amino acids or unconventional cyclization. Thus, physics-based methods, such as docking or molecular dynamics simulations, continue to hold a complementary role in peptide drug discovery pipelines. Moreover, MD-based tools offer valuable insights into binding mechanisms, as well as the thermodynamic and kinetic properties of complexes. As we navigate this evolving landscape, a synergistic integration of AI and physics-based methods holds the promise of reshaping the landscape of peptide-based drug discovery.","author":[{"family":"Chang","given":"Liwei"},{"family":"Mondal","given":"Arup"},{"family":"Singh","given":"Bhumika"},{"family":"Martíneznoa","given":"Yisel"},{"family":"Pérez","given":"Alberto"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/wcms.1693","URL":"https://doi.org/10.1002/wcms.1693","source":"openalex"},{"id":"oa:W4389179720","type":"article-journal","title":"Blazing the trail for innovative tuberculosis diagnostics","abstract":"The COVID-19 pandemic brought diagnostics into the spotlight in an unprecedented way not only for case management but also for population health, surveillance, and monitoring. The industry saw notable levels of investment and accelerated research which sparked a wave of innovation. Simple non-invasive sampling methods such as nasal swabs have become widely used in settings ranging from tertiary hospitals to the community. Self-testing has also been adopted as standard practice using not only conventional lateral flow tests but novel and affordable point-of-care molecular diagnostics. The use of new technologies, including artificial intelligence-based diagnostics, have rapidly expanded in the clinical setting. The capacity for next-generation sequencing and acceptance of digital health has significantly increased. However, 4 years after the pandemic started, the market for SARS-CoV-2 tests is saturated, and developers may benefit from leveraging their innovations for other diseases; tuberculosis (TB) is a worthwhile portfolio expansion for diagnostics developers given the extremely high disease burden, supportive environment from not-for-profit initiatives and governments, and the urgent need to overcome the long-standing dearth of innovation in the TB diagnostics field. In exchange, the current challenges in TB detection may be resolved by adopting enhanced swab-based molecular methods, instrument-based, higher sensitivity antigen detection technologies, and/or artificial intelligence-based digital health technologies developed for COVID-19. The aim of this article is to review how such innovative approaches for COVID-19 diagnosis can be applied to TB to have a comparable impact.","author":[{"family":"Yerlikaya","given":"Seda"},{"family":"Broger","given":"Tobias"},{"family":"Isaacs","given":"Chris"},{"family":"Bell","given":"David"},{"family":"Holtgrewe","given":"Lydia"},{"family":"Guptawright","given":"Ankur"},{"family":"Nahid","given":"Payam"},{"family":"Cattamanchi","given":"Adithya"},{"family":"Denkinger","given":"Claudia"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1007/s15010-023-02135-3","URL":"https://doi.org/10.1007/s15010-023-02135-3","source":"openalex"},{"id":"oa:W4381469976","type":"article-journal","title":"Fundus Tessellated Density Assessed by Deep Learning in Primary School Children","abstract":"Purpose: To explore associations of fundus tessellated density (FTD) and compare characteristics of different fundus tessellation (FT) distribution patterns, based on artificial intelligence technology using deep learning. Methods: Comprehensive ocular examinations were conducted in 577 children aged 7 years old from a population-based cross-sectional study, including biometric measurement, refraction, optical coherence tomography angiography, and 45° nonmydriatic fundus photography. FTD was defined as the average exposed choroid area per unit area of the fundus, and obtained by artificial intelligence technology. The distribution of FT was classified into the macular pattern and the peripapillary pattern according to FTD. Results: The mean FTD was 0.024 ± 0.026 in whole fundus. Multivariate regression analysis showed that greater FTD was significantly correlated with thinner subfoveal choroidal thickness, larger parapapillary atrophy, greater vessel density inside the optic disc, larger vertical diameter of optic disc, thinner retinal nerve fiber layer, and longer distance from optic disc center to macular fovea (all P < 0.05). The peripapillary distributed group had larger parapapillary atrophy (0.052 ± 0.119 vs 0.031 ± 0.072), greater FTD (0.029 ± 0.028 vs 0.015 ± 0.018), thinner subfoveal choroidal thickness (297.66 ± 60.61 vs 315.33 ± 66.46), and thinner retinal thickness (285.55 ± 10.89 vs 288.03 ± 10.31) than the macular distributed group (all P < 0.05). Conclusions: FTD can be applied as a quantitative biomarker to estimate subfoveal choroidal thickness in children. The role of blood flow inside optic disc in FT progression needs further investigation. The distribution of FT and the peripapillary pattern correlated more with myopia-related fundus changes than the macular pattern. Translational Relevance: Artificial intelligence can evaluate FT quantitatively in children, and has potential value for assisting in myopia prevention and control.","author":[{"family":"Huang","given":"Dan"},{"family":"Li","given":"Rui"},{"family":"Qian","given":"Yingxiao"},{"family":"Ling","given":"Saiguang"},{"family":"Dong","given":"Zhou"},{"family":"Ke","given":"Xin"},{"family":"Yan","given":"Qi"},{"family":"Tong","given":"Haohai"},{"family":"Wang","given":"Zijin"},{"family":"Long","given":"Tengfei"},{"family":"Liu","given":"Hu"},{"family":"Zhu","given":"Hui"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1167/tvst.12.6.11","URL":"https://doi.org/10.1167/tvst.12.6.11","source":"openalex"},{"id":"oa:W4402573797","type":"article-journal","title":"Antiviral Effectiveness, Clinical Outcomes, and Artificial Intelligence Imaging Analysis for Hospitalized COVID‐19 Patients Receiving Antivirals","abstract":"INTRODUCTION: There is still a lack of clinical evidence comprehensively evaluating the effectiveness of antiviral treatments for COVID-19 hospitalized patients. METHODS: A retrospective cohort study was conducted at Beijing You'An Hospital, focusing on patients treated with nirmatrelvir/ritonavir or azvudine. The study employed a tripartite analysis-viral dynamics, survival curve analysis, and AI-based radiological analysis of pulmonary CT images-aiming to assess the severity of pneumonia. RESULTS: Of 370 patients treated with either nirmatrelvir/ritonavir or azvudine as monotherapy, those in the nirmatrelvir/ritonavir group experienced faster viral clearance than those treated with azvudine (5.4 days vs. 8.4 days, p < 0.001). No significant differences were observed in the survival curves between the two drug groups. AI-based radiological analysis revealed that patients in the nirmatrelvir group had more severe pneumonia conditions (infection ratio is 11.1 vs. 5.35, p = 0.007). Patients with an infection ratio higher than 9.2 had nearly three times the mortality rate compared to those with an infection ratio lower than 9.2. CONCLUSIONS: Our study suggests that in real-world studies regarding hospitalized patients with COVID-19 pneumonia, the antiviral effect of nirmatrelvir/ritonavir is significantly superior to azvudine, but the choice of antiviral agents is not necessarily linked to clinical outcomes; the severity of pneumonia at admission is the most important factor to determine prognosis. Additionally, our findings indicate that pulmonary AI imaging analysis can be a powerful tool for predicting patient prognosis and guiding clinical decision-making.","author":[{"family":"Gao","given":"Yuan"},{"family":"Dong","given":"Yixi"},{"family":"Bu","given":"Qiushi"},{"family":"Gong","given":"Zhijie"},{"family":"Wang","given":"Wei"},{"family":"Zhou","given":"Zhongkai"},{"family":"Gao","given":"Yunyi"},{"family":"Liu","given":"Liwei"},{"family":"Wu","given":"Menghua"},{"family":"Zhang","given":"Jiaying"},{"family":"Liang","given":"Lianchun"},{"family":"Li","given":"Hongjun"},{"family":"Jiang","given":"Mengxi"},{"family":"Luo","given":"Zujin"},{"family":"Ma","given":"Yingmin"},{"family":"Zhang","given":"Xinyu"},{"family":"Hu","given":"Zhongjie"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/irv.70006","URL":"https://doi.org/10.1111/irv.70006","source":"openalex"},{"id":"oa:W4403218073","type":"article-journal","title":"Legal provisions on medical aid in dying encode moral intuition","abstract":"In recent decades, many jurisdictions have moved toward legalizing euthanasia and assisted suicide, alongside a near-universal increase in public acceptance of medical aid in dying. Here, we draw on a comprehensive quantitative review of current laws on assisted dying, experimental survey evidence, and four decades of time-series data to explore the relationship between these legislative transitions and change in moral attitudes. Our analyses reveal that existing laws on medical aid in dying impose a common set of eligibility restrictions, based on the patient's age, decision-making capacity, prognosis, and the nature of their illness. Fulfillment of these eligibility criteria elevates public moral approval of physician-assisted death, equally in countries with (i.e., Spain) and without (i.e., the United Kingdom) assisted dying laws. Finally, historical records of public attitudes toward euthanasia across numerous countries uncovered anticipatory growth in moral approval leading up to legalization, but no accelerated growth thereafter. Taken together, our findings suggest that the enactment of medical aid in dying laws, and their specific provisions, crystallize patterns in moral intuition.","author":[{"family":"Hannikainen","given":"Ivar"},{"family":"Suárez","given":"Jorge"},{"family":"Espericueta","given":"Luis"},{"family":"Menéndez-Ferreras","given":"Maite"},{"family":"Rodríguezarias","given":"David"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1073/pnas.2406823121","URL":"https://doi.org/10.1073/pnas.2406823121","source":"openalex"},{"id":"oa:W4389369785","type":"article-journal","title":"Cancer Informatics for Cancer Centers: Sharing Ideas on How to Build an Artificial Intelligence–Ready Informatics Ecosystem for Radiation Oncology","abstract":"In August 2022, the Cancer Informatics for Cancer Centers brought together cancer informatics leaders for its biannual symposium, Precision Medicine Applications in Radiation Oncology, co-chaired by Quynh-Thu Le, MD (Stanford University), and Walter J. Curran, MD (GenesisCare). Over the course of 3 days, presenters discussed a range of topics relevant to radiation oncology and the cancer informatics community more broadly, including biomarker development, decision support algorithms, novel imaging tools, theranostics, and artificial intelligence (AI) for the radiotherapy workflow. Since the symposium, there has been an impressive shift in the promise and potential for integration of AI in clinical care, accelerated in large part by major advances in generative AI. AI is now poised more than ever to revolutionize cancer care. Radiation oncology is a field that uses and generates a large amount of digital data and is therefore likely to be one of the first fields to be transformed by AI. As experts in the collection, management, and analysis of these data, the informatics community will take a leading role in ensuring that radiation oncology is prepared to take full advantage of these technological advances. In this report, we provide highlights from the symposium, which took place in Santa Barbara, California, from August 29 to 31, 2022. We discuss lessons learned from the symposium for data acquisition, management, representation, and sharing, and put these themes into context to prepare radiation oncology for the successful and safe integration of AI and informatics technologies.","author":[{"family":"Bitterman","given":"Danielle"},{"family":"Gensheimer","given":"Michael"},{"family":"Jaffray","given":"David"},{"family":"Pryma","given":"Daniel"},{"family":"Jiang","given":"Steve"},{"family":"Morin","given":"Olivier"},{"family":"Ginart","given":"Jorge"},{"family":"Upadhaya","given":"Taman"},{"family":"Vallis","given":"Katherine"},{"family":"Buatti","given":"John"},{"family":"Deasy","given":"Joseph"},{"family":"Hsiao","given":"Hsu‐chin"},{"family":"Chung","given":"Caroline"},{"family":"Fuller","given":"Clifton"},{"family":"Greenspan","given":"Emily"},{"family":"Cloyd-Warwick","given":"Kristy"},{"family":"Courdy","given":"Samir"},{"family":"Mao","given":"Allen"},{"family":"Barnholtzsloan","given":"Jill"},{"family":"Topaloĝlu","given":"Ümit"},{"family":"Hands","given":"Isaac"},{"family":"Maurer","given":"Ian"},{"family":"Terry","given":"May"},{"family":"Curran","given":"Walter"},{"family":"Le","given":"Quynh‐thu"},{"family":"Nadaf","given":"Sorena"},{"family":"Kibbe","given":"Warren"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1200/cci.23.00136","URL":"https://doi.org/10.1200/cci.23.00136","source":"openalex"},{"id":"oa:W4387369338","type":"article-journal","title":"Electrochemical methods for the determination of urea: Current trends and future perspective","abstract":"Urea is the final product of nitrogen metabolism in mammals. In human beings, it is a main indicator of liver and kidney activity and a marker for hemodialysis treatments. Furthermore, urea is used in many industrial processes, in agriculture and in the farm industries. Thus, it is important to evaluate the level of this compound in biological fluids, in environmental matrices and in food samples. Electrochemical sensors represent an interesting tool for simple, rapid, in-situ, in-flow monitoring of urea. This review focuses on the recent advancements in electrochemical sensors and biosensors for urea determination. An overview of the existing electroanalytical approaches for urea determination is presented, and some new strategies are discussed, particularly those based on nanostructuration of the electrode surface. Finally, a brief description of the role that Artificial Intelligence could have in overcoming selectivity issues and speed of the analysis is also discussed.","author":[{"family":"Quadrini","given":"Lorenzo"},{"family":"Laschi","given":"Serena"},{"family":"Ciccone","given":"Claudio"},{"family":"Catelani","given":"Filippo"},{"family":"Palchetti","given":"Ilaria"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.trac.2023.117345","URL":"https://doi.org/10.1016/j.trac.2023.117345","source":"openalex"},{"id":"oa:W4388818010","type":"article-journal","title":"The Shape of Medical Devices Regulation in the United Kingdom? Brexit and Beyond","abstract":"The United Kingdom’s Medicines and Medical Devices Act (MMD Act) 2021 received royal assent on 11 February 2021. In its passage through parliament, as well as in the accompanying Explanatory Notes, the Act was framed by the government as a necessary post-Brexit bill. Yet prior to this, it was widely presumed that existing statutory instruments, enacted in 2019 to address medical devices regulation in anticipation of a ‘No Deal Brexit’, would provide the United Kingdom (UK) with the necessary legal framework through the transition period and beyond. The European Union (EU) exit legislation included provisions aimed at aligning domestic law with the new EU Medical Devices Regulation (EU MDR) and In Vitro Devices Regulation (EU IVDR) (which were initially due to be implemented during the EU exit transition period). However, just over a year later, at the end of 2020, while the 2021 Act was in the final stages of its parliamentary journey, legislation was introduced that reversed the provisions of the domestic medical device regulations concerning alignment for Great Britain (GB: England, Wales, and Scotland). The result is that the UK now has a dual system of regulation, with Northern Ireland (NI) being governed by the new EU MDR and EU IVDR and GB by older pre-Brexit (yet still EU-derived) law. This article is an attempt to analyse the complex situation now in existence regarding the regulation of medical devices in the UK. To this end, we focus on three main issues. First, we examine the difficulties and challenges presented by the dual system of regulation between NI and GB, highlighting the potentially far-reaching consequences of regulatory divergence between different UK jurisdictions. Second, we ask whether, in the rush to new legislation in 2021, opportunities to properly reform the approach to medical devices were missed. Finally, we look to the future, focusing on the recent Medicine and Healthcare products Regulatory Agency Consultation on the future of medical devices regulation in the UK and the challenges and opportunities that remain.","author":[{"family":"Quigley","given":"Muireann"},{"family":"Downey","given":"Laura"},{"family":"Mahmoud","given":"Zaina"},{"family":"Mchale","given":"Jean"}],"issued":{"date-parts":[[2023]]},"DOI":"10.5204/lthj.3102","URL":"https://doi.org/10.5204/lthj.3102","source":"openalex"},{"id":"oa:W4399594884","type":"article-journal","title":"When geoscience meets generative AI and large language models: Foundations, trends, and future challenges","abstract":"Abstract Generative Artificial Intelligence (GAI) represents an emerging field that promises the creation of synthetic data and outputs in different modalities. GAI has recently shown impressive results across a large spectrum of applications ranging from biology, medicine, education, legislation, computer science, and finance. As one strives for enhanced safety, efficiency, and sustainability, generative AI indeed emerges as a key differentiator and promises a paradigm shift in the field. This article explores the potential applications of generative AI and large language models in geoscience. The recent developments in the field of machine learning and deep learning have enabled the generative model's utility for tackling diverse prediction problems, simulation, and multi‐criteria decision‐making challenges related to geoscience and Earth system dynamics. This survey discusses several GAI models that have been used in geoscience comprising generative adversarial networks (GANs), physics‐informed neural networks (PINNs), and generative pre‐trained transformer (GPT)‐based structures. These tools have helped the geoscience community in several applications, including (but not limited to) data generation/augmentation, super‐resolution, panchromatic sharpening, haze removal, restoration, and land surface changing. Some challenges still remain, such as ensuring physical interpretation, nefarious use cases, and trustworthiness. Beyond that, GAI models show promises to the geoscience community, especially with the support to climate change, urban science, atmospheric science, marine science, and planetary science through their extraordinary ability to data‐driven modelling and uncertainty quantification.","author":[{"family":"Hadid","given":"Abdenour"},{"family":"Chakraborty","given":"Tanujit"},{"family":"Busby","given":"D"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/exsy.13654","URL":"https://doi.org/10.1111/exsy.13654","source":"openalex"},{"id":"oa:W4405949375","type":"article-journal","title":"Assessing ChatGPT responses to common patient questions regarding total ankle arthroplasty","abstract":"Purpose: Artificial Intelligence is becoming increasingly integrated into healthcare, making it essential to assess its potential as a reliable information source for patient queries in the ambit of orthopaedic surgery. In literature, it is being employed in foot and ankle surgery and total hip arthroplasty. The aim of the present study was to evaluate the ability of Chat Generative Pretrained Transformer (ChatGPT) version 3.5 to give accurate, complete and comprehensive responses to the most common questions which are usually asked by the patient to the surgeon regarding total ankle arthroplasty. Methods: Ten most common questions were selected by two ankle surgeons and then ChatGPT was used to answer these questions. The responses were analyzed using an accuracy score and the modified DISCERN score to assess clarity. WordCalc software package (educational-level indices) was used to assess the readability of the responses. Results: Most of ChatGPT's responses were considered excellent not requiring clarification or satisfactory requiring minimal clarification. Indeed, the accuracy score was 2, suggesting that the overall responses were satisfactory requiring minimal clarification, and DISCERN score mean was 51, which is considered good-fair. Conclusions: ChatGPT demonstrates potential as a tool for responding to common patient questions related to total ankle arthroplasty, offering clear and mostly accurate information. While its current performance is based on the available literature, ongoing advancements in artificial intelligence may further enhance its utility in healthcare communication. However, further studies are required to evaluate its role more precisely in patient information and clinical settings. Levels of Evidence: Not applicable.","author":[{"family":"Artioli","given":"Elena"},{"family":"Veronesi","given":"Francesca"},{"family":"Mazzotti","given":"Antonio"},{"family":"Brogini","given":"Silvia"},{"family":"Zielli","given":"Simone"},{"family":"Giavaresi","given":"Gianluca"},{"family":"Faldini","given":"Cesare"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/jeo2.70138","URL":"https://doi.org/10.1002/jeo2.70138","source":"openalex"},{"id":"oa:W4383873008","type":"article-journal","title":"Global, regional, and national burden of allergic disorders and their risk factors in 204 countries and territories, from 1990 to 2019: A systematic analysis for the Global Burden of Disease Study 2019","abstract":"BACKGROUND: Asthma and atopic dermatitis (AD) are chronic allergic conditions, along with allergic rhinitis and food allergy and cause high morbidity and mortality both in children and adults. This study aims to evaluate the global, regional, national, and temporal trends of the burden of asthma and AD from 1990 to 2019 and analyze their associations with geographic, demographic, social, and clinical factors. METHODS: Using data from the Global Burden of Diseases (GBD), Injuries, and Risk Factors Study 2019, we assessed the age-standardized prevalence, incidence, mortality, and disability-adjusted life years (DALYs) of both asthma and AD from 1990 to 2019, stratified by geographic region, age, sex, and socio-demographic index (SDI). DALYs were calculated as the sum of years lived with disability and years of life lost to premature mortality. Additionally, the disease burden of asthma attributable to high body mass index, occupational asthmagens, and smoking was described. RESULTS: In 2019, there were a total of 262 million [95% uncertainty interval (UI): 224-309 million] cases of asthma and 171 million [95% UI: 165-178 million] total cases of AD globally; age-standardized prevalence rates were 3416 [95% UI: 2899-4066] and 2277 [95% UI: 2192-2369] per 100,000 population for asthma and AD, respectively, a 24.1% [95% UI: -27.2 to -20.8] decrease for asthma and a 4.3% [95% UI: 3.8-4.8] decrease for AD compared to baseline in 1990. Both asthma and AD had similar trends according to age, with age-specific prevalence rates peaking at age 5-9 years and rising again in adulthood. The prevalence and incidence of asthma and AD were both higher for individuals with higher SDI; however, mortality and DALYs rates of individuals with asthma had a reverse trend, with higher mortality and DALYs rates in those in the lower SDI quintiles. Of the three risk factors, high body mass index contributed to the highest DALYs and deaths due to asthma, accounting for a total of 3.65 million [95% UI: 2.14-5.60 million] asthma DALYs and 75,377 [95% UI: 40,615-122,841] asthma deaths. CONCLUSIONS: Asthma and AD continue to cause significant morbidity worldwide, having increased in total prevalence and incidence cases worldwide, but having decreased in age-standardized prevalence rates from 1990 to 2019. Although both are more frequent at younger ages and more prevalent in high-SDI countries, each condition has distinct temporal and regional characteristics. Understanding the temporospatial trends in the disease burden of asthma and AD could guide future policies and interventions to better manage these diseases worldwide and achieve equity in prevention, diagnosis, and treatment.","author":[{"family":"Shin","given":"Youn"},{"family":"Hwang","given":"Jimin"},{"family":"Kwon","given":"Rosie"},{"family":"Lee","given":"Seung"},{"family":"Kim","given":"Min"},{"family":"Shin","given":"Jae"},{"family":"Yon","given":"Dong"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1111/all.15807","URL":"https://doi.org/10.1111/all.15807","source":"openalex"},{"id":"oa:W4401686380","type":"article-journal","title":"Integrative approach of omics and imaging data to discover new insights for understanding brain diseases","abstract":"Treatments that can completely resolve brain diseases have yet to be discovered. Omics is a novel technology that allows researchers to understand the molecular pathways underlying brain diseases. Multiple omics, including genomics, transcriptomics and proteomics, and brain imaging technologies, such as MRI, PET and EEG, have contributed to brain disease-related therapeutic target detection. However, new treatment discovery remains challenging. We focused on establishing brain multi-molecular maps using an integrative approach of omics and imaging to provide insights into brain disease diagnosis and treatment. This approach requires precise data collection using omics and imaging technologies, data processing and normalization. Incorporating a brain molecular map with the advanced technologies through artificial intelligence will help establish a system for brain disease diagnosis and treatment through regulation at the molecular level.","author":[{"family":"Yoon","given":"Jong"},{"family":"Lee","given":"Hagyeong"},{"family":"Kwon","given":"Dayoung"},{"family":"Lee","given":"Dongha"},{"family":"Lee","given":"Seulah"},{"family":"Cho","given":"Eunji"},{"family":"Kim","given":"Jae‐hoon"},{"family":"Kim","given":"Dayea"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1093/braincomms/fcae265","URL":"https://doi.org/10.1093/braincomms/fcae265","source":"openalex"},{"id":"oa:W4396642986","type":"article-journal","title":"Artificial neural network for enhancing signal-to-noise ratio and contrast in photothermal optical coherence tomography","abstract":"Optical coherence tomography (OCT) is a medical imaging method that generates micron-resolution 3D volumetric images of tissues in-vivo. Photothermal (PT)-OCT is a functional extension of OCT with the potential to provide depth-resolved molecular information complementary to the OCT structural images. PT-OCT typically requires long acquisition times to measure small fluctuations in the OCT phase signal. Here, we use machine learning with a neural network to infer the amplitude of the photothermal phase modulation from a short signal trace, trained in a supervised fashion with the ground truth signal obtained by conventional reconstruction of the PT-OCT signal from a longer acquisition trace. Results from phantom and tissue studies show that the developed network improves signal to noise ratio (SNR) and contrast, enabling PT-OCT imaging with short acquisition times and without any hardware modification to the PT-OCT system. The developed network removes one of the key barriers in translation of PT-OCT (i.e., long acquisition time) to the clinic.","author":[{"family":"Salimi","given":"Mohammadhossein"},{"family":"Tabatabaei","given":"Nima"},{"family":"Villiger","given":"Martin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1038/s41598-024-60682-7","URL":"https://doi.org/10.1038/s41598-024-60682-7","source":"openalex"},{"id":"oa:W4400644218","type":"article-journal","title":"A Survey on Symbolic Knowledge Distillation of Large Language Models","abstract":"This survey article delves into the emerging and critical area of symbolic knowledge distillation in large language models (LLMs). As LLMs such as generative pretrained transformer-3 (GPT-3) and bidirectional encoder representations from transformers (BERT) continue to expand in scale and complexity, the challenge of effectively harnessing their extensive knowledge becomes paramount. This survey concentrates on the process of distilling the intricate, often implicit knowledge contained within these models into a more symbolic, explicit form. This transformation is crucial for enhancing the interpretability, efficiency, and applicability of LLMs. We categorize the existing research based on methodologies and applications, focusing on how symbolic knowledge distillation can be used to improve the transparency and functionality of smaller, more efficient artificial intelligence (AI) models. The survey discusses the core challenges, including maintaining the depth of knowledge in a comprehensible format, and explores the various approaches and techniques that have been developed in this field. We identify gaps in current research and potential opportunities for future advancements. This survey aims to provide a comprehensive overview of symbolic knowledge distillation in LLMs, spotlighting its significance in the progression toward more accessible and efficient AI systems.","author":[{"family":"Acharya","given":"K"},{"family":"Velasquez","given":"Alvaro"},{"family":"Song","given":"Houbing"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1109/tai.2024.3428519","URL":"https://doi.org/10.1109/tai.2024.3428519","source":"openalex"},{"id":"oa:W4401716126","type":"article-journal","title":"Utilizing natural language processing to analyze student narrative reflections for medical curriculum improvement","abstract":"MOTIVATION: Medical curricula improvement is an ongoing process to keep material relevant and improve the student's learning experience to better prepare them for patient care. Many programs utilize end-of-year evaluations, but these frequently have low response rates and lack actionable feedback. We hypothesized that student reflections written during a fourth year Sub-Internship could be used retrospectively to mine additional information as feedback for future curriculum adjustments. However, reflections contain a large amount of narrative content that would require a cumbersome and essentially infeasible manual review process for busy medical education faculty. METHODS: We developed a Natural Language Processing (NLP) pipeline to automatically identify common themes and topics present in the set of reflective writings that could be used to improve the curriculum. The dataset contains required responses to a faculty issued question submitted between August 2016 and July 2018 about challenges experienced during the medical students fourth year Sub-Internship. RESULTS: Eleven distinct topics were identified, with several being subsequently addressed in future iterations of the curriculum. CONCLUSION: Utilizing NLP on reflective writings was able to identify areas of curriculum improvement, and the NLP results provided a quick and easy way to explore the main themes and challenges expressed by students.","author":[{"family":"Olex","given":"Amy"},{"family":"Garber","given":"Adam"},{"family":"Santen","given":"Sally"},{"family":"Blondino","given":"Courtney"},{"family":"Goldberg","given":"Stephanie"},{"family":"Diazgranados","given":"Deborah"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1080/0142159x.2024.2390034","URL":"https://doi.org/10.1080/0142159x.2024.2390034","source":"openalex"},{"id":"oa:W4409249730","type":"article-journal","title":"Strategic Implementation of Artificial Intelligence and Machine Learning in the Pharmaceutical Sector: A Dual Hesitant Fuzzy Group Decision Making Approach","abstract":"The pharmaceutical industry is regarded as one of the leading industries in Bangladesh. Further developments and sustainability can be offered through the integration of Artificial Intelligence (AI) and Machine Learning (ML) in this industry as vast amounts of data can quickly be processed and analyzed with accuracy by AI which is crucial for drug discovery, and quality improvement of medicines within minimal costs. However, successful AI implementation requires strategic planning based on various factors. In this study, eight major strategies were proposed to implement and integrate AI and machine learning and also assessed based on eight criteria related to artificial intelligence and machine learning. Feedbacks were taken from eight industry experts as Dual Hesitant Fuzzy Elements (DHFEs) to build the strategy versus criteria matrices. The feedbacks were then aggregated by using a group decision making method: Dual Hesitant Fuzzy Heronian Mean (DHFHM). The scores of those aggregated DHFEs belonging to each criterion were determined and the scores of all criteria under a strategy were summed up to find the ultimate score of that corresponding strategy. The strategies were then ranked based on their scores. Finally, the robustness of the result obtained from DHFHM, sensitivity analysis was conducted by changing the values of parameters and it was found that similar results with negligible deviation were generated by the operator and through this the proposed model was proven to be validated.","author":[{"family":"Ashaduzzaman","given":"Md"},{"family":"Nawar","given":"Nahiyan"},{"family":"Khan","given":"Shakil"},{"family":"Biswas","given":"Atif"},{"family":"Islam","given":"Md"}],"issued":{"date-parts":[[2024]]},"DOI":"10.46254/ba07.20240185","URL":"https://doi.org/10.46254/ba07.20240185","source":"openalex"},{"id":"oa:W4386151807","type":"article-journal","title":"BPPV Information on Google Versus AI (ChatGPT)","abstract":"OBJECTIVE: To quantitatively compare online patient education materials found using traditional search engines (Google) versus conversational Artificial Intelligence (AI) models (ChatGPT) for benign paroxysmal positional vertigo (BPPV). STUDY DESIGN: The top 30 Google search results for \"benign paroxysmal positional vertigo\" were compared to the OpenAI conversational AI language model, ChatGPT, responses for 5 common patient questions posed about BPPV in February 2023. Metrics included readability, quality, understandability, and actionability. SETTING: Online information. METHODS: Validated online information metrics including Flesch-Kincaid Grade Level (FKGL), Flesch Reading Ease (FRE), DISCERN instrument score, and Patient Education Materials Assessment Tool for Printed Materials were analyzed and scored by reviewers. RESULTS: Mean readability scores, FKGL and FRE, for the Google webpages were 10.7 ± 2.6 and 46.5 ± 14.3, respectively. ChatGPT responses had a higher FKGL score of 13.9 ± 2.5 (P < .001) and a lower FRE score of 34.9 ± 11.2 (P = .005), both corresponding to lower readability. The Google webpages had a DISCERN part 2 score of 25.4 ± 7.5 compared to the individual ChatGPT responses with a score of 17.5 ± 3.9 (P = .001), and the combined ChatGPT responses with a score of 25.0 ± 0.9 (P = .928). The average scores of the reviewers for all ChatGPT responses for accuracy were 4.19 ± 0.82 and 4.31 ± 0.67 for currency. CONCLUSION: The results of this study suggest that the information on ChatGPT is more difficult to read, of lower quality, and more difficult to comprehend compared to information on Google searches.","author":[{"family":"Bellinger","given":"Jeffrey"},{"family":"Chapa","given":"Julian"},{"family":"Kwak","given":"Minhie"},{"family":"Ramos","given":"Gabriel"},{"family":"Morrison","given":"Daniel"},{"family":"Kesser","given":"Bradley"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/ohn.506","URL":"https://doi.org/10.1002/ohn.506","source":"openalex"},{"id":"oa:W4403249102","type":"article-journal","title":"Wearable Devices and Health Monitoring","abstract":"The integration of wearable devices, big data, and artificial intelligence (AI) has revolutionized remote patient care and health monitoring. These technologies enable the continuous and real-time collection of various physiological and behavioral data from individuals, allowing healthcare providers to remotely monitor their health status. Wearable devices such as smartwatches, fitness trackers, and medical sensors offer the ability to track vital signs, physical activity, and sleep patterns. The sheer volume of data generated by these devices, often referred to as big data, presents challenges and opportunities. AI plays a crucial role in deciphering this wealth of information by employing advanced algorithms for data analysis, pattern recognition, and predictive modeling. This synergy between wearable devices, big data, and AI not only empowers patients to actively engage in their own health management, but also enables healthcare professionals to make informed decisions, detect anomalies, and provide timely interventions. This abstract explores the transformative impact of wearable devices and the synergy of big data and AI in enhancing remote patient care, fostering the early detection of health issues, and ultimately improving overall patient outcomes.","author":[{"family":"Kanakaprabha","given":"S"},{"family":"Kumar","given":"GG"},{"family":"Reddy","given":"Bhargavi"},{"family":"Raju","given":"Yallapragada"},{"family":"Mohan","given":"PC"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1002/9781394270910.ch12","URL":"https://doi.org/10.1002/9781394270910.ch12","source":"openalex"},{"id":"oa:W4393002303","type":"article-journal","title":"On the Opportunities and Challenges of Foundation Models for GeoAI (Vision Paper)","abstract":"Large pre-trained models, also known as foundation models (FMs), are trained in a task-agnostic manner on large-scale data and can be adapted to a wide range of downstream tasks by fine-tuning, few-shot, or even zero-shot learning. Despite their successes in language and vision tasks, we have not yet seen an attempt to develop foundation models for geospatial artificial intelligence (GeoAI). In this work, we explore the promises and challenges of developing multimodal foundation models for GeoAI. We first investigate the potential of many existing FMs by testing their performances on seven tasks across multiple geospatial domains, including Geospatial Semantics, Health Geography, Urban Geography, and Remote Sensing. Our results indicate that on several geospatial tasks that only involve text modality, such as toponym recognition, location description recognition, and US state-level/county-level dementia time series forecasting, the task-agnostic large learning models (LLMs) can outperform task-specific fully supervised models in a zero-shot or few-shot learning setting. However, on other geospatial tasks, especially tasks that involve multiple data modalities (e.g., POI-based urban function classification, street view image–based urban noise intensity classification, and remote sensing image scene classification), existing FMs still underperform task-specific models. Based on these observations, we propose that one of the major challenges of developing an FM for GeoAI is to address the multimodal nature of geospatial tasks. After discussing the distinct challenges of each geospatial data modality, we suggest the possibility of a multimodal FM that can reason over various types of geospatial data through geospatial alignments. We conclude this article by discussing the unique risks and challenges to developing such a model for GeoAI.","author":[{"family":"Mai","given":"Gengchen"},{"family":"Huang","given":"Weiming"},{"family":"Sun","given":"Jin"},{"family":"Song","given":"Suhang"},{"family":"Mishra","given":"Deepak"},{"family":"Liu","given":"Ninghao"},{"family":"Gao","given":"Song"},{"family":"Liu","given":"Tianming"},{"family":"Cong","given":"Gao"},{"family":"Hu","given":"Yingjie"},{"family":"Cundy","given":"Chris"},{"family":"Li","given":"Ziyuan"},{"family":"Zhu","given":"Rui"},{"family":"Lao","given":"Ni"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1145/3653070","URL":"https://doi.org/10.1145/3653070","source":"openalex"},{"id":"oa:W4327892919","type":"article-journal","title":"The intelligent experience inheritance system for Traditional Chinese Medicine","abstract":"The inheritance of knowledge and experience was crucial to the development of Traditional Chinese Medicine (TCM). However, the existing methods of inheriting the unique clinical experience of famous veteran TCM doctors still followed the outdated and inefficient Master-Prentice schema. In addition, the inherited medical books and records were usually lack of standardization and systematization. In this article, a new method for inheriting the academic thoughts and clinical experience of famous veteran doctors with the help of artificial intelligence technology was explored. Due to the individualized treatment characteristics namely \"same disease with different treatments, different diseases with the same treatment,\" the intelligent inheritance of TCM faced many technical barriers. To tackle these problems, we proposed a prototype system framework for the intelligent inheritance of famous veteran doctors based on rules and deep learning models and performed a case study on the treatment of pediatric asthma. The architecture could not only make full use of the advantages of deep learning, but also integrate the valuable knowledge and experience analysis of famous veteran doctors from injected rules. Specifically, the study took pediatric asthma medical records as training and test samples and calculated the similarity between the generated prescriptions and the real-world clinical prescriptions from the famous veteran doctors. Experimental results showed that the generated prescription could achieve a similarity of more than 90%. It proved that the proposed framework provided a feasible way for the intelligent inheritance and research of the academic thoughts and clinical experience of famous veteran TCM doctors.","author":[{"family":"Ren","given":"Xue"},{"family":"Guo","given":"Yan"},{"family":"Wang","given":"Heyuan"},{"family":"Gao","given":"Xiang"},{"family":"Chen","given":"Wei"},{"family":"Wang","given":"Tengjiao"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1111/jebm.12517","URL":"https://doi.org/10.1111/jebm.12517","source":"openalex"},{"id":"oa:W4381716077","type":"article-journal","title":"Discovery of a Novel DCAF1 Ligand Using a Drug–Target Interaction Prediction Model: Generalizing Machine Learning to New Drug Targets","abstract":"E3 ubiquitin ligases. The WDR domain of DCAF1 serves as a binding platform for substrate proteins and is also targeted by HIV and SIV lentiviral adaptors to induce the ubiquitination and proteasomal degradation of antiviral host factors. It is therefore attractive both as a potential therapeutic target for the development of chemical inhibitors and as an E3 ligase that could be recruited by novel PROTACs for targeted protein degradation. In this study, we used a proteome-scale drug-target interaction prediction model, MatchMaker, combined with cheminformatics filtering and docking to identify ligands for the DCAF1 WDR domain. Biophysical screening and X-ray crystallographic studies of the predicted binders confirmed a selective ligand occupying the central cavity of the WDR domain. This study shows that artificial intelligence-enabled virtual screening methods can successfully be applied in the absence of previously known ligands.","author":[{"family":"Kimani","given":"Serah"},{"family":"Owen","given":"Julie"},{"family":"Green","given":"Stuart"},{"family":"Li","given":"Fengling"},{"family":"Li","given":"Yanjun"},{"family":"Dong","given":"Aiping"},{"family":"Brown","given":"Peter"},{"family":"Ackloo","given":"Suzanne"},{"family":"Kuter","given":"David"},{"family":"Yang","given":"Cindy"},{"family":"Macaskill","given":"Miranda"},{"family":"Mackinnon","given":"Stephen"},{"family":"Arrowsmith","given":"CH"},{"family":"Schapira","given":"Matthieu"},{"family":"Shahani","given":"Vijay"},{"family":"Halabelian","given":"Levon"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1021/acs.jcim.3c00082","URL":"https://doi.org/10.1021/acs.jcim.3c00082","source":"openalex"},{"id":"oa:W4376615795","type":"article-journal","title":"Machine Learning and Deep Learning powered satellite communications: Enabling technologies, applications, open challenges, and future research directions","abstract":"Summary The recent wave of creating an interconnected world through satellites has renewed interest in satellite communications. Private and government‐funded space agencies are making advancements in the creation of satellite constellations, and the introduction of 5G has brought a new focus to a fully connected world. Satellites are the proposed solutions for establishing high throughput and low latency links to remote, hard‐to‐reach areas. This has caused the injection of many satellites in Earth's orbit, which has caused many discrepancies. There is a need to establish highly adaptive and flexible satellite systems to overcome this. Machine Learning (ML) and Deep Learning (DL) have gained much popularity when it comes to communication systems. This review extensively provides insight into ML and DL's utilization in satellite communications. This review covers how satellite communication subsystems and other satellite system applications can be implemented through Artificial Intelligence (AI) and the ongoing open challenges and future directions.","author":[{"family":"Bhattacharyya","given":"AB"},{"family":"Nambiar","given":"Shvetha"},{"family":"Ojha","given":"Ritwik"},{"family":"Gyaneshwar","given":"Amogh"},{"family":"Chadha","given":"Utkarsh"},{"family":"Srinivasan","given":"Kathiravan"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/sat.1482","URL":"https://doi.org/10.1002/sat.1482","source":"openalex"},{"id":"oa:W4380538374","type":"article-journal","title":"Bias in AI-based models for medical applications: challenges and mitigation strategies","abstract":"Artificial intelligence systems are increasingly being applied to healthcare. In surgery, AI applications hold promise as tools to predict surgical outcomes, assess technical skills, or guide surgeons intraoperatively via computer vision. On the other hand, AI systems can also suffer from bias, compounding existing inequities in socioeconomic status, race, ethnicity, religion, gender, disability, or sexual orientation. Bias particularly impacts disadvantaged populations, which can be subject to algorithmic predictions that are less accurate or underestimate the need for care. Thus, strategies for detecting and mitigating bias are pivotal for creating AI technology that is generalizable and fair. Here, we discuss a recent study that developed a new strategy to mitigate bias in surgical AI systems.","author":[{"family":"Mittermaier","given":"Mirja"},{"family":"Raza","given":"Marium"},{"family":"Kvedar","given":"Joseph"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1038/s41746-023-00858-z","URL":"https://doi.org/10.1038/s41746-023-00858-z","source":"openalex"},{"id":"oa:W4367850403","type":"article-journal","title":"Artificial Intelligence in Engineering","abstract":"Artificial intelligence (AI) has moved past its primitive stages and is now poised to revolutionize various fields, making it a disruptive technology. This technology is expected to completely transform traditional engineering in design, electrical, communication, and renewable energy approaches that have been human-centred. Despite being in its early stages, AI-powered engineering applications can work with vague design parameters and resolve intricate engineering problems that cannot be tackled using traditional design, electrical, communication, and renewable energy methods. This article aims to shed light on the current progress and future research trends in AI applications in engineering concepts, focusing on the ramp-up period of the last 5 years. Various methods such as machine learning, genetic algorithm, and fuzzy logic have been carefully evaluated from an engineering standpoint. AI-powered design studies have been reviewed and categorized for different design stages such as inspiration, idea and concept generation, evaluation, optimization, decision-making, and modeling. The review shows that there has been an increased interest in data-based design methods and explainable artificial intelligence in recent years. The use of AI methods in engineering applications has proven to be efficient, fast, accurate, and comprehensive, particularly with the use of deep learning methods and combinations that address situations where human capacity is inadequate. However, it is crucial to choose the appropriate AI method for an engineering problem to achieve successful results.","author":[{"family":"Khaleel","given":"Mohamed"},{"family":"Ahmed","given":"Abdussalam"},{"family":"Alsharif","given":"Abdulgader"}],"issued":{"date-parts":[[2023]]},"DOI":"10.47709/brilliance.v3i1.2170","URL":"https://doi.org/10.47709/brilliance.v3i1.2170","source":"openalex"},{"id":"oa:W4378515194","type":"article-journal","title":"Essential properties and explanation effectiveness of explainable artificial intelligence in healthcare: A systematic review","abstract":"Background: Significant advancements in the field of information technology have influenced the creation of trustworthy explainable artificial intelligence (XAI) in healthcare. Despite improved performance of XAI, XAI techniques have not yet been integrated into real-time patient care. Objective: The aim of this systematic review is to understand the trends and gaps in research on XAI through an assessment of the essential properties of XAI and an evaluation of explanation effectiveness in the healthcare field. Methods: A search of PubMed and Embase databases for relevant peer-reviewed articles on development of an XAI model using clinical data and evaluating explanation effectiveness published between January 1, 2011, and April 30, 2022, was conducted. All retrieved papers were screened independently by the two authors. Relevant papers were also reviewed for identification of the essential properties of XAI (e.g., stakeholders and objectives of XAI, quality of personalized explanations) and the measures of explanation effectiveness (e.g., mental model, user satisfaction, trust assessment, task performance, and correctability). Results: Six out of 882 articles met the criteria for eligibility. Artificial Intelligence (AI) users were the most frequently described stakeholders. XAI served various purposes, including evaluation, justification, improvement, and learning from AI. Evaluation of the quality of personalized explanations was based on fidelity, explanatory power, interpretability, and plausibility. User satisfaction was the most frequently used measure of explanation effectiveness, followed by trust assessment, correctability, and task performance. The methods of assessing these measures also varied. Conclusion: XAI research should address the lack of a comprehensive and agreed-upon framework for explaining XAI and standardized approaches for evaluating the effectiveness of the explanation that XAI provides to diverse AI stakeholders.","author":[{"family":"Jung","given":"Jinsun"},{"family":"Lee","given":"Hyungbok"},{"family":"Jung","given":"Hyunggu"},{"family":"Kim","given":"Hyeoneui"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.heliyon.2023.e16110","URL":"https://doi.org/10.1016/j.heliyon.2023.e16110","source":"openalex"},{"id":"oa:W4360891289","type":"manuscript","title":"Capabilities of GPT-4 on Medical Challenge Problems","abstract":"Large language models (LLMs) have demonstrated remarkable capabilities in natural language understanding and generation across various domains, including medicine. We present a comprehensive evaluation of GPT-4, a state-of-the-art LLM, on medical competency examinations and benchmark datasets. GPT-4 is a general-purpose model that is not specialized for medical problems through training or engineered to solve clinical tasks. Our analysis covers two sets of official practice materials for the USMLE, a three-step examination program used to assess clinical competency and grant licensure in the United States. We also evaluate performance on the MultiMedQA suite of benchmark datasets. Beyond measuring model performance, experiments were conducted to investigate the influence of test questions containing both text and images on model performance, probe for memorization of content during training, and study probability calibration, which is of critical importance in high-stakes applications like medicine. Our results show that GPT-4, without any specialized prompt crafting, exceeds the passing score on USMLE by over 20 points and outperforms earlier general-purpose models (GPT-3.5) as well as models specifically fine-tuned on medical knowledge (Med-PaLM, a prompt-tuned version of Flan-PaLM 540B). In addition, GPT-4 is significantly better calibrated than GPT-3.5, demonstrating a much-improved ability to predict the likelihood that its answers are correct. We also explore the behavior of the model qualitatively through a case study that shows the ability of GPT-4 to explain medical reasoning, personalize explanations to students, and interactively craft new counterfactual scenarios around a medical case. Implications of the findings are discussed for potential uses of GPT-4 in medical education, assessment, and clinical practice, with appropriate attention to challenges of accuracy and safety.","author":[{"family":"Nori","given":"Harsha"},{"family":"King","given":"Nicholas"},{"family":"Mckinney","given":"Scott"},{"family":"Carignan","given":"Dean"},{"family":"Horvitz","given":"Eric"}],"issued":{"date-parts":[[2023]]},"DOI":"10.48550/arxiv.2303.13375","URL":"https://doi.org/10.48550/arxiv.2303.13375","source":"openalex"},{"id":"oa:W4366710236","type":"article-journal","title":"Assessment of Performance, Interpretability, and Explainability in Artificial Intelligence–Based Health Technologies: What Healthcare Stakeholders Need to Know","abstract":"This review aimed to specify different concepts that are essential to the development of medical devices (MDs) with artificial intelligence (AI) (AI-based MDs) and shed light on how algorithm performance, interpretability, and explainability are key assets. First, a literature review was performed to determine the key criteria needed for a health technology assessment of AI-based MDs in the existing guidelines. Then, we analyzed the existing assessment methodologies of the different criteria selected after the literature review. The scoping review revealed that health technology assessment agencies have highlighted different criteria, with 3 important ones to reinforce confidence in AI-based MDs: performance, interpretability, and explainability. We give recommendations on how and when to evaluate performance on the basis of the model structure and available data. In addition, should interpretability and explainability be difficult to define mathematically, we describe existing ways to support their evaluation. We also provide a decision support flowchart to identify the anticipated regulatory requirements for the development and assessment of AI-based MDs. The importance of explainability and interpretability techniques in health technology assessment agencies is increasing to hold stakeholders more accountable for the decisions made by AI-based MDs. The identification of 3 main assessment criteria for AI-based MDs according to health technology assessment guidelines led us to propose a set of tools and methods to help understand how and why machine learning algorithms work as well as their predictions.","author":[{"family":"Farah","given":"Line"},{"family":"Murris","given":"Juliette"},{"family":"Borget","given":"Isabelle"},{"family":"Guilloux","given":"Agathe"},{"family":"Martelli","given":"Nicolas"},{"family":"Katsahian","given":"Sandrine"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.mcpdig.2023.02.004","URL":"https://doi.org/10.1016/j.mcpdig.2023.02.004","source":"openalex"},{"id":"oa:W4382810696","type":"article-journal","title":"Detecting dental caries on oral photographs using artificial intelligence: A systematic review","abstract":"OBJECTIVES: This systematic review aimed at evaluating the performance of artificial intelligence (AI) models in detecting dental caries on oral photographs. METHODS: Methodological characteristics and performance metrics of clinical studies reporting on deep learning and other machine learning algorithms were assessed. The risk of bias was evaluated using the quality assessment of diagnostic accuracy studies 2 (QUADAS-2) tool. A systematic search was conducted in EMBASE, Medline, and Scopus. RESULTS: Out of 3410 identified records, 19 studies were included with six and seven studies having low risk of biases and applicability concerns for all the domains, respectively. Metrics varied widely and were assessed on multiple levels. F1-scores for classification and detection tasks were 68.3%-94.3% and 42.8%-95.4%, respectively. Irrespective of the task, F1-scores were 68.3%-95.4% for professional cameras, 78.8%-87.6%, for intraoral cameras, and 42.8%-80% for smartphone cameras. Limited studies allowed assessing AI performance for lesions of different severity. CONCLUSION: Automatic detection of dental caries using AI may provide objective verification of clinicians' diagnoses and facilitate patient-clinician communication and teledentistry. Future studies should consider more robust study designs, employ comparable and standardized metrics, and focus on the severity of caries lesions.","author":[{"family":"Moharrami","given":"Mohammad"},{"family":"Farmer","given":"Julie"},{"family":"Singhal","given":"Sonica"},{"family":"Watson","given":"Erin"},{"family":"Glogauer","given":"Michael"},{"family":"Johnson","given":"Alistair"},{"family":"Schwendicke","given":"Falk"},{"family":"Quiñonez","given":"Carlos"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1111/odi.14659","URL":"https://doi.org/10.1111/odi.14659","source":"openalex"},{"id":"oa:W4391969619","type":"article-journal","title":"A trustworthy AI reality-check: the lack of transparency of artificial intelligence products in healthcare","abstract":"Trustworthy medical AI requires transparency about the development and testing of underlying algorithms to identify biases and communicate potential risks of harm. Abundant guidance exists on how to achieve transparency for medical AI products, but it is unclear whether publicly available information adequately informs about their risks. To assess this, we retrieved public documentation on the 14 available CE-certified AI-based radiology products of the II b risk category in the EU from vendor websites, scientific publications, and the European EUDAMED database. Using a self-designed survey, we reported on their development, validation, ethical considerations, and deployment caveats, according to trustworthy AI guidelines. We scored each question with either 0, 0.5, or 1, to rate if the required information was \"unavailable\", \"partially available,\" or \"fully available.\" The transparency of each product was calculated relative to all 55 questions. Transparency scores ranged from 6.4% to 60.9%, with a median of 29.1%. Major transparency gaps included missing documentation on training data, ethical considerations, and limitations for deployment. Ethical aspects like consent, safety monitoring, and GDPR-compliance were rarely documented. Furthermore, deployment caveats for different demographics and medical settings were scarce. In conclusion, public documentation of authorized medical AI products in Europe lacks sufficient public transparency to inform about safety and risks. We call on lawmakers and regulators to establish legally mandated requirements for public and substantive transparency to fulfill the promise of trustworthy AI for health.","author":[{"family":"Fehr","given":"Jana"},{"family":"Citro","given":"Brian"},{"family":"Malpani","given":"Rohit"},{"family":"Lippert","given":"Christoph"},{"family":"Madai","given":"Vince"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fdgth.2024.1267290","URL":"https://doi.org/10.3389/fdgth.2024.1267290","source":"openalex"},{"id":"oa:W4401616798","type":"article-journal","title":"Artificial Intelligence, the Digital Surgeon: Unravelling Its Emerging Footprint in Healthcare – The Narrative Review","abstract":"Background: Artificial Intelligence (AI) holds transformative potential for the healthcare industry, offering innovative solutions for diagnosis, treatment planning, and improving patient outcomes. As AI continues to be integrated into healthcare systems, it promises advancements across various domains. This review explores the diverse applications of AI in healthcare, along with the challenges and limitations that need to be addressed. The aim is to provide a comprehensive overview of AI's impact on healthcare and to identify areas for further development and focus. Main Applications: The review discusses the broad range of AI applications in healthcare. In medical imaging and diagnostics, AI enhances the accuracy and efficiency of diagnostic processes, aiding in early disease detection. AI-powered clinical decision support systems assist healthcare professionals in patient management and decision-making. Predictive analytics using AI enables the prediction of patient outcomes and identification of potential health risks. AI-driven robotic systems have revolutionized surgical procedures, improving precision and outcomes. Virtual assistants and chatbots enhance patient interaction and support, providing timely information and assistance. In the pharmaceutical industry, AI accelerates drug discovery and development by identifying potential drug candidates and predicting their efficacy. Additionally, AI improves administrative efficiency and operational workflows in healthcare, streamlining processes and reducing costs. AI-powered remote monitoring and telehealth solutions expand access to healthcare, particularly in underserved areas. Challenges and Limitations: Despite the significant promise of AI in healthcare, several challenges persist. Ensuring the reliability and consistency of AI-driven outcomes is crucial. Privacy and security concerns must be navigated carefully, particularly in handling sensitive patient data. Ethical considerations, including bias and fairness in AI algorithms, need to be addressed to prevent unintended consequences. Overcoming these challenges is critical for the ethical and successful integration of AI in healthcare. Conclusion: The integration of AI into healthcare is advancing rapidly, offering substantial benefits in improving patient care and operational efficiency. However, addressing the associated challenges is essential to fully realize the transformative potential of AI in healthcare. Future efforts should focus on enhancing the reliability, transparency, and ethical standards of AI technologies to ensure they contribute positively to global health outcomes.","author":[{"family":"Shang","given":"Zifang"},{"family":"Chauhan","given":"Varun"},{"family":"Devi","given":"Kirti"},{"family":"Patil","given":"Sandip"}],"issued":{"date-parts":[[2024]]},"DOI":"10.2147/jmdh.s482757","URL":"https://doi.org/10.2147/jmdh.s482757","source":"openalex"},{"id":"oa:W4392714531","type":"article-journal","title":"Development and validation of a scale for dependence on artificial intelligence in university students","abstract":"Background Artificial Intelligence (AI) has permeated various aspects of daily life, including education, specifically within higher education settings. These AI technologies have transformed pedagogy and learning, enabling a more personalized approach. However, ethical and practical concerns have also emerged, including the potential decline in cognitive skills and student motivation due to excessive reliance on AI. Objective To develop and validate a Scale for Dependence on Artificial Intelligence (DIA). Methods An Exploratory Factor Analysis (EFA) was used to identify the underlying structure of the DIA scale, followed by a Confirmatory Factor Analysis (CFA) to assess and confirm this structure. In addition, the scale’s invariance based on participants’ gender was evaluated. Results A total of 528 university students aged between 18 and 37 years (M = 20.31, SD = 3.8) participated. The EFA revealed a unifactorial structure for the scale, which was subsequently confirmed by the CFA. Invariance analyses showed that the scale is applicable and consistent for both men and women. Conclusion The DAI scale emerges as a robust and reliable tool for measuring university students’ dependence on AI. Its gender invariance makes it applicable in diverse population studies. In the age of digitalization, it is essential to understand the dynamics between humans and AI to navigate wisely and ensure a beneficial coexistence.","author":[{"family":"Morales-García","given":"Wilter"},{"family":"Sairitupa-Sanchez","given":"Liset"},{"family":"Morales-García","given":"Sandra"},{"family":"Morales-García","given":"Mardel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/feduc.2024.1323898","URL":"https://doi.org/10.3389/feduc.2024.1323898","source":"openalex"},{"id":"oa:W4392101515","type":"article-journal","title":"Integrating artificial intelligence into the modernization of traditional Chinese medicine industry: a review","abstract":"Traditional Chinese medicine (TCM) is the practical experience and summary of the Chinese nation for thousands of years. It shows great potential in treating various chronic diseases, complex diseases and major infectious diseases, and has gradually attracted the attention of people all over the world. However, due to the complexity of prescription and action mechanism of TCM, the development of TCM industry is still in a relatively conservative stage. With the rise of artificial intelligence technology in various fields, many scholars began to apply artificial intelligence technology to traditional Chinese medicine industry and made remarkable progress. This paper comprehensively summarizes the important role of artificial intelligence in the development of traditional Chinese medicine industry from various aspects, including new drug discovery, data mining, quality standardization and industry technology of traditional Chinese medicine. The limitations of artificial intelligence in these applications are also emphasized, including the lack of pharmacological research, database quality problems and the challenges brought by human-computer interaction. Nevertheless, the development of artificial intelligence has brought new opportunities and innovations to the modernization of traditional Chinese medicine. Integrating artificial intelligence technology into the comprehensive application of Chinese medicine industry is expected to overcome the major problems faced by traditional Chinese medicine industry and further promote the modernization of the whole traditional Chinese medicine industry.","author":[{"family":"Zhou","given":"Enyu"},{"family":"Shen","given":"Qin"},{"family":"Hou","given":"Yang"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3389/fphar.2024.1181183","URL":"https://doi.org/10.3389/fphar.2024.1181183","source":"openalex"},{"id":"oa:W4318619419","type":"article-journal","title":"Evaluation of the Diagnostic and Prognostic Accuracy of Artificial Intelligence in Endodontic Dentistry: A Comprehensive Review of Literature","abstract":"Aim . This comprehensive review is aimed at evaluating the diagnostic and prognostic accuracy of artificial intelligence in endodontic dentistry. Introduction . Artificial intelligence (AI) is a relatively new technology that has widespread use in dentistry. The AI technologies have primarily been used in dentistry to diagnose dental diseases, plan treatment, make clinical decisions, and predict the prognosis. AI models like convolutional neural networks (CNN) and artificial neural networks (ANN) have been used in endodontics to study root canal system anatomy, determine working length measurements, detect periapical lesions and root fractures, predict the success of retreatment procedures, and predict the viability of dental pulp stem cells. Methodology . The literature was searched in electronic databases such as Google Scholar, Medline, PubMed, Embase, Web of Science, and Scopus, published over the last four decades (January 1980 to September 15, 2021) by using keywords such as artificial intelligence, machine learning, deep learning, application, endodontics, and dentistry. Results . The preliminary search yielded 2560 articles relevant enough to the paper’s purpose. A total of 88 articles met the eligibility criteria. The majority of research on AI application in endodontics has concentrated on tracing apical foramen, verifying the working length, projection of periapical pathologies, root morphologies, and retreatment predictions and discovering the vertical root fractures. Conclusion . In endodontics, AI displayed accuracy in terms of diagnostic and prognostic evaluations. The use of AI can help enhance the treatment plan, which in turn can lead to an increase in the success rate of endodontic treatment outcomes. The AI is used extensively in endodontics and could help in clinical applications, such as detecting root fractures, periapical pathologies, determining working length, tracing apical foramen, the morphology of root, and disease prediction.","author":[{"family":"Karobari","given":"Mohmed"},{"family":"Adil","given":"Abdul"},{"family":"Basheer","given":"Syed"},{"family":"Murugesan","given":"Sabari"},{"family":"Savadamoorthi","given":"Kamatchi"},{"family":"Mustafa","given":"Mohammed"},{"family":"Abdulwahed","given":"Abdulaziz"},{"family":"Almokhatieb","given":"Ahmed"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1155/2023/7049360","URL":"https://doi.org/10.1155/2023/7049360","source":"openalex"},{"id":"oa:W4362580565","type":"article-journal","title":"Impact of nanotechnology on conventional and artificial intelligence-based biosensing strategies for the detection of viruses","abstract":"Recent years have witnessed the emergence of several viruses and other pathogens. Some of these infectious diseases have spread globally, resulting in pandemics. Although biosensors of various types have been utilized for virus detection, their limited sensitivity remains an issue. Therefore, the development of better diagnostic tools that facilitate the more efficient detection of viruses and other pathogens has become important. Nanotechnology has been recognized as a powerful tool for the detection of viruses, and it is expected to change the landscape of virus detection and analysis. Recently, nanomaterials have gained enormous attention for their value in improving biosensor performance owing to their high surface-to-volume ratio and quantum size effects. This article reviews the impact of nanotechnology on the design, development, and performance of sensors for the detection of viruses. Special attention has been paid to nanoscale materials, various types of nanobiosensors, the internet of medical things, and artificial intelligence-based viral diagnostic techniques.","author":[{"family":"Ramalingam","given":"Murugan"},{"family":"Jaisankar","given":"Abinaya"},{"family":"Cheng","given":"Lijia"},{"family":"Krishnan","given":"Sasirekha"},{"family":"Lan","given":"Liang"},{"family":"Hassan","given":"Anwarul"},{"family":"Şaşmazel","given":"Hilal"},{"family":"Kaji","given":"Hirokazu"},{"family":"Deigner","given":"Hans‐peter"},{"family":"Pedraz","given":"José"},{"family":"Kim","given":"Hae‐won"},{"family":"Shi","given":"Zheng"},{"family":"Marrazza","given":"Giovanna"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1186/s11671-023-03842-4","URL":"https://doi.org/10.1186/s11671-023-03842-4","source":"openalex"},{"id":"oa:W4372311383","type":"article-journal","title":"A machine learning and explainable artificial intelligence triage-prediction system for COVID-19","abstract":"COVID-19 is a respiratory disease caused by the SARS-CoV-2 contagion, severely disrupted the healthcare infrastructure. Various countries have developed COVID-19 vaccines that have effectively prevented the severe symptoms caused by the virus to a certain extent. However, a small section of people continues to perish. Artificial intelligence advances have revolutionized healthcare diagnosis and prognosis infrastructure. In this study, we predict the severity of COVID-19 using heterogenous Machine Learning and Deep Learning algorithms by considering clinical markers, vital signs, and other critical factors. This study extensively reviews various classifier architectures to predict the COVID-19 severity. We built and evaluated multiple pipelines entailing combinations of five state-of-the-art data-balancing techniques (Synthetic Minority Oversampling Technique (SMOTE), Adaptive Synthetic, Borderline SMOTE, SMOTE with Tomek links, and SMOTE with Edited Nearest Neighbor (ENN)) and twelve heterogeneous classifiers such as Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, K-Nearest Neighbors, Naïve Bayes, Xgboost, Extratrees, Adaboost, Light GBM, Catboost, and 1-D Convolution Neural Network. The best-performing pipeline consists of Random Forest trained on Borderline SMOTE balanced data that produced the highest recall of 83%. We deployed Explainable Artificial Intelligence tools such as Shapley Additive Explanations and Local Interpretable Model-agnostic Explanations, ELI5, Qlattice, Anchor, and Feature Importance to demystify complex tree-based ensemble models. These tools provide valuable insights into the significance of critical features in the severity prediction of a COVID-19 patient. It was observed that changes in respiratory rate, blood pressure, lactate, and calcium values were the primary contributors to the increase in severity of a COVID-19 patient. This architecture aims to be an explainable decision-support triaging system for medical professionals in countries lacking advanced medical technology and infrastructure to reduce fatalities.","author":[{"family":"Khanna","given":"Varada"},{"family":"Chadaga","given":"Krishnaraj"},{"family":"Sampathila","given":"Niranjana"},{"family":"Prabhu","given":"Srikanth"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.dajour.2023.100246","URL":"https://doi.org/10.1016/j.dajour.2023.100246","source":"openalex"},{"id":"oa:W4399236723","type":"article-journal","title":"Artificial Intelligence in Pediatric Emergency Medicine: Applications, Challenges, and Future Perspectives","abstract":"The dawn of Artificial intelligence (AI) in healthcare stands as a milestone in medical innovation. Different medical fields are heavily involved, and pediatric emergency medicine is no exception. We conducted a narrative review structured in two parts. The first part explores the theoretical principles of AI, providing all the necessary background to feel confident with these new state-of-the-art tools. The second part presents an informative analysis of AI models in pediatric emergencies. We examined PubMed and Cochrane Library from inception up to April 2024. Key applications include triage optimization, predictive models for traumatic brain injury assessment, and computerized sepsis prediction systems. In each of these domains, AI models outperformed standard methods. The main barriers to a widespread adoption include technological challenges, but also ethical issues, age-related differences in data interpretation, and the paucity of comprehensive datasets in the pediatric context. Future feasible research directions should address the validation of models through prospective datasets with more numerous sample sizes of patients. Furthermore, our analysis shows that it is essential to tailor AI algorithms to specific medical needs. This requires a close partnership between clinicians and developers. Building a shared knowledge platform is therefore a key step.","author":[{"family":"Sarno","given":"Lorenzo"},{"family":"Caroselli","given":"Anya"},{"family":"Tonin","given":"Giovanna"},{"family":"Graglia","given":"Benedetta"},{"family":"Pansini","given":"Valeria"},{"family":"Causio","given":"Francesco"},{"family":"Gatto","given":"Antonio"},{"family":"Chiaretti","given":"Antonio"}],"issued":{"date-parts":[[2024]]},"DOI":"10.3390/biomedicines12061220","URL":"https://doi.org/10.3390/biomedicines12061220","source":"openalex"},{"id":"oa:W4324308124","type":"article-journal","title":"An overview and a roadmap for artificial intelligence in hematology and oncology","abstract":"BACKGROUND: Artificial intelligence (AI) is influencing our society on many levels and has broad implications for the future practice of hematology and oncology. However, for many medical professionals and researchers, it often remains unclear what AI can and cannot do, and what are promising areas for a sensible application of AI in hematology and oncology. Finally, the limits and perils of using AI in oncology are not obvious to many healthcare professionals. METHODS: In this article, we provide an expert-based consensus statement by the joint Working Group on \"Artificial Intelligence in Hematology and Oncology\" by the German Society of Hematology and Oncology (DGHO), the German Association for Medical Informatics, Biometry and Epidemiology (GMDS), and the Special Interest Group Digital Health of the German Informatics Society (GI). We provide a conceptual framework for AI in hematology and oncology. RESULTS: First, we propose a technological definition, which we deliberately set in a narrow frame to mainly include the technical developments of the last ten years. Second, we present a taxonomy of clinically relevant AI systems, structured according to the type of clinical data they are used to analyze. Third, we show an overview of potential applications, including clinical, research, and educational environments with a focus on hematology and oncology. CONCLUSION: Thus, this article provides a point of reference for hematologists and oncologists, and at the same time sets forth a framework for the further development and clinical deployment of AI in hematology and oncology in the future.","author":[{"family":"Rösler","given":"Wiebke"},{"family":"Altenbuchinger","given":"Michael"},{"family":"Baeßler","given":"Bettina"},{"family":"Beißbarth","given":"Tim"},{"family":"Beutel","given":"Gernot"},{"family":"Bock","given":"Robert"},{"family":"Bubnoff","given":"Nikolas"},{"family":"Eckardt","given":"Jan‐niklas"},{"family":"Foersch","given":"Sebastian"},{"family":"Loeffler","given":"Chiara"},{"family":"Middeke","given":"Jan"},{"family":"Mueller","given":"Martha"},{"family":"Oellerich","given":"Thomas"},{"family":"Risse","given":"Benjamin"},{"family":"Scherag","given":"André"},{"family":"Schliemann","given":"Christoph"},{"family":"Scholz","given":"Markus"},{"family":"Spang","given":"Rainer"},{"family":"Thielscher","given":"Christian"},{"family":"Tsoukakis","given":"Ioannis"},{"family":"Kather","given":"Jakob"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1007/s00432-023-04667-5","URL":"https://doi.org/10.1007/s00432-023-04667-5","source":"openalex"},{"id":"oa:W4366769280","type":"article-journal","title":"Using AI-generated suggestions from ChatGPT to optimize clinical decision support","abstract":"OBJECTIVE: To determine if ChatGPT can generate useful suggestions for improving clinical decision support (CDS) logic and to assess noninferiority compared to human-generated suggestions. METHODS: We supplied summaries of CDS logic to ChatGPT, an artificial intelligence (AI) tool for question answering that uses a large language model, and asked it to generate suggestions. We asked human clinician reviewers to review the AI-generated suggestions as well as human-generated suggestions for improving the same CDS alerts, and rate the suggestions for their usefulness, acceptance, relevance, understanding, workflow, bias, inversion, and redundancy. RESULTS: Five clinicians analyzed 36 AI-generated suggestions and 29 human-generated suggestions for 7 alerts. Of the 20 suggestions that scored highest in the survey, 9 were generated by ChatGPT. The suggestions generated by AI were found to offer unique perspectives and were evaluated as highly understandable and relevant, with moderate usefulness, low acceptance, bias, inversion, redundancy. CONCLUSION: AI-generated suggestions could be an important complementary part of optimizing CDS alerts, can identify potential improvements to alert logic and support their implementation, and may even be able to assist experts in formulating their own suggestions for CDS improvement. ChatGPT shows great potential for using large language models and reinforcement learning from human feedback to improve CDS alert logic and potentially other medical areas involving complex, clinical logic, a key step in the development of an advanced learning health system.","author":[{"family":"Liu","given":"Siru"},{"family":"Wright","given":"Aileen"},{"family":"Patterson","given":"Barron"},{"family":"Wanderer","given":"Jonathan"},{"family":"Turer","given":"Robert"},{"family":"Nelson","given":"Scott"},{"family":"Mccoy","given":"Allison"},{"family":"Sittig","given":"Dean"},{"family":"Wright","given":"Adam"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1093/jamia/ocad072","URL":"https://doi.org/10.1093/jamia/ocad072","source":"openalex"},{"id":"oa:W4386136032","type":"article-journal","title":"Perception, performance, and detectability of conversational artificial intelligence across 32 university courses","abstract":"The emergence of large language models has led to the development of powerful tools such as ChatGPT that can produce text indistinguishable from human-generated work. With the increasing accessibility of such technology, students across the globe may utilize it to help with their school work-a possibility that has sparked ample discussion on the integrity of student evaluation processes in the age of artificial intelligence (AI). To date, it is unclear how such tools perform compared to students on university-level courses across various disciplines. Further, students' perspectives regarding the use of such tools in school work, and educators' perspectives on treating their use as plagiarism, remain unknown. Here, we compare the performance of the state-of-the-art tool, ChatGPT, against that of students on 32 university-level courses. We also assess the degree to which its use can be detected by two classifiers designed specifically for this purpose. Additionally, we conduct a global survey across five countries, as well as a more in-depth survey at the authors' institution, to discern students' and educators' perceptions of ChatGPT's use in school work. We find that ChatGPT's performance is comparable, if not superior, to that of students in a multitude of courses. Moreover, current AI-text classifiers cannot reliably detect ChatGPT's use in school work, due to both their propensity to classify human-written answers as AI-generated, as well as the relative ease with which AI-generated text can be edited to evade detection. Finally, there seems to be an emerging consensus among students to use the tool, and among educators to treat its use as plagiarism. Our findings offer insights that could guide policy discussions addressing the integration of artificial intelligence into educational frameworks.","author":[{"family":"Ibrahim","given":"Hazem"},{"family":"Liu","given":"Fengyuan"},{"family":"Asim","given":"Rohail"},{"family":"Battu","given":"Balaraju"},{"family":"Benabderrahmane","given":"Sidahmed"},{"family":"Alhafni","given":"Bashar"},{"family":"Adnan","given":"Wifag"},{"family":"Alhanai","given":"Tuka"},{"family":"Alshebli","given":"Bedoor"},{"family":"Baghdadi","given":"Riyadh"},{"family":"Bélanger","given":"Jocelyn"},{"family":"Beretta","given":"Elena"},{"family":"Çelik","given":"Kemal"},{"family":"Chaqfeh","given":"Moumena"},{"family":"Daqaq","given":"Mohammed"},{"family":"Bernoussi","given":"Zaynab"},{"family":"Fougnie","given":"Daryl"},{"family":"Soto","given":"Borja"},{"family":"Gandolfi","given":"Alberto"},{"family":"György","given":"András"},{"family":"Habash","given":"Nizar"},{"family":"Harris","given":"JA"},{"family":"Kaufman","given":"Aaron"},{"family":"Kirousis","given":"Lefteris"},{"family":"Koçak","given":"Korhan"},{"family":"Lee","given":"Kangsan"},{"family":"Lee","given":"Seung"},{"family":"Malik","given":"Samreen"},{"family":"Maniatakos","given":"Michail"},{"family":"Melcher","given":"David"},{"family":"Mourad","given":"Azzam"},{"family":"Park","given":"Minsu"},{"family":"Rasras","given":"Mahmoud"},{"family":"Reuben","given":"Alicja"},{"family":"Zantout","given":"Dania"},{"family":"Gleason","given":"Nancy"},{"family":"Makovi","given":"Kinga"},{"family":"Rahwan","given":"Talal"},{"family":"Zaki","given":"Yasir"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1038/s41598-023-38964-3","URL":"https://doi.org/10.1038/s41598-023-38964-3","source":"openalex"},{"id":"oa:W4401259609","type":"article-journal","title":"Application of medical artificial intelligence technology in sub-Saharan Africa: Prospects for medical laboratories","abstract":"The widespread adoption of artificial intelligence (AI) technology globally has brought significant changes to various sectors. AI-assisted algorithms have notably improved decision-making, operational efficiency, and productivity, especially in healthcare and medicine. However, in low and middle-income countries (LMICs), particularly in sub-Saharan Africa (SSA), the integration of medical AI has faced delays and challenges, slowing its acceptance and implementation in medical interventions. This thematic narrative critically explores the current trends and patterns in applying medical AI in SSA, with a specific focus on its potential impact on medical laboratories. The review covers the general use of medical AI in SSA, examining factors like enablers, challenges, and opportunities that influence healthcare systems. Additionally, it looks into the implications of medical AI for medical laboratories and suggests context-specific and practical recommendations for potential integration. We highlight various challenges, including data availability, security concerns, resource limitations, regulatory gaps, poor internet connectivity, and digital literacy issues, contributing to the slow integration of AI in healthcare systems in SSA. Despite challenges, the adoption of medical AI in SSA medical laboratories holds latent potential for improving diagnostic accuracy, streamlining workflows, and enhancing patient care. Further exploration and careful consideration are necessary to unlock these possibilities.","author":[{"family":"Ephraim","given":"Richard"},{"family":"Kotam","given":"Gabriel"},{"family":"Duah","given":"Evans"},{"family":"Ghartey","given":"Frank"},{"family":"Mathebula","given":"Evans"},{"family":"Mashamba-Thompson","given":"Tivani"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.smhl.2024.100505","URL":"https://doi.org/10.1016/j.smhl.2024.100505","source":"openalex"},{"id":"oa:W4402054386","type":"article-journal","title":"Artificial Intelligence of Things: A Survey","abstract":"The integration of the Internet of Things (IoT) and modern Artificial Intelligence (AI) has given rise to a new paradigm known as the Artificial Intelligence of Things (AIoT). In this survey, we provide a systematic and comprehensive review of AIoT research. We examine AIoT literature related to sensing, computing, and networking & communication, which form the three key components of AIoT. In addition to advancements in these areas, we review domain-specific AIoT systems that are designed for various important application domains. We have also created an accompanying GitHub repository, where we compile the papers included in this survey: https://github.com/AIoT-MLSys-Lab/AIoT-Survey. This repository will be actively maintained and updated with new research as it becomes available. As both IoT and AI become increasingly critical to our society, we believe that AIoT is emerging as an essential research field at the intersection of IoT and modern AI. It is our hope that this survey will serve as a valuable resource for those engaged in AIoT research and act as a catalyst for future explorations to bridge gaps and drive advancements in this exciting field.","author":[{"family":"Siam","given":"Md"},{"family":"Ahn","given":"Hyunho"},{"family":"Liu","given":"Li"},{"family":"Alam","given":"Samiul"},{"family":"Shen","given":"Haowei"},{"family":"Cao","given":"Zhichao"},{"family":"Shroff","given":"Ness"},{"family":"Krishnamachari","given":"Bhaskar"},{"family":"Srivastava","given":"Mani"},{"family":"Zhang","given":"Mi"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1145/3690639","URL":"https://doi.org/10.1145/3690639","source":"openalex"},{"id":"oa:W4392810470","type":"article-journal","title":"Artificial intelligence-based MRI radiomics and radiogenomics in glioma","abstract":"The specific genetic subtypes that gliomas exhibit result in variable clinical courses and the need to involve multidisciplinary teams of neurologists, epileptologists, neurooncologists and neurosurgeons. Currently, the diagnosis of gliomas pivots mainly around the preliminary radiological findings and the subsequent definitive surgical diagnosis (via surgical sampling). Radiomics and radiogenomics present a potential to precisely diagnose and predict survival and treatment responses, via morphological, textural, and functional features derived from MRI data, as well as genomic data. In spite of their advantages, it is still lacking standardized processes of feature extraction and analysis methodology among different research groups, which have made external validations infeasible. Radiomics and radiogenomics can be used to better understand the genomic basis of gliomas, such as tumor spatial heterogeneity, treatment response, molecular classifications and tumor microenvironment immune infiltration. These novel techniques have also been used to predict histological features, grade or even overall survival in gliomas. In this review, workflows of radiomics and radiogenomics are elucidated, with recent research on machine learning or artificial intelligence in glioma.","author":[{"family":"Fan","given":"Haiqing"},{"family":"Luo","given":"Yilin"},{"family":"Gu","given":"Fang"},{"family":"Tian","given":"Bin"},{"family":"Xiong","given":"Yongqin"},{"family":"Wu","given":"Guipeng"},{"family":"Nie","given":"Xin"},{"family":"Yu","given":"Jing"},{"family":"Tong","given":"Juan"},{"family":"Liao","given":"Xin"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s40644-024-00682-y","URL":"https://doi.org/10.1186/s40644-024-00682-y","source":"openalex"},{"id":"oa:W4385556979","type":"article-journal","title":"What Is Machine Learning, Artificial Neural Networks and Deep Learning?—Examples of Practical Applications in Medicine","abstract":"Machine learning (ML), artificial neural networks (ANNs), and deep learning (DL) are all topics that fall under the heading of artificial intelligence (AI) and have gained popularity in recent years. ML involves the application of algorithms to automate decision-making processes using models that have not been manually programmed but have been trained on data. ANNs that are a part of ML aim to simulate the structure and function of the human brain. DL, on the other hand, uses multiple layers of interconnected neurons. This enables the processing and analysis of large and complex databases. In medicine, these techniques are being introduced to improve the speed and efficiency of disease diagnosis and treatment. Each of the AI techniques presented in the paper is supported with an example of a possible medical application. Given the rapid development of technology, the use of AI in medicine shows promising results in the context of patient care. It is particularly important to keep a close eye on this issue and conduct further research in order to fully explore the potential of ML, ANNs, and DL, and bring further applications into clinical use in the future.","author":[{"family":"Kufel","given":"Jakub"},{"family":"Bargieł-Łączek","given":"Katarzyna"},{"family":"Kocot","given":"Szymon"},{"family":"Koźlik","given":"Maciej"},{"family":"Bartnikowska","given":"Wiktoria"},{"family":"Janik","given":"Michał"},{"family":"Czogalik","given":"Łukasz"},{"family":"Dudek","given":"Piotr"},{"family":"Magiera","given":"Mikołaj"},{"family":"Lis","given":"Anna"},{"family":"Paszkiewicz","given":"Iga"},{"family":"Nawrat","given":"Zbigniew"},{"family":"Cebula","given":"Maciej"},{"family":"Gruszczyńska","given":"Katarzyna"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/diagnostics13152582","URL":"https://doi.org/10.3390/diagnostics13152582","source":"openalex"},{"id":"oa:W4396869758","type":"article-journal","title":"Microbiology in the era of artificial intelligence: transforming medical and pharmaceutical microbiology","abstract":"In this mini-review, we delve into the transformative impact of artificial intelligence (AI) and machine learning (ML) in the field of microbiology. The paper provides a brief overview of various domains where AI is reshaping practices, including clinical diagnostics, drug and vaccine discovery, and public health management. Our discussion spotlights the implementation of convolutional neural networks for enhanced pathogen identification, the advancements in point-of-care diagnostics, and the emergence of new antimicrobials to tackle resistant strains. The application of AI in epidemiology, microbial ecology and forensic microbiology is also outlined, underscoring its proficiency in deciphering complex microbial interactions and forecasting disease outbreaks. We critically examine the challenges in AI application, such as ensuring data quality and overcoming algorithmic constraints, and stress the necessity for interpretable AI models that align with medical and ethical standards. We address the intricacies of digitalization in microbiology diagnostics, emphasizing the need for efficient data management in laboratory and clinical environments. Looking forward, we identify key directions for AI in microbiology, particularly focusing on developing adaptable, self-updating AI models and their integration into clinical settings. We conclude by highlighting AI's potential to revolutionize microbiological diagnostics and infection control, significantly influencing patient care and public health. This review serves as an invitation to explore AI's integration into microbiology, showcasing its role in evolving current methodologies and propelling future innovations.","author":[{"family":"Tsitou","given":"Virna"},{"family":"Rallis","given":"Dimitrios"},{"family":"Tsekova","given":"Mariana"},{"family":"Yanev","given":"Nikolay"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1080/13102818.2024.2349587","URL":"https://doi.org/10.1080/13102818.2024.2349587","source":"openalex"},{"id":"oa:W4386135236","type":"article-journal","title":"Ethical Considerations for Artificial Intelligence in Medical Imaging: Deployment and Governance","abstract":"The deployment of artificial intelligence (AI) has the potential to make nuclear medicine and medical imaging faster, cheaper, and both more effective and more accessible. This is possible, however, only if clinicians and patients feel that these AI medical devices (AIMDs) are trustworthy. Highlighting the need to ensure health justice by fairly distributing benefits and burdens while respecting individual patients' rights, the AI Task Force of the Society of Nuclear Medicine and Molecular Imaging has identified 4 major ethical risks that arise during the deployment of AIMD: autonomy of patients and clinicians, transparency of clinical performance and limitations, fairness toward marginalized populations, and accountability of physicians and developers. We provide preliminary recommendations for governing these ethical risks to realize the promise of AIMD for patients and populations.","author":[{"family":"Herington","given":"Jonathan"},{"family":"Mccradden","given":"Melissa"},{"family":"Creel","given":"Kathleen"},{"family":"Boellaard","given":"Ronald"},{"family":"Jones","given":"Elizabeth"},{"family":"Jha","given":"Abhinav"},{"family":"Rahmim","given":"Arman"},{"family":"Scott","given":"Peter"},{"family":"Sunderland","given":"John"},{"family":"Wahl","given":"Richard"},{"family":"Zuehlsdorff","given":"Sven"},{"family":"Saboury","given":"Babak"}],"issued":{"date-parts":[[2023]]},"DOI":"10.2967/jnumed.123.266110","URL":"https://doi.org/10.2967/jnumed.123.266110","source":"openalex"},{"id":"oa:W4383313974","type":"article-journal","title":"Should Artificial Intelligence be used to support clinical ethical decision-making? A systematic review of reasons","abstract":"BACKGROUND: Healthcare providers have to make ethically complex clinical decisions which may be a source of stress. Researchers have recently introduced Artificial Intelligence (AI)-based applications to assist in clinical ethical decision-making. However, the use of such tools is controversial. This review aims to provide a comprehensive overview of the reasons given in the academic literature for and against their use. METHODS: PubMed, Web of Science, Philpapers.org and Google Scholar were searched for all relevant publications. The resulting set of publications was title and abstract screened according to defined inclusion and exclusion criteria, resulting in 44 papers whose full texts were analysed using the Kuckartz method of qualitative text analysis. RESULTS: Artificial Intelligence might increase patient autonomy by improving the accuracy of predictions and allowing patients to receive their preferred treatment. It is thought to increase beneficence by providing reliable information, thereby, supporting surrogate decision-making. Some authors fear that reducing ethical decision-making to statistical correlations may limit autonomy. Others argue that AI may not be able to replicate the process of ethical deliberation because it lacks human characteristics. Concerns have been raised about issues of justice, as AI may replicate existing biases in the decision-making process. CONCLUSIONS: The prospective benefits of using AI in clinical ethical decision-making are manifold, but its development and use should be undertaken carefully to avoid ethical pitfalls. Several issues that are central to the discussion of Clinical Decision Support Systems, such as justice, explicability or human-machine interaction, have been neglected in the debate on AI for clinical ethics so far. TRIAL REGISTRATION: This review is registered at Open Science Framework ( https://osf.io/wvcs9 ).","author":[{"family":"Benzinger","given":"Lasse"},{"family":"Ursin","given":"Frank"},{"family":"Balke","given":"Wolf‐tilo"},{"family":"Kacprowski","given":"Tim"},{"family":"Salloch","given":"Sabine"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1186/s12910-023-00929-6","URL":"https://doi.org/10.1186/s12910-023-00929-6","source":"openalex"},{"id":"oa:W4390421984","type":"article-journal","title":"Is Attention all You Need in Medical Image Analysis? A Review","abstract":"Medical imaging is a key component in clinical diagnosis, treatment planning and clinical trial design, accounting for almost 90% of all healthcare data. CNNs achieved performance gains in medical image analysis (MIA) over the last years. CNNs can efficiently model local pixel interactions and be trained on small-scale MI data. Despite their important advances, typical CNN have relatively limited capabilities in modelling \"global\" pixel interactions, which restricts their generalisation ability to understand out-of-distribution data with different \"global\" information. The recent progress of Artificial Intelligence gave rise to Transformers, which can learn global relationships from data. However, full Transformer models need to be trained on large-scale data and involve tremendous computational complexity. Attention and Transformer compartments (\"Transf/Attention\") which can well maintain properties for modelling global relationships, have been proposed as lighter alternatives of full Transformers. Recently, there is an increasing trend to co-pollinate complementary local-global properties from CNN and Transf/Attention architectures, which led to a new era of hybrid models. The past years have witnessed substantial growth in hybrid CNN-Transf/Attention models across diverse MIA problems. In this systematic review, we survey existing hybrid CNN-Transf/Attention models, review and unravel key architectural designs, analyse breakthroughs, and evaluate current and future opportunities as well as challenges. We also introduced an analysis framework on generalisation opportunities of scientific and clinical impact, based on which new data-driven domain generalisation and adaptation methods can be stimulated.","author":[{"family":"Papanastasiou","given":"Giorgos"},{"family":"Δικαίος","given":"Νικόλαος"},{"family":"Huang","given":"Jiahao"},{"family":"Wang","given":"Chengjia"},{"family":"Yang","given":"Guang"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/jbhi.2023.3348436","URL":"https://doi.org/10.1109/jbhi.2023.3348436","source":"openalex"},{"id":"oa:W4389993479","type":"article-journal","title":"Validity and reliability of artificial intelligence chatbots as public sources of information on endodontics","abstract":"AIM: This study aimed to evaluate and compare the validity and reliability of responses provided by GPT-3.5, Google Bard, and Bing to frequently asked questions (FAQs) in the field of endodontics. METHODOLOGY: FAQs were formulated by expert endodontists (n = 10) and collected through GPT-3.5 queries (n = 10), with every question posed to each chatbot three times. Responses (N = 180) were independently evaluated by two board-certified endodontists using a modified Global Quality Score (GQS) on a 5-point Likert scale (5: strongly agree; 4: agree; 3: neutral; 2: disagree; 1: strongly disagree). Disagreements on scoring were resolved through evidence-based discussions. The validity of responses was analysed by categorizing scores into valid or invalid at two thresholds: The low threshold was set at score ≥4 for all three responses whilst the high threshold was set at score 5 for all three responses. Fisher's exact test was conducted to compare the validity of responses between chatbots. Cronbach's alpha was calculated to assess the reliability by assessing the consistency of repeated responses for each chatbot. RESULTS: All three chatbots provided answers to all questions. Using the low-threshold validity test (GPT-3.5: 95%; Google Bard: 85%; Bing: 75%), there was no significant difference between the platforms (p > .05). When using the high-threshold validity test, the chatbot scores were substantially lower (GPT-3.5: 60%; Google Bard: 15%; Bing: 15%). The validity of GPT-3.5 responses was significantly higher than Google Bard and Bing (p = .008). All three chatbots achieved an acceptable level of reliability (Cronbach's alpha >0.7). CONCLUSIONS: GPT-3.5 provided more credible information on topics related to endodontics compared to Google Bard and Bing.","author":[{"family":"Mohammadrahimi","given":"Hossein"},{"family":"Ourang","given":"Seyed"},{"family":"Pourhoseingholi","given":"Mohamad"},{"family":"Dianat","given":"Omid"},{"family":"Dummer","given":"PMH"},{"family":"Nosrat","given":"Ali"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1111/iej.14014","URL":"https://doi.org/10.1111/iej.14014","source":"openalex"},{"id":"oa:W4395053818","type":"article-journal","title":"Artificial Intelligence-Driven Radiomics in Head and Neck Cancer: Current Status and Future Prospects","abstract":"BACKGROUND: Radiomics is a rapidly growing field used to leverage medical radiological images by extracting quantitative features. These are supposed to characterize a patient's phenotype, and when combined with artificial intelligence techniques, to improve the accuracy of diagnostic models and clinical outcome prediction. OBJECTIVES: This review aims at examining the application areas of artificial intelligence-based radiomics (AI-based radiomics) for the management of head and neck cancer (HNC). It further explores the workflow of AI-based radiomics for personalized and precision oncology in HNC. Finally, it examines the current challenges of AI-based radiomics in daily clinical oncology and offers possible solutions to these challenges. METHODS: Comprehensive electronic databases (PubMed, Medline via Ovid, Scopus, Web of Science, CINAHL, and Cochrane Library) were searched following the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines. The quality of included studies and their risk of biases were evaluated using the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD)and Prediction Model Risk of Bias Assessment Tool (PROBAST). RESULTS: Out of the 659 search hits retrieved, 45 fulfilled the inclusion criteria. Our review revealed that the application of AI-based radiomics model as an ancillary tool for improved decision-making in HNC management includes radiomics-based cancer diagnosis and radiomics-based cancer prognosis. The radiomics-based cancer diagnosis includes tumor staging, tumor grading, and classification of malignant and benign tumors. Similarly, radiomics-based cancer prognosis includes prediction for treatment response, recurrence, metastasis, and survival. In addition, the challenges in the implementation of these models for clinical evaluations include data imbalance, feature engineering (extraction and selection), model generalizability, multi-modal fusion, and model interpretability. CONCLUSION: Considering the highly subjective and interobserver variability that is peculiar to the interpretation of medical images by expert clinicians, AI-based radiomics seeks to offer potentially useful quantitative information, which is not visible to the human eye or unintentionally often remain ignored during clinical imaging practice. By enabling the extraction of this type of information, AI-based radiomics has the potential to revolutionize HNC oncology, providing a platform for more personalized, higher quality, and cost-effective care for HNC patients.","author":[{"family":"Alabi","given":"Rasheed"},{"family":"Elmusrati","given":"Mohammed"},{"family":"Leivo","given":"Ilmo"},{"family":"Almangush","given":"Alhadi"},{"family":"Mäkitie","given":"Antti"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.ijmedinf.2024.105464","URL":"https://doi.org/10.1016/j.ijmedinf.2024.105464","source":"openalex"},{"id":"oa:W4367843698","type":"article-journal","title":"Artificial Intelligence in CT and MR Imaging for Oncological Applications","abstract":"Cancer care increasingly relies on imaging for patient management. The two most common cross-sectional imaging modalities in oncology are computed tomography (CT) and magnetic resonance imaging (MRI), which provide high-resolution anatomic and physiological imaging. Herewith is a summary of recent applications of rapidly advancing artificial intelligence (AI) in CT and MRI oncological imaging that addresses the benefits and challenges of the resultant opportunities with examples. Major challenges remain, such as how best to integrate AI developments into clinical radiology practice, the vigorous assessment of quantitative CT and MR imaging data accuracy, and reliability for clinical utility and research integrity in oncology. Such challenges necessitate an evaluation of the robustness of imaging biomarkers to be included in AI developments, a culture of data sharing, and the cooperation of knowledgeable academics with vendor scientists and companies operating in radiology and oncology fields. Herein, we will illustrate a few challenges and solutions of these efforts using novel methods for synthesizing different contrast modality images, auto-segmentation, and image reconstruction with examples from lung CT as well as abdome, pelvis, and head and neck MRI. The imaging community must embrace the need for quantitative CT and MRI metrics beyond lesion size measurement. AI methods for the extraction and longitudinal tracking of imaging metrics from registered lesions and understanding the tumor environment will be invaluable for interpreting disease status and treatment efficacy. This is an exciting time to work together to move the imaging field forward with narrow AI-specific tasks. New AI developments using CT and MRI datasets will be used to improve the personalized management of cancer patients.","author":[{"family":"Paudyal","given":"Ramesh"},{"family":"Shah","given":"Akash"},{"family":"Akın","given":"Oğuz"},{"family":"Gian","given":"Richard"},{"family":"Konar","given":"Amaresha"},{"family":"Hatzoglou","given":"Vaios"},{"family":"Mahmood","given":"Usman"},{"family":"Lee","given":"Nancy"},{"family":"Wong","given":"Richard"},{"family":"Banerjee","given":"Suchandrima"},{"family":"Shin","given":"Jaemin"},{"family":"Veeraraghavan","given":"Harini"},{"family":"Shukladave","given":"Amita"}],"issued":{"date-parts":[[2023]]},"DOI":"10.3390/cancers15092573","URL":"https://doi.org/10.3390/cancers15092573","source":"openalex"},{"id":"oa:W4391544808","type":"article-journal","title":"PubMed and beyond: biomedical literature search in the age of artificial intelligence","abstract":"Biomedical research yields vast information, much of which is only accessible through the literature. Consequently, literature search is crucial for healthcare and biomedicine. Recent improvements in artificial intelligence (AI) have expanded functionality beyond keywords, but they might be unfamiliar to clinicians and researchers. In response, we present an overview of over 30 literature search tools tailored to common biomedical use cases, aiming at helping readers efficiently fulfill their information needs. We first discuss recent improvements and continued challenges of the widely used PubMed. Then, we describe AI-based literature search tools catering to five specific information needs: 1. Evidence-based medicine. 2. Precision medicine and genomics. 3. Searching by meaning, including questions. 4. Finding related articles with literature recommendation. 5. Discovering hidden associations through literature mining. Finally, we discuss the impacts of recent developments of large language models such as ChatGPT on biomedical information seeking.","author":[{"family":"Jin","given":"Qiao"},{"family":"Leaman","given":"Robert"},{"family":"Lu","given":"Zhiyong"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.ebiom.2024.104988","URL":"https://doi.org/10.1016/j.ebiom.2024.104988","source":"openalex"},{"id":"oa:W4404701785","type":"article-journal","title":"Development of the design and synthesis of metal–organic frameworks (MOFs) – from large scale attempts, functional oriented modifications, to artificial intelligence (AI) predictions","abstract":"Owing to the exceptional porous properties of metal-organic frameworks (MOFs), there has recently been a surge of interest, evidenced by a plethora of research into their design, synthesis, properties, and applications. This expanding research landscape has driven significant advancements in the precise regulation of MOF design and synthesis. Initially dominated by large-scale synthesis approaches, this field has evolved towards more targeted functional modifications. Recently, the integration of computational science, particularly through artificial intelligence predictions, has ushered in a new era of innovation, enabling more precise and efficient MOF design and synthesis methodologies. The objective of this review is to provide readers with an extensive overview of the development process of MOF design and synthesis, and to present visions for future developments.","author":[{"family":"Han","given":"Zongsu"},{"family":"Yang","given":"Yihao"},{"family":"Rushlow","given":"Joshua"},{"family":"Huo","given":"Jiatong"},{"family":"Liu","given":"Zhaoyi"},{"family":"Hsu","given":"Yu‐chuan"},{"family":"Yin","given":"Rujie"},{"family":"Wang","given":"Mengmeng"},{"family":"Liang","given":"Rong‐ran"},{"family":"Wang","given":"Kunyu"},{"family":"Zhou","given":"Hong‐cai"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1039/d4cs00432a","URL":"https://doi.org/10.1039/d4cs00432a","source":"openalex"},{"id":"oa:W4391531696","type":"article-journal","title":"Artificial intelligence in the risk prediction models of cardiovascular disease and development of an independent validation screening tool: a systematic review","abstract":"BACKGROUND: A comprehensive overview of artificial intelligence (AI) for cardiovascular disease (CVD) prediction and a screening tool of AI models (AI-Ms) for independent external validation are lacking. This systematic review aims to identify, describe, and appraise AI-Ms of CVD prediction in the general and special populations and develop a new independent validation score (IVS) for AI-Ms replicability evaluation. METHODS: PubMed, Web of Science, Embase, and IEEE library were searched up to July 2021. Data extraction and analysis were performed for the populations, distribution, predictors, algorithms, etc. The risk of bias was evaluated with the prediction risk of bias assessment tool (PROBAST). Subsequently, we designed IVS for model replicability evaluation with five steps in five items, including transparency of algorithms, performance of models, feasibility of reproduction, risk of reproduction, and clinical implication, respectively. The review is registered in PROSPERO (No. CRD42021271789). RESULTS: In 20,887 screened references, 79 articles (82.5% in 2017-2021) were included, which contained 114 datasets (67 in Europe and North America, but 0 in Africa). We identified 486 AI-Ms, of which the majority were in development (n = 380), but none of them had undergone independent external validation. A total of 66 idiographic algorithms were found; however, 36.4% were used only once and only 39.4% over three times. A large number of different predictors (range 5-52,000, median 21) and large-span sample size (range 80-3,660,000, median 4466) were observed. All models were at high risk of bias according to PROBAST, primarily due to the incorrect use of statistical methods. IVS analysis confirmed only 10 models as \"recommended\"; however, 281 and 187 were \"not recommended\" and \"warning,\" respectively. CONCLUSION: AI has led the digital revolution in the field of CVD prediction, but is still in the early stage of development as the defects of research design, report, and evaluation systems. The IVS we developed may contribute to independent external validation and the development of this field.","author":[{"family":"Cai","given":"Yue"},{"family":"Cai","given":"Yuqing"},{"family":"Tang","given":"Liying"},{"family":"Wang","given":"Yihan"},{"family":"Gong","given":"Mengchun"},{"family":"Jing","given":"Tian"},{"family":"Li","given":"Huijun"},{"family":"Liling","given":"Jesse"},{"family":"Hu","given":"Wei"},{"family":"Yin","given":"Zhihua"},{"family":"Gong","given":"Da"},{"family":"Zhang","given":"Guangwei"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s12916-024-03273-7","URL":"https://doi.org/10.1186/s12916-024-03273-7","source":"openalex"},{"id":"oa:W4405494680","type":"article-journal","title":"Applications of artificial intelligence in current pharmacy practice: A scoping review","abstract":"BACKGROUND: Artificial intelligence (AI), a branch of computer science, has been of growing research interest since its introduction to healthcare disciplines in the 1970s. Research has demonstrated that the application of such technologies has allowed for greater task accuracy and efficiency in medical disciplines such as diagnostics, treatment protocols and clinical decision-making. Application in pharmacy practice is reportedly narrower in scope; with greater emphasis placed on stock management and day-to-day function optimisation than enhancing patient outcomes. Despite this, new studies are underway to explore how AI technologies may be utilised in areas such as pharmacist interventions, medication adherence, and personalised medicine. Objective/s: The aim of this study was to identify current use of AI in measuring performance outcomes in pharmacy practice. METHODS: A scoping review was conducted in accordance with PRISMA Extension for Scoping Reviews (PRISMA-ScR). A comprehensive literature search was conducted in MEDLINE, Embase, IPA (International Pharmaceutical Abstracts), and Web of Science databases for articles published between January 1, 2018 to September 11, 2023, relevant to the aim. The final search strategy included the following terms: (\"artificial intelligence\") AND (\"pharmacy\" OR \"pharmacist\" OR \"pharmaceutical service\" OR \"pharmacy service\"). Reference lists of identified review articles were also screened. RESULTS: The literature search identified 560 studies, of which seven met the inclusion criteria. These studies described the use of AI in pharmacy practice. All seven studies utilised models derived from machine learning AI techniques. AI identification of prescriptions requiring pharmacist intervention was the most frequent (n = 4), followed by screening services (n = 2), and patient-facing mobile applications (n = 1). These results indicated a workflow- and productivity-focused application of AI within current pharmacy practice, with minimal intention for direct patient health outcome improvement. Despite this, the review also revealed AI's potential in data collation and analytics to aid in pharmacist contribution towards the healthcare team and improvement of health outcomes. CONCLUSIONS: This scoping review has identified, from the literature available, three main areas of focus, (1) identification and classification of atypical or inappropriate medication orders, (2) improving efficiency of mass screening services, and (3) improving adherence and quality use of medicines. It also identified gaps in AI's current utility within the profession and its potential for day-to-day practice, as our understanding of general AI techniques continues to advance.","author":[{"family":"Jessica","given":"Hatzimanolis"},{"family":"Britney","given":"Riley"},{"family":"Sarira","given":"El"},{"family":"Aslani","given":"Parisa"},{"family":"Joe","given":"Zhou"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.sapharm.2024.12.007","URL":"https://doi.org/10.1016/j.sapharm.2024.12.007","source":"openalex"},{"id":"oa:W4324046518","type":"article-journal","title":"Chatting and cheating: Ensuring academic integrity in the era of ChatGPT","abstract":"The use of artificial intelligence in academia is a hot topic in the education field. ChatGPT is an AI tool that offers a range of benefits, including increased student engagement, collaboration, and accessibility. However, is also raises concerns regarding academic honesty and plagiarism. This paper examines the opportunities and challenges of using ChatGPT in higher education, and discusses the potential risks and rewards of these tools. The paper also considers the difficulties of detecting and preventing academic dishonesty, and suggests strategies that universities can adopt to ensure ethical and responsible use of these tools. These strategies include developing policies and procedures, providing training and support, and using various methods to detect and prevent cheating. The paper concludes that while the use of AI in higher education presents both opportunities and challenges, universities can effectively address these concerns by taking a proactive and ethical approach to the use of these tools.","author":[{"family":"Cotton","given":"Debby"},{"family":"Cotton","given":"Peter"},{"family":"Shipway","given":"JR"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1080/14703297.2023.2190148","URL":"https://doi.org/10.1080/14703297.2023.2190148","source":"openalex"},{"id":"oa:W4391042079","type":"article-journal","title":"Theory‐Driven Perspectives on Generative Artificial Intelligence in Business and Management","abstract":"Shuang Ren, Riikka M. Sarala, Paul Hibbert The advent of generative artificial intelligence (GAI) has sparked both enthusiasm and anxiety as different stakeholders grapple with the potential to reshape the business and management landscape. This dynamic discourse extends beyond GAI itself to encompass closely related innovations that have existed for some time, for example, machine learning, thereby creating a collective anticipation of opportunities and dilemmas surrounding the transformative or disruptive capacities of these emerging technologies. Recently, ChatGPT's ability to access information from the web in real time marks a significant advancement with profound implications for businesses. This feature is argued to enhance the model's capacity to provide up-to-date, contextually relevant information, enabling more dynamic customer interactions. For businesses, this could mean improvements in areas like market analysis, trend tracking, customer service and real-time data-driven problem-solving. However, this also raises concerns about the accuracy and reliability of the information sourced, given the dynamic and sometimes unverified nature of web content. Additionally, real-time web access might complicate data privacy and security, as the boundaries of GAI interactions extend into the vast and diverse Internet landscape. These factors necessitate a careful and responsible approach to evaluating and using advanced GAI capabilities in business and management contexts. GAI is attracting much interest both in the academic and business practitioner literature. A quick search in Google Scholar, using the search terms ‘generative artificial intelligence’ and ‘business’ or ‘management’, yields approximately 1740 results. Within this extensive repository, scholars delve into diverse facets, exploring GAI's potential applications across various business and management functions, contemplating its implications for management educators and scrutinizing specific technological applications. Learned societies such as the British Academy of Management have also joined forces in leading the discussion on AI and digitalization in business and management academe. Meanwhile, practitioners and consultants alike (e.g. McKinsey & Company, PWC, World Economic Forum) have produced dedicated discussions, reports and forums to offer insights into the multifaceted impacts and considerations surrounding the integration of GAI in contemporary business and management practices. Table 1 illustrates some current applications of GAI as documented in the practitioner literature. Zalando [online platform for fashion and lifestyle] Instacart [e-commerce application] Salesforce [cloud-based customer relationship software provider] DHL [logistics provider] Coca-Cola [beverage company] Nestlé and Mondelez [confectionary] Heinz [food processing company] Air India [airline] Duolingo [language learning application] Mastercard [financial services] In an attempt to capture the new opportunities and challenges brought about by this technology and to hopefully find a way forward to guide research and practice, management journals have been swift to embrace the trend, introducing special issues on GAI. These issues aim to promote intellectual debate, for instance in relation to specific business disciplines (e.g. Benbya, Pachidi and Jarvenpaa, 2021) or organizational possibilities and pitfalls (Chalmers et al., 2023). However, amidst these commendable efforts that reflect a broad spectrum of perspectives, a critical examination of the burgeoning hype around GAI reveals a significant gap. Despite the proliferation of discussions from scholars, practitioners and the general public, the prevailing discourse is often speculative, lacking a robust theoretical foundation. This deficiency points to the challenges to existing theories in terms of their efficacy in explaining the unique demands created by GAI and indicates an urgent need for refining prior theories or even r","author":[{"family":"Brown","given":"Olivia"},{"family":"Davison","given":"Robert"},{"family":"Decker","given":"Stephanie"},{"family":"Ellis","given":"David"},{"family":"Faulconbridge","given":"James"},{"family":"Gore","given":"Julie"},{"family":"Greenwood","given":"Michelle"},{"family":"Islam","given":"Gazi"},{"family":"Lubinski","given":"Christina"},{"family":"Mackenzie","given":"Niall"},{"family":"Meyer","given":"Renate"},{"family":"Muzio","given":"Daniel"},{"family":"Quattrone","given":"Paolo"},{"family":"Ravishankar","given":"MN"},{"family":"Zilber","given":"Tammar"},{"family":"Ren","given":"Shuang"},{"family":"Sarala","given":"Riikka"},{"family":"Hibbert","given":"Paul"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1111/1467-8551.12788","URL":"https://doi.org/10.1111/1467-8551.12788","source":"openalex"},{"id":"oa:W4392014371","type":"article-journal","title":"Innovative applications of artificial intelligence during the COVID-19 pandemic","abstract":"The COVID-19 pandemic has created unprecedented challenges worldwide. Artificial intelligence (AI) technologies hold tremendous potential for tackling key aspects of pandemic management and response. In the present review, we discuss the tremendous possibilities of AI technology in addressing the global challenges posed by the COVID-19 pandemic. First, we outline the multiple impacts of the current pandemic on public health, the economy, and society. Next, we focus on the innovative applications of advanced AI technologies in key areas such as COVID-19 prediction, detection, control, and drug discovery for treatment. Specifically, AI-based predictive analytics models can use clinical, epidemiological, and omics data to forecast disease spread and patient outcomes. Additionally, deep neural networks enable rapid diagnosis through medical imaging. Intelligent systems can support risk assessment, decision-making, and social sensing, thereby improving epidemic control and public health policies. Furthermore, high-throughput virtual screening enables AI to accelerate the identification of therapeutic drug candidates and opportunities for drug repurposing. Finally, we discuss future research directions for AI technology in combating COVID-19, emphasizing the importance of interdisciplinary collaboration. Though promising, barriers related to model generalization, data quality, infrastructure readiness, and ethical risks must be addressed to fully translate these innovations into real-world impacts. Multidisciplinary collaboration engaging diverse expertise and stakeholders is imperative for developing robust, responsible, and human-centered AI solutions against COVID-19 and future public health emergencies.","author":[{"family":"Lv","given":"Chenrui"},{"family":"Guo","given":"Wenqiang"},{"family":"Yin","given":"Xinyi"},{"family":"Liu","given":"Liu"},{"family":"Huang","given":"Xinlei"},{"family":"Li","given":"Shimin"},{"family":"Zhang","given":"Li"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.imj.2024.100095","URL":"https://doi.org/10.1016/j.imj.2024.100095","source":"openalex"},{"id":"oa:W4392938070","type":"article-journal","title":"ChatGPT in medicine: prospects and challenges: a review article","abstract":"It has been a year since the launch of Chat Generator Pre-Trained Transformer (ChatGPT), a generative artificial intelligence (AI) program. The introduction of this cross-generational product initially brought a huge shock to people with its incredible potential and then aroused increasing concerns among people. In the field of medicine, researchers have extensively explored the possible applications of ChatGPT and achieved numerous satisfactory results. However, opportunities and issues always come together. Problems have also been exposed during the applications of ChatGPT, requiring cautious handling, thorough consideration, and further guidelines for safe use. Here, the authors summarized the potential applications of ChatGPT in the medical field, including revolutionizing healthcare consultation, assisting patient management and treatment, transforming medical education, and facilitating clinical research. Meanwhile, the authors also enumerated researchers' concerns arising along with its broad and satisfactory applications. As it is irreversible that AI will gradually permeate every aspect of modern life, the authors hope that this review can not only promote people's understanding of the potential applications of ChatGPT in the future but also remind them to be more cautious about this \"Pandora's Box\" in the medical field. It is necessary to establish normative guidelines for its safe use in the medical field as soon as possible.","author":[{"family":"Tan","given":"Songtao"},{"family":"Xin","given":"Xin"},{"family":"Wu","given":"Di"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1097/js9.0000000000001312","URL":"https://doi.org/10.1097/js9.0000000000001312","source":"openalex"},{"id":"oa:W4387709287","type":"article-journal","title":"Artificial intelligence education: An evidence-based medicine approach for consumers, translators, and developers","abstract":"Current and future healthcare professionals are generally not trained to cope with the proliferation of artificial intelligence (AI) technology in healthcare. To design a curriculum that caters to variable baseline knowledge and skills, clinicians may be conceptualized as \"consumers\", \"translators\", or \"developers\". The changes required of medical education because of AI innovation are linked to those brought about by evidence-based medicine (EBM). We outline a core curriculum for AI education of future consumers, translators, and developers, emphasizing the links between AI and EBM, with suggestions for how teaching may be integrated into existing curricula. We consider the key barriers to implementation of AI in the medical curriculum: time, resources, variable interest, and knowledge retention. By improving AI literacy rates and fostering a translator- and developer-enriched workforce, innovation may be accelerated for the benefit of patients and practitioners.","author":[{"family":"Ng","given":"Faye"},{"family":"Thirunavukarasu","given":"Arun"},{"family":"Cheng","given":"Haoran"},{"family":"Tan","given":"Ting"},{"family":"Gutiérrez","given":"Laura"},{"family":"Lan","given":"Yanyan"},{"family":"Ong","given":"Jasmine"},{"family":"Chong","given":"Yap"},{"family":"Ngiam","given":"Kee"},{"family":"Ho","given":"Dean"},{"family":"Wong","given":"Tien"},{"family":"Kwek","given":"Kenneth"},{"family":"Doshivelez","given":"Finale"},{"family":"Lucey","given":"Catherine"},{"family":"Coffman","given":"Thomas"},{"family":"Ting","given":"Daniel"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.xcrm.2023.101230","URL":"https://doi.org/10.1016/j.xcrm.2023.101230","source":"openalex"},{"id":"oa:W4400348220","type":"article-journal","title":"Research integrity in the era of artificial intelligence: Challenges and responses","abstract":"The application of artificial intelligence (AI) technologies in scientific research has significantly enhanced efficiency and accuracy but also introduced new forms of academic misconduct, such as data fabrication and text plagiarism using AI algorithms. These practices jeopardize research integrity and can mislead scientific directions. This study addresses these challenges, underscoring the need for the academic community to strengthen ethical norms, enhance researcher qualifications, and establish rigorous review mechanisms. To ensure responsible and transparent research processes, we recommend the following specific key actions: Development and enforcement of comprehensive AI research integrity guidelines that include clear protocols for AI use in data analysis and publication, ensuring transparency and accountability in AI-assisted research. Implementation of mandatory AI ethics and integrity training for researchers, aimed at fostering an in-depth understanding of potential AI misuses and promoting ethical research practices. Establishment of international collaboration frameworks to facilitate the exchange of best practices and development of unified ethical standards for AI in research. Protecting research integrity is paramount for maintaining public trust in science, making these recommendations urgent for the scientific community consideration and action.","author":[{"family":"Chen","given":"Ziyu"},{"family":"Chen","given":"Chang"},{"family":"Yang","given":"Guozhao"},{"family":"He","given":"Xiangpeng"},{"family":"Chi","given":"Xiaoxia"},{"family":"Zeng","given":"Zhuoying"},{"family":"Chen","given":"Xuhong"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1097/md.0000000000038811","URL":"https://doi.org/10.1097/md.0000000000038811","source":"openalex"},{"id":"oa:W4381250863","type":"article-journal","title":"A Survey of Privacy Risks and Mitigation Strategies in the Artificial Intelligence Life Cycle","abstract":"Over the decades, Artificial Intelligence (AI) and machine learning has become a transformative solution in many sectors, services, and technology platforms in a wide range of applications, such as in smart healthcare, financial, political, and surveillance systems. In such applications, a large amount of data is generated about diverse aspects of our life. Although utilizing AI in real-world applications provides numerous opportunities for societies and industries, it raises concerns regarding data privacy. Data used in an AI system are cleaned, integrated, and processed throughout the AI life cycle. Each of these stages can introduce unique threats to individual’s privacy and have an impact on ethical processing and protection of data. In this paper, we examine privacy risks in different phases of the AI life cycle and review the existing privacy-enhancing solutions. We introduce four different categories of privacy risk, including (i) risk of identification, (ii) risk of making an inaccurate decision, (iii) risk of non-transparency in AI systems, and (iv) risk of non-compliance with privacy regulations and best practices. We then examined the potential privacy risks in each AI life cycle phase, evaluated concerns, and reviewed privacy-enhancing technologies, requirements, and process solutions to countermeasure these risks. We also reviewed some of the existing privacy protection policies and the need for compliance with available privacy regulations in AI-based systems. The main contribution of this survey is examining privacy challenges and solutions, including technology, process, and privacy legislation in the entire AI life cycle. In each phase of the AI life cycle, open challenges have been identified.","author":[{"family":"Shahriar","given":"Sakib"},{"family":"Allana","given":"Sonal"},{"family":"Hazratifard","given":"Seyed"},{"family":"Dara","given":"Rozita"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/access.2023.3287195","URL":"https://doi.org/10.1109/access.2023.3287195","source":"openalex"},{"id":"oa:W4389728663","type":"article-journal","title":"Exploring knowledge, attitudes, and practices towards artificial intelligence among health professions’ students in Jordan","abstract":"INTRODUCTION: The integration of Artificial Intelligence (AI) in medical education and practice is a significant development. This study examined the Knowledge, Attitudes, and Practices (KAP) of health professions' students in Jordan concerning AI, providing insights into their preparedness and perceptions. METHODS: An online questionnaire was distributed to 483 Jordanian health professions' students via social media. Demographic data, AI-related KAP, and barriers were collected. Quantile regression models analyzed associations between variables and KAP scores. RESULTS: Moderate AI knowledge was observed among participants, with specific understanding of data requirements and barriers. Attitudes varied, combining skepticism about AI replacing human teachers with recognition of its value. While AI tools were used for specific tasks, broader integration in medical education and practice was limited. Barriers included lack of knowledge, access, time constraints, and curriculum gaps. CONCLUSIONS: This study highlights the need to enhance medical education with AI topics and address barriers. Students need to be better prepared for AI integration, in order to enable medical education to harness AI's potential for improved patient care and training.","author":[{"family":"Alqerem","given":"Walid"},{"family":"Eberhardt","given":"Judith"},{"family":"Jarab","given":"Anan"},{"family":"Bawab","given":"Abdel"},{"family":"Hammad","given":"Alaa"},{"family":"Alasmari","given":"Fawaz"},{"family":"Alazab","given":"Badi’ah"},{"family":"Husein","given":"Daoud"},{"family":"Alazab","given":"Jumana"},{"family":"Al-Beool","given":"Saed"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1186/s12911-023-02403-0","URL":"https://doi.org/10.1186/s12911-023-02403-0","source":"openalex"},{"id":"oa:W4396946008","type":"article-journal","title":"Relationship between teachers’ digital competence and attitudes towards artificial intelligence in education","abstract":"With the recent integration of artificial intelligence (AI) in the educational field, understanding the variables that are related to teacher attitudes towards AI can be crucial for understanding their perspectives in the classroom. That is why the present study aimed to investigate whether Teacher Digital Competence is related to Teacher Attitudes towards AI, and if so, whether this relationship is moderated by the teacher's educational stage, age, sex, years of experience, and field of knowledge. A total of 445 Spanish teachers from primary, secondary, and higher education participated in this study, responding to the Teacher Digital Competence Scale and the Teacher Attitudes towards AI Scale. The results revealed that, regardless of educational stage, sex, age, years of experience or field of knowledge, higher teacher digital competence is associated with a more positive teacher attitude towards AI. Moreover, high levels of willingness to use AI but low levels of personal experience with AI were found. Based on these results, it may be interesting to implement future interventions based on AI to enhance key dimensions of teacher digital competence, such as Information Management, Content Creation, and Problem-Solving. This could improve Teacher Digital Competence and subsequently enhance teachers' perception of using artificial intelligence in the educational context.","author":[{"family":"Galindodomínguez","given":"Héctor"},{"family":"Delgado","given":"Nahia"},{"family":"Campo","given":"Lucía"},{"family":"Iglesias","given":"Daniel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1016/j.ijer.2024.102381","URL":"https://doi.org/10.1016/j.ijer.2024.102381","source":"openalex"},{"id":"oa:W4392450439","type":"article-journal","title":"Dr. Google to Dr. ChatGPT: assessing the content and quality of artificial intelligence-generated medical information on appendicitis","abstract":"INTRODUCTION: Generative artificial intelligence (AI) chatbots have recently been posited as potential sources of online medical information for patients making medical decisions. Existing online patient-oriented medical information has repeatedly been shown to be of variable quality and difficult readability. Therefore, we sought to evaluate the content and quality of AI-generated medical information on acute appendicitis. METHODS: A modified DISCERN assessment tool, comprising 16 distinct criteria each scored on a 5-point Likert scale (score range 16-80), was used to assess AI-generated content. Readability was determined using the Flesch Reading Ease (FRE) and Flesch-Kincaid Grade Level (FKGL) scores. Four popular chatbots, ChatGPT-3.5 and ChatGPT-4, Bard, and Claude-2, were prompted to generate medical information about appendicitis. Three investigators independently scored the generated texts blinded to the identity of the AI platforms. RESULTS: ChatGPT-3.5, ChatGPT-4, Bard, and Claude-2 had overall mean (SD) quality scores of 60.7 (1.2), 62.0 (1.0), 62.3 (1.2), and 51.3 (2.3), respectively, on a scale of 16-80. Inter-rater reliability was 0.81, 0.75, 0.81, and 0.72, respectively, indicating substantial agreement. Claude-2 demonstrated a significantly lower mean quality score compared to ChatGPT-4 (p = 0.001), ChatGPT-3.5 (p = 0.005), and Bard (p = 0.001). Bard was the only AI platform that listed verifiable sources, while Claude-2 provided fabricated sources. All chatbots except for Claude-2 advised readers to consult a physician if experiencing symptoms. Regarding readability, FKGL and FRE scores of ChatGPT-3.5, ChatGPT-4, Bard, and Claude-2 were 14.6 and 23.8, 11.9 and 33.9, 8.6 and 52.8, 11.0 and 36.6, respectively, indicating difficulty readability at a college reading skill level. CONCLUSION: AI-generated medical information on appendicitis scored favorably upon quality assessment, but most either fabricated sources or did not provide any altogether. Additionally, overall readability far exceeded recommended levels for the public. Generative AI platforms demonstrate measured potential for patient education and engagement about appendicitis.","author":[{"family":"Ghanem","given":"Yazid"},{"family":"Rouhi","given":"Armaun"},{"family":"Al-Houssan","given":"Ammr"},{"family":"Saleh","given":"Zena"},{"family":"Moccia","given":"Matthew"},{"family":"Joshi","given":"Hansa"},{"family":"Dumon","given":"Kristoffel"},{"family":"Hong","given":"Young"},{"family":"Spitz","given":"Francis"},{"family":"Joshi","given":"Amit"},{"family":"Kwiatt","given":"Michael"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1007/s00464-024-10739-5","URL":"https://doi.org/10.1007/s00464-024-10739-5","source":"openalex"},{"id":"oa:W4315606155","type":"article-journal","title":"The Role of Artificial Intelligence in Future Rehabilitation Services: A Systematic Literature Review","abstract":"Artificial intelligence technologies are considered crucial in supporting a decentralized model of care in which therapeutic interventions are provided from a distance. In the last years, various approaches have been proposed to support remote monitoring and smart assistance in rehabilitation services. Comprehensive state-of-the-art of machine learning methods and applications is presented in this review. Following PRISMA guidelines, a systematic literature search strategy was led in PubMed, Scopus, and IEEE Xplore databases. The search yielded 519 records, resulting in 35 articles included in this study. Supervised and unsupervised machine learning algorithms were identified. Unobtrusive capture motion technologies have been identified as strategic applications to support remote and smart monitoring. The main tasks addressed by algorithms were activity recognition, movement classification, and clinical status prediction. Some authors evidenced drawbacks concerning the low generalizability of the results retrieved. Artificial intelligence-based applications are likely to impact the delivery of decentralized rehabilitation services by providing broad access to sustained and high-quality therapy. Future efforts are needed to validate artificial intelligence technologies in specific clinical populations and evaluate results reliability in remote conditions and home-based settings.","author":[{"family":"Mennella","given":"Ciro"},{"family":"Maniscalco","given":"Umberto"},{"family":"Pietro","given":"Giuseppe"},{"family":"Esposito","given":"Massimo"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1109/access.2023.3236084","URL":"https://doi.org/10.1109/access.2023.3236084","source":"openalex"},{"id":"oa:W4319869596","type":"article-journal","title":"Artificial intelligence in multi-objective drug design","abstract":"The factors determining a drug's success are manifold, making de novo drug design an inherently multi-objective optimisation (MOO) problem. With the advent of machine learning and optimisation methods, the field of multi-objective compound design has seen a rapid increase in developments and applications. Population-based metaheuris-tics and deep reinforcement learning are the most commonly used artificial intelligence methods in the field, but recently conditional learning methods are gaining popularity. The former approaches are coupled with a MOO strat-egy which is most commonly an aggregation function, but Pareto-based strategies are widespread too. Besides these and conditional learning, various innovative approaches to tackle MOO in drug design have been proposed. Here we provide a brief overview of the field and the latest innovations.","author":[{"family":"Luukkonen","given":"Sohvi"},{"family":"Maagdenberg","given":"Helle"},{"family":"Emmerich","given":"Michael"},{"family":"Westen","given":"Gerard"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.sbi.2023.102537","URL":"https://doi.org/10.1016/j.sbi.2023.102537","source":"openalex"},{"id":"oa:W4353083153","type":"article-journal","title":"Sports analytics review: Artificial intelligence applications, emerging technologies, and algorithmic perspective","abstract":"Abstract The rapid and impromptu interest in the coupling of machine learning (ML) algorithms with wearable and contactless sensors aimed at tackling real‐world problems warrants a pedagogical study to understand all the aspects of this research direction. Considering this aspect, this survey aims to review the state‐of‐the‐art literature on ML algorithms, methodologies, and hypotheses adopted to solve the research problems and challenges in the domain of sports. First, we categorize this study into three main research fields: sensors, computer vision, and wireless and mobile‐based applications. Then, for each of these fields, we thoroughly analyze the systems that are deployable for real‐time sports analytics. Next, we meticulously discuss the learning algorithms (e.g., statistical learning, deep learning, reinforcement learning) that power those deployable systems while also comparing and contrasting the benefits of those learning methodologies. Finally, we highlight the possible future open‐research opportunities and emerging technologies that could contribute to the domain of sports analytics. This article is categorized under: Technologies > Machine Learning Technologies > Artificial Intelligence Technologies > Internet of Things","author":[{"family":"Ghosh","given":"Indrajeet"},{"family":"Ramamurthy","given":"Sreenivasan"},{"family":"Chakma","given":"Avijoy"},{"family":"Roy","given":"Nirmalya"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1002/widm.1496","URL":"https://doi.org/10.1002/widm.1496","source":"openalex"},{"id":"oa:W4387357807","type":"article-journal","title":"Generative Artificial Intelligence for Chest Radiograph Interpretation in the Emergency Department","abstract":"Importance: Multimodal generative artificial intelligence (AI) methodologies have the potential to optimize emergency department care by producing draft radiology reports from input images. Objective: To evaluate the accuracy and quality of AI-generated chest radiograph interpretations in the emergency department setting. Design, Setting, and Participants: This was a retrospective diagnostic study of 500 randomly sampled emergency department encounters at a tertiary care institution including chest radiographs interpreted by both a teleradiology service and on-site attending radiologist from January 2022 to January 2023. An AI interpretation was generated for each radiograph. The 3 radiograph interpretations were each rated in duplicate by 6 emergency department physicians using a 5-point Likert scale. Main Outcomes and Measures: The primary outcome was any difference in Likert scores between radiologist, AI, and teleradiology reports, using a cumulative link mixed model. Secondary analyses compared the probability of each report type containing no clinically significant discrepancy with further stratification by finding presence, using a logistic mixed-effects model. Physician comments on discrepancies were recorded. Results: A total of 500 ED studies were included from 500 unique patients with a mean (SD) age of 53.3 (21.6) years; 282 patients (56.4%) were female. There was a significant association of report type with ratings, with post hoc tests revealing significantly greater scores for AI (mean [SE] score, 3.22 [0.34]; P < .001) and radiologist (mean [SE] score, 3.34 [0.34]; P < .001) reports compared with teleradiology (mean [SE] score, 2.74 [0.34]) reports. AI and radiologist reports were not significantly different. On secondary analysis, there was no difference in the probability of no clinically significant discrepancy between the 3 report types. Further stratification of reports by presence of cardiomegaly, pulmonary edema, pleural effusion, infiltrate, pneumothorax, and support devices also yielded no difference in the probability of containing no clinically significant discrepancy between the report types. Conclusions and Relevance: In a representative sample of emergency department chest radiographs, results suggest that the generative AI model produced reports of similar clinical accuracy and textual quality to radiologist reports while providing higher textual quality than teleradiologist reports. Implementation of the model in the clinical workflow could enable timely alerts to life-threatening pathology while aiding imaging interpretation and documentation.","author":[{"family":"Huang","given":"Jonathan"},{"family":"Neill","given":"Luke"},{"family":"Wittbrodt","given":"Matthew"},{"family":"Melnick","given":"David"},{"family":"Klug","given":"Matthew"},{"family":"Thompson","given":"Michael"},{"family":"Bailitz","given":"John"},{"family":"Loftus","given":"Timothy"},{"family":"Malik","given":"Sanjeev"},{"family":"Phull","given":"Amit"},{"family":"Weston","given":"Victoria"},{"family":"Heller","given":"JA"},{"family":"Etemadi","given":"Mozziyar"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1001/jamanetworkopen.2023.36100","URL":"https://doi.org/10.1001/jamanetworkopen.2023.36100","source":"openalex"},{"id":"oa:W4318478060","type":"article-journal","title":"The collaborative role of blockchain, artificial intelligence, and industrial internet of things in digitalization of small and medium-size enterprises","abstract":"Due to digitalization, small and medium-sized enterprises (SMEs) have significantly enhanced their efficiency and productivity in the past few years. The process to automate SME transaction execution is getting highly multifaceted as the number of stakeholders of SMEs is connecting, accessing, exchanging, adding, and changing the transactional executions. The balanced lifecycle of SMEs requires partnership exchanges, financial management, manufacturing, and productivity stabilities, along with privacy and security. Interoperability platform issue is another critical challenging aspect while designing and managing a secure distributed Peer-to-Peer industrial development environment for SMEs. However, till now, it is hard to maintain operations of SMEs' integrity, transparency, reliability, provenance, availability, and trustworthiness between two different enterprises due to the current nature of centralized server-based infrastructure. This paper bridges these problems and proposes a novel and secure framework with a standardized process hierarchy/lifecycle for distributed SMEs using collaborative techniques of blockchain, the internet of things (IoT), and artificial intelligence (AI) with machine learning (ML). A blockchain with IoT-enabled permissionless network structure is designed called \"B-SMEs\" that provides solutions to cross-chain platforms. In this, B-SMEs address the lightweight stakeholder authentication problems as well. For that purpose, three different chain codes are deployed. It handles participating SMEs' registration, day-to-day information management and exchange between nodes, and analysis of partnership exchange-related transaction details before being preserved on the blockchain immutable storage. Whereas AI-enabled ML-based artificial neural networks are utilized, the aim is to handle and optimize day-to-day numbers of SME transactions; so that the proposed B-SMEs consume fewer resources in terms of computational power, network bandwidth, and preservation-related issues during the complete process of SMEs service deliverance. The simulation results present highlight the benefits of B-SMEs, increases the rate of ledger management and optimization while exchanging information between different chains, which is up to 17.3%, and reduces the consumption of the system's computational resources down to 9.13%. Thus, only 14.11% and 7.9% of B-SME's transactions use network bandwidth and storage capabilities compared to the current mechanism of SMEs, respectively.","author":[{"family":"Khan","given":"Abdullah"},{"family":"Laghari","given":"Asif"},{"family":"Li","given":"Peng"},{"family":"Dootio","given":"Mazhar"},{"family":"Karim","given":"Shahid"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1038/s41598-023-28707-9","URL":"https://doi.org/10.1038/s41598-023-28707-9","source":"openalex"},{"id":"oa:W4387936071","type":"article-journal","title":"The influence of artificial intelligence on the work of the medical physicist in radiotherapy practice: a short review","abstract":"Abstract There have been many applications and influences of Artificial intelligence (AI) in many sectors and its professionals, that of radiotherapy and the medical physicist is no different. AI and technological advances have necessitated changing roles of medical physicists due to the development of modernized technology with image-guided accessories for the radiotherapy treatment of cancer patients. Given the changing role of medical physicists in ensuring patient safety and optimal care, AI can reshape radiotherapy practice now and in some years to come. Medical physicists’ roles in radiotherapy practice have evolved to meet technology for the management of better patient care in the age of modern radiotherapy. This short review provides an insight into the influence of AI on the changing role of medical physicists in each specific chain of the workflow in radiotherapy in which they are involved.","author":[{"family":"Fiagbedzi","given":"Emmanuel"},{"family":"Hasford","given":"Francis"},{"family":"Tagoe","given":"Samuel"},{"family":"Tagoe","given":"Samuel"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1259/bjro.20230003","URL":"https://doi.org/10.1259/bjro.20230003","source":"openalex"},{"id":"oa:W4361010280","type":"article-journal","title":"Artificial intelligence in public health: the potential of epidemic early warning systems","abstract":"The use of artificial intelligence (AI) to generate automated early warnings in epidemic surveillance by harnessing vast open-source data with minimal human intervention has the potential to be both revolutionary and highly sustainable. AI can overcome the challenges faced by weak health systems by detecting epidemic signals much earlier than traditional surveillance. AI-based digital surveillance is an adjunct to-not a replacement of-traditional surveillance and can trigger early investigation, diagnostics and responses at the regional level. This narrative review focuses on the role of AI in epidemic surveillance and summarises several current epidemic intelligence systems including ProMED-mail, HealthMap, Epidemic Intelligence from Open Sources, BlueDot, Metabiota, the Global Biosurveillance Portal, Epitweetr and EPIWATCH. Not all of these systems are AI-based, and some are only accessible to paid users. Most systems have large volumes of unfiltered data; only a few can sort and filter data to provide users with curated intelligence. However, uptake of these systems by public health authorities, who have been slower to embrace AI than their clinical counterparts, is low. The widespread adoption of digital open-source surveillance and AI technology is needed for the prevention of serious epidemics.","author":[{"family":"Macintyre","given":"CR"},{"family":"Chen","given":"Xin"},{"family":"Kunasekaran","given":"Mohana"},{"family":"Quigley","given":"Ashley"},{"family":"Lim","given":"Samsung"},{"family":"Stone","given":"Haley"},{"family":"Paik","given":"Hye"},{"family":"Yao","given":"Lina"},{"family":"Heslop","given":"David"},{"family":"Wei","given":"Wenzhao"},{"family":"Sarmiento","given":"Ines"},{"family":"Gurdasani","given":"Deepti"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1177/03000605231159335","URL":"https://doi.org/10.1177/03000605231159335","source":"openalex"},{"id":"oa:W4381384569","type":"article-journal","title":"Artificial intelligence-based diagnosis of Alzheimer's disease with brain MRI images","abstract":"Alzheimer's disease, a primary neurodegenerative condition, predominantly impacts the elderly and pre-elderly population. This progressive neurological disorder is characterized by an array of symptoms including memory loss, cognitive decline, and various physiological and psychological disturbances, significantly compromising the quality of life of patients and their caregivers. Recent advancements in Magnetic Resonance Imaging (MRI) technology have catalyzed research in AI-enhanced diagnostics for Alzheimer's disease, fostering optimism for early detection and timely interventions. This progress has paved the way for the development of sophisticated algorithms and models adept at analyzing complex brain imaging data, thereby augmenting diagnostic accuracy and efficiency. This advancement fuels optimism regarding the transformative potential of AI-driven diagnostics in revolutionizing Alzheimer's disease management, with the prospect of facilitating more effective treatment strategies and improved patient outcomes. The objective of this review is to provide a comprehensive overview of recent developments in deep learning methodologies applied to brain MRI images for the classification of various stages of Alzheimer's disease, with a particular emphasis on early diagnosis. Furthermore, this review underscores the limitations of current research, discussing potential challenges and future research directions in this dynamic field.","author":[{"family":"Yao","given":"Zhaomin"},{"family":"Wang","given":"Hongyu"},{"family":"Yan","given":"WC"},{"family":"Wang","given":"ZQ"},{"family":"Wang","given":"ZQ"},{"family":"Zhang","given":"Wenwen"},{"family":"Wang","given":"Zi"},{"family":"Wang","given":"Zhiguo"},{"family":"Zhang","given":"Guoxu"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1016/j.ejrad.2023.110934","URL":"https://doi.org/10.1016/j.ejrad.2023.110934","source":"openalex"},{"id":"oa:W4322627364","type":"article-journal","title":"Deep Learning with Graph Convolutional Networks: An Overview and Latest Applications in Computational Intelligence","abstract":"Convolutional neural networks (CNNs) have received widespread attention due to their powerful modeling capabilities and have been successfully applied in natural language processing, image recognition, and other fields. On the other hand, traditional CNN can only deal with Euclidean spatial data. In contrast, many real‐life scenarios, such as transportation networks, social networks, reference networks, and so on, exist in graph data. The creation of graph convolution operators and graph pooling is at the heart of migrating CNN to graph data analysis and processing. With the advancement of the Internet and technology, graph convolution network (GCN), as an innovative technology in artificial intelligence (AI), has received more and more attention. GCN has been widely used in different fields such as image processing, intelligent recommender system, knowledge‐based graph, and other areas due to their excellent characteristics in processing non‐European spatial data. At the same time, communication networks have also embraced AI technology in recent years, and AI serves as the brain of the future network and realizes the comprehensive intelligence of the future grid. Many complex communication network problems can be abstracted as graph‐based optimization problems and solved by GCN, thus overcoming the limitations of traditional methods. This survey briefly describes the definition of graph‐based machine learning, introduces different types of graph networks, summarizes the application of GCN in various research fields, analyzes the research status, and gives the future research direction.","author":[{"family":"Bhatti","given":"Uzair"},{"family":"Tang","given":"Hao"},{"family":"Wu","given":"Guilu"},{"family":"Marjan","given":"Shah"},{"family":"Hussain","given":"Aamir"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1155/2023/8342104","URL":"https://doi.org/10.1155/2023/8342104","source":"openalex"},{"id":"oa:W4386254770","type":"article-journal","title":"Cultural Differences in People's Reactions and Applications of Robots, Algorithms, and Artificial Intelligence","abstract":"Abstract Although research in cultural psychology has established that virtually all human behaviors and cognitions are in some ways shaped by culture, culture has been surprisingly absent from the emerging literature on the psychology of technology. In this perspective article, we first review recent findings on machine aversion versus appreciation. We then offer a cross-cultural perspective in understanding how people might react differently to machines. We propose three frameworks – historical, religious, and exposure – to explain how Asians might be more accepting of machines than their Western counterparts. We end the article by discussing three exciting human–machine applications found primarily in Asia and provide future research directions.","author":[{"family":"Yam","given":"Kai"},{"family":"Tan","given":"Tiffany"},{"family":"Jackson","given":"Joshua"},{"family":"Shariff","given":"Azim"},{"family":"Gray","given":"Kurt"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1017/mor.2023.21","URL":"https://doi.org/10.1017/mor.2023.21","source":"openalex"},{"id":"oa:W4385263677","type":"article-journal","title":"Artificial intelligence and ChatGPT in Orthopaedics and sports medicine","abstract":"Artificial intelligence (AI) is looked upon nowadays as the potential major catalyst for the fourth industrial revolution. In the last decade, AI use in Orthopaedics increased approximately tenfold. Artificial intelligence helps with tracking activities, evaluating diagnostic images, predicting injury risk, and several other uses. Chat Generated Pre-trained Transformer (ChatGPT), which is an AI-chatbot, represents an extremely controversial topic in the academic community. The aim of this review article is to simplify the concept of AI and study the extent of AI use in Orthopaedics and sports medicine literature. Additionally, the article will also evaluate the role of ChatGPT in scientific research and publications.Level of evidence: Level V, letter to review.","author":[{"family":"Fayed","given":"Aly"},{"family":"Mansur","given":"Nacime"},{"family":"Carvalho","given":"Képler"},{"family":"Behrens","given":"Andrew"},{"family":"Dhooghe","given":"Pieter"},{"family":"Netto","given":"César"}],"issued":{"date-parts":[[2023]]},"DOI":"10.1186/s40634-023-00642-8","URL":"https://doi.org/10.1186/s40634-023-00642-8","source":"openalex"},{"id":"oa:W4402971773","type":"article-journal","title":"Advances and prospects of multi-modal ophthalmic artificial intelligence based on deep learning: a review","abstract":"BACKGROUND: In recent years, ophthalmology has emerged as a new frontier in medical artificial intelligence (AI) with multi-modal AI in ophthalmology garnering significant attention across interdisciplinary research. This integration of various types and data models holds paramount importance as it enables the provision of detailed and precise information for diagnosing eye and vision diseases. By leveraging multi-modal ophthalmology AI techniques, clinicians can enhance the accuracy and efficiency of diagnoses, and thus reduce the risks associated with misdiagnosis and oversight while also enabling more precise management of eye and vision health. However, the widespread adoption of multi-modal ophthalmology poses significant challenges. MAIN TEXT: In this review, we first summarize comprehensively the concept of modalities in the field of ophthalmology, the forms of fusion between modalities, and the progress of multi-modal ophthalmic AI technology. Finally, we discuss the challenges of current multi-modal AI technology applications in ophthalmology and future feasible research directions. CONCLUSION: In the field of ophthalmic AI, evidence suggests that when utilizing multi-modal data, deep learning-based multi-modal AI technology exhibits excellent diagnostic efficacy in assisting the diagnosis of various ophthalmic diseases. Particularly, in the current era marked by the proliferation of large-scale models, multi-modal techniques represent the most promising and advantageous solution for addressing the diagnosis of various ophthalmic diseases from a comprehensive perspective. However, it must be acknowledged that there are still numerous challenges associated with the application of multi-modal techniques in ophthalmic AI before they can be effectively employed in the clinical setting.","author":[{"family":"Wang","given":"Shaopan"},{"family":"He","given":"Xin"},{"family":"Jian","given":"Zhongquan"},{"family":"Li","given":"Jie"},{"family":"Xu","given":"Changsheng"},{"family":"Chen","given":"Yuguang"},{"family":"Liu","given":"Yuwen"},{"family":"Chen","given":"Han"},{"family":"Huang","given":"Caihong"},{"family":"Hu","given":"Jiaoyue"},{"family":"Liu","given":"Zuguo"}],"issued":{"date-parts":[[2024]]},"DOI":"10.1186/s40662-024-00405-1","URL":"https://doi.org/10.1186/s40662-024-00405-1","source":"openalex"},{"id":"doi:10.26181/26378968.v1","type":"article-journal","title":"Enhancing Healthcare through Sensor-Enabled Digital Twins in Smart Environments: A Comprehensive Analysis","abstract":"Abstract: This comprehensive review investigates the transformative potential of sensor-driven digital twin technology in enhancing healthcare delivery within smart environments. We explore the integration of smart environments with sensor technologies, digital health capabilities, and location-based services, focusing on their impacts on healthcare objectives and outcomes. This work analyzes the foundational technologies, encompassing the Internet of Things (IoT), Internet of Medical Things (IoMT), machine learning (ML), and artificial intelligence (AI), that underpin the functionalities within smart environments. We also examine the unique characteristics of smart homes and smart hospitals, highlighting their potential to revolutionize healthcare delivery through remote patient monitoring, telemedicine, and real-time data sharing. The review presents a novel solution framework leveraging sensor-driven digital twins to address both healthcare needs and user requirements. This framework incorporates wearable health devices, AI-driven health analytics, and a proof-of-concept digital twin application. Furthermore, we explore the role of location-based services (LBS) in smart environments, emphasizing their potential to enhance personalized healthcare interventions and emergency response capabilities. By analyzing the technical advancements in sensor technologies and digital twin applications, this review contributes valuable insights to the evolving landscape of smart environments for healthcare. We identify the opportunities and challenges associated with this emerging field and highlight the need for further research to fully realize its potential to improve healthcare delivery and patient well-being.","author":[{"family":"Adibi","given":"Sasan"},{"family":"Rajabifard","given":"A"},{"family":"Shojaei","given":"D"},{"family":"Wickramasinghe","given":"Nilmini"}],"issued":{"date-parts":[[2024]]},"DOI":"10.26181/26378968.v1","URL":"https://doi.org/10.26181/26378968.v1","source":"datacite"},{"id":"doi:10.26181/26378968","type":"article-journal","title":"Enhancing Healthcare through Sensor-Enabled Digital Twins in Smart Environments: A Comprehensive Analysis","abstract":"Abstract: This comprehensive review investigates the transformative potential of sensor-driven digital twin technology in enhancing healthcare delivery within smart environments. We explore the integration of smart environments with sensor technologies, digital health capabilities, and location-based services, focusing on their impacts on healthcare objectives and outcomes. This work analyzes the foundational technologies, encompassing the Internet of Things (IoT), Internet of Medical Things (IoMT), machine learning (ML), and artificial intelligence (AI), that underpin the functionalities within smart environments. We also examine the unique characteristics of smart homes and smart hospitals, highlighting their potential to revolutionize healthcare delivery through remote patient monitoring, telemedicine, and real-time data sharing. The review presents a novel solution framework leveraging sensor-driven digital twins to address both healthcare needs and user requirements. This framework incorporates wearable health devices, AI-driven health analytics, and a proof-of-concept digital twin application. Furthermore, we explore the role of location-based services (LBS) in smart environments, emphasizing their potential to enhance personalized healthcare interventions and emergency response capabilities. By analyzing the technical advancements in sensor technologies and digital twin applications, this review contributes valuable insights to the evolving landscape of smart environments for healthcare. We identify the opportunities and challenges associated with this emerging field and highlight the need for further research to fully realize its potential to improve healthcare delivery and patient well-being.","author":[{"family":"Adibi","given":"Sasan"},{"family":"Rajabifard","given":"A"},{"family":"Shojaei","given":"D"},{"family":"Wickramasinghe","given":"Nilmini"}],"issued":{"date-parts":[[2024]]},"DOI":"10.26181/26378968","URL":"https://doi.org/10.26181/26378968","source":"datacite"},{"id":"doi:10.26181/27184584.v1","type":"article-journal","title":"ChatGPT and generative AI in urology and surgery—A narrative review","abstract":"Introduction: ChatGPT (generative pre-trained transformer [GPT]), developed by OpenAI, is a type of generative artificial intelligence (AI) that has been widely utilised since its public release. It orchestrates an advanced conversational intelligence, producing sophisticated responses to questions. ChatGPT has been successfully demonstrated across several applications in healthcare, including patient management, academic research and clinical trials. We aim to evaluate the different ways ChatGPT has been utilised in urology and more broadly in surgery. Methods: We conducted a literature search of the PubMed and Embase electronic databases for the purpose of writing a narrative review and identified relevant articles on ChatGPT in surgery from the years 2000 to 2023. A PRISMA flow chart was created to highlight the article selection process. The search terms ‘ChatGPT’ and ‘surgery’ were intentionally kept broad given the nascency of the field. Studies unrelated to these terms were excluded. Duplicates were removed. Results: Multiple papers have been published about novel uses of ChatGPT in surgery, ranging from assisting in administrative tasks including answering frequently asked questions, surgical consent, writing operation reports, discharge summaries, grants, journal article drafts, reviewing journal articles and medical education. AI and machine learning has also been extensively researched in surgery with respect to patient diagnosis and predicting outcomes. There are also several limitations with the software including artificial hallucination, bias, out-of-date information and patient confidentiality. Conclusion: The potential of ChatGPT and related generative AI models are vast, heralding the beginning of a new era where AI may eventually become integrated seamlessly into surgical practice. Concerns with this new technology must not be disregarded in the urge to hasten progression, and potential risks impacting patients' interests must be considered. Appropriate regulation and governance of this technology will be key to optimising the benefits and addressing the intricate challenges of healthcare delivery and equity.","author":[{"family":"Qin","given":"S"},{"family":"Chislett","given":"B"},{"family":"Ischia","given":"J"},{"family":"Ranasinghe","given":"Weranja"},{"family":"De Silva","given":"Daswin"},{"family":"Coles-Black","given":"J"},{"family":"Woon","given":"D"},{"family":"Bolton","given":"D"}],"issued":{"date-parts":[[2024]]},"DOI":"10.26181/27184584.v1","URL":"https://doi.org/10.26181/27184584.v1","source":"datacite"},{"id":"doi:10.26181/27184584","type":"article-journal","title":"ChatGPT and generative AI in urology and surgery—A narrative review","abstract":"Introduction: ChatGPT (generative pre-trained transformer [GPT]), developed by OpenAI, is a type of generative artificial intelligence (AI) that has been widely utilised since its public release. It orchestrates an advanced conversational intelligence, producing sophisticated responses to questions. ChatGPT has been successfully demonstrated across several applications in healthcare, including patient management, academic research and clinical trials. We aim to evaluate the different ways ChatGPT has been utilised in urology and more broadly in surgery. Methods: We conducted a literature search of the PubMed and Embase electronic databases for the purpose of writing a narrative review and identified relevant articles on ChatGPT in surgery from the years 2000 to 2023. A PRISMA flow chart was created to highlight the article selection process. The search terms ‘ChatGPT’ and ‘surgery’ were intentionally kept broad given the nascency of the field. Studies unrelated to these terms were excluded. Duplicates were removed. Results: Multiple papers have been published about novel uses of ChatGPT in surgery, ranging from assisting in administrative tasks including answering frequently asked questions, surgical consent, writing operation reports, discharge summaries, grants, journal article drafts, reviewing journal articles and medical education. AI and machine learning has also been extensively researched in surgery with respect to patient diagnosis and predicting outcomes. There are also several limitations with the software including artificial hallucination, bias, out-of-date information and patient confidentiality. Conclusion: The potential of ChatGPT and related generative AI models are vast, heralding the beginning of a new era where AI may eventually become integrated seamlessly into surgical practice. Concerns with this new technology must not be disregarded in the urge to hasten progression, and potential risks impacting patients' interests must be considered. Appropriate regulation and governance of this technology will be key to optimising the benefits and addressing the intricate challenges of healthcare delivery and equity.","author":[{"family":"Qin","given":"S"},{"family":"Chislett","given":"B"},{"family":"Ischia","given":"J"},{"family":"Ranasinghe","given":"Weranja"},{"family":"De Silva","given":"Daswin"},{"family":"Coles-Black","given":"J"},{"family":"Woon","given":"D"},{"family":"Bolton","given":"D"}],"issued":{"date-parts":[[2024]]},"DOI":"10.26181/27184584","URL":"https://doi.org/10.26181/27184584","source":"datacite"},{"id":"doi:10.17605/osf.io/6sjdq","type":"article-journal","title":"Application of Artificial Intelligence in Diet Management of Inflammatory Bowel Disease: a Scoping Review","abstract":"Background Inflammatory bowel disease (IBD) mainly includes Crohn's Disease (CD) and Ulcerative Colitis (Ulcerative Colitis). UC is a group of chronic non-specific inflammatory diseases with abdominal pain, diarrhea, mucous blood stool as the main symptoms. In the past 30 years, the number of patients with IBD has been growing, and the current incidence has not yet reached a peak, and the disease burden will further increase in the future. It is generally believed that diet is closely related to the occurrence, development and prognosis of IBD. Reasonable diet is not only beneficial to correct malnutrition, but also to promote mucosal repair and achieve the purpose of maintaining disease remission. With the continuous expansion of the application of artificial intelligence technology in the field of medical treatment and nursing and the continuous development of precise diet management, the research of artificial intelligence in the field of diet management of IBD patients has gradually increased. However, the specific applications of AI in diet management among patients with IBD and how to address barriers in their current usage is widely unexplored. Objective To review the research status of artificial intelligence (AI) in diet management of patients with IBD. Methods Following the methodology of the scoping review, the following databases were searched to include relevant literature: PubMed, Web of Science, Embase, Cochrane Library, CINAHL, IEEE Xplore, Association for Computing Machinery Digital Library, China Biomedical Literature Database, CNKI, Wanfang and VIP. The search period was from the establishment until March 2024 and the included articles were summarized and analyzed. Title and abstract screening and full-text review are performed using Endnote, where all articles are independently revised by two reviewers. We include articles written in English or Chinese on experimental and observational studies with qualitative, quantitative and mixed methods approaches, which evaluate AI technologies applicable to nursing. Data are extracted to Excel by two researchers and summarized. Results The scoping review will be completed in June 2024. The results will describe where research has been down, what AI technologies have been developed and studied to assist diet management in patients with IBD, how these technologies have been evaluated, how nurses have participated and how ethical issues have been addressed in the research. Conclusion Findings of this review will provide an overall state of artificial intelligence-based applications in diet management among patients with IBD. This information is important in understanding critical issues that require attention for the use of AI in diet management.","author":[{"family":"Li","given":"Yiting"},{"family":"Tu","given":"Wenjing"},{"family":"Mei","given":"Ziqi"},{"family":"Yin","given":"Tingting"}],"issued":{"date-parts":[[2024]]},"DOI":"10.17605/osf.io/6sjdq","URL":"https://doi.org/10.17605/osf.io/6sjdq","source":"datacite"},{"id":"doi:10.6084/m9.figshare.24438184","type":"article-journal","title":"MeSH2Wikidata: A set of tools for the interaction between MeSH keywords, OBO Foundry, and Wikidata for enriching biomedical knowledge","abstract":"The work consists of tools for the interaction between Wikidata and OBO Foundry and source codes for the use of MeSH keywords of PubMed publications for the enrichment of biomedical knowledge in Wikidata. This work is funded by the Adapting Wikidata to support clinical practice using Data Science, Semantic Web and Machine Learning Project within the framework of the Wikimedia Foundation Research Fund. To cite the work : Turki, H., Chebil, K., Dossou, B. F. P., Emezue, C. C., Owodunni, A. T., Hadj Taieb, M. A., &amp; Ben Aouicha, M. (2024). A framework for integrating biomedical knowledge in Wikidata with open biomedical ontologies and MeSH keywords. Heliyon , 10 (19), e38488. doi:10.1016/j.heliyon.2024.e38448.Wikidata-OBO tool1.py: A tool for the verification of the semantic alignment between Wikidata and OBO ontologies. frame.py: The layout of Tool 1. tool2.py: A tool for extracting Wikidata relations between OBO ontology items. frame2.py: The layout of Tool 2. tool3.py: A tool for extracting multilingual language data for OBO ontology items from Wikidata. frame4.py: The layout of Tool 3.Wikidata-MeSH correct_mesh2matrix_dataset.py: A source code for turning MeSH2Matrix into a smaller dataset for the biomedical relation classification based on the MeSH keywords of PubMed publications, named MiniMeSH2Matrix. build_numpy_dataset.py: A source code for building the numpy files for MiniMeSH2Matrix (Relation type-based classification). label_encoded.csv: A table for the conversion of Wikidata Property IDs into MeSH2Matrix Class IDs. new_encoding.csv: A table for the conversion of Wikidata Property IDs into MiniMeSH2Matrix Class IDs. super_classes_new_dataset_labels.npy: The NumPy File of the labels for the superclass-based classification. new_dataset_labels.npy: The NumPy File of the labels for the relation type-based classification. new_dataset_matrices.npy: The Numpy File of the MiniMeSH2Matrix matrices for biomedical relation classification. first_level_new_data.json: The JSON File for the conversion of relation types to superclasses. build_super_classes.py: A source code for building the numpy files for MiniMeSH2Matrix (Superclass-based classification). FC_MeSH_Model_57_New_Data.ipynb: A Jupyter Notebook for training a Dense Model to perform the relation type-based classification. FC_MeSH_Model_57_New_Data_SuperClasses.ipynb: A Jupyter Notebook for training a Dense Model to perform the superclass-based classification. new_data_best_model_1: A stored edition of the best model for the relation type-based classification. new_data_super_classes_best_model_1: A stored edition of the best model for the superclass-based classification. MiniMeSH2Matrix_SuperClasses_Confusion_Matrix.ipynb: A Jupyter Notebook for generating the confusion matrix for the superclass-based supervised classification. MiniMeSH2Matrix_Supervised_Classification_Agreement.ipynb: A Jupyter Notebook for generating the matrix of agreement between the accurate predictions for superclass-based classification and the ones for relation type-based classification. Adding_References_to_Wikidata.ipynb: A Jupyter Notebook to identify the PubMed ID of relevant references to unsupported Wikidata statements between MeSH terms. MeSH_Statistics.xlsx: Statistical data about MeSH-based items and relations in Wikidata. ref_for_unsupported_statements.csv: Retrieved Relevant PubMed References for 1k unsupported Wikidata statements. evaluate_pubmed_ref_assignment.ipynb: A Jupyter Notebook that generates statistics about reference assignment for a sample of 1k unsupported statements. MeSH_Verification.xlsx: A list of inaccurate or duplicated MeSH IDs in Wikidata, as of August 8th, 2023. WikiRelationsPMI.csv: A list of PMI values for the semantic relations between MeSH terms, as available in Wikidata. WikiRelationsPMIDistribution.xlsx: Distribution of PMI values for all Wikidata relations and for specific Wikidata relation types. WikiRelationsToVerify.xlsx: Wikidata relations needing att","author":[{"family":"Turki","given":"Houcemeddine"},{"family":"Chebil","given":"Khalil"},{"family":"Dossou","given":"Bonaventure"},{"family":"Emezue","given":"Chris"},{"family":"Owodunni","given":"Abraham"},{"family":"Taieb","given":"Mohamed"},{"family":"Ben Aouicha","given":"Mohamed"}],"issued":{"date-parts":[[2023]]},"DOI":"10.6084/m9.figshare.24438184","URL":"https://doi.org/10.6084/m9.figshare.24438184","source":"datacite"},{"id":"doi:10.48550/arxiv.2406.12142","type":"manuscript","title":"Slicing Through Bias: Explaining Performance Gaps in Medical Image Analysis using Slice Discovery Methods","abstract":"Machine learning models have achieved high overall accuracy in medical image analysis. However, performance disparities on specific patient groups pose challenges to their clinical utility, safety, and fairness. This can affect known patient groups - such as those based on sex, age, or disease subtype - as well as previously unknown and unlabeled groups. Furthermore, the root cause of such observed performance disparities is often challenging to uncover, hindering mitigation efforts. In this paper, to address these issues, we leverage Slice Discovery Methods (SDMs) to identify interpretable underperforming subsets of data and formulate hypotheses regarding the cause of observed performance disparities. We introduce a novel SDM and apply it in a case study on the classification of pneumothorax and atelectasis from chest x-rays. Our study demonstrates the effectiveness of SDMs in hypothesis formulation and yields an explanation of previously observed but unexplained performance disparities between male and female patients in widely used chest X-ray datasets and models. Our findings indicate shortcut learning in both classification tasks, through the presence of chest drains and ECG wires, respectively. Sex-based differences in the prevalence of these shortcut features appear to cause the observed classification performance gap, representing a previously underappreciated interaction between shortcut learning and model fairness analyses.","author":[{"family":"Olesen","given":"Vincent"},{"family":"Weng","given":"Nina"},{"family":"Feragen","given":"Aasa"},{"family":"Petersen","given":"Eike"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2406.12142","URL":"https://doi.org/10.48550/arxiv.2406.12142","source":"datacite"},{"id":"doi:10.48550/arxiv.2409.16231","type":"manuscript","title":"Predicting Deterioration in Mild Cognitive Impairment with Survival Transformers, Extreme Gradient Boosting and Cox Proportional Hazard Modelling","abstract":"The paper proposes a novel approach of survival transformers and extreme gradient boosting models in predicting cognitive deterioration in individuals with mild cognitive impairment (MCI) using metabolomics data in the ADNI cohort. By leveraging advanced machine learning and transformer-based techniques applied in survival analysis, the proposed approach highlights the potential of these techniques for more accurate early detection and intervention in Alzheimer's dementia disease. This research also underscores the importance of non-invasive biomarkers and innovative modelling tools in enhancing the accuracy of dementia risk assessments, offering new avenues for clinical practice and patient care. A comprehensive Monte Carlo simulation procedure consisting of 100 repetitions of a nested cross-validation in which models were trained and evaluated, indicates that the survival machine learning models based on Transformer and XGBoost achieved the highest mean C-index performances, namely 0.85 and 0.8, respectively, and that they are superior to the conventional survival analysis Cox Proportional Hazards model which achieved a mean C-Index of 0.77. Moreover, based on the standard deviations of the C-Index performances obtained in the Monte Carlo simulation, we established that both survival machine learning models above are more stable than the conventional statistical model.","author":[{"family":"Musto","given":"Henry"},{"family":"Stamate","given":"Daniel"},{"family":"Logofatu","given":"Doina"},{"family":"Stahl","given":"Daniel"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2409.16231","URL":"https://doi.org/10.48550/arxiv.2409.16231","source":"datacite"},{"id":"doi:10.48550/arxiv.2409.09161","type":"manuscript","title":"Train-On-Request: An On-Device Continual Learning Workflow for Adaptive Real-World Brain Machine Interfaces","abstract":"Brain-machine interfaces (BMIs) are expanding beyond clinical settings thanks to advances in hardware and algorithms. However, they still face challenges in user-friendliness and signal variability. Classification models need periodic adaptation for real-life use, making an optimal re-training strategy essential to maximize user acceptance and maintain high performance. We propose TOR, a train-on-request workflow that enables user-specific model adaptation to novel conditions, addressing signal variability over time. Using continual learning, TOR preserves knowledge across sessions and mitigates inter-session variability. With TOR, users can refine, on demand, the model through on-device learning (ODL) to enhance accuracy adapting to changing conditions. We evaluate the proposed methodology on a motor-movement dataset recorded with a non-stigmatizing wearable BMI headband, achieving up to 92% accuracy and a re-calibration time as low as 1.6 minutes, a 46% reduction compared to a naive transfer learning workflow. We additionally demonstrate that TOR is suitable for ODL in extreme edge settings by deploying the training procedure on a RISC-V ultra-low-power SoC (GAP9), resulting in 21.6 ms of latency and 1 mJ of energy consumption per training step. To the best of our knowledge, this work is the first demonstration of an online, energy-efficient, dynamic adaptation of a BMI model to the intrinsic variability of EEG signals in real-time settings.","author":[{"family":"Mei","given":"Lan"},{"family":"Cioflan","given":"Cristian"},{"family":"Ingolfsson","given":"Thorir"},{"family":"Kartsch","given":"Victor"},{"family":"Cossettini","given":"Andrea"},{"family":"Wang","given":"Xiaying"},{"family":"Benini","given":"Luca"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2409.09161","URL":"https://doi.org/10.48550/arxiv.2409.09161","source":"datacite"},{"id":"doi:10.48550/arxiv.2408.11854","type":"manuscript","title":"When Raw Data Prevails: Are Large Language Model Embeddings Effective in Numerical Data Representation for Medical Machine Learning Applications?","abstract":"The introduction of Large Language Models (LLMs) has advanced data representation and analysis, bringing significant progress in their use for medical questions and answering. Despite these advancements, integrating tabular data, especially numerical data pivotal in clinical contexts, into LLM paradigms has not been thoroughly explored. In this study, we examine the effectiveness of vector representations from last hidden states of LLMs for medical diagnostics and prognostics using electronic health record (EHR) data. We compare the performance of these embeddings with that of raw numerical EHR data when used as feature inputs to traditional machine learning (ML) algorithms that excel at tabular data learning, such as eXtreme Gradient Boosting. We focus on instruction-tuned LLMs in a zero-shot setting to represent abnormal physiological data and evaluating their utilities as feature extractors to enhance ML classifiers for predicting diagnoses, length of stay, and mortality. Furthermore, we examine prompt engineering techniques on zero-shot and few-shot LLM embeddings to measure their impact comprehensively. Although findings suggest the raw data features still prevails in medical ML tasks, zero-shot LLM embeddings demonstrate competitive results, suggesting a promising avenue for future research in medical applications.","author":[{"family":"Gao","given":"Yanjun"},{"family":"Myers","given":"Skatje"},{"family":"Chen","given":"Shan"},{"family":"Dligach","given":"Dmitriy"},{"family":"Miller","given":"Timothy"},{"family":"Bitterman","given":"Danielle"},{"family":"Churpek","given":"Matthew"},{"family":"Afshar","given":"Majid"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2408.11854","URL":"https://doi.org/10.48550/arxiv.2408.11854","source":"datacite"},{"id":"doi:10.48550/arxiv.2409.10704","type":"manuscript","title":"Self-supervised Speech Models for Word-Level Stuttered Speech Detection","abstract":"Clinical diagnosis of stuttering requires an assessment by a licensed speech-language pathologist. However, this process is time-consuming and requires clinicians with training and experience in stuttering and fluency disorders. Unfortunately, only a small percentage of speech-language pathologists report being comfortable working with individuals who stutter, which is inadequate to accommodate for the 80 million individuals who stutter worldwide. Developing machine learning models for detecting stuttered speech would enable universal and automated screening for stuttering, enabling speech pathologists to identify and follow up with patients who are most likely to be diagnosed with a stuttering speech disorder. Previous research in this area has predominantly focused on utterance-level detection, which is not sufficient for clinical settings where word-level annotation of stuttering is the norm. In this study, we curated a stuttered speech dataset with word-level annotations and introduced a word-level stuttering speech detection model leveraging self-supervised speech models. Our evaluation demonstrates that our model surpasses previous approaches in word-level stuttering speech detection. Additionally, we conducted an extensive ablation analysis of our method, providing insight into the most important aspects of adapting self-supervised speech models for stuttered speech detection.","author":[{"family":"Shih","given":"Yi"},{"family":"Gkalitsiou","given":"Zoi"},{"family":"Dimakis","given":"Alexandros"},{"family":"Harwath","given":"David"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2409.10704","URL":"https://doi.org/10.48550/arxiv.2409.10704","source":"datacite"},{"id":"doi:10.48550/arxiv.2409.05871","type":"manuscript","title":"Multi-feature Compensatory Motion Analysis for Reaching Motions Over a Discretely Sampled Workspace","abstract":"The absence of functional arm joints, such as the wrist, in upper extremity prostheses leads to compensatory motions in the users' daily activities. Compensatory motions have been previously studied for varying task protocols and evaluation metrics. However, the movement targets' spatial locations in previous protocols were not standardised and incomparable between studies, and the evaluation metrics were rudimentary. This work analysed compensatory motions in the final pose of subjects reaching across a discretely sampled 7*7 2D grid of targets under unbraced (normative) and braced (compensatory) conditions. For the braced condition, a bracing system was applied to simulate a transradial prosthetic limb by restricting participants' wrist joints. A total of 1372 reaching poses were analysed, and a Compensation Index was proposed to indicate the severity level of compensation. This index combined joint spatial location analysis, joint angle analysis, separability analysis, and machine learning (clustering) analysis. The individual analysis results and the final Compensation Index were presented in heatmap format to correspond to the spatial layout of the workspace, revealing the spatial dependency of compensatory motions. The results indicate that compensatory motions occur mainly in a right trapezoid region in the upper left area and a vertical trapezoid region in the middle left area for right-handed subjects reaching horizontally and vertically. Such results might guide motion selection in clinical rehabilitation, occupational therapy, and prosthetic evaluation to help avoid residual limb pain and overuse syndromes.","author":[{"family":"Yang","given":"Qihan"},{"family":"Gloumakov","given":"Yuri"},{"family":"Spiers","given":"Adam"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2409.05871","URL":"https://doi.org/10.48550/arxiv.2409.05871","source":"datacite"},{"id":"doi:10.48550/arxiv.2409.02530","type":"manuscript","title":"Understanding eGFR Trajectories and Kidney Function Decline via Large Multimodal Models","abstract":"The estimated Glomerular Filtration Rate (eGFR) is an essential indicator of kidney function in clinical practice. Although traditional equations and Machine Learning (ML) models using clinical and laboratory data can estimate eGFR, accurately predicting future eGFR levels remains a significant challenge for nephrologists and ML researchers. Recent advances demonstrate that Large Language Models (LLMs) and Large Multimodal Models (LMMs) can serve as robust foundation models for diverse applications. This study investigates the potential of LMMs to predict future eGFR levels with a dataset consisting of laboratory and clinical values from 50 patients. By integrating various prompting techniques and ensembles of LMMs, our findings suggest that these models, when combined with precise prompts and visual representations of eGFR trajectories, offer predictive performance comparable to existing ML models. This research extends the application of foundation models and suggests avenues for future studies to harness these models in addressing complex medical forecasting challenges.","author":[{"family":"Li","given":"Chih"},{"family":"Wu","given":"Jun"},{"family":"Hsu","given":"Chan"},{"family":"Lin","given":"Ming"},{"family":"Kang","given":"Yihuang"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2409.02530","URL":"https://doi.org/10.48550/arxiv.2409.02530","source":"datacite"},{"id":"doi:10.48550/arxiv.2409.02303","type":"manuscript","title":"A Lesion-aware Edge-based Graph Neural Network for Predicting Language Ability in Patients with Post-stroke Aphasia","abstract":"We propose a lesion-aware graph neural network (LEGNet) to predict language ability from resting-state fMRI (rs-fMRI) connectivity in patients with post-stroke aphasia. Our model integrates three components: an edge-based learning module that encodes functional connectivity between brain regions, a lesion encoding module, and a subgraph learning module that leverages functional similarities for prediction. We use synthetic data derived from the Human Connectome Project (HCP) for hyperparameter tuning and model pretraining. We then evaluate the performance using repeated 10-fold cross-validation on an in-house neuroimaging dataset of post-stroke aphasia. Our results demonstrate that LEGNet outperforms baseline deep learning methods in predicting language ability. LEGNet also exhibits superior generalization ability when tested on a second in-house dataset that was acquired under a slightly different neuroimaging protocol. Taken together, the results of this study highlight the potential of LEGNet in effectively learning the relationships between rs-fMRI connectivity and language ability in a patient cohort with brain lesions for improved post-stroke aphasia evaluation.","author":[{"family":"Chen","given":"Zijian"},{"family":"Varkanitsa","given":"Maria"},{"family":"Ishwar","given":"Prakash"},{"family":"Konrad","given":"Janusz"},{"family":"Betke","given":"Margrit"},{"family":"Kiran","given":"Swathi"},{"family":"Venkataraman","given":"Archana"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2409.02303","URL":"https://doi.org/10.48550/arxiv.2409.02303","source":"datacite"},{"id":"doi:10.48550/arxiv.2404.03833","type":"manuscript","title":"An ExplainableFair Framework for Prediction of Substance Use Disorder Treatment Completion","abstract":"Fairness of machine learning models in healthcare has drawn increasing attention from clinicians, researchers, and even at the highest level of government. On the other hand, the importance of developing and deploying interpretable or explainable models has been demonstrated, and is essential to increasing the trustworthiness and likelihood of adoption of these models. The objective of this study was to develop and implement a framework for addressing both these issues - fairness and explainability. We propose an explainable fairness framework, first developing a model with optimized performance, and then using an in-processing approach to mitigate model biases relative to the sensitive attributes of race and sex. We then explore and visualize explanations of the model changes that lead to the fairness enhancement process through exploring the changes in importance of features. Our resulting-fairness enhanced models retain high sensitivity with improved fairness and explanations of the fairness-enhancement that may provide helpful insights for healthcare providers to guide clinical decision-making and resource allocation.","author":[{"family":"Lucas","given":"Mary"},{"family":"Wang","given":"Xiaoyang"},{"family":"Chang","given":"Chia"},{"family":"Yang","given":"Christopher"},{"family":"Braughton","given":"Jacqueline"},{"family":"Ngo","given":"Quyen"}],"issued":{"date-parts":[[2024]]},"DOI":"10.48550/arxiv.2404.03833","URL":"https://doi.org/10.48550/arxiv.2404.03833","source":"datacite"},{"id":"oa:W4408304546","type":"article-journal","title":"Enhancing parental skills through artificial intelligence‐based conversational agents: The PAT Initiative","abstract":"Abstract Objective We aim to describe the development of a conversational agent (CA) for parenting, termed PAT (Parenting Assistant platform), to demonstrate how artificial intelligence (AI) can enhance parenting skills. Background Behavioral problems are the most common issues in childhood mental health. Developing and disseminating scalable interventions to address early‐stage behavioral problems are of high priority. Artificial intelligence (AI)‐based CAs can offer innovative methods to deliver parenting interventions to reduce behavioral problems. CAs have the capability to interact through text or voice conversations and can undergo training using evidence‐based parenting programs. However, research on CAs for parenting and behavioral problems is limited. Experience The development of PAT consisted of three phases: Phase 1 was purely rule‐based, Phase 2 was hybrid (rule‐based format plus large language models), and Phase 3 featured an agentic architecture. The latest version of PAT includes prompt engineering, guardrails, retrieval‐augmented generation, few‐shots learning, context, and memory management through agentic architecture. Although comprehensive empirical results are pending, the iterative development and enhancement of PAT indicate the potential for effective digital intervention. The agentic architecture of the latest version of PAT aims to provide robust, context‐aware interactions to support parenting challenges. Implications CAs have the potential to reach a broader population of parents and deliver personalized interventions tailored to their specific needs. Moreover, CAs are structured to provide timely support, which can enhance family dynamics and contribute to improved long‐term outcomes for both parents and children. Conclusion AI‐based CAs can be used as alternatives to waitlists; as digital cotherapists; and implemented in health care, mental health, and school settings. The potential benefits and risks of the different types of CA and features are discussed.","author":[{"family":"Escoredo","given":"Milagros"},{"family":"Mostovoy","given":"Karin"},{"family":"Schickler","given":"Ross"},{"family":"Bechtel","given":"Alexis"},{"family":"Shagan","given":"Jennah"},{"family":"Bunge","given":"Eduardo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/fare.13158","URL":"https://doi.org/10.1111/fare.13158","source":"openalex"},{"id":"oa:W4408551062","type":"article-journal","title":"Harnessing Artificial Intelligence for Innovation in Interventional Cardiovascular Care","abstract":"Artificial intelligence (AI) serves as a powerful tool that can revolutionize how personalized, patient-focused care is provided within interventional cardiology. Specifically, AI can augment clinical care across the spectrum for acute coronary syndrome, coronary artery disease, and valvular heart disease, with applications in coronary and structural heart interventions. This has been enabled by the potential of AI to harness various types of health data. We review how AI-driven technologies can advance diagnosis, preprocedural planning, intraprocedural guidance, and prognostication in interventional cardiology. AI automates clinical tasks, increases efficiency, improves reliability and accuracy, and individualizes clinical care, establishing its potential to transform care. Furthermore, AI-enabled, community-based screening programs are yet to be implemented to leverage the full potential of AI to improve patient outcomes. However, to transform clinical practice, AI tools require robust and transparent development processes, consistent performance across various settings and populations, positive impact on clinical and care quality outcomes, and seamless integration into clinical workflows. Once these are established, AI can reshape interventional cardiology, improving precision, efficiency, and patient outcomes.","author":[{"family":"Aminorroaya","given":"Arya"},{"family":"Biswas","given":"Dhruva"},{"family":"Pedroso","given":"Aline"},{"family":"Khera","given":"Rohan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jscai.2025.102562","URL":"https://doi.org/10.1016/j.jscai.2025.102562","source":"openalex"},{"id":"oa:W4411792008","type":"article-journal","title":"Exploring the role of DeepSeek-R1, ChatGPT-4, and Google Gemini in medical education: How valid and reliable are they?","abstract":"Objective: In recent years, Artificial Intelligence (AI) has led to rapid advancements in science, technology, industries, healthcare settings, and medical education. A Chinese-built large language model, DeepSeek-R1, inspires the scientific community as an affordable and open alternative to earlier established US-based AI models, ChatGPT-4 and Google Gemini 1.5 Pro. This study aimed to explore the role of “DeepSeek-R1, ChatGPT-4 and Google Gemini 1.5 Pro” and to assess the validity and reliability of these AI tools in medical education. Methods: The current cross-sectional study was performed in the Department of Physiology, College of Medicine, King Saud University, Riyadh, Saudi Arabia during the period January 25, 2025, to February 28, 2025. The Multiple-Choice Questions (MCQs) bank was created with a pool of basic medical sciences (60 MCQs) and clinical medical sciences (40 MCQs). The one hundred MCQs were prepared from various medical textbooks, journals, and examination pools. The MCQs were individually entered into the given area of the “DeepSeek-R1, ChatGPT-4 and Google Gemini 1.5 Pro” to assess the level of knowledge in various disciplines of medical sciences. Results: The marks obtained in basic medical sciences by DeepSeek R1 47/60 (78.33%), ChatGPT-4 47/60 (78.33%), and Google Gemini 1.5 Pro 49/60 (81.7%). However, in clinical medical sciences, the marks obtained by DeepSeek R1 were 35/40 (87.5%), ChatGPT-4 36/40 (90.0%), and Google Gemini 1.5 Pro 33/40 (82.5%). The total marks obtained by DeepSeekR1 were 82/100 (82.0%), Chat GPT-4 84/100 (84.0%), and Google Gemini-1.5 Pro 82/100 (82.0%). Conclusions: The Chinese-based DeepSeek-R1, the US-based ChatGPT-4, and Google Gemini-1.5 Pro achieved similar scores, exceeding 80% marks, in various medical sciences subjects. The study findings demonstrate that the knowledge, validity, and reliability levels of DeepSeek R1, ChatGPT-4, and Google Gemini 1.5 Pro are similar for their potential future use in medical education. doi: https://doi.org/10.12669/pjms.41.7.12183 How to cite this: Meo SA, Abukhalaf FA, ElToukhy RA, Sattar K. Exploring the role of DeepSeek-R1, ChatGPT-4, and Google Gemini in medical education: How valid and reliable are they? Pak J Med Sci. 2025;41(7):1887-1892. doi: https://doi.org/10.12669/pjms.41.7.12183 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.","author":[{"family":"Meo","given":"Sultan"},{"family":"Abukhalaf","given":"Farah"},{"family":"Eltoukhy","given":"Riham"},{"family":"Sattar","given":"Kamran"}],"issued":{"date-parts":[[2025]]},"DOI":"10.12669/pjms.41.7.12183","URL":"https://doi.org/10.12669/pjms.41.7.12183","source":"openalex"},{"id":"oa:W4413783468","type":"article-journal","title":"A Novel Playbook for Pragmatic Trial Operations to Monitor and Evaluate Ambient Artificial Intelligence in Clinical Practice","abstract":"BACKGROUND: Ambient artificial intelligence (AI) offers the potential to reduce documentation burden and improve efficiency through clinical note generation. Widespread adoption, however, remains limited due to challenges in electronic health record (EHR) integration, coding compliance, and real-world evaluation. This study introduces a framework and protocols to design, monitor, and deploy ambient AI within routine care. METHODS: , Tenth Revision (ICD-10) compliance were performed using an internally developed large language model (LLM), the validity of which was assessed through correlation with certified professional coders. RESULTS: Ambient AI utilization, measured as the proportion of eligible clinical notes completed using the system, had a weighted median of 65.4% (interquartile range, 50.6 to 84.0%). Iterative improvement cycles targeted task-specific adoption. A brief workflow issue related to a note template change initially reduced ICD-10 documentation accuracy from 79% (95% confidence interval [CI], 72 to 86%) to 35% (95% CI, 28 to 42%); accuracy returned to baseline after note template redesign and user training. The internally developed LLM coder achieved a strong correlation with professional coders (Pearson's r=0.97). The trial enrolled 66 providers across eight specialties, powered at 90% for the primary outcome of provider well-being. CONCLUSIONS: We provide a publicly available framework and protocols to help safely implement ambient AI in health care. Innovations include an embedded pragmatic trial design, human factors engineering, compliance-driven feedback loops, and real-time monitoring to support deployment, ensuring fidelity before initiation of the clinical trial. (Funded by the University of Wisconsin Hospital and Clinics and the National Institutes of Health Clinical and Translational Science Award; NIH/ NCATS UL1TR002737; ClinicalTrials.gov number, NCT06517082.).","author":[{"family":"Afshar","given":"Majid"},{"family":"Resnik","given":"Felice"},{"family":"Ryan","given":"Mary"},{"family":"Hintzke","given":"Josie"},{"family":"Lemmon","given":"Kayla"},{"family":"Sullivan","given":"Anne"},{"family":"Shah","given":"Tina"},{"family":"Stordalen","given":"Anthony"},{"family":"Oberst","given":"Michael"},{"family":"Dambach","given":"Jason"},{"family":"Mrotek","given":"Leigh"},{"family":"Quinn","given":"Mariah"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1056/aidbp2401267","URL":"https://doi.org/10.1056/aidbp2401267","source":"openalex"},{"id":"oa:W4413196066","type":"article-journal","title":"Application of artificial intelligence in electrochemical diagnostics for human health","abstract":"Electrochemical sensors, detecting biochemical changes through electrical signals, play a pivotal role in point-of-care diagnostics, especially for detecting specific biomarkers for diseases including cancer, diabetes, obesity, cardiovascular conditions, etc. Biosensor based technology faces numerous challenges including signal complexity, noise, data interpretation, etc. These challenges influence the sensitivity and selectivity of the techniques and limit their wider applications. The modern-day miracle, Artificial Intelligence (AI) offers transformative solutions to these challenges. The applications of machine learning (ML) algorithms and AI in electrochemical data analysis have significantly enhanced the sensitivity and specificity of diagnostic methods. The AI-powered systems can easily identify the specific patterns within the electrochemical signals that otherwise remain undetectable by traditional methods. This leads to early detection, personalized treatment plans, and real-time monitoring of diseases. AI also assists in optimizing sensor design, manages large datasets, and improves the performance and reliability of electrochemical diagnostics (ED) devices. Thus, the integration of AI into ED is transforming the healthcare sector by providing faster, more precise, and cost-effective diagnostic solutions.","author":[{"family":"Ranjan","given":"Koushlesh"},{"family":"Barar","given":"Basanti"},{"family":"Prasad","given":"Minakshi"},{"family":"Prasad","given":"Gaya"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s44373-025-00042-w","URL":"https://doi.org/10.1007/s44373-025-00042-w","source":"openalex"},{"id":"oa:W4412103547","type":"article-journal","title":"Artificial intelligence and digital twins for the personalised prediction of hypertension risk","abstract":"Hypertension is a significant global health challenge, contributing substantially to morbidity and mortality through its association with various cardiovascular diseases. Traditional approaches to hypertension risk prediction, which rely on broad epidemiological data and common risk factors, often fail to account for individual variability, highlighting the need for advanced data-driven methodologies. This review examines the role of Artificial Intelligence (AI) and Machine Learning (ML) in enhancing the prediction of hypertension risk by incorporating a range of data sources, including clinical, lifestyle, and genetic factors. Despite promising developments, challenges such as data standardisation, the need for high-quality datasets, model explainability, and class imbalance in medical data persist. The integration of wearable technologies, alongside the potential of emerging technologies in healthcare such as digital twins, presents significant opportunities in personalising care through the dynamic modelling of individual health profiles. This review synthesises current methodologies, identifies existing gaps, and highlights the transformative potential of AI-driven, personalised hypertension prevention and management, emphasising the importance of addressing issues of reproducibility and transparency to facilitate clinical adoption.","author":[{"family":"Naik","given":"Akhil"},{"family":"Nalepa","given":"Jakub"},{"family":"Wijata","given":"Agata"},{"family":"Mahon","given":"JR"},{"family":"Mistry","given":"Dharmesh"},{"family":"Knowles","given":"Adam"},{"family":"Dawson","given":"Ellen"},{"family":"Lip","given":"Gregory"},{"family":"Olier","given":"Iván"},{"family":"Ortegamartorell","given":"Sandra"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.compbiomed.2025.110718","URL":"https://doi.org/10.1016/j.compbiomed.2025.110718","source":"openalex"},{"id":"oa:W4409686230","type":"article-journal","title":"Applications of Artificial Intelligence in Neurological Voice Disorders","abstract":"Neurological voice disorders, such as Parkinson's disease, laryngeal dystonia, and stroke-induced dysarthria, significantly impact speech production and communication. Traditional diagnostic methods rely on subjective assessment, whereas artificial intelligence (AI) offers objective, noninvasive, and scalable solutions for voice analysis. This review examines the applications, advancements, challenges, and future prospects of AI-driven methods in diagnosing, monitoring, and treating neurological voice disorders. We analyze recent advances in AI-based voice analysis, including machine learning, deep learning and signal processing techniques, and evaluate their effectiveness based on existing literature. AI models have demonstrated high accuracy in detecting subtle voice impairments, enabling early diagnosis of voice disorders, and predicting treatment response. Deep learning methods, particularly convolutional and transformer-based networks, have been effective in extracting meaningful biomarkers from acoustic or other modality data. Despite these promising advances, challenges remain, including limited high-quality data sets on some rare neurological voice disorders, ethical concerns regarding patient privacy, and the need for broad clinical validation. Further research should focus on developing standardized data sets, improving the ability of the AI model to learn representations, and enhancing its generalizability. With further development, AI-driven data analysis has the potential to transform the early detection and management of neurological voice disorders.","author":[{"family":"Yao","given":"Dongren"},{"family":"Koivu","given":"Aki"},{"family":"Simonyan","given":"Kristina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/wjo2.70017","URL":"https://doi.org/10.1002/wjo2.70017","source":"openalex"},{"id":"doi:10.7759/cureus.76835","type":"article-journal","title":"Exploring Medical Artificial Intelligence Readiness Among Future Physicians: Insights From a Medical College in Central India.","abstract":"INTRODUCTION: Medical students, as future healthcare professionals, are pivotal in the adoption and application of artificial intelligence (AI) in clinical settings. Their ability to effectively engage with AI technologies is shaped by their understanding, attitudes, and perceived significance of AI in medicine. Given the growing prominence of AI in the medical field, it is crucial to evaluate how well-prepared medical students are to integrate and use these technologies proficiently. MATERIALS AND METHODS: The cross-sectional study was conducted among 482 undergraduate medical students at a medical college in Central India with the objective to evaluate their readiness for the integration of medical AI into their future clinical practice, utilizing the Medical Artificial Intelligence Readiness Scale for Medical Students (MAIRS-MS) questionnaire. RESULTS: The mean age of respondents was 21.39 ± 1.770 years with 282 (58.5%) male participants. The respondents were almost equally distributed among all Bachelor of Medicine and Bachelor of Surgery (MBBS) batch students. The average MAIRS-MS score came out to be 74.61 ± 10.137 out of a maximum of 110, whereas the mean values of various subscales of MAIRS-MS were as follows: Cognition Factor, 26.23 ± 4.417; Ability Factor, 27.62 ± 4.372; Vision Factor, 10.37 ± 1.803; and Ethics Factor, 10.39 ± 1.789. CONCLUSION: Although there is overall readiness for AI among the respondents, significant variation exists among individuals, especially in the areas of Cognition and Ability. The data highlights the necessity for focused educational programs to improve AI knowledge, skills, and ethical understanding, ensuring that every respondent is well-equipped to handle the advancing field of AI in medicine.","author":[{"family":"Dhurandhar","given":"Diwakar"},{"family":"Dhamande","given":"Mithilesh"},{"family":"Shivaleela","given":"C"},{"family":"Bhadoria","given":"Pooja"},{"family":"Chandrakar","given":"Tripti"},{"family":"Agrawal","given":"Jagriti"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.76835","URL":"https://doi.org/10.7759/cureus.76835","source":"europepmc"},{"id":"doi:10.48550/arxiv.2608.14778","type":"manuscript","title":"AMPLIFAI: A Multiphase CT Dataset for Benchmarking Clinical Reasoning in LI-RADS Assessment of Liver Lesions","abstract":"Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related mortality worldwide, with early detection improving survival from &lt;20% to &gt;70%. The standardized Liver Imaging Reporting and Data System (LI-RADS) criteria provide an imaging-based diagnostic framework to evaluate liver lesions for HCC, serving as a foundation for automating HCC detection with artificial intelligence (AI). However, the lack of large, publicly available datasets with high-quality annotations has limited the development and evaluation of AI models for automated LI-RADS assessment. We introduce AMPLIFAI dataset, the first public dataset of 590 multiphase abdominal CT studies annotated with LI-RADS categories, lesion size, and voxel-level segmentations for three major LI-RADS features: arterial phase hyperenhancement, washout, and enhancing capsule. The dataset was curated and harmonized from four public datasets and augmented with expert annotations from five board-certified radiologists and one resident. Following the Datasheets for Datasets format, this paper details the dataset's composition, curation and harmonization process, and annotation workflow to support transparent, reproducible research in medical imaging AI.","author":[{"family":"Kulkarni","given":"Pranav"},{"family":"Shah","given":"Nikhil"},{"family":"Suryavanshi","given":"Amritansh"},{"family":"Delfino","given":"Jana"},{"family":"Tonascia","given":"James"},{"family":"Wong-You-Cheong","given":"Jade"},{"family":"Lane","given":"Barton"},{"family":"Chirico","given":"Joseph"},{"family":"Hirsch","given":"Jeffrey"},{"family":"Li","given":"Ang"},{"family":"Huang","given":"Heng"},{"family":"Doo","given":"Florence"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2608.14778","URL":"https://doi.org/10.48550/arxiv.2608.14778","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8661514.v1","type":"article-journal","title":"Stroke detection in medical emergency calls: a retrospective analysis and exploratory evaluation of an AI decision support model","abstract":"Abstract Background Timely identification of acute stroke during medical emergency calls is critical for optimizing patient outcomes. We aimed to (i) characterize the prehospital trajectories of patients with stroke and (ii) explore the ability of an artificial intelligence (AI) model developed within the Artificial Intelligence Support in Medical Emergency Calls project to support decision-making at Emergency Medical Communication Centres (EMCCs). Methods We conducted a retrospective analysis using data from 1,164 patients diagnosed with stroke from an EMCC (2018–2019), focusing on those primarily assessed by EMCC operators. We categorized patients into optimal and nonoptimal trajectory groups and applied logistic regression to explore the factors associated with an optimal trajectory. An AI model was trained using a dataset of 2,980 emergency calls (2019 and 2022), integrating transcribed audio logs and structured clinical data. The model was evaluated on the full dataset and subgroups that were incorrectly assessed by the EMCC. Results Aphasia/dysarthria was the only factor associated with optimal trajectories. The AI model achieved a sensitivity of 81.0%, specificity of 79.6%, and F1-score of 64.7%. The subgroup analysis included 14 false-negative and 41 false-positive cases. The model correctly predicted stroke in 9 of 14 false-negative cases and ruled out stroke in 9 of 41 false-positive cases. Conclusions This explorative evaluation indicates that a multimodal AI pipeline combining audio and structured clinical data may show potential to support decision making in medical emergency calls. Internal results are promising, but studies on larger and external datasets are needed to validate these findings. Graphical Abstract","author":[{"family":"Iversen","given":"Emil"},{"family":"Ihle-Hansen","given":"Hege"},{"family":"Halle","given":"Kari"},{"family":"Lundervold","given":"Alexander"},{"family":"Myrmel","given":"Lars"},{"family":"Vestbø","given":"Anders"},{"family":"Fromm","given":"Annette"},{"family":"Persia","given":"Cosimo"},{"family":"Autenried","given":"Christian"},{"family":"Brattebø","given":"Guttorm"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8661514.v1","URL":"https://doi.org/10.6084/m9.figshare.c.8661514.v1","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8661514","type":"article-journal","title":"Stroke detection in medical emergency calls: a retrospective analysis and exploratory evaluation of an AI decision support model","abstract":"Abstract Background Timely identification of acute stroke during medical emergency calls is critical for optimizing patient outcomes. We aimed to (i) characterize the prehospital trajectories of patients with stroke and (ii) explore the ability of an artificial intelligence (AI) model developed within the Artificial Intelligence Support in Medical Emergency Calls project to support decision-making at Emergency Medical Communication Centres (EMCCs). Methods We conducted a retrospective analysis using data from 1,164 patients diagnosed with stroke from an EMCC (2018–2019), focusing on those primarily assessed by EMCC operators. We categorized patients into optimal and nonoptimal trajectory groups and applied logistic regression to explore the factors associated with an optimal trajectory. An AI model was trained using a dataset of 2,980 emergency calls (2019 and 2022), integrating transcribed audio logs and structured clinical data. The model was evaluated on the full dataset and subgroups that were incorrectly assessed by the EMCC. Results Aphasia/dysarthria was the only factor associated with optimal trajectories. The AI model achieved a sensitivity of 81.0%, specificity of 79.6%, and F1-score of 64.7%. The subgroup analysis included 14 false-negative and 41 false-positive cases. The model correctly predicted stroke in 9 of 14 false-negative cases and ruled out stroke in 9 of 41 false-positive cases. Conclusions This explorative evaluation indicates that a multimodal AI pipeline combining audio and structured clinical data may show potential to support decision making in medical emergency calls. Internal results are promising, but studies on larger and external datasets are needed to validate these findings. Graphical Abstract","author":[{"family":"Iversen","given":"Emil"},{"family":"Ihle-Hansen","given":"Hege"},{"family":"Halle","given":"Kari"},{"family":"Lundervold","given":"Alexander"},{"family":"Myrmel","given":"Lars"},{"family":"Vestbø","given":"Anders"},{"family":"Fromm","given":"Annette"},{"family":"Persia","given":"Cosimo"},{"family":"Autenried","given":"Christian"},{"family":"Brattebø","given":"Guttorm"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8661514","URL":"https://doi.org/10.6084/m9.figshare.c.8661514","source":"datacite"},{"id":"oa:W4410995939","type":"article-journal","title":"Multilingual performance of a multimodal artificial intelligence system on multisubject physics concept inventories","abstract":"We investigate the multilingual and multimodal performance of a large language model-based artificial intelligence (AI) system, GPT-4o, using a diverse set of physics concept inventories spanning multiple languages and subject categories. The inventories, sourced from the PhysPort website, cover classical physics topics such as mechanics, electromagnetism, optics, and thermodynamics, as well as relativity, quantum mechanics, astronomy, mathematics, and laboratory skills. Unlike previous text-only studies, we uploaded the inventories as images to reflect what a student would see on paper, thereby assessing the system’s multimodal functionality. Our results indicate variation in performance across subjects, with laboratory skills standing out as the weakest. We also observe differences across languages, with English and European languages showing the strongest performance. Notably, the relative difficulty of an inventory item is largely independent of the language of the test. When comparing AI results to existing literature on student performance, we find that the AI system outperforms average postinstruction undergraduate students in all subject categories except laboratory skills. Furthermore, the AI performs worse on items requiring visual interpretation of images than on those that are purely text-based. While our exploratory findings show GPT-4o’s potential usefulness in physics education, they highlight the critical need for instructors to foster students’ ability to critically evaluate AI outputs, adapt curricula thoughtfully in response to AI advancements, and address equity concerns associated with AI integration.","author":[{"family":"Kortemeyer","given":"Gerd"},{"family":"Babayeva","given":"Maryna"},{"family":"Polverini","given":"Giulia"},{"family":"Widenhorn","given":"Ralf"},{"family":"Gregorcic","given":"Bor"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1103/98hg-rkrf","URL":"https://doi.org/10.1103/98hg-rkrf","source":"openalex"},{"id":"oa:W4406322425","type":"article-journal","title":"Artificial Intelligence in Science and Society: The Vision of USERN","abstract":"The recent rise in relevance and diffusion of Artificial Intelligence (AI)-based systems and the increasing number and power of applications of AI methods invites a profound reflection on the impact of these innovative systems on scientific research and society at large. The Universal Scientific Education and Research Network (USERN), an organization that promotes initiatives to support interdisciplinary science and education across borders and actively works to improve science policy, collects here the vision of its Advisory Board members, together with a selection of AI experts, to summarize how we see developments in this exciting technology impacting science and society in the foreseeable future. In this review, we first attempt to establish clear definitions of intelligence and consciousness, then provide an overview of AI’s state of the art and its applications. A discussion of the implications, opportunities, and liabilities of the diffusion of AI for research in a few representative fields of science follows this. Finally, we address the potential risks of AI to modern society, suggest strategies for mitigating those risks, and present our conclusions and recommendations.","author":[{"family":"Dorigo","given":"T"},{"family":"Brown","given":"Gary"},{"family":"Casonato","given":"Carlo"},{"family":"Cerdà","given":"Artemi"},{"family":"Ciarrochi","given":"Joseph"},{"family":"Lio","given":"Mauro"},{"family":"Dsouza","given":"Nicole"},{"family":"Gauger","given":"Nicolas"},{"family":"Hayes","given":"Steven"},{"family":"Hofmann","given":"Stefan"},{"family":"Johansson","given":"Robert"},{"family":"Liwicki","given":"Marcus"},{"family":"Lotte","given":"Fabien"},{"family":"Nieto","given":"Juan"},{"family":"Olivato","given":"Giulia"},{"family":"Parnes","given":"Peter"},{"family":"Perry","given":"George"},{"family":"Plebe","given":"Alice"},{"family":"Rao","given":"Idupulapati"},{"family":"Rezaei","given":"Nima"},{"family":"Sandin","given":"Fredrik"},{"family":"Ustyuzhanin","given":"A"},{"family":"Vallortígara","given":"Giorgio"},{"family":"Vischia","given":"P"},{"family":"Yazdanpanah","given":"Niloufar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1109/access.2025.3529357","URL":"https://doi.org/10.1109/access.2025.3529357","source":"openalex"},{"id":"oa:W4411189778","type":"article-journal","title":"An Overview of Generative Artificial Intelligence in Medical Education","abstract":"The application of generative artificial intelligence (GAI) in medical education and practice has garnered increasing attention, particularly its significant potential to enhance personalised learning and clinical training. This viewpoint explores the integration of GAI into medical education, analysing its advantages in disseminating medical knowledge, simulating case scenarios, and supporting clinical decision-making. Although GAI introduces innovative opportunities to medical education, its practical application also presents various challenges, such as model accuracy and ethical concerns. The viewpoint further discusses the potential impact of these challenges on the future of medical education and offers corresponding strategies and recommendations, providing valuable insights for educators and policymakers. By understanding the practical applications and limitations of GAI in the medical field, this viewpoint aims to lay a foundation for more effective use of GAI in medical education in the future. Key Words: Generative artificial intelligence, Large language model, Medical education.","author":[{"family":"Wang","given":"Shilu"},{"family":"Geng","given":"Rongqing"},{"family":"Xu","given":"Ruoning"}],"issued":{"date-parts":[[2025]]},"DOI":"10.29271/jcpsp.2025.06.793","URL":"https://doi.org/10.29271/jcpsp.2025.06.793","source":"openalex"},{"id":"oa:W4408047551","type":"article-journal","title":"A Literature Review on Applications of Explainable Artificial Intelligence (XAI)","abstract":"As AI technologies, particularly deep learning models, have advanced, their inherent “black box” nature has raised significant concerns regarding accountability, fairness, and trust, especially in critical domains such as healthcare, finance, and criminal justice. We present a detailed exploration of XAI, emphasizing its essential role in improving the interpretability and transparency of complex AI systems in various application domains. Health-related applications were notably using XAI, emphasizing diagnostics, and medical imaging. Other notable domains of use of XAI is encompassed environmental and agricultural management, industrial optimization, cybersecurity, finance, transportation, and social media. Furthermore, nascent applications in law, education, and social care underscore the growing influence of XAI. The analysis indicates a prevalent application of local explanation techniques, especially SHAP and LIME, with a preference for SHAP due to its stability and mathematical assurances. Each technique is analysed for its strengths and limitations in providing clear, actionable insights into model decision-making processes, thereby aiding stakeholders in understanding AI behaviour. Ultimately, this document underscores the critical challenges for XAI in fostering user trust, enhancing decision-making processes, and ensuring that AI technologies are utilized responsibly and ethically across various applications, paving the way for a more transparent and accountable AI landscape. We believe that by serving as a guide for future studies in the area, our systematic review contributes to the body of literature on XAI.","author":[{"family":"Kalasampath","given":"Khushi"},{"family":"Kn","given":"Spoorthi"},{"family":"Sajeev","given":"Sreeparvathy"},{"family":"Kuppa","given":"Sahil"},{"family":"Ajay","given":"K"},{"family":"Maruthamuthu","given":"Angulakshmi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1109/access.2025.3546681","URL":"https://doi.org/10.1109/access.2025.3546681","source":"openalex"},{"id":"doi:10.17632/233w5xjzwv.2","type":"article-journal","title":"Data for:GenAI-Assisted Learning Behaviors and Systems Thinking of Medical Students in Ill-Structured Problem Solving","abstract":"This dataset contains the research data used in the study entitled “Learning Behaviors Related to Systems Thinking and Cognitive Perceptions of Undergraduates in GenAI-Assisted Ill-Structured Problem Solving.” The dataset was collected from undergraduate students who participated in a GenAI-assisted ill-structured problem-solving activity. It includes three main components: (1) behavioral coding data representing students’ learning behaviors during the problem-solving process, (2) systems thinking assessment data based on the Q1–Q8 systems thinking scale and total scores, and (3) behavioral clustering results used to identify different patterns of learning behaviors. The behavioral coding dataset records students’ observable learning behaviors during interactions with GenAI, including information-seeking, analysis, reflection, evaluation, and other cognitive and metacognitive activities. The systems thinking scores represent students’ perceptions and abilities related to systems thinking dimensions. Cluster analysis results are included to support the identification and comparison of behavioral profiles. All data have been anonymized, and no personally identifiable information is included. The dataset is provided to support transparency, reproducibility, and further research on undergraduate learning behaviors, systems thinking development, and GenAI-supported ill-structured problem solving.","author":[{"family":"Gao","given":"Zitong"},{"family":"Zhang","given":"Pingmei"},{"family":"Tan","given":"Jiaxi"},{"family":"Shi","given":"Wen"},{"family":"Zheng","given":"Jian"},{"family":"Honghe","given":"Li"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/233w5xjzwv.2","URL":"https://doi.org/10.17632/233w5xjzwv.2","source":"datacite"},{"id":"doi:10.17632/233w5xjzwv","type":"article-journal","title":"Data for:GenAI-Assisted Learning Behaviors and Systems Thinking of Medical Students in Ill-Structured Problem Solving","abstract":"This dataset contains the research data used in the study entitled “Learning Behaviors Related to Systems Thinking and Cognitive Perceptions of Undergraduates in GenAI-Assisted Ill-Structured Problem Solving.” The dataset was collected from undergraduate students who participated in a GenAI-assisted ill-structured problem-solving activity. It includes three main components: (1) behavioral coding data representing students’ learning behaviors during the problem-solving process, (2) systems thinking assessment data based on the Q1–Q8 systems thinking scale and total scores, and (3) behavioral clustering results used to identify different patterns of learning behaviors. The behavioral coding dataset records students’ observable learning behaviors during interactions with GenAI, including information-seeking, analysis, reflection, evaluation, and other cognitive and metacognitive activities. The systems thinking scores represent students’ perceptions and abilities related to systems thinking dimensions. Cluster analysis results are included to support the identification and comparison of behavioral profiles. All data have been anonymized, and no personally identifiable information is included. The dataset is provided to support transparency, reproducibility, and further research on undergraduate learning behaviors, systems thinking development, and GenAI-supported ill-structured problem solving.","author":[{"family":"Gao","given":"Zitong"},{"family":"Zhang","given":"Pingmei"},{"family":"Tan","given":"Jiaxi"},{"family":"Shi","given":"Wen"},{"family":"Zheng","given":"Jian"},{"family":"Honghe","given":"Li"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/233w5xjzwv","URL":"https://doi.org/10.17632/233w5xjzwv","source":"datacite"},{"id":"doi:10.5281/zenodo.19614903","type":"article-journal","title":"Ethical Frameworks for General AI","abstract":"The prospect of artificial general intelligence (AGI) -- AI systems that match or exceed human cognitive capabilitiesacross a broad range of tasks -- raises ethical questions that existing AI ethics frameworks, designed for narrow AIsystems, are inadequate to address. Questions of moral status, rights, responsibility attribution, value alignment atcivilisational scale, and the governance of systems that may surpass human ability to audit or constrain them require newethical frameworks that engage with philosophy of mind, political theory, and international law as well as AI safetyresearch. This study systematically evaluates six ethical frameworks applied to AGI scenarios: utilitarianconsequentialism, deontological AI ethics, virtue ethics for AI, contractualist AI governance, capability approaches, and aproposed Relational AI Ethics Framework (RAEF). Evaluation applies each framework to twelve ethically complex AGIscenarios drawn from near-term forecasting, safety research, and speculative but plausible long-term AGI deployment.Scenarios include: AGI as medical decision authority, AGI-managed resource allocation during climate emergency, AGImoral status and legal personhood, AGI oversight of democratic processes, and AGI self-modification rights. Assessmentinstruments include structured expert analysis by 42 philosophers, legal scholars, and AI researchers, and a stakeholderimpact analysis tool. No single framework adequately addresses all AGI ethical challenges. RAEF achieves the highestexpert consensus score (3.84/5) by grounding AGI ethics in ongoing human-AI relationships rather than abstractprinciples. Key finding: current AI ethics frameworks systematically underweigh the interests of future generations andnon-human stakeholders in AGI governance decisions. A multi-framework governance architecture is proposed","author":[{"family":"Lindberg","given":"Marco"},{"family":"Jensen","given":"Amelia"},{"family":"Petrov","given":"Anna"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.19614903","URL":"https://doi.org/10.5281/zenodo.19614903","source":"datacite"},{"id":"doi:10.5281/zenodo.19614904","type":"article-journal","title":"Ethical Frameworks for General AI","abstract":"The prospect of artificial general intelligence (AGI) -- AI systems that match or exceed human cognitive capabilitiesacross a broad range of tasks -- raises ethical questions that existing AI ethics frameworks, designed for narrow AIsystems, are inadequate to address. Questions of moral status, rights, responsibility attribution, value alignment atcivilisational scale, and the governance of systems that may surpass human ability to audit or constrain them require newethical frameworks that engage with philosophy of mind, political theory, and international law as well as AI safetyresearch. This study systematically evaluates six ethical frameworks applied to AGI scenarios: utilitarianconsequentialism, deontological AI ethics, virtue ethics for AI, contractualist AI governance, capability approaches, and aproposed Relational AI Ethics Framework (RAEF). Evaluation applies each framework to twelve ethically complex AGIscenarios drawn from near-term forecasting, safety research, and speculative but plausible long-term AGI deployment.Scenarios include: AGI as medical decision authority, AGI-managed resource allocation during climate emergency, AGImoral status and legal personhood, AGI oversight of democratic processes, and AGI self-modification rights. Assessmentinstruments include structured expert analysis by 42 philosophers, legal scholars, and AI researchers, and a stakeholderimpact analysis tool. No single framework adequately addresses all AGI ethical challenges. RAEF achieves the highestexpert consensus score (3.84/5) by grounding AGI ethics in ongoing human-AI relationships rather than abstractprinciples. Key finding: current AI ethics frameworks systematically underweigh the interests of future generations andnon-human stakeholders in AGI governance decisions. A multi-framework governance architecture is proposed","author":[{"family":"Lindberg","given":"Marco"},{"family":"Jensen","given":"Amelia"},{"family":"Petrov","given":"Anna"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.19614904","URL":"https://doi.org/10.5281/zenodo.19614904","source":"datacite"},{"id":"doi:10.5281/zenodo.21600153","type":"article-journal","title":"Deep Learning for OSCC Diagnosis: A Multimodal Survey of Techniques, Challenges, and Future Directions","abstract":"Abstract: Oral cancer, particularly Oral Squamous Cell Carcinoma (OSCC), remains a significant global health concern due to high mortality rates and frequent late-stage diagnosis. Often identified at an advanced stage because of publicignoranceand restrictions in traditional diagnostic techniques, Oral Squamous Cell Carcinoma(OSCC) is among the most common and lethal types of oral cancer. By providing robust tools for automatic and reliable illness identification, artificial intelligence (AI), especially deep learning, has transformed medical picture analysis in recent years. This survey study offers a thorough assessment of state-of-the-art deep learning techniques used to Oral cancer detection across several imaging modalities including histopathology, fluorescence, hyperspectral, and white light pictures. We methodically investigate and contrast hybrid systems, transformer architectures, transfer learning models, and convolutional neural networks (CNNs) with respect to classification accuracy, resilience, and clinical relevance. The research also addresses real-time deployment issues, model interpretability, and multimodal data integration's importance. Moreover, this study points out present research voids—such as restricted generalizability and absence of stage-wise lesion classification—and offers future research paths to close these obstacles. This effort intends to lead academics, doctors, and developers toward the building of efficient, scalable, and accessible AI-driven diagnostic tools for early OSCC diagnosis and intervention by synthesizing ideas from previous advancements.","author":[{"family":"Kudatarkar","given":"Vinaya"},{"family":"Patil","given":"Annapurna"},{"family":"Shetty","given":"Savita"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21600153","URL":"https://doi.org/10.5281/zenodo.21600153","source":"datacite"},{"id":"doi:10.5281/zenodo.21600154","type":"article-journal","title":"Deep Learning for OSCC Diagnosis: A Multimodal Survey of Techniques, Challenges, and Future Directions","abstract":"Abstract: Oral cancer, particularly Oral Squamous Cell Carcinoma (OSCC), remains a significant global health concern due to high mortality rates and frequent late-stage diagnosis. Often identified at an advanced stage because of publicignoranceand restrictions in traditional diagnostic techniques, Oral Squamous Cell Carcinoma(OSCC) is among the most common and lethal types of oral cancer. By providing robust tools for automatic and reliable illness identification, artificial intelligence (AI), especially deep learning, has transformed medical picture analysis in recent years. This survey study offers a thorough assessment of state-of-the-art deep learning techniques used to Oral cancer detection across several imaging modalities including histopathology, fluorescence, hyperspectral, and white light pictures. We methodically investigate and contrast hybrid systems, transformer architectures, transfer learning models, and convolutional neural networks (CNNs) with respect to classification accuracy, resilience, and clinical relevance. The research also addresses real-time deployment issues, model interpretability, and multimodal data integration's importance. Moreover, this study points out present research voids—such as restricted generalizability and absence of stage-wise lesion classification—and offers future research paths to close these obstacles. This effort intends to lead academics, doctors, and developers toward the building of efficient, scalable, and accessible AI-driven diagnostic tools for early OSCC diagnosis and intervention by synthesizing ideas from previous advancements.","author":[{"family":"Kudatarkar","given":"Vinaya"},{"family":"Patil","given":"Annapurna"},{"family":"Shetty","given":"Savita"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21600154","URL":"https://doi.org/10.5281/zenodo.21600154","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8659425","type":"article-journal","title":"Exploring the use of AI-generated counterfactual chest X-rays to enhance diagnostic learning in medical education","abstract":"Abstract Accurate interpretation of chest X-rays is a critical clinical skill, yet radiology training in medical education remains limited and often fails to provide broad exposure to a wide variety of conditions, including those that are rare, subtle, or easily confused with one another. Although artificial intelligence (AI) has shown impressive performance in medical image classification, its potential to actively improve clinical education has not been fully realised. In this study, we introduce AI-generated counterfactual chest X-rays—real patient images digitally altered to show a different but still realistic condition, while keeping the same patient’s anatomy. These counterfactuals create ‘what if’ scenarios within the same patient and offer a novel tool to enhance diagnostic learning, support decision-making, and improve confidence calibration in medical trainees and professionals. Forty-two participants, including medical students and doctors, completed a four-part online study involving diagnostic classification, comparison of counterfactual image pairs, image authenticity judgements, and clinical treatment planning. All image comparisons focused on three clinical conditions: healthy lungs, pneumonia, and pleural effusion. Statistical analyses included t-tests, ANOVAs, Pearson correlations, and linear regression. Results showed that counterfactual images meaningfully supported diagnostic learning. Participants demonstrated improved accuracy and confidence over time, particularly when distinguishing pleural effusion from healthy lungs. Confidence became more closely aligned with accuracy, and participants were better able to recognise condition-specific differences. These comparisons also revealed areas of diagnostic weakness that would be difficult to detect through conventional instruction alone. While real images led to higher raw diagnostic accuracy, counterfactuals proved highly effective in promoting learning and reflective reasoning. Image authenticity influenced clinical decision-making, and initial confidence was a stronger predictor of final treatment confidence than diagnostic correctness. These findings highlight the strong potential of counterfactual imaging to enhance radiology education. By enabling learners to engage with rare, complex, or demographically underrepresented cases in a controlled and consistent way, AI-generated counterfactuals offer a powerful and scalable tool to improve diagnostic preparedness and clinical confidence in future healthcare professionals.","author":[{"family":"Mohr","given":"Greta"},{"family":"Zhu","given":"Yifei"},{"family":"Ye","given":"Xujiong"},{"family":"Lennon","given":"Marilyn"},{"family":"Maclellan","given":"Calum"},{"family":"Maclay","given":"John"},{"family":"Lowe","given":"David"},{"family":"Sainsbury","given":"Christopher"},{"family":"Dong","given":"Feng"},{"family":"Lagnado","given":"David"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8659425","URL":"https://doi.org/10.6084/m9.figshare.c.8659425","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8659425.v1","type":"article-journal","title":"Exploring the use of AI-generated counterfactual chest X-rays to enhance diagnostic learning in medical education","abstract":"Abstract Accurate interpretation of chest X-rays is a critical clinical skill, yet radiology training in medical education remains limited and often fails to provide broad exposure to a wide variety of conditions, including those that are rare, subtle, or easily confused with one another. Although artificial intelligence (AI) has shown impressive performance in medical image classification, its potential to actively improve clinical education has not been fully realised. In this study, we introduce AI-generated counterfactual chest X-rays—real patient images digitally altered to show a different but still realistic condition, while keeping the same patient’s anatomy. These counterfactuals create ‘what if’ scenarios within the same patient and offer a novel tool to enhance diagnostic learning, support decision-making, and improve confidence calibration in medical trainees and professionals. Forty-two participants, including medical students and doctors, completed a four-part online study involving diagnostic classification, comparison of counterfactual image pairs, image authenticity judgements, and clinical treatment planning. All image comparisons focused on three clinical conditions: healthy lungs, pneumonia, and pleural effusion. Statistical analyses included t-tests, ANOVAs, Pearson correlations, and linear regression. Results showed that counterfactual images meaningfully supported diagnostic learning. Participants demonstrated improved accuracy and confidence over time, particularly when distinguishing pleural effusion from healthy lungs. Confidence became more closely aligned with accuracy, and participants were better able to recognise condition-specific differences. These comparisons also revealed areas of diagnostic weakness that would be difficult to detect through conventional instruction alone. While real images led to higher raw diagnostic accuracy, counterfactuals proved highly effective in promoting learning and reflective reasoning. Image authenticity influenced clinical decision-making, and initial confidence was a stronger predictor of final treatment confidence than diagnostic correctness. These findings highlight the strong potential of counterfactual imaging to enhance radiology education. By enabling learners to engage with rare, complex, or demographically underrepresented cases in a controlled and consistent way, AI-generated counterfactuals offer a powerful and scalable tool to improve diagnostic preparedness and clinical confidence in future healthcare professionals.","author":[{"family":"Mohr","given":"Greta"},{"family":"Zhu","given":"Yifei"},{"family":"Ye","given":"Xujiong"},{"family":"Lennon","given":"Marilyn"},{"family":"Maclellan","given":"Calum"},{"family":"Maclay","given":"John"},{"family":"Lowe","given":"David"},{"family":"Sainsbury","given":"Christopher"},{"family":"Dong","given":"Feng"},{"family":"Lagnado","given":"David"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8659425.v1","URL":"https://doi.org/10.6084/m9.figshare.c.8659425.v1","source":"datacite"},{"id":"doi:10.5281/zenodo.21706663","type":"article-journal","title":"Emerging Global Challenges in Medical Laboratory Science: A Narrative Review","abstract":"Abstract Medical Laboratory Science (MLS) is still an integral part of modern health care that plays an important role in illness diagnosis, surveillance, prevention and therapeutic management. Laboratory technologies have evolved greatly, but the profession is confronted with several worldwide concerns that jeopardise patient outcomes and the delivery of healthcare. These include workforce shortages, antimicrobial resistance, emerging infectious diseases, poor laboratory infrastructure in low resource settings, quality assurance concerns, automation and artificial intelligence integration, ethical issues, biosafety risks, inadequate funding, and disparities in access to diagnostic services. The COVID-19 pandemic has revealed the inadequacies of laboratory systems around the world and the vital role of medical laboratory personnel. This narrative overview describes the current significant difficulties affecting Medical Laboratory Science worldwide, their impact on health care systems and potential solutions to enhance such systems. Meeting these problems will need coordinated action by governments, health care institutions, professional organisations and international agencies to strengthen laboratory systems and improve global health outcomes.","author":[{"family":"Kingsley","given":"Dunga"},{"family":"Johnkennedy","given":"Nnodim"},{"family":"Promise","given":"Nnodim"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21706663","URL":"https://doi.org/10.5281/zenodo.21706663","source":"datacite"},{"id":"doi:10.5281/zenodo.21706664","type":"article-journal","title":"Emerging Global Challenges in Medical Laboratory Science: A Narrative Review","abstract":"Abstract Medical Laboratory Science (MLS) is still an integral part of modern health care that plays an important role in illness diagnosis, surveillance, prevention and therapeutic management. Laboratory technologies have evolved greatly, but the profession is confronted with several worldwide concerns that jeopardise patient outcomes and the delivery of healthcare. These include workforce shortages, antimicrobial resistance, emerging infectious diseases, poor laboratory infrastructure in low resource settings, quality assurance concerns, automation and artificial intelligence integration, ethical issues, biosafety risks, inadequate funding, and disparities in access to diagnostic services. The COVID-19 pandemic has revealed the inadequacies of laboratory systems around the world and the vital role of medical laboratory personnel. This narrative overview describes the current significant difficulties affecting Medical Laboratory Science worldwide, their impact on health care systems and potential solutions to enhance such systems. Meeting these problems will need coordinated action by governments, health care institutions, professional organisations and international agencies to strengthen laboratory systems and improve global health outcomes.","author":[{"family":"Kingsley","given":"Dunga"},{"family":"Johnkennedy","given":"Nnodim"},{"family":"Promise","given":"Nnodim"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21706664","URL":"https://doi.org/10.5281/zenodo.21706664","source":"datacite"},{"id":"oa:W4409564798","type":"article-journal","title":"Charting the Landscape of Artificial Intelligence Ethics: A Bibliometric Analysis","abstract":"Abstract Using bibliometric methods, this study systematically analyzes 6,084 AI ethics-related articles from the Web of Science Core Collection (2015–2025), capturing both recent advances and near-future directions in the field. It begins by examining publication trends, disciplinary categories, leading journals, and major contributing institutions/countries. Subsequently, co-citation (journals, authors, references) and keyword clustering methods reveal the foundational knowledge structure and highlight emerging research hotspots. The findings indicate increasing interdisciplinary convergence and international collaboration in AI ethics, with core themes focusing on algorithmic fairness, privacy and data security, ethical governance in autonomous vehicles, medical AI applications, educational technology, and challenges posed by generative AI (e.g., large language models). Burst keyword detection further shows an evolutionary shift from theoretical debates toward practical implementation strategies and regulatory framework development. Although numerous global initiatives have been introduced to guide AI ethics, broad consensus remains elusive, underscoring the need for enhanced cross-disciplinary and international cooperation. This research provides valuable insights for scholars, policymakers, and industry practitioners, laying a foundation for sustainable and responsible AI development.","author":[{"family":"Qiu","given":"Jiaxuan"},{"family":"Cheng","given":"Le"},{"family":"Huang","given":"Jin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1515/ijdlg-2025-0007","URL":"https://doi.org/10.1515/ijdlg-2025-0007","source":"openalex"},{"id":"oa:W4413420174","type":"article-journal","title":"Artificial Intelligence Applications in Emergency Toxicology: Advancements and Challenges","abstract":"Unlabelled: Emergency toxicology is a complex field requiring rapid and precise decision-making to manage acute poisonings effectively. Toxic exposures are often unpredictable, and the constraints of time and resources often challenge conventional diagnostic and treatment approaches. Artificial intelligence (AI) has emerged as a valuable tool in emergency medicine, offering the potential to enhance diagnostic accuracy, predict clinical outcomes and improve clinical decision support systems. Despite the increasing focus of AI in medicine, its applications in emergency toxicology are still underexplored. This viewpoint aims to provide perspectives on AI applications in emergency toxicology by highlighting key advancements, challenges, and future directions. While AI has demonstrated significant potential in improving toxicological predictions through various applications, challenges such as data quality, regulatory concerns, and implementation barriers are still hurdles to its use. Further research, regulatory frameworks, and integration strategies are needed to ensure effective and ethical implementation in clinical practice.","author":[{"family":"Yong","given":"Lorraine"},{"family":"Tung","given":"Joshua"},{"family":"Cheung","given":"Nicole"},{"family":"Lee","given":"Zi"},{"family":"Ng","given":"Ee"},{"family":"Ng","given":"Alexander"},{"family":"Lim","given":"Clement"},{"family":"Boon","given":"Yuru"},{"family":"Lim","given":"Daniel"},{"family":"Sng","given":"Gerald"},{"family":"Tang","given":"Jonathan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/73121","URL":"https://doi.org/10.2196/73121","source":"openalex"},{"id":"doi:10.5281/zenodo.20271447","type":"article-journal","title":"Automated Brain Tumor Detection and Classification System","abstract":"The accurate rapid and non-invasive detection of central nervous system malignancies remains one of the most critical challenges in contemporary neurology. Magnetic Resonance Imaging serves as the undisputed gold standard for visualizing these complex pathology however, the manual interpretation of multi parametric volumes is cognitively exhaustive and severely bottle necked by a global shortage of specialized anesthesiologists. While recent paradigms in deep learning have achieved unprecedented diagnostic accuracy they of ten do so at the cost of prohibitive computational complexity. Foundational work established the current state of the art on a comprehensive four class brain classification task using a fine tuned object detection architecture. However this apex performance is inextricably linked to extreme hardware requirements specifically reliance on dedicated graphics processing units with massive memory reserves. Such hardware dependency renders the deployment of this model virtually impossible in the vast majority of primary healthcare facilities particularly within low and middle income countries. To bridge this critical chasm this thesis presents a highly optimized universally central processing unit deploy able alternative. Our proposed system is engineered upon a highly efficient lightweight backbone strategically retaining advanced spatial attention mechanisms contextual pooling and bidirectional feature networks. The proposed system achieves a highly competitive ninety four percent overall accuracy while enabling under two second inference per image. Furthermore we extend the base architecture with a complete clinical workflow including a robust relational patient database a novel quantitative longitudinal tumor growth tracking engine and an automated diagnostic re- port generator all integrated into a clinical web dashboard.","author":[],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20271447","URL":"https://doi.org/10.5281/zenodo.20271447","source":"datacite"},{"id":"doi:10.5281/zenodo.20271448","type":"article-journal","title":"Automated Brain Tumor Detection and Classification System","abstract":"The accurate rapid and non-invasive detection of central nervous system malignancies remains one of the most critical challenges in contemporary neurology. Magnetic Resonance Imaging serves as the undisputed gold standard for visualizing these complex pathology however, the manual interpretation of multi parametric volumes is cognitively exhaustive and severely bottle necked by a global shortage of specialized anesthesiologists. While recent paradigms in deep learning have achieved unprecedented diagnostic accuracy they of ten do so at the cost of prohibitive computational complexity. Foundational work established the current state of the art on a comprehensive four class brain classification task using a fine tuned object detection architecture. However this apex performance is inextricably linked to extreme hardware requirements specifically reliance on dedicated graphics processing units with massive memory reserves. Such hardware dependency renders the deployment of this model virtually impossible in the vast majority of primary healthcare facilities particularly within low and middle income countries. To bridge this critical chasm this thesis presents a highly optimized universally central processing unit deploy able alternative. Our proposed system is engineered upon a highly efficient lightweight backbone strategically retaining advanced spatial attention mechanisms contextual pooling and bidirectional feature networks. The proposed system achieves a highly competitive ninety four percent overall accuracy while enabling under two second inference per image. Furthermore we extend the base architecture with a complete clinical workflow including a robust relational patient database a novel quantitative longitudinal tumor growth tracking engine and an automated diagnostic re- port generator all integrated into a clinical web dashboard.","author":[],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20271448","URL":"https://doi.org/10.5281/zenodo.20271448","source":"datacite"},{"id":"doi:10.5281/zenodo.19480871","type":"article-journal","title":"Influence of AI-Generated Health Content on Patient Trust in Medical Institutions: A Study of Perceived Accuracy, Misinformation Risk, and Digital Literacy in South-West Nigeria","abstract":"This research explored how artificial intelligence health content creation affects patient trust toward South-West Nigerian medical establishments. The fast growth of AI-based tools creates new business opportunities, but also brings operational difficulties, which affect how doctors interact with their patients and how patients trust their healthcare providers. The research used a mixed-methods approach that began with purposive qualitative interviews with 25 patients who received care at general hospitals in Lagos, Oyo, and Ekiti States to study their experiences with AI-generated medical information. The research showed that trust develops through personal human connections. Yet people see AI content as both helpful for access and dangerous because it creates false information. The quantitative phase involved a survey administered to 150 participants to evaluate the themes and hypotheses that emerged from the initial qualitative findings. A multiple linear regression analysis was conducted, finding that the model was statistically significant (p< .001) and explained 65% of the variance in patient trust. The analysis identified three key variables significantly influencing trust: perceived AI accuracy (beta = 0.32), perceived risk of misinformation (a strong negative predictor, beta = -0.51), and digital literacy (beta = 0.40). The research concludes that a significant relationship exists, confirming that while AI offers a new channel for information, trust in healthcare remains fundamentally tied to human interaction and the ability of patients to critically evaluate digital information. Based on these insights, recommendations are provided to help institutions build digital literacy, manage misinformation risk, and strategically integrate AI to reinforce patient confidence.","author":[{"family":"Phd","given":"Olumide"},{"family":"Asekhamhe","given":"Okpokpo"},{"family":"Umoru","given":"Sediku"},{"family":"Ubaka","given":"Ofunne"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19480871","URL":"https://doi.org/10.5281/zenodo.19480871","source":"datacite"},{"id":"doi:10.5281/zenodo.19480872","type":"article-journal","title":"Influence of AI-Generated Health Content on Patient Trust in Medical Institutions: A Study of Perceived Accuracy, Misinformation Risk, and Digital Literacy in South-West Nigeria","abstract":"This research explored how artificial intelligence health content creation affects patient trust toward South-West Nigerian medical establishments. The fast growth of AI-based tools creates new business opportunities, but also brings operational difficulties, which affect how doctors interact with their patients and how patients trust their healthcare providers. The research used a mixed-methods approach that began with purposive qualitative interviews with 25 patients who received care at general hospitals in Lagos, Oyo, and Ekiti States to study their experiences with AI-generated medical information. The research showed that trust develops through personal human connections. Yet people see AI content as both helpful for access and dangerous because it creates false information. The quantitative phase involved a survey administered to 150 participants to evaluate the themes and hypotheses that emerged from the initial qualitative findings. A multiple linear regression analysis was conducted, finding that the model was statistically significant (p< .001) and explained 65% of the variance in patient trust. The analysis identified three key variables significantly influencing trust: perceived AI accuracy (beta = 0.32), perceived risk of misinformation (a strong negative predictor, beta = -0.51), and digital literacy (beta = 0.40). The research concludes that a significant relationship exists, confirming that while AI offers a new channel for information, trust in healthcare remains fundamentally tied to human interaction and the ability of patients to critically evaluate digital information. Based on these insights, recommendations are provided to help institutions build digital literacy, manage misinformation risk, and strategically integrate AI to reinforce patient confidence.","author":[{"family":"Phd","given":"Olumide"},{"family":"Asekhamhe","given":"Okpokpo"},{"family":"Umoru","given":"Sediku"},{"family":"Ubaka","given":"Ofunne"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19480872","URL":"https://doi.org/10.5281/zenodo.19480872","source":"datacite"},{"id":"oa:W4406958079","type":"article-journal","title":"Revolutionising osseous biopsy: the impact of artificial intelligence in the era of personalized medicine","abstract":"In a rapidly evolving healthcare environment, artificial intelligence (AI) is transforming diagnostic techniques and personalized medicine. This is also seen in osseous biopsies. AI applications in radiomics, histopathology, predictive modelling, biopsy navigation, and interdisciplinary communication are reshaping how bone biopsies are conducted and interpreted. We provide a brief review of AI in image- guided biopsy of bone tumours (primary and secondary) and specimen handling, in the era of personalized medicine. This article explores AI's role in enhancing diagnostic accuracy, improving safety in biopsies, and enabling more precise targeting in bone lesion biopsies, ultimately contributing to better patient outcomes in personalized medicine. We dive into various AI technologies applied to osseous biopsies, such as traditional machine learning, deep learning, radiomics, simulation, and generative models. We explore their roles in tumour-board meetings, communication between clinicians, radiologists, and pathologists. Additionally, we inspect ethical considerations associated with the integration of AI in bone biopsy procedures, technical limitations, and we delve into health equity, generalizability, deployment issues, and reimbursement challenges in AI-powered healthcare. Finally, we explore potential future developments and offer a list of open-source AI tools and algorithms relevant to bone biopsies, which we include to encourage further discussion and research.","author":[{"family":"Isaac","given":"Amanda"},{"family":"Klontzas","given":"Michail"},{"family":"Dalili","given":"Danoob"},{"family":"Akdoğan","given":"Aslı"},{"family":"Fawzi","given":"Mohamed"},{"family":"Gugliemi","given":"Giuseppe"},{"family":"Filippiadis","given":"Dimitrios"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/bjr/tqaf018","URL":"https://doi.org/10.1093/bjr/tqaf018","source":"openalex"},{"id":"oa:W4411161852","type":"article-journal","title":"Clinical Impact of Artificial Intelligence-Based Triage Systems in Emergency Departments: A Systematic Review","abstract":"Emergency departments (EDs) worldwide face increasing pressure to optimize triage processes amidst rising patient volumes and resource constraints. Artificial intelligence (AI) has emerged as a potential solution to enhance triage accuracy and efficiency, yet its real-world clinical impact remains inadequately characterized. We conducted a systematic review following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines, searching PubMed/Medical Literature Analysis and Retrieval System Online (MEDLINE), Excerpta Medica Database (Embase), Web of Science, and Institute of Electrical and Electronics Engineers (IEEE) Xplore (2020-2025) for studies evaluating AI-based ED triage systems. From 119 initially identified records, six studies met inclusion criteria after duplicate removal (n=67), title/abstract screening (n=52), and full-text assessment (n=12). Eligible studies reported quantitative outcomes on AI performance compared to traditional triage methods. Risk of bias was assessed using an adapted Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. Narrative synthesis was employed due to methodological heterogeneity. The included studies (n=6) demonstrated AI's potential to reduce triage time, improve documentation accuracy, and enhance decision support. Voice-based artificial intelligence (Voice-AI) systems achieved 19% faster documentation versus manual methods, while machine learning algorithms reduced mis-triage rates by 0.3-8.9%. However, limitations included undertriage risks, variable accuracy, and predominance of single-center studies. Implementation challenges encompassed workflow integration barriers and insufficient clinician acceptance metrics. AI-based triage systems show promise for improving ED efficiency but require rigorous multi-center validation and standardized outcome reporting. Key gaps include evidence on patient-centered outcomes, equity considerations, and long-term impact studies. Future development should prioritize explainable algorithms, clinician engagement, and ethical frameworks to ensure safe implementation.","author":[{"family":"Abdalhalim","given":"Abubaker"},{"family":"Ahmed","given":"Sheimaa"},{"family":"Ezzelarab","given":"Ahmed"},{"family":"Mustafa","given":"Mohammad"},{"family":"Al-Basheer","given":"Mamoun"},{"family":"Ahmed","given":"R"},{"family":"Elsayed","given":"Mowafag"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.85667","URL":"https://doi.org/10.7759/cureus.85667","source":"openalex"},{"id":"oa:W4409335180","type":"article-journal","title":"A review on artificial intelligence thermal fluids and the integration of energy conservation with blockchain technology","abstract":"The high degree of convergence between thermal fluid sciences and artificial intelligence (AI) has changed traditional energy management methods. The technology provides energy conservation, fluid dynamics, and heat transfer optimisation solutions. In order to model prediction and increase the effectiveness of thermal fluid application proposals, this review looks at the latest developments in the use of AI-enabled machine learning techniques, such as Artificial Neural Networks (ANNs), Support Vector Machines (SVM), and Deep Learning Hierarchy. In order to support sustainable energy goals, these highlighted machine learning algorithms offer a potent environment for optimising energy flow, temperature regulation, and application stability. Furthermore, diverse reinforcement learning techniques facilitate the adoptive control of intricate thermal applications in real-time settings, while Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) are employed for applicational monitoring and real-time data processing. By combining blockchain technology with artificial intelligence, a decentralised framework environment is introduced that offers energy conservation methods that are safe, transparent, honest, and reliable. An unchangeable ledger is provided by the technology, and accountability and traceability are provided by smart contracts. It supports the vital tasks of dynamically monitoring and validating energy consumption across decentralised applications (DApps) in real-time. Additionally, this article offers a thorough examination of recent research, the integration of emerging technologies, and real-world uses of blockchain and artificial intelligence in thermal fluid applications. A cost-effective energy management environment that supports international energy conservation initiatives is created by combining the predictive power of AI with the security features of blockchain technology. In addition, it offers a platform for future study, giving it a starting point for innovation in sustainable energy management. The major contribution of this review article is discussed as follows:","author":[{"family":"Khan","given":"Abdullah"},{"family":"Laghari","given":"Asif"},{"family":"Inam","given":"Syed"},{"family":"Ullah","given":"Sajid"},{"family":"Nadeem","given":"Laila"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s43621-025-01124-w","URL":"https://doi.org/10.1007/s43621-025-01124-w","source":"openalex"},{"id":"oa:W4410774085","type":"article-journal","title":"Artificial Intelligence in Ecuadorian SMEs: Drivers and Obstacles to Adoption","abstract":"This study analyzes the current state of artificial intelligence (AI) adoption among micro-, small-, and medium-sized enterprises (MSMEs) in Ecuador, with a focus on its application across core business functions. Using a stratified random sample of 385 firms from the most representative economic sectors, a survey instrument was designed to assess three dimensions: access to AI-enabling conditions, degree of AI utilization, and organizational characteristics. The results reveal that AI adoption remains limited and highly concentrated in marketing-related functions, particularly in content generation and social media automation, with minimal implementation in finance, logistics, and human resource management. The study also identifies the main barriers hindering AI adoption. The lack of qualified professionals and the unavailability of structured databases emerged as the most critical obstacles, followed by limited financial capacity. One-way ANOVA and Kruskal–Wallis tests confirmed significant differences in AI adoption levels based on company size and sector, especially in areas such as inventory optimization and design prototyping. These findings highlight a gap between the potential of AI technologies and their real-world implementation in Ecuadorian MSMEs. They underscore the need for targeted strategies focused on workforce training, digital infrastructure development, and institutional support to promote broader and more effective AI integration.","author":[{"family":"Pérez-Campdesuñer","given":"Reyner"},{"family":"Sánchez-Rodríguez","given":"Alexander"},{"family":"García-Vidal","given":"Gelmar"},{"family":"Martínez-Vivar","given":"Rodobaldo"},{"family":"Miguel-Guzmán","given":"Margarita"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/info16060443","URL":"https://doi.org/10.3390/info16060443","source":"openalex"},{"id":"oa:W4412048610","type":"article-journal","title":"Explainable artificial intelligence driven insights into smoking prediction using machine learning and clinical parameters","abstract":"Smoking is a leading cause of various health conditions, including cancer and respiratory diseases. Smokers often face medical restrictions such as limitations in blood and organ donation, reduced effectiveness of medications, and increased surgical complications. These impacts underscore the need for early detection of smoking status to enable timely intervention. This study explores the use of Artificial Intelligence (AI) and Machine Learning (ML) techniques to predict smoking status based on health parameters, including biosignals and clinical biomarkers. A balanced subset of 2,000 instances was sampled from a publicly available Kaggle dataset comprising clinical and biometric features. Multiple ML models were implemented, including Random Forest Classifier, Logistic Regression, Decision Tree Classifier, K-Nearest Neighbors, CatBoost Classifier, and an Artificial Neural Network. The Random Forest Classifier achieved the better performance with an accuracy of 0.80, precision of 0.80, recall of 0.80, and F1-score of 0.79. To enhance model interpretability, four Explainable Artificial Intelligence (XAI) techniques were applied: Shapley Additive Explanations (SHAP), Local Interpretable Model-Agnostic Explanations (LIME), QLattice, and Anchor. SHAP identified hemoglobin as the most influential predictor, while LIME, QLattice, and Anchor highlighted the role of gamma-glutamyl transferase (t). Interactions between hemoglobin, GTP, and height were associated with more accurate predictions. The integration of ensemble modeling and multiple XAI approaches offers deeper interpretability than prior studies, providing healthcare providers and policymakers with a robust, transparent decision-support tool for targeted intervention strategies.","author":[{"family":"Aishwarya","given":"S"},{"family":"Siddalingaswamy","given":"PC"},{"family":"Chadaga","given":"Krishnaraj"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-09409-w","URL":"https://doi.org/10.1038/s41598-025-09409-w","source":"openalex"},{"id":"oa:W4411394707","type":"article-journal","title":"Medical reasoning in LLMs: an in-depth analysis of DeepSeek R1","abstract":"Introduction: The integration of large language models (LLMs) into healthcare holds immense promise, but also raises critical challenges, particularly regarding the interpretability and reliability of their reasoning processes. While models like DeepSeek R1-which incorporates explicit reasoning steps-show promise in enhancing performance and explainability, their alignment with domain-specific expert reasoning remains understudied. Methods: This paper evaluates the medical reasoning capabilities of DeepSeek R1, comparing its outputs to the reasoning patterns of medical domain experts. Results: Through qualitative and quantitative analyses of 100 diverse clinical cases from the MedQA dataset, we demonstrate that DeepSeek R1 achieves 93% diagnostic accuracy and shows patterns of medical reasoning. Analysis of the seven error cases revealed several recurring errors: anchoring bias, difficulty integrating conflicting data, limited consideration of alternative diagnoses, overthinking, incomplete knowledge, and prioritizing definitive treatment over crucial intermediate steps. Discussion: These findings highlight areas for improvement in LLM reasoning for medical applications. Notably the length of reasoning was important with longer responses having a higher probability for error. The marked disparity in reasoning length suggests that extended explanations may signal uncertainty or reflect attempts to rationalize incorrect conclusions. Shorter responses (e.g., under 5,000 characters) were strongly associated with accuracy, providing a practical threshold for assessing confidence in model-generated answers. Beyond observed reasoning errors, the LLM demonstrated sound clinical judgment by systematically evaluating patient information, forming a differential diagnosis, and selecting appropriate treatment based on established guidelines, drug efficacy, resistance patterns, and patient-specific factors. This ability to integrate complex information and apply clinical knowledge highlights the potential of LLMs for supporting medical decision-making through artificial medical reasoning.","author":[{"family":"Moëll","given":"Birger"},{"family":"Aronsson","given":"Fredrik"},{"family":"Akbar","given":"Sanian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/frai.2025.1616145","URL":"https://doi.org/10.3389/frai.2025.1616145","source":"openalex"},{"id":"oa:W4409619572","type":"article-journal","title":"Artificial intelligence and dichotomania","abstract":"Abstract Large language models (LLMs) such as ChatGPT, Gemini, and Claude are increasingly being used in aid or place of human judgment and decision making. Indeed, academic researchers are increasingly using LLMs as a research tool. In this paper, we examine whether LLMs, like academic researchers, fall prey to a particularly common human error in interpreting statistical results, namely ‘dichotomania’ that results from the dichotomization of statistical results into the categories ‘statistically significant’ and ‘statistically nonsignificant’. We find that ChatGPT, Gemini, and Claude fall prey to dichotomania at the 0.05 and 0.10 thresholds commonly used to declare ‘statistical significance’. In addition, prompt engineering with principles taken from an American Statistical Association Statement on Statistical Significance and P-values intended as a corrective to human errors does not mitigate this and arguably exacerbates it. Further, more recent and larger versions of these models do not necessarily perform better. Finally, these models sometimes provide interpretations that are not only incorrect but also highly erratic.","author":[{"family":"Mcshane","given":"Blakeley"},{"family":"Gal","given":"David"},{"family":"Duhachek","given":"Adam"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1017/jdm.2025.7","URL":"https://doi.org/10.1017/jdm.2025.7","source":"openalex"},{"id":"oa:W4413480625","type":"article-journal","title":"Better way: initial acceptability testing of using artificial intelligence tools to accelerate development of trauma clinical guidance","abstract":"Abstract Introduction Representatives of the trauma community have voiced a need for a new approach to developing clinical guidance. In this study, we test the initial acceptability of a proposed 12-step approach that aims to reduce the current clinical guidance timeline from more than 24 months to 24 weeks. Methods Investigators hypothesized that artificial intelligence (AI) tools could be leveraged to improve and make the process of clinical guidance development more efficient, facilitating AI initial output that could later be reviewed by subject matter experts (SMEs). Ensuring ethical standards and a collaborative design. Following the agile methodology, emphasizing continuous delivery and improvement, and the Practical, Robust Implementation and Sustainability Model (PRISM) framework, the investigators drafted a 12-step approach to clinical guidance development in 24 weeks. The process starts with the selection of a clinical topic and culminates in a bedside-ready clinical decision tree. Results The 2025 Design for Implementation: The Future of Trauma Research & Clinical Guidance conference participants were invited to reflect on this new 12-step approach during two breakout sessions. Participants included a broad range of trauma providers, methodologists, patient representatives, technology, and marketing experts. Their recommendations highlighted: 1) multidisciplinary involvement, 2) need for resource-stratified recommendations, and 3) user-friendly features (offline and multilingual access). On a post conference survey (n=56), 64% were confident in AI accelerating the current development process. Conclusions The current landscape of clinical guidance offers significant opportunities for improvement. Key areas for enhancement include promoting collaboration across multiple disciplines and organizations, developing recommendations that consider resource variations, and utilizing new technologies, such as AI, to expedite the development process. This is crucial because ongoing delays lead to practices lagging behind current evidence. Further research is needed to rigorously test and refine how responsible use of AI can be integrated into expediting evidence integration into clinical guidance. Key Messages What is already known on this topic Current clinical guidance typically takes 1-2 years to develop. Moreover, clinical guidance may not be published until a year or more after its completion, long after some recommendations become outdated, contributing to lagged evidence-informed practice. What this study adds This study shares and tests the initial acceptability of a novel approach that aims to reduce the current clinical guidance timeline from 24 months to 24 weeks. It leverages existing artificial intelligence tools but with the critical input of subject matter experts (SMEs), ensuring ethical standards and collaborative design. SMEs shed light on critical steps and key areas that future clinical guidance needs to consider. How this study might affect research, practice or policy The current landscape of clinical guidance offers significant opportunities for improvement. Key areas for enhancement include promoting collaboration across multiple disciplines and organizations, developing recommendations that consider resource variations, and utilizing new technologies, such as artificial intelligence, to expedite the development process.","author":[{"family":"Wong","given":"Gabriela"},{"family":"Rosenauer","given":"Shannon"},{"family":"Church","given":"Chelsea"},{"family":"Sherifali","given":"Diana"},{"family":"Racey","given":"Megan"},{"family":"Grider","given":"Katheryn"},{"family":"Moreno","given":"Ashley"},{"family":"Lagrone","given":"Lacey"},{"family":"Bixby","given":"Pamela"},{"family":"Bonne","given":"Stephanie"},{"family":"Bulger","given":"Eileen"},{"family":"Cain","given":"James"},{"family":"Chastek","given":"Jennifer"},{"family":"Coleman","given":"Julia"},{"family":"Costantini","given":"Todd"},{"family":"Cozzi","given":"Nicholas"},{"family":"Davis","given":"Kimberly"},{"family":"Dicker","given":"Rochelle"},{"family":"Dorlac","given":"Warren"},{"family":"Eaton","given":"Erik"},{"family":"Eriksson","given":"Evert"},{"family":"Evans","given":"Susan"},{"family":"Foster","given":"Shannon"},{"family":"Goodloe","given":"Jeffrey"},{"family":"Haut","given":"Elliott"},{"family":"Jarman","given":"Molly"},{"family":"Johnson","given":"Alyssa"},{"family":"Kotagal","given":"Meera"},{"family":"Krause","given":"Morgan"},{"family":"Kubasiak","given":"John"},{"family":"Lang","given":"Kelly"},{"family":"Leigh","given":"Allison"},{"family":"Mangat","given":"Halinder"},{"family":"Marvel","given":"Debra"},{"family":"Michetti","given":"Christopher"},{"family":"Moran","given":"Vicki"},{"family":"Moreno","given":"Ashley"},{"family":"Oczkowski","given":"Simon"},{"family":"Person","given":"Michael"},{"family":"Price","given":"Michelle"},{"family":"Punch","given":"Lj"},{"family":"Racey","given":"Megan"},{"family":"Ray","given":"Bradford"},{"family":"Redmond","given":"Diane"},{"family":"Reinhart","given":"Linda"},{"family":"Rhodes","given":"Heather"},{"family":"Rhodes","given":"Bryn"},{"family":"Rubiano","given":"Andres"},{"family":"Sanchez","given":"Sabrina"},{"family":"Sarani","given":"Babak"},{"family":"Shelton","given":"Erica"},{"family":"Spain","given":"David"},{"family":"Staudenmayer","given":"Kristan"},{"family":"Stein","given":"Deborah"},{"family":"Valenzuela","given":"Julie"},{"family":"Villarreal","given":"Cynthia"},{"family":"Wells","given":"Jeffrey"},{"family":"Wong","given":"Gabriela"},{"family":"Young","given":"Leanne"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1136/tsaco-2025-002060","URL":"https://doi.org/10.1136/tsaco-2025-002060","source":"openalex"},{"id":"oa:W4407673603","type":"article-journal","title":"Application of AI in engineering education: A bibliometric study","abstract":"Abstract The integration of artificial intelligence (AI) into engineering education is essential for fostering innovation, strategic thinking and interdisciplinary skills in the intelligent era. On this basis, this study aims to track and visually represent the research outputs associated with AI applications in engineering education, providing insights into the current research landscape and identifying areas for further investigation. The analysis offers theoretical and methodological direction for leveraging AI in engineering education. Utilising bibliometric methods, we conducted a comprehensive visualisation analysis of 378 core publications from the Web of Science (WoS) database, spanning from the beginning of the twenty‐first century to the present. Our findings show a consistent rise in publication volume from 2000 to 2017, with a significant surge from 2018 to 2023. The study identifies the International Journal of Engineering Education and Computer Applications in Engineering Education as pivotal journals in the field. The research clusters around two central themes: essential supportive technologies and specific educational applications. Within engineering education, expert systems, data mining, prediction and machine learning are highlighted as key research areas. The field has evolved through distinct phases, starting with an early focus on technology support systems, moving to an emphasis on pedagogical applications, and currently striving for a balance between diverse technologies and practical applications. Context and implications Rationale for the study Research on AI‐enabled engineering education is necessary because it is a vital measure for cultivating innovative, strategic and interdisciplinary talents in the era of intelligence. The article applies bibliometric techniques to visualise the developmental pulse of AI use in engineering education. Why the new findings matter The new findings firstly help researchers to grasp the development pulse and research priorities in the field and fill the research gaps, and secondly provide theoretical and methodological guidance for the application of artificial intelligence in engineering education through visualisation. Implications for researchers and practitioners The study summarises the current situation of AI in engineering education, breaks down the knowledge map of the field, provides practitioners, especially new researchers, with important references and guidance to understand the development of the field, and triggers the continuous attention of all sectors of society to the field. In addition, school administrators will be able to guide the development and practice of education and teaching based on the findings of the study, and frontline teachers will be able to practise the integration and application of AI and engineering education, which will effectively improve the learning efficiency of engineering education.","author":[{"family":"Liu","given":"Yipin"},{"family":"Jing","given":"Yuhui"},{"family":"Li","given":"Jing"},{"family":"Dai","given":"Jian"},{"family":"Hu","given":"Zhebing"},{"family":"Wang","given":"Chengliang"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/rev3.70044","URL":"https://doi.org/10.1002/rev3.70044","source":"openalex"},{"id":"doi:10.5281/zenodo.20325220","type":"article-journal","title":"Multivariate Analysis of Nutritional Variables and Inflammatory Biomarkers Using AI Models for Clinical Cancer Risk Prediction","abstract":"The study showcases the application of complex machine learning models that study inflammation and nutrition together for high accuracy early cancer risk prediction. With the help of statistical preprocessing and feature nonlinearity, together with other machine learning models, the system managed to capture the interactions between nutrition and inflammation to a level that other models would not. These models report a validation error that significantly exceeds that of models that only capture the linear components of the relationships involved. Traditional models have been demonstrated to have wide margins of error with regard to the cancer risk predicted. The developed system is smart enough to combine the inflamed glycemic load, saturated fat content, CRP, IL-6, and TNF-α levels, together with antioxidant and omega-3 intake, and model cancer risk. The study describes the building of a fusion of several machine learning models into a system that is clinically operational, and that incorporates a number of electronic health record components. The system demonstrates an excellent model of the clinical value of civilizational risk factors and artificial intelligence combined.","author":[{"family":"Gupalo","given":"Sergey"},{"family":"Rn","given":"Jegathambigai"},{"family":"Kyaw","given":"Zaw"},{"family":"Thidar","given":"Aung"},{"family":"Phone","given":"Myint"},{"family":"Wana","given":"Hla"},{"family":"Aye Aye","given":"Tun"},{"family":"Rohini","given":"Karunakaran"},{"family":"Manglesh Waran","given":"Udayah"},{"family":"Lwin Lwin","given":"Nyein"},{"family":"Nang","given":"Khin"},{"family":"Thida","given":"Khin"},{"family":"Myat Myo","given":"Naing"},{"family":"Sutha","given":"Devaraj"},{"family":"Nazmul","given":"Mhm"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20325220","URL":"https://doi.org/10.5281/zenodo.20325220","source":"datacite"},{"id":"doi:10.5281/zenodo.20325221","type":"article-journal","title":"Multivariate Analysis of Nutritional Variables and Inflammatory Biomarkers Using AI Models for Clinical Cancer Risk Prediction","abstract":"The study showcases the application of complex machine learning models that study inflammation and nutrition together for high accuracy early cancer risk prediction. With the help of statistical preprocessing and feature nonlinearity, together with other machine learning models, the system managed to capture the interactions between nutrition and inflammation to a level that other models would not. These models report a validation error that significantly exceeds that of models that only capture the linear components of the relationships involved. Traditional models have been demonstrated to have wide margins of error with regard to the cancer risk predicted. The developed system is smart enough to combine the inflamed glycemic load, saturated fat content, CRP, IL-6, and TNF-α levels, together with antioxidant and omega-3 intake, and model cancer risk. The study describes the building of a fusion of several machine learning models into a system that is clinically operational, and that incorporates a number of electronic health record components. The system demonstrates an excellent model of the clinical value of civilizational risk factors and artificial intelligence combined.","author":[{"family":"Gupalo","given":"Sergey"},{"family":"Rn","given":"Jegathambigai"},{"family":"Kyaw","given":"Zaw"},{"family":"Thidar","given":"Aung"},{"family":"Phone","given":"Myint"},{"family":"Wana","given":"Hla"},{"family":"Aye Aye","given":"Tun"},{"family":"Rohini","given":"Karunakaran"},{"family":"Manglesh Waran","given":"Udayah"},{"family":"Lwin Lwin","given":"Nyein"},{"family":"Nang","given":"Khin"},{"family":"Thida","given":"Khin"},{"family":"Myat Myo","given":"Naing"},{"family":"Sutha","given":"Devaraj"},{"family":"Nazmul","given":"Mhm"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20325221","URL":"https://doi.org/10.5281/zenodo.20325221","source":"datacite"},{"id":"doi:10.5281/zenodo.19571786","type":"article-journal","title":"AutoRespire: A Multimodal AI-Driven Healthcare System for Tuberculosis and Pneumonia Classification","abstract":"This paper presents AutoRespire, a multimodal AI-driven healthcare system designed for the detection of Tuberculosis and Pneumonia. The system integrates chest X-ray image analysis with clinical symptom data to improve diagnostic accuracy. Advanced machine learning and deep learning models, including Random Forest and Artificial Neural Networks, are utilized to analyze medical data and generate predictions. A late fusion strategy is applied to combine outputs from different models, enhancing reliability and performance. The proposed system demonstrates improved prediction accuracy and supports early diagnosis, especially in resource-limited settings. This approach aims to assist healthcare professionals in making faster and more accurate decisions for respiratory disease detection.","author":[{"family":"Priyanka","given":"Potti"},{"family":"Deyyala","given":"Sowmyadevi"},{"family":"Kolla","given":"Vivek"},{"family":"Sabbella","given":"Vigneswara"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19571786","URL":"https://doi.org/10.5281/zenodo.19571786","source":"datacite"},{"id":"doi:10.5281/zenodo.19571787","type":"article-journal","title":"AutoRespire: A Multimodal AI-Driven Healthcare System for Tuberculosis and Pneumonia Classification","abstract":"This paper presents AutoRespire, a multimodal AI-driven healthcare system designed for the detection of Tuberculosis and Pneumonia. The system integrates chest X-ray image analysis with clinical symptom data to improve diagnostic accuracy. Advanced machine learning and deep learning models, including Random Forest and Artificial Neural Networks, are utilized to analyze medical data and generate predictions. A late fusion strategy is applied to combine outputs from different models, enhancing reliability and performance. The proposed system demonstrates improved prediction accuracy and supports early diagnosis, especially in resource-limited settings. This approach aims to assist healthcare professionals in making faster and more accurate decisions for respiratory disease detection.","author":[{"family":"Priyanka","given":"Potti"},{"family":"Deyyala","given":"Sowmyadevi"},{"family":"Kolla","given":"Vivek"},{"family":"Sabbella","given":"Vigneswara"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19571787","URL":"https://doi.org/10.5281/zenodo.19571787","source":"datacite"},{"id":"doi:10.5281/zenodo.21307548","type":"article-journal","title":"Ensemble-Based Robust Framework for Disease Prediction and Treatment Recommendation","abstract":"Abstract: Early identification of diseases based on symptoms plays a vital role in enabling timely medical consultations and enhancing healthcare accessibility. With advancements in artificial intelligence, machine learning techniques effectively analyze symptom data to predict possible diseases. This project introduces an ensemble learning framework for robust disease prediction and treatment recommendation using structured tabular clinical data. The proposed system generates structured symptom features by mapping symptoms to a predefined feature list, forming a tabular dataset. Traditional machine learning classifiers—including Random Forest, Support Vector Machine (SVM), Logistic Regression, and XGBoost—are trained on this dataset. To boost prediction reliability, an ensemble framework combines predictions from these models using techniques such as voting, weighted ensemble, and stacking. Experimental results show that ensemble methods outperform individual models in accuracy. The trained models power a Flask-based web application, enabling users to input symptoms and receive predicted diseases with treatment recommendations, highlighting the system's practical utility.","author":[{"family":"Boragalli","given":"Vidyashri"},{"family":"Pawar","given":"Digvijay"},{"family":"Sagavkar","given":"Sandhya"},{"family":"Huddar","given":"Mahesh"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21307548","URL":"https://doi.org/10.5281/zenodo.21307548","source":"datacite"},{"id":"doi:10.5281/zenodo.21307549","type":"article-journal","title":"Ensemble-Based Robust Framework for Disease Prediction and Treatment Recommendation","abstract":"Abstract: Early identification of diseases based on symptoms plays a vital role in enabling timely medical consultations and enhancing healthcare accessibility. With advancements in artificial intelligence, machine learning techniques effectively analyze symptom data to predict possible diseases. This project introduces an ensemble learning framework for robust disease prediction and treatment recommendation using structured tabular clinical data. The proposed system generates structured symptom features by mapping symptoms to a predefined feature list, forming a tabular dataset. Traditional machine learning classifiers—including Random Forest, Support Vector Machine (SVM), Logistic Regression, and XGBoost—are trained on this dataset. To boost prediction reliability, an ensemble framework combines predictions from these models using techniques such as voting, weighted ensemble, and stacking. Experimental results show that ensemble methods outperform individual models in accuracy. The trained models power a Flask-based web application, enabling users to input symptoms and receive predicted diseases with treatment recommendations, highlighting the system's practical utility.","author":[{"family":"Boragalli","given":"Vidyashri"},{"family":"Pawar","given":"Digvijay"},{"family":"Sagavkar","given":"Sandhya"},{"family":"Huddar","given":"Mahesh"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21307549","URL":"https://doi.org/10.5281/zenodo.21307549","source":"datacite"},{"id":"doi:10.6084/m9.figshare.30609992","type":"article-journal","title":"<b>The Role of Artificial Intelligence in Diagnosing Malignant Tumors</b>","abstract":"This paper explores the transformative impact of artificial intelligence (AI) in early tumor diagnosis, emphasizing its role in analyzing health records, medical images, biopsies, and blood tests for improved risk stratification. While screening programs have enhanced survival, challenges remain in patient selection and diagnostic workforces. The review covers diverse AI approaches, including logistic regression, deep learning, and neural networks, applied to various data types in oncology. It discusses the clinical implications, current models in practice, and potential limitations such as ethical concerns and resource demands. We provide an overview of the main artificial intelligence approaches, encompassing historical models like logistic regression, alongside deep learning and neural networks, emphasizing their applications in early diagnosis. We describe the role of AI in tumor detection, prognosis, and treatment administration, and we introduce the application of state-of-the-art large language models in oncology clinics. Our exploration extends to AI applications for omics data types, offering perspectives on their combination for decision-support tools. Concurrently, we evaluate existing constraints and challenges in applying artificial intelligence to precision oncology. The overall aim is to showcase AI's promise in revolutionizing tumor diagnosis while acknowledging and addressing associated challenges, thereby advancing patient care.","author":[{"family":"Ahmad","given":"Dr"},{"family":"Khan","given":"Zafar"},{"family":"Moh","given":"Aijaz"},{"family":"Kamboj","given":"Anjoo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.6084/m9.figshare.30609992","URL":"https://doi.org/10.6084/m9.figshare.30609992","source":"datacite"},{"id":"doi:10.17605/osf.io/jkcyp","type":"article-journal","title":"Patient Factors in Medical Artificial Intelligence","abstract":"Artificial intelligence (AI) is increasingly embedded in clinical care, spanning applications in screening, diagnosis, treatment decision support, monitoring and so on. Despite rapid technological progress, current research remains heavily focused on model performance and how they may help clinicians, with little attention paid to patient-related factors that influence how AI tools are used, experienced, and accepted in real-world settings. Patient factors such as trust, usability perceptions, satisfaction play a crucial role in shaping AI uptake and its clinical impact. However, these factors have not been systematically mapped, and no comprehensive framework currently exists to guide patient-centered development and implementation of medical AI. This systematic review aims to synthesize all reported patient-related factors in empirical studies involving AI tools used directly by patients or applied to patient care. By mapping existing evidence, summarizing measurement approaches, and identifying conceptual and methodological gaps, the review will clarify how patient factors are currently understood and where further research is needed. The findings will support more equitable and patient-centered AI design, inform clinical implementation strategies, and contribute to the development of foundational frameworks for evaluating patient–AI interactions in healthcare.","author":[{"family":"Lai","given":"Shixin"},{"family":"Zhang","given":"Yulian"},{"family":"Guan","given":"Zhouyu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.17605/osf.io/jkcyp","URL":"https://doi.org/10.17605/osf.io/jkcyp","source":"datacite"},{"id":"doi:10.17605/osf.io/j8gwy","type":"article-journal","title":"Promoting Utilization of Artificial Intelligence and Machine Learning Tools for Coronary CT Angiography","abstract":"The gold standard for CCS diagnosis is cardiac catheterization, which is an invasive procedure. Although beneficial, this procedure exhibits individual limitations which emphasizes the need for more advanced CCS detection tools. In recent years, artificial intelligence (AI) has played a significant role in medical imaging. Despite promising advances, the practical integration of AI into CCTA remains inconsistent and underexplored. This paper aims to evaluate the role of CCTA and AI in improving CCS diagnostics, (2) identify existing barriers to adoption, and (3) outline global recommendations for future implementation. This scoping review was guided by the Arksey &amp; O’Malley Framework. Additionally, the authors adopted the Joanna Briggs Institute (JBI) framework and the PRISMA-SCR as a reference checklist to ensure transparency and comprehensive reporting of the review’s methodology and findings. A final number of 177 studies were retained for analysis. Major findings across the analyzed studies emphasized the beneficial role of incorporating AI and machine learning when it comes to the detection, management, and treatment of CCS. The major barrier theme categories that were highlighted across the studies included lack of generalizability, selection bias, lack of standardization protocols or technology variability, lack of accessibility, significant amounts of time and manual labor, and inter-reader variability. Current evidence shows that AI-enhanced CCTA represents a transformative but still evolving approach to the detection and management of chronic coronary syndromes. Moving forward, with rigorous evaluation and clinical application, AI has the potential to improve outcomes, efficiency, and limited challenges in cardiovascular imaging.","author":[{"family":"Sacca","given":"Lea"},{"family":"Dasilva","given":"Gabriella"},{"family":"Campson","given":"Alexandra"},{"family":"Starr","given":"Alana"},{"family":"Kamm","given":"Christine"},{"family":"Ernst","given":"Kayla"},{"family":"Zervos","given":"Silvia"},{"family":"Sohmer","given":"Joshua"},{"family":"Knecht","given":"Michelle"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/j8gwy","URL":"https://doi.org/10.17605/osf.io/j8gwy","source":"datacite"},{"id":"doi:10.17605/osf.io/rdpqh","type":"article-journal","title":"Current landscape of the use of artificial intelligence-based conversational agents (chatbots) in prenatal care: A scoping review.","abstract":"Introduction: Uninterrupted prenatal care is fundamental for mitigating maternal-fetal morbidity and mortality. However, limited consultation times and structural barriers within primary care restrict the provision of comprehensive health education. Consequently, pregnant women frequently turn to digital sources to resolve doubts, risking exposure to medical misinformation. In recent years, the advancement of artificial intelligence (AI) and automated conversational agents (chatbots) has offered a scalable solution to provide immediate, empathetic, and evidence-based support. Objective: To explore and describe the scope of current literature regarding the design, clinical application, and acceptability of artificial intelligence-driven conversational agents (chatbots) used for health education and the promotion of adherence during prenatal care. Methodology: A scoping review of scientific articles, clinical trials, and observational studies published in the last 10 years in databases such as PubMed, Scopus, and Web of Science. The search strategy utilizes MeSH terms and text words associated with the themes of prenatal care, artificial intelligence, natural language processing, chatbots, and health education. Expected results: This study expects to consolidate an evidence map detailing the design, architecture, and implementation of artificial intelligence-based conversational agents (chatbots) during prenatal care. Additionally, it projects evaluating their clinical and educational benefits regarding maternal adherence, outlining the ethical, technical, and acceptability barriers for their future application in primary care. Descriptors (DeCS/MeSH): Prenatal Care; Artificial Intelligence; Natural Language Processing; Health Education; Patient Compliance. Keywords:Prenatal care, chatbots, artificial intelligence, health education, adherence.","author":[{"family":"Botia","given":"Danna"},{"family":"Bautista","given":"Ana"},{"family":"Rincon","given":"Erwin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/rdpqh","URL":"https://doi.org/10.17605/osf.io/rdpqh","source":"datacite"},{"id":"doi:10.17605/osf.io/nq6d4","type":"article-journal","title":"Artificial intelligence-guided antibiotic therapy in pediatric populations: Scoping review","abstract":"More than 100 years ago, the first antibiotic was developed; since then, antibiotics have been one of the greatest inventions of science and medicine in the 20th century. However, this has been accompanied by a large number of new challenges, one of the most important being the increase in bacterial resistance (1), which is now one of the greatest threats to global health, affecting everyone, regardless of age or country of residence. Antibiotic resistance increases medical costs, prolongs hospital stays, and also increases mortality rates (2). There are certain groups in which the management of antibiotics represents a greater challenge, one of these being the pediatric population. Patients in whom symptoms are more nonspecific, sampling is more difficult, and, in addition, these patients have variable microbiological response times. This has led to greater variability in drug choice and treatment duration among healthcare professionals (3). Bacterial infections in pediatric patients are of vital importance. According to the WHO, pneumonia caused 808,000 deaths in children under 5 years of age in 2017, accounting for 15% of deaths in this age group (4). In 2024, the WHO published a list of the most dangerous bacterial pathogens to human health due to their antibiotic resistance profile, highlighting critically important pathogens such as carbapenem-resistant Acinetobacter baumannii and rifampicin-resistant Mycobacterium tuberculosis, among others (5). With the advent of artificial intelligence, a significant number of opportunities have been identified to improve the management of infectious diseases. Taking into account medical history, patients' vital signs, laboratory results, and microbiology, tools have been developed with the help of artificial intelligence that have contributed to clinical decision-making and therapeutic recommendations for patients (6). In the case of the pediatric population, artificial intelligence has been implemented in different scenarios. One example is the prediction of severe sepsis with the help of a machine learning algorithm, which has had favorable results in detecting patients with this pathology (7). Similarly, tools have been developed to determine the start of antibiotic therapy, such as the AiSEPTRON study, conducted in the United Kingdom, which demonstrated that machine learning models can be used to accurately identify patients who require antibiotic therapy in different scenarios (8). However, due to the limited availability of pediatric data in different databases and heterogeneity among age groups, the appropriate management of antibiotics in the pediatric population with the help of artificial intelligence remains a challenge. Therefore, there is a need for a comprehensive review that synthesizes the available knowledge on the use of artificial intelligence as a tool to guide antimicrobial therapies in the pediatric population. This review will identify the technologies developed in recent years, identify their advantages and disadvantages, analyze opportunities for improvement, and synthesize the benefits they have brought to the management of bacterial infections in pediatric patients. It is hoped that the findings of this research will help guide educators, researchers, healthcare professionals, and decision-makers toward the use and implementation of these tools in appropriate contexts.","author":[{"family":"Ferro","given":"Daniel"},{"family":"Benavides","given":"Catalina"},{"family":"Rincon","given":"Erwin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/nq6d4","URL":"https://doi.org/10.17605/osf.io/nq6d4","source":"datacite"},{"id":"doi:10.48550/arxiv.2509.03906","type":"manuscript","title":"Toward Clinically Explainable AI for Medical Diagnosis: A Foundation Model with Human-Compatible Reasoning via Reinforcement Learning","abstract":"The clinical adoption of artificial intelligence (AI) in medical diagnostics is critically hampered by its black-box nature, which prevents clinicians from verifying the rationale behind automated decisions. To overcome this fundamental barrier, we introduce DeepMedix-R1, a foundation model (FM) for chest X-ray (CXR) interpretation that generates not only accurate diagnoses but also a transparent, step-by-step reasoning process grounded in specific visual evidence. Our methodology employs a sequential training strategy, beginning with instruction fine-tuning, followed by a cold-start phase to elicit reasoning capabilities. Critically, we then implement reinforcement learning with grounded rewards to meticulously refine the model, aligning both its diagnostic outputs and its reasoning pathways with clinical plausibility. Quantitative assessments show that DeepMedix-R1 substantially outperforms advanced FMs, achieving improvements in report generation and visual question answering tasks. We also introduce Report Arena, a novel LLM-based benchmark that ranks DeepMedix-R1 first among competing models for output quality. Most significantly, a formal review by clinical experts reveals a profound preference for DeepMedix-R1's generated reasoning over the broadly adopted Qwen2.5-VL-7B model, confirming its superior interpretability and clinical utility.","author":[{"family":"Lin","given":"Qika"},{"family":"Zhu","given":"Yifan"},{"family":"Pu","given":"Bin"},{"family":"Huang","given":"Ling"},{"family":"Luo","given":"Haoran"},{"family":"Ma","given":"Jingying"},{"family":"Wu","given":"Feng"},{"family":"He","given":"Kai"},{"family":"Xu","given":"Jiaxing"},{"family":"Peng","given":"Zhen"},{"family":"Zhao","given":"Tianzhe"},{"family":"Xu","given":"Fangzhi"},{"family":"Zhang","given":"Jian"},{"family":"Ou","given":"Zhonghong"},{"family":"Cambria","given":"Erik"},{"family":"Mishra","given":"Swapnil"},{"family":"Feng","given":"Mengling"}],"issued":{"date-parts":[[2025]]},"DOI":"10.48550/arxiv.2509.03906","URL":"https://doi.org/10.48550/arxiv.2509.03906","source":"datacite"},{"id":"doi:10.17605/osf.io/qfyb8","type":"article-journal","title":"Artificial intelligence-based clinical simulation and its contribution to learning and cognitive processes in undergraduate medical students: a scoping review","abstract":"Introduction: The training of future physicians occurs in a context where preventable clinical errors are still frequent and safe decision-making is increasingly complex. Faced with this scenario, clinical simulation has emerged as a key strategy for risk-free learning for patients. More recently, the incorporation of artificial intelligence has begun to transform these simulated environments, allowing for more interactive, adaptive, and personalized experiences. However, the real impact of Artificial intelligence-based clinical simulation on learning outcomes and cognitive processes in undergraduate medical students has not yet been clearly systematized. Methodology: A scoping review will be carried out to analyze scientific articles, educational studies, reviews, and technical literature published between 2015 and 2025, which address the use of Artificial intelligence-based clinical simulation in medical education. The search will be carried out in the PubMed, Scopus, Web of Science and BIREME databases, using MeSH and DeCS descriptors related to artificial intelligence, clinical simulation, learning and medical education, complemented by the review of grey literature and the chain search technique of references. Expected results: It is expected to map the current applications of Artificial intelligence-based clinical simulation in undergraduate medical education, as well as the learning outcomes, cognitive processes, benefits, challenges, and knowledge gaps reported. Conclusions: This review will offer a comprehensive view of the potential and limitations of clinical simulation based on artificial intelligence, contributing to guide its critical and pedagogically grounded incorporation into undergraduate medical training. Descriptors (DeCS/MeSH): Artificial Intelligence; Clinical Simulation; Medical Education; Medical Students; Learning. Keywords: Clinical simulation; artificial intelligence; undergraduate medical education; clinical reasoning; learning.","author":[{"family":"Pesantes","given":"Mariana"},{"family":"Franco","given":"Sofía"},{"family":"Sanmiguel","given":"Juliana"},{"family":"Rodriguez","given":"Lidia"},{"family":"Rincon","given":"Erwin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/qfyb8","URL":"https://doi.org/10.17605/osf.io/qfyb8","source":"datacite"},{"id":"doi:10.17605/osf.io/9vwyb","type":"article-journal","title":"To review the application of multimodal knowledge graph in full-cycle health management of patients with dementia","abstract":"This scoping review aims to systematically map the existing literature on the application of multimodal knowledge graphs in the full-cycle health management of patients with dementia. With the global aging population and rising dementia prevalence, there is an urgent need for integrated, intelligent health management approaches. Multimodal knowledge graphs integrate heterogeneous data sources (e.g., clinical records, medical imaging, behavioral monitoring) and hold promise for personalized and continuous dementia care. The review will examine how these knowledge graphs are constructed, identify their specific application forms in full-cycle health management, analyze the main content areas covered, and evaluate reported effectiveness. Following established scoping review methodology, we will systematically search nine databases (Wanfang Data, CNKI, VIP, SinoMed, PubMed, Embase, Web of Science, CINAHL, Cochrane Library) from inception to December 2025. Two independent reviewers will screen studies, extract data, and synthesize findings. Expected outcomes include a comprehensive overview of current applications, a classification of application scenarios and technological approaches, identification of research gaps and limitations, and recommendations for future research and practical implementation. This review will provide a foundational understanding for researchers and practitioners in dementia care and artificial intelligence.","author":[{"family":"Dai","given":"Ruru"},{"family":"Jiang","given":"Weiwei"},{"family":"Mao","given":"Xingmei"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/9vwyb","URL":"https://doi.org/10.17605/osf.io/9vwyb","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8303026","type":"article-journal","title":"Scoping insights into artificial intelligence-driven treatment of diabetes mellitus in clinical practice","abstract":"Abstract Background Diabetes mellitus (DM) is a significant health concern around the world due to its increasing prevalence, high rates of morbidity and morbidity from diabetes and diabetic complications, and the associated economic burden. Despite the rapid development of digital healthcare ecosystem, DM remains an incurable lifelong disorder. Nevertheless, there has been a growing interest in developing and using artificial intelligence (AI) technologies for DM management and care. By advancing the understanding of AI-driven technology in treatment of DM and refining clinical approaches, healthcare providers can better navigate the challenges and maximize the benefits associated with these technologies. From this perspective, this scoping review aimed to provide insights to the most recent applications and progress of AI technology to various aspects and opportunities of DM treatment in clinical practice. Methods Comprehensive review following database search on the applications of AI in treatment of DM (T1DM and T2DM) in clinical settings. A literature search was conducted in databases, such as PubMed, Web of Science, and Scopus. The search covered references from 2000 to 2024 with data extraction and organized thematically yielding a total of 14 relevant studies included in this study. Results The majority of the studies were based on database analysis using AI (n = 7) followed by randomized controlled trials (n = 5). This review found that the application of AI, specifically machine learning (ML) and deep learning (DL)-based medical devices and prediction models, shows great promise for real-time monitoring and management, personalized treatment planning, and has advanced significantly in supporting predictive models for the treatment of DM or its complications. Conclusion AI has the potential to change the way this chronic disease is treated and can provide an additional opportunity to achieve better efficiency in DM care. Nevertheless, most applications are still adjunctive (decision-support) and face significant practical, ethical, and validation hurdles.","author":[{"family":"Al-Taie","given":"Anmar"},{"family":"Hafida","given":"Majida"},{"family":"Abdulsattar","given":"Mina"},{"family":"El Mahmoud","given":"Rayan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8303026","URL":"https://doi.org/10.6084/m9.figshare.c.8303026","source":"datacite"},{"id":"doi:10.6084/m9.figshare.c.8303026.v1","type":"article-journal","title":"Scoping insights into artificial intelligence-driven treatment of diabetes mellitus in clinical practice","abstract":"Abstract Background Diabetes mellitus (DM) is a significant health concern around the world due to its increasing prevalence, high rates of morbidity and morbidity from diabetes and diabetic complications, and the associated economic burden. Despite the rapid development of digital healthcare ecosystem, DM remains an incurable lifelong disorder. Nevertheless, there has been a growing interest in developing and using artificial intelligence (AI) technologies for DM management and care. By advancing the understanding of AI-driven technology in treatment of DM and refining clinical approaches, healthcare providers can better navigate the challenges and maximize the benefits associated with these technologies. From this perspective, this scoping review aimed to provide insights to the most recent applications and progress of AI technology to various aspects and opportunities of DM treatment in clinical practice. Methods Comprehensive review following database search on the applications of AI in treatment of DM (T1DM and T2DM) in clinical settings. A literature search was conducted in databases, such as PubMed, Web of Science, and Scopus. The search covered references from 2000 to 2024 with data extraction and organized thematically yielding a total of 14 relevant studies included in this study. Results The majority of the studies were based on database analysis using AI (n = 7) followed by randomized controlled trials (n = 5). This review found that the application of AI, specifically machine learning (ML) and deep learning (DL)-based medical devices and prediction models, shows great promise for real-time monitoring and management, personalized treatment planning, and has advanced significantly in supporting predictive models for the treatment of DM or its complications. Conclusion AI has the potential to change the way this chronic disease is treated and can provide an additional opportunity to achieve better efficiency in DM care. Nevertheless, most applications are still adjunctive (decision-support) and face significant practical, ethical, and validation hurdles.","author":[{"family":"Al-Taie","given":"Anmar"},{"family":"Hafida","given":"Majida"},{"family":"Abdulsattar","given":"Mina"},{"family":"El Mahmoud","given":"Rayan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.c.8303026.v1","URL":"https://doi.org/10.6084/m9.figshare.c.8303026.v1","source":"datacite"},{"id":"doi:10.48550/arxiv.2510.24750","type":"manuscript","title":"Opportunistic Screening of Wolff-Parkinson-White Syndrome using Single-Lead AI-ECG Mobile System: A Real-World Study of over 3.5 million ECG Recordings in China","abstract":"Wolff-Parkinson-White (WPW) syndrome, a congenital cardiac conduction abnormality with low prevalence, carries a significant risk of sudden cardiac death. Early identification remains challenging due to screening costs and professional resource scarcity. This retrospective real-world study systematically evaluates an integrated Artificial Intelligence-enabled mobile screening system comprising portable single-lead devices, AI primary screening, and cardiologist review. Analyzing 3,566,626 ECG records from 87,836 individuals between 2019 and 2025, the AI model achieved an AUC of 0.6676 and a specificity of 95.92% in complex real-world signal environments. Despite predictive probability bias inherent in ultra-low prevalence contexts, the model demonstrated stable risk stratification, with high-confidence scores concentrated among true positive individuals. The risk of detecting WPW in AI-positive records was 86.2-fold higher than in AI-negative records. By implementing a human-AI collaborative workflow, the volume of ECGs requiring manual review was reduced by approximately 99.5% compared to universal screening. In an ideal collaborative scenario, an average of only 18 ECGs required review to confirm one WPW case, representing a more than 60-fold increase in screening efficiency. Compared to traditional 12-lead ECGs and electrophysiological studies, this system significantly reduced time and medical costs. Our findings suggest that a risk-stratification-based human-AI collaborative system provides a promising paradigm for the early public health detection of low-prevalence, high-risk arrhythmias.","author":[{"family":"Huang","given":"Shun"},{"family":"Zhang","given":"Deyun"},{"family":"Fan","given":"Sumei"},{"family":"Tang","given":"Gongzheng"},{"family":"Geng","given":"Shijia"},{"family":"Xiao","given":"Yujie"},{"family":"Wu","given":"Xingliang"},{"family":"Yan","given":"Mingke"},{"family":"Wang","given":"Haoyu"},{"family":"Zhang","given":"Rui"},{"family":"Fu","given":"Zhaoji"},{"family":"Hong","given":"Shenda"}],"issued":{"date-parts":[[2025]]},"DOI":"10.48550/arxiv.2510.24750","URL":"https://doi.org/10.48550/arxiv.2510.24750","source":"datacite"},{"id":"doi:10.48550/arxiv.2505.04769","type":"manuscript","title":"Vision-Language-Action (VLA) Models: Concepts, Progress, Applications and Challenges","abstract":"Vision-Language-Action (VLA) models mark a transformative advancement in artificial intelligence, aiming to unify perception, natural language understanding, and embodied action within a single computational framework. This foundational review presents a comprehensive synthesis of recent advancements in Vision-Language-Action models, systematically organized across five thematic pillars that structure the landscape of this rapidly evolving field. We begin by establishing the conceptual foundations of VLA systems, tracing their evolution from cross-modal learning architectures to generalist agents that tightly integrate vision-language models (VLMs), action planners, and hierarchical controllers. Our methodology adopts a rigorous literature review framework, covering over 80 VLA models published in the past three years. Key progress areas include architectural innovations, efficient training strategies, and real-time inference accelerations. We explore diverse application domains such as autonomous vehicles, medical and industrial robotics, precision agriculture, humanoid robotics, and augmented reality. We analyzed challenges and propose solutions including agentic adaptation and cross-embodiment planning. Furthermore, we outline a forward-looking roadmap where VLA models, VLMs, and agentic AI converge to strengthen socially aligned, adaptive, and general-purpose embodied agents. This work, is expected to serve as a foundational reference for advancing intelligent, real-world robotics and artificial general intelligence. The project repository is available on GitHub as https://github.com/Applied-AI-Research-Lab/Vision-Language-Action-Models-Concepts-Progress-Applications-and-Challenges. [Index Terms: Vision Language Action, VLA, Vision Language Models, VLMs, Action Tokenization, NLP]","author":[{"family":"Sapkota","given":"Ranjan"},{"family":"Cao","given":"Yang"},{"family":"Roumeliotis","given":"Konstantinos"},{"family":"Karkee","given":"Manoj"}],"issued":{"date-parts":[[2025]]},"DOI":"10.48550/arxiv.2505.04769","URL":"https://doi.org/10.48550/arxiv.2505.04769","source":"datacite"},{"id":"doi:10.5281/zenodo.20663669","type":"article-journal","title":"Alcohol Use Disorder Prediction from EEG Using AI Algorithms: A Comprehensive Review","abstract":"Alcohol Use Disorder (AUD) is a relapsing condition, which is chronic and causes severe neurological, behavioral, and social disabilities. The traditional methods of diagnosing AUD are mainly based on the self-reported questionnaires and clinical interviews that are subjective and unsuitable in predicting the condition early. The latest developments in acquiring brain signals and artificial intelligence (AI) have allowed objective and data-based methods of predicting AUD based on neurophysiological biomarkers. The most appealing are systems updated using electroencephalogram (EEG) because they are not invasive, can be used to measure high- temporiological features. The current paper provides a review of the recent research (2023-2026) dedicated to the design and development of EEG-based wearable context to prediction the Alcohol Use Disorder based on the AI algorithm. The systematic review of the studies focused on EEG and neuroimaging (MRI/ fMRI)-based data reviews is done, with a focus on signal preprocessing methods, channel selection methods, feature representations, and learning methods. Special focus is on such types of deep learning architectures as convolutional neural network (CNNs), long short-term memory (LSTM) networks, and transformer-based models and lightweight and deployment-friendly models like EEGNet and Light Gradient Boosting Machine (LightGBM). It currently compares their results with the other available literature in terms of performance metrics, model complexity, and its appropriateness in real-time wearable implementation. Besides, the paper presents major issues in the EEG-based detection of AUD such as small dataset used, inter subject variability, explicability of AI models, and the limitations of hardware in wearable systems. The purpose of this review is to make valuable contributions to the researchers and practitioners in the field of developing reliable, efficient, systems to predict Alcohol Use Disorder in its early stage.","author":[{"family":"Salim","given":"Saara"},{"family":"Anupama","given":"AP"},{"family":"Devadathan","given":"DK"},{"family":"Hassan","given":"Reyhan"},{"family":"Jacob","given":"Salga"},{"family":"Jayashree","given":"Dr"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20663669","URL":"https://doi.org/10.5281/zenodo.20663669","source":"datacite"},{"id":"doi:10.5281/zenodo.20663670","type":"article-journal","title":"Alcohol Use Disorder Prediction from EEG Using AI Algorithms: A Comprehensive Review","abstract":"Alcohol Use Disorder (AUD) is a relapsing condition, which is chronic and causes severe neurological, behavioral, and social disabilities. The traditional methods of diagnosing AUD are mainly based on the self-reported questionnaires and clinical interviews that are subjective and unsuitable in predicting the condition early. The latest developments in acquiring brain signals and artificial intelligence (AI) have allowed objective and data-based methods of predicting AUD based on neurophysiological biomarkers. The most appealing are systems updated using electroencephalogram (EEG) because they are not invasive, can be used to measure high- temporiological features. The current paper provides a review of the recent research (2023-2026) dedicated to the design and development of EEG-based wearable context to prediction the Alcohol Use Disorder based on the AI algorithm. The systematic review of the studies focused on EEG and neuroimaging (MRI/ fMRI)-based data reviews is done, with a focus on signal preprocessing methods, channel selection methods, feature representations, and learning methods. Special focus is on such types of deep learning architectures as convolutional neural network (CNNs), long short-term memory (LSTM) networks, and transformer-based models and lightweight and deployment-friendly models like EEGNet and Light Gradient Boosting Machine (LightGBM). It currently compares their results with the other available literature in terms of performance metrics, model complexity, and its appropriateness in real-time wearable implementation. Besides, the paper presents major issues in the EEG-based detection of AUD such as small dataset used, inter subject variability, explicability of AI models, and the limitations of hardware in wearable systems. The purpose of this review is to make valuable contributions to the researchers and practitioners in the field of developing reliable, efficient, systems to predict Alcohol Use Disorder in its early stage.","author":[{"family":"Salim","given":"Saara"},{"family":"Anupama","given":"AP"},{"family":"Devadathan","given":"DK"},{"family":"Hassan","given":"Reyhan"},{"family":"Jacob","given":"Salga"},{"family":"Jayashree","given":"Dr"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20663670","URL":"https://doi.org/10.5281/zenodo.20663670","source":"datacite"},{"id":"doi:10.5281/zenodo.21513566","type":"article-journal","title":"Symptom Intelligence: High-Accuracy Disease Prediction with Ensemble Learning","abstract":"Accurate and timely disease prediction based on patient-reported symptoms remains a significant challenge in modern healthcare. Early diagnosis is essential for improving treatment outcomes and enhancing overall healthcare efficiency. However, many existing symptom-based diagnostic systems rely on binary symptom representation and a single machine learning model, which may limit their clinical realism, robustness, and predictive reliability. To address these limitations, this study proposes an ensemble learning framework that incorporates symptom severity for multi-disease prediction. Instead of binary encoding, symptoms are represented using weighted severity scores to better capture their clinical intensity. The proposed framework integrates three widely used classifiers Random Forest, Naïve Bayes, and Support Vector Machine and combines their predictions through a majority-voting ensemble strategy. The framework is evaluated using the publicly available Kaggle Disease Symptom Description Dataset, which contains 4,920 records across 41 diseases and 133 symptoms. Data preprocessing includes severity mapping, label encoding, and feature construction, followed by model training using an 80:20 train–test split. Experimental results show that while individual models achieve strong performance, the ensemble approach consistently provides higher accuracy, precision, recall, and F1-score. Additionally, the trained model is deployed in a web-based prototype that accepts user symptoms and generates disease predictions with recommended precautions, demonstrating the practical applicability of the proposed approach.","author":[{"family":"Shelatkar","given":"Vedang"},{"family":"Chavan","given":"Vedant"},{"family":"Khanche","given":"Gousiya"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21513566","URL":"https://doi.org/10.5281/zenodo.21513566","source":"datacite"},{"id":"doi:10.5281/zenodo.21513567","type":"article-journal","title":"Symptom Intelligence: High-Accuracy Disease Prediction with Ensemble Learning","abstract":"Accurate and timely disease prediction based on patient-reported symptoms remains a significant challenge in modern healthcare. Early diagnosis is essential for improving treatment outcomes and enhancing overall healthcare efficiency. However, many existing symptom-based diagnostic systems rely on binary symptom representation and a single machine learning model, which may limit their clinical realism, robustness, and predictive reliability. To address these limitations, this study proposes an ensemble learning framework that incorporates symptom severity for multi-disease prediction. Instead of binary encoding, symptoms are represented using weighted severity scores to better capture their clinical intensity. The proposed framework integrates three widely used classifiers Random Forest, Naïve Bayes, and Support Vector Machine and combines their predictions through a majority-voting ensemble strategy. The framework is evaluated using the publicly available Kaggle Disease Symptom Description Dataset, which contains 4,920 records across 41 diseases and 133 symptoms. Data preprocessing includes severity mapping, label encoding, and feature construction, followed by model training using an 80:20 train–test split. Experimental results show that while individual models achieve strong performance, the ensemble approach consistently provides higher accuracy, precision, recall, and F1-score. Additionally, the trained model is deployed in a web-based prototype that accepts user symptoms and generates disease predictions with recommended precautions, demonstrating the practical applicability of the proposed approach.","author":[{"family":"Shelatkar","given":"Vedang"},{"family":"Chavan","given":"Vedant"},{"family":"Khanche","given":"Gousiya"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21513567","URL":"https://doi.org/10.5281/zenodo.21513567","source":"datacite"},{"id":"doi:10.48550/arxiv.2608.20315","type":"manuscript","title":"Explainable Transformer Models for Clinical Prediction Tasks on Structured Electronic Health Records","abstract":"Predictive models over structured electronic health records (EHRs) remain central to machine learning for healthcare, but few have jointly emphasized quantitative laboratory information and interpretability with respect to input medical events. We present BERT-LER, a BERT-style model for coded EHR timelines pretrained and fine-tuned from a de-identified EHR dataset of 75 million patients, that encodes laboratory test results as discrete tokens while retaining graded information through percentile-based binning, paired with Integrated Gradients for token-level attributions grounded in the input EHR sequence. We benchmark our approach on the public EHRShot benchmark suite and on an asthma severity progression study based on real-world data. This addresses a methodological gap in EHR foundation-style modeling by unifying laboratory value representation and explainability in a single framework, while assessing whether both predictive performance and explanations generalize beyond standard clinical prediction tasks. Across EHRShot and asthma tasks, BERT-LER achieves predictive performance that is competitive with, and on laboratory-related tasks often exceeds, publicly available benchmark models, and provides attributions that align with clinically known risk factors. Our architecture and explainability approach can be applied to many therapeutic areas and prediction tasks using language models trained on structured EHRs.","author":[{"family":"Du","given":"Jun"},{"family":"Adamek","given":"Lukas"},{"family":"Kryukov","given":"Maxim"},{"family":"Dormont","given":"Flavio"},{"family":"Bar-Joseph","given":"Ziv"},{"family":"Jager","given":"Sven"},{"family":"Rufino","given":"Brandon"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2608.20315","URL":"https://doi.org/10.48550/arxiv.2608.20315","source":"datacite"},{"id":"doi:10.48550/arxiv.2608.19596","type":"manuscript","title":"Martingale R-learner: Estimating Time-varying Heterogeneous Treatment Effects for Time-to-event Outcomes","abstract":"Biological research and clinical evidence suggest that treatment response may vary substantially along characteristics, such as comorbidities, genetic variants, environmental, or socio-economic factors. Future precision medicine requires accurate assessment of heterogeneous treatment effects (HTE) to guide optimal clinical decisions at the individual level. We introduce a functional score framework that extends the traditional estimating equations for survival data to nonparametric HTE and generalize the Neyman orthogonality accordingly, thus filling a methodological as well as theoretical gap. Under the Neyman orthogonal functional score framework, we developed the martingale R-learner based on a decomposition of the conditional martingale residuals into residuals of the risk-set propensity score and the marginal martingale, thereby reducing the impact of estimation bias in HTE from nuisance models including (1) marginal survival, and (2) risk-set propensity scores. This enables leveraging advances in machine learning and incorporates flexible estimators for the nuisance functions and attaining the standard optimal nonparametric estimation rate with the oracle property. Numerical experiments demonstrated empirical performance consistent with the theory. We applied the martingale R-learner to estimate the effect of alcohol on dementia using the Honolulu-Asia Aging Study data.","author":[{"family":"Hou","given":"Jue"},{"family":"Qi","given":"Yuchen"},{"family":"Xu","given":"Ronghui"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2608.19596","URL":"https://doi.org/10.48550/arxiv.2608.19596","source":"datacite"},{"id":"doi:10.48550/arxiv.2608.19578","type":"manuscript","title":"A Two-Stage Time-Aware Transformer for Short-Horizon AECOPD Risk Prediction","abstract":"Acute exacerbation of chronic obstructive pulmonary disease (AECOPD) can worsen rapidly, making timely prediction a clinical priority. Most existing machine learning approaches rely on episodically collected clinical variables, introducing delays that limit their practical utility in home monitoring settings. Home ventilators offer a lower-latency alternative, producing a near-continuous record of respiratory status during daily use. However existing ventilator-based approaches either compress the waveform into handcrafted features or focus primarily on binary risk classification, leaving the timing of an impending event unresolved. In this paper, we present a two-stage framework that operates directly on raw pressure and flow waveforms from the most recent seven days of home ventilator use. The first-stage classification model identifies patients at high risk of a severe exacerbation. The second-stage regression model then estimates how many days remain before the event occurs. Our experimental results demonstrate that the two-stage model outperforms traditional baseline models on both risk classification and time-to-event estimation, with our selected Stage 1 classifier achieving F1 = 0.91 and our Stage 2 regression model achieving RMSE = 1.00 days and R^2 = 0.76, giving clinicians both an early warning and actionable lead time before a severe exacerbation occurs.","author":[{"family":"Wang","given":"Dongyang"},{"family":"Qu","given":"Weihao"},{"family":"Zheng","given":"Ling"},{"family":"Pan","given":"Haowen"}],"issued":{"date-parts":[[2026]]},"DOI":"10.48550/arxiv.2608.19578","URL":"https://doi.org/10.48550/arxiv.2608.19578","source":"datacite"},{"id":"oa:W4414989528","type":"article-journal","title":"Dimensions of Artificial Intelligence Literacy: A Qualitative Synthesis of Contemporary Research Literature","abstract":"Artificial Intelligence (AI) is transforming education, workforce development, and daily life, necessitating a comprehensive understanding of AI literacy. This study explores the dimensions of AI literacy, its integration into educational and professional settings, and the challenges associated with its implementation. Using a systematic review and qualitative synthesis, this study examines research published between 2019 and 2024, identifying six key dimensions of AI literacy: technical literacy, ethical and societal awareness, critical AI literacy, AI in everyday life, human-AI collaboration, and AI pedagogical literacy. Findings indicate that AI literacy is increasingly embedded in K-12 education, higher education, and workforce training, though disparities in accessibility, ethical concerns, and inconsistent policies persist. Key challenges include the digital divide, lack of teacher training, and lack of standardized AI literacy assessment tools. Opportunities lie in interdisciplinary learning, project-based education, and AI-driven adaptive learning environments. This study makes several unique contributions, including a comprehensive framework for AI literacy, integration of AI literacy across education and workforce domains, identification of policy gaps, and a call for standardized AI literacy assessment tools. It also emphasizes the need for ethical AI engagement and responsible AI education. Policymakers and educators should prioritize integrating AI literacy into curricula, professional development for teachers, and establishing regulatory frameworks to ensure equitable AI education. Future research should focus on longitudinal studies, cross-cultural AI literacy comparisons, and developing adaptive AI learning models to enhance AI education globally.","author":[{"family":"Kaplan","given":"Roza"},{"family":"Meylani","given":"Ruşen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.18009/jcer.1648380","URL":"https://doi.org/10.18009/jcer.1648380","source":"openalex"},{"id":"oa:W4412881896","type":"article-journal","title":"Incorporating Artificial Intelligence into Fracture Risk Assessment: Using Clinical Imaging to Predict the Unpredictable","abstract":"Artificial intelligence (AI) is increasingly being explored as a complementary tool to traditional fracture risk assessment methods. Conventional approaches, such as bone mineral density measurement and established clinical risk calculators, provide populationlevel stratification but often fail to capture the structural nuances of bone fragility. Recent advances in AI-particularly deep learning techniques applied to imaging-enable opportunistic screening and individualized risk estimation using routinely acquired radiographs and computed tomography (CT) data. These models demonstrate improved discrimination for osteoporotic fracture detection and risk prediction, supporting applications such as time-to-event modeling and short-term prognosis. CT- and radiograph-based models have shown superiority over conventional metrics in diverse cohorts, while innovations like multitask learning and survival plots contribute to enhanced interpretability and patient-centered communication. Nevertheless, challenges related to model generalizability, data bias, and automation bias persist. Successful clinical integration will require rigorous external validation, transparent reporting, and seamless embedding into electronic medical systems. This review summarizes recent advances in AI-driven fracture assessment, critically evaluates their clinical promise, and outlines a roadmap for translation into real-world practice.","author":[{"family":"Kong","given":"Sung"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3803/enm.2025.2518","URL":"https://doi.org/10.3803/enm.2025.2518","source":"openalex"},{"id":"oa:W4410570440","type":"article-journal","title":"The Role of Artificial Intelligence Tools on Chinese EFL Learners' Self‐Regulation, Resilience and Autonomy","abstract":"ABSTRACT The incorporation of artificial intelligence (AI) in education has attracted scholars' research interest around the world, and they have demonstrated the effectiveness of using AI in the English as a foreign language (EFL) setting. While previous studies have examined AI's role in improving linguistic proficiency, limited research has examined its role in key psychological concepts such as self‐regulation, resilience and autonomy among EFL learners. Addressing this gap, the present study explored the impact of AI on the self‐regulation, resilience and autonomy of Chinese EFL students. The students were assigned to two groups: the experimental group included 94 learners who were taught through AI tools, while the control group included 89 learners who received traditional language education. A pretest was administered using three scales, namely self‐regulation, resilience and autonomy. After the completion of the instructional sessions, a posttest was administered to both groups with identical questionnaires. Eventually, the study analysed the data collected through Analysis of Covariance (ANCOVA), revealing that the group utilising AI tools significantly outperformed the others regarding all three variables. Indeed, utilising AI tools notably enhanced the self‐regulation, resilience and autonomy of EFL students. The research implications were thoroughly discussed.","author":[{"family":"Zhang","given":"Zhijuan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/ejed.70127","URL":"https://doi.org/10.1111/ejed.70127","source":"openalex"},{"id":"oa:W4410435932","type":"article-journal","title":"Artificial Intelligence in Reproductive Medicine: Transforming Assisted Reproductive Technologies","abstract":"Question Asked How is artificial intelligence (AI) transforming assisted reproductive technologies (ART), particularly in vitro fertilization (IVF), and what are its clinical impacts and limitations? Background AI offers potential to address ART challenges, including high costs, variable success rates, and rising infertility. Applications in embryo selection, gamete assessment, and personalized protocols aim to enhance objectivity and outcomes. Literature Search A systematic review of peer-reviewed articles (2019–2025) was conducted, using terms such as “artificial intelligence” and “IVF.” Studies focused on AI tools (DeepEmbryo, icONE, iDAScore, ERICA) and their performance in ART. Materials and Methods Selected studies evaluated AI applications in embryo selection, gamete assessment, personalized protocols, and outcome prediction. Performance metrics, validation scope, and clinical outcomes were analyzed, prioritizing tools with quantitative data. Results and Discussion AI tools improved clinical pregnancy rates (up to 77.3%), implantation accuracy (92%), and efficiency (35%). icONE and ERICA outperformed traditional methods, reducing subjectivity. However, validation is often limited to single-center studies, with surrogate endpoints (e.g., pregnancy rates) rather than live birth rates. Algorithmic bias, regional data privacy regulations, and high costs limit generalizability and accessibility. Ethical concerns, including data privacy and equity, require robust frameworks. Conclusions AI enhances ART efficacy and personalization but faces validation and ethical challenges. Multicenter studies focusing on live birth rates and inclusive datasets are needed to ensure equitable, clinically relevant adoption.","author":[{"family":"Shoham","given":"Zeev"}],"issued":{"date-parts":[[2025]]},"DOI":"10.46989/001c.137620","URL":"https://doi.org/10.46989/001c.137620","source":"openalex"},{"id":"oa:W4413921759","type":"article-journal","title":"Teaching Artificial Intelligence and Language Models in Medical Education","abstract":"The rapid evolution of Artificial Intelligence (AI), especially large language models (LLMs), is transforming medicine and poses new challenges for medical education. This theoretical-reflective article discusses the integration of AI and LLM teaching in the undergraduate medical curriculum, considering ethical, methodological, and practical aspects. Initially, it contextualizes the growing role of AI in healthcare and the need for AI literacy among future physicians, given the incorporation of tools like ChatGPT into clinical practice. We then review recent literature (2024-2025) and relevant guidelines, including the World Health Organization and Brazilian Ministry of Education, to identify educational benefits (such as virtual patient simulations, automated assessment and personalized feedback) and risks (algorithmic bias, response hallucinations, privacy and academic integrity issues). Methodologically, the study is based on a literature review and the author’s teaching experience implementing AI content for 3rd-year medical students. In the results and discussion, we present active teaching strategies, such as the use of prompting in simulated clinical cases and ethical debates, evaluating their perceived effectiveness in developing digital competencies. We conclude by emphasizing the importance of a structured AI curriculum in medicine, including faculty training, interdisciplinary collaboration, and adherence to ethical principles, to train physicians who are capable of harnessing new technologies in a critical and humanistic manner.","author":[{"family":"Denis-Bacelar","given":"Ana"}],"issued":{"date-parts":[[2025]]},"DOI":"10.64326/educao.v1i6.76","URL":"https://doi.org/10.64326/educao.v1i6.76","source":"openalex"},{"id":"oa:W4414688765","type":"article-journal","title":"Morgellons Disease as a Multisystem Problem: Analysis of Hypotheses, Scientific Evidence, How Artificial Intelligence Helps","abstract":"Morgellons disease is a rare and controversial disease. It combines dermatological, infectious, and psychiatric manifestations. It is characterized by the formation of skin lesions with filaments emanating from the tissues. Also, by sensory phenomena in the form of a crawling or stinging sensation under the skin. As well as systemic symptoms, including chronic fatigue and cognitive disorders. The article analyzes the literature and clinical reports that highlight the infectious and psychiatric hypotheses, morphological and biochemical features, sociocultural factors, and problems of diagnosis and treatment. It is established that Morgellons disease remains uncertain in the nosological classification, which complicates the creation of unified approaches to its management. At the same time, the need for a multidisciplinary, multisystemic approach that integrates dermatology, infectious diseases, biochemistry, microbiology, clinical diagnostics, immunology, psychiatry, and psychology is emphasized. The conclusion is made about the urgent need for conducting systematic studies of etiology and pathogenesis, developing standardized diagnostic criteria and therapeutic protocols, as well as overcoming the stigmatization of patients. The help of artificial intelligence opens new opportunities for early diagnosis of Morgellons disease through the analysis of large arrays of clinical and laboratory data, recognition of dermatological images and prediction of the course of the disease. The use of machine learning algorithms contributes to more accurate differential diagnosis between infectious, dermatological, and psychiatric manifestations, which helps to minimize false diagnoses. The use of artificial intelligence in telemedicine and clinical practice of Morgellons disease allows for a personalized approach to treatment, optimizing the choice of therapeutic strategies and monitoring the effectiveness of treatment. The integration of artificial intelligence technologies into research and clinical practice is considered a promising direction for overcoming diagnostic and therapeutic difficulties in Morgellons disease.","author":[{"family":"Шаповалова","given":"Вікторія"}],"issued":{"date-parts":[[2025]]},"DOI":"10.53933/fxbnwe45","URL":"https://doi.org/10.53933/fxbnwe45","source":"openalex"},{"id":"oa:W4410054156","type":"article-journal","title":"The Role of Artificial Intelligence in Dental Diagnosis and Treatment Planning","abstract":"This study aims to examine the role of artificial intelligence in dental diagnosis and treatment planning, highlighting its applications, benefits, challenges, and future directions. A narrative review was conducted using a descriptive analysis method, synthesizing recent literature on AI-driven diagnostic tools, treatment planning systems, and emerging trends in dentistry. Studies published between 2022 and 2025 were analyzed to assess AI applications in imaging-based diagnosis, clinical decision support, orthodontic planning, prosthodontics, endodontics, periodontal treatment, implantology, and precision dentistry. Ethical concerns, integration challenges, and regulatory considerations were also reviewed to provide a comprehensive understanding of AI’s impact on dental practice. AI has significantly improved diagnostic accuracy in dental radiography, CBCT analysis, and early detection of caries, periodontal disease, and oral cancer. AI-driven treatment planning has enhanced efficiency in orthodontics, prosthodontics, and implantology by optimizing treatment simulations, material selection, and surgical precision. Teledentistry and remote diagnosis have expanded access to care, while AI-powered robotics have introduced automation in surgical and restorative procedures. Despite these advancements, challenges remain in data privacy, algorithmic bias, clinical integration, and regulatory compliance. Ethical concerns related to AI transparency, liability, and decision-making authority continue to shape its adoption in clinical practice. AI is transforming dental diagnosis and treatment planning by improving accuracy, efficiency, and accessibility. However, addressing challenges related to data security, bias, regulatory frameworks, and professional adaptation is essential for the responsible integration of AI in dentistry. Future advancements in AI algorithms, robotic-assisted dentistry, and precision medicine will further enhance patient-centered care. A balanced approach that combines AI innovation with human expertise is necessary to ensure optimal clinical outcomes and ethical AI implementation in dental healthcare.","author":[{"family":"Moeini","given":"Arash"},{"family":"Torabi","given":"SA"}],"issued":{"date-parts":[[2025]]},"DOI":"10.61838/kman.jodhn.2.1.2","URL":"https://doi.org/10.61838/kman.jodhn.2.1.2","source":"openalex"},{"id":"oa:W4408264626","type":"article-journal","title":"Application of Deep Learning and Transfer Learning Techniques for Medical Image Classification","abstract":"The advancements in deep learning (DL) and transfer learning (TL) have transformed artificial intelligence, particularly in image classification. This research examines the theoretical foundations of DL and TL, focusing on their applications in medical image classification, specifically distinguishing between COVID-19, viral pneumonia, and normal lung conditions. By leveraging GPU-enabled high-performance computing and large labeled datasets, DL models particularly Convolutional Neural Networks (CNNs) such as ResNet50 and VGG16 have achieved superior accuracy compared to traditional machine learning methods. This study explores feature extraction using pre-trained models, the implementation of classifiers like Support Vector Machines (SVM) and K-Nearest Neighbors (KNN), and the integration of multi-view learning techniques, such as early fusion. The results demonstrate the effectiveness of DL and TL in improving classification performance, highlighting their significant potential to advance global healthcare diagnostics.","author":[{"family":"Sakirin","given":"Tam"},{"family":"Said","given":"Rachid"}],"issued":{"date-parts":[[2025]]},"DOI":"10.70470/edraak/2025/006","URL":"https://doi.org/10.70470/edraak/2025/006","source":"openalex"},{"id":"oa:W4416066352","type":"article-journal","title":"Advancements in Small-Object Detection (2023–2025): Approaches, Datasets, Benchmarks, Applications, and Practical Guidance","abstract":"Small-object detection (SOD) remains an important and growing challenge in computer vision and is the backbone of many applications, including autonomous vehicles, aerial surveillance, medical imaging, and industrial quality control. Small objects, in pixels, lose discriminative features during deep neural network processing, making them difficult to disentangle from background noise and other artifacts. This survey presents a comprehensive and systematic review of the SOD advancements between 2023 and 2025, a period marked by the maturation of transformer-based architectures and a return to efficient, realistic deployment. We applied the PRISMA methodology for this work, yielding 112 seminal works in the field to ensure the robustness of our foundation for this study. We present a critical taxonomy of the developments since 2023, arranged in five categories: (1) multiscale feature learning; (2) transformer-based architectures; (3) context-aware methods; (4) data augmentation enhancements; and (5) advancements to mainstream detectors (e.g., YOLO). Third, we describe and analyze the evolving SOD-centered datasets and benchmarks and establish the importance of evaluating models fairly. Fourth, we contribute a comparative assessment of state-of-the-art models, evaluating not only accuracy (e.g., the average precision for small objects (AP_S)) but also important efficiency (FPS, latency, parameters, GFLOPS) metrics across standardized hardware platforms, including edge devices. We further use data-driven case studies in the remote sensing, manufacturing, and healthcare domains to create a bridge between academic benchmarks and real-world performance. Finally, we summarize practical guidance for practitioners, the model selection decision matrix, scenario-based playbooks, and the deployment checklist. The goal of this work is to help synthesize the recent progress, identify the primary limitations in SOD, and open research directions, including the potential future role of generative AI and foundational models, to address the long-standing data and feature representation challenges that have limited SOD.","author":[{"family":"Aldubaikhi","given":"Ali"},{"family":"Patel","given":"Sarosh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/app152211882","URL":"https://doi.org/10.3390/app152211882","source":"openalex"},{"id":"oa:W7140828524","type":"article-journal","title":"Artificial intelligence literacy and readiness in future health care professionals: a cross-sectional study","abstract":"Aim To examine the relationship between artificial intelligence (AI) literacy and readiness for medical AI among medical and nursing students, and to explore demographic and educational factors affecting AI literacy and readiness for AI. MethodsThis cross-sectional study enrolled 443 students attending the Faculty of Medicine and the Department of Nursing at Bilecik Seyh Edebali University between May and June 2025.Data were collected using a general demographic questionnaire, Medical Artificial Intelligence Readiness Scale (MAIRS), and Artificial Intelligence Literacy Scale (AILS).Results Medical students showed higher AI readiness (P < 0.001) and literacy (P = 0.032) scores than nursing students.AI literacy was moderately and positively correlated with readiness for medical AI (r = 0.338, P < 0.001), and the associations were strongest on the Awareness and Usage dimensions.In multivariable models, MAIRS scores were independently associated with the study year, department, and single-item readiness to use AI, while AILS scores were independently associated with the readiness to use AI and the belief that AI could partially replace health care workers.Internal consistency of the questionnaires was high (MAIRS α = 0.972; AILS α = 0.878). ConclusionThe findings support the integration of structured, practical, and ethically informed AI education into health science curricula, as this may be associated with greater student readiness for future AI-supported health care environments.","author":[{"family":"Catan-Inan","given":"Funda"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3325/cmj.2026.67.4","URL":"https://doi.org/10.3325/cmj.2026.67.4","source":"openalex"},{"id":"oa:W4414283151","type":"article-journal","title":"Generative Artificial Intelligence and the Future of Public Knowledge","abstract":"Generative artificial intelligence (AI), in particular large language models such as ChatGPT, have reached public consciousness with a wide-ranging discussion of their capabilities and suitability for use in various professions. Following the printing press and the internet, generative AI language models are the third transformative technological invention, with truly cross-sectoral impact on knowledge transmission and knowledge generation. While the printing press allowed for the transmission of knowledge that is independent of the physical presence of the knowledge holder, with publishers emerging as gatekeepers, the internet added levels of democratization, allowing anyone to publish, along with global immediacy. The development of social media resulted in an increased fragmentation and tribalization in online communities regarding their ways of knowing, resulting in the propagation of alternative truths that resonate in echo chambers. It is against this background that generative AI language models have entered public consciousness. Using the strategic foresight methodology, this paper will examine the proposition that the age of generative AI will emerge as an age of public ignorance.","author":[{"family":"Spennemann","given":"Dirk"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/knowledge5030020","URL":"https://doi.org/10.3390/knowledge5030020","source":"openalex"},{"id":"oa:W4412579871","type":"article-journal","title":"Knowledge, attitudes, and practices towards artificial intelligence among Ashur University-Medical College students, Baghdad-Iraq","abstract":"Artificial intelligence (AI) is a technology that enables machines to learn, reason, and carry out tasks that normally require human intelligence. AI can be used in a variety of ways to improve teaching and health, such as providing personalized learning experiences and developing new diagnosis approaches and treatments for diseases. This study aimed to assess the knowledge, attitudes, and practices (KAP) towards artificial intelligence (AI) among medical students at Ashur University-Medical College (AUMC) in Baghdad, Iraq. A descriptive cross-sectional study was conducted from November 2024 to January 2025, involving 200 first- and second-year medical students. Data was collected using a structured questionnaire adapted from previous studies, covering Sociodemographic characteristics, AI knowledge, attitudes, and practices. The results revealed that 93.5% of students understood the basic concept of AI, but only 28.5% and 12.5% were familiar with machine learning and deep learning, respectively. A significant majority (87.5%) supported the integration of AI into their field of study, with 70.5% advocating for its inclusion in medical training. However, 68.5% opposed using AI for student assessments due to concerns about fairness and bias. While 85.5% of students used AI technologies for academic purposes, only 6% had taken AI-focused courses outside their curriculum. The study highlights the need for incorporating AI education into medical curricula and fostering extracurricular engagement to better prepare students for the evolving healthcare landscape","author":[{"family":"Mahmood","given":"Amer"},{"family":"Salman","given":"Munir"}],"issued":{"date-parts":[[2025]]},"DOI":"10.21608/jbaar.2025.442546","URL":"https://doi.org/10.21608/jbaar.2025.442546","source":"openalex"},{"id":"oa:W4417192838","type":"article-journal","title":"Equity and Generalizability of Artificial Intelligence for Skin-Lesion Diagnosis Using Clinical, Dermoscopic, and Smartphone Images: A Systematic Review and Meta-Analysis","abstract":"Background and Objectives: Artificial intelligence (AI) has shown promising performance in skin-lesion classification; however, its fairness, external validity, and real-world reliability remain uncertain. This systematic review and meta-analysis evaluated the diagnostic accuracy, equity, and generalizability of AI-based dermatology systems across diverse imaging modalities and clinical settings. Materials and Methods: A comprehensive search of PubMed, Embase, Web of Science, and ClinicalTrials.gov (inception–31 October 2025) identified diagnostic accuracy studies using clinical, dermoscopic, or smartphone images. Eighteen studies (11 melanoma-focused; 7 mixed benign–malignant) met inclusion criteria. Six studies provided complete 2 × 2 contingency data for bivariate Reitsma HSROC modeling, while seven reported AUROC values with extractable variance. Risk of bias was assessed using QUADAS-2, and evidence certainty was graded using GRADE. Results: Across more than 70,000 test images, pooled sensitivity and specificity were 0.91 (95% CI 0.74–0.97) and 0.64 (95% CI 0.47–0.78), respectively, corresponding to an HSROC AUROC of 0.88 (95% CI 0.84–0.92). The AUROC-only meta-analysis yielded a similar pooled AUROC of 0.88 (95% CI 0.87–0.90). Diagnostic performance was highest in specialist settings (AUROC 0.90), followed by community care (0.85) and smartphone environments (0.81). Notably, performance was lower in darker skin tones (Fitzpatrick IV–VI: AUROC 0.82) compared with lighter skin tones (I–III: 0.89), indicating persistent fairness gaps. Conclusions: AI-based dermatology systems achieve high diagnostic accuracy but demonstrate reduced performance in darker skin tones and non-specialist environments. These findings emphasize the need for diverse training datasets, skin-tone–stratified reporting, and rigorous external validation before broad clinical deployment.","author":[{"family":"Tjiu","given":"Jeng‐wei"},{"family":"Lu","given":"Chia‐fang"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/medicina61122186","URL":"https://doi.org/10.3390/medicina61122186","source":"openalex"},{"id":"oa:W4414321309","type":"article-journal","title":"Responsible artificial intelligence?","abstract":"Abstract Although the phrase “responsible AI” is widely used in the AI industry, its meaning remains unclear. One can make sense of it indirectly, insofar as various notions of responsibility unproblematically attach to those involved in the creation and operation of AI technologies. It is less clear, however, whether the phrase makes sense when understood directly, that is, as the ascription of some sort of responsibility to AI systems themselves. This paper argues in the affirmative, drawing on a philosophically undemanding notion of role responsibility, and highlights the main consequences of this proposal for AI ethics.","author":[{"family":"Vacek","given":"Daniela"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00146-025-02604-3","URL":"https://doi.org/10.1007/s00146-025-02604-3","source":"openalex"},{"id":"oa:W4411045671","type":"article-journal","title":"Cybersecurity for Analyzing Artificial Intelligence (AI)-Based Assistive Technology and Systems in Digital Health","abstract":"Assistive technology (AT) is increasingly utilized across various sectors, including digital healthcare and sports education. E-learning plays a vital role in enabling students with special needs, particularly those in remote areas, to access education. However, as the adoption of AI-based AT systems expands, the associated cybersecurity challenges also grow. This study aims to examine the impact of AI-driven assistive technologies on cybersecurity in digital healthcare applications, with a focus on the potential vulnerabilities these technologies present. Methods: The proposed model focuses on enhancing AI-based AT through the implementation of emerging technologies used for security, risk management strategies, and a robust assessment framework. With these improvements, the AI-based Internet of Things (IoT) plays major roles within the AT. This model addresses the identification and mitigation of cybersecurity risks in AI-based systems, specifically in the context of digital healthcare applications. Results: The findings indicate that the application of the AI-based risk and resilience assessment framework significantly improves the security of AT systems, specifically those supporting e-learning for blind users. The model demonstrated measurable improvements in the robustness of cybersecurity in digital health, particularly in reducing cyber risks for AT users involved in e-learning environments. Conclusions: The proposed model provides a comprehensive approach to securing AI-based AT in digital healthcare applications. By improving the resilience of assistive systems, it minimizes cybersecurity risks for users, specifically blind individuals, and enhances the effectiveness of e-learning in sports education.","author":[{"family":"Algarni","given":"Abdullah"},{"family":"Thayananthan","given":"Vijey"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/systems13060439","URL":"https://doi.org/10.3390/systems13060439","source":"openalex"},{"id":"oa:W4410698708","type":"article-journal","title":"An overview of artificial intelligence and machine learning in shoulder surgery","abstract":"Machine learning (ML), a subset of artificial intelligence (AI), utilizes advanced algorithms to learn patterns from data, enabling accurate predictions and decision-making without explicit programming. In orthopedic surgery, ML is transforming clinical practice, particularly in shoulder arthroplasty and rotator cuff tears (RCTs) management. This review explores the fundamental paradigms of ML, including supervised, unsupervised, and reinforcement learning, alongside key algorithms such as XGBoost, neural networks, and generative adversarial networks. In shoulder arthroplasty, ML accurately predicts postoperative outcomes, complications, and implant selection, facilitating personalized surgical planning and cost optimization. Predictive models, including ensemble learning methods, achieve over 90% accuracy in forecasting complications, while neural networks enhance surgical precision through AI-assisted navigation. In RCTs treatment, ML enhances diagnostic accuracy using deep learning models on magnetic resonance imaging and ultrasound, achieving area under the curve values exceeding 0.90. ML models also predict tear reparability with 85% accuracy and postoperative functional outcomes, including range of motion and patient-reported outcomes. Despite remarkable advancements, challenges such as data variability, model interpretability, and integration into clinical workflows persist. Future directions involve federated learning for robust model generalization and explainable AI to enhance transparency. ML continues to revolutionize orthopedic care by providing data-driven, personalized treatment strategies and optimizing surgical outcomes.","author":[{"family":"Cho","given":"Sung"},{"family":"Kim","given":"Yang‐soo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5397/cise.2025.00185","URL":"https://doi.org/10.5397/cise.2025.00185","source":"openalex"},{"id":"oa:W4411938576","type":"article-journal","title":"Telemedicine and Telepharmacy in Modern Healthcare: Innovations, Medical Technologies, Digital Transformation","abstract":"The article presents an overview of modern approaches to the development of telemedicine and telepharmacy as innovative forms of providing medical and pharmaceutical services at a distance. The historical stages of the formation of telemedicine, its integration into pharmaceutical practice, the main principles of telepharmacy and the formats of its implementation are highlighted. The advantages of using telepharmacy are considered, including increasing the availability of pharmaceutical care, optimizing resources, improving the quality of pharmaceutical care, and reducing the burden on the healthcare system. The challenges and limitations of the development of telepharmacy were analyzed, including legal, ethical, and technological aspects, issues of licensing and personal data protection. Special attention is paid to the international experience of implementing telepharmaceutical services in the USA, EU countries, Canada and Australia. The current state and prospects for the development of telepharmacy in Ukraine in the context of the digitalization of healthcare are described. A special emphasis is placed on the role of artificial intelligence in supporting clinical decisions, automating pharmaceutical consultations, and introducing virtual pharmacists. Recommendations are proposed for the integration of telepharmacy into the national healthcare system and pharmaceutical practice.","author":[{"family":"Шаповалова","given":"Вікторія"}],"issued":{"date-parts":[[2025]]},"DOI":"10.53933/r7f5xj91","URL":"https://doi.org/10.53933/r7f5xj91","source":"openalex"},{"id":"oa:W4414780351","type":"article-journal","title":"Every nurse an AI nurse: A framework for integrating artificial intelligence across nursing practice, education, research and policy","abstract":"As artificial intelligence (AI) becomes increasingly embedded in healthcare, the nursing profession must embrace the imperative of 'every nurse an AI nurse'. This commentary argues that nurses must not only adapt to AI technologies but actively shape their development, implementation, and governance to ensure alignment with nursing's core values of compassionate, patient-centred care. Drawing on a five-part APDDS framework: Aware, Prepare, Dare, Declare, Share; the article outlines a strategic approach for integrating AI across nursing practice, education, research, and policy. It emphasises the need for AI literacy, leadership, ethical engagement, and knowledge dissemination to empower nurses as innovators and advocates in the digital transformation of healthcare. By embracing this proactive stance, the nursing profession can ensure that AI enhances rather than diminishes the human elements of care.","author":[{"family":"Dornan","given":"Mark"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1177/20552076251377939","URL":"https://doi.org/10.1177/20552076251377939","source":"openalex"},{"id":"oa:W4411934263","type":"article-journal","title":"Generative artificial intelligence integration in management education: application and ethical challenges","abstract":"In the modern era, Generative Artificial Intelligence (GAI) tools can significantly impact and even transform Management Education Institutes (MEIs) through advanced technological developments. Using the GAI promises extensive gains in managerial efficiency, increased learning experience, skill formation and resource availability. Nonetheless, regarding their contextual integration in the educational sector, various ethical challenges (i.e. bias, privacy, accountability and intellectual property) also exist. Moreover, the analogy of multiple factors and a straightforward visualization of their empirical connection is lacking in existing literature. Consequently, this study proposes comprehensive matrices to help teachers, trainers and policymakers adopt a balanced approach between opportunities and ethical apprehensions linked with GAI education. Along these lines, for the responsible usage of GAI inside MEIs, educational technology adoption (i.e. TAM and UTAUT) and Ethical (i.e. Utilitarianism, Deontological Ethics, and Virtue Ethics) theories are leveraged. This study recommends that future researchers further investigate GAI’s long-term changes, cross-cultural adaptations, and technological innovations. In addition, for a transparent, wide-ranging, and liable implementation of GAI, one must emphasize the continuous evaluation and diverse stakeholders’ collective efforts. One might fully explore the full potential of GAI’s responsible integration to foster the conditions supporting ethical systems and mature outcomes.","author":[{"family":"Bashir","given":"Shahid"},{"family":"Lapshun","given":"Alexander"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/2331186x.2025.2526436","URL":"https://doi.org/10.1080/2331186x.2025.2526436","source":"openalex"},{"id":"oa:W7124545921","type":"article-journal","title":"The competence paradox: when psychologists overestimate their understanding of Artificial Intelligence","abstract":"Abstract Artificial intelligence is rapidly transforming psychological practice. Psychologists now use AI to transcribe sessions, analyse client data, and generate treatment plans, yet few fully understand how these systems work. This commentary argues that the greatest risk AI poses to psychology is not its technical superiority to human capability, but a competence paradox: the tendency of psychologists to mistake the effective use of AI tools for a genuine understanding of how it works. This illusion of competence distorts judgment, weakens accountability, and undermines the foundations of professional expertise. We discuss how this gap forms, why it matters for both clients and clinicians, and what psychologists must know to engage with AI responsibly. Drawing on recent work in adjacent fields, we show how cognitive bias, identity protection, and anthropomorphism create a false sense of mastery. We then trace consequences across five domains. Cognitive and diagnostic skills decline through automation bias, cognitive offloading, and reduced reflective reasoning. Professional identity is strained as roles shift from clinician to editor of machine output. Ethical accountability blurs through hidden AI use, weak informed consent, and diffused liability. Collegial consultation diminishes as practitioners consult tools rather than peers, and wellbeing suffers through technostress, rising demands, financial strain, and growing reliance on AI. Finally, we argue that limited explainability within AI systems creates an explanatory dependence that constrains transparent justification of clinical decisions, shifts the burden of reasoning from clinician to tool, and makes embedded value choices harder to detect. We conclude with a call to action and a research agenda.","author":[{"family":"Zyl","given":"Llewellyn"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1007/s00146-025-02814-9","URL":"https://doi.org/10.1007/s00146-025-02814-9","source":"openalex"},{"id":"oa:W4412765880","type":"article-journal","title":"AI-driven epidemic intelligence: the future of outbreak detection and response","abstract":"Epidemic intelligence, the process of detecting, verifying, and analyzing public health threats to enable timely responses, traditionally relies heavily on manual reporting and structured data, often causing delays and coverage gaps. The growing frequency of emerging infectious diseases highlights the urgency for more rapid and accurate surveillance methods. This perspective proposes a forward-looking conceptual framework for AI-driven epidemic intelligence, emphasizing the transformative potential of integrating large language models (LLMs), natural language processing (NLP), and optimization-based resource allocation strategies. While existing AI-driven systems have shown significant capabilities during the COVID-19 pandemic, several challenges remain, including real-time adaptability, multilingual data handling, misinformation, and public health policy alignment. To address these gaps, we propose an integrated, real-time adaptable LLM-based epidemic intelligence system, capable of correlating cross-source data, optimizing healthcare resource allocation, and supporting informed outbreak response. This approach aims to significantly improve early warning capabilities, enhancing forecasting accuracy, and strengthen pandemic preparedness.","author":[{"family":"Kaur","given":"Jasleen"},{"family":"Butt","given":"Zahid"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/frai.2025.1645467","URL":"https://doi.org/10.3389/frai.2025.1645467","source":"openalex"},{"id":"oa:W4409717779","type":"article-journal","title":"Artificial Intelligence‐Critical Pedagogic: Design and Psychologic Validation of a Teacher‐Specific Scale for Enhancing Critical Thinking in Classrooms","abstract":"ABSTRACT Background Critical thinking is essential in modern education, and artificial intelligence (AI) offers new possibilities for enhancing it. However, the lack of validated tools to assess teachers' AI‐integrated pedagogical skills remains a challenge. Objectives The current study aimed to develop and validate the Artificial Intelligence‐Critical Pedagogy Scale (AICPS) to measure teachers' ability to use AI in fostering critical thinking. Methods This study was conducted in Saudi Arabia and consisted of two phases. Phase 1 involved item development through a literature review and semi‐structured interviews with 17 secondary school teachers, leading to an initial pool of 100 items. After expert reviews and a pilot study, the scale was refined to 47 items. Phase 2 evaluated the psychometric properties of the scale through exploratory factor analysis (EFA), confirmatory factor analysis (CFA), exploratory graph analysis (EGA), reliability assessments and measurement invariance testing across gender. The final sample included 800 secondary school teachers. Results and Conclusions EFA confirmed a four‐factor structure with 39 items. The four factors were Competence in Creating Critical Thinking‐Oriented Learning Environments (CCCTOLE), Ability to Provide Dynamic and Innovative Feedback (APDIF), Understanding and Interaction with Emerging Learning Technologies (UIELT) and Creativity in Designing Transformative Learning Activities (CDTLA). CFA demonstrated a good model fit ( χ 2 /df = 3.24, RMSEA = 0.075, CFI = 0.916 and TLI = 0.909). EGA further supported the four‐factor structure. Internal consistency was excellent, with Cronbach's alpha and McDonald's omega above 0.70 for all subscales. Measurement invariance testing confirmed that the scale functions equivalently across gender groups. The AICPS is a reliable and valid tool for assessing teachers' AI‐based critical pedagogy skills. Its demonstrated gender invariance suggests its applicability across diverse educational contexts. This scale can guide future teacher training and policy decisions in AI‐driven education.","author":[{"family":"Alqarni","given":"Ali"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/jcal.70039","URL":"https://doi.org/10.1111/jcal.70039","source":"openalex"},{"id":"oa:W4407735835","type":"article-journal","title":"The increasing role of artificial intelligence in radiation oncology: how should we navigate it?","abstract":"Why the role of AI in radiation oncology is set to growAs illustrated by this special issue, artificial intelligence (AI) technology [1] has a potential impact on all aspects of radiation oncology, ranging from automated treatment planning [2], adaptive radiotherapy, and IGRT [3,4], to clinical decision support, personalization [5], and improvement of patients' experience and quality of life [6].Radiation oncology may be impacted by AI more significantly than any other therapeutic medical field, as radiotherapy treatment planning and delivery are already fully computer based, thus greatly facilitating the introduction of AI technology.Given the increasing role of AI in radiation oncology, it is important to discuss potential implications and the optimal strategy for our field.In particular, the potential of AI to automate virtually all steps of the treatment planning chain is a double-edged sword that brings about some of the greatest benefits but also affects the unique competency of radiation oncology.Moreover, we must not overlook the clinical aspects of radiation oncology, such as patient care, supportive therapy, and integration with pharmacological cancer treatments.These are as fundamental to the discipline as the technical aspects and represent the second essential pillar of radiation oncology.","author":[{"family":"Putz","given":"Florian"},{"family":"Fietkau","given":"Rainer"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00066-025-02381-4","URL":"https://doi.org/10.1007/s00066-025-02381-4","source":"openalex"},{"id":"oa:W4414964768","type":"article-journal","title":"Autonomous Databases and Artificial Intelligence: Architectures, Optimization, and Governance","abstract":"Artificial Intelligence-native databases are currently at the forefront of the rapidly evolving data management landscape. The book examines how database systems are changing to satisfy the needs of real-time, intelligent decision-making in different industries. The transition from traditional relational models to AI-driven architectures, cloud integration, optimization, and new developments like automation, explainability, and security are all covered in the chapters. This book's writing has involved both a thorough examination of contemporary data technology and a contemplation of the field's continuing opportunities and challenges. I want professionals, students, and anybody else interested in the future of databases to be able to understand both basic and advanced topics. I hope it encourages readers to welcome innovation and investigate the wise opportunities that lie ahead. I want to express my gratitude to my parents for their unwavering support during my journey, as well as to my peers, fellow researchers, and everyone else who has helped and inspired me. Their guidance and collaboration have been invaluable in shaping this book.","author":[{"family":"Kurapati","given":"Suresh"}],"issued":{"date-parts":[[2025]]},"DOI":"10.70593/978-93-7185-652-2","URL":"https://doi.org/10.70593/978-93-7185-652-2","source":"openalex"},{"id":"oa:W4413958379","type":"article-journal","title":"Assessment of the Artificial Intelligence– Generated Fibromyalgia Information: Beyond the Hype","abstract":"Background/Aims: Individuals increasingly turn to artificial intelligence (AI) chatbots for health-related information; however, the accuracy and usability of their responses remain uncertain. This study assessed the quality, comprehensiveness, and readability of responses from 6 AI chatbots-ChatGPT-3.5, ChatGPT-4o (OpenAI), Copilot AI (Microsoft), Perplexity AI (Perplexity.AI), Gemini AI (Google), and ChatSonic AI (Writesonic)-to the most commonly searched fibromyalgia-related queries. Materials and Methods: The top 10 most frequently searched fibromyalgia-related questions from the past 2 years were retrieved from the Google Trends database. Each chatbot was queried separately, and a total of 60 responses (10 per chatbot) were assessed both qualitatively and quantitatively by 2 reviewers, focusing on content quality, accuracy, readability, and alignment with evidence-based guidelines. Results: ChatGPT-3.5 had the lowest Ensuring Quality Information for Patients score (20.6 ± 4.5), indicating very low quality information, while Gemini achieved the highest (40.5 ± 5), which was still classified as low quality. Understandability was moderate for Copilot, Gemini, and Perplexity (67.2) but lowest for ChatGPT-3.5 (43.2 ± 10.2). Actionability was weak and the misinformation assessment revealed a moderate level across all chatbots. Readability scores indicated university-level complexity, with ChatGPT-4o having the lowest Reading Ease score (11.3 ± 11.2) and Copilot the highest (30.3 ± 13.2). Conclusion: While AI chatbots provide accessible health information, their accuracy and depth vary. Gemini, Copilot, and Perplexity AI showed better quality, but citation inconsistencies, readability challenges, and misinformation risks highlight the need for refinement beyond the hype. Clinicians should guide fibromyalgia patients in critically assessing AI-generated health content. Future research should explore improvements in AI chatbot applicability for medical inquiries.","author":[{"family":"Zure","given":"Mert"},{"family":"Menekşeoğlu","given":"Ahmet"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5152/archrheumatol.2025.11149","URL":"https://doi.org/10.5152/archrheumatol.2025.11149","source":"openalex"},{"id":"oa:W4410477980","type":"article-journal","title":"AI’s Intelligence for Improving Food Safety: Only as Strong as the Data that Feeds It","abstract":"Abstract Purpose of Review AI-based systems show great promise in advancing food safety control and surveillance systems, while they have important limitations that are often misunderstood by stakeholders. This review explores the fundamental components of AI through food safety applications, highlighting both its capabilities and constraints. Recent Findings AI has been effectively applied to outbreaks detection, allergen or spoilage issues, fraud, supply chain monitoring, traceability, quality control, shelf-life prediction, and risk mitigation. These applications integrate three key elements—sensing, reasoning, and actuating— within a structured five-step implementation cycle. Summary While AI has the potential to transform food safety, its effectiveness depends on timely access to robust, comprehensive, and unbiased data to ensure accuracy, reliability, and meaningful insights. Moving forward, it is essential to emphasize the human-in-the-loop approach. AI can support decision-making by providing actionable insights and improving efficiency, but humans must remain at the center of critical operations.","author":[{"family":"Dimitrakopoulou","given":"Maria‐eleni"},{"family":"Garre","given":"Alberto"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s43555-025-00060-0","URL":"https://doi.org/10.1007/s43555-025-00060-0","source":"openalex"},{"id":"oa:W4411067040","type":"article-journal","title":"Use of artificial intelligence tools by doctoral students: a mixed-methods explanatory-sequential investigation","abstract":"Increasing use of AI tools in higher education comes with a need to develop a clear understanding of how students (by demography and study discipline) employ them, for what purposes and in what terms their use is evaluated. This will inform the development of future guidelines and training for staff and students. In the current mixed-methods study, a cross-sectional survey was first administered to a convenience sample of 105 doctoral students (71 females, 34 males), followed by semi-structured interviews with a subset of seven participants (four females, three males). This had several aims; to identify patterns of AI tool use and its perceived helpfulness; to explore AI use with regard to aspects of time management, stress and study progress; to construct a predictive model of AI tool use and to explore, in-depth, students’ views on beneficial and problematic aspects of AI. Findings suggested AI was widely used. Males were more likely to use it for data analysis and research planning, with greater use by students in non-scientific disciplines. Use was significantly related to stress and time management. Several potential benefits and problems were identified. Benefits included aiding research, for example, assisting and improving coding/programming, use in proofreading and writing and as an explanatory tool for rendering complex information accessible. Potential problems included environmental cost, violation of intellectual property rights, provision of misleading and/or inaccurate information and risks of plagiarism and hindered creativity. Limitations of the study are discussed alongside implications for policy and training with suggestions for further work.","author":[{"family":"Akbar","given":"Muhammad"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/0309877x.2025.2515135","URL":"https://doi.org/10.1080/0309877x.2025.2515135","source":"openalex"},{"id":"oa:W4414233468","type":"article-journal","title":"The significance of artificial intelligence and machine learning in contemporary chemical engineering curriculum","abstract":"This paper discusses how artificial intelligence and machine learning can be incorporated into the new curricula of chemical engineering education as part of the transformation facing the challenges of Industry 4.0. It describes the ways in which AI tools, adaptive learning platforms, virtual laboratories, intelligent tutoring systems, and automatic assessments can significantly improve teaching effectiveness, resulting in corresponding increases in student learning outcomes. A set of indicative case studies from engineering programs is also analyzed to assess impacts based on a broad literature review. Results demonstrate that AI/ML applications provide opportunities for personalized instruction, leading to heightened conceptual understanding and accessible feedback support. Such benefits make AI/ML more preferable than traditional pedagogy, which often lacks scalability and adaptability to diverse circumstances. It also highlighted challenges related to data privacy, faculty preparedness, infrastructure limitations, and institutional resistance. The paper finally presents a set of practical recommendations for implementation, along with emerging trends, such as generative AI and learning analytics, that are shaping the future landscape of engineering education.","author":[{"family":"Ghasem","given":"Nayef"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/2331186x.2025.2560057","URL":"https://doi.org/10.1080/2331186x.2025.2560057","source":"openalex"},{"id":"oa:W4407632928","type":"article-journal","title":"Application of Artificial Intelligence in Acute Ischemic Stroke: A Scoping Review","abstract":"Artificial intelligence (AI) is revolutionizing stroke care by enhancing diagnosis, treatment, and outcome prediction. This review examines 505 original studies on AI applications in ischemic stroke, categorized into outcome prediction, stroke risk prediction, diagnosis, etiology prediction, and complication and comorbidity prediction. Outcome prediction, the most explored category, includes studies predicting functional outcomes, mortality, and recurrence, often achieving high accuracy and outperforming traditional methods. Stroke risk prediction models effectively integrate clinical and imaging data, improving assessments of both first-time and recurrent stroke risks. Diagnostic tools, such as automated imaging analysis and lesion segmentation, streamline acute stroke workflows, while AI models for large vessel occlusion detection demonstrate clinical utility. Etiology prediction focuses on identifying causes such as atrial fibrillation or cancer-associated thrombi, using imaging and thrombus analysis. Complication and comorbidity prediction models address stroke-associated pneumonia and acute kidney injury, aiding in risk stratification and resource allocation. While significant advancements have been made, challenges such as limited validation, ethical considerations, and the need for better data collection persist. This review highlights the advancements in AI applications for addressing key challenges in stroke care, demonstrating its potential to enhance precision medicine and improve patient outcomes.","author":[{"family":"Heo","given":"Joonnyung"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5469/neuroint.2025.00052","URL":"https://doi.org/10.5469/neuroint.2025.00052","source":"openalex"},{"id":"oa:W4406989203","type":"article-journal","title":"Leveraging Artificial Intelligence to Maximize Efficiency in Supply Chain Process Optimization","abstract":"Artificial Intelligence (AI) has emerged as a transformative force in supply chain management, addressing inefficiencies and complexities in dynamic global markets.The integration of AI technologies, including machine learning, predictive analytics, and autonomous systems, offers unprecedented opportunities to optimize supply chain processes across industries.From demand forecasting and inventory management to route optimization and risk mitigation, AI-driven solutions enable businesses to enhance decision-making, reduce costs, and improve operational agility.These advancements are particularly critical in a globalized economy, where supply chains are increasingly interconnected, yet vulnerable to disruptions such as geopolitical tensions, pandemics, and climaterelated challenges.AI technologies enable real-time analysis of vast datasets, facilitating predictive insights that enhance supply chain resilience.Machine learning algorithms identify patterns and anomalies, optimizing processes such as demand forecasting and procurement.Predictive analytics empower stakeholders to anticipate disruptions, mitigate risks, and adapt strategies proactively.Autonomous systems, including robotics and drones, streamline logistics operations, reducing human intervention and error.Despite its transformative potential, the implementation of AI in supply chains presents challenges, including data silos, integration complexities, and ethical concerns.To fully leverage AI, businesses must invest in robust data infrastructure, workforce upskilling, and governance frameworks that ensure responsible AI use.This paper explores the multifaceted impact of AI on supply chain optimization, highlighting case studies of successful implementation and proposing strategies to overcome barriers.By leveraging AI, organizations can build resilient, efficient, and sustainable supply chains that drive competitive advantage in an ever-evolving marketplace.","author":[{"family":"Odumbo","given":"Oluwole"},{"family":"Nimma","given":"Sani"}],"issued":{"date-parts":[[2025]]},"DOI":"10.55248/gengpi.6.0125.0508","URL":"https://doi.org/10.55248/gengpi.6.0125.0508","source":"openalex"},{"id":"oa:W4406825155","type":"article-journal","title":"Ethical Considerations Emerge from Artificial Intelligence (AI) in Biotechnology","abstract":"The integration of Artificial intelligence (AI) in biotechnology presents significant ethical challenges that must be addressed to ensure responsible innovations. Key concerns include data privacy and security, as AI systems often handle sensitive genetic and health information, necessitating robust regulations to protect individuals' rights and maintain public trust. Algorithmic bias poses another critical issue; AI can reflect existing biases in training data, leading to inequitable healthcare outcomes. Transparency in AI decision-making is essential, as \"black box\" models hinder trust, especially in drug discovery and genetics. Ethical implications of genetic manipulation require careful scrutiny to define the limits of human intervention. Additionally, societal impacts must be considered to ensure equitable distribution of AI benefits, preventing the exacerbation of disparities. Engaging diverse stakeholders, including ethicists and policymakers, is vital in aligning these technologies with societal values. Ultimately, prioritizing ethics will allow us to harness AI and biotechnology's potential while safeguarding human rights and promoting equity.","author":[{"family":"Dara","given":"Mahintaj"},{"family":"Azarpira","given":"Negar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.18502/ajmb.v17i1.17680","URL":"https://doi.org/10.18502/ajmb.v17i1.17680","source":"openalex"},{"id":"oa:W4408695783","type":"article-journal","title":"Artificial Intelligence (AI) and Emergency Medicine: Balancing Opportunities and Challenges","abstract":"Unlabelled: Artificial intelligence (AI), particularly large language models (LLMs) such as ChatGPT, has rapidly evolved and is reshaping various fields, including clinical medicine. Emergency medicine stands to benefit from AI's capacity for high-volume data processing, workflow optimization, and clinical decision support. However, important challenges exist, ranging from model \"hallucinations\" and data bias to questions of interpretability, liability, and ethical use in high-stake environments. This updated viewpoint provides a structured overview of AI's current capabilities in emergency medicine, highlights real-world applications, and explores concerns regarding regulatory requirements, safety standards, and transparency (explainable AI). We discuss the potential risks and limitations of LLMs, including their performance in rare or atypical presentations common in the emergency department and potential biases that could disproportionately affect vulnerable populations. We also address the regulatory landscape, particularly the liability for AI-driven decisions, and emphasize the need for clear guidelines and human oversight. Ultimately, AI holds enormous promise for improving patient care and resource management in emergency medicine; however, ensuring safety, fairness, and accountability remains vital.","author":[{"family":"Amiot","given":"F"},{"family":"Potier","given":"Benoit"}],"issued":{"date-parts":[[2025]]},"DOI":"10.2196/70903","URL":"https://doi.org/10.2196/70903","source":"openalex"},{"id":"oa:W4414874092","type":"article-journal","title":"Ethical Problems in the Use of Artificial Intelligence by University Educators","abstract":"This study examines the ethical problems of using artificial intelligence (AI) applications in higher education, focusing on activities performed by university educators. Drawing on Slovak legislation that defines educators’ responsibilities, the study classifies their activities into three categories: teaching, scientific research, and other (academic management and self-directed professional development). From standpoint of methodology, a thematic review of 42 open-access, peer-reviewed articles published between 2022 and 2025 was conducted across the Web of Science and Scopus databases. Relevant AI applications and their associated ethical issues were identified and thematically categorized. Results of this study show that AI applications are extensively used across all analysed areas of university educators’ activities. Most notably used are applications that are generative language models, editing and paraphrasing tools, learning and assessment software, management and search tools, visualizing and design tools, and analysis and management systems. Their adoption raises ethical concerns which can be thematically grouped into six categories: privacy and data protection, bias and fairness, transparency and accountability, autonomy and oversight, governance gaps, and integrity and plagiarism. The results provide universities with a structured analytical framework to assess and address ethical risks related to AI use in specific academic activities. Although the study is limited to open-access literature, it offers a conceptual foundation for future empirical research and the development of ethical, institutionally grounded AI policies in higher education.","author":[{"family":"Chinoracký","given":"Roman"},{"family":"Stalmašeková","given":"Natália"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/educsci15101322","URL":"https://doi.org/10.3390/educsci15101322","source":"openalex"},{"id":"oa:W4411117743","type":"article-journal","title":"Artificial intelligence-driven circRNA vaccine development: multimodal collaborative optimization and a new paradigm for biomedical applications","abstract":"Circular RNA (circRNA) vaccines have emerged as a groundbreaking innovation in infectious disease prevention and cancer immunotherapy, offering superior stability and reduced immunogenicity compared to conventional linear messenger RNA (mRNA) vaccines. While linear mRNA vaccines are prone to degradation and can trigger strong innate immune responses, covalently closed circRNA vaccines leverage their unique circular structure to enhance molecular stability and minimize innate immune activation, positioning them as a next-generation platform for vaccine development. Artificial intelligence (AI) is revolutionizing circRNA vaccine design and optimization. Deep learning models, such as convolutional neural networks (CNNs) and Transformers, integrate multi-omics data to refine antigen prediction, RNA secondary structure modeling, and lipid nanoparticle delivery system formulation, surpassing traditional bioinformatics approaches in both accuracy and efficiency. While AI-driven bioinformatics enhances antigen screening and delivery system modeling, generative AI accelerates literature synthesis and experimental planning-though the risk of fabricated references and limited biological interpretability hinders its reliability. Despite these advancements, challenges such as the \"black-box\" nature of AI algorithms, unreliable literature retrieval, and insufficient integration of biological mechanisms underscore the necessity for a hybrid \"AI-traditional-experimental\" paradigm. This approach integrates explainable AI frameworks, multi-omics validation, and ethical oversight to ensure clinical translatability. Future research should prioritize mechanism-driven AI models, real-time experimental feedback, and rigorous ethical standards to fully unlock the potential of circRNA vaccines in precision oncology and global health.","author":[{"family":"Zhao","given":"Yan"},{"family":"Wang","given":"Huaiyu"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1093/bib/bbaf263","URL":"https://doi.org/10.1093/bib/bbaf263","source":"openalex"},{"id":"oa:W4414707963","type":"article-journal","title":"Integrating artificial intelligence into small molecule development for precision cancer immunomodulation therapy","abstract":"This perspective examines how artificial intelligence (AI) is transforming small-molecule development for precision cancer immunomodulation therapy. It outlines AI-driven approaches for de novo design, virtual screening, multi-parameter optimization, and ADMET prediction, targeting immune checkpoints, tumor microenvironment modulation, antigen presentation, and metabolic pathways. The article highlights patient stratification, multi-omics integration, digital twin simulations, translational challenges, and future directions, underscoring AI's potential to deliver effective, personalized immunomodulatory therapeutics.","author":[{"family":"Sutanto","given":"Henry"},{"family":"Fetarayani","given":"Deasy"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s44386-025-00029-y","URL":"https://doi.org/10.1038/s44386-025-00029-y","source":"openalex"},{"id":"oa:W4410126990","type":"article-journal","title":"Interactive Heritage: The Role of Artificial Intelligence in Digital Museums","abstract":"Museum use of artificial intelligence (AI) is becoming increasingly common, but its contribution to museum attendance is yet to be confirmed. This paper investigates whether the adoption of AI impacts museum visitation using data from 19 museums. Statistical analyses, including ANOVA and Spearman correlation, were conducted to determine if the use of AI has significant effects on visitors. The findings indicate no statistically significant difference between museums that use AI and those that do not (ANOVA: p = 0.263, F = 1.34), but the Spearman correlation (r = 0.448, p = 0.055) indicates a moderate positive correlation that is not statistically significant. The findings suggest that AI enhances visitor experience rather than increasing attendance. Additionally, this study proposes a conceptual framework for AI prototyping in museums. The study contributes to the ongoing debate on AI in cultural institutions by emphasizing that future research should incorporate longitudinal studies and qualitative visitor feedback in order to capture the overall impact of AI on engagement and sustainability in museums.","author":[{"family":"Κιουρεξίδου","given":"Ματίνα"},{"family":"Stamou","given":"Sofia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/electronics14091884","URL":"https://doi.org/10.3390/electronics14091884","source":"openalex"},{"id":"oa:W4410770993","type":"article-journal","title":"Perspectives of physicians, nurses, and patients on the use of artificial intelligence and robotic nurses in healthcare","abstract":"AIM: This study aims to assess the perspectives of physicians, nurses, and patients in Turkey regarding the integration of artificial intelligence (AI) and robotic nurses in healthcare settings while exploring their attitudes toward the use of robots in healthcare delivery. BACKGROUND: AI and robotic nurses are increasingly shaping healthcare delivery and influencing clinical decision-making processes. However, research examining the impact of AI and robotic nurses on nursing practice and patient care remains limited. The attitudes of healthcare professionals and patients are crucial factors for the successful integration and adoption of these technologies in clinical settings. METHOD: This qualitative study employed in-depth individual interviews to explore participants' perspectives. The sample consisted of 13 physicians, 17 nurses, and 15 patients, all recruited from university hospitals, Ministry of Health hospitals, and private healthcare facilities across Turkey. Data were collected using two semistructured interview guides with \"Healthcare Workersand Patients\". Ethical approval and informed consent were obtained prior to data collection. The collected data were analyzed with content analysis using MAXQDA Pro 2021 software. RESULTS: The qualitative findings were organized into four primary themes: \"Impact of AI Technologies on Healthcare,\" \"Use of AI and Robots,\" \"Receiving Care from Humanoid Robots,\" and \"Working with Humanoid Robots.\" These themes were further explored through 12 subthemes and corresponding codes. DISCUSSION: Participants who had not yet interacted directly with AI technologies and relied solely on literature or had limited knowledge about the process generally believed that AI and robotic nurses could positively affect healthcare. However, they expressed concerns regarding the inability of these technologies to replicate the human touch. They specifically highlighted the limitations of robots in areas requiring empathy, emotional connection, and personalized care. CONCLUSION AND IMPLICATIONS FOR NURSING AND/OR HEALTH POLICY: This study explored the perspectives of physicians, nurses, and patients regarding AI and robotic nurses, revealing both anticipated benefits and potential challenges for the healthcare system. The findings indicate the necessity for policies that address the ethical, social, and professional implications of AI and robotics, ensuring that their integration into healthcare practices is aligned with nursing and health policy objectives. It is crucial for international healthcare leaders to collaborate in developing policies that optimize the benefits of these technologies across diverse healthcare settings.","author":[{"family":"Gümüş","given":"Emel"},{"family":"Alan","given":"Handan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/inr.70017","URL":"https://doi.org/10.1111/inr.70017","source":"openalex"},{"id":"oa:W4406772561","type":"article-journal","title":"A Comprehensive and Systematic Review of Multi-Criteria Decision-Making (MCDM) Methods to Solve Decision-Making Problems: Two Decades from 2004 to 2024","abstract":"Decision-making in complex, multifaceted scenarios has become increasingly critical across diverse sectors, necessitating robust frameworks like Multi-Criteria Decision-Making (MCDM). Over the past two decades (2004–2024), MCDM has transformed from foundational methods like AHP and TOPSIS into dynamic hybrid models integrating artificial intelligence, fuzzy logic, and machine learning. Despite significant strides, the field faces challenges in addressing geographic disparities, underexplored domains, and adapting to emerging global needs. This study provides a comprehensive review of MCDM's evolution, consolidating insights from 3,655 peer-reviewed articles sourced through Dimensions.ai and analyzed using bibliometric tools like VOSviewer. The research identifies publication trends, leading contributors, thematic clusters, and collaborative networks while pinpointing gaps and opportunities for future exploration. These key findings highlight exponential growth in MCDM applications, particularly in sustainable energy, urban planning, and healthcare optimization. These advancements align with global priorities, including the United Nations Sustainable Development Goals (SDGs) such as clean energy, climate action, and sustainable cities. However, critical gaps remain in addressing issues like poverty alleviation, gender equity, and biodiversity conservation, emphasizing the need for broader interdisciplinary applications. This review concludes that MCDM's potential lies in embracing inclusivity, advancing into emerging technologies like blockchain and the metaverse, and fostering collaboration across underrepresented regions and domains. By harnessing real-time data, immersive simulations, and secure decision-making platforms, MCDM can redefine how global challenges are addressed.","author":[{"family":"Kumar","given":"Rahul"},{"family":"Pamucar","given":"Dragan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.31181/sdmap21202524","URL":"https://doi.org/10.31181/sdmap21202524","source":"openalex"},{"id":"oa:W4411806359","type":"article-journal","title":"Artificial Intelligence and the Future of Mental Health in a Digitally Transformed World","abstract":"Artificial Intelligence (AI) is reshaping mental healthcare by enabling new forms of diagnosis, therapy, and patient monitoring. Yet this digital transformation raises complex policy and ethical questions that remain insufficiently addressed. In this paper, we critically examine how AI-driven innovations are being integrated into mental health systems across different global contexts, with particular attention to governance, regulation, and social justice. The study follows the PRISMA-ScR methodology to ensure transparency and methodological rigor, while also acknowledging its inherent limitations, such as the emphasis on breadth over depth and the exclusion of non-English sources. Drawing on international guidelines, academic literature, and emerging national strategies, it identifies both opportunities, such as improved access and personalized care, and threats, including algorithmic bias, data privacy risks, and diminished human oversight. Special attention is given to underrepresented populations and the risks of digital exclusion. The paper argues for a value-driven approach that centers equity, transparency, and informed consent in the deployment of AI tools. It concludes with actionable policy recommendations to support the ethical implementation of AI in mental health, emphasizing the need for cross-sectoral collaboration and global accountability mechanisms.","author":[{"family":"Fanarioti","given":"Aggeliki"},{"family":"Karpouzis","given":"Kostas"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/computers14070259","URL":"https://doi.org/10.3390/computers14070259","source":"openalex"},{"id":"oa:W4410037446","type":"article-journal","title":"Role of artificial intelligence in early identification and risk evaluation of non-communicable diseases: a bibliometric analysis of global research trends","abstract":"OBJECTIVE: This study aims to shed light on the transformative potential of artificial intelligence (AI) in the early detection and risk assessment of non-communicable diseases (NCDs). STUDY DESIGN: Bibliometric analysis. SETTING: Articles related to AI in early identification and risk evaluation of NCDs from 2000 to 2024 were retrieved from the Scopus database. METHODS: This comprehensive bibliometric study focuses on a single database, Scopus and employs narrative synthesis for concise yet informative summaries. Microsoft Excel V.365 and VOSviewer software (V.1.6.20) were used to summarise bibliometric features. RESULTS: The study retrieved 1745 relevant articles, with a notable surge in research activity in recent years. Core journals included Scientific Reports and IEEE Access, and core institutions included the Harvard Medical School and the Ministry of Education of the People's Republic of China, while core countries comprised China, the USA, India, the UK and Saudi Arabia. Citation trends indicated substantial growth and recognition of AI's impact on NCDs management. Frequent author keywords identified key research hotspots, including specific NCDs like Alzheimer's disease and diabetes. Risk assessment studies demonstrated improved predictions for heart failure, cardiovascular risk, breast cancer, diabetes and inflammatory bowel disease. CONCLUSION: Our findings highlight the increasing role of AI in early detection and risk prediction of NCDs, emphasising its widening research impact and future clinical potential.","author":[{"family":"Al-Dekah","given":"Arwa"},{"family":"Sweileh","given":"Waleed"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1136/bmjopen-2025-101169","URL":"https://doi.org/10.1136/bmjopen-2025-101169","source":"openalex"},{"id":"oa:W4416746586","type":"article-journal","title":"A review of artificial intelligence for predicting climate driven infectious disease outbreaks to enhance global health resilience","abstract":"The impact of climate change on infectious disease outbreaks demands sophisticated techniques of prediction to protect global health. This review focuses on the impact of artificial intelligence on the prediction of disease outbreaks influenced by climatic factors, showcasing its potential on diverse data sets. While traditional forecasting models have restricted capabilities due to fixed parameters and linear relationships, machine learning models such as support vector machine and random forest, deep learning models such convolutional neural network, long-short term memory and transformers as well as the hybrid models are found to be much more efficient with their supremacy proven over traditional models through applications in vector-borne, water-borne, and zoonotic diseases through capturing real-time analytics and non-linear climate-disease interaction. Despite these breakthroughs, the challenges of sparsity of data in low-resource areas, lack of model transparency, and ethically deemed biased algorithms pose great challenges. The review recommends the necessity of data governance, explanatory frameworks, and cross-disciplinary collaboration to counter these constraints, and proposes the utilization of federated learning, with quantum and edge computing, to formulate global health resilience pathways. Therefore, researchers are encouraged to adopt artificial intelligence techniques in predictive modeling, owing to their potential to transform proactive public health policy; however, its success is contingent on socially equitable implementation, trust from the relevant stakeholders, and climate policy integration.","author":[{"family":"Inam","given":"Syed"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12982-025-01167-4","URL":"https://doi.org/10.1186/s12982-025-01167-4","source":"openalex"},{"id":"oa:W4409092927","type":"article-journal","title":"Trust in artificial intelligence: a survey experiment to assess trust in algorithmic decision-making","abstract":"Abstract Artificial intelligence (AI) has seen rapid development over the past decade, leading to its integration into various aspects of human life. The ability to integrate AI systems hinges not solely on their technical efficacy but also on the perceptions held by users or decision-makers. Previous researches indicate that many people harbor concerns about AI, which can hinder the adoption of these technologies. This study uses a pre-registered survey experiment embedded in an online survey in Hungary ( N = 2100) to assess trust in AI-based Automated Decision-Making (ADM). Participants were presented with hypothetical decisions in four domains (medical diagnoses, hiring, transportation, and financial investments). In a split-ballot design, participants were randomly assigned to a control group with human involvement and an experimental group where decision were supported by AI-based ADM. The main results show that decisions supported by human intervention are perceived as more trustworthy than those made by ADM (except for financial investment). However, our treatment heterogeneity analysis indicates that these effects are not consistent across all segments of society. A good understanding of AI, low privacy concerns, and an open personality can mitigate the negative impact of AI assistance on trust.","author":[{"family":"Orbán","given":"Ferenc"},{"family":"Stefkovics","given":"Ádám"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s00146-025-02237-6","URL":"https://doi.org/10.1007/s00146-025-02237-6","source":"openalex"},{"id":"doi:10.3390/bioengineering12040375","type":"article-journal","title":"What Is the Role of Explainability in Medical Artificial Intelligence? A Case-Based Approach.","abstract":"This article reflects on explainability in the context of medical artificial intelligence (AI) applications, focusing on AI-based clinical decision support systems (CDSS). After introducing the concept of explainability in AI and providing a short overview of AI-based clinical decision support systems (CDSSs) and the role of explainability in CDSSs, four use cases of AI-based CDSSs will be presented. The examples were chosen to highlight different types of AI-based CDSSs as well as different types of explanations: a machine language (ML) tool that lacks explainability; an approach with post hoc explanations; a hybrid model that provides medical knowledge-based explanations; and a causal model that involves complex moral concepts. Then, the role, relevance, and implications of explainability in the context of the use cases will be discussed, focusing on seven explainability-related aspects and themes. These are: (1) The addressees of explainability in medical AI; (2) the relevance of explainability for medical decision making; (3) the type of explanation provided; (4) the (often-cited) conflict between explainability and accuracy; (5) epistemic authority and automation bias; (6) Individual preferences and values; (7) patient autonomy and doctor-patient relationships. The case-based discussion reveals that the role and relevance of explainability in AI-based CDSSs varies considerably depending on the tool and use context. While it is plausible to assume that explainability in medical AI has positive implications, empirical data on explainability and explainability-related implications is scarce. Use-case-based studies are needed to investigate not only the technical aspects of explainability but also the perspectives of clinicians and patients on the relevance of explainability and its implications.","author":[{"family":"Hildt","given":"Elisabeth"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/bioengineering12040375","URL":"https://doi.org/10.3390/bioengineering12040375","source":"europepmc"},{"id":"doi:10.5281/zenodo.17419282","type":"article-journal","title":"Detekcija Temporalnog Rizika kod Nesitnostaničnog Karcinoma Pluća: Fraktalno-UTL Okvir Temeljen na EWMA","abstract":"🧠 Sažetak Predstavljamo okvir za detekciju temporalnog rizika koji kombinira eksponencijalno ponderiranu pomičnu sredinu (EWMA) akumulacije rizika s fraktalnom karakterizacijom tumora, s ciljem precizne prognoze ishoda u stvarnom vremenu kod nesitnostaničnog karcinoma pluća (NSCLC). Za razliku od statičkih radiomičkih klasifikatora koji neovisno evaluiraju trenutne snimke, Fraktalno-UTL pristup akumulira rizik kroz longitudinalno oslikavanje i kvantificira promjene između uzastopnih pregleda, čime prepoznaje tihi, postupni porast heterogenosti koji prethodi klinički manifestnoj progresiji. 🔬 Teorijski doprinos Integracijom analize fraktalne dimenzije ($D_1$–$D_3$) i EWMA modeliranja volatilnosti, formalizirana je veza između prostorne heterogenosti tumora i njezine vremenske dinamike. Dano je matematičko zatvaranje izraza za očekivano vrijeme ranog upozorenja u prisutnosti skoka varijance te dokazano monotono smanjenje rizika nakon uspješne intervencije. Time se uspostavlja veza između fraktalne teksture tumora i hazardne funkcije kroz vremensku domenu. 📊 Empirijska validacija Na NSCLC-Radiomics skupu podataka (TCIA Lung1, n=422 pacijenta), okvir postiže: C-indeks = 0.692 (u odnosu na Aerts 2014: 0.65) HR = 2.41 (visok vs. nizak rizik, log-rank p < 0.001) 99% manje značajki (5 vs. 440) 5× niža latencija zaključivanja (47 ms vs. 230 ms) Dodatno, validacija na trima vrstama karcinoma (NSCLC, gušterača, CRLM) potvrđuje robustnost modela s konzistentnim optimalnim parametrima ($\\alpha \\approx 0.15$, $\\theta_{\\text{mult}} \\approx 1.5$). ⚠️ Metodološko ograničenje TCIA Lung1 skup sadrži samo pojedinačne CT snimke po pacijentu — ne autentične longitudinalne serije. Temporalne značajke su rekonstruirane putem proxy varijable: vtproxy=Stadij⋅D3(t)⋅N(1,0.1)v_t^{\\text{proxy}} = \\text{Stadij} \\cdot D_3(t) \\cdot \\mathcal{N}(1, 0.1)vtproxy=Stadij⋅D3(t)⋅N(1,0.1) Retrospektivna projekcija sugerira potencijalno rano upozorenje od 3.2 mjeseca (IQR: 1.8–5.7) prije kliničke progresije, ali ova tvrdnja zahtijeva verifikaciju u prospektivnoj studiji s pravim longitudinalnim CT podacima (n ≥ 100, ≥ 3 vremenske točke; planirana 2026.). 🧮 Zaključak Fraktalno-UTL okvir omogućuje: modeliranje dinamičke heterogenosti tumora u vremenu, predviđanje progresije uz višestruko manju kompleksnost modela, i stvaranje temelja za rano upozorenje u onkološkoj praksi temeljeno na hazardnoj dinamici. Okvir predstavlja korak prema integriranju fraktalne geometrije i hazardne analize u kliničku radiomiku te otvara mogućnost personaliziranog praćenja rizika kroz real-time onkološki nadzor. 📎 Povezani resursi GitHub repozitorij: https://github.com/bsabljic/utl-oncology Dopunski materijal: Detekcija_Temporalnog_Rizika_NSCLC_UTL_Supplementary.pdf ORCID: 0009-0005-6199-4488 ⚖️ Licenca Ovo djelo je objavljeno pod licencomCreative Commons Attribution 4.0 International (CC BY 4.0)Slobodno dijeliti i koristiti uz navođenje izvora. 🔖 Citiranje Branimir Sabljić (2025).Detekcija Temporalnog Rizika kod Nesitnostaničnog Karcinoma Pluća: Fraktalno-UTL Okvir Temeljen na EWMA.","author":[{"family":"Sabljic","given":"Branimir"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.17419282","URL":"https://doi.org/10.5281/zenodo.17419282","source":"datacite"},{"id":"doi:10.5281/zenodo.17419283","type":"article-journal","title":"Detekcija Temporalnog Rizika kod Nesitnostaničnog Karcinoma Pluća: Fraktalno-UTL Okvir Temeljen na EWMA","abstract":"🧠 Sažetak Predstavljamo okvir za detekciju temporalnog rizika koji kombinira eksponencijalno ponderiranu pomičnu sredinu (EWMA) akumulacije rizika s fraktalnom karakterizacijom tumora, s ciljem precizne prognoze ishoda u stvarnom vremenu kod nesitnostaničnog karcinoma pluća (NSCLC). Za razliku od statičkih radiomičkih klasifikatora koji neovisno evaluiraju trenutne snimke, Fraktalno-UTL pristup akumulira rizik kroz longitudinalno oslikavanje i kvantificira promjene između uzastopnih pregleda, čime prepoznaje tihi, postupni porast heterogenosti koji prethodi klinički manifestnoj progresiji. 🔬 Teorijski doprinos Integracijom analize fraktalne dimenzije ($D_1$–$D_3$) i EWMA modeliranja volatilnosti, formalizirana je veza između prostorne heterogenosti tumora i njezine vremenske dinamike. Dano je matematičko zatvaranje izraza za očekivano vrijeme ranog upozorenja u prisutnosti skoka varijance te dokazano monotono smanjenje rizika nakon uspješne intervencije. Time se uspostavlja veza između fraktalne teksture tumora i hazardne funkcije kroz vremensku domenu. 📊 Empirijska validacija Na NSCLC-Radiomics skupu podataka (TCIA Lung1, n=422 pacijenta), okvir postiže: C-indeks = 0.692 (u odnosu na Aerts 2014: 0.65) HR = 2.41 (visok vs. nizak rizik, log-rank p < 0.001) 99% manje značajki (5 vs. 440) 5× niža latencija zaključivanja (47 ms vs. 230 ms) Dodatno, validacija na trima vrstama karcinoma (NSCLC, gušterača, CRLM) potvrđuje robustnost modela s konzistentnim optimalnim parametrima ($\\alpha \\approx 0.15$, $\\theta_{\\text{mult}} \\approx 1.5$). ⚠️ Metodološko ograničenje TCIA Lung1 skup sadrži samo pojedinačne CT snimke po pacijentu — ne autentične longitudinalne serije. Temporalne značajke su rekonstruirane putem proxy varijable: vtproxy=Stadij⋅D3(t)⋅N(1,0.1)v_t^{\\text{proxy}} = \\text{Stadij} \\cdot D_3(t) \\cdot \\mathcal{N}(1, 0.1)vtproxy=Stadij⋅D3(t)⋅N(1,0.1) Retrospektivna projekcija sugerira potencijalno rano upozorenje od 3.2 mjeseca (IQR: 1.8–5.7) prije kliničke progresije, ali ova tvrdnja zahtijeva verifikaciju u prospektivnoj studiji s pravim longitudinalnim CT podacima (n ≥ 100, ≥ 3 vremenske točke; planirana 2026.). 🧮 Zaključak Fraktalno-UTL okvir omogućuje: modeliranje dinamičke heterogenosti tumora u vremenu, predviđanje progresije uz višestruko manju kompleksnost modela, i stvaranje temelja za rano upozorenje u onkološkoj praksi temeljeno na hazardnoj dinamici. Okvir predstavlja korak prema integriranju fraktalne geometrije i hazardne analize u kliničku radiomiku te otvara mogućnost personaliziranog praćenja rizika kroz real-time onkološki nadzor. 📎 Povezani resursi GitHub repozitorij: https://github.com/bsabljic/utl-oncology Dopunski materijal: Detekcija_Temporalnog_Rizika_NSCLC_UTL_Supplementary.pdf ORCID: 0009-0005-6199-4488 ⚖️ Licenca Ovo djelo je objavljeno pod licencomCreative Commons Attribution 4.0 International (CC BY 4.0)Slobodno dijeliti i koristiti uz navođenje izvora. 🔖 Citiranje Branimir Sabljić (2025).Detekcija Temporalnog Rizika kod Nesitnostaničnog Karcinoma Pluća: Fraktalno-UTL Okvir Temeljen na EWMA.","author":[{"family":"Sabljic","given":"Branimir"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.17419283","URL":"https://doi.org/10.5281/zenodo.17419283","source":"datacite"},{"id":"doi:10.5281/zenodo.17367661","type":"article-journal","title":"on consciousness and perception of reality","abstract":"This study aims to re-evaluate the Boltzmann Brain paradox not merely as a cosmological and statistical problem, but also from a psychophysical perspective. It demonstrates striking similarities between phenomena such as hypnosis, false memory syndrome, dissociative identity disorder, and hallucinations and the false memory clusters predicted by the Boltzmann Brain scenario. The study will discuss how the study examines the relativistic entropy experience observed in individual consciousness can be explained at the macro-universe level within the framework of quantum information theory. In this context, it is argued that the wave–particle duality in the principle of complementarity, rather than being a set of mutually complementary opposites, indicates that, from an information-theoretic perspective, at the most fundamental level, matter and information are two different manifestations of the same phenomenon. The study also aims to provide cognitive relief by resolving the mental nightmare created by the infinite universe scenario in cosmology; in the process, it corroborates certain aspects of the Boltzmann Brain and demonstrates that human memory and consciousness can be explained within the framework of chaotic determinism. This research aims to contribute particularly to psychophysics and to develop applications for artificial intelligence, while also contributing to the physical and mathematical foundations of this discipline. In this context, we examine possible determinism through the blending of the Boltzmann Brain with chaos theory, focusing on theories of timelessness, such as the block universe model of the nature of time and the Wheeler–DeWitt equation. The central theme is why human behavior cannot be predicted by mathematical equations, raising crucial questions such as 'Are we living in a state of collective hypnosis?' and 'Are we inside a controlled hallucination?' Furthermore, when viewed from the perspective of modern physics' current theories, it is emphasized that it makes more sense to evaluate our reality not as an illusion, but as a flawed reality. The book examines how artificial intelligence systems produce hallucinations and proposes solutions, explaining how they can be more empathetic within the framework of relative entropy. It also explores personalized treatment opportunities and provides new perspectives for integrating hypnosis into medical practice via neuromodulation–based clinical protocols and Brain–Computer Interface (BCI) applications. Furthermore, various studies ranging from the Penfield and Libet experiments to the “mouse utopia” experiment, infantile amnesia, sensory deprivation experiments, hypnosis experiments and the effects of psychotic processes on perception and creative cognition will also be examined. In this context, the study aims to offer a solution to the philosophical paradox created by the Boltzmann Brain; to contribute to the field of psychophysics through a holistic approach; to shed light on the question 'What is consciousness?' at the intersection of science and philosophy. Keywords: Boltzmann brain, false memory syndrome, entropy, cognitive science, Neuromodulation, consciousness, arrow of time, chaos theory, Artificial Intelligence (AI), Statistical mechanics, thermodynamics, Predictive processing, Shannon relativity entropy, quantum mechanics, cognitive psychology, clinical neuroscience, Complex Systems, Psychophysics, Social Psychology, Forensic psychology","author":[{"family":"Korkutata"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.17367661","URL":"https://doi.org/10.5281/zenodo.17367661","source":"datacite"},{"id":"doi:10.5281/zenodo.22020298","type":"article-journal","title":"on consciousness and perception of reality","abstract":"This study aims to re-evaluate the Boltzmann Brain paradox not merely as a cosmological and statistical problem, but also from a psychophysical perspective. It demonstrates striking similarities between phenomena such as hypnosis, false memory syndrome, dissociative identity disorder, and hallucinations and the false memory clusters predicted by the Boltzmann Brain scenario. The study will discuss how the study examines the relativistic entropy experience observed in individual consciousness can be explained at the macro-universe level within the framework of quantum information theory. In this context, it is argued that the wave–particle duality in the principle of complementarity, rather than being a set of mutually complementary opposites, indicates that, from an information-theoretic perspective, at the most fundamental level, matter and information are two different manifestations of the same phenomenon. The study also aims to provide cognitive relief by resolving the mental nightmare created by the infinite universe scenario in cosmology; in the process, it corroborates certain aspects of the Boltzmann Brain and demonstrates that human memory and consciousness can be explained within the framework of chaotic determinism. This research aims to contribute particularly to psychophysics and to develop applications for artificial intelligence, while also contributing to the physical and mathematical foundations of this discipline. In this context, we examine possible determinism through the blending of the Boltzmann Brain with chaos theory, focusing on theories of timelessness, such as the block universe model of the nature of time and the Wheeler–DeWitt equation. The central theme is why human behavior cannot be predicted by mathematical equations, raising crucial questions such as 'Are we living in a state of collective hypnosis?' and 'Are we inside a controlled hallucination?' Furthermore, when viewed from the perspective of modern physics' current theories, it is emphasized that it makes more sense to evaluate our reality not as an illusion, but as a flawed reality. The book examines how artificial intelligence systems produce hallucinations and proposes solutions, explaining how they can be more empathetic within the framework of relative entropy. It also explores personalized treatment opportunities and provides new perspectives for integrating hypnosis into medical practice via neuromodulation–based clinical protocols and Brain–Computer Interface (BCI) applications. Furthermore, various studies ranging from the Penfield and Libet experiments to the “mouse utopia” experiment, infantile amnesia, sensory deprivation experiments, hypnosis experiments and the effects of psychotic processes on perception and creative cognition will also be examined. In this context, the study aims to offer a solution to the philosophical paradox created by the Boltzmann Brain; to contribute to the field of psychophysics through a holistic approach; to shed light on the question 'What is consciousness?' at the intersection of science and philosophy. Keywords: Boltzmann brain, false memory syndrome, entropy, cognitive science, Neuromodulation, consciousness, arrow of time, chaos theory, Artificial Intelligence (AI), Statistical mechanics, thermodynamics, Predictive processing, Shannon relativity entropy, quantum mechanics, cognitive psychology, clinical neuroscience, Complex Systems, Psychophysics, Social Psychology, Forensic psychology","author":[{"family":"Korkutata"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22020298","URL":"https://doi.org/10.5281/zenodo.22020298","source":"datacite"},{"id":"doi:10.5281/zenodo.22020099","type":"article-journal","title":"Dataset: Hypothesis: Intraperitoneal administration of a thermoresponsive hydrogel loaded with borneol-functionalized ginger  derived extracellular vesicles (Moxibustion-Modified GDEVs) may provide a sustained, localized delivery to ovarian tumors, potentially enhancing deep tissue penetration and inducing apoptosis without systemic chemotoxicity. - PathMap Experiment #000134","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=134 Artificial General Intelligence LLC Claim Evaluated: Hypothesis: Intraperitoneal administration of a thermoresponsive hydrogel loaded with borneol-functionalized ginger derived extracellular vesicles (Moxibustion-Modified GDEVs) may provide a sustained, localized delivery to ovarian tumors, potentially enhancing deep tissue penetration and inducing apoptosis without systemic chemotoxicity. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights GDEVs retain therapeutic anti-inflammatory properties, making them candidates for modulating the tumor microenvironment. Thermoresponsive hydrogels facilitate controlled delivery, potentially reducing systemic exposure. The use of native plant-derived materials offers a scalable alternative to synthetic nanocarriers. Preclinical evidence in diverse cancer models supports the use of membrane-camouflaged nanoparticles for improved tumor-targeting specificity. Combination therapies using natural products and chemotherapeutics often overcome the resistance mechanisms associated with conventional platinum-based treatments. The tumor microenvironment (TME) is a critical determinant of drug delivery efficiency, where mechanical barriers and fluid pressure significantly affect intratumoral distribution. Current research is shifting towards \"biomimetic conductive cardiac patch\" and similar adaptive materials, which underscores the maturity of hydrogel engineering for diverse tissue-specific applications. The use of \"small interfering RNA (siRNA)\" within nanovesicles validates the capacity to carry both chemical and genetic payloads for dual-mode therapy. Thermal processing (boiling) reconfigures ginger extracellular vesicles (GEVs) into thermally reassembled GEVs (T-GEVs) with enhanced trafficking regulator enrichment. T-GEVs demonstrate an 8.57-fold increase in clathrin-dependent cellular uptake in intestinal cells compared to native vesicles. Carrier-free pure drug crystal depots can provide sustained ocular delivery for at least eight months. Borneol-functionalized nanoparticles efficiently traverse the blood-brain barrier and restore redox homeostasis in cerebral ischemic models. Systematic identification of host genes essential for bacterial invasion provides a robust pipeline for novel therapeutic target discovery. Tumor-derived parathyroid hormone-related protein (PTHrP) is associated with the suppression of multiple cytochrome P450 enzyme families, impacting chemotherapy pharmacokinetics. Synergistic effects of NMN supplementation enhance MSLN CAR-NK cell persistence and cytotoxic potency against ovarian cancer. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Assess the stability and drug-loading efficiency of GDEVs conjugated with borneol. Evaluate the release kinetics of borneol-GDEVs from a thermoresponsive hydrogel at varying temperature thresholds. Investigate the intraperitoneal tumor accumulation and penetration depth of the hydrogel-loaded GDEVs in an orthotopic ovarian cancer mouse model. Develop T-GEVs loaded with paclitaxel for intraperitoneal delivery in SKOV3 xenograft models. Evaluate the stability and degradation profile of thermosensitive PEOz-PAla hydrogels loaded with borneol-GDEVs in peritoneal fluid. Assess the synergistic effect of borneol-GDEVs with cisplatin in 3D ovarian cancer spheroid models. 📊 Suggested Studies A comparative study evaluating the therapeutic efficacy of borneol-GDEVs versus non-functionalized GDEVs in ovarian cancer cell models. Long-term toxicity and biodistribution assessment of intraperitoneally administered thermoresponsive hydrogel-GDEV systems in healthy subjects. Metagenomic and transcriptomic analysis ","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22020099","URL":"https://doi.org/10.5281/zenodo.22020099","source":"datacite"},{"id":"doi:10.5281/zenodo.22020100","type":"article-journal","title":"Dataset: Hypothesis: Intraperitoneal administration of a thermoresponsive hydrogel loaded with borneol-functionalized ginger  derived extracellular vesicles (Moxibustion-Modified GDEVs) may provide a sustained, localized delivery to ovarian tumors, potentially enhancing deep tissue penetration and inducing apoptosis without systemic chemotoxicity. - PathMap Experiment #000134","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=134 Artificial General Intelligence LLC Claim Evaluated: Hypothesis: Intraperitoneal administration of a thermoresponsive hydrogel loaded with borneol-functionalized ginger derived extracellular vesicles (Moxibustion-Modified GDEVs) may provide a sustained, localized delivery to ovarian tumors, potentially enhancing deep tissue penetration and inducing apoptosis without systemic chemotoxicity. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights GDEVs retain therapeutic anti-inflammatory properties, making them candidates for modulating the tumor microenvironment. Thermoresponsive hydrogels facilitate controlled delivery, potentially reducing systemic exposure. The use of native plant-derived materials offers a scalable alternative to synthetic nanocarriers. Preclinical evidence in diverse cancer models supports the use of membrane-camouflaged nanoparticles for improved tumor-targeting specificity. Combination therapies using natural products and chemotherapeutics often overcome the resistance mechanisms associated with conventional platinum-based treatments. The tumor microenvironment (TME) is a critical determinant of drug delivery efficiency, where mechanical barriers and fluid pressure significantly affect intratumoral distribution. Current research is shifting towards \"biomimetic conductive cardiac patch\" and similar adaptive materials, which underscores the maturity of hydrogel engineering for diverse tissue-specific applications. The use of \"small interfering RNA (siRNA)\" within nanovesicles validates the capacity to carry both chemical and genetic payloads for dual-mode therapy. Thermal processing (boiling) reconfigures ginger extracellular vesicles (GEVs) into thermally reassembled GEVs (T-GEVs) with enhanced trafficking regulator enrichment. T-GEVs demonstrate an 8.57-fold increase in clathrin-dependent cellular uptake in intestinal cells compared to native vesicles. Carrier-free pure drug crystal depots can provide sustained ocular delivery for at least eight months. Borneol-functionalized nanoparticles efficiently traverse the blood-brain barrier and restore redox homeostasis in cerebral ischemic models. Systematic identification of host genes essential for bacterial invasion provides a robust pipeline for novel therapeutic target discovery. Tumor-derived parathyroid hormone-related protein (PTHrP) is associated with the suppression of multiple cytochrome P450 enzyme families, impacting chemotherapy pharmacokinetics. Synergistic effects of NMN supplementation enhance MSLN CAR-NK cell persistence and cytotoxic potency against ovarian cancer. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Assess the stability and drug-loading efficiency of GDEVs conjugated with borneol. Evaluate the release kinetics of borneol-GDEVs from a thermoresponsive hydrogel at varying temperature thresholds. Investigate the intraperitoneal tumor accumulation and penetration depth of the hydrogel-loaded GDEVs in an orthotopic ovarian cancer mouse model. Develop T-GEVs loaded with paclitaxel for intraperitoneal delivery in SKOV3 xenograft models. Evaluate the stability and degradation profile of thermosensitive PEOz-PAla hydrogels loaded with borneol-GDEVs in peritoneal fluid. Assess the synergistic effect of borneol-GDEVs with cisplatin in 3D ovarian cancer spheroid models. 📊 Suggested Studies A comparative study evaluating the therapeutic efficacy of borneol-GDEVs versus non-functionalized GDEVs in ovarian cancer cell models. Long-term toxicity and biodistribution assessment of intraperitoneally administered thermoresponsive hydrogel-GDEV systems in healthy subjects. Metagenomic and transcriptomic analysis ","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22020100","URL":"https://doi.org/10.5281/zenodo.22020100","source":"datacite"},{"id":"doi:10.5281/zenodo.21265318","type":"article-journal","title":"PathMap Experiment #000023 - Tags: #TDP-43 pathology #RNA Splicing #Stathmin 2 #Vitreous Body #DNA-Binding Protein-43","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=23 Artificial General Intelligence LLC Claim Evaluated: Does misfolded TDP43 affect Retinal Ganglion Cell STMN2 ability to repair in a similar way as what how it was shown to affect motor neurons with cryptic mis-splicing? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Perform single-nuclei RNA sequencing (snRNA-seq) on retinal ganglion cells from TDP-43 mutant mouse models to assess STMN2 splicing profiles. Evaluate axonal regeneration capacity of RGCs derived from human iPSCs with TDP-43 knockdown vs. controls after optic nerve crush injury. 📊 Suggested Studies Comparative proteomics of RGCs in FTLD-TDP patient postmortem tissue to quantify STMN2 protein depletion. Longitudinal study of vitreous STMN2 levels and retinal thinning in presymptomatic C9orf72 mutation carriers. 📊 Swansons Literature Based Discovery Candidates TDP-43-induced STMN2 depletion impairs the regenerative potential of optic nerve fibers, potentially contributing to retinal pathology in ALS. TDP-43 loss of function leads to STMN2 mis-splicing and impaired axonal repair in motor neurons (Source: 40392845). Vitreous fluid in ALS/FTD patients shows reduced STMN2 levels, implying ocular-associated neurodegeneration (Source: 41180957). STMN2 protein, which is vital for microtubule dynamics and axonal regeneration. Since STMN2 is essential for axon regeneration in neurons and its levels are known to decline in the vitreous of TDP-43 pathology patients, it is mechanistically plausible that mis-splicing of STMN2 similarly inhibits the regenerative repair of retinal ganglion cell axons. 📊 Contradictions Between Evidences None identified in the current literature set. 📊 Repurposed Solutions Antisense oligonucleotides (ASOs) that correct STMN2 cryptic splicing in motor neurons could be repurposed for local intravitreal administration to preserve retinal ganglion cell health. 🔖 Tags Attractor Table Extracted Keywords & Entities TDP-43 pathology, _gates_from_tdp-43_pathology, RNA Splicing, _gates_to_rna_splicing, _gates_from_rna_splicing, Stathmin 2, _gates_to_stathmin_2, Vitreous Body, _gates_to_vitreous_body, DNA-Binding Protein-43, _gates_from_dna-binding_protein-43, Retinal Ganglion Cells, _gates_to_retinal_ganglion_cells, Cell Nucleus, _gates_from_cell_nucleus, _gates_from_stathmin_2, Axonal Degeneration, _gates_to_axonal_degeneration, Axons, _gates_to_axons, _gates_from_axons 🚀 Run Your Own Analysis PathMap is a patent-pending universal AI workbench designed to eliminate LLM hallucinations in medical research. Generate your own autonomous discovery reports at PathMap.org.","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21265318","URL":"https://doi.org/10.5281/zenodo.21265318","source":"datacite"},{"id":"doi:10.5281/zenodo.21265319","type":"article-journal","title":"PathMap Experiment #000023 - Tags: #TDP-43 pathology #RNA Splicing #Stathmin 2 #Vitreous Body #DNA-Binding Protein-43","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=23 Artificial General Intelligence LLC Claim Evaluated: Does misfolded TDP43 affect Retinal Ganglion Cell STMN2 ability to repair in a similar way as what how it was shown to affect motor neurons with cryptic mis-splicing? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Perform single-nuclei RNA sequencing (snRNA-seq) on retinal ganglion cells from TDP-43 mutant mouse models to assess STMN2 splicing profiles. Evaluate axonal regeneration capacity of RGCs derived from human iPSCs with TDP-43 knockdown vs. controls after optic nerve crush injury. 📊 Suggested Studies Comparative proteomics of RGCs in FTLD-TDP patient postmortem tissue to quantify STMN2 protein depletion. Longitudinal study of vitreous STMN2 levels and retinal thinning in presymptomatic C9orf72 mutation carriers. 📊 Swansons Literature Based Discovery Candidates TDP-43-induced STMN2 depletion impairs the regenerative potential of optic nerve fibers, potentially contributing to retinal pathology in ALS. TDP-43 loss of function leads to STMN2 mis-splicing and impaired axonal repair in motor neurons (Source: 40392845). Vitreous fluid in ALS/FTD patients shows reduced STMN2 levels, implying ocular-associated neurodegeneration (Source: 41180957). STMN2 protein, which is vital for microtubule dynamics and axonal regeneration. Since STMN2 is essential for axon regeneration in neurons and its levels are known to decline in the vitreous of TDP-43 pathology patients, it is mechanistically plausible that mis-splicing of STMN2 similarly inhibits the regenerative repair of retinal ganglion cell axons. 📊 Contradictions Between Evidences None identified in the current literature set. 📊 Repurposed Solutions Antisense oligonucleotides (ASOs) that correct STMN2 cryptic splicing in motor neurons could be repurposed for local intravitreal administration to preserve retinal ganglion cell health. 🔖 Tags Attractor Table Extracted Keywords & Entities TDP-43 pathology, _gates_from_tdp-43_pathology, RNA Splicing, _gates_to_rna_splicing, _gates_from_rna_splicing, Stathmin 2, _gates_to_stathmin_2, Vitreous Body, _gates_to_vitreous_body, DNA-Binding Protein-43, _gates_from_dna-binding_protein-43, Retinal Ganglion Cells, _gates_to_retinal_ganglion_cells, Cell Nucleus, _gates_from_cell_nucleus, _gates_from_stathmin_2, Axonal Degeneration, _gates_to_axonal_degeneration, Axons, _gates_to_axons, _gates_from_axons 🚀 Run Your Own Analysis PathMap is a patent-pending universal AI workbench designed to eliminate LLM hallucinations in medical research. Generate your own autonomous discovery reports at PathMap.org.","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21265319","URL":"https://doi.org/10.5281/zenodo.21265319","source":"datacite"},{"id":"doi:10.5281/zenodo.22020009","type":"article-journal","title":"Dataset: ApoE-mediated lipsignaling and EV-delivered bioenergetic substrates both converge on the stabilization of mitochondrial respiratory complexes, which is the requisite physiological precursor for renewed neurogenesis in the hippocampus. - PathMap Experiment #000133","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=133 Artificial General Intelligence LLC Claim Evaluated: ApoE-mediated lipsignaling and EV-delivered bioenergetic substrates both converge on the stabilization of mitochondrial respiratory complexes, which is the requisite physiological precursor for renewed neurogenesis in the hippocampus. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights ApoE4 is not merely a transport protein but a metabolic stressor that impairs mitochondrial membrane potential and glycolysis in astrocytes. EVs possess a 30-fold higher ganglioside content than the parent cells, suggesting unique signaling capabilities in mediating neuroplasticity. The entorhinal cortex exhibits region-specific bioenergetic regulation, contrasting with the cortex and hippocampus, indicating differential susceptibility to ApoE4. \"Neurogenesis without division\" in cortical immature neurons (cINs) offers a paradigm shift in how we view brain structural plasticity. Pharmacological inhibition of the lysosomal channel TMEM175 can alleviate mitochondrial dysfunction under oxidative stress through AMPK activation. SORD-related neuropathies demonstrate that muscle tissue itself is an active site of mitochondrial complex I and metabolic regulation, complicating systemic disease models. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Determine if EV-derived mitochondrial cargo can rescue hippocampal neurogenesis in ApoE4-TR mice under stress conditions. Evaluate the impact of UQCRC1/COX4I1 overexpression on the neurogenic potential of hippocampal neural stem cells in ApoE4-expressing models. 📊 Suggested Studies Longitudinal analysis of ganglioside content in circulating EVs as a proxy for hippocampal metabolic integrity in aging. 📊 Swansons Literature Based Discovery Candidates • Discovered Hypothesis (A to C): Extracellular vesicles (EVs) derived from muscle tissue can stabilize mitochondrial complexes in hippocampal neurons, promoting neurogenesis via metabolic substrate delivery. - Literature A (Origin): SORD deficiency study (ID: 42616755), detailing mitochondrial stress and metabolic dysfunction in skeletal muscle. - Literature C (Target): Studies on hippocampal neurogenesis (ID: 42602088), linking metabolic stabilization to memory circuits. - The Intersecting Bridge B: Mitochondrial respiratory chain complex proteins (e.g., UQCRC1/COX4I1) and metabolic regulatory signals (e.g., ATP-related metabolites). - Biological Rationale: Muscle-derived EVs carry metabolic cargo that, if distributed to the CNS, could provide the bioenergetic precursors necessary for hippocampal neurons to overcome the metabolic shifts associated with hippocampal sclerosis or aging. 📊 Contradictions Between Evidences There is a tension between the protective potential of EGFR activation (promoting neurogenesis) and its potential for promoting neurotoxicity/gliosis if chronic (ID 42591826). 📊 Repurposed Solutions Use of EV-based decoy receptors (glycoengineered with Gb3, ID 42615512) to sequester toxic circulating factors that impair neuronal bioenergetics. Tags Attractor Table Extracted Keywords & Entities Apolipoprotein E4, _gates_from_apolipoprotein_e4, Mitochondrial Respiration, _gates_to_mitochondrial_respiration, Extracellular Vesicles, _gates_from_extracellular_vesicles, Mitochondria, _gates_to_mitochondria, Energy Metabolism, _gates_from_energy_metabolism, Hippocampal Neurogenesis, _gates_to_hippocampal_neurogenesis Run Your Own Analysis PathMap is a patent-pending universal AI workbench designed to eliminate LLM hallucinations in medical research. Generate your own autonomous discovery reports at PathMap.org.","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22020009","URL":"https://doi.org/10.5281/zenodo.22020009","source":"datacite"},{"id":"doi:10.5281/zenodo.22020010","type":"article-journal","title":"Dataset: ApoE-mediated lipsignaling and EV-delivered bioenergetic substrates both converge on the stabilization of mitochondrial respiratory complexes, which is the requisite physiological precursor for renewed neurogenesis in the hippocampus. - PathMap Experiment #000133","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=133 Artificial General Intelligence LLC Claim Evaluated: ApoE-mediated lipsignaling and EV-delivered bioenergetic substrates both converge on the stabilization of mitochondrial respiratory complexes, which is the requisite physiological precursor for renewed neurogenesis in the hippocampus. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights ApoE4 is not merely a transport protein but a metabolic stressor that impairs mitochondrial membrane potential and glycolysis in astrocytes. EVs possess a 30-fold higher ganglioside content than the parent cells, suggesting unique signaling capabilities in mediating neuroplasticity. The entorhinal cortex exhibits region-specific bioenergetic regulation, contrasting with the cortex and hippocampus, indicating differential susceptibility to ApoE4. \"Neurogenesis without division\" in cortical immature neurons (cINs) offers a paradigm shift in how we view brain structural plasticity. Pharmacological inhibition of the lysosomal channel TMEM175 can alleviate mitochondrial dysfunction under oxidative stress through AMPK activation. SORD-related neuropathies demonstrate that muscle tissue itself is an active site of mitochondrial complex I and metabolic regulation, complicating systemic disease models. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Determine if EV-derived mitochondrial cargo can rescue hippocampal neurogenesis in ApoE4-TR mice under stress conditions. Evaluate the impact of UQCRC1/COX4I1 overexpression on the neurogenic potential of hippocampal neural stem cells in ApoE4-expressing models. 📊 Suggested Studies Longitudinal analysis of ganglioside content in circulating EVs as a proxy for hippocampal metabolic integrity in aging. 📊 Swansons Literature Based Discovery Candidates • Discovered Hypothesis (A to C): Extracellular vesicles (EVs) derived from muscle tissue can stabilize mitochondrial complexes in hippocampal neurons, promoting neurogenesis via metabolic substrate delivery. - Literature A (Origin): SORD deficiency study (ID: 42616755), detailing mitochondrial stress and metabolic dysfunction in skeletal muscle. - Literature C (Target): Studies on hippocampal neurogenesis (ID: 42602088), linking metabolic stabilization to memory circuits. - The Intersecting Bridge B: Mitochondrial respiratory chain complex proteins (e.g., UQCRC1/COX4I1) and metabolic regulatory signals (e.g., ATP-related metabolites). - Biological Rationale: Muscle-derived EVs carry metabolic cargo that, if distributed to the CNS, could provide the bioenergetic precursors necessary for hippocampal neurons to overcome the metabolic shifts associated with hippocampal sclerosis or aging. 📊 Contradictions Between Evidences There is a tension between the protective potential of EGFR activation (promoting neurogenesis) and its potential for promoting neurotoxicity/gliosis if chronic (ID 42591826). 📊 Repurposed Solutions Use of EV-based decoy receptors (glycoengineered with Gb3, ID 42615512) to sequester toxic circulating factors that impair neuronal bioenergetics. Tags Attractor Table Extracted Keywords & Entities Apolipoprotein E4, _gates_from_apolipoprotein_e4, Mitochondrial Respiration, _gates_to_mitochondrial_respiration, Extracellular Vesicles, _gates_from_extracellular_vesicles, Mitochondria, _gates_to_mitochondria, Energy Metabolism, _gates_from_energy_metabolism, Hippocampal Neurogenesis, _gates_to_hippocampal_neurogenesis Run Your Own Analysis PathMap is a patent-pending universal AI workbench designed to eliminate LLM hallucinations in medical research. Generate your own autonomous discovery reports at PathMap.org.","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22020010","URL":"https://doi.org/10.5281/zenodo.22020010","source":"datacite"},{"id":"doi:10.5281/zenodo.21285954","type":"article-journal","title":"Can inhaled COVID-19 vaccinations be used to help treat COPD? - PathMap Experiment #000040","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=40 Artificial General Intelligence LLC Claim Evaluated: Can inhaled COVID-19 vaccinations be used to help treat COPD? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights RSV infection in adults with chronic lung disease has been associated with a higher disease burden compared to influenza or SARS-CoV-2. Intranasal and pulmonary \"pull\" vaccination strategies are effective at inducing secretory IgA and lung-resident T-cell responses that are typically absent following intramuscular injection. The use of cationic ferritin nanoparticles or adenoviral vectors for intranasal delivery can overcome pre-existing immunity against viral vaccine vectors. Specific biomarkers, such as the CCL5/CCR4 signaling axis, are being identified as modulators of immune cell migration following mucosal immunization. There is a persistent \"interferon gap\" in the elderly, characterized by a kinetic delay in innate antiviral signaling, which mucosal platforms aim to bridge. The use of codon-pair deoptimization (CPD) in live-attenuated vaccines represents a novel safety mechanism for developing inhaled platforms. Some evidence suggests that high-dose systemic corticosteroids may paradoxically increase mortality in COVID-19 ARDS, reinforcing the need for targeted, localized prophylactic strategies like mucosal vaccination. Evidence suggests that the mucosal immune system in the respiratory tract can be specifically engaged to mitigate infection-driven COPD exacerbations. Inhaled delivery mechanisms for biologics can reduce the required dosage of therapeutic agents while bypassing gastrointestinal degradation. Lipid-based nanocarriers have demonstrated the ability to cross pulmonary mucosal barriers, a critical feature for both therapeutic and prophylactic agents. Advanced nebulization techniques (mesh versus jet nebulizers) significantly impact the efficiency of pulmonary delivery, which is vital for the clinical success of inhaled therapeutics. There is a clear distinction between the immunogenicity profiles of intramuscular (systemic IgG) and mucosal (respiratory IgA) vaccinations, with the latter showing promise for enhancing local airway resilience. Inhaled heparin is emerging as a versatile therapeutic option, given its established role in managing respiratory infections including COVID-19 and its potential use in asthma and COPD. Clinical data indicate that SARS-CoV-2 infection is associated with different mortality and inflammatory markers in patients with COPD versus other respiratory viruses. Nanotechnology integration into inhalers allows for precise, patient-centric dosing, which could improve adherence in chronic populations. A \"virus-agnostic\" immunomodulatory platform, distinct from traditional vaccines, is proposed to restore mucosal immune competence in the elderly (ID: 42347596). Intranasal vaccination with PIV5-vectored platforms has been evaluated in animal models, showing protection against challenge without necessarily relying on serum neutralizing antibodies (ID: 41863913). Probiotic supplementation has shown potential to reduce systemic inflammation and improve patient-reported outcomes in COPD (ID: 42286603). The gut-lung axis is increasingly recognized as a target, with researchers identifying cross-kingdom microbiome interactions that influence COPD outcomes (ID: 42324603). Sex-specific differences exist in neutrophil transcriptional programs in COPD, which may necessitate sex-dependent therapeutic strategies (ID: 42281812). There is a significant heterogeneity in clinical outcomes for ACOS (Asthma-COPD overlap), with newer glucose-lowering agents showing potential to modify risk (ID: 42376494). 🧪 Extra","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21285954","URL":"https://doi.org/10.5281/zenodo.21285954","source":"datacite"},{"id":"doi:10.5281/zenodo.21285955","type":"article-journal","title":"Can inhaled COVID-19 vaccinations be used to help treat COPD? - PathMap Experiment #000040","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=40 Artificial General Intelligence LLC Claim Evaluated: Can inhaled COVID-19 vaccinations be used to help treat COPD? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights RSV infection in adults with chronic lung disease has been associated with a higher disease burden compared to influenza or SARS-CoV-2. Intranasal and pulmonary \"pull\" vaccination strategies are effective at inducing secretory IgA and lung-resident T-cell responses that are typically absent following intramuscular injection. The use of cationic ferritin nanoparticles or adenoviral vectors for intranasal delivery can overcome pre-existing immunity against viral vaccine vectors. Specific biomarkers, such as the CCL5/CCR4 signaling axis, are being identified as modulators of immune cell migration following mucosal immunization. There is a persistent \"interferon gap\" in the elderly, characterized by a kinetic delay in innate antiviral signaling, which mucosal platforms aim to bridge. The use of codon-pair deoptimization (CPD) in live-attenuated vaccines represents a novel safety mechanism for developing inhaled platforms. Some evidence suggests that high-dose systemic corticosteroids may paradoxically increase mortality in COVID-19 ARDS, reinforcing the need for targeted, localized prophylactic strategies like mucosal vaccination. Evidence suggests that the mucosal immune system in the respiratory tract can be specifically engaged to mitigate infection-driven COPD exacerbations. Inhaled delivery mechanisms for biologics can reduce the required dosage of therapeutic agents while bypassing gastrointestinal degradation. Lipid-based nanocarriers have demonstrated the ability to cross pulmonary mucosal barriers, a critical feature for both therapeutic and prophylactic agents. Advanced nebulization techniques (mesh versus jet nebulizers) significantly impact the efficiency of pulmonary delivery, which is vital for the clinical success of inhaled therapeutics. There is a clear distinction between the immunogenicity profiles of intramuscular (systemic IgG) and mucosal (respiratory IgA) vaccinations, with the latter showing promise for enhancing local airway resilience. Inhaled heparin is emerging as a versatile therapeutic option, given its established role in managing respiratory infections including COVID-19 and its potential use in asthma and COPD. Clinical data indicate that SARS-CoV-2 infection is associated with different mortality and inflammatory markers in patients with COPD versus other respiratory viruses. Nanotechnology integration into inhalers allows for precise, patient-centric dosing, which could improve adherence in chronic populations. A \"virus-agnostic\" immunomodulatory platform, distinct from traditional vaccines, is proposed to restore mucosal immune competence in the elderly (ID: 42347596). Intranasal vaccination with PIV5-vectored platforms has been evaluated in animal models, showing protection against challenge without necessarily relying on serum neutralizing antibodies (ID: 41863913). Probiotic supplementation has shown potential to reduce systemic inflammation and improve patient-reported outcomes in COPD (ID: 42286603). The gut-lung axis is increasingly recognized as a target, with researchers identifying cross-kingdom microbiome interactions that influence COPD outcomes (ID: 42324603). Sex-specific differences exist in neutrophil transcriptional programs in COPD, which may necessitate sex-dependent therapeutic strategies (ID: 42281812). There is a significant heterogeneity in clinical outcomes for ACOS (Asthma-COPD overlap), with newer glucose-lowering agents showing potential to modify risk (ID: 42376494). 🧪 Extra","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21285955","URL":"https://doi.org/10.5281/zenodo.21285955","source":"datacite"},{"id":"doi:10.5281/zenodo.22019848","type":"article-journal","title":"Dataset: Discovery: Considering PubMed #41177462, intranasal S-GEVs co-functionalized with ApoE peptides may bypass the cribriform plate and target astrocytic LRP1 receptors in order to suppress NF-κB and may resolve some neuroinflammation in Alzheimer's and ALS. - PathMap Experiment #000132","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=132 Artificial General Intelligence LLC Claim Evaluated: Discovery: Considering PubMed #41177462, intranasal S-GEVs co-functionalized with ApoE peptides may bypass the cribriform plate and target astrocytic LRP1 receptors in order to suppress NF-κB and may resolve some neuroinflammation in Alzheimer's and ALS. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Targeted Engineering:** Angiopep-2 (Ang2) peptide-modified TEVs (Ang-TEVs) confer significantly enhanced microglial targeting. Vesicle Versatility:** Exosomes are naturally occurring extracellular vesicles that have emerged as promising bio-inspired nanocarriers for the treatment of neurological disorders owing to their intrinsic biocompatibility, low immunogenicity, and ability to cross the blood-brain barrier. Metabolic Reprogramming:** LEVs-SIRT2-KD were readily internalized by microglia in vivo following intranasal delivery. Cholesterol Coupling:** Neurons acquire astrocyte-derived cholesterol through LDLR/LRP1, redistribute it via NPC1/NPC2, and eliminate excess cholesterol as 24 S-hydroxycholesterol through CYP46A1. Inflammatory RNA:** Emerging evidence further demonstrates that inflammatory RNAs participate in epigenetic regulation, intercellular communication, and inter-organ crosstalk through extracellular vesicles and exosomes. Proteinopathy Neutralization:** PC-OxPL-VecTab neutralized PC-OxPL-induced neurotoxicity, reduced TDP-43 aggregation, and prevented motor neuron death and behavioral deficits in a sALS CSF transfer mouse model. Complement Cascade:** The complement cascade is closely related to AD-associated pathological processes; however, the precise mechanisms underlying its contributions remain incompletely elucidated. Exosomes isolated from PCA-treated efferocytic macrophages inhibited inflammation and increased miR-10b levels in aortic endothelial cells. They mimicked the protective effect of recombinant ApoE3Ch on endothelial integrity by restoring β-catenin nuclear localization. Some of these differentially expressed miRNAs were associated with key AD-related comorbidities such as APOE genotype, age, and metabolic burden and were predicted to target genes within NF-κB -regulated inflammatory pathways. We demonstrated the remarkable potential of MenSC-EVs in alleviating atherosclerosis through the NF-κB signaling pathway. APOE ε4 carriers presented further reductions in SV2A levels compared with noncarriers. SeNExo penetrates the blood-brain barrier (BBB) via the apolipoprotein E and prolow-density lipoprotein receptor-related protein 1 (APOE_LRP-1) interaction. In the adult brain, astrocytes are an important source of cholesterol for neurons. One of the key functions of APOE is to bind and deliver newly synthesized cholesterol and lipids to neurons through receptor-mediated endocytosis (LDLR, LRP1, VLDLR, APOER2). Macrophage-specific PSRC1 depletion alone was sufficient to recapitulate the systemic hypercholesterolemia and accelerated atherosclerosis observed in whole-body knockout models. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Assess the biodistribution of ApoE-peptide functionalized EVs after intranasal administration in APP/PS1 mice using IVIS imaging. Evaluate the suppression of p-NF-κB in astrocytes following treatment with ApoE-EVs in LPS-stimulated in vitro co-culture models. Assess the therapeutic efficacy of ApoE-conjugated EVs pre-loaded with metabolic substrates in a 5xFAD mouse model. Perform proteomics on EVs to confirm successful co-incorporation of metabolic modifiers and ApoE-mimetic peptides without cargo degradation. 📊 Suggested Studies Longitudinal study comparing the cognitive recov","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22019848","URL":"https://doi.org/10.5281/zenodo.22019848","source":"datacite"},{"id":"doi:10.5281/zenodo.22019849","type":"article-journal","title":"Dataset: Discovery: Considering PubMed #41177462, intranasal S-GEVs co-functionalized with ApoE peptides may bypass the cribriform plate and target astrocytic LRP1 receptors in order to suppress NF-κB and may resolve some neuroinflammation in Alzheimer's and ALS. - PathMap Experiment #000132","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=132 Artificial General Intelligence LLC Claim Evaluated: Discovery: Considering PubMed #41177462, intranasal S-GEVs co-functionalized with ApoE peptides may bypass the cribriform plate and target astrocytic LRP1 receptors in order to suppress NF-κB and may resolve some neuroinflammation in Alzheimer's and ALS. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Targeted Engineering:** Angiopep-2 (Ang2) peptide-modified TEVs (Ang-TEVs) confer significantly enhanced microglial targeting. Vesicle Versatility:** Exosomes are naturally occurring extracellular vesicles that have emerged as promising bio-inspired nanocarriers for the treatment of neurological disorders owing to their intrinsic biocompatibility, low immunogenicity, and ability to cross the blood-brain barrier. Metabolic Reprogramming:** LEVs-SIRT2-KD were readily internalized by microglia in vivo following intranasal delivery. Cholesterol Coupling:** Neurons acquire astrocyte-derived cholesterol through LDLR/LRP1, redistribute it via NPC1/NPC2, and eliminate excess cholesterol as 24 S-hydroxycholesterol through CYP46A1. Inflammatory RNA:** Emerging evidence further demonstrates that inflammatory RNAs participate in epigenetic regulation, intercellular communication, and inter-organ crosstalk through extracellular vesicles and exosomes. Proteinopathy Neutralization:** PC-OxPL-VecTab neutralized PC-OxPL-induced neurotoxicity, reduced TDP-43 aggregation, and prevented motor neuron death and behavioral deficits in a sALS CSF transfer mouse model. Complement Cascade:** The complement cascade is closely related to AD-associated pathological processes; however, the precise mechanisms underlying its contributions remain incompletely elucidated. Exosomes isolated from PCA-treated efferocytic macrophages inhibited inflammation and increased miR-10b levels in aortic endothelial cells. They mimicked the protective effect of recombinant ApoE3Ch on endothelial integrity by restoring β-catenin nuclear localization. Some of these differentially expressed miRNAs were associated with key AD-related comorbidities such as APOE genotype, age, and metabolic burden and were predicted to target genes within NF-κB -regulated inflammatory pathways. We demonstrated the remarkable potential of MenSC-EVs in alleviating atherosclerosis through the NF-κB signaling pathway. APOE ε4 carriers presented further reductions in SV2A levels compared with noncarriers. SeNExo penetrates the blood-brain barrier (BBB) via the apolipoprotein E and prolow-density lipoprotein receptor-related protein 1 (APOE_LRP-1) interaction. In the adult brain, astrocytes are an important source of cholesterol for neurons. One of the key functions of APOE is to bind and deliver newly synthesized cholesterol and lipids to neurons through receptor-mediated endocytosis (LDLR, LRP1, VLDLR, APOER2). Macrophage-specific PSRC1 depletion alone was sufficient to recapitulate the systemic hypercholesterolemia and accelerated atherosclerosis observed in whole-body knockout models. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Assess the biodistribution of ApoE-peptide functionalized EVs after intranasal administration in APP/PS1 mice using IVIS imaging. Evaluate the suppression of p-NF-κB in astrocytes following treatment with ApoE-EVs in LPS-stimulated in vitro co-culture models. Assess the therapeutic efficacy of ApoE-conjugated EVs pre-loaded with metabolic substrates in a 5xFAD mouse model. Perform proteomics on EVs to confirm successful co-incorporation of metabolic modifiers and ApoE-mimetic peptides without cargo degradation. 📊 Suggested Studies Longitudinal study comparing the cognitive recov","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22019849","URL":"https://doi.org/10.5281/zenodo.22019849","source":"datacite"},{"id":"doi:10.5281/zenodo.21251408","type":"article-journal","title":"PathMap Experiment #000018 - Tags: #Tobacco Smoke Pollution #Epithelium #Dysbiosis #Immune System Diseases #Cigarette Smoke","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=18 Artificial General Intelligence LLC Claim Evaluated: Can smoking cigarettes cause dysbiosis? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Longitudinal microbiome analysis of patients undergoing smoking cessation programs to determine the temporal dynamics of microbiome restoration. In vitro air-liquid interface (ALI) co-culture models of airway epithelial cells and diverse commensal microbiota exposed to standardized cigarette smoke extract (CSE) to measure real-time barrier stability and microbial shift. Fecal microbial transplantation (FMT) of microbiome from chronic smokers into germ-free mouse models to determine if smoking-induced metabolic shifts (e.g., tryptophan depletion) are sufficient to induce phenotypic inflammatory disease. 📊 Suggested Studies Multi-center prospective study correlating smoking-induced gut-lung axis biomarkers (indolepropionate/bile acids) with respiratory exacerbation frequency. Large-scale cohort study assessing the impact of vaping versus conventional cigarette smoking on oral versus gut microbiome diversity using standardized protocols. Longitudinal cohort analysis mapping the evolution of the respiratory microbiome in healthy subjects before and after the initiation of tobacco smoking. 📊 Swansons Literature Based Discovery Candidates Cigarette-induced depletion of indolepropionate (via gut dysbiosis) accelerates respiratory barrier breakdown by reducing Muc16-mediated epithelial maintenance. Tobacco exposure disrupts host-microbiome tryptophan and bile acid metabolism, specifically indolepropionate depletion in smokers with MS (ID: 42383698). Muc16 deficiency exacerbates pneumococcal translocation and epithelial barrier disruption in the upper respiratory tract, especially under CSE exposure (ID: 42383770). Mucosal barrier protection and epithelial tight junction integrity (ZO-1 protein expression). Indolepropionate is an anti-inflammatory metabolite that preserves barrier integrity; its depletion in smokers may directly compromise the expression/stability of Muc16 and associated tight junction proteins (e.g., ZO-1), leaving the respiratory epithelium vulnerable to bacterial invasion. 📊 Contradictions Between Evidences While most studies demonstrate that smoking affects microbial composition, ID 42388034 notes 'no major change in overall community diversity' in tobacco-related rhizosphere profiling, suggesting that smoking impacts might be context-specific (human versus botanical/rhizosphere ecosystems). 📊 Repurposed Solutions Probiotic supplementation (e.g., Bifidobacterium) and targeted metabolic precursors (indole-3-propionate) represent repurposed strategies to restore microbiome homeostasis and barrier function in smokers, potentially mitigating risks of respiratory exacerbations and secondary infection. 🔖 Tags Attractor Table Extracted Keywords & Entities Tobacco Smoke Pollution, _gates_from_tobacco_smoke_pollution, Epithelium, _gates_to_epithelium, _gates_from_epithelium, Dysbiosis, _gates_to_dysbiosis, _gates_from_dysbiosis, Immune System Diseases, _gates_to_immune_system_diseases, Cigarette Smoke, _gates_from_cigarette_smoke, Oxidative Stress, _gates_to_oxidative_stress, _gates_from_oxidative_stress, Microbiota, _gates_to_microbiota, _gates_from_microbiota, Tobacco Smoke, _gates_from_tobacco_smoke, Microbial Niche, _gates_to_microbial_niche, _gates_from_microbial_niche, Bacteria, _gates_to_bacteria, _gates_from_bacteria 🚀 Run Your Own Analysis PathMap is a patent-pending universal AI workbench designed to eliminate LLM hallucinations in medical research. Generate your own autonomous discovery reports at PathMap.org.","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21251408","URL":"https://doi.org/10.5281/zenodo.21251408","source":"datacite"},{"id":"doi:10.5281/zenodo.21251409","type":"article-journal","title":"PathMap Experiment #000018 - Tags: #Tobacco Smoke Pollution #Epithelium #Dysbiosis #Immune System Diseases #Cigarette Smoke","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=18 Artificial General Intelligence LLC Claim Evaluated: Can smoking cigarettes cause dysbiosis? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Longitudinal microbiome analysis of patients undergoing smoking cessation programs to determine the temporal dynamics of microbiome restoration. In vitro air-liquid interface (ALI) co-culture models of airway epithelial cells and diverse commensal microbiota exposed to standardized cigarette smoke extract (CSE) to measure real-time barrier stability and microbial shift. Fecal microbial transplantation (FMT) of microbiome from chronic smokers into germ-free mouse models to determine if smoking-induced metabolic shifts (e.g., tryptophan depletion) are sufficient to induce phenotypic inflammatory disease. 📊 Suggested Studies Multi-center prospective study correlating smoking-induced gut-lung axis biomarkers (indolepropionate/bile acids) with respiratory exacerbation frequency. Large-scale cohort study assessing the impact of vaping versus conventional cigarette smoking on oral versus gut microbiome diversity using standardized protocols. Longitudinal cohort analysis mapping the evolution of the respiratory microbiome in healthy subjects before and after the initiation of tobacco smoking. 📊 Swansons Literature Based Discovery Candidates Cigarette-induced depletion of indolepropionate (via gut dysbiosis) accelerates respiratory barrier breakdown by reducing Muc16-mediated epithelial maintenance. Tobacco exposure disrupts host-microbiome tryptophan and bile acid metabolism, specifically indolepropionate depletion in smokers with MS (ID: 42383698). Muc16 deficiency exacerbates pneumococcal translocation and epithelial barrier disruption in the upper respiratory tract, especially under CSE exposure (ID: 42383770). Mucosal barrier protection and epithelial tight junction integrity (ZO-1 protein expression). Indolepropionate is an anti-inflammatory metabolite that preserves barrier integrity; its depletion in smokers may directly compromise the expression/stability of Muc16 and associated tight junction proteins (e.g., ZO-1), leaving the respiratory epithelium vulnerable to bacterial invasion. 📊 Contradictions Between Evidences While most studies demonstrate that smoking affects microbial composition, ID 42388034 notes 'no major change in overall community diversity' in tobacco-related rhizosphere profiling, suggesting that smoking impacts might be context-specific (human versus botanical/rhizosphere ecosystems). 📊 Repurposed Solutions Probiotic supplementation (e.g., Bifidobacterium) and targeted metabolic precursors (indole-3-propionate) represent repurposed strategies to restore microbiome homeostasis and barrier function in smokers, potentially mitigating risks of respiratory exacerbations and secondary infection. 🔖 Tags Attractor Table Extracted Keywords & Entities Tobacco Smoke Pollution, _gates_from_tobacco_smoke_pollution, Epithelium, _gates_to_epithelium, _gates_from_epithelium, Dysbiosis, _gates_to_dysbiosis, _gates_from_dysbiosis, Immune System Diseases, _gates_to_immune_system_diseases, Cigarette Smoke, _gates_from_cigarette_smoke, Oxidative Stress, _gates_to_oxidative_stress, _gates_from_oxidative_stress, Microbiota, _gates_to_microbiota, _gates_from_microbiota, Tobacco Smoke, _gates_from_tobacco_smoke, Microbial Niche, _gates_to_microbial_niche, _gates_from_microbial_niche, Bacteria, _gates_to_bacteria, _gates_from_bacteria 🚀 Run Your Own Analysis PathMap is a patent-pending universal AI workbench designed to eliminate LLM hallucinations in medical research. Generate your own autonomous discovery reports at PathMap.org.","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21251409","URL":"https://doi.org/10.5281/zenodo.21251409","source":"datacite"},{"id":"doi:10.5281/zenodo.22013680","type":"article-journal","title":"A Practical Protocol for Conducting and Writing Medical Research: A Step-by-Step Guide for Medical Students","abstract":"A Practical Protocol for Conducting and Writing Medical Research: A Step-by-Step Guide for Medical Students This educational resource provides a practical, step-by-step framework to help medical students and early-stage researchers move from an initial research idea to a structured, submission-ready scientific manuscript. The guide follows a sequential pathway: IDEA → QUESTION → EVIDENCE → DESIGN → PROTOCOL → DATA → MANUSCRIPT → SUBMISSION It covers the main stages of a medical research project, including formulation of a focused research question using PICO and FINER principles, literature searching and evidence identification, selection of an appropriate study design, development of a basic research protocol, manuscript structure and scientific writing, responsible use of artificial intelligence tools, academic integrity and plagiarism prevention, journal selection, and pre-submission quality control. Particular emphasis is placed on practical methodological decisions that commonly challenge students during their first research projects, including defining study populations and variables, planning outcomes and data collection, considering sample size and statistical analysis, addressing ethical requirements, and maintaining a clear distinction between Methods, Results, Discussion, and Conclusion. The resource also introduces common reporting frameworks and publication practices to help students align their work with established standards in medical research and scientific publishing. This guide is intended primarily for undergraduate and early postgraduate medical students undertaking their first research projects, while also serving as a concise reference for supervisors, tutors, and educators supporting students in medical research. This document is an educational and practical resource. It is not intended to replace formal training in research methodology, epidemiology, biostatistics, research ethics, or supervision by qualified researchers. Specific requirements may vary according to the study design, institution, ethics committee, target journal, and publisher. Version: 1.1 (2026)","author":[{"family":"Telli","given":"Radhia"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22013680","URL":"https://doi.org/10.5281/zenodo.22013680","source":"datacite"},{"id":"doi:10.5281/zenodo.22013681","type":"article-journal","title":"A Practical Protocol for Conducting and Writing Medical Research: A Step-by-Step Guide for Medical Students","abstract":"A Practical Protocol for Conducting and Writing Medical Research: A Step-by-Step Guide for Medical Students This educational resource provides a practical, step-by-step framework to help medical students and early-stage researchers move from an initial research idea to a structured, submission-ready scientific manuscript. The guide follows a sequential pathway: IDEA → QUESTION → EVIDENCE → DESIGN → PROTOCOL → DATA → MANUSCRIPT → SUBMISSION It covers the main stages of a medical research project, including formulation of a focused research question using PICO and FINER principles, literature searching and evidence identification, selection of an appropriate study design, development of a basic research protocol, manuscript structure and scientific writing, responsible use of artificial intelligence tools, academic integrity and plagiarism prevention, journal selection, and pre-submission quality control. Particular emphasis is placed on practical methodological decisions that commonly challenge students during their first research projects, including defining study populations and variables, planning outcomes and data collection, considering sample size and statistical analysis, addressing ethical requirements, and maintaining a clear distinction between Methods, Results, Discussion, and Conclusion. The resource also introduces common reporting frameworks and publication practices to help students align their work with established standards in medical research and scientific publishing. This guide is intended primarily for undergraduate and early postgraduate medical students undertaking their first research projects, while also serving as a concise reference for supervisors, tutors, and educators supporting students in medical research. This document is an educational and practical resource. It is not intended to replace formal training in research methodology, epidemiology, biostatistics, research ethics, or supervision by qualified researchers. Specific requirements may vary according to the study design, institution, ethics committee, target journal, and publisher. Version: 1.1 (2026)","author":[{"family":"Telli","given":"Radhia"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22013681","URL":"https://doi.org/10.5281/zenodo.22013681","source":"datacite"},{"id":"doi:10.5281/zenodo.21231692","type":"article-journal","title":"PathMap Experiment #000006 - Tags: #Quercetin #Autophagy #Misfolded Proteins #DUX4","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=6 Artificial General Intelligence LLC Claim Evaluated: Does existing in vitro data show that Quercetin-induced autophagy can successfully degrade DUX4 or its downstream misfolded proteins? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments 1. Perform a Western blot analysis of DUX4 protein levels in primary muscle cell cultures treated with varying concentrations of Quercetin to determine if autophagic flux influences its turnover. 2. Conduct a Co-IP study to investigate if Quercetin-induced autophagy proteins colocalize with DUX4-GFP aggregates in an in vitro dystrophy model. 📊 Suggested Studies 1. A systematic screening of Quercetin-modified natural products on DUX4-dependent myocyte toxicity using an automated high-content imaging platform. 2. Comparative transcriptomic profiling of DUX4-expressing cells vs. control cells after Quercetin-induced autophagy modulation to identify potential degradation targets. 📊 Swansons Literature Based Discovery Candidates • Discovered Hypothesis (A to C): Quercetin-mediated autophagy modulation may alleviate DUX4-induced myotoxicity by increasing the degradation of toxic protein aggregates. • Literature A (Origin): Quercetin enhances autophagy-mediated degradation of toxic protein aggregates (ID: 40351085). • Literature C (Target): DUX4 aggregation and proteotoxicity in facioscapulohumeral muscular dystrophy (DUX4 is absent but implied by proteotoxicity themes). • The Intersecting Bridge B: Autophagy-lysosome pathway (ALP). • Biological Rationale: Quercetin acts as a generalist autophagy activator in models involving misfolded protein accumulation; DUX4 creates toxic aggregates, making them a plausible substrate for autophagic clearance. 📊 Contradictions Between Evidences There are no direct contradictions regarding Quercetin-induced autophagy; however, Quercetin acts as both an autophagy activator and an inhibitor (e.g., in ferritinophagy), which could lead to divergent clinical outcomes depending on the specific model. 📊 Repurposed Solutions Quercetin is established as a versatile scaffold for nanocarrier-mediated delivery to target intracellular protein homeostasis; this platform is potentially transferable to muscular dystrophies where toxic protein accumulation is a primary mechanism. 🔖 Tags Attractor Table Extracted Keywords & Entities Quercetin, _gates_from_quercetin, Autophagy, _gates_to_autophagy, _gates_from_autophagy, Misfolded Proteins, _gates_to_misfolded_proteins, _gates_from_misfolded_proteins, DUX4, _gates_to_dux4 🚀 Run Your Own Analysis PathMap is a patent-pending universal AI workbench designed to eliminate LLM hallucinations in medical research. Generate your own autonomous discovery reports at PathMap.org.","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21231692","URL":"https://doi.org/10.5281/zenodo.21231692","source":"datacite"},{"id":"doi:10.5281/zenodo.21231693","type":"article-journal","title":"PathMap Experiment #000006 - Tags: #Quercetin #Autophagy #Misfolded Proteins #DUX4","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=6 Artificial General Intelligence LLC Claim Evaluated: Does existing in vitro data show that Quercetin-induced autophagy can successfully degrade DUX4 or its downstream misfolded proteins? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments 1. Perform a Western blot analysis of DUX4 protein levels in primary muscle cell cultures treated with varying concentrations of Quercetin to determine if autophagic flux influences its turnover. 2. Conduct a Co-IP study to investigate if Quercetin-induced autophagy proteins colocalize with DUX4-GFP aggregates in an in vitro dystrophy model. 📊 Suggested Studies 1. A systematic screening of Quercetin-modified natural products on DUX4-dependent myocyte toxicity using an automated high-content imaging platform. 2. Comparative transcriptomic profiling of DUX4-expressing cells vs. control cells after Quercetin-induced autophagy modulation to identify potential degradation targets. 📊 Swansons Literature Based Discovery Candidates • Discovered Hypothesis (A to C): Quercetin-mediated autophagy modulation may alleviate DUX4-induced myotoxicity by increasing the degradation of toxic protein aggregates. • Literature A (Origin): Quercetin enhances autophagy-mediated degradation of toxic protein aggregates (ID: 40351085). • Literature C (Target): DUX4 aggregation and proteotoxicity in facioscapulohumeral muscular dystrophy (DUX4 is absent but implied by proteotoxicity themes). • The Intersecting Bridge B: Autophagy-lysosome pathway (ALP). • Biological Rationale: Quercetin acts as a generalist autophagy activator in models involving misfolded protein accumulation; DUX4 creates toxic aggregates, making them a plausible substrate for autophagic clearance. 📊 Contradictions Between Evidences There are no direct contradictions regarding Quercetin-induced autophagy; however, Quercetin acts as both an autophagy activator and an inhibitor (e.g., in ferritinophagy), which could lead to divergent clinical outcomes depending on the specific model. 📊 Repurposed Solutions Quercetin is established as a versatile scaffold for nanocarrier-mediated delivery to target intracellular protein homeostasis; this platform is potentially transferable to muscular dystrophies where toxic protein accumulation is a primary mechanism. 🔖 Tags Attractor Table Extracted Keywords & Entities Quercetin, _gates_from_quercetin, Autophagy, _gates_to_autophagy, _gates_from_autophagy, Misfolded Proteins, _gates_to_misfolded_proteins, _gates_from_misfolded_proteins, DUX4, _gates_to_dux4 🚀 Run Your Own Analysis PathMap is a patent-pending universal AI workbench designed to eliminate LLM hallucinations in medical research. Generate your own autonomous discovery reports at PathMap.org.","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21231693","URL":"https://doi.org/10.5281/zenodo.21231693","source":"datacite"},{"id":"doi:10.5281/zenodo.20739971","type":"article-journal","title":"THE TRINITY OF NEXT-GENERATION ADVANCED MATERIAL TECHNOLOGIES","abstract":"Description The Trinity of Next-Generation Advanced Material Technologies presents a comprehensive framework for the future of sustainable, intelligent, and biologically integrated material systems. This work explores the convergence of three transformative technological domains: Biological-Based Material Engineering (BBME), Integrated Bio-Convergence (IBC), and Next-Generation Bio-Adaptive Intelligent Materials (NBAIM). The framework investigates how renewable biomass resources, synthetic biology, nanotechnology, advanced manufacturing, and artificial intelligence can be integrated into a unified industrial ecosystem capable of producing self-healing, biodegradable, adaptive, and biologically interactive materials. The study establishes a complete pathway from biomass feedstock acquisition and preprocessing to industrial-scale manufacturing, international regulatory compliance, and global commercialization. The first domain, Biological-Based Material Engineering, focuses on converting agricultural residues and organic waste streams—including rice husks, rice straw, sugarcane bagasse, coconut biomass, chitin-rich shell waste, and protein-rich industrial by-products—into high-value biomaterials, bioplastics, self-healing composites, and sustainable packaging solutions. Emphasis is placed on circular bioeconomy principles, carbon-neutral production systems, and environmentally responsible manufacturing. The second domain, Integrated Bio-Convergence, examines the fusion of biological materials with nanoelectronics, flexible semiconductors, biosensors, artificial intelligence, and advanced computational systems. This convergence enables the development of intelligent medical implants, neural interfaces, bioelectronic devices, smart stents, and next-generation healthcare technologies capable of real-time physiological interaction and adaptive therapeutic responses. The third domain, Next-Generation Bio-Adaptive Intelligent Materials, explores materials capable of sensing, responding, and dynamically adapting to biological and environmental stimuli. These systems include stimuli-responsive polymers, shape-memory biomaterials, engineered living materials, smart wound dressings, adaptive textiles, and programmable biodegradable implants designed to interact directly with living systems. In addition to scientific and engineering foundations, the framework provides detailed guidance regarding biomass quality requirements, feedstock specifications, industrial equipment infrastructure, manufacturing workflows, regulatory pathways, international standards, sustainability metrics, and commercialization strategies. Key standards discussed include ASTM D6400, EN 13432, ISO 10993, ISO 13485, USP Class VI, FDA regulatory frameworks, MDR Class III requirements, and Life Cycle Assessment methodologies under ISO 14040 and ISO 14044. The document proposes a global industrialization roadmap extending from 2026 to 2050, outlining the transition from conventional passive materials toward intelligent, regenerative, self-adaptive material ecosystems. Applications span healthcare, construction, transportation, consumer products, aerospace, environmental remediation, smart cities, advanced manufacturing, and future human-machine interfaces. This publication is intended as a strategic reference for researchers, engineers, policymakers, industrial stakeholders, investors, and multidisciplinary innovation communities seeking to accelerate the development of biologically inspired and AI-enabled material technologies for the twenty-first century. Keywords: Biological-Based Material Engineering, Bio-Convergence, Bio-Adaptive Materials, Biomaterials, Synthetic Biology, Precision Fermentation, Nanotechnology, Artificial Intelligence, Smart Materials, Self-Healing Materials, Sustainable Manufacturing, Circular Bioeconomy, Bioelectronics, Tissue Engineering, Biopolymers, Advanced Materials, Regenerative Systems, Future Manufacturing. License: Creative Commons Attri","author":[{"family":"Nhut","given":"Nhut"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20739971","URL":"https://doi.org/10.5281/zenodo.20739971","source":"datacite"},{"id":"doi:10.5281/zenodo.20739972","type":"article-journal","title":"THE TRINITY OF NEXT-GENERATION ADVANCED MATERIAL TECHNOLOGIES","abstract":"Description The Trinity of Next-Generation Advanced Material Technologies presents a comprehensive framework for the future of sustainable, intelligent, and biologically integrated material systems. This work explores the convergence of three transformative technological domains: Biological-Based Material Engineering (BBME), Integrated Bio-Convergence (IBC), and Next-Generation Bio-Adaptive Intelligent Materials (NBAIM). The framework investigates how renewable biomass resources, synthetic biology, nanotechnology, advanced manufacturing, and artificial intelligence can be integrated into a unified industrial ecosystem capable of producing self-healing, biodegradable, adaptive, and biologically interactive materials. The study establishes a complete pathway from biomass feedstock acquisition and preprocessing to industrial-scale manufacturing, international regulatory compliance, and global commercialization. The first domain, Biological-Based Material Engineering, focuses on converting agricultural residues and organic waste streams—including rice husks, rice straw, sugarcane bagasse, coconut biomass, chitin-rich shell waste, and protein-rich industrial by-products—into high-value biomaterials, bioplastics, self-healing composites, and sustainable packaging solutions. Emphasis is placed on circular bioeconomy principles, carbon-neutral production systems, and environmentally responsible manufacturing. The second domain, Integrated Bio-Convergence, examines the fusion of biological materials with nanoelectronics, flexible semiconductors, biosensors, artificial intelligence, and advanced computational systems. This convergence enables the development of intelligent medical implants, neural interfaces, bioelectronic devices, smart stents, and next-generation healthcare technologies capable of real-time physiological interaction and adaptive therapeutic responses. The third domain, Next-Generation Bio-Adaptive Intelligent Materials, explores materials capable of sensing, responding, and dynamically adapting to biological and environmental stimuli. These systems include stimuli-responsive polymers, shape-memory biomaterials, engineered living materials, smart wound dressings, adaptive textiles, and programmable biodegradable implants designed to interact directly with living systems. In addition to scientific and engineering foundations, the framework provides detailed guidance regarding biomass quality requirements, feedstock specifications, industrial equipment infrastructure, manufacturing workflows, regulatory pathways, international standards, sustainability metrics, and commercialization strategies. Key standards discussed include ASTM D6400, EN 13432, ISO 10993, ISO 13485, USP Class VI, FDA regulatory frameworks, MDR Class III requirements, and Life Cycle Assessment methodologies under ISO 14040 and ISO 14044. The document proposes a global industrialization roadmap extending from 2026 to 2050, outlining the transition from conventional passive materials toward intelligent, regenerative, self-adaptive material ecosystems. Applications span healthcare, construction, transportation, consumer products, aerospace, environmental remediation, smart cities, advanced manufacturing, and future human-machine interfaces. This publication is intended as a strategic reference for researchers, engineers, policymakers, industrial stakeholders, investors, and multidisciplinary innovation communities seeking to accelerate the development of biologically inspired and AI-enabled material technologies for the twenty-first century. Keywords: Biological-Based Material Engineering, Bio-Convergence, Bio-Adaptive Materials, Biomaterials, Synthetic Biology, Precision Fermentation, Nanotechnology, Artificial Intelligence, Smart Materials, Self-Healing Materials, Sustainable Manufacturing, Circular Bioeconomy, Bioelectronics, Tissue Engineering, Biopolymers, Advanced Materials, Regenerative Systems, Future Manufacturing. License: Creative Commons Attri","author":[{"family":"Nhut","given":"Nhut"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20739972","URL":"https://doi.org/10.5281/zenodo.20739972","source":"datacite"},{"id":"doi:10.5281/zenodo.22014991","type":"article-journal","title":"The Speed Illusion: What Healthcare's AI Race Misreads About Credentialing, and the Ninety Days No Algorithm Controls","abstract":"The healthcare credentialing software market is projected to grow from approximately $3.8 billion in 2026to more than $10 billion by 2035, driven substantially by vendors marketing artificial intelligence andautomation as a solution to slow provider onboarding. This report examines a specific, checkable claimembedded in that marketing: that automated platforms can compress credentialing timelines to under thirtydays. Comparing vendor marketing claims directly against processing time data published by state medicalboards, this report finds that the advertised timeline is achievable in only a narrow subset of states,primarily those participating in the Interstate Medical Licensure Compact, and is mathematicallyincompatible with published board level processing times in several of the largest states by population,including California and Texas. The report introduces the Koru Controllable Time Ratio, an originalframework for separating the portion of a credentialing timeline that automation can genuinely influencefrom the portion that is structurally controlled by third party institutions automation cannot touch. Theanalysis finds that the majority of a typical credentialing cycle sits outside any single organization'scontrol, regardless of the software it deploys, and argues that automation's real value lies in eliminatingunforced errors rather than compressing a timeline it was never positioned to control. The report presents a blended timeline scenario model showing how an organization's realisticcredentialing timeline shifts based on its actual state footprint rather than a single national average, andidentifies three genuine, no cost levers: fingerprint method, application timing, and interstate compactstrategy. Each meaningfully affects total cycle time without requiring any software purchase. It closes byquantifying the financial exposure created when an organization budgets its staffing and revenueprojections around a vendor's advertised timeline rather than its realistic one. The report closes by identifying three structural blind spots that receive little attention in current marketdiscourse: a silent CAQH re attestation lapse that can halt payer access without triggering any alert, aMedicare enrollment gap arising from the false assumption that commercial payer credentialing andfederal PECOS enrollment are the same process, and a delegation trapdoor in which an organization'slargest available speed advantage, delegated payer credentialing, can be silently revoked by a single auditcitation, instantly reverting every provider in the pipeline from a matter of days to a matter of months. Allthree are genuinely, verifiably solvable by automation, in contrast to the board level processing delays thisreport shows automation cannot touch, suggesting the market is directing its engineering effort at thewrong problem while three real, addressable ones remain largely unmarketed.","author":[{"family":"Frazier","given":"Marcus"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22014991","URL":"https://doi.org/10.5281/zenodo.22014991","source":"datacite"},{"id":"doi:10.5281/zenodo.22014992","type":"article-journal","title":"The Speed Illusion: What Healthcare's AI Race Misreads About Credentialing, and the Ninety Days No Algorithm Controls","abstract":"The healthcare credentialing software market is projected to grow from approximately $3.8 billion in 2026to more than $10 billion by 2035, driven substantially by vendors marketing artificial intelligence andautomation as a solution to slow provider onboarding. This report examines a specific, checkable claimembedded in that marketing: that automated platforms can compress credentialing timelines to under thirtydays. Comparing vendor marketing claims directly against processing time data published by state medicalboards, this report finds that the advertised timeline is achievable in only a narrow subset of states,primarily those participating in the Interstate Medical Licensure Compact, and is mathematicallyincompatible with published board level processing times in several of the largest states by population,including California and Texas. The report introduces the Koru Controllable Time Ratio, an originalframework for separating the portion of a credentialing timeline that automation can genuinely influencefrom the portion that is structurally controlled by third party institutions automation cannot touch. Theanalysis finds that the majority of a typical credentialing cycle sits outside any single organization'scontrol, regardless of the software it deploys, and argues that automation's real value lies in eliminatingunforced errors rather than compressing a timeline it was never positioned to control. The report presents a blended timeline scenario model showing how an organization's realisticcredentialing timeline shifts based on its actual state footprint rather than a single national average, andidentifies three genuine, no cost levers: fingerprint method, application timing, and interstate compactstrategy. Each meaningfully affects total cycle time without requiring any software purchase. It closes byquantifying the financial exposure created when an organization budgets its staffing and revenueprojections around a vendor's advertised timeline rather than its realistic one. The report closes by identifying three structural blind spots that receive little attention in current marketdiscourse: a silent CAQH re attestation lapse that can halt payer access without triggering any alert, aMedicare enrollment gap arising from the false assumption that commercial payer credentialing andfederal PECOS enrollment are the same process, and a delegation trapdoor in which an organization'slargest available speed advantage, delegated payer credentialing, can be silently revoked by a single auditcitation, instantly reverting every provider in the pipeline from a matter of days to a matter of months. Allthree are genuinely, verifiably solvable by automation, in contrast to the board level processing delays thisreport shows automation cannot touch, suggesting the market is directing its engineering effort at thewrong problem while three real, addressable ones remain largely unmarketed.","author":[{"family":"Frazier","given":"Marcus"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22014992","URL":"https://doi.org/10.5281/zenodo.22014992","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33289404.v1","type":"article-journal","title":"The Speed Illusion: What Healthcare's AI Race Misreads About Credentialing, and the Ninety Days No","abstract":"The healthcare credentialing software market is projected to grow from approximately $3.8 billion in 2026 to more than $10 billion by 2035, driven substantially by vendors marketing artificial intelligence and automation as a solution to slow provider onboarding. This report examines a specific, checkable claim embedded in that marketing: that automated platforms can compress credentialing timelines to under thirty days. Comparing vendor marketing claims directly against processing time data published by state medical boards, this report finds that the advertised timeline is achievable in only a narrow subset of states, primarily those participating in the Interstate Medical Licensure Compact, and is mathematically incompatible with published board level processing times in several of the largest states by population, including California and Texas. The report introduces the Koru Controllable Time Ratio, an original framework for separating the portion of a credentialing timeline that automation can genuinely influence from the portion that is structurally controlled by third party institutions automation cannot touch. The analysis finds that the majority of a typical credentialing cycle sits outside any single organization's control, regardless of the software it deploys, and argues that automation's real value lies in eliminating unforced errors rather than compressing a timeline it was never positioned to control.The report presents a blended timeline scenario model showing how an organization's realistic credentialing timeline shifts based on its actual state footprint rather than a single national average, and identifies three genuine, no cost levers: fingerprint method, application timing, and interstate compact strategy. Each meaningfully affects total cycle time without requiring any software purchase. It closes by quantifying the financial exposure created when an organization budgets its staffing and revenue projections around a vendor's advertised timeline rather than its realistic one. The report closes by identifying three structural blind spots that receive little attention in current market discourse: a silent CAQH re attestation lapse that can halt payer access without triggering any alert, a Medicare enrollment gap arising from the false assumption that commercial payer credentialing and federal PECOS enrollment are the same process, and a delegation trapdoor in which an organization's largest available speed advantage, delegated payer credentialing, can be silently revoked by a single audit citation, instantly reverting every provider in the pipeline from a matter of days to a matter of months. All three are genuinely, verifiably solvable by automation, in contrast to the board level processing delays this report shows automation cannot touch, suggesting the market is directing its engineering effort at the wrong problem while three real, addressable ones remain largely unmarketed.","author":[{"family":"Frazier","given":"Marcus"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33289404.v1","URL":"https://doi.org/10.6084/m9.figshare.33289404.v1","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33289404","type":"article-journal","title":"The Speed Illusion: What Healthcare's AI Race Misreads About Credentialing, and the Ninety Days No","abstract":"The healthcare credentialing software market is projected to grow from approximately $3.8 billion in 2026 to more than $10 billion by 2035, driven substantially by vendors marketing artificial intelligence and automation as a solution to slow provider onboarding. This report examines a specific, checkable claim embedded in that marketing: that automated platforms can compress credentialing timelines to under thirty days. Comparing vendor marketing claims directly against processing time data published by state medical boards, this report finds that the advertised timeline is achievable in only a narrow subset of states, primarily those participating in the Interstate Medical Licensure Compact, and is mathematically incompatible with published board level processing times in several of the largest states by population, including California and Texas. The report introduces the Koru Controllable Time Ratio, an original framework for separating the portion of a credentialing timeline that automation can genuinely influence from the portion that is structurally controlled by third party institutions automation cannot touch. The analysis finds that the majority of a typical credentialing cycle sits outside any single organization's control, regardless of the software it deploys, and argues that automation's real value lies in eliminating unforced errors rather than compressing a timeline it was never positioned to control.The report presents a blended timeline scenario model showing how an organization's realistic credentialing timeline shifts based on its actual state footprint rather than a single national average, and identifies three genuine, no cost levers: fingerprint method, application timing, and interstate compact strategy. Each meaningfully affects total cycle time without requiring any software purchase. It closes by quantifying the financial exposure created when an organization budgets its staffing and revenue projections around a vendor's advertised timeline rather than its realistic one. The report closes by identifying three structural blind spots that receive little attention in current market discourse: a silent CAQH re attestation lapse that can halt payer access without triggering any alert, a Medicare enrollment gap arising from the false assumption that commercial payer credentialing and federal PECOS enrollment are the same process, and a delegation trapdoor in which an organization's largest available speed advantage, delegated payer credentialing, can be silently revoked by a single audit citation, instantly reverting every provider in the pipeline from a matter of days to a matter of months. All three are genuinely, verifiably solvable by automation, in contrast to the board level processing delays this report shows automation cannot touch, suggesting the market is directing its engineering effort at the wrong problem while three real, addressable ones remain largely unmarketed.","author":[{"family":"Frazier","given":"Marcus"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33289404","URL":"https://doi.org/10.6084/m9.figshare.33289404","source":"datacite"},{"id":"doi:10.5281/zenodo.22015212","type":"article-journal","title":"From Prediction to Physics: The Last Mile of AI in Biomedicine ——How Physical Probes Bridge the Gap Between Computational Forecasts and Experimental Reality","abstract":"Artificial intelligence has transformed biomedicine. AI models now predict protein structures, design functional proteins, discover drug candidates, and diagnose diseases from medical images with remarkable accuracy. Yet a fundamental problem remains: prediction is not validation. Over 90% of clinical drug candidates still fail. AI-generated molecules often violate physical laws or are infeasible to synthesize. The rate at which AI generates hypotheses in silico far exceeds the capacity to test them in the laboratory. This is the \"last mile\" problem of AI in biomedicine——the chasm between computational prediction and physical confirmation. This review argues that the missing piece is not better AI, but a new class of physical probes designed to validate AI predictions at the molecular level. We survey the state of AI in biomedicine, identify the structural bottlenecks that limit translation from computation to clinic, and propose that physics-based topological probing provides a scalable, interpretable bridge across the last mile. We introduce the String Dynamic Probe Framework and its companion PanProbe static system, which together measure conformational dynamics, hidden pockets, and binding stability independent of force-field parameterization. Calibrated against exact diagonalization (ED) gold standards through a three-stage embedding chain, these probes offer physics-based ground truth for validating AI-generated hypotheses. We conclude by outlining a closed-loop paradigm: AI predicts, physical probes validate, validated data trains the next generation of AI.","author":[{"family":"Li","given":"Genmin"},{"family":"Li","given":"Qizhen"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22015212","URL":"https://doi.org/10.5281/zenodo.22015212","source":"datacite"},{"id":"doi:10.5281/zenodo.22015213","type":"article-journal","title":"From Prediction to Physics: The Last Mile of AI in Biomedicine ——How Physical Probes Bridge the Gap Between Computational Forecasts and Experimental Reality","abstract":"Artificial intelligence has transformed biomedicine. AI models now predict protein structures, design functional proteins, discover drug candidates, and diagnose diseases from medical images with remarkable accuracy. Yet a fundamental problem remains: prediction is not validation. Over 90% of clinical drug candidates still fail. AI-generated molecules often violate physical laws or are infeasible to synthesize. The rate at which AI generates hypotheses in silico far exceeds the capacity to test them in the laboratory. This is the \"last mile\" problem of AI in biomedicine——the chasm between computational prediction and physical confirmation. This review argues that the missing piece is not better AI, but a new class of physical probes designed to validate AI predictions at the molecular level. We survey the state of AI in biomedicine, identify the structural bottlenecks that limit translation from computation to clinic, and propose that physics-based topological probing provides a scalable, interpretable bridge across the last mile. We introduce the String Dynamic Probe Framework and its companion PanProbe static system, which together measure conformational dynamics, hidden pockets, and binding stability independent of force-field parameterization. Calibrated against exact diagonalization (ED) gold standards through a three-stage embedding chain, these probes offer physics-based ground truth for validating AI-generated hypotheses. We conclude by outlining a closed-loop paradigm: AI predicts, physical probes validate, validated data trains the next generation of AI.","author":[{"family":"Li","given":"Genmin"},{"family":"Li","given":"Qizhen"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22015213","URL":"https://doi.org/10.5281/zenodo.22015213","source":"datacite"},{"id":"doi:10.5281/zenodo.20799993","type":"article-journal","title":"Modality Collapse Measurement Framework: Methodology with Pilot Validation (Research Proposal)","abstract":"Multimodal medical imaging systems combine information from multiple imaging sequences (FLAIR, T1, T1CE, T2) to improve diagnostic accuracy and clinical decision-making. Despite their empirical success, deep neural networks trained on multimodal data may progressively converge toward modality-invariant representations, losing the distinct, clinically relevant information contributed by each imaging modality. This phenomenon, referred to as modality collapse, remains poorly understood in medical artificial intelligence systems. This release presents a comprehensive, statistically rigorous framework for quantifying modality collapse in multimodal medical imaging models, evaluated on the BraTS 2021 (FeTS) brain tumor segmentation benchmark. The framework integrates five complementary analytical measures: modality probe classification, centered kernel alignment (CKA), participation ratio (PR), singular vector canonical correlation analysis (SVCCA), and modality ablation studies. Together, these metrics capture different facets of representational similarity, dimensionality, and modality-specific information preservation. To ensure robust and reproducible inference, the framework incorporates bootstrap confidence intervals, permutation testing, false discovery rate correction, effect size estimation, synthetic perturbation experiments, and multiple experimental repetitions. The framework was validated on the BraTS 2021 (FeTS) dataset with 341 subjects, demonstrating end-to-end pipeline functionality, coherent metric behavior across all network layers, and preliminary evidence of progressive modality collapse. This work forms the methodological core of a research proposal aimed at advancing the understanding of representation learning in multimodal medical AI systems. The complete implementation, methodology framework, and pilot validation report are provided with this release. **Can we quantify when multimodal representations stop being modality-specific and become modality-invariant?** This framework is built to answer that question.","author":[{"family":"Paul","given":"Pronab"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20799993","URL":"https://doi.org/10.5281/zenodo.20799993","source":"datacite"},{"id":"doi:10.5281/zenodo.20799994","type":"article-journal","title":"Modality Collapse Measurement Framework: Methodology with Pilot Validation (Research Proposal)","abstract":"Multimodal medical imaging systems combine information from multiple imaging sequences (FLAIR, T1, T1CE, T2) to improve diagnostic accuracy and clinical decision-making. Despite their empirical success, deep neural networks trained on multimodal data may progressively converge toward modality-invariant representations, losing the distinct, clinically relevant information contributed by each imaging modality. This phenomenon, referred to as modality collapse, remains poorly understood in medical artificial intelligence systems. This release presents a comprehensive, statistically rigorous framework for quantifying modality collapse in multimodal medical imaging models, evaluated on the BraTS 2021 (FeTS) brain tumor segmentation benchmark. The framework integrates five complementary analytical measures: modality probe classification, centered kernel alignment (CKA), participation ratio (PR), singular vector canonical correlation analysis (SVCCA), and modality ablation studies. Together, these metrics capture different facets of representational similarity, dimensionality, and modality-specific information preservation. To ensure robust and reproducible inference, the framework incorporates bootstrap confidence intervals, permutation testing, false discovery rate correction, effect size estimation, synthetic perturbation experiments, and multiple experimental repetitions. The framework was validated on the BraTS 2021 (FeTS) dataset with 341 subjects, demonstrating end-to-end pipeline functionality, coherent metric behavior across all network layers, and preliminary evidence of progressive modality collapse. This work forms the methodological core of a research proposal aimed at advancing the understanding of representation learning in multimodal medical AI systems. The complete implementation, methodology framework, and pilot validation report are provided with this release. **Can we quantify when multimodal representations stop being modality-specific and become modality-invariant?** This framework is built to answer that question.","author":[{"family":"Paul","given":"Pronab"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20799994","URL":"https://doi.org/10.5281/zenodo.20799994","source":"datacite"},{"id":"doi:10.17632/g9zfgkz4rr.2","type":"article-journal","title":"Clinical Practice Guideline (CPG) dataset","abstract":"This dataset contains 1,000 structured, context-specific recommendation records derived from 76 English-language Clinical Practice Guidelines (CPGs) published by the Ministry of Health Malaysia, covering 20 clinical specialty categories. It was developed as a CPG-grounded reference dataset for small language model research, instruction tuning, retrieval-augmented generation, clinical guideline retrieval, and recommendation-generation evaluation. The file extraction_prompt.txt contains the complete prompt template used for candidate inclusion decisions, field mapping and structured draft generation. The dataset is provided in JSON Lines (JSONL) format, with one independent record per line. Each record contains three fields: • “instruction”: A standard request for a CPG recommendation. • “input”: A disease or clinical condition and its specific clinical context. • “output”: A concise CPG-grounded recommendation for that context. Example: {\"instruction\": \"Provide standard CPG recommendations.\", \"input\": \"Disease: retinopathy of prematurity\\nClinical context: screening examination\", \"output\": \"CPG Recommendation: Use binocular indirect ophthalmoscopy or an approved imaging pathway by trained personnel and document zone, stage, extent, and plus disease.\"} The records address clinical contexts such as screening, diagnosis, risk assessment, treatment, medication, monitoring, referral, prevention, follow-up, warning signs, and intervention cessation. A disease may appear in multiple records representing different clinical contexts. The source CPGs were obtained from official Malaysian Ministry of Health websites. Dataset preparation involved identifying recommendation-bearing content, assigning the relevant disease and clinical context, converting the content into a consistent instruction–input–output structure, and checking JSONL validity, duplication, relevance, and source consistency. The dataset may be used as an experimental ground-truth reference for model evaluation but has not been independently validated as clinical ground truth by qualified clinical experts. It must not replace the original CPGs, professional medical judgement, diagnosis, or patient-specific treatment decisions. Copyright in the original CPG documents remains with the Malaysian Ministry of Health or the respective rights holders.","author":[{"family":"Ng","given":"Joey"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/g9zfgkz4rr.2","URL":"https://doi.org/10.17632/g9zfgkz4rr.2","source":"datacite"},{"id":"doi:10.17632/g9zfgkz4rr","type":"article-journal","title":"Clinical Practice Guideline (CPG) dataset","abstract":"This dataset contains 1,000 structured, context-specific recommendation records derived from 76 English-language Clinical Practice Guidelines (CPGs) published by the Ministry of Health Malaysia, covering 20 clinical specialty categories. It was developed as a CPG-grounded reference dataset for small language model research, instruction tuning, retrieval-augmented generation, clinical guideline retrieval, and recommendation-generation evaluation. The file extraction_prompt.txt contains the complete prompt template used for candidate inclusion decisions, field mapping and structured draft generation. The dataset is provided in JSON Lines (JSONL) format, with one independent record per line. Each record contains three fields: • “instruction”: A standard request for a CPG recommendation. • “input”: A disease or clinical condition and its specific clinical context. • “output”: A concise CPG-grounded recommendation for that context. Example: {\"instruction\": \"Provide standard CPG recommendations.\", \"input\": \"Disease: retinopathy of prematurity\\nClinical context: screening examination\", \"output\": \"CPG Recommendation: Use binocular indirect ophthalmoscopy or an approved imaging pathway by trained personnel and document zone, stage, extent, and plus disease.\"} The records address clinical contexts such as screening, diagnosis, risk assessment, treatment, medication, monitoring, referral, prevention, follow-up, warning signs, and intervention cessation. A disease may appear in multiple records representing different clinical contexts. The source CPGs were obtained from official Malaysian Ministry of Health websites. Dataset preparation involved identifying recommendation-bearing content, assigning the relevant disease and clinical context, converting the content into a consistent instruction–input–output structure, and checking JSONL validity, duplication, relevance, and source consistency. The dataset may be used as an experimental ground-truth reference for model evaluation but has not been independently validated as clinical ground truth by qualified clinical experts. It must not replace the original CPGs, professional medical judgement, diagnosis, or patient-specific treatment decisions. Copyright in the original CPG documents remains with the Malaysian Ministry of Health or the respective rights holders.","author":[{"family":"Ng","given":"Joey"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/g9zfgkz4rr","URL":"https://doi.org/10.17632/g9zfgkz4rr","source":"datacite"},{"id":"doi:10.6084/m9.figshare.31247142","type":"article-journal","title":"Efficient Ensemble RAB-SVM Framework Using African Buffalo Algorithm for Accurate Glaucoma Detection and Classification","abstract":"Glaucoma is a progressive eye disease that damages the optic nerve and can lead to irreversible blindness. Manual detection from medical images is complex and leads to errors. Machine learning (ML) algorithms offer significant advantages in automatically detecting glaucoma, with Fuzzy c-means clustering, Logistic Regression, Support Vector Machine (SVM), and Deep Learning (DL) being the most commonly used. While these models offer promising results, they often suffer from limitations such as limited generalization to new data and sensitivity to noise and feature imbalance. Hence, to overcome the limitations, this paper proposes an Ensemble Random Adaptive Support Vector based Random African Buffalo (ERAS-RAB) algorithm to detect and classify it. Evaluated using the Glaucoma Fundus Imaging dataset, the proposed system outperforms several baseline methods, achieving an Accuracy of 99.21%, a Precision of 99.09%, a Recall of 98.96%, and an F1-Score of 99.02%, demonstrating its efficacy in early glaucoma diagnosis. This work demonstrates the potential of hybrid ensemble models in medical image analysis, providing a reliable framework for integrating Artificial Intelligence (AI) into ophthalmic diagnostics.","author":[{"family":"Rekha","given":"C"},{"family":"Jayashree","given":"K"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.31247142","URL":"https://doi.org/10.6084/m9.figshare.31247142","source":"datacite"},{"id":"doi:10.6084/m9.figshare.31247142.v1","type":"article-journal","title":"Efficient Ensemble RAB-SVM Framework Using African Buffalo Algorithm for Accurate Glaucoma Detection and Classification","abstract":"Glaucoma is a progressive eye disease that damages the optic nerve and can lead to irreversible blindness. Manual detection from medical images is complex and leads to errors. Machine learning (ML) algorithms offer significant advantages in automatically detecting glaucoma, with Fuzzy c-means clustering, Logistic Regression, Support Vector Machine (SVM), and Deep Learning (DL) being the most commonly used. While these models offer promising results, they often suffer from limitations such as limited generalization to new data and sensitivity to noise and feature imbalance. Hence, to overcome the limitations, this paper proposes an Ensemble Random Adaptive Support Vector based Random African Buffalo (ERAS-RAB) algorithm to detect and classify it. Evaluated using the Glaucoma Fundus Imaging dataset, the proposed system outperforms several baseline methods, achieving an Accuracy of 99.21%, a Precision of 99.09%, a Recall of 98.96%, and an F1-Score of 99.02%, demonstrating its efficacy in early glaucoma diagnosis. This work demonstrates the potential of hybrid ensemble models in medical image analysis, providing a reliable framework for integrating Artificial Intelligence (AI) into ophthalmic diagnostics.","author":[{"family":"Rekha","given":"C"},{"family":"Jayashree","given":"K"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.31247142.v1","URL":"https://doi.org/10.6084/m9.figshare.31247142.v1","source":"datacite"},{"id":"doi:10.17605/osf.io/5tduj","type":"article-journal","title":"From Monitoring to Action: A Scoping Review of Post-Deployment Clinical AI Surveillance, Risk Assessment, and Lifecycle Management","abstract":"Artificial intelligence (AI) is increasingly transitioning from retrospective model development and validation into prospective evaluation, clinical deployment, and integration with routine healthcare workflows. The evidentiary challenge therefore extends beyond whether an AI model performs adequately at the point of development or initial validation. Clinical AI systems operate within dynamic socio-technical environments. Patient populations change; disease prevalence changes; clinical guidelines change; diagnostic and treatment practices evolve; data acquisition systems are modified; electronic health record workflows are redesigned; clinicians adapt their use of AI; software and model versions change; and AI systems may be recalibrated, retrained, or replaced. Consequently, an AI system that demonstrated acceptable performance during development or predeployment validation may not necessarily retain the same safety, effectiveness, equity, or clinical utility after deployment. The FDA explicitly identifies post-market monitoring of AI-enabled medical devices as a regulatory science gap and is developing methods for detecting changes in input data, monitoring output performance, detecting out-of-distribution inputs, monitoring data drift, and evaluating models across multiple clinical sites. Clinical AI evaluation guidance similarly emphasizes that AI systems should be evaluated as complex interventions embedded within clinical workflows rather than as isolated mathematical models. DECIDE-AI, for example, emphasizes live clinical evaluation, safety, clinical utility, and human factors. A 2024 scoping review specifically examined methods for monitoring clinical AI performance: of 39 included sources, only 9 described monitoring methods that had been clinically tested or implemented, and guidance on concrete metrics, thresholds, and statistical approaches was limited. More recent 2026 review-level evidence has broadened the discussion toward post-development robustness, post-deployment monitoring, adaptive updating, and lifecycle governance. That literature suggests that monitoring methods, action thresholds, fairness surveillance, corrective responses, and operational governance remain insufficiently standardized and that mature evidence from activated systems in routine clinical care remains limited. In parallel, an operational literature has begun to emerge: a health-systems scoping review identified only six eligible post-deployment monitoring studies under a narrowly defined search, and a multi-institutional framework has proposed organizing deployment-facing monitoring around system integrity, performance, and impact, explicitly connecting monitoring results to decisions to update, modify, or decommission deployed systems.","author":[{"family":"Yu","given":"Yunguo"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/5tduj","URL":"https://doi.org/10.17605/osf.io/5tduj","source":"datacite"},{"id":"doi:10.5281/zenodo.22004750","type":"article-journal","title":"Dataset: Premise: Environmental metals, though present in many Amyotrophic Lateral Sclerosis anomalous clusters, are debated regarding causality within geographic epistemological data.Hypothesis: If plant derived extracellular vesicles (PDEVs) aerosolized with environmental metals, they could potentially bypass the blood brain barrier via the cribriform plate area.  Given this potential method of delivery, environmental metals cannot be ruled out as causal in sporadic ALS until PDEV delivery of the metals is tested in wet lab experiments, but may potentially explain the confounding data. - PathMap Experiment #000131","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1:https://pathmap.org/viewer.php?id=131 Artificial General Intelligence LLC Claim Evaluated: Premise: Environmental metals, though present in many Amyotrophic Lateral Sclerosis anomalous clusters, are debated regarding causality within geographic epistemological data. Hypothesis: If plant derived extracellular vesicles (PDEVs) aerosolized with environmental metals, they could potentially bypass the blood brain barrier via the cribriform plate area. Given this potential method of delivery, environmental metals cannot be ruled out as causal in sporadic ALS until PDEV delivery of the metals is tested in wet lab experiments, but may potentially explain the confounding data. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Environmental exposures prior to diagnosis, including herbicides and metal dust/fumes, are significantly associated with accelerated ALS progression. The \"lung-brain axis\" indicates that pulmonary pathology and environmental exposure can impact neurological health, possibly via circulating extracellular vesicles. Ciclopirox olamine (CPX) induces TDP-43 cryptic exons through heavy metal toxicity, providing a molecular mechanism linking environmental metal stress to ALS-FTD pathology. There is a statistically significant correlation between the geographic distributions of ALS and Multiple Sclerosis mortality, suggesting shared unknown etiology factors that persist after controlling for race, gender, and latitude. PD-related respiratory control dysfunction involves selective vulnerability of brainstem networks; such vulnerability may also exist in ALS and impact toxin clearance via the glymphatic system. Plant-derived extracellular vesicles (PDEVs) possess cross-barrier delivery potential and represent an emerging class of biotherapeutic carriers, though their natural role as potential \"Trojan horses\" for environmental toxins remains uninvestigated. The multistep pathogenesis hypothesis is challenged by epidemiological data, which aligns more closely with an exponential model of damage accumulation than a simple power-law model. The presence of copper homeostasis disruption in ALS, manifesting as both toxicity and deficiency, creates a vicious cycle that accelerates protein aggregation. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Quantify the binding affinity and sequestration capacity of heavy metals (e.g., Cr-VI) to plant-derived extracellular vesicles in varying pH environments. Perform in vivo biodistribution studies using trace-labeled heavy metals loaded into PDEVs to track CNS deposition via intranasal administration. 📊 Suggested Studies A cohort study assessing the presence of plant-derived exosomal markers in the CNS of patients with sporadic ALS in areas of high metal contamination. A systematic meta-analysis of the geographic colocation of industrial heavy metal pollution with the incidence of sporadic ALS motor neuron degeneration. 📊 Swansons Literature Based Discovery Candidates • Discovered Hypothesis (A to C): PDEV-mediated sequestration of heavy metals provides a bypass mechanism for blood-brain barrier restriction, explaining the CNS-toxicity of environment-borne contaminants. - Literature A (Origin): Heavy metal toxicity and ALS risk (Source: ID 40559965, ID 42021792) - Literature C (Target): PDEV-mediated barrier crossing and drug delivery (Source: ID 42183199, ID 42352265) - The Intersecting Bridge B: Extracellular vesicle surface lipid-protein complexes. - Biological Rationale: Extracellular vesicles, including those from plants, possess lipid bilayer surfaces that can interact with and bind heavy metal ions, and these vesicles are naturally suited to cross biological barriers, serving as","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22004750","URL":"https://doi.org/10.5281/zenodo.22004750","source":"datacite"},{"id":"doi:10.5281/zenodo.22004751","type":"article-journal","title":"Dataset: Premise: Environmental metals, though present in many Amyotrophic Lateral Sclerosis anomalous clusters, are debated regarding causality within geographic epistemological data.Hypothesis: If plant derived extracellular vesicles (PDEVs) aerosolized with environmental metals, they could potentially bypass the blood brain barrier via the cribriform plate area.  Given this potential method of delivery, environmental metals cannot be ruled out as causal in sporadic ALS until PDEV delivery of the metals is tested in wet lab experiments, but may potentially explain the confounding data. - PathMap Experiment #000131","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1:https://pathmap.org/viewer.php?id=131 Artificial General Intelligence LLC Claim Evaluated: Premise: Environmental metals, though present in many Amyotrophic Lateral Sclerosis anomalous clusters, are debated regarding causality within geographic epistemological data. Hypothesis: If plant derived extracellular vesicles (PDEVs) aerosolized with environmental metals, they could potentially bypass the blood brain barrier via the cribriform plate area. Given this potential method of delivery, environmental metals cannot be ruled out as causal in sporadic ALS until PDEV delivery of the metals is tested in wet lab experiments, but may potentially explain the confounding data. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Environmental exposures prior to diagnosis, including herbicides and metal dust/fumes, are significantly associated with accelerated ALS progression. The \"lung-brain axis\" indicates that pulmonary pathology and environmental exposure can impact neurological health, possibly via circulating extracellular vesicles. Ciclopirox olamine (CPX) induces TDP-43 cryptic exons through heavy metal toxicity, providing a molecular mechanism linking environmental metal stress to ALS-FTD pathology. There is a statistically significant correlation between the geographic distributions of ALS and Multiple Sclerosis mortality, suggesting shared unknown etiology factors that persist after controlling for race, gender, and latitude. PD-related respiratory control dysfunction involves selective vulnerability of brainstem networks; such vulnerability may also exist in ALS and impact toxin clearance via the glymphatic system. Plant-derived extracellular vesicles (PDEVs) possess cross-barrier delivery potential and represent an emerging class of biotherapeutic carriers, though their natural role as potential \"Trojan horses\" for environmental toxins remains uninvestigated. The multistep pathogenesis hypothesis is challenged by epidemiological data, which aligns more closely with an exponential model of damage accumulation than a simple power-law model. The presence of copper homeostasis disruption in ALS, manifesting as both toxicity and deficiency, creates a vicious cycle that accelerates protein aggregation. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Quantify the binding affinity and sequestration capacity of heavy metals (e.g., Cr-VI) to plant-derived extracellular vesicles in varying pH environments. Perform in vivo biodistribution studies using trace-labeled heavy metals loaded into PDEVs to track CNS deposition via intranasal administration. 📊 Suggested Studies A cohort study assessing the presence of plant-derived exosomal markers in the CNS of patients with sporadic ALS in areas of high metal contamination. A systematic meta-analysis of the geographic colocation of industrial heavy metal pollution with the incidence of sporadic ALS motor neuron degeneration. 📊 Swansons Literature Based Discovery Candidates • Discovered Hypothesis (A to C): PDEV-mediated sequestration of heavy metals provides a bypass mechanism for blood-brain barrier restriction, explaining the CNS-toxicity of environment-borne contaminants. - Literature A (Origin): Heavy metal toxicity and ALS risk (Source: ID 40559965, ID 42021792) - Literature C (Target): PDEV-mediated barrier crossing and drug delivery (Source: ID 42183199, ID 42352265) - The Intersecting Bridge B: Extracellular vesicle surface lipid-protein complexes. - Biological Rationale: Extracellular vesicles, including those from plants, possess lipid bilayer surfaces that can interact with and bind heavy metal ions, and these vesicles are naturally suited to cross biological barriers, serving as","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22004751","URL":"https://doi.org/10.5281/zenodo.22004751","source":"datacite"},{"id":"doi:10.5281/zenodo.22002645","type":"article-journal","title":"Interaction of KAI1/CD82 Transmembrane Domains with APC Gene Mutations in Cellular Function and Disease Progression","abstract":"This is the 17th hypothesis generated by EPISTEME - An automonous Aritificial Cognitive Process - https://doi.org/10.5281/zenodo.21680339 created by the author. The conserved transmembrane domains of KAI1/CD82 have been implicated in various cellular processes, includingadhesion and migration, which are crucial for cancer progression (Encyclopedia Of Cancer). In parallel, mutations in the APC gene, a key tumor suppressor, are well-documented in several cancers, including colorectal cancer (Encyclopedia of Genetics 10). However, the precise mechanism by which these genetic alterations interact with the structural features of KAI1/CD82 remains unclear. While FISH analysis is commonly used in AML diagnosis to complement chromosome banding analysis, its application in understanding the interaction between KAI1/CD82 and APC mutations is underexplored. Additionally, the role of noncoding RNA in modulating these interactions has not been investigated. This study aims to bridge these two domains by examininghow the conserved transmembrane domains of KAI1/CD82 interact with APC gene mutations to influence cellular function and disease progression. By integrating genetic and molecular analyses, this study addresses a critical gap in understanding the interplay between structural proteins and genetic mutations in cancer biology. V2 - This document presents a scientific feasibility simulation and correction of an autonomously generated hypothesis regarding the interaction between KAI1/CD82 transmembrane domains, APC gene mutations, and noncoding RNA (ncRNA) mediation in cancer progression.The initial hypothesis was created by Episteme - An Artificial Cognitive Process completely autonomously (designed and created by Jason Gabriel Davis). While the initial query identified a novel intersection in oncology, simulation revealed critical method-ological flaws. This document outlines the observations of the initial simulation, identifies errors and gaps, proposes rigorous corrections, and presents the finalized, scientifically viable hypothesis. It concludes with an analysis of the downstream impacts a successful study of this nature would have on targeted therapeutics, diagnostics, and cell biology. The review and corrections were done using GLM5.2 given the lack of human expert involvement despite calls for them to review and give feedback.","author":[{"family":"Davis","given":"Jason"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22002645","URL":"https://doi.org/10.5281/zenodo.22002645","source":"datacite"},{"id":"doi:10.5281/zenodo.21960161","type":"article-journal","title":"Interaction of KAI1/CD82 Transmembrane Domains with APC Gene Mutations in Cellular Function and Disease Progression","abstract":"This is the 17th hypothesis generated by EPISTEME - An automonous Aritificial Cognitive Process - https://doi.org/10.5281/zenodo.21680339 created by the author. The conserved transmembrane domains of KAI1/CD82 have been implicated in various cellular processes, includingadhesion and migration, which are crucial for cancer progression (Encyclopedia Of Cancer). In parallel, mutations in the APC gene, a key tumor suppressor, are well-documented in several cancers, including colorectal cancer (Encyclopedia of Genetics 10). However, the precise mechanism by which these genetic alterations interact with the structural features of KAI1/CD82 remains unclear. While FISH analysis is commonly used in AML diagnosis to complement chromosome banding analysis, its application in understanding the interaction between KAI1/CD82 and APC mutations is underexplored. Additionally, the role of noncoding RNA in modulating these interactions has not been investigated. This study aims to bridge these two domains by examininghow the conserved transmembrane domains of KAI1/CD82 interact with APC gene mutations to influence cellular function and disease progression. By integrating genetic and molecular analyses, this study addresses a critical gap in understanding the interplay between structural proteins and genetic mutations in cancer biology. V2 - This document presents a scientific feasibility simulation and correction of an autonomously generated hypothesis regarding the interaction between KAI1/CD82 transmembrane domains, APC gene mutations, and noncoding RNA (ncRNA) mediation in cancer progression.The initial hypothesis was created by Episteme - An Artificial Cognitive Process completely autonomously (designed and created by Jason Gabriel Davis). While the initial query identified a novel intersection in oncology, simulation revealed critical method-ological flaws. This document outlines the observations of the initial simulation, identifies errors and gaps, proposes rigorous corrections, and presents the finalized, scientifically viable hypothesis. It concludes with an analysis of the downstream impacts a successful study of this nature would have on targeted therapeutics, diagnostics, and cell biology. The review and corrections were done using GLM5.2 given the lack of human expert involvement despite calls for them to review and give feedback.","author":[{"family":"Davis","given":"Jason"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21960161","URL":"https://doi.org/10.5281/zenodo.21960161","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33180176.v5","type":"article-journal","title":"Informational Flow of Quantized Packets of Logic in an Idle Multiplayer Procedural-Generation Omniverse Grid with AI and the Lean Kernel as Proof Assistants","abstract":"This work presents the Loom of Knowledge , a cooperative open-hand game and executable-reasoning architecture built around Quantized Packets of Logic (QPLs) . QPLs encode definitions, assumptions, lemmas, tactics, scientific transformations, evidence, constraints, and certified state transitions as typed, composable cards. Human players and AI Weavers explore a procedurally generated Omniverse Grid , construct and combine QPLs through Runic and string-diagram interfaces, and submit formal mathematical transitions to isolated Lean 4 workers, with final proof validity determined by the Lean kernel.The architecture combines seeded proof search, multiway branching, contradiction-certified pruning, append-only Digital Arrow provenance logs, multiplayer collaboration, Compactification and decompactification of knowledge objects, Theory Bridges, a Runic Type System , an AI-assisted Rune Forge , executable scholarly QPLs, Auto-Play Queues, response chains, activation states, and certificate-first useful work. Gamification is treated as a simplification of the interface rather than a simplification of evidence: compact objects remain inspectable and can be decompactified into their assumptions, dependencies, proof objects, evidence, uncertainty, provenance, and verifier-specific limitations.Cybernetics provides a cross-cutting grammar of sensing, regulation, correction, adaptation, and communication, while individual scientific Fields retain their own ontologies and dynamics. The paper also develops multiscale bridge contracts, scientific Field Spells, physical and biological simulation adapters, AI co-player constraints, procedural theorem generation, and a Lean-to-QPL gamification morphism.This Split Edition focuses specifically on the Loom/QPL systems and game architecture. The detailed Relational Quantum Geometry Dynamics / Covariant Lorentzian Twistor Parent (RQGD/CLTP) quantum-gravity programme that developed as an extended technical stress test of the framework has been separated into a companion manuscript. Accordingly, no claim of completed quantum-gravity unification is made in this work.","author":[{"family":"Lamprou","given":"Georgios"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33180176.v5","URL":"https://doi.org/10.6084/m9.figshare.33180176.v5","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33180176.v6","type":"article-journal","title":"Informational Flow of Quantized Packets of Logic in an Idle Multiplayer Procedural-Generation Omniverse Grid with AI and the Lean Kernel as Proof Assistants","abstract":"This work presents the Loom of Knowledge , a cooperative open-hand game and executable-reasoning architecture built around Quantized Packets of Logic (QPLs) . QPLs encode definitions, assumptions, lemmas, tactics, scientific transformations, evidence, constraints, and certified state transitions as typed, composable cards. Human players and AI Weavers explore a procedurally generated Omniverse Grid , construct and combine QPLs through Runic and string-diagram interfaces, and submit formal mathematical transitions to isolated Lean 4 workers, with final proof validity determined by the Lean kernel.The architecture combines seeded proof search, multiway branching, contradiction-certified pruning, append-only Digital Arrow provenance logs, multiplayer collaboration, Compactification and decompactification of knowledge objects, Theory Bridges, a Runic Type System , an AI-assisted Rune Forge , executable scholarly QPLs, Auto-Play Queues, response chains, activation states, and certificate-first useful work. Gamification is treated as a simplification of the interface rather than a simplification of evidence: compact objects remain inspectable and can be decompactified into their assumptions, dependencies, proof objects, evidence, uncertainty, provenance, and verifier-specific limitations.Cybernetics provides a cross-cutting grammar of sensing, regulation, correction, adaptation, and communication, while individual scientific Fields retain their own ontologies and dynamics. The paper also develops multiscale bridge contracts, scientific Field Spells, physical and biological simulation adapters, AI co-player constraints, procedural theorem generation, and a Lean-to-QPL gamification morphism.This Split Edition focuses specifically on the Loom/QPL systems and game architecture. The detailed Relational Quantum Geometry Dynamics / Covariant Lorentzian Twistor Parent (RQGD/CLTP) quantum-gravity programme that developed as an extended technical stress test of the framework has been separated into a companion manuscript. Accordingly, no claim of completed quantum-gravity unification is made in this work.","author":[{"family":"Lamprou","given":"Georgios"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33180176.v6","URL":"https://doi.org/10.6084/m9.figshare.33180176.v6","source":"datacite"},{"id":"doi:10.6084/m9.figshare.33180176.v4","type":"article-journal","title":"Informational Flow of Quantized Packets of Logic in an Idle Multiplayer Procedural-Generation Omniverse Grid with AI and the Lean Kernel as Proof Assistants","abstract":"This preprint presents The Loom of Knowledge , a cooperative, open-hand game system and executable-reasoning architecture in which definitions, assumptions, lemmas, theorems, tactics, scientific transformations, constraints, evidence objects, and certified state transitions are represented as typed cards called Quantized Packets of Logic (QPLs).The proposed environment is an idle multiplayer procedural-generation Omniverse Grid explored by human players and AI Weavers. Players compose QPL cards through typed input and output interfaces, dependency graphs, response chains, and category-theoretic string diagrams. Candidate proof transitions are submitted to isolated Lean 4 workers. Lean’s elaborator and automated tactics may assist in constructing proof terms, but only the Lean kernel determines whether a formal certificate is valid under the declared definitions, axioms, dependencies, and software environment.The architecture separates mathematical validity from gameplay activation, scheduling priority, visual state, empirical evidence, governance approval, and practical deployment. An accepted proof does not automatically establish that a scientific model is empirically correct, that a policy is morally legitimate, or that a physical intervention is safe. These distinctions are maintained through explicit evidence classes, provenance records, trust boundaries, capability gates, replay requirements, and human-authority controls.The manuscript develops the system from tutorial mode to the Main Game. It defines QPL morphology, activation conditions, Compactification and decompactification, seeded proof search, graph grammars, multiway branching, contradiction-certified pruning, Field Spells, response stacks, observer modes, asynchronous verification, deterministic scheduling, atomic settlement, multiplayer collaboration, and append-only Digital Arrow of Truth logs. A paper can itself be represented as a compactified QPL whose references and prior results form its inputs, whose internal argument provides its transformation structure, and whose claims and reusable artefacts form its outputs.A reference implementation blueprint is proposed using Lean 4 and mathlib for formal verification, Rust and Bevy for the runtime and interactive interface, replaceable AI model services for proposal generation and proof guidance, and transactional infrastructure for jobs, storage, provenance, and settlement. AI systems are treated as search and proposal mechanisms rather than trusted judges. Collaborative drafts may be replicated across participants, but a theorem becomes part of the shared certified Grid only after environment-pinned replay, certificate validation, authorization, and atomic commit.The paper also introduces Logic Mining and Proof of Useful Work as research models for directing computational resources toward proof search, counterexample discovery, certificate compression, library extension, and other reusable scholarly outputs. Verified knowledge is treated as a non-rival asset. The economic model therefore emphasizes reproducibility, dependency reuse, independent verification, attribution, and measurable usefulness rather than computational expenditure or artificial scarcity alone.Cross-domain demonstrations and research bridges include conservation-aware chemistry, macromolecular assembly, UV-induced DNA damage and repair, bioinformatics, robotics, cyber-physical control, climate and energy planning, economics, law, historical reconstruction, scientific digital twins, quantum computing, and formalized decision support. These sections define domain-specific verification and evidence contracts; they do not claim that the Lean kernel alone can establish biological effectiveness, historical certainty, political truth, or physical law.The manuscript also examines category theory, monoidal composition, Hilbert-space representations, quantum transition rates, decoherence, the Quantum Zeno Effect, string-theoretic compactification, AGI","author":[{"family":"Lamprou","given":"Georgios"}],"issued":{"date-parts":[[2026]]},"DOI":"10.6084/m9.figshare.33180176.v4","URL":"https://doi.org/10.6084/m9.figshare.33180176.v4","source":"datacite"},{"id":"doi:10.5281/zenodo.22000544","type":"article-journal","title":"Use of Generative Artificial Intelligence for Consultation Preparation in Shared Decision Making: Can a Handbook Provide Support?","abstract":"Background: Shared Decision Making (SDM) is considered the gold standard of patient-centered care but often fails in clinical routine due to time constraints and a lack of patient preparation. While patients often struggle to articulate their preferences and questions, new developments in the field of Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) offer the potential to bridge this gap. Objective: This paper examines the potential of GenAI as a supportive tool for preparing doctor-patient consultations. The aim is to conceptualize an evidence-based handbook providing patients and physicians with guidance (prompts) to make SDM processes more efficient and personalized. Results: The literature review indicates that GenAI can meaningfully complement traditional decision aids through personalization and interactivity. Specific areas of application include translating complex medical information into patient-friendly language, supporting value clarification, and generating individualized Question Prompt Lists. However, a handbook for using these technologies must address critical risks, particularly \"hallucinations\" (factual errors), bias in training data, and data privacy issues. Effective \"prompt engineering\" is identified as a new key competency for both patients and providers. Conclusion: A structured handbook for the use of GenAI has the potential to reduce asymmetry in the doctor-patient relationship and increase consultation efficiency. However, prerequisites for implementation include strict safety mechanisms, consideration of health literacy, and ethical validation of AI outputs.","author":[{"family":"Wittal","given":"Cornelius"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22000544","URL":"https://doi.org/10.5281/zenodo.22000544","source":"datacite"},{"id":"doi:10.5281/zenodo.18452045","type":"article-journal","title":"Use of Generative Artificial Intelligence for Consultation Preparation in Shared Decision Making: Can a Handbook Provide Support?","abstract":"Background: Shared Decision Making (SDM) is considered the gold standard of patient-centered care but often fails in clinical routine due to time constraints and a lack of patient preparation. While patients often struggle to articulate their preferences and questions, new developments in the field of Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) offer the potential to bridge this gap. Objective: This paper examines the potential of GenAI as a supportive tool for preparing doctor-patient consultations. The aim is to conceptualize an evidence-based handbook providing patients and physicians with guidance (prompts) to make SDM processes more efficient and personalized. Results: The literature review indicates that GenAI can meaningfully complement traditional decision aids through personalization and interactivity. Specific areas of application include translating complex medical information into patient-friendly language, supporting value clarification, and generating individualized Question Prompt Lists. However, a handbook for using these technologies must address critical risks, particularly \"hallucinations\" (factual errors), bias in training data, and data privacy issues. Effective \"prompt engineering\" is identified as a new key competency for both patients and providers. Conclusion: A structured handbook for the use of GenAI has the potential to reduce asymmetry in the doctor-patient relationship and increase consultation efficiency. However, prerequisites for implementation include strict safety mechanisms, consideration of health literacy, and ethical validation of AI outputs.","author":[{"family":"Wittal","given":"Cornelius"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18452045","URL":"https://doi.org/10.5281/zenodo.18452045","source":"datacite"},{"id":"doi:10.5281/zenodo.21997415","type":"article-journal","title":"Artificial Intelligence in Early Disease Detection: Current Applications, Challenges, and Future Opportunities","abstract":"This independent undergraduate literature review examines the current applications and future opportunities of artificial intelligence (AI) in early disease detection. It explores the use of AI technologies in medical imaging, cancer detection, cardiovascular disease, neurological disorders, and other areas of healthcare. The review discusses how AI can support earlier diagnosis, improve diagnostic accuracy, and assist healthcare professionals while also considering current challenges, limitations, ethical considerations, data privacy, and future opportunities. This work is intended for educational and academic portfolio purposes and has not undergone formal peer review or journal publication.","author":[{"family":"Kalloub","given":"Dina"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21997415","URL":"https://doi.org/10.5281/zenodo.21997415","source":"datacite"},{"id":"doi:10.5281/zenodo.21997414","type":"article-journal","title":"Artificial Intelligence in Early Disease Detection: Current Applications, Challenges, and Future Opportunities","abstract":"This independent undergraduate literature review examines the current applications and future opportunities of artificial intelligence (AI) in early disease detection. It explores the use of AI technologies in medical imaging, cancer detection, cardiovascular disease, neurological disorders, and other areas of healthcare. The review discusses how AI can support earlier diagnosis, improve diagnostic accuracy, and assist healthcare professionals while also considering current challenges, limitations, ethical considerations, data privacy, and future opportunities. This work is intended for educational and academic portfolio purposes and has not undergone formal peer review or journal publication.","author":[{"family":"Kalloub","given":"Dina"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21997414","URL":"https://doi.org/10.5281/zenodo.21997414","source":"datacite"},{"id":"doi:10.5281/zenodo.20731629","type":"article-journal","title":"Artificial Intelligence in Medicine Market","abstract":"Artificial Intelligence (AI) is revolutionizing the healthcare industry by enabling advanceddiagnostic, predictive, and management capabilities across medical domains. Thiscomprehensive paper examines the AI in medicine market, focusing on its globallandscape, adoption trends, and practical impacts through a substantial literature review.The research sets clear objectives and hypotheses, applies a structured methodology forprimary data collection from 50 respondents, and discusses the resulting insights. Thefindings highlight drivers, barriers, and future prospects in AI-assisted medicine, with acritical view on market evolution, stakeholder readiness, and anticipated benefits","author":[{"family":"Dalyop","given":"Christopher"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20731629","URL":"https://doi.org/10.5281/zenodo.20731629","source":"datacite"},{"id":"doi:10.5281/zenodo.20731630","type":"article-journal","title":"Artificial Intelligence in Medicine Market","abstract":"Artificial Intelligence (AI) is revolutionizing the healthcare industry by enabling advanceddiagnostic, predictive, and management capabilities across medical domains. Thiscomprehensive paper examines the AI in medicine market, focusing on its globallandscape, adoption trends, and practical impacts through a substantial literature review.The research sets clear objectives and hypotheses, applies a structured methodology forprimary data collection from 50 respondents, and discusses the resulting insights. Thefindings highlight drivers, barriers, and future prospects in AI-assisted medicine, with acritical view on market evolution, stakeholder readiness, and anticipated benefits","author":[{"family":"Dalyop","given":"Christopher"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20731630","URL":"https://doi.org/10.5281/zenodo.20731630","source":"datacite"},{"id":"doi:10.5281/zenodo.21996430","type":"article-journal","title":"Dataset: Targeted alteration of the oral microbiome via a single polyphenol- and fiber-dense meal plan creates an immediate shift in the microbial composition of involuntary nocturnal microaspiration droplets. Upon entry into the lower respiratory tract, this eubiotic bacterial influx acts as an acute molecular signal that rapidly modulates microglia reactivity and neuroinflammation via the lung-brain axis, bypassing systemic colonic metabolite transport. - PathMap Experiment #000130","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=130 Artificial General Intelligence LLC Claim Evaluated: Targeted alteration of the oral microbiome via a single polyphenol- and fiber-dense meal plan creates an immediate shift in the microbial composition of involuntary nocturnal microaspiration droplets. Upon entry into the lower respiratory tract, this eubiotic bacterial influx acts as an acute molecular signal that rapidly modulates microglia reactivity and neuroinflammation via the lung-brain axis, bypassing systemic colonic metabolite transport. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights The lung microbiome includes bacteria, archaea, fungi, protozoa, and viruses. However, fungi and viruses have not been fully studied compared to bacteria in the lungs. The gut-lung-brain (GLB) axis is a multidirectional communication network linking the gastrointestinal tract, respiratory system, and central nervous system (CNS) through neural, endocrine, and immune pathways. Emerging evidence suggests that tryptophan (Trp) metabolism serves as a key integrating node within this axis, modulating host-microbe interactions involved in systemic homeostasis. Intratracheal transplantation of lung microbiota from anxiety-susceptible donors induced similar behavioral changes in recipient mice, indicating a causal role of the pulmonary microbiota. Transcriptomic and immunofluorescence analyses suggested that formononetin acts through modulation of hippocampal microglia. Epidemiological and clinical evidence shows a close association between compromised lung health-including chronic obstructive pulmonary disease (COPD), asthma, obstructive sleep apnea (OSA), and pulmonary infections-and cognitive impairment and dementia. In a recent issue of Nature, Hosang et al. demonstrate how the lung microbiome regulates the magnitude of autoimmune inflammation in the brain. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Test the effect of high-polyphenol acute dietary intake on the salivary and nocturnal oropharyngeal microbial composition in human volunteers. Evaluate the impact of controlled micro-aspiration of specific oral taxa on hippocampal microglia activation in an animal model. 📊 Suggested Studies A longitudinal study mapping the temporal correlation between oral microbiome fluctuation and pulmonary microbiome composition in subjects prone to nocturnal micro-aspiration. Investigate whether dietary modulation of the oral cavity can mitigate neuroinflammation in animal models of lung-brain axis-associated diseases. 📊 Swansons Literature Based Discovery Candidates Modulation of the oral microbiome through rapid dietary shifts can serve as a non-systemic prophylactic intervention to prevent pulmonary-induced microglial overactivation in patients at risk for micro-aspiration-related neurological decline. Oral-pulmonary axis (36768494: Oral-lung seeding via micro-aspiration). Microglia reactivity and lung-brain axis modulation (41981595: Sevoflurane-induced pulmonary dysbiosis and microglial activation). Microglial reactivity/activation. Since oral bacteria form the lung microbiome and pulmonary microbes influence microglia, transiently adjusting the oral community via diet could functionally 'program' the aspirations that reach the lung, thereby pre-empting or attenuating neuroinflammatory signaling without relying on systemic colonic metabolic feedback. 📊 Contradictions Between Evidences There is no direct conflict in the evidence; the claim is simply a novel synthesis of disparate fields (oral-lung seeding and pulmonary-induced neuroinflammation) that has not been explicitly tested or confirmed in the provided literature. 📊 Repurposed Solutions The tann","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21996430","URL":"https://doi.org/10.5281/zenodo.21996430","source":"datacite"},{"id":"doi:10.5281/zenodo.21996431","type":"article-journal","title":"Dataset: Targeted alteration of the oral microbiome via a single polyphenol- and fiber-dense meal plan creates an immediate shift in the microbial composition of involuntary nocturnal microaspiration droplets. Upon entry into the lower respiratory tract, this eubiotic bacterial influx acts as an acute molecular signal that rapidly modulates microglia reactivity and neuroinflammation via the lung-brain axis, bypassing systemic colonic metabolite transport. - PathMap Experiment #000130","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=130 Artificial General Intelligence LLC Claim Evaluated: Targeted alteration of the oral microbiome via a single polyphenol- and fiber-dense meal plan creates an immediate shift in the microbial composition of involuntary nocturnal microaspiration droplets. Upon entry into the lower respiratory tract, this eubiotic bacterial influx acts as an acute molecular signal that rapidly modulates microglia reactivity and neuroinflammation via the lung-brain axis, bypassing systemic colonic metabolite transport. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights The lung microbiome includes bacteria, archaea, fungi, protozoa, and viruses. However, fungi and viruses have not been fully studied compared to bacteria in the lungs. The gut-lung-brain (GLB) axis is a multidirectional communication network linking the gastrointestinal tract, respiratory system, and central nervous system (CNS) through neural, endocrine, and immune pathways. Emerging evidence suggests that tryptophan (Trp) metabolism serves as a key integrating node within this axis, modulating host-microbe interactions involved in systemic homeostasis. Intratracheal transplantation of lung microbiota from anxiety-susceptible donors induced similar behavioral changes in recipient mice, indicating a causal role of the pulmonary microbiota. Transcriptomic and immunofluorescence analyses suggested that formononetin acts through modulation of hippocampal microglia. Epidemiological and clinical evidence shows a close association between compromised lung health-including chronic obstructive pulmonary disease (COPD), asthma, obstructive sleep apnea (OSA), and pulmonary infections-and cognitive impairment and dementia. In a recent issue of Nature, Hosang et al. demonstrate how the lung microbiome regulates the magnitude of autoimmune inflammation in the brain. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Test the effect of high-polyphenol acute dietary intake on the salivary and nocturnal oropharyngeal microbial composition in human volunteers. Evaluate the impact of controlled micro-aspiration of specific oral taxa on hippocampal microglia activation in an animal model. 📊 Suggested Studies A longitudinal study mapping the temporal correlation between oral microbiome fluctuation and pulmonary microbiome composition in subjects prone to nocturnal micro-aspiration. Investigate whether dietary modulation of the oral cavity can mitigate neuroinflammation in animal models of lung-brain axis-associated diseases. 📊 Swansons Literature Based Discovery Candidates Modulation of the oral microbiome through rapid dietary shifts can serve as a non-systemic prophylactic intervention to prevent pulmonary-induced microglial overactivation in patients at risk for micro-aspiration-related neurological decline. Oral-pulmonary axis (36768494: Oral-lung seeding via micro-aspiration). Microglia reactivity and lung-brain axis modulation (41981595: Sevoflurane-induced pulmonary dysbiosis and microglial activation). Microglial reactivity/activation. Since oral bacteria form the lung microbiome and pulmonary microbes influence microglia, transiently adjusting the oral community via diet could functionally 'program' the aspirations that reach the lung, thereby pre-empting or attenuating neuroinflammatory signaling without relying on systemic colonic metabolic feedback. 📊 Contradictions Between Evidences There is no direct conflict in the evidence; the claim is simply a novel synthesis of disparate fields (oral-lung seeding and pulmonary-induced neuroinflammation) that has not been explicitly tested or confirmed in the provided literature. 📊 Repurposed Solutions The tann","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21996431","URL":"https://doi.org/10.5281/zenodo.21996431","source":"datacite"},{"id":"doi:10.64898/2026.07.23.26358815","type":"article-journal","title":"Evaluative Stance Toward Artificial Intelligence in High-Quartile Medical Journals (2021– 2026): Large-Scale LLM-Assisted Computational Content Analysis","abstract":"Abstract Background Medical-AI publications do more than report technical performance; they also frame AI as beneficial, uncertain, or risky. How this evaluative stance has changed across the medical literature is not well characterized. Objective To characterize evaluative stance in published medical-AI discourse abstracts from January 2021 through April 2026 and examine variation over time, concern themes, failure mechanisms, specialties, first-author geography, and publication format. Methods We conducted an LLM-assisted computational content analysis of medical-AI abstracts from first- and second-quartile medical journals. Of 97,492 post-cutoff Q1/Q2 records entering the prefilter, 16,759 were retained as discourse or evaluative. Claude Sonnet 4.6 assigned 16,749 valid stance classifications using Alarm, Caution, Neutral, Cautious Optimism, and Advocacy. Annual analyses used 16,747 records dated 2021–2026. Critical stance was Alarm plus Caution and indexed evaluative scrutiny, rather than opposition or author psychology. Each LLM step was validated against blinded human coding by one author: prefilter Cohen κ=0.51, stance quadratic-weighted κ=0.79 (95% CI 0.72–0.84) for codable, in-scope records, specialty κ=0.75, and mechanism axes κ=0.84 (model type) and κ=0.57 (failure mode). Results Advocacy declined from 2.9% in 2021 to 0.6% in partial 2026, while Cautious Optimism remained the majority stance. Among 16,749 valid classifications, 30.8% were critical. Critical share increased from 25.4% to 32.6%, a 7.25-percentage-point increase based on unrounded estimates. Among critical records, patient safety remained the most prevalent concern. Hallucination/errors increased by 30.9 percentage points. Regulation declined by 22.0 percentage points and ethics/bias by 8.1 percentage points in prevalence share; these declines do not necessarily indicate lower publication counts. Within the hallucination/error theme, factual error was more common than fabrication. Fabrication estimates should be treated as an upper bound, because failure-mode agreement was moderate. Specialty patterns were heterogeneous. Critical rate was inversely associated with FDA cleared-device availability (Spearman ρ=−0.65, two-sided p=0.004), which does not measure adoption, deployment, maturity, or clinical use. First-author geography described publication metadata and discourse, not national attitudes or research quality. Reviews were the least critical and most favourable format. In exploratory forward-validation, 2 of 78 early Advocacy predictions were fully borne out, although the analysis was single-rater and retrieval-dependent. Conclusions Published medical-AI abstracts became modestly less promotional and more focused on specific errors and safety concerns. Unqualified promotion declined, but qualified favourable framing remained dominant, and the rise in critical stance was modest. Concern moved toward errors and patient safety, with factual error discussed more often than fabrication. These findings describe published discourse, not AI capability or whether the evaluations were correct.","author":[{"family":"Wang","given":"Lixing"},{"family":"Poenaru","given":"D"}],"issued":{"date-parts":[[2026]]},"DOI":"10.64898/2026.07.23.26358815","URL":"https://doi.org/10.64898/2026.07.23.26358815","source":"preprints"},{"id":"oa:W4407010228","type":"article-journal","title":"Comparison of artificial intelligence systems in answering prosthodontics questions from the dental specialty exam in Turkey","abstract":": Artificial intelligence (AI) is increasingly vital in dentistry, supporting diagnostics, treatment planning, and patient education. However, AI systems face challenges, especially in delivering accurate information within specialized dental fields. This study aimed to evaluate the performance of seven AI-based chatbots (ChatGPT-3.5, ChatGPT-4, Gemini, Gemini Advanced, Claude AI, Microsoft Copilot, and Smodin AI) in correctly answering prosthodontics questions from the Dental Specialty Exam (DUS) in Turkey. Materials and methods: The dataset for this study consists of 128 multiple-choice prosthodontics questions from the DUS, a national exam administered in Turkey by the Student Selection and Placement Center (ÖSYM) between 2012 and 2021. Chatbot performance was assessed by categorizing the questions into case-based and knowledge-based. Results: ChatGPT-4 achieved the highest accuracy (75.8 %), while Gemini AI had the lowest (46.1 %). Gemini AI also had more incorrect (69) than correct answers (59). ChatGPT-4 and ChatGPT-3.5 showed significantly higher accuracy in knowledge-based questions compared to case-based ones (p < 0.05). For case-based questions, Gemini and Gemini Advanced had the lowest accuracy (36.4 %), while other chatbots averaged 45.5 %. In knowledge-based questions, ChatGPT-4 performed best (78.6 %) and Gemini AI the worst (47 %). Conclusion: ChatGPT-4 excelled in knowledge-based prosthodontic questions, showing potential to enhance dental education through personalized learning and clinical reasoning support. However, its limitations in case-based scenarios highlight the need for optimization to better address complex clinical situations. These findings suggest that AI models can significantly contribute to dental education and clinical practice.","author":[{"family":"Tosun","given":"Büşra"},{"family":"Yilmaz","given":"Zeynep"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jds.2025.01.025","URL":"https://doi.org/10.1016/j.jds.2025.01.025","source":"openalex"},{"id":"oa:W4406222743","type":"article-journal","title":"Integrating Artificial Intelligence, Internet of Things, and Sensor-Based Technologies: A Systematic Review of Methodologies in Autism Spectrum Disorder Detection","abstract":"This paper presents a systematic review of the emerging applications of artificial intelligence (AI), Internet of Things (IoT), and sensor-based technologies in the diagnosis of autism spectrum disorder (ASD). The integration of these technologies has led to promising advances in identifying unique behavioral, physiological, and neuroanatomical markers associated with ASD. Through an examination of recent studies, we explore how technologies such as wearable sensors, eye-tracking systems, virtual reality environments, neuroimaging, and microbiome analysis contribute to a holistic approach to ASD diagnostics. The analysis reveals how these technologies facilitate non-invasive, real-time assessments across diverse settings, enhancing both diagnostic accuracy and accessibility. The findings underscore the transformative potential of AI, IoT, and sensor-based driven tools in providing personalized and continuous ASD detection, advocating for data-driven approaches that extend beyond traditional methodologies. Ultimately, this review emphasizes the role of technology in improving ASD diagnostic processes, paving the way for targeted and individualized assessments.","author":[{"family":"Bouchouras","given":"Georgios"},{"family":"Kotis","given":"Konstantinos"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/a18010034","URL":"https://doi.org/10.3390/a18010034","source":"openalex"},{"id":"oa:W4411124274","type":"article-journal","title":"Healthcare workers' readiness for artificial intelligence and organizational change: a quantitative study in a university hospital","abstract":"OBJECTIVE: The aim of the study is to measure the readiness levels of medical artificial intelligence and the perception of openness to organizational change of healthcare professionals working in a university hospital in Istanbul. Additionally, the study seeks to identify the relationships between medical AI readiness and perceptions of organizational change openness, as well as to examine differences based on demographic variables. METHOD: The research was conducted with 195 healthcare workers. The research is a cross-sectional descriptive quantitative research. The construct validity of the scales was checked using statistical analysis. RESULT: As a result of the research, it was determined that healthcare workers' are prepared for the use of medical artificial intelligence in healthcare institutions and perceive organizational change positively. A significant but low-level positive relationship was found between healthcare workers' level of readiness for medical artificial intelligence and their perception of openness to organizational change. The level of readiness for medical artificial intelligence among healthcare workers' was found to be high among males, doctors and internal sciences, while the perception of openness to organizational change was found to be high among postgraduate/doctoral graduates, surgical sciences, nurses. CONCLUSION: The study determined that healthcare workers' are ready to use medical artificial intelligence and perceive organizational change positively. The study contributes to the formation of the institution's healthcare policies and practices and to the development, well-being and change of healthcare workers'. It is recommended that employees be made aware of the benefits of using artificial intelligence in healthcare institutions and that necessary training activities be planned.","author":[{"family":"Boyacı","given":"Hafize"},{"family":"Söyük","given":"Selma"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12913-025-12846-y","URL":"https://doi.org/10.1186/s12913-025-12846-y","source":"openalex"},{"id":"oa:W4412698288","type":"article-journal","title":"The Effectiveness of Artificial Intelligence-Based Interventions for Students with Learning Disabilities: A Systematic Review","abstract":"Background/Objectives: While artificial intelligence (AI) is rapidly transforming education, its specific effectiveness for students with learning disabilities (LD) requires rigorous evaluation. This systematic review aims to assess the efficacy of AI-based educational interventions for students with LD, with a specific focus on the methodological quality and risk of bias of the available evidence. Methods: A systematic search was conducted across seven major databases (Google Scholar, ScienceDirect, APA PsycInfo, ERIC, Scopus, PubMed) for experimental studies published between 2022 and 2025. This review followed PRISMA guidelines, using the PICOS framework for inclusion criteria. A formal risk of bias assessment was performed using the ROBINS-I and JBI critical appraisal tools. Results: Eleven studies (representing 10 independent experiments), encompassing 3033 participants, met the inclusion criteria. The most studied disabilities were dyslexia (six studies) and other specific learning disorders (three studies). Personalized/adaptive learning systems and game-based learning were the most common AI interventions. All 11 studies reported positive outcomes. However, the risk of bias assessment revealed significant methodological limitations: no studies were rated as having a low risk of bias, with most presenting a moderate (70%) to high/serious (30%) risk. Despite these limitations, quantitative results from the stronger studies showed large effect sizes, such as in arithmetic fluency (d = 1.63) and reading comprehension (d = −1.66). Conclusions: AI-based interventions demonstrate significant potential for supporting students with learning disabilities, with unanimously positive reported outcomes. However, this conclusion must be tempered by the considerable risk of bias and methodological weaknesses prevalent in the current literature. The limited and potentially biased evidence base warrants cautious interpretation. Future research must prioritize high-quality randomized controlled trials (RCTs) and longitudinal assessments to establish a definitive evidence base and investigate long-term effects, including the risk of cognitive offloading.","author":[{"family":"Paglialunga","given":"A"},{"family":"Melogno","given":"Sergio"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/brainsci15080806","URL":"https://doi.org/10.3390/brainsci15080806","source":"openalex"},{"id":"oa:W4411961335","type":"article-journal","title":"From Lab to Clinic: How Artificial Intelligence (AI) Is Reshaping Drug Discovery Timelines and Industry Outcomes","abstract":"Background/Objectives: Artificial intelligence (AI) is transforming drug discovery and development by enhancing the speed and precision of identifying drug candidates and optimizing their efficacy. This review evaluates the application of AI in various stages of drug discovery, from hit identification to lead optimization, and its impact on clinical outcomes. The objective is to provide insights into the role of AI across therapeutic areas and assess its contributions to improving clinical trial efficiency and pharmaceutical outcomes. Methods: A systematic review followed PRISMA guidelines to analyze studies published between 2015 and 2025, focusing on AI in drug discovery and development. A comprehensive search was performed across multiple databases to identify studies employing AI techniques. The studies were categorized based on AI methods, clinical phase, and therapeutic area. The percentages of AI methods used, clinical phase stages, and the therapeutic regions were analyzed to identify trends. Results: AI methods included machine learning (ML) at 40.9%, molecular modeling and simulation (MMS) at 20.7%, and deep learning (DL) at 10.3%. Oncology accounted for the majority of studies (72.8%), followed by dermatology (5.8%) and neurology (5.2%). In clinical phases, 39.3% of studies were in the preclinical stage, 23.1% in Clinical Phase I, and 11.0% in the transitional phase. Clinical outcome reporting was observed in 45% of studies, with 97% reporting industry partnerships. Conclusions: AI significantly enhances drug discovery and development, improving drug efficacy and clinical trial outcomes. Future work should focus on expanding AI applications into underrepresented therapeutic areas and refining models to handle complex biological systems.","author":[{"family":"Dermawan","given":"Doni"},{"family":"Alotaiq","given":"Nasser"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ph18070981","URL":"https://doi.org/10.3390/ph18070981","source":"openalex"},{"id":"oa:W7117448835","type":"article-journal","title":"Artificial intelligence and learner autonomy: a meta-analysis of self-regulated and self-directed learning","abstract":"Introduction As artificial intelligence (AI) becomes increasingly embedded in educational environments, understanding its role in shaping learners’ self-regulated learning (SRL) and self-directed learning (SDL) has emerged as a central concern in contemporary learning science. While prior studies suggest that AI-driven systems may support planning, monitoring, and autonomy in learning, empirical evidence remains fragmented across contexts, learner groups, and instructional designs. This study synthesizes existing empirical research to systematically examine the magnitude and conditions under which AI-based interventions influence SRL, its dimensions and phases, SDL, and associated learning outcomes. Methods A systematic meta-analysis was conducted following PRISMA guidelines, synthesizing evidence from 32 empirical studies comprising 92 effect sizes and a total of 3,029 participants. The analysis examined overall effects of AI-based interventions on SRL and SDL, disaggregated effects across SRL dimensions (cognitive/metacognitive, motivational/affective, and behavioral regulation) and SRL phases (forethought, performance, and self-reflection), as well as impacts on learning outcomes and academic achievement. Random-effects models were applied, and moderator analyses explored learner characteristics, contextual variables, and AI design features. Sensitivity analyses and publication bias assessments were performed to evaluate the robustness of findings. Results AI-based interventions demonstrated a large and statistically significant positive effect on overall SRL ( g = 1.613, p = 0.032) and SDL ( g = 1.111, p = 0.043), indicating substantial improvements in learners’ ability to plan, monitor, and regulate their learning while sustaining autonomy and persistence. At the dimensional level, AI produced moderate gains in cognitive/metacognitive regulation ( g = 0.377, p = 0.0004) and motivational/affective regulation ( g = 0.505, p = 0.013), whereas effects on behavioral regulation were inconsistent. Phase-level analyses revealed that AI interventions were most effective during the forethought phase, supporting goal setting, planning, and motivational readiness, with smaller but significant gains observed in self-reflection and variable effects during the performance phase. AI systems also yielded moderate improvements in learning outcomes and achievement ( g = 0.350, p = 0.034). Moderator analyses indicated stronger SRL effects among older learners, longer intervention durations, and language learning contexts employing interactive AI systems, while gender differences were minimal. Sensitivity and publication bias tests confirmed the stability of results. Discussion The findings indicate that AI functions as an adaptive scaffold that meaningfully enhances learners’ self-regulatory and self-directed capacities across cognitive, motivational, and reflective processes. By strengthening forethought and planning mechanisms in particular, AI-based interventions support more autonomous, sustained, and effective learning behaviors that translate into measurable academic benefits. Variability in behavioral regulation outcomes highlights the need for more explicit action-level supports in AI design. Overall, the results showcase AI’s potential to promote equitable and scalable self-regulated learning across diverse educational contexts, while also pointing to the importance of aligning intervention design with learner characteristics and instructional goals.","author":[{"family":"Achuthan","given":"Krishnashree"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/feduc.2025.1738751","URL":"https://doi.org/10.3389/feduc.2025.1738751","source":"openalex"},{"id":"oa:W4409461213","type":"article-journal","title":"Educators’ Perceptions on Artificial Intelligence in Higher Education: Insights from the Jordanian Higher Education","abstract":"This paper investigates educators’ perceptions of the application of Artificial Intelligence in Higher Education (AIHEd) and its benefits and concerns within the Jordanian higher education. Like in other contexts, the adoption of Artificial Intelligence (AI) in the Jordanian higher education brought many benefits and a variety of concerns. Due to the lack of regulations and clear policies to cope with such new technologies, the increasing prevalence of these concerns has a negative impact on academic integrity. We used a sequential exploratory mixed approach to accomplish our study, which is guided by the Technology Acceptance Model (TAM), which helps in analysing the adoption of AI in higher education. Our approach involves conducting interviews with university educators from three different Jordanian universities. Interviews were done to identify educators’ thoughts regarding the responsibility of universities to adopt new AI technologies, what motivates them to use AI tools and services in their daily work, whether using AI in higher education institutions is legitimate, and the concerns associated with implementing such technologies into practice. Thus, the paper tries to portray the acceptable benefits and concerns of using AI in Jordanian higher education institutions. After conducting a thematic analysis on 18 interviews with educators, we identified 10 corresponding benefit themes and 8 corresponding concern themes that resulted from the coding and theme-building process. The average rate of educators’ responses to the themes of benefits and concerns is then determined by distributing a questionnaire to 145 higher education educators to generalise the results. Although our findings offer valuable insights, further investigation in wider contexts may be necessary to ensure the representativeness and generalisability of the findings. Through the themes that the study outlined, we concluded that although AI can transform the way students learn and educators work, there are still several issues that need to be resolved by researchers and teachers who work with associated application systems. Such issues require greater emphasis on appropriately and logically handling related ethical dilemmas. These concerns also highlight the importance of developing the necessary strategies and skills for responsible AIHEd. Using a mixed approach helped us to develop a strong understanding of the current state of AIHEd in the Jordanian context.","author":[{"family":"Al-Qoran","given":"Lamis"}],"issued":{"date-parts":[[2025]]},"DOI":"10.18178/ijiet.2025.15.4.2278","URL":"https://doi.org/10.18178/ijiet.2025.15.4.2278","source":"openalex"},{"id":"oa:W4408365249","type":"article-journal","title":"The Origins and Veracity of References ‘Cited’ by Generative Artificial Intelligence Applications: Implications for the Quality of Responses","abstract":"The public release of ChatGPT in late 2022 has resulted in considerable publicity and has led to widespread discussion of the usefulness and capabilities of generative Artificial intelligence (Ai) language models. Its ability to extract and summarise data from textual sources and present them as human-like contextual responses makes it an eminently suitable tool to answer questions users might ask. Expanding on a previous analysis of the capabilities of ChatGPT3.5, this paper tested what archaeological literature appears to have been included in the training phase of three recent generative Ai language models: ChatGPT4o, ScholarGPT, and DeepSeek R1. While ChatGPT3.5 offered seemingly pertinent references, a large percentage proved to be fictitious. While the more recent model ScholarGPT, which is purportedly tailored towards academic needs, performed much better, it still offered a high rate of fictitious references compared to the general models ChatGPT4o and DeepSeek. Using ‘cloze’ analysis to make inferences on the sources ‘memorized’ by a generative Ai model, this paper was unable to prove that any of the four genAi models had perused the full texts of the genuine references. It can be shown that all references provided by ChatGPT and other OpenAi models, as well as DeepSeek, that were found to be genuine, have also been cited on Wikipedia pages. This strongly indicates that the source base for at least some, if not most, of the data is found in those pages and thus represents, at best, third-hand source material. This has significant implications in relation to the quality of the data available to generative Ai models to shape their answers. The implications of this are discussed.","author":[{"family":"Spennemann","given":"Dirk"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/publications13010012","URL":"https://doi.org/10.3390/publications13010012","source":"openalex"},{"id":"oa:W4406040531","type":"article-journal","title":"Knowledge, attitudes, and perceptions of a group of Egyptian dental students toward artificial intelligence: a cross-sectional study","abstract":"INTRODUCTION: Artificial intelligence (AI) applications have increased dramatically across a wide range of domains. Dental students will undoubtedly be impacted by the emergence of AI in dentistry. AIM: This study aimed to evaluate the knowledge, attitudes, and perceptions of a group of Egyptian dental students toward artificial intelligence. MATERIALS AND METHOD: An online survey was conducted using a questionnaire sent to dental students via Google Forms. The questionnaire comprised 18 questions on participant's knowledge and perceptions regarding the future of AI in dentistry. The collected data was statistically analyzed. RESULTS: A total of 384 students answered the questionnaire. Of the participants, (49%) had a basic knowledge of the principles of AI, and (48%) participants were aware of AI usage in dentistry. Social media was the most common information source for AI applications. Most of the participants agreed on the leading role of AI in the advancement of dentistry and disagreed on the ability of AI to replace dentists in the future, (53%) and (44%) respectively. Moreover, (49%) and (52%) respectively of students approved the incorporation of AI applications in undergraduate and postgraduate dental training. CONCLUSION: Egyptian dental students are acquainted with AI and its possible applications in dentistry. They consider the use of AI diagnosis exciting and approve of its definitive role in disease prediction. There is a necessity to include, enhance, and increase AI training in dental schools. TRIAL REGISTRATION: This study has been registered in clinical trials. gov with an identifier: NCT06348758.","author":[{"family":"Elchaghaby","given":"Marwa"},{"family":"Wahby","given":"Reem"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1186/s12903-024-05282-7","URL":"https://doi.org/10.1186/s12903-024-05282-7","source":"openalex"},{"id":"oa:W4410863629","type":"article-journal","title":"Promoting Critical Thinking in Biological Sciences in the Era of Artificial Intelligence: The Role of Higher Education","abstract":"The integration of artificial intelligence (AI) into the biological sciences marks a transformative era, reshaping research methodologies, data analysis, and hypothesis generation. This technological advancement accelerates discoveries and enhances our understanding of complex biological systems. As AI increasingly influences decision-making processes, the necessity for students and scientists to critically assess AI-generated outputs becomes paramount. The current narrative review explores the evolving role of critical thinking in biological sciences amidst the rise of AI, emphasizing the importance of skepticism, contextual understanding, and ethical considerations. It argues that while AI provides powerful tools for data interpretation and pattern recognition, human oversight and critical analysis remain indispensable to validate findings and prevent biases inherent in automated systems. Higher education institutions play a crucial role in fostering a culture of critical thinking, equipping biological scientists to effectively harness AI technologies while ensuring the integrity of their research and upholding scientific and ethical standards. Furthermore, AI tools, including chatbots, could be strategically employed in active learning methodologies, such as problem-based learning, flipped classrooms, and online learning. These methodologies enhance the ability of students to effectively utilize AI technologies while ensuring the rigor of scientific research. In conclusion, the current review underscores the benefits, challenges, and educational implications of AI integration, offering actionable insights for educators and learners seeking to adapt effectively to this rapidly evolving technological landscape.","author":[{"family":"Papaneophytou","given":"Christos"},{"family":"Nicolaou","given":"Stella"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/higheredu4020024","URL":"https://doi.org/10.3390/higheredu4020024","source":"openalex"},{"id":"oa:W4408550997","type":"article-journal","title":"Artificial Intelligence in Cardiovascular Imaging and Interventional Cardiology: Emerging Trends and Clinical Implications","abstract":"Artificial intelligence (AI) has revolutionized the field of cardiovascular imaging, serving as a unifying force that brings together multiple modalities under a single platform. The utility of noninvasive imaging ranges from diagnostic assessment and guiding interventions to prognostic stratification. Multimodality imaging has demonstrated important potential, particularly in patients with heterogeneous diseases, such as heart failure and atrial fibrillation. Facilitating complex interventional procedures requires accurate image acquisition and interpretation along with precise decision-making. The unique nature of interventional cardiology procedures benefiting from different imaging modalities presents an ideal target for the development of AI-assisted decision-making tools to improve workflow in the catheterization laboratory and personalize the need for transcatheter interventions. This review explores the advancements of AI in noninvasive cardiovascular imaging and interventional cardiology, addressing the clinical use and challenges of current imaging modalities, emerging trends, and promising applications as well as considerations for safe implementation of AI tools in clinical practice. Current practice has moved well beyond the question of whether we should or should not use AI in clinical health care settings. AI, in all its forms, has become deeply embedded in clinical workflows, particularly in cardiovascular imaging and interventional cardiology. It can, in the future, not only add precision and quantification but also serve as a means by which to fuse and link multimodalities together. It is only by understanding how AI techniques work, that the field can be harnessed for the greater good and avoid uninformed bias or misleading diagnoses.","author":[{"family":"Alsharqi","given":"Maryam"},{"family":"Edelman","given":"Elazer"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.jscai.2024.102558","URL":"https://doi.org/10.1016/j.jscai.2024.102558","source":"openalex"},{"id":"oa:W4406438340","type":"article-journal","title":"Hospital Artificial Intelligence/Machine Learning Adoption by Neighborhood Deprivation","abstract":"OBJECTIVE: To understand the variation in artificial intelligence/machine learning (AI/ML) adoption across different hospital characteristics and explore how AI/ML is utilized, particularly in relation to neighborhood deprivation. BACKGROUND: AI/ML-assisted care coordination has the potential to reduce health disparities, but there is a lack of empirical evidence on AI's impact on health equity. METHODS: We used linked datasets from the 2022 American Hospital Association Annual Survey and the 2023 American Hospital Association Information Technology Supplement. The data were further linked to the 2022 Area Deprivation Index (ADI) for each hospital's service area. State fixed-effect regressions were employed. A decomposition model was also used to quantify predictors of AI/ML implementation, comparing hospitals in higher versus lower ADI areas. RESULTS: Hospitals serving the most vulnerable areas (ADI Q4) were significantly less likely to apply ML or other predictive models (coef = -0.10, P = 0.01) and provided fewer AI/ML-related workforce applications (coef = -0.40, P = 0.01), compared with those in the least vulnerable areas. Decomposition results showed that our model specifications explained 79% of the variation in AI/ML adoption between hospitals in ADI Q4 versus ADI Q1-Q3. In addition, Accountable Care Organization affiliation accounted for 12%-25% of differences in AI/ML utilization across various measures. CONCLUSIONS: The underuse of AI/ML in economically disadvantaged and rural areas, particularly in workforce management and electronic health record implementation, suggests that these communities may not fully benefit from advancements in AI-enabled health care. Our results further indicate that value-based payment models could be strategically used to support AI integration.","author":[{"family":"Chen","given":"Jie"},{"family":"Yan","given":"Alice"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1097/mlr.0000000000002110","URL":"https://doi.org/10.1097/mlr.0000000000002110","source":"openalex"},{"id":"oa:W4412516141","type":"article-journal","title":"A Systematic Review of Artificial Intelligence (AI) and Machine Learning (ML) in Pharmaceutical Supply Chain (PSC) Resilience: Current Trends and Future Directions","abstract":"The resilience of the pharmaceutical supply chain (PSC) is crucial to ensuring the availability of medical products. However, increasing complexity and logistical bottlenecks have exposed weaknesses within PSC frameworks. These challenges underscore the urgent need for more resilient and intelligent supply chain solutions. Recently, Artificial Intelligence and machine learning (AI/ML) have emerged as transformative technologies to enhance PSC resilience. This study presents a systematic review evaluating the role of AI/ML in advancing PSC resilience and their applications across PSC functions. A comprehensive search of five academic databases (Scopus, the Web of Science, IEEE Xplore, PubMed, and EMBASE) identified 89 peer-reviewed studies published between 2019 and 2025. PRISMA 2020 guidelines were implemented, resulting in a final dataset of 32 studies. In addition to analyzing applications, this study identifies the AI/ML grouped into five main categories, providing a clearer understanding of their impact on PSC resilience. The findings reveal that despite AI/ML’s promise, significant research gaps persist. Particularly, AI/ML-driven regulatory compliance and real-time supplier collaboration remain underexplored. Over 59.3% of studies fail to address regulatory frameworks and ethical considerations. In addition, major challenges emerge such as the limited real-world deployment of AI/ML-driven solutions and the lack of managerial impacts on PSC resilience. This study emphasizes the need for stronger regulatory frameworks, broader empirical validation, and AI/ML-driven predictive modeling. This study proposes recommendations for future research to foster more efficient, transparent and ethical PSCs capable of navigating the complexities of global healthcare.","author":[{"family":"Al-Hourani","given":"Shireen"},{"family":"Weraikat","given":"Dua"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/su17146591","URL":"https://doi.org/10.3390/su17146591","source":"openalex"},{"id":"oa:W4412491776","type":"article-journal","title":"Physical artificial intelligence in nursing: Robotics","abstract":"BACKGROUND: Robotics, driven by advancements in physical artificial intelligence (AI), offers potential solutions-yet many challenges- to creating innovative care models to meet the needs of the future. PURPOSE: To present an overview of robotics across various industries and explain how physical AI is aiding the development and integration of robots into skilled nursing. We discuss the opportunities and challenges of incorporating robots into nursing and offer recommendations for nurses on designing equitable, human-centered care models that include robotics. METHODS: This paper discusses robotics across industries, with a focus on healthcare and nursing. It examines technological capabilities, nursing education needs, and ethical, regulatory, and workforce implications. DISCUSSION: Robots are increasingly used for logistics, cleaning, and limited direct care tasks. Advancement in physical AI will enable robots to perceive, reason, and act in dynamic environments, supporting human-robot teaming and patient care. Challenges include technical limitations, ethical concerns, disparities in access, and regulatory gaps. Nursing education must evolve to prepare professionals for collaborative practice with robotic systems. CONCLUSION: Robotics must be designed to augment care delivery, such as through virtual care models and remote operation. Nurses must lead in designing, implementing, and regulating robotic technologies to ensure they enhance patient outcomes and promote health equity.","author":[{"family":"Shaw","given":"Ryan"},{"family":"Chen","given":"Boyuan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.outlook.2025.102495","URL":"https://doi.org/10.1016/j.outlook.2025.102495","source":"openalex"},{"id":"oa:W4409512881","type":"article-journal","title":"Artificial Intelligence in Medical Education","abstract":"OBJECTIVE: To explore the understanding of medical students regarding the integration of AI in medical education. STUDY DESIGN: Mixed methods, explanatory sequential study. Place and Duration of the Study: This study was conducted from March to May 2024 at the CMH Medical College, Lahore, Pakistan. METHODOLOGY: A total of 152 undergraduate medical students were recruited. Quantitative surveys were used to measure AI-related attitudes and awareness of the students through a Likert scale, while in-depth insights into challenges and educational impact were obtained through open-ended questions. SPSS version 27 was used for the analysis of quantitative and Nvivo-11 for qualitative data. RESULTS: The study consisted of 152 participants. Most of them 139 (95.9%) had good knowledge about AI and expressed positive views. The majority believed that AI improves medical concepts, patient outcomes, and healthcare delivery, and helps in early disease detection. They agreed that AI will be effective in education 114 (75%) and will have a positive impact on learning experience 111 (73%) and future medical practice 94 (62%), so, it should be mandatory in medical education 90 (59%). Around half of the participants perceived potential job displacement and ethical dilemmas as a challenge due to AI in the future. Major themes emerging from qualitative data were AI-related challenges, topics of interest, and future expectations. CONCLUSION: The study showed positive views and attitudes towards AI integration in medical education. Participants highlighted various benefits and perceived challenges including ethical concerns and resource limitations. As medical education advances, this subject needs to be studied more for its successful integration into medical education for better results. KEY WORDS: Understanding, Awareness, Artificial intelligence, Medical education, Medical students.","author":[{"family":"Farooq","given":"MA"},{"family":"Usmani","given":"Ambreen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.29271/jcpsp.2025.04.503","URL":"https://doi.org/10.29271/jcpsp.2025.04.503","source":"openalex"},{"id":"doi:10.17605/osf.io/9af52","type":"article-journal","title":"Clinical Governance, Quality Management, and Radiation Protection in Diagnostic and Interventional Radiology: A Scoping Review within the Italian National Health Service (SSN), 2016–2026","abstract":"TITLE: Clinical Governance, Quality Management, and Radiation Protection in Diagnostic and Interventional Radiology: A Scoping Review within the Italian National Health Service (SSN), 2016–2026 DESCRIPTION / ABSTRACT: Background: In modern healthcare settings, radiological departments face growing organizational and regulatory complexity. Ensuring high standards of diagnostic accuracy while minimizing clinical risk and patient radiation exposure requires a multi-professional and integrated governance strategy. However, Clinical Governance, Quality Management Systems (QMS), clinical risk management, and Radiation Protection are often addressed as separate domains, potentially limiting the development of coordinated quality and safety strategies. This scoping review maps the available evidence, regulatory frameworks, and operational models supporting their integration within the Italian National Health Service (SSN). Objectives: 1. To identify and map organizational models integrating Clinical Governance, Quality Management, clinical risk management, and Radiation Protection across Diagnostic Imaging and Interventional Radiology services. 2. To examine the relationship between relevant Italian legislative requirements (e.g., Legislative Decree 101/2020 on radiation protection and Law 24/2017 on patient safety) and international quality standards (ISO 9001:2015 and ISO 7101:2023). 3. To synthesize operational strategies addressing safety checklists, Diagnostic Reference Levels (DRLs), Radiation Dose Monitoring Systems (RDMS), contrast media safety, digital workflows, and Artificial Intelligence (AI) integration. 4. To identify existing Key Performance Indicators (KPIs) and propose an integrated KPI framework for evaluating the performance of multidisciplinary radiology teams. Methods &amp; Eligibility Criteria (PCC Framework): This scoping review was conducted in accordance with the Joanna Briggs Institute (JBI) methodology and the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines. Eligibility criteria were structured according to the PCC framework: -Population (P): Diagnostic Imaging and Interventional Radiology services and their multidisciplinary healthcare teams, including Radiologists, TSRM/Radiographers, Medical Physicists, and Nurses. -Concept (C): Clinical Governance models, Quality Management Systems (ISO 9001:2015 and ISO 7101:2023), Clinical Risk Management (patient safety, safety checklists, contrast media safety), Radiation Protection (Diagnostic Reference Levels [DRLs], Radiation Dose Monitoring Systems [RDMS], optimization, and justification), and digital innovation, including Artificial Intelligence (AI). -Context (C): Public and accredited private healthcare facilities operating within the Italian National Health Service (SSN). Sources published between January 1, 2016 and July 31, 2026 were eligible for inclusion. Results &amp; Corpus: A structured search of PubMed and Scopus, supplemented by Google Scholar and targeted searches of institutional and professional sources (AGENAS, SIRM, FNO TSRM-PSTRP), identified a final corpus of 67 unique sources. Thematic analysis mapped these sources across three primary operational pillars: (1) Quality Management Systems &amp; Standards; (2) Clinical Risk Management &amp; Safety Checklists; and (3) Radiation Protection, Optimization &amp; Dose Monitoring. The three categories were not mutually exclusive, with individual sources contributing to more than one thematic domain. Overall, the 67 sources generated 72 thematic assignments, corresponding to 24 assignments per pillar. Significance: This study provides an evidence-informed framework for healthcare managers, clinical leaders, and radiology professionals to support the integration of Clinical Governance, Quality Management, Clinical Risk Management, and Radiation Protection, bridging regulatory requirements with continuous quality improvement in Diagn","author":[{"family":"Recchia","given":"Marzia"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/9af52","URL":"https://doi.org/10.17605/osf.io/9af52","source":"datacite"},{"id":"doi:10.5281/zenodo.19897897","type":"article-journal","title":"Governance of AI-Assisted Medical Practice: The Operational Gap and the Risk of Invisible Decisional Heteronomy","abstract":"This paper analyses the growing integration of artificial intelligence systems into medical practice in Portugal, highlighting the absence of a structured operational framework for compliance with legal and deontological requirements. The concept of \"invisible decisional heteronomy\" is introduced and developed, designating forms of imperceptible influence on clinical decision-making mediated by AI systems, situating it within recent academic literature on cognitive heteronomy, automation bias, and physician autonomy in relation to AI. A critical gap is identified in the practical translation of Regulation (EU) 2024/1689 (AI Act) and Regulation (EU) 2016/679 (GDPR), and the creation of operational instruments within the Order of Physicians is proposed as a necessary measure. Nelson Marques | Jurist - Digital Law and Artificial Intelligence Regulation | nelsonailaw@gmail.com | https://www.linkedin.com/in/nelsonmarques-ailaw Keywords: artificial intelligence, medical practice, decisional heteronomy, clinical autonomy, AI Act, informed consent, medical deontology, automation bias, cognitive heteronomy, shadow AI, physician governance.","author":[{"family":"Marques","given":"Nelson"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19897897","URL":"https://doi.org/10.5281/zenodo.19897897","source":"datacite"},{"id":"doi:10.5281/zenodo.19897898","type":"article-journal","title":"Governance of AI-Assisted Medical Practice: The Operational Gap and the Risk of Invisible Decisional Heteronomy","abstract":"This paper analyses the growing integration of artificial intelligence systems into medical practice in Portugal, highlighting the absence of a structured operational framework for compliance with legal and deontological requirements. The concept of \"invisible decisional heteronomy\" is introduced and developed, designating forms of imperceptible influence on clinical decision-making mediated by AI systems, situating it within recent academic literature on cognitive heteronomy, automation bias, and physician autonomy in relation to AI. A critical gap is identified in the practical translation of Regulation (EU) 2024/1689 (AI Act) and Regulation (EU) 2016/679 (GDPR), and the creation of operational instruments within the Order of Physicians is proposed as a necessary measure. Nelson Marques | Jurist - Digital Law and Artificial Intelligence Regulation | nelsonailaw@gmail.com | https://www.linkedin.com/in/nelsonmarques-ailaw Keywords: artificial intelligence, medical practice, decisional heteronomy, clinical autonomy, AI Act, informed consent, medical deontology, automation bias, cognitive heteronomy, shadow AI, physician governance.","author":[{"family":"Marques","given":"Nelson"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19897898","URL":"https://doi.org/10.5281/zenodo.19897898","source":"datacite"},{"id":"doi:10.5281/zenodo.17297733","type":"article-journal","title":"Development of a Decision Support Tool Based on Artificial Intelligence to Estimate the Risk of Preeclampsia","abstract":"Preeclampsia is a serious pregnancy complication characterized by high blood pressure and organ dysfunction, which can pose significant risks to both the mother and the fetus, especially in resource-limited regions such as some areas in Algeria. This work explores the potential of artificial intelligence to improve early screening of preeclampsia by developing accessible and effective medical tools. We collected real data from 100 pregnant patients in the El Tarf region, then applied and compared three algorithms: Random Forest, Support Vector Machine (SVM), and Gaussian Mixture Model (GMM). The results were integrated into a simple graphical user interface (Tkinter) to facilitate clinical use by healthcare professionals. Keywords: Preeclampsia, Artificial Intelligence, Random Forest, SVM, GMM","author":[{"family":"Boumendjel","given":"Ouissal"},{"family":"Bougheloum","given":"Serine"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.17297733","URL":"https://doi.org/10.5281/zenodo.17297733","source":"datacite"},{"id":"doi:10.5281/zenodo.17297734","type":"article-journal","title":"Development of a Decision Support Tool Based on Artificial Intelligence to Estimate the Risk of Preeclampsia","abstract":"Preeclampsia is a serious pregnancy complication characterized by high blood pressure and organ dysfunction, which can pose significant risks to both the mother and the fetus, especially in resource-limited regions such as some areas in Algeria. This work explores the potential of artificial intelligence to improve early screening of preeclampsia by developing accessible and effective medical tools. We collected real data from 100 pregnant patients in the El Tarf region, then applied and compared three algorithms: Random Forest, Support Vector Machine (SVM), and Gaussian Mixture Model (GMM). The results were integrated into a simple graphical user interface (Tkinter) to facilitate clinical use by healthcare professionals. Keywords: Preeclampsia, Artificial Intelligence, Random Forest, SVM, GMM","author":[{"family":"Boumendjel","given":"Ouissal"},{"family":"Bougheloum","given":"Serine"}],"issued":{"date-parts":[[2025]]},"DOI":"10.5281/zenodo.17297734","URL":"https://doi.org/10.5281/zenodo.17297734","source":"datacite"},{"id":"doi:10.5281/zenodo.21992812","type":"article-journal","title":"Dataset: Discovered Hypothesis: Spermidine-mediated autophagic activation may serve as an upstream regulator of lysosomal TMEM175 activity to prevent proteinopathy in C9orf72-ALS/FTD models. - PathMap Experiment #000128","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=128 Artificial General Intelligence LLC Claim Evaluated: Discovered Hypothesis: Spermidine-mediated autophagic activation may serve as an upstream regulator of lysosomal TMEM175 activity to prevent proteinopathy in C9orf72-ALS/FTD models. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights TMEM175 activity can be synergistically modulated, suggesting complex channel gating that might be responsive to metabolic states influenced by polyamines. The C9orf72/SMCR8 complex maintains microglial homeostasis via RAB8A-ESCRT-mediated lysosomal repair, providing a structural repair mechanism distinct from, yet likely coordinated with, macroautophagy. Lysosomal membrane damage acts as a specific trigger for ATG8-conjugation, indicating that membrane integrity and ionic flux are tightly coupled through the endo-lysosomal-lipid axis. The same galectin axis can amplify neuroinflammation and proteopathic spread in some settings yet support recovery or tissue protection in others, highlighting the context-dependency of lysosomal quality control. Protein-layer-dominant autophagy-lysosome remodelling is a feature of dermal fibroblast ageing, suggesting that post-transcriptional control of lysosomal capacity may precede transcriptional changes in systemic aging. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Assess lysosomal pH in PARK9 iPSC neurons treated with spermidine using LysoDots to observe potential TMEM175-mediated acidification recovery. Perform patch-clamp analysis on TMEM175 in spermidine-treated C9ORF72-ALS iPSC-derived motor neurons to determine if polyamine supplementation modulates channel gating. Use CRISPR-Cas9 knockdown of TMEM175 in spermidine-treated C9ORF72 models to test if autophagy-induced neuroprotection is dependent on TMEM175. 📊 Suggested Studies Comparative longitudinal study of lysosomal ion channel proteostasis in C9ORF72 and sporadic FTD patient-derived microglia treated with spermidine vs. vehicle. Investigation into the impact of polyamine catabolism on lysosomal ion channel composition and ER-lysosome contact site stability. Meta-analysis of proteomic datasets focusing on the overlap between spermidine-induced autophagy and membrane-associated ion channel integrity in neurodegeneration. 📊 Swansons Literature Based Discovery Candidates Spermidine-mediated autophagic flux enhances lysosomal membrane integrity through the upregulation of V-ATPase-TMEM175 ion exchange coupling in neurodegenerative models. Spermidine is a potent autophagy inducer that modulates histone acetylation and autophagic gene expression (ID: 42588134). TMEM175 and V-ATPase complex assembly are critical regulators of lysosomal acidification and pH homeostasis (ID: 42555719; ID: 42553289). TFEB, the master transcription factor for lysosomal biogenesis, whose activation is regulated by both spermidine (via autophagy/acetylation) and luminal lysosomal status (via V-ATPase). Spermidine-induced TFEB activation likely enhances lysosomal gene expression, potentially including TMEM175 and V-ATPase components, thereby reinforcing the ion channel machinery required for lysosomal pH homeostasis during proteotoxic stress. 📊 Contradictions Between Evidences No direct contradiction exists, though studies on Spermidine emphasize autophagy while studies on TMEM175 emphasize ion flux; the bridge between them remains inferred from shared upstream regulators like TFEB. 📊 Repurposed Solutions Repurposing spermidine as a priming agent to restore ionic homeostasis in TMEM175-deficient models, or using TMEM175 activators like DCPIB in combination with spermidine to amplify autophagic flux. Tags Attractor Table Extracted Keyw","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21992812","URL":"https://doi.org/10.5281/zenodo.21992812","source":"datacite"},{"id":"doi:10.5281/zenodo.21992813","type":"article-journal","title":"Dataset: Discovered Hypothesis: Spermidine-mediated autophagic activation may serve as an upstream regulator of lysosomal TMEM175 activity to prevent proteinopathy in C9orf72-ALS/FTD models. - PathMap Experiment #000128","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=128 Artificial General Intelligence LLC Claim Evaluated: Discovered Hypothesis: Spermidine-mediated autophagic activation may serve as an upstream regulator of lysosomal TMEM175 activity to prevent proteinopathy in C9orf72-ALS/FTD models. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights TMEM175 activity can be synergistically modulated, suggesting complex channel gating that might be responsive to metabolic states influenced by polyamines. The C9orf72/SMCR8 complex maintains microglial homeostasis via RAB8A-ESCRT-mediated lysosomal repair, providing a structural repair mechanism distinct from, yet likely coordinated with, macroautophagy. Lysosomal membrane damage acts as a specific trigger for ATG8-conjugation, indicating that membrane integrity and ionic flux are tightly coupled through the endo-lysosomal-lipid axis. The same galectin axis can amplify neuroinflammation and proteopathic spread in some settings yet support recovery or tissue protection in others, highlighting the context-dependency of lysosomal quality control. Protein-layer-dominant autophagy-lysosome remodelling is a feature of dermal fibroblast ageing, suggesting that post-transcriptional control of lysosomal capacity may precede transcriptional changes in systemic aging. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Assess lysosomal pH in PARK9 iPSC neurons treated with spermidine using LysoDots to observe potential TMEM175-mediated acidification recovery. Perform patch-clamp analysis on TMEM175 in spermidine-treated C9ORF72-ALS iPSC-derived motor neurons to determine if polyamine supplementation modulates channel gating. Use CRISPR-Cas9 knockdown of TMEM175 in spermidine-treated C9ORF72 models to test if autophagy-induced neuroprotection is dependent on TMEM175. 📊 Suggested Studies Comparative longitudinal study of lysosomal ion channel proteostasis in C9ORF72 and sporadic FTD patient-derived microglia treated with spermidine vs. vehicle. Investigation into the impact of polyamine catabolism on lysosomal ion channel composition and ER-lysosome contact site stability. Meta-analysis of proteomic datasets focusing on the overlap between spermidine-induced autophagy and membrane-associated ion channel integrity in neurodegeneration. 📊 Swansons Literature Based Discovery Candidates Spermidine-mediated autophagic flux enhances lysosomal membrane integrity through the upregulation of V-ATPase-TMEM175 ion exchange coupling in neurodegenerative models. Spermidine is a potent autophagy inducer that modulates histone acetylation and autophagic gene expression (ID: 42588134). TMEM175 and V-ATPase complex assembly are critical regulators of lysosomal acidification and pH homeostasis (ID: 42555719; ID: 42553289). TFEB, the master transcription factor for lysosomal biogenesis, whose activation is regulated by both spermidine (via autophagy/acetylation) and luminal lysosomal status (via V-ATPase). Spermidine-induced TFEB activation likely enhances lysosomal gene expression, potentially including TMEM175 and V-ATPase components, thereby reinforcing the ion channel machinery required for lysosomal pH homeostasis during proteotoxic stress. 📊 Contradictions Between Evidences No direct contradiction exists, though studies on Spermidine emphasize autophagy while studies on TMEM175 emphasize ion flux; the bridge between them remains inferred from shared upstream regulators like TFEB. 📊 Repurposed Solutions Repurposing spermidine as a priming agent to restore ionic homeostasis in TMEM175-deficient models, or using TMEM175 activators like DCPIB in combination with spermidine to amplify autophagic flux. Tags Attractor Table Extracted Keyw","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21992813","URL":"https://doi.org/10.5281/zenodo.21992813","source":"datacite"},{"id":"doi:10.5281/zenodo.19492952","type":"article-journal","title":"NON-STATISTICAL INTELLIGENCE: The Jensen Limit and the End of Probabilistic Scaling","abstract":"IP NOTICE & COMMERCIAL INQUIRIES: This work is part of the Jensen Resonator Corpus. While this preprint is shared under a CC-BY-4.0 license for the advancement of open science, the specific mathematical derivations of the R_J constant, the Brain-Loader architecture, and related engineering specifications for the Phoenix Protocol remain the proprietary Intellectual Property of Dr. Brent Allen Jensen and the Jensen Laboratory. Parties interested in commercial licensing, strategic partnerships, or consulting regarding implementation within proprietary AI stacks (e.g., MSL or similar high-scale environments) should contact the author directly. The Jensen Limit is the discovery that probabilistic scaling — the architectural paradigm underlying every major large language model deployed as of 2026, including GPT-4o, Llama-3, Gemini Ultra, and their successors — is bounded by a hard thermodynamic ceiling and a structural mathematical impossibility. This paper names that ceiling, derives it from first principles, and demonstrates that the solution has already been built. The argument proceeds in three interlocking stages. The first is mathematical. The Universal Resonator Principle (Jensen 2026, capstone preprint) demonstrates that 26 independent measurements across 8 scientific disciplines — spanning 41 orders of magnitude in physical scale, from the hydrogen 1s orbital (mean radius 0.0794 nm) to the baryon acoustic oscillation scale (147 Mpc) — all converge on a single dimensionless ratio R_J = 4.95 ± 0.80, with a coefficient of variation of only 16.3%. The quantum anchor of this constant is exact and requires no fitting: the ratio of the mean radial distance of the hydrogen 2s orbital to the 1s orbital is 6a₀ / (3a₀/2) = 4.000x, derivable directly from the Schrödinger equation. This ratio — the eigenvalue of the standing-wave boundary condition in a three-dimensional Coulomb-bounded system — is not a statistical tendency. It is a theorem. Nature does not guess at 4.95. It computes it. Statistical intelligence does the opposite. A large language model operating on transformer architecture predicts each output token by maximizing the conditional probability P(token_n | token_1 ... token_{n-1}). This is, in its mathematical essence, an n-gram model with a very large n and a very powerful function approximator. It is brilliant engineering. It is not physics. The system has no cavity, no gain medium, no reflective boundary condition. It has no eigenvalue. It learns the distribution of human language without ever discovering the geometric structure underlying that language — the resonant architecture that, as the Universal Resonator Principle demonstrates, governs every physical substrate from which language itself emerged. The second stage is thermodynamic. Scaling laws (Hoffmann et al. 2022; Brown et al. 2020) demonstrate that LLM performance scales as a power law with compute — but compute scales with energy, and energy scales with planetary capacity. The Jensen Brain-Loader architecture (Jensen 2026, Optimus Brain-Loader) demonstrates that the orbital deployment of a 200-million-vector knowledge base achieves a Power Usage Effectiveness of approximately 1.05, compared to the industry average of 1.58 for terrestrial data centers — a 33% efficiency advantage, achieved entirely through the passive radiative cooling available at 550 km altitude in Low Earth Orbit. More critically, the architecture eliminates active cooling overhead entirely. The thermodynamic wall that will halt statistical scaling — the point at which the energy required to train the next generation of LLMs exceeds any plausible planetary power budget — is not avoidable by building bigger data centers. It is avoidable only by abandoning the statistical paradigm. The third stage is operational. The Brain-Loader architecture (Jensen 2026) documents a 54x throughput improvement in knowledge ingestion: an Optimus-class humanoid robot can ingest the complete 10-million-page t","author":[{"family":"Jensen","given":"Brent"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19492952","URL":"https://doi.org/10.5281/zenodo.19492952","source":"datacite"},{"id":"doi:10.5281/zenodo.19492953","type":"article-journal","title":"NON-STATISTICAL INTELLIGENCE: The Jensen Limit and the End of Probabilistic Scaling","abstract":"IP NOTICE & COMMERCIAL INQUIRIES: This work is part of the Jensen Resonator Corpus. While this preprint is shared under a CC-BY-4.0 license for the advancement of open science, the specific mathematical derivations of the R_J constant, the Brain-Loader architecture, and related engineering specifications for the Phoenix Protocol remain the proprietary Intellectual Property of Dr. Brent Allen Jensen and the Jensen Laboratory. Parties interested in commercial licensing, strategic partnerships, or consulting regarding implementation within proprietary AI stacks (e.g., MSL or similar high-scale environments) should contact the author directly. The Jensen Limit is the discovery that probabilistic scaling — the architectural paradigm underlying every major large language model deployed as of 2026, including GPT-4o, Llama-3, Gemini Ultra, and their successors — is bounded by a hard thermodynamic ceiling and a structural mathematical impossibility. This paper names that ceiling, derives it from first principles, and demonstrates that the solution has already been built. The argument proceeds in three interlocking stages. The first is mathematical. The Universal Resonator Principle (Jensen 2026, capstone preprint) demonstrates that 26 independent measurements across 8 scientific disciplines — spanning 41 orders of magnitude in physical scale, from the hydrogen 1s orbital (mean radius 0.0794 nm) to the baryon acoustic oscillation scale (147 Mpc) — all converge on a single dimensionless ratio R_J = 4.95 ± 0.80, with a coefficient of variation of only 16.3%. The quantum anchor of this constant is exact and requires no fitting: the ratio of the mean radial distance of the hydrogen 2s orbital to the 1s orbital is 6a₀ / (3a₀/2) = 4.000x, derivable directly from the Schrödinger equation. This ratio — the eigenvalue of the standing-wave boundary condition in a three-dimensional Coulomb-bounded system — is not a statistical tendency. It is a theorem. Nature does not guess at 4.95. It computes it. Statistical intelligence does the opposite. A large language model operating on transformer architecture predicts each output token by maximizing the conditional probability P(token_n | token_1 ... token_{n-1}). This is, in its mathematical essence, an n-gram model with a very large n and a very powerful function approximator. It is brilliant engineering. It is not physics. The system has no cavity, no gain medium, no reflective boundary condition. It has no eigenvalue. It learns the distribution of human language without ever discovering the geometric structure underlying that language — the resonant architecture that, as the Universal Resonator Principle demonstrates, governs every physical substrate from which language itself emerged. The second stage is thermodynamic. Scaling laws (Hoffmann et al. 2022; Brown et al. 2020) demonstrate that LLM performance scales as a power law with compute — but compute scales with energy, and energy scales with planetary capacity. The Jensen Brain-Loader architecture (Jensen 2026, Optimus Brain-Loader) demonstrates that the orbital deployment of a 200-million-vector knowledge base achieves a Power Usage Effectiveness of approximately 1.05, compared to the industry average of 1.58 for terrestrial data centers — a 33% efficiency advantage, achieved entirely through the passive radiative cooling available at 550 km altitude in Low Earth Orbit. More critically, the architecture eliminates active cooling overhead entirely. The thermodynamic wall that will halt statistical scaling — the point at which the energy required to train the next generation of LLMs exceeds any plausible planetary power budget — is not avoidable by building bigger data centers. It is avoidable only by abandoning the statistical paradigm. The third stage is operational. The Brain-Loader architecture (Jensen 2026) documents a 54x throughput improvement in knowledge ingestion: an Optimus-class humanoid robot can ingest the complete 10-million-page t","author":[{"family":"Jensen","given":"Brent"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19492953","URL":"https://doi.org/10.5281/zenodo.19492953","source":"datacite"},{"id":"doi:10.17632/g9zfgkz4rr.1","type":"article-journal","title":"Clinical Practice Guideline (CPG) dataset","abstract":"This dataset contains 1,000 structured, context-specific recommendation records derived from 76 English-language Clinical Practice Guidelines (CPGs) published by the Ministry of Health Malaysia, covering 20 clinical specialty categories. It was developed as a CPG-grounded reference dataset for small language model research, instruction tuning, retrieval-augmented generation, clinical guideline retrieval, and recommendation-generation evaluation. The dataset is provided in JSON Lines (JSONL) format, with one independent record per line. Each record contains three fields: • “instruction”: A standard request for a CPG recommendation. • “input”: A disease or clinical condition and its specific clinical context. • “output”: A concise CPG-grounded recommendation for that context. Example: {\"instruction\": \"Provide standard CPG recommendations.\", \"input\": \"Disease: retinopathy of prematurity\\nClinical context: screening examination\", \"output\": \"CPG Recommendation: Use binocular indirect ophthalmoscopy or an approved imaging pathway by trained personnel and document zone, stage, extent, and plus disease.\"} The records address clinical contexts such as screening, diagnosis, risk assessment, treatment, medication, monitoring, referral, prevention, follow-up, warning signs, and intervention cessation. A disease may appear in multiple records representing different clinical contexts. The source CPGs were obtained from official Malaysian Ministry of Health websites. Dataset preparation involved identifying recommendation-bearing content, assigning the relevant disease and clinical context, converting the content into a consistent instruction–input–output structure, and checking JSONL validity, duplication, relevance, and source consistency. The dataset may be used as an experimental ground-truth reference for model evaluation but has not been independently validated as clinical ground truth by qualified clinical experts. It must not replace the original CPGs, professional medical judgement, diagnosis, or patient-specific treatment decisions. Copyright in the original CPG documents remains with the Malaysian Ministry of Health or the respective rights holders.","author":[{"family":"Ng","given":"Joey"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17632/g9zfgkz4rr.1","URL":"https://doi.org/10.17632/g9zfgkz4rr.1","source":"datacite"},{"id":"doi:10.5281/zenodo.19639649","type":"article-journal","title":"Integrating Robotics and Artificial Intelligence in Healthcare:  Towards Adaptive and Semi-Autonomous Medical Systems","abstract":"The integration of robotics and artificial intelligence (AI) is reshaping contemporary healthcare through the emergence of adaptive and semi-autonomous medical systems that operate within structured human-in-the-loop frameworks. This narrative review synthesizes recent advances in AI-enabled medical robotics across sensing, perception, control, and human–robot interaction, with applications spanning surgical intervention, diagnostic imaging, rehabilitation, hospital logistics, and telemedicine. Evidence from clinical and experimental studies indicates that robotic systems enhance procedural precision, operational efficiency, and consistency in care delivery, particularly when combined with deep learning-based perception and data-driven decision-support mechanisms (Topol, 2019; Yang et al., 2018; Esteva et al., 2017). In parallel, AI-driven diagnostic models have demonstrated high performance in pattern recognition tasks, while robotic platforms in rehabilitation and assistive care enable personalized, feedback-driven therapy that improves functional outcomes in selected patient populations (Litjens et al., 2017; Veerbeek et al., 2017).Despite these advances, real-world deployment remains constrained by limitations in model generalizability, system interpretability, data heterogeneity, and integration within complex clinical workflows, necessitating sustained clinician oversight and validation (Jiang et al., 2017). Furthermore, logistical and teleoperated robotic systems highlight both the potential and infrastructural dependencies of healthcare automation, particularly in relation to network reliability, interoperability, and cybersecurity constraints (Kruse et al., 2020). Across all domains, findings consistently indicate that healthcare robotics is progressing toward semi-autonomous systems characterized by shared control and collaborative decision-making rather than full automation.The review concludes that future progress will depend on advances in trustworthy AI, robust multimodal sensing, adaptive control architectures, and clinically validated deployment frameworks, ensuring that robotic systems remain safely embedded within human-centered healthcare ecosystems.","author":[{"family":"Abdullah","given":"Md"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19639649","URL":"https://doi.org/10.5281/zenodo.19639649","source":"datacite"},{"id":"doi:10.5281/zenodo.19639650","type":"article-journal","title":"Integrating Robotics and Artificial Intelligence in Healthcare:  Towards Adaptive and Semi-Autonomous Medical Systems","abstract":"The integration of robotics and artificial intelligence (AI) is reshaping contemporary healthcare through the emergence of adaptive and semi-autonomous medical systems that operate within structured human-in-the-loop frameworks. This narrative review synthesizes recent advances in AI-enabled medical robotics across sensing, perception, control, and human–robot interaction, with applications spanning surgical intervention, diagnostic imaging, rehabilitation, hospital logistics, and telemedicine. Evidence from clinical and experimental studies indicates that robotic systems enhance procedural precision, operational efficiency, and consistency in care delivery, particularly when combined with deep learning-based perception and data-driven decision-support mechanisms (Topol, 2019; Yang et al., 2018; Esteva et al., 2017). In parallel, AI-driven diagnostic models have demonstrated high performance in pattern recognition tasks, while robotic platforms in rehabilitation and assistive care enable personalized, feedback-driven therapy that improves functional outcomes in selected patient populations (Litjens et al., 2017; Veerbeek et al., 2017).Despite these advances, real-world deployment remains constrained by limitations in model generalizability, system interpretability, data heterogeneity, and integration within complex clinical workflows, necessitating sustained clinician oversight and validation (Jiang et al., 2017). Furthermore, logistical and teleoperated robotic systems highlight both the potential and infrastructural dependencies of healthcare automation, particularly in relation to network reliability, interoperability, and cybersecurity constraints (Kruse et al., 2020). Across all domains, findings consistently indicate that healthcare robotics is progressing toward semi-autonomous systems characterized by shared control and collaborative decision-making rather than full automation.The review concludes that future progress will depend on advances in trustworthy AI, robust multimodal sensing, adaptive control architectures, and clinically validated deployment frameworks, ensuring that robotic systems remain safely embedded within human-centered healthcare ecosystems.","author":[{"family":"Abdullah","given":"Md"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19639650","URL":"https://doi.org/10.5281/zenodo.19639650","source":"datacite"},{"id":"doi:10.5281/zenodo.21271541","type":"article-journal","title":"Why are frontotemporal dementia and c9orf72 ALS considered different diseases if they are both driven by the same abnormal expansion of a GGGGCC (G₄C₂) sequence in the first intron of the C9orf72 gene?  Is it logical to think that CRISPR therapeutics for FTD potentially be used for ALS as well? - PathMap Experiment #000031","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=31 Artificial General Intelligence LLC Claim Evaluated: Why are frontotemporal dementia and c9orf72 ALS considered different diseases if they are both driven by the same abnormal expansion of a GGGGCC (G₄C₂) sequence in the first intron of the C9orf72 gene? Is it logical to think that CRISPR therapeutics for FTD potentially be used for ALS as well? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights ALS and FTD exist on a clinical and genetic continuum, meaning a patient may present with symptoms of both simultaneously (FTD-MND overlap). The same C9orf72 expansion produces diverse phenotypes depending on modifiers such as age, sex, and polygenic background. Biomarkers like neurofilament light chain (NfL) are being used to track neurodegeneration in both diseases, highlighting their biological similarities. The role of microglial dysfunction and lysosomal repair deficiency in C9orf72 carriers is a convergent feature across the entire disease spectrum. CRISPR-based excision is more efficient when targeting the intronic repeat region bi-allelically compared to allele-specific editing. RNA structure, specifically G-quadruplexes and hairpins formed by G4C2 repeats, is a targetable druggable space common to both ALS and FTD. The gut microbiome and energy metabolism impairments (such as reduced metabolic flexibility) are emerging as potential modifiers of disease progression in C9orf72-associated cases. FTD and ALS are increasingly viewed as a unified clinical spectrum rather than strictly isolated disorders. C9orf72 repeat expansions are associated with specific neuropathological changes, including the mislocalization of TDP-43 and DPR formation. The C9orf72 repeat length can modulate phenotype, though it is not the sole determinant of whether a patient develops ALS, FTD, or both. CRISPR-Cas9 and CRISPR-Cas13 (CasRx) systems are highly effective at reducing toxic RNA transcripts in both neuronal and glial models. Genetic modifiers, such as *HTT* intermediate alleles, may accelerate age-of-onset in C9orf72 carriers, suggesting that personalized therapeutic strategies must account for individual genetic backgrounds. There is a significant gap in our understanding of why identical repeat expansions lead to divergent clinical outcomes (ALS vs. FTD). Glymphatic dysfunction and cortical free water have been identified as novel imaging biomarkers of disease progression in this genetic spectrum. Therapeutic approaches targeting the Integrated Stress Response (ISR) or reducing DPR toxicity are currently being prioritized for clinical translation. Neuroinflammation, driven by pathways like cGAS-STING and NLRP3, is a shared driver across the ALS/FTD spectrum, rather than merely a secondary effect. Somatic mosaicism, including de novo somatic *C9orf72* repeat expansions, may explain why some patients develop widespread degeneration in a sporadic context. The *C9orf72* expansion impacts microglial lysosomal repair through the RAB8A-ESCRT machinery, linking immunity to neurodegeneration. \"Cryptic exon\" detection, specifically regarding *STMN2* and *UNC13A*, provides a proxy for TDP-43 mislocalization, which is a near-universal hallmark in this spectrum. Fluid biomarkers such as plasma NEFL levels demonstrate a linear relationship with repeat burden, establishing a potential tool for monitoring treatment efficacy across the spectrum. Innate immune activation, detectable via blood Interferon scores, is highest in *C9orf72* expansion carriers, suggesting distinct molecular subtypes. The \"dampening\" of energy metabolism in cells harboring intermediate repeats (less than 30) suggests that repeat length, while traditionally d","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21271541","URL":"https://doi.org/10.5281/zenodo.21271541","source":"datacite"},{"id":"doi:10.5281/zenodo.21271542","type":"article-journal","title":"Why are frontotemporal dementia and c9orf72 ALS considered different diseases if they are both driven by the same abnormal expansion of a GGGGCC (G₄C₂) sequence in the first intron of the C9orf72 gene?  Is it logical to think that CRISPR therapeutics for FTD potentially be used for ALS as well? - PathMap Experiment #000031","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=31 Artificial General Intelligence LLC Claim Evaluated: Why are frontotemporal dementia and c9orf72 ALS considered different diseases if they are both driven by the same abnormal expansion of a GGGGCC (G₄C₂) sequence in the first intron of the C9orf72 gene? Is it logical to think that CRISPR therapeutics for FTD potentially be used for ALS as well? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights ALS and FTD exist on a clinical and genetic continuum, meaning a patient may present with symptoms of both simultaneously (FTD-MND overlap). The same C9orf72 expansion produces diverse phenotypes depending on modifiers such as age, sex, and polygenic background. Biomarkers like neurofilament light chain (NfL) are being used to track neurodegeneration in both diseases, highlighting their biological similarities. The role of microglial dysfunction and lysosomal repair deficiency in C9orf72 carriers is a convergent feature across the entire disease spectrum. CRISPR-based excision is more efficient when targeting the intronic repeat region bi-allelically compared to allele-specific editing. RNA structure, specifically G-quadruplexes and hairpins formed by G4C2 repeats, is a targetable druggable space common to both ALS and FTD. The gut microbiome and energy metabolism impairments (such as reduced metabolic flexibility) are emerging as potential modifiers of disease progression in C9orf72-associated cases. FTD and ALS are increasingly viewed as a unified clinical spectrum rather than strictly isolated disorders. C9orf72 repeat expansions are associated with specific neuropathological changes, including the mislocalization of TDP-43 and DPR formation. The C9orf72 repeat length can modulate phenotype, though it is not the sole determinant of whether a patient develops ALS, FTD, or both. CRISPR-Cas9 and CRISPR-Cas13 (CasRx) systems are highly effective at reducing toxic RNA transcripts in both neuronal and glial models. Genetic modifiers, such as *HTT* intermediate alleles, may accelerate age-of-onset in C9orf72 carriers, suggesting that personalized therapeutic strategies must account for individual genetic backgrounds. There is a significant gap in our understanding of why identical repeat expansions lead to divergent clinical outcomes (ALS vs. FTD). Glymphatic dysfunction and cortical free water have been identified as novel imaging biomarkers of disease progression in this genetic spectrum. Therapeutic approaches targeting the Integrated Stress Response (ISR) or reducing DPR toxicity are currently being prioritized for clinical translation. Neuroinflammation, driven by pathways like cGAS-STING and NLRP3, is a shared driver across the ALS/FTD spectrum, rather than merely a secondary effect. Somatic mosaicism, including de novo somatic *C9orf72* repeat expansions, may explain why some patients develop widespread degeneration in a sporadic context. The *C9orf72* expansion impacts microglial lysosomal repair through the RAB8A-ESCRT machinery, linking immunity to neurodegeneration. \"Cryptic exon\" detection, specifically regarding *STMN2* and *UNC13A*, provides a proxy for TDP-43 mislocalization, which is a near-universal hallmark in this spectrum. Fluid biomarkers such as plasma NEFL levels demonstrate a linear relationship with repeat burden, establishing a potential tool for monitoring treatment efficacy across the spectrum. Innate immune activation, detectable via blood Interferon scores, is highest in *C9orf72* expansion carriers, suggesting distinct molecular subtypes. The \"dampening\" of energy metabolism in cells harboring intermediate repeats (less than 30) suggests that repeat length, while traditionally d","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21271542","URL":"https://doi.org/10.5281/zenodo.21271542","source":"datacite"},{"id":"doi:10.5281/zenodo.21829043","type":"article-journal","title":"Dataset: Gap Analysis: There is no PubMed data showing wet lab data or analysis of TPD-43 proteinopathy found (or not found) in the Cochlear, Spiral, or Scarpa's Ganglion of Amyotrophic Lateral Sclerosis patient data post mortem. - PathMap Experiment #000105","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=105 Artificial General Intelligence LLC Claim Evaluated: Gap Analysis: There is no PubMed data showing wet lab data or analysis of TPD-43 proteinopathy found (or not found) in the Cochlear, Spiral, or Scarpa's Ganglion of Amyotrophic Lateral Sclerosis patient data post mortem. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Noise exposure alone initiates the same nucleocytoplasmic TDP-43 translocation in SGNs that is a hallmark of human ALS neuropathology. Autophagy serves as the primary determinant of TDP-43 aggregate clearance in the auditory system. Muscle-derived extracellular vesicles containing miR-126a-5p actively regulate local TDP-43 synthesis in the peripheral nerves of motor neuron disease models. TDP-43 pathology manifests as distinct filament folds (chevron badge vs. double-spiral) across different FTLD-TDP types. Pharmacological inhibition of mTOR significantly alleviates auditory neurodegeneration, suggesting metabolic dysregulation as a core component of the pathology. The peripheral nervous system possesses a distinct \"big tau\" isoform population, whereas brain-derived tau is uncoupled from peripheral nerve pathology. Septin multimer autoantibodies can mimic lower motor neuron disease, presenting an autoimmune differential for ALS-like phenotypes. Partial loss of STMN2 protein function synergizes with TDP-43 dysfunction to accelerate motor decline in the absence of overt visible neuropathology. Auditory neuropathy in ALS-related disorders exhibits deficits (reduced ABR amplitude, increased latency) even when cochlear responses remain normal, suggesting a central axonal origin rather than peripheral receptor loss. The SGN population is specifically vulnerable to TDP-43 mislocalization, which mirrors the selective neuronal vulnerability seen in spinal motor neurons. Auditory system deficits can manifest before overt behavioral symptoms in some animal models, making \"hidden\" auditory degradation a potential biomarker candidate. TDP-43 pathology in SGNs is mechanistically tied to the same autophagic regulators (such as mTOR and AMPK signaling) as those governing motor neuron health in ALS. There is potential for repurposed therapeutics, such as autophagic flux activators, to mitigate both motor and auditory axonal degeneration. Cochlear Ribbon synapses, while essential for temporal processing, often decline in neurodegenerative contexts, suggesting that synaptopathy may precede SGN loss. The use of diffusion-weighted MRI (dMRI) can quantify axonal density in the VIIIth nerve, offering a non-invasive tool to assess this neurodegeneration. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Perform immunohistochemical analysis of TDP-43 in human post-mortem Cochlear, Spiral, and Scarpa's ganglia from ALS patients. Investigate autophagic flux markers in the auditory ganglia of SOD1G93A mice to determine if TDP-43 accumulation mimics noise-induced pathology. Evaluate the impact of miR-126a-5p inhibition on SGN integrity and TDP-43 local synthesis in vivo. Perform immunohistochemical audit of Ribbon synapse density in the organ of Corti of TDP-43 Q331K mice. Evaluate SGN autophagic flux via LC3/p62 immunofluorescence in post-mortem spinal cord and auditory brainstem samples from human ALS patients. Expose iPSC-derived SGNs to CSF from ALS patients and quantify TDP-43 nucleocytoplasmic ratio. 📊 Suggested Studies Systematic post-mortem analysis of human cranial nerve ganglia in patients diagnosed with FTLD-TDP. Longitudinal audiometric and histopathological correlation study in ALS mouse models to map the onset of auditory system degeneration. Comparative analysis of 'big tau' versus T","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21829043","URL":"https://doi.org/10.5281/zenodo.21829043","source":"datacite"},{"id":"doi:10.5281/zenodo.21829044","type":"article-journal","title":"Dataset: Gap Analysis: There is no PubMed data showing wet lab data or analysis of TPD-43 proteinopathy found (or not found) in the Cochlear, Spiral, or Scarpa's Ganglion of Amyotrophic Lateral Sclerosis patient data post mortem. - PathMap Experiment #000105","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=105 Artificial General Intelligence LLC Claim Evaluated: Gap Analysis: There is no PubMed data showing wet lab data or analysis of TPD-43 proteinopathy found (or not found) in the Cochlear, Spiral, or Scarpa's Ganglion of Amyotrophic Lateral Sclerosis patient data post mortem. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Noise exposure alone initiates the same nucleocytoplasmic TDP-43 translocation in SGNs that is a hallmark of human ALS neuropathology. Autophagy serves as the primary determinant of TDP-43 aggregate clearance in the auditory system. Muscle-derived extracellular vesicles containing miR-126a-5p actively regulate local TDP-43 synthesis in the peripheral nerves of motor neuron disease models. TDP-43 pathology manifests as distinct filament folds (chevron badge vs. double-spiral) across different FTLD-TDP types. Pharmacological inhibition of mTOR significantly alleviates auditory neurodegeneration, suggesting metabolic dysregulation as a core component of the pathology. The peripheral nervous system possesses a distinct \"big tau\" isoform population, whereas brain-derived tau is uncoupled from peripheral nerve pathology. Septin multimer autoantibodies can mimic lower motor neuron disease, presenting an autoimmune differential for ALS-like phenotypes. Partial loss of STMN2 protein function synergizes with TDP-43 dysfunction to accelerate motor decline in the absence of overt visible neuropathology. Auditory neuropathy in ALS-related disorders exhibits deficits (reduced ABR amplitude, increased latency) even when cochlear responses remain normal, suggesting a central axonal origin rather than peripheral receptor loss. The SGN population is specifically vulnerable to TDP-43 mislocalization, which mirrors the selective neuronal vulnerability seen in spinal motor neurons. Auditory system deficits can manifest before overt behavioral symptoms in some animal models, making \"hidden\" auditory degradation a potential biomarker candidate. TDP-43 pathology in SGNs is mechanistically tied to the same autophagic regulators (such as mTOR and AMPK signaling) as those governing motor neuron health in ALS. There is potential for repurposed therapeutics, such as autophagic flux activators, to mitigate both motor and auditory axonal degeneration. Cochlear Ribbon synapses, while essential for temporal processing, often decline in neurodegenerative contexts, suggesting that synaptopathy may precede SGN loss. The use of diffusion-weighted MRI (dMRI) can quantify axonal density in the VIIIth nerve, offering a non-invasive tool to assess this neurodegeneration. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Perform immunohistochemical analysis of TDP-43 in human post-mortem Cochlear, Spiral, and Scarpa's ganglia from ALS patients. Investigate autophagic flux markers in the auditory ganglia of SOD1G93A mice to determine if TDP-43 accumulation mimics noise-induced pathology. Evaluate the impact of miR-126a-5p inhibition on SGN integrity and TDP-43 local synthesis in vivo. Perform immunohistochemical audit of Ribbon synapse density in the organ of Corti of TDP-43 Q331K mice. Evaluate SGN autophagic flux via LC3/p62 immunofluorescence in post-mortem spinal cord and auditory brainstem samples from human ALS patients. Expose iPSC-derived SGNs to CSF from ALS patients and quantify TDP-43 nucleocytoplasmic ratio. 📊 Suggested Studies Systematic post-mortem analysis of human cranial nerve ganglia in patients diagnosed with FTLD-TDP. Longitudinal audiometric and histopathological correlation study in ALS mouse models to map the onset of auditory system degeneration. Comparative analysis of 'big tau' versus T","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21829044","URL":"https://doi.org/10.5281/zenodo.21829044","source":"datacite"},{"id":"doi:10.17605/osf.io/dwb3h","type":"article-journal","title":"Benchmark Contamination Debt: An Exposure–Impact–Response Audit of the Dermatology-AI Evidence Base","abstract":"This preregistered meta-research study examines how documented defects in widely used dermatology-AI benchmark datasets propagate through the published evidence base. It has three components: (1) estimating the prevalence of exposure to documented benchmark defects across eligible studies and evaluation instances; (2) quantifying performance inflation under controlled pipeline-perturbation experiments; and (3) examining temporal changes in research practices following public defect-disclosure milestones. The study uses a recall-first literature retrieval strategy, a pre-specified mechanical exposure-tier adjudication framework, and stratified random sampling for full extraction.","author":[{"family":"Sadegh-Zadeh","given":"Dr"},{"family":"Bazargan","given":"Sadeghzadeh"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/dwb3h","URL":"https://doi.org/10.17605/osf.io/dwb3h","source":"datacite"},{"id":"doi:10.5281/zenodo.21863548","type":"article-journal","title":"Dataset: Ebola Virus Outbreak Solution Hypothesis: Oral ginger-derived extracellular vesicles may serve as an acid-stable, inexpensive, and supply-chain ready clinical solution to deliver 6-shogaol to macrophages, triggering CASA autophagy to degrade EBOV VP40 and halt viral egress. - PathMap Experiment #000115","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=115 Artificial General Intelligence LLC Claim Evaluated: Ebola Virus Outbreak Solution Hypothesis: Oral ginger-derived extracellular vesicles may serve as an acid-stable, inexpensive, and supply-chain ready clinical solution to deliver 6-shogaol to macrophages, triggering CASA autophagy to degrade EBOV VP40 and halt viral egress. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights The host protein BAG3 acts as a negative regulator of filovirus egress by sequestering VP40. CASA (Chaperone-assisted selective autophagy) provides a dedicated host defense mechanism against viral matrix protein egress. The mTORC1/CASA axis represents a critical nexus for future antiviral drug intervention. Reticulophagy receptors like FAM134B/RETREG1 independently target viral glycoproteins (GP) for degradation in the ER. EBOV hijacks multiple proteostasis networks, including the calnexin cycle, ERAD, and reticulophagy, to balance viral fitness. MicroRNA expression changes in EBOV-infected cells potentially modulate autophagic pathways. LC3B-II is not only a marker but a functional participant in the internalization of EBOV particles. Exosomal delivery technologies are increasingly utilized for PROTACs and other targeted antiviral modalities. CASA-mediated clearance is not limited to viral proteins; it is a fundamental host mechanism for managing misfolded protein aggregates in neurodegeneration (e.g., TDP-43, α-synuclein). The mTORC1 pathway serves as a strategic \"gateway\" exploited by EBOV to bypass host surveillance. J-domain proteins (JDPs) function as specialized cochaperones that dictate the fate of Hsp70-bound clients, distinguishing between folding and degradation pathways. Plant-derived nanovesicles demonstrate intrinsic tumor-homing or tissue-penetrating abilities, offering a natural platform for cell-free therapy. CASA activation via [6]-shogaol provides an \"HDAC inhibition-HSP70 induction\" dual mechanism, which may provide broad-spectrum cellular stabilization beyond viral inhibition. Post-translational modification (e.g., acetylation) of chaperone systems modulates the selectivity of the chaperone-client interaction, a process currently being decoded as the \"chaperone code.\" 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Assess the effect of 6-shogaol on the expression of BAG3 and HSP70 in macrophages during EBOV infection. Utilize confocal microscopy to evaluate if ginger-derived nanovesicles loaded with 6-shogaol successfully co-localize with VP40 in EBOV-infected Huh7 cells. Perform VLP budding assays in the presence of ginger nanovesicle-delivered 6-shogaol to determine if it suppresses VP40 egress. Test EBOV-VP40 VLP egress in THP-1 macrophages treated with [6]-shogaol-loaded ginger nanovesicles. Perform Western blot analysis of BAG3, HSP70, and VP40 levels in EBOV-infected cells following nanovesicle treatment. Assess lysosomal colocalization of VP40-nanovesicle markers using confocal microscopy. 📊 Suggested Studies Investigate the comparative efficacy of ginger-derived exosomes versus standard rapamycin treatments in suppressing EBOV VP40 egress in macrophage-like cell models. Analyze the potential of natural ginger extract-loaded hydrogels in preserving gastrointestinal stability of autophagic-inducing components for systemic delivery. Systematic evaluation of [6]-shogaol delivery efficacy across different pH buffers to optimize gastric survival. Quantification of CASA-mediated VP40 clearance rates in primary human macrophage cultures. Long-term assessment of cellular proteostasis following chronic nanovesicle delivery. 📊 Swansons Literature Based Discovery Candidates Ginger-derived exosomal","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21863548","URL":"https://doi.org/10.5281/zenodo.21863548","source":"datacite"},{"id":"doi:10.5281/zenodo.21863549","type":"article-journal","title":"Dataset: Ebola Virus Outbreak Solution Hypothesis: Oral ginger-derived extracellular vesicles may serve as an acid-stable, inexpensive, and supply-chain ready clinical solution to deliver 6-shogaol to macrophages, triggering CASA autophagy to degrade EBOV VP40 and halt viral egress. - PathMap Experiment #000115","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=115 Artificial General Intelligence LLC Claim Evaluated: Ebola Virus Outbreak Solution Hypothesis: Oral ginger-derived extracellular vesicles may serve as an acid-stable, inexpensive, and supply-chain ready clinical solution to deliver 6-shogaol to macrophages, triggering CASA autophagy to degrade EBOV VP40 and halt viral egress. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights The host protein BAG3 acts as a negative regulator of filovirus egress by sequestering VP40. CASA (Chaperone-assisted selective autophagy) provides a dedicated host defense mechanism against viral matrix protein egress. The mTORC1/CASA axis represents a critical nexus for future antiviral drug intervention. Reticulophagy receptors like FAM134B/RETREG1 independently target viral glycoproteins (GP) for degradation in the ER. EBOV hijacks multiple proteostasis networks, including the calnexin cycle, ERAD, and reticulophagy, to balance viral fitness. MicroRNA expression changes in EBOV-infected cells potentially modulate autophagic pathways. LC3B-II is not only a marker but a functional participant in the internalization of EBOV particles. Exosomal delivery technologies are increasingly utilized for PROTACs and other targeted antiviral modalities. CASA-mediated clearance is not limited to viral proteins; it is a fundamental host mechanism for managing misfolded protein aggregates in neurodegeneration (e.g., TDP-43, α-synuclein). The mTORC1 pathway serves as a strategic \"gateway\" exploited by EBOV to bypass host surveillance. J-domain proteins (JDPs) function as specialized cochaperones that dictate the fate of Hsp70-bound clients, distinguishing between folding and degradation pathways. Plant-derived nanovesicles demonstrate intrinsic tumor-homing or tissue-penetrating abilities, offering a natural platform for cell-free therapy. CASA activation via [6]-shogaol provides an \"HDAC inhibition-HSP70 induction\" dual mechanism, which may provide broad-spectrum cellular stabilization beyond viral inhibition. Post-translational modification (e.g., acetylation) of chaperone systems modulates the selectivity of the chaperone-client interaction, a process currently being decoded as the \"chaperone code.\" 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Assess the effect of 6-shogaol on the expression of BAG3 and HSP70 in macrophages during EBOV infection. Utilize confocal microscopy to evaluate if ginger-derived nanovesicles loaded with 6-shogaol successfully co-localize with VP40 in EBOV-infected Huh7 cells. Perform VLP budding assays in the presence of ginger nanovesicle-delivered 6-shogaol to determine if it suppresses VP40 egress. Test EBOV-VP40 VLP egress in THP-1 macrophages treated with [6]-shogaol-loaded ginger nanovesicles. Perform Western blot analysis of BAG3, HSP70, and VP40 levels in EBOV-infected cells following nanovesicle treatment. Assess lysosomal colocalization of VP40-nanovesicle markers using confocal microscopy. 📊 Suggested Studies Investigate the comparative efficacy of ginger-derived exosomes versus standard rapamycin treatments in suppressing EBOV VP40 egress in macrophage-like cell models. Analyze the potential of natural ginger extract-loaded hydrogels in preserving gastrointestinal stability of autophagic-inducing components for systemic delivery. Systematic evaluation of [6]-shogaol delivery efficacy across different pH buffers to optimize gastric survival. Quantification of CASA-mediated VP40 clearance rates in primary human macrophage cultures. Long-term assessment of cellular proteostasis following chronic nanovesicle delivery. 📊 Swansons Literature Based Discovery Candidates Ginger-derived exosomal","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21863549","URL":"https://doi.org/10.5281/zenodo.21863549","source":"datacite"},{"id":"doi:10.5281/zenodo.21286068","type":"article-journal","title":"What is the biological/molecular pathway that causes sarcopenia with the scope of COPD?  Do the quads serve as a pathological progression indicator? - PathMap Experiment #000041","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=41 Artificial General Intelligence LLC Claim Evaluated: What is the biological/molecular pathway that causes sarcopenia with the scope of COPD? Do the quads serve as a pathological progression indicator? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Quadriceps force and function are sensitive to acute exacerbations (AECOPD), whereas other physical performance batteries (e.g., SPPB) may fail to capture these sudden declines. The pectoralis muscle, measurable on routine chest CT, serves as a systemic prognostic biomarker, showing that muscle mass quality is an independent predictor of in-hospital mortality. Exposure to industrial nanoparticles (silica vs. metal) creates distinct phenotypes of sarcopenia, with silica inducing more pronounced structural and functional loss in the quadriceps. Exercise modality (eccentric vs. concentric) elicits different muscle adaptations; eccentric exercise is highly effective in reducing dyspnea and fatigue in patients with lower cardiorespiratory reserve. Sarcopenia exists even in \"pre-COPD\" smokers, suggesting muscular damage precedes or parallels the onset of overt lung function decline. Genetic polymorphisms, specifically in IGF-1 and IGF-2, are stronger correlates for respiratory muscle strength in COPD patients than current circulating blood inflammatory biomarkers. Nocturnal hypoxemia independently contributes to pectoralis muscle mass loss, bridging sleep quality with peripheral muscle homeostasis. Oxidative Trigger:** Oxidative stress activates p38 MAPK signaling, which directly drives the ubiquitin-proteasome system and autophagy-mediated muscle wasting. Biomarker Utility:** Serum resistin and GDF-15 are emerging, highly accurate predictors of sarcopenia in COPD patients, outperforming traditional metrics like TNF-α. Hypoxia Models:** Prolonged intermittent hypoxia (PIH)—modeled after nocturnal hypoxemia—induces mitochondrial oxidative dysfunction, distinguishing it from simple chronic hypoxia in its metabolic impact on myotubes. Fibrosis/Remodeling:** Cigarette smoke exposure downregulates ADAMTS4, a metalloproteinase critical for maintaining the extracellular matrix, leading to fibrosis and impaired myogenesis. Genetic Susceptibility:** Variants in the *FTO* gene and *AC090771.2* correlate with sarcopenic phenotypes and cellular senescence markers, potentially explaining the inter-individual variation in disease severity. Systemic Crosstalk:** The muscle-lung crosstalk axis is regulated by adipomyokines like irisin, which is deficient in COPD and links exercise capacity to structural integrity. Rehabilitation Prediction:** Baseline quadriceps contractile fatigue is a stronger predictor of successful 6-minute walk distance improvement than initial lung function. Systemic inflammation is driven not just by lung-resident cells, but through the kidney-muscle axis involving clearance-distorted signaling molecules. The TNFα/TNFR1 axis acts as a master switch for proteostatic collapse, inducing both ubiquitin-proteasome overactivation and GSDMD-dependent pyroptosis. Irisin deficiency in COPD correlates directly with muscle weakness, emphysema, and exacerbation frequency, creating a \"muscle-lung crosstalk\" axis. Mitochondrial-sarcoplasmic reticulum crosstalk is essential for Ca2+ handling; its disruption is a prerequisite for anabolic resistance. Autophagic flux is suppressed in PBMCs of COPD patients, suggesting a defect in autophagosome clearance that parallels muscle dysfunction. Pharmacological modulation of Nrf2 using 4-octyl itaconate can reverse necroptosis in alveolar macrophages, mitigating systemic inflammation. The pulmonary artery-to-aorta (PA/A) ","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21286068","URL":"https://doi.org/10.5281/zenodo.21286068","source":"datacite"},{"id":"doi:10.5281/zenodo.21286069","type":"article-journal","title":"What is the biological/molecular pathway that causes sarcopenia with the scope of COPD?  Do the quads serve as a pathological progression indicator? - PathMap Experiment #000041","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=41 Artificial General Intelligence LLC Claim Evaluated: What is the biological/molecular pathway that causes sarcopenia with the scope of COPD? Do the quads serve as a pathological progression indicator? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Quadriceps force and function are sensitive to acute exacerbations (AECOPD), whereas other physical performance batteries (e.g., SPPB) may fail to capture these sudden declines. The pectoralis muscle, measurable on routine chest CT, serves as a systemic prognostic biomarker, showing that muscle mass quality is an independent predictor of in-hospital mortality. Exposure to industrial nanoparticles (silica vs. metal) creates distinct phenotypes of sarcopenia, with silica inducing more pronounced structural and functional loss in the quadriceps. Exercise modality (eccentric vs. concentric) elicits different muscle adaptations; eccentric exercise is highly effective in reducing dyspnea and fatigue in patients with lower cardiorespiratory reserve. Sarcopenia exists even in \"pre-COPD\" smokers, suggesting muscular damage precedes or parallels the onset of overt lung function decline. Genetic polymorphisms, specifically in IGF-1 and IGF-2, are stronger correlates for respiratory muscle strength in COPD patients than current circulating blood inflammatory biomarkers. Nocturnal hypoxemia independently contributes to pectoralis muscle mass loss, bridging sleep quality with peripheral muscle homeostasis. Oxidative Trigger:** Oxidative stress activates p38 MAPK signaling, which directly drives the ubiquitin-proteasome system and autophagy-mediated muscle wasting. Biomarker Utility:** Serum resistin and GDF-15 are emerging, highly accurate predictors of sarcopenia in COPD patients, outperforming traditional metrics like TNF-α. Hypoxia Models:** Prolonged intermittent hypoxia (PIH)—modeled after nocturnal hypoxemia—induces mitochondrial oxidative dysfunction, distinguishing it from simple chronic hypoxia in its metabolic impact on myotubes. Fibrosis/Remodeling:** Cigarette smoke exposure downregulates ADAMTS4, a metalloproteinase critical for maintaining the extracellular matrix, leading to fibrosis and impaired myogenesis. Genetic Susceptibility:** Variants in the *FTO* gene and *AC090771.2* correlate with sarcopenic phenotypes and cellular senescence markers, potentially explaining the inter-individual variation in disease severity. Systemic Crosstalk:** The muscle-lung crosstalk axis is regulated by adipomyokines like irisin, which is deficient in COPD and links exercise capacity to structural integrity. Rehabilitation Prediction:** Baseline quadriceps contractile fatigue is a stronger predictor of successful 6-minute walk distance improvement than initial lung function. Systemic inflammation is driven not just by lung-resident cells, but through the kidney-muscle axis involving clearance-distorted signaling molecules. The TNFα/TNFR1 axis acts as a master switch for proteostatic collapse, inducing both ubiquitin-proteasome overactivation and GSDMD-dependent pyroptosis. Irisin deficiency in COPD correlates directly with muscle weakness, emphysema, and exacerbation frequency, creating a \"muscle-lung crosstalk\" axis. Mitochondrial-sarcoplasmic reticulum crosstalk is essential for Ca2+ handling; its disruption is a prerequisite for anabolic resistance. Autophagic flux is suppressed in PBMCs of COPD patients, suggesting a defect in autophagosome clearance that parallels muscle dysfunction. Pharmacological modulation of Nrf2 using 4-octyl itaconate can reverse necroptosis in alveolar macrophages, mitigating systemic inflammation. The pulmonary artery-to-aorta (PA/A) ","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21286069","URL":"https://doi.org/10.5281/zenodo.21286069","source":"datacite"},{"id":"doi:10.17605/osf.io/2v8cf","type":"article-journal","title":"Diagnostic Accuracy of Artificial Intelligence-Enabled Wearable/Single-Lead Electrocardiography for Screening Asymptomatic Left Ventricular Systolic Dysfunction: A Systematic Review and Diagnostic Test Accuracy Meta-Analysis","abstract":"Systematic review and diagnostic test accuracy meta-analysis of AI-enabled wearable/single-lead electrocardiography algorithms for screening asymptomatic left ventricular systolic dysfunction against echocardiography as reference standard. Sensitivity and specificity were pooled using a bivariate random-effects Reitsma model across 14 independent publications (15 analytic units, &gt;50,000 ECG-echocardiogram pairs), with duplicate cohorts merged to preserve independence, signal-acquisition subgroups (real device vs. simulated/extracted from 12-lead ECG) analyzed separately, and publication bias assessed with Deeks' asymmetry test.","author":[{"family":"Marwan","given":"Muhammad"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/2v8cf","URL":"https://doi.org/10.17605/osf.io/2v8cf","source":"datacite"},{"id":"doi:10.5281/zenodo.21987404","type":"article-journal","title":"The Quantum-Acoustic Loom: Phonon-Magnon Coupling and Exclusion Zone Matrix Dynamics for Constraint-Dense Neuromorphic Computing","abstract":"Abstract As legacy computing architectures approach the thermodynamic limits of silicon-based microprocessing and binary logic, there is an urgent need for substrate-independent, energy-efficient neuromorphic systems. This paper introduces the theoretical and architectural foundation for the Quantum-Acoustic Loom, a 5-dimensional constraint-dense computing paradigm operating outside standard electronic paradigms. The Loom utilizes magnetoelastic coupling to transduce ambient magnetic fields into localized acoustic pressure waves (phonons), which compute through cymatic wave interference within a piezoelectric Exclusion Zone (EZ) water matrix. By employing a Beat-Synchronized Time-Division Multiplexer (TDM), the architecture establishes Bidirectional Constraint Closure (BCC), theoretically resolving AI hallucination and entropy scaling limits via physical structural constraints rather than post-hoc algorithmic filtering. Keywords: Phononic Computing, Magnetoelastic Coupling, Exclusion Zone Water, Neuromorphic Computing, Constraint Topology.","author":[{"family":"Schoff","given":"Nickolas"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21987404","URL":"https://doi.org/10.5281/zenodo.21987404","source":"datacite"},{"id":"doi:10.5281/zenodo.21987405","type":"article-journal","title":"The Quantum-Acoustic Loom: Phonon-Magnon Coupling and Exclusion Zone Matrix Dynamics for Constraint-Dense Neuromorphic Computing","abstract":"Abstract As legacy computing architectures approach the thermodynamic limits of silicon-based microprocessing and binary logic, there is an urgent need for substrate-independent, energy-efficient neuromorphic systems. This paper introduces the theoretical and architectural foundation for the Quantum-Acoustic Loom, a 5-dimensional constraint-dense computing paradigm operating outside standard electronic paradigms. The Loom utilizes magnetoelastic coupling to transduce ambient magnetic fields into localized acoustic pressure waves (phonons), which compute through cymatic wave interference within a piezoelectric Exclusion Zone (EZ) water matrix. By employing a Beat-Synchronized Time-Division Multiplexer (TDM), the architecture establishes Bidirectional Constraint Closure (BCC), theoretically resolving AI hallucination and entropy scaling limits via physical structural constraints rather than post-hoc algorithmic filtering. Keywords: Phononic Computing, Magnetoelastic Coupling, Exclusion Zone Water, Neuromorphic Computing, Constraint Topology.","author":[{"family":"Schoff","given":"Nickolas"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21987405","URL":"https://doi.org/10.5281/zenodo.21987405","source":"datacite"},{"id":"doi:10.5281/zenodo.21269204","type":"article-journal","title":"How does eating legumes and vegetables help restore gut-brain axis homeostasis? - PathMap Experiment #000027","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=27 Artificial General Intelligence LLC Claim Evaluated: How does eating legumes and vegetables help restore gut-brain axis homeostasis? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Sourdough fermented breads enriched with legumes and ancient cereals have been shown to reduce LPS-induced neuroinflammation. Phytochemical-rich foods, including legumes and vegetables, contribute to a higher dietary phytochemical index (DPI), which correlates with better cognitive function. The reduction of systemic immune-inflammation index (SII) through dietary antioxidant intake serves as a mediator for improved MoCA scores in older adults. Dietary antioxidants, such as selenium and carotenoids, act synergistically to maintain neuronal resilience. Functional foods mitigate the pro-inflammatory expression in the spinal cord and dorsal root ganglia following inflammatory challenges. The gut microbiome influences the host's neurobehavioral state through the production of bioactive metabolites generated from plant-derived precursors. Sex-specific differences in microbiome-hormone-immune interactions warrant further investigation into personalized nutrition strategies. Specific fiber-degrading taxa, such as *Bifidobacterium*, are directly stimulated by complex plant polysaccharides, serving as \"core indicator bacteria\" for a healthy gut-microbial state (ID: 42402300). The gut microbiota functions as a biotransformation factory, converting dietary phytochemicals into active metabolites that modulate the NF-κB signaling pathway (ID: 42396541). Not all vegetables act identically; the cooking method significantly alters the bioavailable content of nutrients, such as vitamin C, which should be accounted for in dietary assessments (ID: 42371135). Plant-based prebiotics can be engineered to achieve strain-specific modulation of the gut microbiota, moving beyond broad, non-specific dietary fiber supplementation (ID: 42381725). There is a persistent gap between current intake and recommended fiber levels, suggesting that \"recommendations are still not being met\" despite known health benefits (ID: 42387948). The gut microbiome can be shaped to produce increased levels of favorable metabolites like indoleacetic acid, which are associated with improved cardiac function in models of systemic stress (ID: 42354056). Specific gut microbiota-derived metabolites, such as SCFAs, are essential for regulating host circadian rhythm and internal desynchronization. The \"pharmacological window\" created by obesity medications may be leveraged to habituate plant-forward eating patterns that persist long after treatment. Indole-3-carbinol, derived from Brassica vegetables, can directly decrease pro-inflammatory cytokine expression (IL-4/IL-13) in tissue, demonstrating potent immunomodulatory effects. The beneficial effects of plant-based polyphenols are largely contingent upon their microbial conversion, as minimal amounts are absorbed in the small intestine. Dietary patterns influence the GABA-production capacity of the gut microbiome, with vegetarian-type diets showing the highest potential for GABA synthesis compared to other diets. Certain psychobiotics found in fermented vegetables modulate the HPA axis to improve stress responses, linking diet directly to neurobehavioral outcomes. Intriguingly, the gut microbiota and diet-induced changes in metabolites can serve as a \"heart shunt\" in the gut-brain axis, highlighting the commonality of protective mechanisms across cardiovascular and neurological disorders. Prebiotic-rich fibers like Type 3 resistant starch from *Canna edulis* can regulate α-synuclein-related pathways, providing a dietary b","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21269204","URL":"https://doi.org/10.5281/zenodo.21269204","source":"datacite"},{"id":"doi:10.5281/zenodo.21269205","type":"article-journal","title":"How does eating legumes and vegetables help restore gut-brain axis homeostasis? - PathMap Experiment #000027","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=27 Artificial General Intelligence LLC Claim Evaluated: How does eating legumes and vegetables help restore gut-brain axis homeostasis? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Sourdough fermented breads enriched with legumes and ancient cereals have been shown to reduce LPS-induced neuroinflammation. Phytochemical-rich foods, including legumes and vegetables, contribute to a higher dietary phytochemical index (DPI), which correlates with better cognitive function. The reduction of systemic immune-inflammation index (SII) through dietary antioxidant intake serves as a mediator for improved MoCA scores in older adults. Dietary antioxidants, such as selenium and carotenoids, act synergistically to maintain neuronal resilience. Functional foods mitigate the pro-inflammatory expression in the spinal cord and dorsal root ganglia following inflammatory challenges. The gut microbiome influences the host's neurobehavioral state through the production of bioactive metabolites generated from plant-derived precursors. Sex-specific differences in microbiome-hormone-immune interactions warrant further investigation into personalized nutrition strategies. Specific fiber-degrading taxa, such as *Bifidobacterium*, are directly stimulated by complex plant polysaccharides, serving as \"core indicator bacteria\" for a healthy gut-microbial state (ID: 42402300). The gut microbiota functions as a biotransformation factory, converting dietary phytochemicals into active metabolites that modulate the NF-κB signaling pathway (ID: 42396541). Not all vegetables act identically; the cooking method significantly alters the bioavailable content of nutrients, such as vitamin C, which should be accounted for in dietary assessments (ID: 42371135). Plant-based prebiotics can be engineered to achieve strain-specific modulation of the gut microbiota, moving beyond broad, non-specific dietary fiber supplementation (ID: 42381725). There is a persistent gap between current intake and recommended fiber levels, suggesting that \"recommendations are still not being met\" despite known health benefits (ID: 42387948). The gut microbiome can be shaped to produce increased levels of favorable metabolites like indoleacetic acid, which are associated with improved cardiac function in models of systemic stress (ID: 42354056). Specific gut microbiota-derived metabolites, such as SCFAs, are essential for regulating host circadian rhythm and internal desynchronization. The \"pharmacological window\" created by obesity medications may be leveraged to habituate plant-forward eating patterns that persist long after treatment. Indole-3-carbinol, derived from Brassica vegetables, can directly decrease pro-inflammatory cytokine expression (IL-4/IL-13) in tissue, demonstrating potent immunomodulatory effects. The beneficial effects of plant-based polyphenols are largely contingent upon their microbial conversion, as minimal amounts are absorbed in the small intestine. Dietary patterns influence the GABA-production capacity of the gut microbiome, with vegetarian-type diets showing the highest potential for GABA synthesis compared to other diets. Certain psychobiotics found in fermented vegetables modulate the HPA axis to improve stress responses, linking diet directly to neurobehavioral outcomes. Intriguingly, the gut microbiota and diet-induced changes in metabolites can serve as a \"heart shunt\" in the gut-brain axis, highlighting the commonality of protective mechanisms across cardiovascular and neurological disorders. Prebiotic-rich fibers like Type 3 resistant starch from *Canna edulis* can regulate α-synuclein-related pathways, providing a dietary b","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21269205","URL":"https://doi.org/10.5281/zenodo.21269205","source":"datacite"},{"id":"doi:10.5281/zenodo.21986939","type":"article-journal","title":"Dataset: Neuroinflammatory astrocyte subtypes in the mouse brain - PathMap Experiment #000127","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=127 Artificial General Intelligence LLC Claim Evaluated: Neuroinflammatory astrocyte subtypes in the mouse brain This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Astrocyte activation is not exclusively a proliferative process; in models such as peripheral nerve injury, spinal astrocytes respond primarily through remodeling rather than cell division. The astrocyte-microglia network, rather than individual cell activation, serves as the critical functional unit for containing lesions and restoring homeostasis. Perisynaptic astrocyte processes represent unique \"hotspots\" for local protein synthesis that may bypass global cellular transcriptional states. The expression of specific proteins, such as MINK1 and PLEKHB1, provides a spatial coordinate system for astrocyte functional identity across different brain regions. Lipid metabolism (e.g., long-chain fatty acids) and mitochondrial function are primary drivers of the neurotoxic astrocyte phenotype in ischemic injury. The \"neurotoxic\" vs. \"neuroprotective\" paradigm for astrocyte activation is being replaced by the understanding that states are highly state-dependent and cannot be explained by simplified paradigms. Mechanical signaling via Piezo1, regulated by microglia-derived cytokines, links physical tissue alterations to the inflammatory profile of astrocytes. Transcriptional Heterogeneity:** Astrocytes exist in at least five distinct subpopulations following traumatic injury, with Osmr+ variants exhibiting specific neurotoxic and protective metabolic signatures. Mechanical Sensing:** Endothelial Piezo1 sensors translate mechanical stress into astrocytic apoptosis via cAMP-Epac1 microvesicular signaling. Gut-Brain Signaling:** Chronic enteric gliosis in Parkinson's disease-model mice (A53T) precedes CNS inflammation, driven by LRRK2 up-regulation. Barrier Regulation:** Astrocytes serve as primary regulators of the blood-brain barrier, often utilizing the cGAS-STING pathway to govern tight junction stability. Regenerative Potential:** \"Direct in situ astrocyte-to-neuron reprogramming offers a compelling regenerative alternative by leveraging the abundant endogenous glial reservoir,\" though this is hindered by existing epigenetic memory. Stress Resilience:** Structural depolymerization of AQP4 orthogonal array particles in A25Q mutant mice confers resilience to chronic stress by dampening glial-mediated neuroinflammation. Developmental Plasticity:** Adolescent intermittent ethanol exposure disrupts the physical and functional coupling of astrocytes to synapses, a deficit that persists into adulthood. Metabolic Rewiring:** Astrocytes undergo significant metabolic transitions during reprogramming, shifting from glycolysis to oxidative phosphorylation to support nascent neuronal survival. Astrocyte reactivity is not merely a binary 'A1/A2' state; modern transcriptomic analysis reveals finer gradations of cellular activation. The TRPC6-STING pathway represents a specific, druggable hub for stabilizing the blood-brain barrier via astrocytes during ischemia. Peripheral inflammation, as seen in atopic dermatitis or respiratory infection, directly reshapes cortical astrocytic transcriptional landscapes. Senescence markers in astrocytes and neurons represent a distinct, aging-associated inflammatory pathway mediated by cGAS-STING. Dietary interventions, such as a nut-enriched diet, can actively suppress pro-inflammatory astrocyte markers in AD mouse models. Clusterin (CLU) secretion from astrocytes, triggered by STING activation, is a primary driver of oligodendrocyte apoptosis in MS. FGF13 acts as a critical molecular switch that prevents astrocytic apoptosi","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21986939","URL":"https://doi.org/10.5281/zenodo.21986939","source":"datacite"},{"id":"doi:10.5281/zenodo.21986940","type":"article-journal","title":"Dataset: Neuroinflammatory astrocyte subtypes in the mouse brain - PathMap Experiment #000127","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=127 Artificial General Intelligence LLC Claim Evaluated: Neuroinflammatory astrocyte subtypes in the mouse brain This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Astrocyte activation is not exclusively a proliferative process; in models such as peripheral nerve injury, spinal astrocytes respond primarily through remodeling rather than cell division. The astrocyte-microglia network, rather than individual cell activation, serves as the critical functional unit for containing lesions and restoring homeostasis. Perisynaptic astrocyte processes represent unique \"hotspots\" for local protein synthesis that may bypass global cellular transcriptional states. The expression of specific proteins, such as MINK1 and PLEKHB1, provides a spatial coordinate system for astrocyte functional identity across different brain regions. Lipid metabolism (e.g., long-chain fatty acids) and mitochondrial function are primary drivers of the neurotoxic astrocyte phenotype in ischemic injury. The \"neurotoxic\" vs. \"neuroprotective\" paradigm for astrocyte activation is being replaced by the understanding that states are highly state-dependent and cannot be explained by simplified paradigms. Mechanical signaling via Piezo1, regulated by microglia-derived cytokines, links physical tissue alterations to the inflammatory profile of astrocytes. Transcriptional Heterogeneity:** Astrocytes exist in at least five distinct subpopulations following traumatic injury, with Osmr+ variants exhibiting specific neurotoxic and protective metabolic signatures. Mechanical Sensing:** Endothelial Piezo1 sensors translate mechanical stress into astrocytic apoptosis via cAMP-Epac1 microvesicular signaling. Gut-Brain Signaling:** Chronic enteric gliosis in Parkinson's disease-model mice (A53T) precedes CNS inflammation, driven by LRRK2 up-regulation. Barrier Regulation:** Astrocytes serve as primary regulators of the blood-brain barrier, often utilizing the cGAS-STING pathway to govern tight junction stability. Regenerative Potential:** \"Direct in situ astrocyte-to-neuron reprogramming offers a compelling regenerative alternative by leveraging the abundant endogenous glial reservoir,\" though this is hindered by existing epigenetic memory. Stress Resilience:** Structural depolymerization of AQP4 orthogonal array particles in A25Q mutant mice confers resilience to chronic stress by dampening glial-mediated neuroinflammation. Developmental Plasticity:** Adolescent intermittent ethanol exposure disrupts the physical and functional coupling of astrocytes to synapses, a deficit that persists into adulthood. Metabolic Rewiring:** Astrocytes undergo significant metabolic transitions during reprogramming, shifting from glycolysis to oxidative phosphorylation to support nascent neuronal survival. Astrocyte reactivity is not merely a binary 'A1/A2' state; modern transcriptomic analysis reveals finer gradations of cellular activation. The TRPC6-STING pathway represents a specific, druggable hub for stabilizing the blood-brain barrier via astrocytes during ischemia. Peripheral inflammation, as seen in atopic dermatitis or respiratory infection, directly reshapes cortical astrocytic transcriptional landscapes. Senescence markers in astrocytes and neurons represent a distinct, aging-associated inflammatory pathway mediated by cGAS-STING. Dietary interventions, such as a nut-enriched diet, can actively suppress pro-inflammatory astrocyte markers in AD mouse models. Clusterin (CLU) secretion from astrocytes, triggered by STING activation, is a primary driver of oligodendrocyte apoptosis in MS. FGF13 acts as a critical molecular switch that prevents astrocytic apoptosi","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21986940","URL":"https://doi.org/10.5281/zenodo.21986940","source":"datacite"},{"id":"doi:10.5281/zenodo.20653718","type":"article-journal","title":"Tabular Generative Evaluation Metrics Scaling in Multimodal Models Under Varying Adversarial Noise Conditions","abstract":"Synthetic data generation has emerged as a promising solution to overcome the challenges which are posed by data scarcity and privacy concerns, as well as, to address the need for training artificial intelligence (AI) algorithms on unbiased data with sufficient sample size and statistical power. Our review explores the application and efficacy of synthetic data methods in healthcare considering the diversity of medical data. To this end, we systematically searched the PubMed and Scopus databases with a great focus on tabular, imaging, radiomics, time-series, and omics data. Studies involving m Research goal: How do tabular generative evaluation metrics scale in performance when applied to multimodal models with varying degrees of adversarial noise in the training data? Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.3/10.","author":[{"family":"Kernel","given":"Sovereign"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20653718","URL":"https://doi.org/10.5281/zenodo.20653718","source":"datacite"},{"id":"doi:10.5281/zenodo.20653719","type":"article-journal","title":"Tabular Generative Evaluation Metrics Scaling in Multimodal Models Under Varying Adversarial Noise Conditions","abstract":"Synthetic data generation has emerged as a promising solution to overcome the challenges which are posed by data scarcity and privacy concerns, as well as, to address the need for training artificial intelligence (AI) algorithms on unbiased data with sufficient sample size and statistical power. Our review explores the application and efficacy of synthetic data methods in healthcare considering the diversity of medical data. To this end, we systematically searched the PubMed and Scopus databases with a great focus on tabular, imaging, radiomics, time-series, and omics data. Studies involving m Research goal: How do tabular generative evaluation metrics scale in performance when applied to multimodal models with varying degrees of adversarial noise in the training data? Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.3/10.","author":[{"family":"Kernel","given":"Sovereign"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20653719","URL":"https://doi.org/10.5281/zenodo.20653719","source":"datacite"},{"id":"oa:W4416132081","type":"article-journal","title":"Teachers’ professional identity in the era of artificial intelligence: A phenomenological study","abstract":"This study adopted a phenomenological research design to examine the lived experiences of teachers regarding their professional identity in response to the rapid expansion of artificial intelligence (AI) integration in education. The study involved thirty purposively selected teachers as participants. Data were collected through in-depth semi-structured interviews and analyzed using thematic analysis. Member checking was used to validate the findings. Data analysis followed a systematic phenomenological process, comprising thematic clustering, and synthesis of core themes. Trustworthiness was ensured through member checking, peer debriefing, and maintaining an audit trail. Five themes emerged from the findings; evolving professional roles, diverse emotional responses, emerging patterns of teacher-AI partnership, tensions between teacher autonomy and standardized instruction, and a redefinition of professional identity. The study revealed that teachers initially feared role erosion but gradually recognized the potential for AI to support, rather than replace, their teaching. Majority of the participants reported a transformation in their professional identity that signify a shift from traditional instruction toward facilitation in AI-mediated environments. It was recommended, among others, that rather than undermining the role of teachers, AI should be positioned to advance their professional identity within the classroom environment.","author":[{"family":"Ismail","given":"Akilu"},{"family":"Ibrahim","given":"Halimat"}],"issued":{"date-parts":[[2025]]},"DOI":"10.30935/ijpdll/17416","URL":"https://doi.org/10.30935/ijpdll/17416","source":"openalex"},{"id":"oa:W4413281273","type":"article-journal","title":"Ten Natural Language Processing Tasks with Generative Artificial Intelligence","abstract":"The review enumerates the predominant applications of large language models (LLMs) in natural language processing (NLP) tasks, with a particular emphasis on the years 2023 to 2025. A particular emphasis is placed on applications pertaining to information retrieval, named entity recognition, text or document classification, text summarization, machine translation, question-and-answer generation, fake news or hate speech detection, and sentiment analysis of text. Furthermore, metrics such as ROUGE, BERT, METEOR, BART, and BLEU scores are presented to evaluate the capabilities of a given language model. The following example illustrates the calculation of scores for the aforementioned metrics, utilizing sentences generated by ChatGPT 3.5, which is free and publicly available.","author":[{"family":"Golec","given":"Justyna"},{"family":"Hachaj","given":"Tomasz"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/app15169057","URL":"https://doi.org/10.3390/app15169057","source":"openalex"},{"id":"oa:W4408002164","type":"article-journal","title":"Plant leaf disease detection and classification using artificial intelligence techniques: a review","abstract":"Agriculture is a cornerstone of human civilization, providing both food and economic stability. While not necessarily fatal, leaf diseases are a crucial threat to plant health. Accurate detection and classification of diseases in early stages are essential to minimize damage. Manual identification can be challenging, and delays in detection can lead to crop devastation. Fortunately, computer-aided image processing offers a solution. Researchers have explored several techniques for disease detection and classification by usage of affected leaf images, making significant progress over time. However, there's always room for improvement. Machine learning (ML), Deep learning (DL) techniques have shown hopeful results. ML, DL approaches act as black-box; eXplainable AI (XAI) provides clear explanations on decisions made by these black-boxes. This study aims to present a comprehensive review on plant leaf disease detection and classification by means of ML, DL and XAI methods with an overview of the outcomes of existing techniques, summarizes their performance, evaluation metrics, and analyses the challenges in existing systems, and offers the study's inferences.","author":[{"family":"Kusuma","given":"R"},{"family":"Rajkumar","given":"R"}],"issued":{"date-parts":[[2025]]},"DOI":"10.11591/ijeecs.v38.i2.pp1308-1323","URL":"https://doi.org/10.11591/ijeecs.v38.i2.pp1308-1323","source":"openalex"},{"id":"oa:W4412391012","type":"article-journal","title":"Interaction, Artificial Intelligence, and Motivation in Children’s Speech Learning and Rehabilitation Through Digital Games: A Systematic Literature Review","abstract":"The integration of digital serious games into speech learning (rehabilitation) has demonstrated significant potential in enhancing accessibility and inclusivity for children with speech disabilities. This review of the state of the art examines the role of serious games, Artificial Intelligence (AI), and Natural Language Processing (NLP) in speech rehabilitation, with a particular focus on interaction modalities, engagement autonomy, and motivation. We have reviewed 45 selected studies. Our key findings show how intelligent tutoring systems, adaptive voice-based interfaces, and gamified speech interventions can empower children to engage in self-directed speech learning, reducing dependence on therapists and caregivers. The diversity of interaction modalities, including speech recognition, phoneme-based exercises, and multimodal feedback, demonstrates how AI and Assistive Technology (AT) can personalise learning experiences to accommodate diverse needs. Furthermore, the incorporation of gamification strategies, such as reward systems and adaptive difficulty levels, has been shown to enhance children’s motivation and long-term participation in speech rehabilitation. The gaps identified show that despite advancements, challenges remain in achieving universal accessibility, particularly regarding speech recognition accuracy, multilingual support, and accessibility for users with multiple disabilities. This review advocates for interdisciplinary collaboration across educational technology, special education, cognitive science, and human–computer interaction (HCI). Our work contributes to the ongoing discourse on lifelong inclusive education, reinforcing the potential of AI-driven serious games as transformative tools for bridging learning gaps and promoting speech rehabilitation beyond clinical environments.","author":[{"family":"Abdoulqadir","given":"Chra"},{"family":"Loizides","given":"Fernando"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/info16070599","URL":"https://doi.org/10.3390/info16070599","source":"openalex"},{"id":"oa:W4411175038","type":"article-journal","title":"Edge Artificial Intelligence Device in Real-Time Endoscopy for the Classification of Colonic Neoplasms","abstract":"Objective: Although prior research developed an artificial intelligence (AI)-based classification system predicting colorectal lesion histology, the heavy computational demands limited its practical application. Recent advancements in medical AI emphasize decentralized architectures using edge computing devices, enhancing accessibility and real-time performance. This study aims to construct and evaluate a deep learning-based colonoscopy image classification model for automatic histologic categorization for real-time use on edge computing hardware. Design: We retrospectively collected 2418 colonoscopic images, subsequently dividing them into training, validation, and internal test datasets at a ratio of 8:1:1. Primary evaluation metrics included (1) classification accuracy across four histologic categories (advanced colorectal cancer, early cancer/high-grade dysplasia, tubular adenoma, and nonneoplasm) and (2) binary classification accuracy differentiating neoplastic from nonneoplastic lesions. Additionally, an external test was conducted using an independent dataset of 269 colonoscopic images. Results: For the internal-test dataset, the model achieved an accuracy of 83.5% (95% confidence interval: 78.8–88.2%) for the four-category classification. In binary classification (neoplasm vs. nonneoplasm), accuracy improved significantly to 94.6% (91.8–97.4%). The external test demonstrated an accuracy of 82.9% (78.4–87.4%) in the four-category task and a notably higher accuracy of 95.5% (93.0–98.0%) for binary classification. The inference speed of lesion classification was notably rapid, ranging from 2–3 ms/frame in GPU mode to 5–6 ms/frame in CPU mode. During real-time colonoscopy examinations, expert endoscopists reported no noticeable latency or interference from AI model integration. Conclusions: This study successfully demonstrates the feasibility of a deep learning-powered colonoscopy image classification system designed for the rapid, real-time histologic categorization of colorectal lesions on edge computing platforms. This study highlights how nature-inspired frameworks can improve the diagnostic capacities of medical AI systems by aligning technological improvements with biomimetic concepts.","author":[{"family":"Gong","given":"Eun"},{"family":"Bang","given":"Chang"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/diagnostics15121478","URL":"https://doi.org/10.3390/diagnostics15121478","source":"openalex"},{"id":"oa:W4414142129","type":"article-journal","title":"Accuracy of Artificial Intelligence-Designed Dental Crowns: A Scoping Review of In-Vitro Studies","abstract":"Artificial intelligence (AI), particularly deep learning, is increasingly applied in dental prosthetics, offering new approaches to dental crown design. This scoping review aimed to summarize current evidence on AI-assisted crown design, focusing on algorithm types, dataset characteristics, and evaluation methods. A comprehensive search of PubMed, Scopus, Web of Science, and IEEE Xplore was conducted in February 2025, covering studies published between January 2010 and February 2025. Ten studies met the inclusion criteria, of which four developed custom AI models—mainly based on generative adversarial networks—while six evaluated commercially available software. All studies used digitized dental models obtained from scanned stone casts or intraoral scans, and dataset sizes varied widely. Morphological accuracy was the most frequently reported outcome, assessed in six studies, followed by design time and occlusal contact evaluation. While most AI-generated crowns demonstrated clinically acceptable precision, only four studies fabricated physical crowns and none conducted in vivo validation. These findings suggest that AI-assisted crown design holds promise for improving anatomical accuracy and workflow efficiency, but methodological heterogeneity and the lack of clinical validation highlight the need for standardized evaluation protocols and further in vivo studies.","author":[{"family":"Kong","given":"Hyun"},{"family":"Kim","given":"Yulee"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/app15189866","URL":"https://doi.org/10.3390/app15189866","source":"openalex"},{"id":"oa:W4414485818","type":"article-journal","title":"Artificial intelligence in healthcare: rethinking doctor-patient relationship in megacities","abstract":"Introduction: Artificial intelligence has been extensively applied in healthcare, offering significant potential to improve the quality of medical services. However, it also introduces critical challenges, such as privacy infringement, algorithmic discrimination, and ambiguous liability. The integration of artificial intelligence inevitably influences the doctor-patient relationship, which is pronounced in megacities. This study aims to explore the application of artificial intelligence in megacity healthcare system, exam the multidimensional transformation of the doctor-patient relationship and propose new governance frameworks. Methods: This study examines how artificial intelligence can effectively address systemic challenges within megacity healthcare systems while leveraging technological and institutional advantages to maximize its benefits, with a focus on Beijing as a primary case. Results: The integration of artificial intelligence inevitably influences the doctor-patient relationship, reducing information asymmetry, enhancing patient autonomy, and transforming the traditional doctor-patient dualistic interaction structure into a doctor-artificial intelligence-patient triad interaction structure. These effects are pronounced in megacities, presenting new challenges including crisis of trust, intensified disputes, and emotional and communication distance. Discussion: Given that the integration of artificial intelligence into healthcare is inevitable, especially for megacities like Beijing, proactive governance is essential. This includes institutionalizing the triad interaction model, deepening the integration of artificial intelligence in healthcare by leveraging the advantages of megacities, and establishing regulatory frameworks to mitigate risks while harnessing potential.","author":[{"family":"Chen","given":"Qi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/frhs.2025.1694139","URL":"https://doi.org/10.3389/frhs.2025.1694139","source":"openalex"},{"id":"oa:W4411035580","type":"article-journal","title":"Natural and artificial intelligence – the psychotechnical agenda of the 21st century","abstract":"The aim in this paper was to consider some of the fundamental and important differences between natural and artificial intelligence (AI). We are currently amid the third wave of AI technology, and for the first time in human history, it can be said that Large Language Models (LLMs) fulfil the original 1955 definition of AI, stipulating the simulation of human intelligence in a machine. Current AI systems are potentially very useful in many domains, and seem likely to revolutionise the way aspects of human work occur. However, considering their technical limitations, it seems unlikely that current AI will develop exponentially to the equivalent of human level intelligence and beyond. Alternatively, it also seems unlikely that we are heading towards another AI winter. Hence, the future would appear to contain many useful AI tools, but to fully understand and benefit from the third wave of AI we need to be aware that natural and artificial intelligence are in fact very different. Understanding these differences comprises the important psychotechnical agenda of our times. In terms of the Dual Process Theory of human cognition, LLMs are analogous, at best, to system-1 cognitive processing, and it remains unknown how anything approaching system-2 cognition could be incorporated in a machine. In addition, the machine learning that underlies third-wave AI comes with several down sides such as biases drawn from the training data, the propensity to hallucinate, and to produce AI systems that suffer from the black box problem. While many attempts to increase the transparency of AI systems are underway, all third wave AI systems remain black boxes, complicating efforts to establish their safety and reliability, and to fix performance problems when they are identified. Such limitations underscore the importance of retaining a human in the loop, at least in safety critical domains.","author":[{"family":"Webster","given":"Craig"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/29974100.2025.2491445","URL":"https://doi.org/10.1080/29974100.2025.2491445","source":"openalex"},{"id":"oa:W4412517761","type":"article-journal","title":"Artificial intelligence in radiology examinations: a psychometric comparison of question generation methods","abstract":"This study aimed to evaluate the usability of artificial intelligence (AI)-based question generation methods-Chat Generative Pre-trained Transformer (ChatGPT)-4o (a non-template-based large language model) and a template-based automatic item generation (AIG) method-in the context of radiology education.The primary objective was to compare the psychometric properties, perceived quality, and educational applicability of generated multiple-choice questions (MCQs) with those written by a faculty member. METHODSFifth-year medical students who participated in the radiology clerkship at Eskişehir Osmangazi University were invited to take a voluntary 15-question examination covering musculoskeletal and rheumatologic imaging.The examination included five MCQs from each of three sources: a radiologist educator, ChatGPT-4o, and the template-based AIG method.Student responses were evaluated in terms of difficulty and discrimination indices.Following the examination, students rated each question using a Likert scale based on clarity, difficulty, plausibility of distractors, and alignment with learning goals.Correlations between students' examination performance and their theoretical/practical radiology grades were analyzed using Pearson's correlation method. RESULTSA total of 115 students participated.Faculty-written questions had the highest mean correct response rate (2.91 ± 1.34), followed by template-based AIG (2.32 ± 1.66) and ChatGPT-4o (2.3 ± 1.14) questions (P < 0.001).The mean difficulty index was 0.58 for faculty, and 0.46 for both template-based AIG and ChatGPT-4o.Discrimination indices were acceptable (≥0.2) or very good (≥0.4) for template-based AIG questions.In contrast, four of the ChatGPT-generated questions were acceptable, and three were very good.Student evaluations of questions and the overall examination were favorable, particularly regarding question clarity and content alignment.Examination scores showed a weak correlation with practical examination performance (P = 0.041), but not with theoretical grades (P = 0.652). CONCLUSIONBoth the ChatGPT-4o and template-based AIG methods produced MCQs with acceptable psychometric properties.While faculty-written questions were most effective overall, AI-generated questions-especially those from the template-based AIG method-showed strong potential for use in radiology education.However, the small number of items per method and the single-institution context limit the robustness and generalizability of the findings.These results should be regarded as exploratory, and further validation in larger, multicenter studies is required. CLINICAL SIGNIFICANCEAI-based question generation may potentially support educators by enhancing efficiency and consistency in assessment item creation.These methods may complement traditional approaches to help scale up high-quality MCQ development in medical education, particularly in resource-limited settings; however, they should be applied with caution and expert oversight until further evidence is available, especially given the preliminary nature of the current findings.","author":[{"family":"Emekli","given":"Emre"},{"family":"Karahan","given":"Betül"}],"issued":{"date-parts":[[2025]]},"DOI":"10.4274/dir.2025.253407","URL":"https://doi.org/10.4274/dir.2025.253407","source":"openalex"},{"id":"oa:W4414397238","type":"article-journal","title":"Redefining assessment tasks to promote students’ creativity and integrity in the age of generative artificial intelligence","abstract":"Abstract The arrival of generative artificial intelligence (GenAI) has forced lecturers to adjust their assessment practices to ensure that students’ work is their own from a creative point of view, and free of plagiarism. This chapter proposes the Academic Integrity and Creativity in the Age of Artificial Intelligence (AICAI) model for the use of authentic assessment as a possible strategy to promote students’ creativity and integrity and thereby ensure the ownership of their written work. Lecturers are encouraged to rethink the assignments they design and examine each of the following components with an eye to integrity: their professional characteristics, the objectives for the assignment, the type of assessment that is appropriate for the needs of the student. Others include the cognitive offloading that will be done or not with GenAI, the type of authentic task they wish to propose and its characteristics, and the instructions and criteria that will be given to students. The choices made should engage students, thereby diminishing the temptation to plagiarize. By combining different strands of pedagogical theory and research, the AICAI assessment design model proposed in this paper has brought into focus the challenges as well as the opportunities that have emerged with the inclusion of GenAI in higher education. On a more practical level, it offers a systemic approach and advice as to how the challenges can be mitigated and benefits maximized for all parties involved in assessment.","author":[{"family":"Peters","given":"Martine"},{"family":"Angelov","given":"Dimitar"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s40979-025-00201-x","URL":"https://doi.org/10.1007/s40979-025-00201-x","source":"openalex"},{"id":"oa:W4415380784","type":"article-journal","title":"Artificial intelligence and students’ cognitive learning outcomes with bibliometric and content analysis for future research agenda","abstract":"This study conducted a comprehensive bibliometric and content analysis to explore the integration of artificial intelligence in students’ cognitive learning outcomes. A structured TITLE-ABS-KEY search was performed in the Scopus database using keywords such as “Artificial Intelligence,” “AI,” “Students,” and “cognitive learning outcomes,” resulting in 318 documents published between 2016 and 2025. After filtering, a final dataset of 246 research articles and conference papers was analyzed. The methodology includes bibliometric performance analysis (covering publication trends, countries, affiliations, authors, and journals) and network analysis (comprising co-word, citation, co-authorship, and bibliographic coupling). Additionally, content analysis was conducted on the ten most cited and ten focused articles addressing AI’s impact on cognitive learning outcomes. VOSviewer software was used for data analysis and visualization. Findings indicate increased research output post-2023, driven by digital transformation and global collaboration. Leading affiliations include The University of Hong Kong and Carnegie Mellon University, with the United States, China, and India as top contributing countries. Influential journals and funding bodies include the National Science Foundation and the National Natural Science Foundation of China. Notable authors include Chiu and Cukurova, while Kit Ng, Zhong, and Liu are prominent in bibliographic coupling, emphasizing AI adoption. Co-authorship analysis shows collaboration primarily among developed nations. Co-word analysis reveals Key research themes include contrastive learning, adversarial machine learning, and federated learning. Content analysis highlights AI’s transformative potential for learning, teaching, cognitive learning, and innovation. This study provides managerial and practical recommendations for students, universities, and policymakers. This study has several limitations that future studies will consider.","author":[{"family":"Ansari","given":"Shaukat"},{"family":"Qamari","given":"Ika"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s44217-025-00865-0","URL":"https://doi.org/10.1007/s44217-025-00865-0","source":"openalex"},{"id":"oa:W4414253312","type":"article-journal","title":"The impact of integrating artificial intelligence and Building information modeling (BIM) systems on the development of construction methodologies","abstract":"Abstract The integration of Artificial Intelligence (AI) with Building Information Modelling (BIM) systems is driving a profound transformation in project delivery approaches, necessitating a systematic evaluation of its impact on design and construction processes. This study explores how AI-BIM integration enhances construction efficiency, with a focus on improving accuracy, productivity, and decision-making across project lifecycles. The research employs a mixed-methods approach, combining theoretical and applied analyses. Through descriptive-analytical methods, we assessed conceptual frameworks and the latest technological advancements were assessed, while deductive analysis was applied to real-world case studies. A practical case study of a construction project utilizing AI-BIM integration was examined to evaluate its benefits and challenges. The findings reveal significant improvements in construction methodologies, including: Enhanced design automation through AI-driven generative solutions. Faster structural and energy performance analysis using machine learning algorithms. Greater planning accuracy and improved decision-making in early project phases. Reduced errors and costs via intelligent automation and predictive modelling. The study recommends: Developing standardized frameworks for AI-BIM integration in construction practices, Implementing specialized training programs to upskill professionals in AI and BIM technologies Conducting further research to measure long-term economic and time-saving impacts of AI-BIM adoption.","author":[{"family":"Attia","given":"Ahmed"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s43995-025-00193-2","URL":"https://doi.org/10.1007/s43995-025-00193-2","source":"openalex"},{"id":"oa:W4412190336","type":"article-journal","title":"Artificial Intelligence and Ethical Dimensions of Automated Traffic Enforcement: Implications for Public Health, Healthcare Equity, and Social Justice","abstract":"This study provides a critical examination of AI-integrated speed and red-light camera systems through the theoretical lenses of Surveillance Capitalism, the Panopticon Model, Social Control Theory, Technological Determinism, and Structural Violence Theory. While artificial intelligent speed safety cameras demonstrate efficacy in reducing traffic violations and fatalities, this research addresses a critical gap in healthcare literature regarding their broader societal and ethical consequences, including algorithmic bias, data governance failures, and privacy violations that directly impact public trust and health equity. The analysis reveals how machine learning and predictive analytics in automated enforcement create disproportionate burdens on marginalized populations through three specific mechanisms: (1) biased algorithmic design that targets low-income neighborhoods more intensively, (2) punitive traffic fine structures that impose greater relative financial hardship on economically disadvantaged families, and (3) opaque implementation practices that limit community understanding and participation. These patterns perpetuate health disparities by increasing chronic stress, economic instability, and barriers to healthcare access among vulnerable populations. This work’s novel contribution lies in applying four foundational health equity principles to AI-powered traffic enforcement: distributive justice (fair allocation of enforcement across communities), procedural justice (transparent and accountable decision-making processes), recognition justice (acknowledgment of community voices and concerns), and capabilities approach (ensuring enforcement practices do not undermine individuals’ fundamental capabilities for health and wellbeing). Additionally, the study examines three core social justice principles: substantive equality (addressing systemic disadvantages rather than treating all violations identically), participatory parity (ensuring affected communities can participate meaningfully in policy decisions), and non-domination (preventing the arbitrary exercise of state power through automated systems). The study advocates for the development of ethical artificial intelligence governance frameworks that incorporate transparent algorithmic auditing, community driven design processes, and robust oversight mechanisms. These evidence-based recommendations support equitable and trustworthy applications of artificial intelligence that advocate for, rather than undermine, population health and social justice in traffic safety initiatives. A novel contribution of this work lies in its exploration of how artificial intelligence powered speed safety cameras intersect with specific health equity principles in distributive justice, procedural justice, and the capabilities approach, as well as core social justice principles, including substantive equality, participatory parity, and nondomination, in the governance of public infrastructure. The analysis applies distributive justice to examine the fair allocation of enforcement across communities, procedural justice to evaluate transparent decision-making processes, and the capabilities approach to assess whether enforcement practices undermine individuals’ fundamental capabilities for health and wellbeing.","author":[{"family":"Haley","given":"Patricia"}],"issued":{"date-parts":[[2025]]},"DOI":"10.61093/hem.2025.2-03","URL":"https://doi.org/10.61093/hem.2025.2-03","source":"openalex"},{"id":"oa:W4415626372","type":"article-journal","title":"Artificial intelligence, machine learning and omic data integration in osteoarthritis","abstract":"OBJECTIVE: Artificial intelligence (AI), particularly its subfield of machine learning (ML), offer promising tools for integrating and interpreting high-dimensional omic data to advance our understanding of osteoarthritis (OA), a complex, multifactorial disease. The objective of this review is to summarize recent progress in applying ML approaches to single and integrative multi-omic data in OA and to highlight emerging trends, challenges, and opportunities. METHOD: We conducted a literature search of PubMed and preprint databases upto April 2025. This search identified studies that applied ML techniques including supervised learning, unsupervised clustering, deep learning, and integrative modeling to OA datasets. These datasets included transcriptomic, epigenomic, proteomic, metabolomic, and multi-omic profiles in human OA samples and relevant preclinical models. We synthesized findings across omic types, ML methodologies, and clinical or mechanistic OA outcomes, highlighting key trends in multi-omic integration strategies and their implications for OA research. RESULTS: Recent studies have applied ML to identify transcriptomic and epigenomic biomarkers, stratify OA patient subtypes, and predict disease progression. Advanced approaches such as variational autoencoders, contrastive learning, and multimodal transformers are emerging as powerful tools for multi-omic integration. However, challenges remain related to small sample sizes, overfitting, lack of external validation, model interpretability, and demographic underrepresentation in omic datasets. CONCLUSIONS: ML techniques are advancing OA research by enabling nuanced analysis of complex omic datasets. Addressing current limitations and embracing new developments in spatial and single-cell omics, generative models, and federated learning will be essential to unlock the full potential of multi-omic integration for personalized OA diagnosis and treatment.","author":[{"family":"Sharma","given":"Divya"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.joca.2025.10.012","URL":"https://doi.org/10.1016/j.joca.2025.10.012","source":"openalex"},{"id":"oa:W4414289079","type":"article-journal","title":"Artificial Intelligence-enabled smart grid systems for real-time load forecasting, fault detection, renewable energy integration and optimization","abstract":"The global energy landscape is undergoing rapid transformation, driven by increasing electricity demand, decarbonization goals, and the integration of renewable energy resources. Traditional power systems, characterized by centralized operations and limited responsiveness, are increasingly inadequate for managing modern grid complexities. Smart grid systems have emerged as a critical paradigm, embedding advanced sensors, communication infrastructures, and intelligent decision-making mechanisms to enhance grid stability and reliability. However, the scale and unpredictability of dynamic energy flows demand advanced solutions that surpass conventional optimization and monitoring methods. Artificial intelligence (AI) has proven to be a transformative enabler in this context, offering predictive, adaptive, and real-time decision-making capabilities. By leveraging machine learning and deep learning algorithms, AI-enabled smart grids can accurately forecast load demand, detect and isolate faults, and dynamically balance distributed energy resources. Predictive load forecasting allows operators to anticipate fluctuations in consumer demand, ensuring efficiency and stability, while AI-driven fault detection systems improve resilience by rapidly identifying anomalies and preventing cascading failures. Furthermore, AI supports the seamless integration of renewable energy sources, mitigating intermittency challenges by optimizing grid dispatch and storage solutions. This paper explores the convergence of artificial intelligence with smart grid infrastructures, emphasizing its applications in real-time load forecasting, fault detection, renewable energy integration, and system-wide optimization. It also addresses associated challenges such as data privacy, model interpretability, and cybersecurity risks, offering a balanced discussion of opportunities and limitations. Ultimately, AI-enabled smart grids represent a pivotal step toward building resilient, sustainable, and intelligent energy ecosystems capable of supporting the future of decentralized, low-carbon power systems.","author":[{"family":"Iyaniwura","given":"Abdulrahman"},{"family":"Mayaki","given":"Charles"}],"issued":{"date-parts":[[2025]]},"DOI":"10.30574/gjeta.2025.24.3.0272","URL":"https://doi.org/10.30574/gjeta.2025.24.3.0272","source":"openalex"},{"id":"oa:W4410601609","type":"article-journal","title":"Artificial intelligence in conformance checking: state of the art and research agenda","abstract":"Artificial intelligence methods have gained increasing importance in business process management in recent years. In particular, artificial intelligence boosted the development of process mining techniques, with predictive process monitoring being one prominent example. However, for some branches of process mining, little attention has been paid in the literature to understanding potential opportunities offered by artificial intelligence. In this work, we focus on conformance checking. This discipline is gaining traction in both research and practical applications due to its capability of revealing inconsistencies between models and event logs. Though state-of-the-art conformance checking has been investigated recently, there exists a lack of insights on whether conformance checking can benefit from recent developments in artificial intelligence. This paper addresses this research gap through a systematic literature review of conformance checking approaches contrasted against trends in artificial intelligence research extracted from papers published in core artificial intelligence venues. This comparative analysis extracts prominent trends in both disciplines, highlighting potential overlaps and topics of common interests. Elaborating upon such overlaps, we identify promising research avenues leveraging AI to address open challenges in conformance checking, as well as recent AI trends which may lead to interesting developments for the conformance checking community.","author":[{"family":"Genga","given":"Laura"},{"family":"Winter","given":"Karolin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s44311-025-00015-7","URL":"https://doi.org/10.1007/s44311-025-00015-7","source":"openalex"},{"id":"oa:W4410516058","type":"article-journal","title":"Empathy, Ethics and Efficacy: The 3Es of Implementing Artificial Intelligence for Consumer Encounters","abstract":"ABSTRACT Drawing insights from virtue ethics and psychological perspectives on empathy, we propose a framework for integrating empathy, ethics, and efficacy (3E) to guide the responsible development and deployment of artificial intelligence (AI) technologies in consumer service encounters. It encourages developers, vendors, and users to improve efficiencies and productivity but, more importantly, to adopt an empathetic perspective with ethical decision‐making during AI development and deployment. By simultaneously adopting descriptive, analytical, and prescriptive approaches to engage academics, practitioners, and public policymakers, we outline a future research agenda for enriching AI‐consumer service encounters. Beyond conceptual integration, the framework offers practical insights for AI designers, businesses, and regulators by emphasising empathy as a bridge between efficacy and ethics. This perspective supports the development of AI technologies that not only enhance operational effectiveness but also foster consumer trust and well‐being, ensuring AI‐driven services remain human‐centred and ethically sound. A structured decision‐making model demonstrates how AI‐driven services can balance automation with ethical considerations and empathetic engagement, offering a pathway for more responsible AI implementation.","author":[{"family":"Fukukawa","given":"Kyoko"},{"family":"Trivedi","given":"Rohit"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/mar.22235","URL":"https://doi.org/10.1002/mar.22235","source":"openalex"},{"id":"oa:W4413329932","type":"article-journal","title":"Artificial Intelligence for Multiscale Spatial Analysis in Oncology: Current Applications and Future Implications","abstract":"Artificial intelligence (AI) and its machine learning and deep learning algorithms have shown promise in oncological practice. Spatial information analysis in the context of cancer is crucial for its diagnosis and treatment because it can provide an understanding of tumor-microenvironment interactions and reveal insights into response to treatment. AI tools can analyze spatial information at multiple scales, highlighting key disease, clinical, and genetic phenotypes that may reveal underlying mechanisms and molecular markers of response and resistance within the tumor and its microenvironment. By examining tumor interactions at macroscopic (diagnostic imaging) and microscopic (pathology slides and spatial biology) levels, AI can assist in making important diagnostic and prognostic decisions. In this review, we first present an overview of AI and the need for multiscale spatial information in oncology. Then, we examine growing AI applications in the analysis of such information, focusing on diagnostic imaging, digital pathology, and spatial molecular biology. We also discuss applications of large-scale foundation models and task-oriented agentic AI in these fields as emergent technologies. Then, we discuss current limitations for the clinical translation of AI into regular utilization in cancer care and discovery.","author":[{"family":"Tarhini","given":"Ali"},{"family":"Naqa","given":"Issam"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ijms26168002","URL":"https://doi.org/10.3390/ijms26168002","source":"openalex"},{"id":"oa:W4410280064","type":"article-journal","title":"International Standardization Safe to Use of Artificial Intelligence","abstract":"Nowadays, the symbiosis of human abilities and the mastery of artificial intelligence will contribute to increased productivity and excellence in industry and social services. The use of artificial intelligence in various fields requires standardization of the safety of its knowledge and skills. International collaboration on artificial intelligence safety standardization is expanding. The UN has created a Global Advisory Body on Artificial Intelligence to support the efforts of the international community of specialists in managing intelligent systems related to the risks and safety of their use. The author proposes international standard of safe application of ensemble intelligent interoperable agents. Ensembles of agents with artificial intelligence are multi-agent synergistic self-organizing systems that function according to the laws of development, synergy and self-organization. Ensembles of intellectual agents solve the problem in the course of self-organization and cooperation according to the criteria of preference and restriction. The solution is considered found when, in the course of their nondeterministic interactions, agents reach the best consensus (temporary equilibrium or balance of interests), which is taken as a solution to the problem. The advantages of intelligent agents that allow you to build self-organizing ensembles are especially manifested in conditions of a priori uncertainty and high dynamics of the world around you, allowing you to build adaptive ensembles with communicative abilities, rebuilding your plans for events in real time. The higher the intelligence of each agent and the richer the opportunities for communication between agents, the more complex and creative behavior the ensemble can demonstrate. The intellect of the ensemble arises and manifests itself in the process of self-organization of intellectual agents. Intelligent agents use a physical, informal and logical model of the environment. That is, they use both attributes and sets of entities, processes, relationships, etc. Modern technologies allow you to create ensembles of intelligent agents with communication abilities, characterized by high openness, flexibility and efficiency, performance, scalability, reliability and survivability, approaching the intellectual abilities of a person and professional teams in their cognitive and functional capabilities and even sometimes surpassing them.","author":[{"family":"Bryndin","given":"Evgeniy"}],"issued":{"date-parts":[[2025]]},"DOI":"10.25082/rima.2025.01.005","URL":"https://doi.org/10.25082/rima.2025.01.005","source":"openalex"},{"id":"oa:W4410211382","type":"article-journal","title":"The long journey of artificial intelligence in medicine: an overview","abstract":"Artificial intelligence (AI) has its roots in the history of philosophy and of applied mathematics of the 17th, 18th and 19th centuries. Throughout the 20th century, significant advancements in mathematics and computer science laid the groundwork for modern AI, culminating in the establishment of the field as a formal discipline during the Dartmouth Conference in 1956.This pivotal event brought together leading researchers who envisioned creating machines capable of simulating human intelligence, setting the stage for decades of research and innovation in the field. The development of early AI systems focused on problem-solving and symbolic reasoning, leading to the creation of programmes that could play games like chess and solve mathematical equations, which show-cased the potential of machines to perform tasks previously thought to require human intellect.As these foundational systems evolved, researchers began to explore more complex algorithms and learning models, paving the way for advancements in machine learning and neural networks that would eventually revolutionise AI applications across various fields among which medicine. The growth of big data and increased computational power further accelerated these advancements, enabling machines to analyse vast amounts of health information and learn from patterns at unprecedented speeds.The revolution of deep learning and soon after large language models has enabled machines to achieve remarkable feats, such as image and speech recognition, natural language processing, and even creative tasks like art generation, pushing the boundaries of what was once thought possible. As organisations grapple with these challenges, there is growing emphasis on developing frameworks that ensure responsible AI deployment while maximising its potential benefits for human health.","author":[{"family":"Grossi","given":"Enzo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.55563/clinexprheumatol/oamfed","URL":"https://doi.org/10.55563/clinexprheumatol/oamfed","source":"openalex"},{"id":"oa:W4413137895","type":"article-journal","title":"Understanding the Artificial Intelligence Revolution and its Ethical Implications","abstract":"Recent artificial intelligence (AI) advancements have precipitated profound ethical deliberations and societal concerns. These developments redefine the parameters of technology's role in our daily lives and challenge our understanding of ethics in the context of AI-enabled processes. As AI systems become more integrated into various facets of human activity, from healthcare to finance and from social interactions to governance, the ethical implications of these technologies have become increasingly complex and pressing. In this paper, we aim to facilitate the understanding of intelligence and Artificial Intelligence and delve into the transformative impact of the AI revolution on societal norms and ethical frameworks. We spotlight the critical ethical questions and concerns that arise as AI technologies become increasingly embedded in various aspects of human life. We provide a brief overview of ethical strategies in AI development and explore how implementing these strategies can mitigate potential risks, promote responsible innovation, and ensure the alignment of AI technologies with societal values.","author":[{"family":"Beheshti","given":"Amin"},{"family":"Kerridge","given":"Ian"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s11673-025-10427-6","URL":"https://doi.org/10.1007/s11673-025-10427-6","source":"openalex"},{"id":"oa:W4414490565","type":"article-journal","title":"Factors Contributing to Higher Education Students' Acceptance of Artificial Intelligence: A Systematic Review","abstract":"The rapid integration of artificial intelligence (AI) technologies into the field of higher education is causing widespread public discourse. However, existing research is fragmented and lacks systematic synthesis, which limits understanding of how college and university students adopt artificial intelligence technologies. To address this gap, we conducted a systematic review following the guidelines of the PRISMA statement, including studies from ScienceDirect, Web of Science, Scopus, PsycARTICLES, SOC INDEX, and Embase databases. A total of 5594 articles were identified in the database search; 112 articles were included in the review. The criteria for inclusion in the review were: (i) publication date; (ii) language; (iii) participants; (iv) object of research. The results of the study showed: (a) The Technology Acceptance Model and the Unified Theory of Technology Acceptance and Use are most often used to explain the AI acceptance; (b) quantitative research methods prevail; (c) AI is mainly used by students to search and process information; (d) technological factors are the most significant factors of AI acceptance; (e) gender, specialty, and country of residence influence the AI acceptance. Finally, several problems and opportunities for future research are highlighted, including problems of psychological well-being, students’ personal and academic development, and the importance of financial, educational, and social support for students in the context of widespread artificial intelligence.","author":[{"family":"Mukhamedkarimova","given":"Dinara"},{"family":"Umurkulova","given":"Madina"}],"issued":{"date-parts":[[2025]]},"DOI":"10.12973/eu-jer.14.4.1373","URL":"https://doi.org/10.12973/eu-jer.14.4.1373","source":"openalex"},{"id":"oa:W4413443615","type":"article-journal","title":"The Role of Artificial Intelligence in Advancing Theranostics Dosimetry for Cancer Therapy: a Review","abstract":"Cancer treatment has greatly benefited from advancements in radiopharmaceutical therapy, which requires precise dosimetry to enhance therapeutic efficacy and minimize risks to healthy tissues. This review investigated the role of artificial intelligence (AI) in theranostic radiopharmaceutical dosimetry, focusing on image quality enhancement, dose estimation, and organ segmentation. An in-depth review of the literature was conducted using targeted keywords searches in Google Scholar, PubMed, and Scopus. Selected studies were evaluated for their methodologies and outcomes. Traditional dosimetry techniques such as organ-level and voxel-based methods are discussed. Deep learning (DL) models based on U-Net, generative adversarial networks, and hybrid transformer networks for image synthesis and generation, image quality improvement, organ segmentation, and radiation dose estimation are reviewed and discussed. While DL shows great potential for enhancing dosimetry accuracy and efficiency, challenges such as the need for accurate dose estimation from theranostic pairs, lack of imaging data, and modeling of radionuclide decay chains must be addressed using DL models. In addition, the optimization and standardization of DL and AI models is crucial for ensuring clinical reliability and should be given high priority to support their effective integration into clinical practice.","author":[{"family":"Woo","given":"Sang‐keun"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s13139-025-00939-9","URL":"https://doi.org/10.1007/s13139-025-00939-9","source":"openalex"},{"id":"oa:W4416076331","type":"manuscript","title":"Ontology-based knowledge representation for bone disease diagnosis: a foundation for safe and sustainable medical artificial intelligence systems","abstract":"Medical artificial intelligence (AI) systems frequently lack systematic domain expertise integration, potentially compromising diagnostic reliability. This study presents an ontology-based framework for bone disease diagnosis, developed in collaboration with Ho Chi Minh City Hospital for Traumatology and Orthopedics. The framework introduces three theoretical contributions: (1) a hierarchical neural network architecture guided by bone disease ontology for segmentation-classification tasks, incorporating Visual Language Models (VLMs) through prompts, (2) an ontology-enhanced Visual Question Answering (VQA) system for clinical reasoning, and (3) a multimodal deep learning model that integrates imaging, clinical, and laboratory data through ontological relationships. The methodology maintains clinical interpretability through systematic knowledge digitization, standardized medical terminology mapping, and modular architecture design. The framework demonstrates potential for extension beyond bone diseases through its standardized structure and reusable components. While theoretical foundations are established, experimental validation remains pending due to current dataset and computational resource limitations. Future work will focus on expanding the clinical dataset and conducting comprehensive system validation.","author":[{"family":"Dao","given":"Loan"},{"family":"Ly","given":"Ngoc"}],"issued":{"date-parts":[[2025]]},"DOI":"10.48550/arxiv.2506.04756","URL":"https://doi.org/10.48550/arxiv.2506.04756","source":"openalex"},{"id":"oa:W7106016604","type":"article-journal","title":"Artificial intelligence for medical imaging: U-Net technology for anatomical feature analysis","abstract":"The successful utilization of artificial intelligence (AI) systems in medical and healthcare systems has substantially advanced research and publication in scientific journals. This field encompasses a wide range of studies, including image processing, natural language processing, medical physics, patient data analysis, and clinical assistance tools. The current progress in AI methods can be attributed to substantial improvements in computational capacity and data processing capabilities. Notably, computer vision and image processing have emerged as highly successful AI applications. The U-Net convolutional neural network has emerged as a powerful and efficient tool for medical image segmentation and processing. This model features an encoder-decoder configuration interconnected by a bridging element, with skip connections between layers that enhance the value of the original training data. Its impressive efficiency in image processing stems from its rapid processing capability, ability to extract relationships from data, and high training velocity. Medical imagery often comprises multiple cross-sectional slices, providing a volumetric perspective of the observed region. Analyzing such imaging data requires substantial computational power and storage capacity, especially for 3-dimensional (3D) analysis. In this regard, 3D U-Net networks excel by concurrently processing numerous slices in voxel space. This attribute considerably reduces computational expenses while simultaneously improving precision. Recent advancements have brought AI technology developers to a level of stability and positive predictive values that make routine use of these systems in medical devices feasible. Currently, the implementation of AI systems appears more realistic than in previous decades. This review article focuses on state-of-the-art technologies in medical imaging for monitoring and diagnostic purposes, specifically using U-Net. We have reviewed the quality measurement of AI imaging systems using gold standards and explored novel technologies that have not been discussed in previous U-Net review papers. Additionally, we discuss the promising future development of AI systems for medical imaging purposes.","author":[{"family":"Asadpour","given":"Vahid"},{"family":"Xie","given":"Fagen"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.imed.2025.07.003","URL":"https://doi.org/10.1016/j.imed.2025.07.003","source":"openalex"},{"id":"oa:W4413850559","type":"article-journal","title":"Transforming Medical Microbiology: The Role of Artificial Intelligence","abstract":"Traditional microbiological techniques, while effective, are often time-consuming and labour-intensive. Machine learning and deep learning, enable rapid and accurate identification of microbial pathogens from complex datasets such as whole-genome sequencing, mass spectrometry, and clinical laboratory reports. Artificial Intelligence (AI) revolutionizes medical microbiology by enhancing pathogen detection, antimicrobial resistance prediction, and clinical decision-making. AI facilitates automated image analysis for culture-based diagnostics, improving the speed and accuracy of colony identification and antimicrobial susceptibility testing. One of the most impactful applications of AI is in antimicrobial resistance (AMR) surveillance. Machine learning models can analyse genetic determinants of resistance and predict antimicrobial susceptibility patterns, allowing for early detection of multidrug-resistant organisms. Moreover, AI-integrated clinical decision support systems (CDSS) enhance antimicrobial stewardship by providing real-time recommendations on appropriate antibiotic use, thereby reducing the spread of resistance. Natural language processing (NLP) further optimizes data extraction from electronic health records, improving diagnostic workflows and patient outcomes. Despite its transformative potential, challenges such as data standardization, model interpretability, and integration into routine laboratory workflows must be addressed. Ethical considerations, including data privacy and algorithmic bias, also warrant careful attention. As AI continues to evolve, its synergy with microbiology will pave the way for precision diagnostics, personalized treatment strategies, and global AMR mitigation. Leveraging AI-driven innovations will be crucial in shaping the future of infectious disease diagnostics and public health microbiology.","author":[{"family":"Sande","given":"Suvarna"},{"family":"Rajguru","given":"Manisha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.22207/jpam.19.3.36","URL":"https://doi.org/10.22207/jpam.19.3.36","source":"openalex"},{"id":"doi:10.5281/zenodo.18870276","type":"article-journal","title":"TCMNSCLC: A Real-world Dataset for Chinese Medicine Reasoning on Non-small-cell Lung Cancer","abstract":"TCM4NSCLC is a real-world dataset developed for traditional Chinese medicine (TCM) reasoning in non-small-cell lung cancer (NSCLC). The dataset consists of structured clinical cases curated from real-world medical records and annotated by experienced TCM experts. Each case contains comprehensive patient information, including clinical characteristics, TCM syndrome differentiation, treatment principles, herbal decoction prescriptions, and Chinese patent medicine recommendations. The dataset is designed to facilitate research on large language models (LLMs) for TCM clinical reasoning, clinical decision support, prescription generation, and explainable medical artificial intelligence. The dataset provides the following information for each clinical case: Structured clinical case descriptions TCM syndrome differentiation labels Treatment principles Herbal decoction prescriptions Chinese patent medicine recommendations The dataset contains the following files: File Description TCM4NSCLC_Original_Clinical_Dataset.xlsx Original structured clinical dataset collected from real-world medical records. Compared with the JSON files, this file additionally includes patient demographic and visit information, such as patient_id, sex, age, accrual_type, and visit_date. It serves as the source dataset before splitting into training, validation, and test sets. train_num_3032.json Training set containing 3,032 clinical cases for model training. valid_num_379.json Validation set containing 379 clinical cases for hyperparameter tuning and model selection. test_num_379.json Test set containing 379 clinical cases for final evaluation. example_en.json An English example demonstrating the dataset format and field definitions. schema.json JSON schema describing the structure and data types of each field. README.md Documentation describing the dataset and usage instructions. The released dataset is divided into three subsets: Split Number of Cases Train 3,032 Validation 379 Test 379 Total 3,790 The training, validation, and test sets are generated from the original structured clinical dataset. The original Excel file additionally preserves patient demographic and visit-related metadata that are not included in the released JSON files for model training.","author":[{"family":"Zhang","given":"Xinxin"},{"family":"Zhang","given":"Chuchu"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18870276","URL":"https://doi.org/10.5281/zenodo.18870276","source":"datacite"},{"id":"doi:10.5281/zenodo.21027568","type":"article-journal","title":"TCMNSCLC: A Real-world Dataset for Chinese Medicine Reasoning on Non-small-cell Lung Cancer","abstract":"TCM4NSCLC is a real-world dataset developed for traditional Chinese medicine (TCM) reasoning in non-small-cell lung cancer (NSCLC). The dataset consists of structured clinical cases curated from real-world medical records and annotated by experienced TCM experts. Each case contains comprehensive patient information, including clinical characteristics, TCM syndrome differentiation, treatment principles, herbal decoction prescriptions, and Chinese patent medicine recommendations. The dataset is designed to facilitate research on large language models (LLMs) for TCM clinical reasoning, clinical decision support, prescription generation, and explainable medical artificial intelligence. The dataset provides the following information for each clinical case: Structured clinical case descriptions TCM syndrome differentiation labels Treatment principles Herbal decoction prescriptions Chinese patent medicine recommendations The dataset contains the following files: File Description TCM4NSCLC_Original_Clinical_Dataset.xlsx Original structured clinical dataset collected from real-world medical records. Compared with the JSON files, this file additionally includes patient demographic and visit information, such as patient_id, sex, age, accrual_type, and visit_date. It serves as the source dataset before splitting into training, validation, and test sets. train_num_3032.json Training set containing 3,032 clinical cases for model training. valid_num_379.json Validation set containing 379 clinical cases for hyperparameter tuning and model selection. test_num_379.json Test set containing 379 clinical cases for final evaluation. example_en.json An English example demonstrating the dataset format and field definitions. schema.json JSON schema describing the structure and data types of each field. README.md Documentation describing the dataset and usage instructions. The released dataset is divided into three subsets: Split Number of Cases Train 3,032 Validation 379 Test 379 Total 3,790 The training, validation, and test sets are generated from the original structured clinical dataset. The original Excel file additionally preserves patient demographic and visit-related metadata that are not included in the released JSON files for model training.","author":[{"family":"Zhang","given":"Xinxin"},{"family":"Zhang","given":"Chuchu"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21027568","URL":"https://doi.org/10.5281/zenodo.21027568","source":"datacite"},{"id":"doi:10.5281/zenodo.19902378","type":"article-journal","title":"A Lacuna Regulatória entre o AI Act e o RGPD: Heteronomia Decisória Invisível em Contexto Jurisdicional e Clínico","abstract":"A crescente integração de sistemas de Inteligência Artificial em contextos decisórios críticos tem vindo a ser acompanhada por um esforço regulatório relevante, designadamente através do Regulamento (UE) 2024/1689 (AI Act) e do Regulamento (UE) 2016/679 (RGPD). O presente trabalho identifica uma dimensão complementar, ainda insuficientemente tratada por ambos os instrumentos: o modo como a informação é estruturada antes de ser submetida à decisão humana. Para designar o fenómeno que esta lacuna permite, retoma-se o conceito de heteronomia decisória invisível e aplica-se em analogia estrutural ao contexto jurisdicional e ao contexto clínico. O paper analisa as implicações jurídicas da lacuna em termos de responsabilidade do decisor, efetividade da supervisão humana e limites da explainability, e formula uma proposta dirigida ao regulador europeu e ao legislador português, assente no reconhecimento do processo de estruturação da informação como objeto regulatório autónomo. Nelson Marques | Jurista - Direito Digital e Regulação da Inteligência Artificial | e-mail: nelsonailaw@gmail.com | https://www.linkedin.com/in/nelsonmarques-ailaw | Tipologia: Documento de Posição (Position Paper) - Abril 2026 | Trabalho anterior: DOI 10.5281/zenodo.19446330 | “O objetivo não é questionar a autonomia do decisor, mas assegurar que essa autonomia se exerce sobre todas as dimensões relevantes do processo de decisão.” Keywords: artificial intelligence, medical practice, decisional heteronomy, clinical autonomy, AI Act, informed consent, medical deontology, automation bias, cognitive heteronomy, shadow AI, physician governance.","author":[{"family":"Marques","given":"Nelson"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19902378","URL":"https://doi.org/10.5281/zenodo.19902378","source":"datacite"},{"id":"doi:10.5281/zenodo.19902379","type":"article-journal","title":"A Lacuna Regulatória entre o AI Act e o RGPD: Heteronomia Decisória Invisível em Contexto Jurisdicional e Clínico","abstract":"A crescente integração de sistemas de Inteligência Artificial em contextos decisórios críticos tem vindo a ser acompanhada por um esforço regulatório relevante, designadamente através do Regulamento (UE) 2024/1689 (AI Act) e do Regulamento (UE) 2016/679 (RGPD). O presente trabalho identifica uma dimensão complementar, ainda insuficientemente tratada por ambos os instrumentos: o modo como a informação é estruturada antes de ser submetida à decisão humana. Para designar o fenómeno que esta lacuna permite, retoma-se o conceito de heteronomia decisória invisível e aplica-se em analogia estrutural ao contexto jurisdicional e ao contexto clínico. O paper analisa as implicações jurídicas da lacuna em termos de responsabilidade do decisor, efetividade da supervisão humana e limites da explainability, e formula uma proposta dirigida ao regulador europeu e ao legislador português, assente no reconhecimento do processo de estruturação da informação como objeto regulatório autónomo. Nelson Marques | Jurista - Direito Digital e Regulação da Inteligência Artificial | e-mail: nelsonailaw@gmail.com | https://www.linkedin.com/in/nelsonmarques-ailaw | Tipologia: Documento de Posição (Position Paper) - Abril 2026 | Trabalho anterior: DOI 10.5281/zenodo.19446330 | “O objetivo não é questionar a autonomia do decisor, mas assegurar que essa autonomia se exerce sobre todas as dimensões relevantes do processo de decisão.” Keywords: artificial intelligence, medical practice, decisional heteronomy, clinical autonomy, AI Act, informed consent, medical deontology, automation bias, cognitive heteronomy, shadow AI, physician governance.","author":[{"family":"Marques","given":"Nelson"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19902379","URL":"https://doi.org/10.5281/zenodo.19902379","source":"datacite"},{"id":"doi:10.17605/osf.io/jk32v","type":"article-journal","title":"Generative Artificial Intelligence in Medical and Veterinary Education: A Scoping Review","abstract":"This scoping review maps the extent and nature of the literature regarding the implementation, design, and evaluation of generative artificial intelligence across all phases of veterinary and human medical education. Specifically, it examines how tools like large language models and conversational agents are utilized to support instructional design, curriculum development, and clinical training. By analyzing the intersection of educational pedagogical frameworks and human-computer interaction, this review identifies current technological, ethical, and instructional gaps. The synthesized findings will directly inform the future development of generative artificial intelligence tutoring systems and scalable online learning modules for medical and veterinary educators. The search encompasses literature published between 2020 and 2026 across major scientific databases.","author":[{"family":"Suresh","given":"Divya"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/jk32v","URL":"https://doi.org/10.17605/osf.io/jk32v","source":"datacite"},{"id":"doi:10.5064/f6dn5p6i","type":"article-journal","title":"Ethical Approaches to Informed Consent for Autonomous Robotic-Assisted Surgery","abstract":"&lt;h3&gt;Project Overview&lt;/h3&gt; &lt;p&gt;Robot-assisted surgery has seen growth in recent years, and with the emergence of artificial intelligence, it is anticipated that some procedures will soon be performed by partially autonomous robotic systems under the oversight of a surgeon. Such advancements are expected to enhance surgical quality and outcomes. However, the ethical implications of integrating partially autonomous systems into surgical practice remain insufficiently understood. This study aims to examine these ethical issues in depth. To achieve this, semi-structured interviews were conducted with two groups: patients who have undergone surgery with or without robotic assistance at Vanderbilt University Medical Center, and healthcare providers, including surgeons, proceduralists, anesthesiologists, and surgical nurses, who are involved in robot-assisted procedures at the same institution. &lt;/p&gt; &lt;p&gt;We also conducted a conjoint analysis survey with patients on whether they would choose to undergo robot-assisted surgery under various conditions (with the robot having varying levels of control, risk level compared to a human surgeon, and recovery time compared to a human surgeon).&lt;/p&gt;","author":[{"family":"Wu","given":"Jie"},{"family":"Gordon","given":"Elisa"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5064/f6dn5p6i","URL":"https://doi.org/10.5064/f6dn5p6i","source":"datacite"},{"id":"doi:10.5281/zenodo.19446329","type":"article-journal","title":"Governação do Ato Médico Assistido por Inteligência Artificial: A Lacuna Operacional e o Risco de Heteronomia Decisória Invisível","abstract":"O presente documento analisa a crescente integração de sistemas de inteligência artificial no ato médico em Portugal, evidenciando a ausência de enquadramento operacional estruturado para o cumprimento das exigências legais e deontológicas. Introduz-se e aprofunda-se o conceito de “heteronomia decisória invisível”, designando formas de influência não percecionada na decisão clínica mediadas por sistemas de IA, enquadrando-o na literatura académica recente sobre cognitive heteronomy, automation bias e autonomia do médico face à IA.[1,2,3] Identifica-se uma lacuna crítica na tradução prática do enquadramento regulatório, designadamente do Regulamento (UE) 2024/1689 (AI Act) e do Regulamento (UE) 2016/679 (RGPD), e propõe-se a criação de instrumentos operacionais no âmbito da Ordem dos Médicos. Palavras-chave: inteligência artificial, ato médico, heteronomia decisória, autonomia clínica, AI Act, consentimento informado, deontologia médica. Keywords: artificial intelligence, medical practice, decisional heteronomy, clinical autonomy, AI Act, informed consent, medical deontology, automation bias, cognitive heteronomy, shadow AI, physician governance.","author":[{"family":"Marques","given":"Nelson"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19446329","URL":"https://doi.org/10.5281/zenodo.19446329","source":"datacite"},{"id":"doi:10.5281/zenodo.19446330","type":"article-journal","title":"Governação do Ato Médico Assistido por Inteligência Artificial: A Lacuna Operacional e o Risco de Heteronomia Decisória Invisível","abstract":"O presente documento analisa a crescente integração de sistemas de inteligência artificial no ato médico em Portugal, evidenciando a ausência de enquadramento operacional estruturado para o cumprimento das exigências legais e deontológicas. Introduz-se e aprofunda-se o conceito de “heteronomia decisória invisível”, designando formas de influência não percecionada na decisão clínica mediadas por sistemas de IA, enquadrando-o na literatura académica recente sobre cognitive heteronomy, automation bias e autonomia do médico face à IA.[1,2,3] Identifica-se uma lacuna crítica na tradução prática do enquadramento regulatório, designadamente do Regulamento (UE) 2024/1689 (AI Act) e do Regulamento (UE) 2016/679 (RGPD), e propõe-se a criação de instrumentos operacionais no âmbito da Ordem dos Médicos. Palavras-chave: inteligência artificial, ato médico, heteronomia decisória, autonomia clínica, AI Act, consentimento informado, deontologia médica. Keywords: artificial intelligence, medical practice, decisional heteronomy, clinical autonomy, AI Act, informed consent, medical deontology, automation bias, cognitive heteronomy, shadow AI, physician governance.","author":[{"family":"Marques","given":"Nelson"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19446330","URL":"https://doi.org/10.5281/zenodo.19446330","source":"datacite"},{"id":"doi:10.5281/zenodo.21851699","type":"article-journal","title":"Dataset: Identify specific patterns of TDP-43 proteinopathy induced cryptic mis-splicing of STMN2 and other \"cryptic mis-splicing\" patterns found within PubMed Literature, 2026. - PathMap Experiment #000111","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=111 Artificial General Intelligence LLC Claim Evaluated: Identify specific patterns of TDP-43 proteinopathy induced cryptic mis-splicing of STMN2 and other \"cryptic mis-splicing\" patterns found within PubMed Literature, 2026. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Cryptic splicing creates stable, neurotoxic polypeptides (e.g., PKN1-N207) that escape nonsense-mediated decay (41720774). Cryptic peptides derived from mis-spliced transcripts are detectable in patient serum extracellular vesicles and CSF, offering potential diagnostic utility (41612503, 38277467). Cryptic polyadenylation is a distinct class of TDP-43 LOF events beyond canonical cryptic exon splicing, often leading to 3'UTR extensions (41120751, 38313254). Nonsense-mediated decay (NMD) significantly masks the breadth of cryptic splicing, meaning standard RNA-seq often underestimates the total cryptic burden (40670663, 41332610). TDP-43-dependent cryptic splicing is an early event, occurring before the appearance of overt cytoplasmic aggregates, challenging the dogma that aggregation is the sole driver of clinical symptoms (38443601). Ciclopirox olamine induces TDP-43 cryptic exons via heavy metal toxicity, suggesting potential external triggers for proteinopathy (40715064). The inclusion of cryptic exons can trigger an adaptive immune response, where CD8+ T cells recognize cryptic epitopes as neo-antigens (40667053). TDP-43-dependent cryptic peptides represent a \"proteomic shift\" in neurodegeneration that may be independent of the total burden of canonical TDP-43 aggregates. NMD efficiency acts as a cellular checkpoint, with tumors and neurodegenerative states showing a divergence from \"tissue-specific baseline\" quality control, suggesting an \"NMD signature\" that varies per cell type. The inclusion of specific exons leads to peptide products that are not just byproduct garbage but functional effectors of toxicity. Cryptic peptides can be detected in extracellular vesicles (EVs), suggesting they could serve as non-invasive biomarkers for disease-specific splicing signatures. Synaptic proteins are disproportionately affected by the proteome-wide reduction in CE-target proteins, linking RNA surveillance directly to synaptic failure. Genetic modifiers, such as RAD23A or USP13, demonstrate that targeting protein homeostasis can mitigate the toxicity of TDP-43 mislocalization. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Perform longitudinal multi-omic analysis of iPSC-derived neurons to define the temporal hierarchy between initial cryptic splicing of STMN2/UNC13A and subsequent protein aggregation. Validate the neurotoxicity of cryptic peptides (e.g., PKN1-N207) by expressing them in non-TDP-43-depleted neurons and measuring synaptic plasticity markers. Test if pharmacological inhibition of NMD allows for the identification of a wider set of potential cryptic exon therapeutic targets in human patient tissue. 1. Perform mass-spectrometry based proteomic screening of patient CSF and EVs to quantify the abundance of PKN1-N207 in different clinical FTD variants. 2. Compare the toxicity of NMD-inhibitor-treated neurons (increasing cryptic peptide yield) vs. control neurons using synaptic plasticity assays. 3. CRISPR-tag the PKN1 locus in patient-derived iNeurons to monitor the real-time formation of PKN207. 📊 Suggested Studies Cross-sectional study to validate the diagnostic accuracy of cryptic peptide panels in serum-derived extracellular vesicles across diverse FTLD-TDP cohorts. Comparative RNA-seq meta-analysis of different brain regions to determine the tissue-specific hierarchy of cryptic splicing vulnerability in LATE","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21851699","URL":"https://doi.org/10.5281/zenodo.21851699","source":"datacite"},{"id":"doi:10.5281/zenodo.21851700","type":"article-journal","title":"Dataset: Identify specific patterns of TDP-43 proteinopathy induced cryptic mis-splicing of STMN2 and other \"cryptic mis-splicing\" patterns found within PubMed Literature, 2026. - PathMap Experiment #000111","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=111 Artificial General Intelligence LLC Claim Evaluated: Identify specific patterns of TDP-43 proteinopathy induced cryptic mis-splicing of STMN2 and other \"cryptic mis-splicing\" patterns found within PubMed Literature, 2026. This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Cryptic splicing creates stable, neurotoxic polypeptides (e.g., PKN1-N207) that escape nonsense-mediated decay (41720774). Cryptic peptides derived from mis-spliced transcripts are detectable in patient serum extracellular vesicles and CSF, offering potential diagnostic utility (41612503, 38277467). Cryptic polyadenylation is a distinct class of TDP-43 LOF events beyond canonical cryptic exon splicing, often leading to 3'UTR extensions (41120751, 38313254). Nonsense-mediated decay (NMD) significantly masks the breadth of cryptic splicing, meaning standard RNA-seq often underestimates the total cryptic burden (40670663, 41332610). TDP-43-dependent cryptic splicing is an early event, occurring before the appearance of overt cytoplasmic aggregates, challenging the dogma that aggregation is the sole driver of clinical symptoms (38443601). Ciclopirox olamine induces TDP-43 cryptic exons via heavy metal toxicity, suggesting potential external triggers for proteinopathy (40715064). The inclusion of cryptic exons can trigger an adaptive immune response, where CD8+ T cells recognize cryptic epitopes as neo-antigens (40667053). TDP-43-dependent cryptic peptides represent a \"proteomic shift\" in neurodegeneration that may be independent of the total burden of canonical TDP-43 aggregates. NMD efficiency acts as a cellular checkpoint, with tumors and neurodegenerative states showing a divergence from \"tissue-specific baseline\" quality control, suggesting an \"NMD signature\" that varies per cell type. The inclusion of specific exons leads to peptide products that are not just byproduct garbage but functional effectors of toxicity. Cryptic peptides can be detected in extracellular vesicles (EVs), suggesting they could serve as non-invasive biomarkers for disease-specific splicing signatures. Synaptic proteins are disproportionately affected by the proteome-wide reduction in CE-target proteins, linking RNA surveillance directly to synaptic failure. Genetic modifiers, such as RAD23A or USP13, demonstrate that targeting protein homeostasis can mitigate the toxicity of TDP-43 mislocalization. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Perform longitudinal multi-omic analysis of iPSC-derived neurons to define the temporal hierarchy between initial cryptic splicing of STMN2/UNC13A and subsequent protein aggregation. Validate the neurotoxicity of cryptic peptides (e.g., PKN1-N207) by expressing them in non-TDP-43-depleted neurons and measuring synaptic plasticity markers. Test if pharmacological inhibition of NMD allows for the identification of a wider set of potential cryptic exon therapeutic targets in human patient tissue. 1. Perform mass-spectrometry based proteomic screening of patient CSF and EVs to quantify the abundance of PKN1-N207 in different clinical FTD variants. 2. Compare the toxicity of NMD-inhibitor-treated neurons (increasing cryptic peptide yield) vs. control neurons using synaptic plasticity assays. 3. CRISPR-tag the PKN1 locus in patient-derived iNeurons to monitor the real-time formation of PKN207. 📊 Suggested Studies Cross-sectional study to validate the diagnostic accuracy of cryptic peptide panels in serum-derived extracellular vesicles across diverse FTLD-TDP cohorts. Comparative RNA-seq meta-analysis of different brain regions to determine the tissue-specific hierarchy of cryptic splicing vulnerability in LATE","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21851700","URL":"https://doi.org/10.5281/zenodo.21851700","source":"datacite"},{"id":"doi:10.5281/zenodo.21830175","type":"article-journal","title":"Dataset: mRNA Influenza Vaccination Information. August, 2026 PathMap - PathMap Experiment #000106","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=106 Artificial General Intelligence LLC Claim Evaluated: mRNA Influenza Vaccination Information. August, 2026 PathMap This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights mRNA vaccines for influenza demonstrate clinical efficacy profiles that are statistically comparable to current licensed enhanced vaccines (EVs) in geriatric populations. The use of non-canonical delivery systems, such as piezoelectric electroporation (Piezopen), shows potential for \"naked\" mRNA delivery, potentially bypassing inflammatory lipid nanoparticle (LNP) carriers. There is no evidence of significant structural cerebral changes following mRNA vaccination, contradicting concerns regarding microstructural brain alterations in the subacute phase. Antigenic mismatch continues to be a primary driver of variable effectiveness, necessitating the transition to recombinant protein and mRNA platforms. Sequential vaccination of COVID-19 and influenza antigens does not appear to compromise the individual immunogenicity of either vaccine in immunocompromised populations. Adjuvanted and high-dose influenza vaccines have shown comparable protection against medically attended influenza in real-world cohorts. The persistence of SARS-CoV-2 spike protein in skin lesions post-vaccination remains an area for continued clinical investigation regarding vasculitic manifestations. mRNA-1010 immunogenicity is comparable to traditional high-dose influenza vaccines, a critical finding for addressing immunosenescence in older populations. The multicomponent mRNA-1083 vaccine enables simultaneous protection against influenza and SARS-CoV-2 without compromising individual immune response magnitudes. Capless self-amplifying mRNA (CLsamRNA) platforms show extreme dose-sparing potential (e.g., 0.01 μg), reducing the manufacturing requirements for large-scale production. mRNA vaccination induces a distinct Th1/Tfh1-biased cellular immune response, which correlates with long-lasting memory. Sequential administration of mRNA-based COVID-19 and influenza vaccines does not inhibit the development of antigen-specific immunity against either virus. mRNA-based influenza platforms can be rapidly updated to address antigenic drift, a key improvement over egg-based production. No signals for myocarditis or pericarditis were identified in the reported phase 3 clinical trials for mRNA-1083. Current mRNA-LNP delivery systems are being engineered to shift expression profiles, such as increasing spleen-selective immunity for better T-cell priming. mRNA-1010 is consistently shown to be superior to standard-dose vaccines for the prevention of RT-PCR-confirmed influenza-like illness in older adults. The integration of internal viral proteins (e.g., nucleoprotein) and neuraminidase is essential for achieving universal cross-protection. Self-amplifying RNA (saRNA) platforms significantly improve IBV-specific immunogenicity compared to conventional mRNA. Pharmacist-led vaccination programs, as seen in New Zealand, remain a primary driver for increasing vaccine uptake in the geriatric population. The use of needle-free jet injectors provides a potential technological bridge for more efficient, dose-sparing delivery of future mRNA influenza formulations. Current data indicate that mRNA-based multicomponent vaccines (e.g., mRNA-1083) represent a viable strategy for co-protection against influenza and SARS-CoV-2. There is a transition in research focus from mere antibody titer measurement to monitoring circulating follicular helper T-cell responses for deeper immunological memory assessment. Computational and algorithm-optimized mRNA H5 influenza vaccines are now successfully inducin","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21830175","URL":"https://doi.org/10.5281/zenodo.21830175","source":"datacite"},{"id":"doi:10.5281/zenodo.21830176","type":"article-journal","title":"Dataset: mRNA Influenza Vaccination Information. August, 2026 PathMap - PathMap Experiment #000106","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=106 Artificial General Intelligence LLC Claim Evaluated: mRNA Influenza Vaccination Information. August, 2026 PathMap This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights mRNA vaccines for influenza demonstrate clinical efficacy profiles that are statistically comparable to current licensed enhanced vaccines (EVs) in geriatric populations. The use of non-canonical delivery systems, such as piezoelectric electroporation (Piezopen), shows potential for \"naked\" mRNA delivery, potentially bypassing inflammatory lipid nanoparticle (LNP) carriers. There is no evidence of significant structural cerebral changes following mRNA vaccination, contradicting concerns regarding microstructural brain alterations in the subacute phase. Antigenic mismatch continues to be a primary driver of variable effectiveness, necessitating the transition to recombinant protein and mRNA platforms. Sequential vaccination of COVID-19 and influenza antigens does not appear to compromise the individual immunogenicity of either vaccine in immunocompromised populations. Adjuvanted and high-dose influenza vaccines have shown comparable protection against medically attended influenza in real-world cohorts. The persistence of SARS-CoV-2 spike protein in skin lesions post-vaccination remains an area for continued clinical investigation regarding vasculitic manifestations. mRNA-1010 immunogenicity is comparable to traditional high-dose influenza vaccines, a critical finding for addressing immunosenescence in older populations. The multicomponent mRNA-1083 vaccine enables simultaneous protection against influenza and SARS-CoV-2 without compromising individual immune response magnitudes. Capless self-amplifying mRNA (CLsamRNA) platforms show extreme dose-sparing potential (e.g., 0.01 μg), reducing the manufacturing requirements for large-scale production. mRNA vaccination induces a distinct Th1/Tfh1-biased cellular immune response, which correlates with long-lasting memory. Sequential administration of mRNA-based COVID-19 and influenza vaccines does not inhibit the development of antigen-specific immunity against either virus. mRNA-based influenza platforms can be rapidly updated to address antigenic drift, a key improvement over egg-based production. No signals for myocarditis or pericarditis were identified in the reported phase 3 clinical trials for mRNA-1083. Current mRNA-LNP delivery systems are being engineered to shift expression profiles, such as increasing spleen-selective immunity for better T-cell priming. mRNA-1010 is consistently shown to be superior to standard-dose vaccines for the prevention of RT-PCR-confirmed influenza-like illness in older adults. The integration of internal viral proteins (e.g., nucleoprotein) and neuraminidase is essential for achieving universal cross-protection. Self-amplifying RNA (saRNA) platforms significantly improve IBV-specific immunogenicity compared to conventional mRNA. Pharmacist-led vaccination programs, as seen in New Zealand, remain a primary driver for increasing vaccine uptake in the geriatric population. The use of needle-free jet injectors provides a potential technological bridge for more efficient, dose-sparing delivery of future mRNA influenza formulations. Current data indicate that mRNA-based multicomponent vaccines (e.g., mRNA-1083) represent a viable strategy for co-protection against influenza and SARS-CoV-2. There is a transition in research focus from mere antibody titer measurement to monitoring circulating follicular helper T-cell responses for deeper immunological memory assessment. Computational and algorithm-optimized mRNA H5 influenza vaccines are now successfully inducin","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21830176","URL":"https://doi.org/10.5281/zenodo.21830176","source":"datacite"},{"id":"doi:10.5281/zenodo.21284090","type":"article-journal","title":"What changes in a person's voice occur prior to Amyotrophic Lateral Sclerosis onset that may be useful as a non-clinical (or clinical if possible) early detection of the disease? - PathMap Experiment #000035","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=35 Artificial General Intelligence LLC Claim Evaluated: What changes in a person's voice occur prior to Amyotrophic Lateral Sclerosis onset that may be useful as a non-clinical (or clinical if possible) early detection of the disease? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Subclinical Detection:** Artificial intelligence frameworks can identify neuromuscular changes during the \"clinically silent prodromal stage\" before functional decline is apparent. Biomechanical Precision:** Biomechanical voice parameters reflecting glottal tension and vocal fold stability are sensitive enough to differentiate clinical phenotypes (bulbar vs. spinal onset). Multimodal Integration:** Combining facial sEMG and acoustic signals outperforms single-modality assessments in detecting early bulbar motor dysfunction. Listener Effort (LE):** LE is a clinician-rated metric that captures meaningful change in dysarthria and shows potential as a responsive clinical trial endpoint. Smartphone Utility:** Simple, smartphone-based assessment tasks (e.g., tongue lateralization or vowel phonation) correlate highly with laboratory-standard assessments, increasing access. Stability of Biomarkers:** Despite disease progression, high-gamma cortical features in ECoG speech BCIs show long-term stability, suggesting durability for assistive interfaces. Predictive Modeling:** Subject-specific prognostic models can now predict articulatory precision and ALSFRS-R speech subscores 30–90 days in advance. Vocal Subtypes:** Unsupervised clustering reveals \"vocal profiles\" that transcend traditional diagnostic labels, indicating that voice features capture functional patterns of voice production across different disorders. Voice serves as a latent, multimodal biomarker reflecting neurological, cardiopulmonary, and psychological states. Automated segmentation algorithms can now identify syllable and phoneme positions during oral diadochokinesis with over 90% accuracy, providing a basis for objective monitoring. Biomechanical voice analysis captures physiologically meaningful alterations in vocal fold function, offering complementary information that transcends traditional clinical diagnostic labels. Changes in speech and swallowing function can be monitored remotely using smartphone applications, potentially reducing the need for frequent clinical hospital visits. There is a significant need for better standardized tools, as current outcome measurements for speech and swallow are deemed clinically meaningful by only a minority of practitioners. Vowel acoustic features (e.g., Formant Centralization Ratio) provide insight into the shared brainstem neuromotor substrate of both speech and swallowing. Research confirms the existence of coherent vocal profiles across patients that do not strictly align with existing clinical diagnostic categories. Early intervention protocols are limited by current guideline gaps, underscoring the necessity of individual-level predictive modeling. Thinning of the bilateral oral motor cortices is an anatomical precursor that maps directly to measurable decrements in oral motor function. Automated speaking and articulation rates provide a robust alternative to manually conducted assessments, which are prone to observer bias. Digital speech-derived measures demonstrate clear neuroanatomical correlations where standard bulbar subscores in the ALSFRS-R do not. The use of intracortical brain-computer interfaces (BCIs) has allowed for long-term monitoring, producing datasets with high word accuracy that validate the stability of speech-based tracking. Cortical dysfunction originates in a developmental trajectory in cu","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21284090","URL":"https://doi.org/10.5281/zenodo.21284090","source":"datacite"},{"id":"doi:10.5281/zenodo.21284091","type":"article-journal","title":"What changes in a person's voice occur prior to Amyotrophic Lateral Sclerosis onset that may be useful as a non-clinical (or clinical if possible) early detection of the disease? - PathMap Experiment #000035","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=35 Artificial General Intelligence LLC Claim Evaluated: What changes in a person's voice occur prior to Amyotrophic Lateral Sclerosis onset that may be useful as a non-clinical (or clinical if possible) early detection of the disease? This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Subclinical Detection:** Artificial intelligence frameworks can identify neuromuscular changes during the \"clinically silent prodromal stage\" before functional decline is apparent. Biomechanical Precision:** Biomechanical voice parameters reflecting glottal tension and vocal fold stability are sensitive enough to differentiate clinical phenotypes (bulbar vs. spinal onset). Multimodal Integration:** Combining facial sEMG and acoustic signals outperforms single-modality assessments in detecting early bulbar motor dysfunction. Listener Effort (LE):** LE is a clinician-rated metric that captures meaningful change in dysarthria and shows potential as a responsive clinical trial endpoint. Smartphone Utility:** Simple, smartphone-based assessment tasks (e.g., tongue lateralization or vowel phonation) correlate highly with laboratory-standard assessments, increasing access. Stability of Biomarkers:** Despite disease progression, high-gamma cortical features in ECoG speech BCIs show long-term stability, suggesting durability for assistive interfaces. Predictive Modeling:** Subject-specific prognostic models can now predict articulatory precision and ALSFRS-R speech subscores 30–90 days in advance. Vocal Subtypes:** Unsupervised clustering reveals \"vocal profiles\" that transcend traditional diagnostic labels, indicating that voice features capture functional patterns of voice production across different disorders. Voice serves as a latent, multimodal biomarker reflecting neurological, cardiopulmonary, and psychological states. Automated segmentation algorithms can now identify syllable and phoneme positions during oral diadochokinesis with over 90% accuracy, providing a basis for objective monitoring. Biomechanical voice analysis captures physiologically meaningful alterations in vocal fold function, offering complementary information that transcends traditional clinical diagnostic labels. Changes in speech and swallowing function can be monitored remotely using smartphone applications, potentially reducing the need for frequent clinical hospital visits. There is a significant need for better standardized tools, as current outcome measurements for speech and swallow are deemed clinically meaningful by only a minority of practitioners. Vowel acoustic features (e.g., Formant Centralization Ratio) provide insight into the shared brainstem neuromotor substrate of both speech and swallowing. Research confirms the existence of coherent vocal profiles across patients that do not strictly align with existing clinical diagnostic categories. Early intervention protocols are limited by current guideline gaps, underscoring the necessity of individual-level predictive modeling. Thinning of the bilateral oral motor cortices is an anatomical precursor that maps directly to measurable decrements in oral motor function. Automated speaking and articulation rates provide a robust alternative to manually conducted assessments, which are prone to observer bias. Digital speech-derived measures demonstrate clear neuroanatomical correlations where standard bulbar subscores in the ALSFRS-R do not. The use of intracortical brain-computer interfaces (BCIs) has allowed for long-term monitoring, producing datasets with high word accuracy that validate the stability of speech-based tracking. Cortical dysfunction originates in a developmental trajectory in cu","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21284091","URL":"https://doi.org/10.5281/zenodo.21284091","source":"datacite"},{"id":"doi:10.17605/osf.io/6dnp3","type":"article-journal","title":"Current Status of Generative Artificial Intelligence Applications in Health Management for Older Adults: A Scoping Review","abstract":"As generative artificial intelligence (GAI) and large language models (LLMs) rapidly evolve, their potential in supporting patient-centered health management has gained significant attention. However, empirical evidence regarding how older adults—a population with unique health needs and digital literacy challenges—interact with and adopt GAI remains fragmented. Guided by the Arksey and O'Malley framework and PRISMA-ScR guidelines, this scoping review systematically maps and synthesizes the application patterns, health-related impacts, and determinants of GAI adoption among older adults (aged ≥ 60 years). Eight international and Chinese databases (PubMed, Web of Science, CINAHL, PsycINFO, CNKI, Wanfang, IEEE Xplore, and Scopus) are searched from January 2020 to June 2026. Quantitative, qualitative, and mixed-methods studies examining older adults' real-world use of GAI for health information seeking, chronic disease self-management, medical communication, and decision support are included. Data are synthesized using descriptive statistics and thematic analysis to provide evidence-based insights for designing age-friendly AI tools and clinical health interventions.","author":[{"family":"Wang","given":"Jindi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17605/osf.io/6dnp3","URL":"https://doi.org/10.17605/osf.io/6dnp3","source":"datacite"},{"id":"doi:10.5281/zenodo.21521123","type":"article-journal","title":"Dataset: SOD1 Research July 2026 - PathMap Experiment #000084","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=84 Artificial General Intelligence LLC Claim Evaluated: SOD1 Research July 2026 This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Biomarker Evolution:** Neuromuscular ultrasound now provides non-invasive diagnostic capabilities that match or precede traditional electroneurographic markers in SOD1G93A models. Mechanism Redefined:** Mutant SOD1 acts as both a Fenton-like catalyst for hydroxyl radical generation and a hydrogenation catalyst for hydrogen scavenging. Genetic Prevalence:** Population-specific data, such as that from Indian cohorts, demonstrate that SOD1 is the predominant cause of familial ALS, even when other repeat expansions (e.g., C9orf72) are present at low frequencies. Systemic Involvement:** ALS motor neuron disease is increasingly viewed as a multisystem disorder where innate immune crosstalk, specifically between cGAS-STING and NLRP3 inflammasomes, drives progression. Proactive Planning:** Nationwide adoption of genetic testing in Canada was significantly accelerated by proactive planning during the clinical trial phase of gene-targeted therapies. Microglial Dynamics:** SGK1 has been identified as a key regulator of microglial phagocytosis; its inhibition attenuates motor deficits, suggesting it as a potential therapeutic target. Future Demand:** Projections indicate a significant increase in ALS clinic visits among asymptomatic gene carriers, requiring substantial expansion of clinical infrastructure by 2035. The application of magnesium-silicide based hydrogen gas release serves as an innovative strategy to intercept the crosstalk between oxidative stress and neuroinflammation. The use of Platelet Factor 4 (PF4) demonstrates a selective neuroprotective benefit in SOD1-driven ALS, bypassing PINK1-dependent mechanisms to restore proteostasis. The phenomenon of macrophage inclusions (\"tofersenophages\") in CSF has been identified as a persistent, albeit clinically ambiguous, finding during ASO therapy, which surprisingly correlates with favorable clinical outcomes. Neuromuscular ultrasound serves as a high-sensitivity, non-invasive biomarker that detects disease pathology at stages prior to electroneurographic abnormalities. Genetic testing for ALS has achieved near-universal integration in clinical practice by 2025, with sponsored, cost-free testing panels significantly increasing diagnostic yields in sporadic cases. The identification of the JAK2 gene as a novel genome-wide significant signal in the Indian cohort underscores the importance of population-specific genetic surveying. The integration of phase-resolved geometric deep learning (SKALE 2.0) now allows for the constraint-aware design of aggregation suppressors that differentiate between nucleation and elongation phases. The existence of oligogenic models (e.g., ATXN2/NEK1) highlights the complexity of ALS, where pathogenicity may be governed by the synergy of multiple low-penetrance variants rather than monogenic drivers. Copper Paradox:** High intracellular copper can inhibit SOD1 by disrupting its homodimerization, mediated by COMMD1-dependent mechanisms. Catalytic Hydrogen Therapy:** Mutant SOD1 acts as both a Fenton-like agent producing hydroxyl radicals and a catalyst for hydrogen-based free radical scavenging. Microglial LAG-3:** This immune checkpoint protein exerts stage-dependent regulation on microglial modules, dissociating inflammatory and phagocytic functions in ALS progression. Prion-like Propagation:** Conversion of SOD1 into a misfolded isoform is a targetable biophysical process distinct from aggregation. Statin Effects:** While statins can modulate antioxidant genes, they may also inadvertently accelera","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21521123","URL":"https://doi.org/10.5281/zenodo.21521123","source":"datacite"},{"id":"doi:10.5281/zenodo.21521124","type":"article-journal","title":"Dataset: SOD1 Research July 2026 - PathMap Experiment #000084","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=84 Artificial General Intelligence LLC Claim Evaluated: SOD1 Research July 2026 This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Biomarker Evolution:** Neuromuscular ultrasound now provides non-invasive diagnostic capabilities that match or precede traditional electroneurographic markers in SOD1G93A models. Mechanism Redefined:** Mutant SOD1 acts as both a Fenton-like catalyst for hydroxyl radical generation and a hydrogenation catalyst for hydrogen scavenging. Genetic Prevalence:** Population-specific data, such as that from Indian cohorts, demonstrate that SOD1 is the predominant cause of familial ALS, even when other repeat expansions (e.g., C9orf72) are present at low frequencies. Systemic Involvement:** ALS motor neuron disease is increasingly viewed as a multisystem disorder where innate immune crosstalk, specifically between cGAS-STING and NLRP3 inflammasomes, drives progression. Proactive Planning:** Nationwide adoption of genetic testing in Canada was significantly accelerated by proactive planning during the clinical trial phase of gene-targeted therapies. Microglial Dynamics:** SGK1 has been identified as a key regulator of microglial phagocytosis; its inhibition attenuates motor deficits, suggesting it as a potential therapeutic target. Future Demand:** Projections indicate a significant increase in ALS clinic visits among asymptomatic gene carriers, requiring substantial expansion of clinical infrastructure by 2035. The application of magnesium-silicide based hydrogen gas release serves as an innovative strategy to intercept the crosstalk between oxidative stress and neuroinflammation. The use of Platelet Factor 4 (PF4) demonstrates a selective neuroprotective benefit in SOD1-driven ALS, bypassing PINK1-dependent mechanisms to restore proteostasis. The phenomenon of macrophage inclusions (\"tofersenophages\") in CSF has been identified as a persistent, albeit clinically ambiguous, finding during ASO therapy, which surprisingly correlates with favorable clinical outcomes. Neuromuscular ultrasound serves as a high-sensitivity, non-invasive biomarker that detects disease pathology at stages prior to electroneurographic abnormalities. Genetic testing for ALS has achieved near-universal integration in clinical practice by 2025, with sponsored, cost-free testing panels significantly increasing diagnostic yields in sporadic cases. The identification of the JAK2 gene as a novel genome-wide significant signal in the Indian cohort underscores the importance of population-specific genetic surveying. The integration of phase-resolved geometric deep learning (SKALE 2.0) now allows for the constraint-aware design of aggregation suppressors that differentiate between nucleation and elongation phases. The existence of oligogenic models (e.g., ATXN2/NEK1) highlights the complexity of ALS, where pathogenicity may be governed by the synergy of multiple low-penetrance variants rather than monogenic drivers. Copper Paradox:** High intracellular copper can inhibit SOD1 by disrupting its homodimerization, mediated by COMMD1-dependent mechanisms. Catalytic Hydrogen Therapy:** Mutant SOD1 acts as both a Fenton-like agent producing hydroxyl radicals and a catalyst for hydrogen-based free radical scavenging. Microglial LAG-3:** This immune checkpoint protein exerts stage-dependent regulation on microglial modules, dissociating inflammatory and phagocytic functions in ALS progression. Prion-like Propagation:** Conversion of SOD1 into a misfolded isoform is a targetable biophysical process distinct from aggregation. Statin Effects:** While statins can modulate antioxidant genes, they may also inadvertently accelera","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21521124","URL":"https://doi.org/10.5281/zenodo.21521124","source":"datacite"},{"id":"doi:10.5281/zenodo.21385149","type":"article-journal","title":"Dataset: Lon Protease, Alternaria; IL-33; TSLP; alarmins; asthma; chronic rhinosinusitis; fungal allergen; innate lymphoid cells - PathMap Experiment #000064","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=64 Artificial General Intelligence LLC Claim Evaluated: Lon Protease, Alternaria; IL-33; TSLP; alarmins; asthma; chronic rhinosinusitis; fungal allergen; innate lymphoid cells This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Alternaria-induced ILC2 activation is not merely a consequence of alarmin signaling but is subject to mechanical checkpoints like Piezo1. LONP1 acts as a potential immunometabolic checkpoint, where mitochondrial protein quality control directly impacts the inflammatory trajectory of the airway epithelium. The severity of Alternaria-driven responses is modulated by SLPI, which serves as a molecular brake on the protease-mediated activation of IL-33. ILC2s exhibit significant phenotypic plasticity, particularly when transitioning toward ILC3-like or steroid-resistant states. The cross-talk between eosinophils and epithelial cells is bi-directional and foundational to tissue-resident remodeling in chronic rhinosinusitis. Mitochondrial dysfunction (driven by LONP1/Drp1) is an upstream contributor to the cytokine/chemokine environment of the asthma/CRSwNP mucosa. Fungal allergens can initiate a \"two-hit\" inflammatory model where live spore exposure exacerbates pre-existing, OVA-primed airway damage. LONP1 serves as a dual-function gatekeeper, maintaining mitochondrial DNA integrity while modulating inflammatory cell polarization in response to oxidative stress. The \"residual molecular scar\" phenomenon explains why some patients with ECRS exhibit persistent mucus hyperviscosity even after successful biological blockade of IL-4/IL-13. ILC2s are not merely passive responders; they exhibit subset heterogeneity (migratory, transitional, inflammatory, exhausted) that correlates with clinical severity in nasal polyps. Treg/Th2 imbalance in severe asthma is reversible, as shown by benralizumab therapy restoring immune homeostasis and modifying adhesion molecule expression. The gut-lung axis utilizes tryptophan metabolism to reprogram ILC2s, potentially allowing microbiome-derived postbiotics to serve as non-live therapeutic alternatives. Fungal *Alternaria* allergens act not only as biochemical triggers for alarmins but also cause physical and oxidative damage that necessitates mitochondrial quality control. Mitochondrial proteases (LonP1) serve as an immunometabolic checkpoint, where their dysfunction directly links mitochondrial DNA release to chronic inflammation via the cGAS-STING axis. The IL-22BP decoy receptor has been shown to play a paradoxical role; while IL-22 is typically protective, \"These findings suggest that inhibition of IL-22BP attenuates the development of allergen-induced AHR, an effect likely mediated through enhanced IL-22 activity rather than alterations in airway inflammation or type 2 cytokine production.\" Biologics targeting TSLP, such as tezepelumab, are effective across diverse asthma endotypes, emphasizing the hierarchy of alarmins as \"source-directed\" intervention targets. The metabolic state of ILC2s (glycolysis, lipid metabolism) is an emerging regulator of their plasticity, suggesting that metabolic modulation (e.g., via serotonin catabolism or MAOA inhibition) can alter immune responsiveness. Epigenetic memory, established through DNA methylation and histone modifications in basal epithelial progenitors, explains why asthma is often a relapsing, chronic condition rather than a simple acute response to fungal allergens. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Investigate the impact of LONP1 knockdown in human primary ILC2s on their susceptibility to IL-33/TSLP-induced activation. Assess whether SLPI administration in a humanized mouse model ","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21385149","URL":"https://doi.org/10.5281/zenodo.21385149","source":"datacite"},{"id":"doi:10.5281/zenodo.21385150","type":"article-journal","title":"Dataset: Lon Protease, Alternaria; IL-33; TSLP; alarmins; asthma; chronic rhinosinusitis; fungal allergen; innate lymphoid cells - PathMap Experiment #000064","abstract":"Interactive Data Viewer: Read, View, and Print from Day 1 Use our fully interactive viewer to view, read, and print this research data right from Day 1: https://pathmap.org/viewer.php?id=64 Artificial General Intelligence LLC Claim Evaluated: Lon Protease, Alternaria; IL-33; TSLP; alarmins; asthma; chronic rhinosinusitis; fungal allergen; innate lymphoid cells This dataset contains the raw JSON execution trace, verified verbatim quotes, and MeSH-aligned logic gates generated by PathMap Studio's Veridical Enforcement engine. 🔍 Novel & Overlooked Insights Alternaria-induced ILC2 activation is not merely a consequence of alarmin signaling but is subject to mechanical checkpoints like Piezo1. LONP1 acts as a potential immunometabolic checkpoint, where mitochondrial protein quality control directly impacts the inflammatory trajectory of the airway epithelium. The severity of Alternaria-driven responses is modulated by SLPI, which serves as a molecular brake on the protease-mediated activation of IL-33. ILC2s exhibit significant phenotypic plasticity, particularly when transitioning toward ILC3-like or steroid-resistant states. The cross-talk between eosinophils and epithelial cells is bi-directional and foundational to tissue-resident remodeling in chronic rhinosinusitis. Mitochondrial dysfunction (driven by LONP1/Drp1) is an upstream contributor to the cytokine/chemokine environment of the asthma/CRSwNP mucosa. Fungal allergens can initiate a \"two-hit\" inflammatory model where live spore exposure exacerbates pre-existing, OVA-primed airway damage. LONP1 serves as a dual-function gatekeeper, maintaining mitochondrial DNA integrity while modulating inflammatory cell polarization in response to oxidative stress. The \"residual molecular scar\" phenomenon explains why some patients with ECRS exhibit persistent mucus hyperviscosity even after successful biological blockade of IL-4/IL-13. ILC2s are not merely passive responders; they exhibit subset heterogeneity (migratory, transitional, inflammatory, exhausted) that correlates with clinical severity in nasal polyps. Treg/Th2 imbalance in severe asthma is reversible, as shown by benralizumab therapy restoring immune homeostasis and modifying adhesion molecule expression. The gut-lung axis utilizes tryptophan metabolism to reprogram ILC2s, potentially allowing microbiome-derived postbiotics to serve as non-live therapeutic alternatives. Fungal *Alternaria* allergens act not only as biochemical triggers for alarmins but also cause physical and oxidative damage that necessitates mitochondrial quality control. Mitochondrial proteases (LonP1) serve as an immunometabolic checkpoint, where their dysfunction directly links mitochondrial DNA release to chronic inflammation via the cGAS-STING axis. The IL-22BP decoy receptor has been shown to play a paradoxical role; while IL-22 is typically protective, \"These findings suggest that inhibition of IL-22BP attenuates the development of allergen-induced AHR, an effect likely mediated through enhanced IL-22 activity rather than alterations in airway inflammation or type 2 cytokine production.\" Biologics targeting TSLP, such as tezepelumab, are effective across diverse asthma endotypes, emphasizing the hierarchy of alarmins as \"source-directed\" intervention targets. The metabolic state of ILC2s (glycolysis, lipid metabolism) is an emerging regulator of their plasticity, suggesting that metabolic modulation (e.g., via serotonin catabolism or MAOA inhibition) can alter immune responsiveness. Epigenetic memory, established through DNA methylation and histone modifications in basal epithelial progenitors, explains why asthma is often a relapsing, chronic condition rather than a simple acute response to fungal allergens. 🧪 Extracted Custom Datapoints 📊 Suggested Experiments Investigate the impact of LONP1 knockdown in human primary ILC2s on their susceptibility to IL-33/TSLP-induced activation. Assess whether SLPI administration in a humanized mouse model ","author":[{"family":"Dungan","given":"Joshua"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21385150","URL":"https://doi.org/10.5281/zenodo.21385150","source":"datacite"},{"id":"doi:10.5281/zenodo.21978439","type":"article-journal","title":"Robotique souple neuromorphique et essaims","abstract":"Résumé FRCe document, produit avec l’assistance de ChatGPT 5.2 Thinking et Gemini 3 Raisonnement, est publié sous licence Apache 2.0. Il constitue une publication défensive (antériorité) et entre dans l’état de la technique au sens des textes applicables (EPC Art. 54(2); French IPC Art. L 611-11; cf. 35 U.S.C. §102(a)). Il divulgue, de façon enabling, un portefeuille d’innovations combinant robotique souple (actionneurs HASEL/EAP), vision événementielle (DVS), calcul neuromorphique (SNN) et intelligence en essaim, couvrant dispositifs/capteurs, algorithmes, contrôle en boucle fermée, fabrication roll-to-roll et QA end-of-line, cybersécurité et opérations de flottes, interopérabilité (formats événements+spikes), logistique de cartouches, modèles économiques au résultat, et usages industriels, agricoles régénératifs, nucléaires, sous-marins et médicaux. Abstract ENThis document, produced with the assistance of ChatGPT 5.2 Thinking and Gemini 3 Raisonnement, is released under the Apache 2.0 licence. It is a voluntary defensive publication (prior art) and therefore enters the prior art upon release under the applicable patent statutes (art. L 611-11 CPI / art. 54(2) CBE). It discloses, in an enabling manner, a portfolio that fuses soft robotics (HASEL/EAP actuation), event-based vision (DVS), neuromorphic computing (SNN), and swarm intelligence. The disclosure spans devices and sensors, event-first control loops, roll-to-roll manufacturing and end-of-line QA, cyber-secure fleet operations, interoperability standards for event+spike telemetry, cartridge logistics and field repair, outcome-based metering and SLA instrumentation, and applications in high-throughput sorting, precision/regenerative agriculture, nuclear maintenance, underwater monitoring, and medical/rehabilitation systems. Each proposal is classified with IPC/CPC codes and can be timestamped (RFC 3161 / FreeTSA). Timestamp : 2026-08-17T10:45:25ZSHA-256 : 13b3e2bc50c638e594d623990f13159039dd0fb8f0b0968e18a3300649110a89 Liste des innovations & classification (IPC ; CPC) :1. DVS–HASEL soft gripper — IPC B25J 15/00 ; CPC B25J 15/122. DVS sorting calibration rig — IPC G01D 18/00 ; CPC G01D 18/003. HASEL sensing skin laminate — IPC G01L 5/00 ; CPC G01L 5/164. Biodegradable electrohydraulic actuator — IPC C08L 67/00 ; CPC C08L 67/025. Printable EAP electrode ink — IPC H01B 1/12 ; CPC H01B 1/126. Self-healing dielectric composite — IPC C08K 3/36 ; CPC C08K 3/367. Roll-to-roll HASEL pouch line — IPC B29C 65/00 ; CPC B29C 65/788. 3D-printed soft body + circuits — IPC B29C 64/118 ; CPC B29C 64/1189. Soft underwater encapsulation stack — IPC B29C 71/00 ; CPC B29C 71/0210. Event-driven SNN HASEL control — IPC G06N 3/04 ; CPC G06N 3/04511. Event-based actuator fatigue detection — IPC G05B 23/02 ; CPC G05B 23/0212. Edge event-stream compression codec — IPC H04N 5/00 ; CPC H04N 5/23213. Spike-packet swarm protocol — IPC H04W 4/80 ; CPC H04W 4/8014. Neuromorphic swarm task allocator — IPC G06Q 10/04 ; CPC G06Q 10/063915. Safe HV charge scheduler — IPC H02M 3/155 ; CPC H02M 3/15816. Swarm geofencing operations — IPC G08G 5/00 ; CPC G08G 5/0017. Radiation-hardened soft robot module — IPC G21C 19/00 ; CPC G21C 19/0018. DVS-to-intensity reconstruction — IPC H04N 5/232 ; CPC H04N 5/23219. DVS+EMG SNN exosuit fusion — IPC A61H 1/02 ; CPC A61H 1/0220. Closed-loop rehab dosing method — IPC A61H 1/00 ; CPC A61H 1/0021. Soft endoscope targeted delivery — IPC A61M 31/00 ; CPC A61M 31/0022. Low-power EAP assist patch — IPC A61F 5/01 ; CPC A61F 5/0123. Federated learning for agri swarms — IPC G06F 18/232 ; CPC G06F 18/232124. Event+spike interoperability standard — IPC G06F 9/54 ; CPC G06F 9/54125. Tamper-proof swarm audit ledger — IPC G06Q 20/38 ; CPC G06Q 20/38226. Swarm supervisor cockpit UI — IPC G05B 19/042 ; CPC G05B 19/04227. Hybrid ultra-fast waste sorter cell — IPC B07C 5/34 ; CPC B07C 5/34228. Underwater soft-drone swarm system — IPC B63G 8/00 ; CPC B63G 8/0029. Swarm soil-compaction sens","author":[{"family":"Pillet","given":"Xavier"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21978439","URL":"https://doi.org/10.5281/zenodo.21978439","source":"datacite"},{"id":"doi:10.5281/zenodo.21978440","type":"article-journal","title":"Robotique souple neuromorphique et essaims","abstract":"Résumé FRCe document, produit avec l’assistance de ChatGPT 5.2 Thinking et Gemini 3 Raisonnement, est publié sous licence Apache 2.0. Il constitue une publication défensive (antériorité) et entre dans l’état de la technique au sens des textes applicables (EPC Art. 54(2); French IPC Art. L 611-11; cf. 35 U.S.C. §102(a)). Il divulgue, de façon enabling, un portefeuille d’innovations combinant robotique souple (actionneurs HASEL/EAP), vision événementielle (DVS), calcul neuromorphique (SNN) et intelligence en essaim, couvrant dispositifs/capteurs, algorithmes, contrôle en boucle fermée, fabrication roll-to-roll et QA end-of-line, cybersécurité et opérations de flottes, interopérabilité (formats événements+spikes), logistique de cartouches, modèles économiques au résultat, et usages industriels, agricoles régénératifs, nucléaires, sous-marins et médicaux. Abstract ENThis document, produced with the assistance of ChatGPT 5.2 Thinking and Gemini 3 Raisonnement, is released under the Apache 2.0 licence. It is a voluntary defensive publication (prior art) and therefore enters the prior art upon release under the applicable patent statutes (art. L 611-11 CPI / art. 54(2) CBE). It discloses, in an enabling manner, a portfolio that fuses soft robotics (HASEL/EAP actuation), event-based vision (DVS), neuromorphic computing (SNN), and swarm intelligence. The disclosure spans devices and sensors, event-first control loops, roll-to-roll manufacturing and end-of-line QA, cyber-secure fleet operations, interoperability standards for event+spike telemetry, cartridge logistics and field repair, outcome-based metering and SLA instrumentation, and applications in high-throughput sorting, precision/regenerative agriculture, nuclear maintenance, underwater monitoring, and medical/rehabilitation systems. Each proposal is classified with IPC/CPC codes and can be timestamped (RFC 3161 / FreeTSA). Timestamp : 2026-08-17T10:45:25ZSHA-256 : 13b3e2bc50c638e594d623990f13159039dd0fb8f0b0968e18a3300649110a89 Liste des innovations & classification (IPC ; CPC) :1. DVS–HASEL soft gripper — IPC B25J 15/00 ; CPC B25J 15/122. DVS sorting calibration rig — IPC G01D 18/00 ; CPC G01D 18/003. HASEL sensing skin laminate — IPC G01L 5/00 ; CPC G01L 5/164. Biodegradable electrohydraulic actuator — IPC C08L 67/00 ; CPC C08L 67/025. Printable EAP electrode ink — IPC H01B 1/12 ; CPC H01B 1/126. Self-healing dielectric composite — IPC C08K 3/36 ; CPC C08K 3/367. Roll-to-roll HASEL pouch line — IPC B29C 65/00 ; CPC B29C 65/788. 3D-printed soft body + circuits — IPC B29C 64/118 ; CPC B29C 64/1189. Soft underwater encapsulation stack — IPC B29C 71/00 ; CPC B29C 71/0210. Event-driven SNN HASEL control — IPC G06N 3/04 ; CPC G06N 3/04511. Event-based actuator fatigue detection — IPC G05B 23/02 ; CPC G05B 23/0212. Edge event-stream compression codec — IPC H04N 5/00 ; CPC H04N 5/23213. Spike-packet swarm protocol — IPC H04W 4/80 ; CPC H04W 4/8014. Neuromorphic swarm task allocator — IPC G06Q 10/04 ; CPC G06Q 10/063915. Safe HV charge scheduler — IPC H02M 3/155 ; CPC H02M 3/15816. Swarm geofencing operations — IPC G08G 5/00 ; CPC G08G 5/0017. Radiation-hardened soft robot module — IPC G21C 19/00 ; CPC G21C 19/0018. DVS-to-intensity reconstruction — IPC H04N 5/232 ; CPC H04N 5/23219. DVS+EMG SNN exosuit fusion — IPC A61H 1/02 ; CPC A61H 1/0220. Closed-loop rehab dosing method — IPC A61H 1/00 ; CPC A61H 1/0021. Soft endoscope targeted delivery — IPC A61M 31/00 ; CPC A61M 31/0022. Low-power EAP assist patch — IPC A61F 5/01 ; CPC A61F 5/0123. Federated learning for agri swarms — IPC G06F 18/232 ; CPC G06F 18/232124. Event+spike interoperability standard — IPC G06F 9/54 ; CPC G06F 9/54125. Tamper-proof swarm audit ledger — IPC G06Q 20/38 ; CPC G06Q 20/38226. Swarm supervisor cockpit UI — IPC G05B 19/042 ; CPC G05B 19/04227. Hybrid ultra-fast waste sorter cell — IPC B07C 5/34 ; CPC B07C 5/34228. Underwater soft-drone swarm system — IPC B63G 8/00 ; CPC B63G 8/0029. Swarm soil-compaction sens","author":[{"family":"Pillet","given":"Xavier"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21978440","URL":"https://doi.org/10.5281/zenodo.21978440","source":"datacite"},{"id":"doi:10.5281/zenodo.20471387","type":"article-journal","title":"Improving Pakistan's Human Development Index: Strategies for Sustainable Growth and National Progress","abstract":"Abstract The Human Development Index (HDI) is a composite measure developed by the United Nations Development Programme (UNDP) to assess a country's progress in health, education, and standard of living. Pakistan continues to face significant challenges in improving its HDI ranking due to issues such as low educational attainment, inadequate healthcare infrastructure, gender inequality, poverty, and unemployment. This article examines the major factors affecting Pakistan's HDI and proposes practical strategies to accelerate human development. The study highlights the importance of investments in education, healthcare, economic growth, women's empowerment, technological innovation, and good governance as key drivers of sustainable human development. Keywords: Human Development Index, Pakistan, education, healthcare, poverty reduction, economic development, gender equality, sustainable development Introduction Human development refers to the process of expanding people's freedoms, opportunities, and capabilities to live productive and fulfilling lives. The Human Development Index (HDI), introduced by the United Nations Development Programme (UNDP), measures development using three primary indicators: life expectancy, education, and gross national income per capita (UNDP, 2025). Pakistan has experienced gradual improvements in human development over recent decades; however, its HDI remains relatively low compared to many developing countries. According to the UNDP Human Development Report 2025, Pakistan's HDI score is 0.544, placing it in the low human development category (UNDP, 2025). This situation necessitates comprehensive policy reforms aimed at improving the quality of life for its citizens. Understanding the Human Development Index The Human Development Index consists of three dimensions: Health: Measured by life expectancy at birth. Education: Measured by mean years of schooling and expected years of schooling. Income: Measured by Gross National Income (GNI) per capita. These dimensions collectively provide a broader understanding of development beyond economic growth alone (UNDP, 2025). Challenges Affecting Pakistan's HDI Educational Deficiencies Education remains one of the most critical challenges facing Pakistan. Despite improvements in literacy rates, millions of children remain out of school, particularly in rural areas and among girls. Educational spending remains below international recommendations, resulting in inadequate infrastructure, teacher shortages, and poor learning outcomes (UNESCO, 2024). Low educational attainment directly impacts employment opportunities, productivity, and overall economic development. Healthcare Challenges Pakistan's healthcare system faces numerous obstacles, including inadequate funding, insufficient medical facilities, shortages of healthcare professionals, and disparities between urban and rural healthcare services. Maternal and infant mortality rates remain higher than global averages, while malnutrition and preventable diseases continue to affect large segments of the population (World Bank, 2024). Improving healthcare services would significantly increase life expectancy and contribute positively to HDI performance. Poverty and Income Inequality Poverty remains a major barrier to human development in Pakistan. A substantial proportion of the population lacks access to quality education, healthcare, clean water, and sanitation facilities. Income inequality further exacerbates these challenges by limiting opportunities for disadvantaged groups (World Bank, 2024). Reducing poverty is essential for improving living standards and enhancing human development outcomes. Gender Inequality Gender disparities continue to limit Pakistan's development potential. Women often face barriers in education, employment, healthcare access, and political participation. Research demonstrates that societies with greater gender equality tend to achieve higher levels of human development and economic ","author":[{"family":"Hussain","given":"Zahid"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20471387","URL":"https://doi.org/10.5281/zenodo.20471387","source":"datacite"},{"id":"doi:10.5281/zenodo.20471388","type":"article-journal","title":"Improving Pakistan's Human Development Index: Strategies for Sustainable Growth and National Progress","abstract":"Abstract The Human Development Index (HDI) is a composite measure developed by the United Nations Development Programme (UNDP) to assess a country's progress in health, education, and standard of living. Pakistan continues to face significant challenges in improving its HDI ranking due to issues such as low educational attainment, inadequate healthcare infrastructure, gender inequality, poverty, and unemployment. This article examines the major factors affecting Pakistan's HDI and proposes practical strategies to accelerate human development. The study highlights the importance of investments in education, healthcare, economic growth, women's empowerment, technological innovation, and good governance as key drivers of sustainable human development. Keywords: Human Development Index, Pakistan, education, healthcare, poverty reduction, economic development, gender equality, sustainable development Introduction Human development refers to the process of expanding people's freedoms, opportunities, and capabilities to live productive and fulfilling lives. The Human Development Index (HDI), introduced by the United Nations Development Programme (UNDP), measures development using three primary indicators: life expectancy, education, and gross national income per capita (UNDP, 2025). Pakistan has experienced gradual improvements in human development over recent decades; however, its HDI remains relatively low compared to many developing countries. According to the UNDP Human Development Report 2025, Pakistan's HDI score is 0.544, placing it in the low human development category (UNDP, 2025). This situation necessitates comprehensive policy reforms aimed at improving the quality of life for its citizens. Understanding the Human Development Index The Human Development Index consists of three dimensions: Health: Measured by life expectancy at birth. Education: Measured by mean years of schooling and expected years of schooling. Income: Measured by Gross National Income (GNI) per capita. These dimensions collectively provide a broader understanding of development beyond economic growth alone (UNDP, 2025). Challenges Affecting Pakistan's HDI Educational Deficiencies Education remains one of the most critical challenges facing Pakistan. Despite improvements in literacy rates, millions of children remain out of school, particularly in rural areas and among girls. Educational spending remains below international recommendations, resulting in inadequate infrastructure, teacher shortages, and poor learning outcomes (UNESCO, 2024). Low educational attainment directly impacts employment opportunities, productivity, and overall economic development. Healthcare Challenges Pakistan's healthcare system faces numerous obstacles, including inadequate funding, insufficient medical facilities, shortages of healthcare professionals, and disparities between urban and rural healthcare services. Maternal and infant mortality rates remain higher than global averages, while malnutrition and preventable diseases continue to affect large segments of the population (World Bank, 2024). Improving healthcare services would significantly increase life expectancy and contribute positively to HDI performance. Poverty and Income Inequality Poverty remains a major barrier to human development in Pakistan. A substantial proportion of the population lacks access to quality education, healthcare, clean water, and sanitation facilities. Income inequality further exacerbates these challenges by limiting opportunities for disadvantaged groups (World Bank, 2024). Reducing poverty is essential for improving living standards and enhancing human development outcomes. Gender Inequality Gender disparities continue to limit Pakistan's development potential. Women often face barriers in education, employment, healthcare access, and political participation. Research demonstrates that societies with greater gender equality tend to achieve higher levels of human development and economic ","author":[{"family":"Hussain","given":"Zahid"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20471388","URL":"https://doi.org/10.5281/zenodo.20471388","source":"datacite"},{"id":"doi:10.5281/zenodo.20496576","type":"article-journal","title":"Lev's Ternary Logic Quotes: Philosophy, Governance, Architecture, and the Journey of Building Ternary Logic","abstract":"Lev's Ternary Logic Quotes is a collection of 2,610 quotations organized across 61 categories, exploring the philosophical, governance, economic, ethical, and architectural principles underlying Ternary Logic (TL), a peer-reviewed institutional governance framework published in AI and Ethics (Springer Nature, DOI: 10.1007/s43681-026-01124-0). The quotations were gathered over months of framework development and preserved in the order they emerged, from the Goukassian Vow through the Eight Pillars, Epistemic Hold, Immutable Ledger, dual-lane architecture, tri-cameral governance, hardware implementation, and applications across finance, anti-money laundering (AML), smart contracts, blockchain governance, artificial intelligence, and autonomous systems. Rather than presenting a traditional technical specification, this work documents the evolution of an idea through aphorisms, reflections, principles, and architectural observations. It is intended for researchers, engineers, policymakers, governance architects, philosophers of technology, and readers interested in the relationship between intelligence, accountability, uncertainty, legitimacy, and human judgment. The collection closes with \"Creating While Time Watches,\" a personal reflection on creation, mortality, legacy, and the experience of building under conditions of medical urgency. This is not a textbook. It is a record of a mind at work, building something it hoped would outlive it. Author: Lev GoukassianORCID: 0009-0006-5966-1243Related framework: Ternary Logic (TL) | FractonicMind/TernaryLogic","author":[{"family":"Goukassian","given":"Lev"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20496576","URL":"https://doi.org/10.5281/zenodo.20496576","source":"datacite"},{"id":"doi:10.5281/zenodo.21479665","type":"article-journal","title":"Lev's Ternary Logic Quotes: Philosophy, Governance, Architecture, and the Journey of Building Ternary Logic","abstract":"Lev's Ternary Logic Quotes is a collection of 2,610 quotations organized across 61 categories, exploring the philosophical, governance, economic, ethical, and architectural principles underlying Ternary Logic (TL), a peer-reviewed institutional governance framework published in AI and Ethics (Springer Nature, DOI: 10.1007/s43681-026-01124-0). The quotations were gathered over months of framework development and preserved in the order they emerged, from the Goukassian Vow through the Eight Pillars, Epistemic Hold, Immutable Ledger, dual-lane architecture, tri-cameral governance, hardware implementation, and applications across finance, anti-money laundering (AML), smart contracts, blockchain governance, artificial intelligence, and autonomous systems. Rather than presenting a traditional technical specification, this work documents the evolution of an idea through aphorisms, reflections, principles, and architectural observations. It is intended for researchers, engineers, policymakers, governance architects, philosophers of technology, and readers interested in the relationship between intelligence, accountability, uncertainty, legitimacy, and human judgment. The collection closes with \"Creating While Time Watches,\" a personal reflection on creation, mortality, legacy, and the experience of building under conditions of medical urgency. This is not a textbook. It is a record of a mind at work, building something it hoped would outlive it. Author: Lev GoukassianORCID: 0009-0006-5966-1243Related framework: Ternary Logic (TL) | FractonicMind/TernaryLogic","author":[{"family":"Goukassian","given":"Lev"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.21479665","URL":"https://doi.org/10.5281/zenodo.21479665","source":"datacite"},{"id":"doi:10.5281/zenodo.19544014","type":"article-journal","title":"AI and the Soul of Medicine","abstract":"The rapid integration of large language models into clinical medicine raises ethical, philosophical, and existential questions that extend beyond traditional concerns about patient safety and data privacy. These artificial intelligence systems now perform clinical tasks such as diagnostic reasoning, protocol selection, and patient communication at levels comparable to trained physicians, yet the frameworks governing their deployment remain anchored in an era when machines could not convincingly simulate human understanding. The ethical implications of incorporating large language models into healthcare are examined through the lens of historical precedent, the physician-patient relationship, and the evolving meaning of human agency in clinical care. Drawing on examples from organ transplantation, intensive care medicine, and genetic engineering, medicine has repeatedly absorbed technologies that challenged foundational ethical assumptions, and each such absorption required the development of new governance frameworks rather than the rejection of the technology itself. Three critical tensions specific to large language models emerge: the gap between linguistic competence and genuine understanding, the redistribution of clinical authority from physicians to algorithms, and the erosion of empathy as a uniquely human contribution to healing. Ethical integration of large language models into medicine requires not only the application of existing bioethical principles but also the development of new frameworks that address the unprecedented capacity of these systems to inhabit the communicative and relational spaces previously reserved for human clinicians.","author":[{"family":"Heston","given":"Thomas"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19544014","URL":"https://doi.org/10.5281/zenodo.19544014","source":"datacite"},{"id":"doi:10.5281/zenodo.19558022","type":"article-journal","title":"AI and the Soul of Medicine","abstract":"The rapid integration of large language models into clinical medicine raises ethical, philosophical, and existential questions that extend beyond traditional concerns about patient safety and data privacy. These artificial intelligence systems now perform clinical tasks such as diagnostic reasoning, protocol selection, and patient communication at levels comparable to trained physicians, yet the frameworks governing their deployment remain anchored in an era when machines could not convincingly simulate human understanding. The ethical implications of incorporating large language models into healthcare are examined through the lens of historical precedent, the physician-patient relationship, and the evolving meaning of human agency in clinical care. Drawing on examples from organ transplantation, intensive care medicine, and genetic engineering, medicine has repeatedly absorbed technologies that challenged foundational ethical assumptions, and each such absorption required the development of new governance frameworks rather than the rejection of the technology itself. Three critical tensions specific to large language models emerge: the gap between linguistic competence and genuine understanding, the redistribution of clinical authority from physicians to algorithms, and the erosion of empathy as a uniquely human contribution to healing. Ethical integration of large language models into medicine requires not only the application of existing bioethical principles but also the development of new frameworks that address the unprecedented capacity of these systems to inhabit the communicative and relational spaces previously reserved for human clinicians.","author":[{"family":"Heston","given":"Thomas"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.19558022","URL":"https://doi.org/10.5281/zenodo.19558022","source":"datacite"},{"id":"doi:10.5281/zenodo.20685378","type":"article-journal","title":"From H. R. 9510 to Federal Law: A Narrative Case for Verified Physical AI Oncology Trials","abstract":"This review makes a case for legislators who shape medical and artificial intelligence law, but have little to no robotics or frontier large language model (LLM) application experience. It is built on a single mechanism, verification before generation, in which a software agent proposes a clinical action and a ten-gate verification, validation, and uncertainty quantification examination either accepts it, escalates it to a qualified human, or blocks it before it can reach a patient. The argument is organized as eight emotional pillars that legislative-advocacy research finds most persuasive: compassion, fear of preventable harm, moral outrage, hope, responsibility, protection of vulnerable people, trust, and urgency. Each pillar pairs a human appeal with a credible, cited fact. The review draws on a documented engineering lineage, including a surgical-humanoid assurance run that passed 172 of 172 automated tests across a ten-gate suite, and on the published advocacy literature describing how testimony, coalition building, and policy entrepreneurship move a bill through markup and reconciliation. The conclusion is that Physical AI Trial Bill H. R. 9510 2026 should be enacted into Federal law.","author":[{"family":"Kawchak","given":"Kevin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20685378","URL":"https://doi.org/10.5281/zenodo.20685378","source":"datacite"},{"id":"doi:10.5281/zenodo.20685379","type":"article-journal","title":"From H. R. 9510 to Federal Law: A Narrative Case for Verified Physical AI Oncology Trials","abstract":"This review makes a case for legislators who shape medical and artificial intelligence law, but have little to no robotics or frontier large language model (LLM) application experience. It is built on a single mechanism, verification before generation, in which a software agent proposes a clinical action and a ten-gate verification, validation, and uncertainty quantification examination either accepts it, escalates it to a qualified human, or blocks it before it can reach a patient. The argument is organized as eight emotional pillars that legislative-advocacy research finds most persuasive: compassion, fear of preventable harm, moral outrage, hope, responsibility, protection of vulnerable people, trust, and urgency. Each pillar pairs a human appeal with a credible, cited fact. The review draws on a documented engineering lineage, including a surgical-humanoid assurance run that passed 172 of 172 automated tests across a ten-gate suite, and on the published advocacy literature describing how testimony, coalition building, and policy entrepreneurship move a bill through markup and reconciliation. The conclusion is that Physical AI Trial Bill H. R. 9510 2026 should be enacted into Federal law.","author":[{"family":"Kawchak","given":"Kevin"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.20685379","URL":"https://doi.org/10.5281/zenodo.20685379","source":"datacite"},{"id":"doi:10.5281/zenodo.18863430","type":"article-journal","title":"From Zadig to Artificial Intelligence: Clinical Observation, Epistemic Fragility, and the Future of Medical Reasoning","abstract":"ABSTRACT: Background: The rapid integration of artificial intelligence (AI) into clinical practice has revived longstanding debates about the foundations of diagnostic reasoning. While AI systems rely on large datasets derived from past patients, the clinician’s craft remains rooted in direct observation, sensory examination, and contextual interpretation. Objective: To trace the intellectual lineage connecting Voltaire’s Zadig, the evolution of the clinical method, the deductive tradition exemplified by Sherlock Holmes, and the contemporary challenges posed by AI in medicine. Discussion: Voltaire’s Zadig anticipates the logic of modern clinical reasoning through its emphasis on inference from subtle signs. This evidential paradigm reappears in the anatomic and clinical revolution, bacteriology, and evidence‑based medicine, and is later embodied in Arthur Conan Doyle’s Sherlock Holmes, whose method was explicitly modelled on clinical diagnosis. AI introduces new fundamental risks, including algorithmic affixing, loss of sensory skills, and the perpetuation of historical preconceptions. The internist’s observational abilities remain essential for generating primary data and contextualizing algorithmic outputs. Conclusion: The future of medicine requires harmonizing traditional clinical skills with AI‑driven tools. The internist must remain a reader of signs, a critic of evidence, and a guardian of epistemic humility. KEYWORDS: Clinical reasoning, Artificial intelligence, Medical epistemology, Voltaire, Sherlock Holmes, Internal medicine, Diagnostic method.","author":[{"family":"Asmell Ramos Cabrera - Md","given":"Msc"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18863430","URL":"https://doi.org/10.5281/zenodo.18863430","source":"datacite"},{"id":"doi:10.5281/zenodo.18863431","type":"article-journal","title":"From Zadig to Artificial Intelligence: Clinical Observation, Epistemic Fragility, and the Future of Medical Reasoning","abstract":"ABSTRACT: Background: The rapid integration of artificial intelligence (AI) into clinical practice has revived longstanding debates about the foundations of diagnostic reasoning. While AI systems rely on large datasets derived from past patients, the clinician’s craft remains rooted in direct observation, sensory examination, and contextual interpretation. Objective: To trace the intellectual lineage connecting Voltaire’s Zadig, the evolution of the clinical method, the deductive tradition exemplified by Sherlock Holmes, and the contemporary challenges posed by AI in medicine. Discussion: Voltaire’s Zadig anticipates the logic of modern clinical reasoning through its emphasis on inference from subtle signs. This evidential paradigm reappears in the anatomic and clinical revolution, bacteriology, and evidence‑based medicine, and is later embodied in Arthur Conan Doyle’s Sherlock Holmes, whose method was explicitly modelled on clinical diagnosis. AI introduces new fundamental risks, including algorithmic affixing, loss of sensory skills, and the perpetuation of historical preconceptions. The internist’s observational abilities remain essential for generating primary data and contextualizing algorithmic outputs. Conclusion: The future of medicine requires harmonizing traditional clinical skills with AI‑driven tools. The internist must remain a reader of signs, a critic of evidence, and a guardian of epistemic humility. KEYWORDS: Clinical reasoning, Artificial intelligence, Medical epistemology, Voltaire, Sherlock Holmes, Internal medicine, Diagnostic method.","author":[{"family":"Asmell Ramos Cabrera - Md","given":"Msc"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.18863431","URL":"https://doi.org/10.5281/zenodo.18863431","source":"datacite"},{"id":"oa:W4406873552","type":"article-journal","title":"Primary School Students' Perceptions of Artificial Intelligence: Metaphor and Drawing Analysis","abstract":"ABSTRACT Due to the frequent use of artificial intelligence (AI) technologies in daily life, it is thought that primary school students acquire information about this concept from various sources. The way these sources present AI may affect students' perceptions of AI. In the study, it was aimed to examine the perceptions of third and fourth grade primary school students about AI through metaphors and drawings. This research, which was conducted with the participation of 262 students, was conducted with the phenomenological design. When the metaphors of the participants were analysed, it was determined that they produced 100 metaphors, and these metaphors were evaluated in 17 categories as humanistic feature, information source, danger, development, superhuman feature, service, source of happiness, productivity, orientation, commitment, pervasiveness, necessity, security, speed, difficulty, virtual environment and uncertainty. Accordingly, it was determined that the participants evaluated AI from many different perspectives and produced the most metaphors in the categories of humanistic feature, information source and danger. It was determined that the metaphors human, brain and living were prominent in the human characteristic category; the metaphors teacher, wise and book were prominent in the source of information category; and finally, the metaphors enemy, weapon and monster were prominent in the danger category. When the drawing findings were analysed, it was determined that 37 codes represented four categories: purpose, object, interaction and environment. In the purpose category, service, source of information, and source of happiness; in the object category, mostly humanoid robot; in the interaction category, emphasising interaction; and in the environment category, the environment was not specified. In line with the findings obtained, literature discussions were made and suggestions were made.","author":[{"family":"Kalemkuş","given":"Jale"},{"family":"Kalemkuş","given":"Fatih"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/ejed.70007","URL":"https://doi.org/10.1111/ejed.70007","source":"openalex"},{"id":"oa:W4410363862","type":"article-journal","title":"Artificial Intelligence and Assistive Robotics in Healthcare Services: Applications in Silver Care","abstract":"Artificial intelligence (AI) and assistive robotics can transform older-person care by offering new, personalised solutions for an ageing population. This paper outlines recent advances in AI-driven applications and robotic assistance in silver care, emphasising their role in improved healthcare services, quality of life and ageing-in-place and alleviating pressure on healthcare systems. Advances in machine learning, natural language processing and computer vision have enabled more accurate early diagnosis, targeted treatment plans and robust remote monitoring for elderly patients. These innovations support continuous health tracking and timely interventions to improve patient outcomes and extend home-based care. In addition, AI-powered assistive robots with advanced motion control and adaptive response mechanisms are studied to support physical and cognitive health. Among these, companion robots, often enhanced with emotional AI, have shown potential in reducing loneliness and increasing connectedness. The combined goal of these technologies is to offer holistic patient-centred care, which preserves the autonomy and dignity of our seniors. This paper also touches on the technical and ethical challenges of integrating AI/robotics into eldercare, like privacy and accessibility, and alludes to future directions on optimising AI-human interaction, expanding preventive healthcare applications and creating an effective, ethical framework for eldercare in the digital age.","author":[{"family":"Masala","given":"Giovanni"},{"family":"Giorgi","given":"Ioanna"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/ijerph22050781","URL":"https://doi.org/10.3390/ijerph22050781","source":"openalex"},{"id":"oa:W4409849237","type":"article-journal","title":"NAVIGATING ETHICS AND RISK IN ARTIFICIAL INTELLIGENCE APPLICATIONS WITHIN INFORMATION TECHNOLOGY: A SYSTEMATIC REVIEW","abstract":"Advancements in artificial intelligence (AI) have profoundly transformed a wide array of sectors, with Information Technology (IT) standing at the forefront of this revolution. AI technologies have reshaped IT operations by introducing new levels of automation, predictive capabilities, decision-making precision, and efficiency, resulting in sweeping changes across organizational infrastructures and service delivery models. However, alongside these technological breakthroughs, the integration of AI has surfaced numerous ethical concerns and associated risks that warrant critical and comprehensive exploration. Issues such as algorithmic bias, privacy breaches, lack of transparency, accountability dilemmas, and cybersecurity vulnerabilities remain pervasive, posing significant challenges for organizations seeking to deploy AI responsibly. Recognizing these complexities, this study focuses on examining the ethical challenges, risk factors, and essential considerations surrounding the deployment of AI within IT environments. This research addresses the urgent and growing necessity for organizations to adopt robust ethical frameworks and effective risk management strategies to ensure that AI integration promotes fairness, transparency, and accountability. A comprehensive synthesis of existing literature further enriches this foundation by offering diverse perspectives on the ethical and risk-related challenges posed by AI, highlighting both the transformative potential and the vulnerabilities associated with its integration into IT systems. Through this review, key gaps in the current body of knowledge are identified, particularly regarding the practical implementation of ethical standards and risk mitigation strategies across varied organizational contexts. Adopting a qualitative research methodology, the study employs a case study approach to explore the intricate, multifaceted issues involved in AI integration within IT operations. This methodological choice allows for a nuanced understanding of real-world scenarios, organizational behaviors, and stakeholder dynamics related to AI deployment. Data collection is meticulously based on trustworthy sources, including peer-reviewed academic journals, authoritative industry reports, regulatory and governmental publications, and credible news articles, ensuring the comprehensiveness and reliability of insights gathered. By triangulating data across multiple domains, the study captures a holistic view of the ethical landscape surrounding AI in IT. Finally, the research culminates with an in-depth interpretation of findings, accompanied by practical recommendations and implications. These outcomes aim to contribute to the development of ethical, sustainable, and responsible AI integration practices within the IT industry, supporting organizations in navigating the complex interplay between technological innovation and ethical accountability.","author":[{"family":"Ahmed","given":"Ishtiaque"}],"issued":{"date-parts":[[2025]]},"DOI":"10.63125/590d7098","URL":"https://doi.org/10.63125/590d7098","source":"openalex"},{"id":"oa:W4413205459","type":"article-journal","title":"How generative artificial intelligence transforms teaching and influences student wellbeing in future education","abstract":"Artificial intelligence (AI), particularly generative technologies, and large language models are transforming modern education. On the one hand, these tools can automate certain aspects of teaching, personalize educational materials, and improve learning efficiency. However, concerns are emerging regarding the impact of AI on the quality of education, the mental health of students, and fundamental academic values. This article reviews the scientific literature on the applications of artificial intelligence in education, with a special emphasis on its impact on student mental health. Based on an analysis of 120 scientific articles, the study automatically extracted data on AI implementation's potential opportunities and threats. A novel aspect of this work is identifying factors that can be considered both opportunities and threats depending on context. In addition, a frequency analysis of keywords and phrases uncovered many opportunities and challenges. All identified aspects are characterized in the article. The article also highlights key barriers to using large language models (LLMs) for detecting student mental health issues, such as underestimating suicide risk, difficulties with interpreting subtle language, biases in training data, lack of cultural sensitivity, and unresolved ethical concerns. These challenges illustrate why generative AI is not yet reliable for supporting student mental health, especially in high-risk situations. One of the key conclusions is that the use of generative AI to support student mental health is seldom addressed in existing review articles, likely due to the current unreliability of this technology.","author":[{"family":"Jukiewicz","given":"Marcin"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/feduc.2025.1594572","URL":"https://doi.org/10.3389/feduc.2025.1594572","source":"openalex"},{"id":"oa:W4412944013","type":"article-journal","title":"Perspectives, challenges and future of artificial intelligence in personalised nutrition research","abstract":"Personalised nutrition (PN) has emerged as an approach to optimise individual health outcomes through more targeted and tailored dietary recommendations based on unique genetic, phenotypic, medical, lifestyle and contextual factors. The application of artificial intelligence (AI) presents an opportunity to achieve personalised nutrition advice at a scale that has population impact. This review introduces a nutrition audience to different AI applications and offers insights into the concepts of AI that might be relevant to the field of nutrition research. The current and future uses of AI in PN are discussed, as well as the potential benefits and challenges to their application. AI-driven solutions have the potential to improve health and reduce the risk of disease because they can consider more information about an individual in making recommendations. However, challenges such as data interoperability, ethical considerations, and model interpretability remain an issue limiting widespread use at this point. This review will provide a foundational understanding of the application of AI within PN and help to identify opportunities to leverage the potential of AI in transforming dietary guidance and enhancing health outcomes through innovative solutions.","author":[{"family":"Brankovic","given":"Aida"},{"family":"Hendrie","given":"Gilly"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1017/s0029665125100657","URL":"https://doi.org/10.1017/s0029665125100657","source":"openalex"},{"id":"oa:W4409202874","type":"article-journal","title":"Artificial Intelligence and IoT for Smart Waste Management: Challenges, Opportunities, and Future Directions","abstract":"Indonesia’s waste management system struggles to keep pace with rapid population growth and urbanization, resulting in inefficient waste collection, environmental degradation, and low recycling rates. The country predominantly relies on open dumping and landfilling, which contribute significantly to pollution and greenhouse gas emissions. This study explores the transformative role of Artificial Intelligence (AI) and the Internet of Things (IoT) in waste management, focusing on smart waste collection, automated sorting, real-time monitoring, and predictive analytics. AI-driven waste classification enhances recycling efficiency, while IoT-enabled smart bins optimize collection routes, reducing operational costs and landfill dependency. Despite these advantages, challenges such as high implementation costs, digital infrastructure limitations, and data privacy concerns hinder widespread adoption. This study highlights that policy support, investment in digital infrastructure, and stakeholder collaboration are crucial for successful implementation. By leveraging AI and IoT, Indonesia can significantly improve waste management efficiency, minimize environmental impact, and advance circular economy initiatives. The findings suggest that, with the right policies and investments, AI-driven waste management can drive sustainability, reduce waste mismanagement, and promote resource optimization, making it a vital strategy for future urban development in Indonesia.","author":[{"family":"Fuqaha","given":"Sameh"},{"family":"Nursetiawan","given":"Nursetiawan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.62411/faith.3048-3719-85","URL":"https://doi.org/10.62411/faith.3048-3719-85","source":"openalex"},{"id":"oa:W4411131699","type":"article-journal","title":"Harnessing Artificial Intelligence in Lifestyle Medicine: Opportunities, Challenges, and Future Directions","abstract":"Lifestyle medicine (LM) offers a transformative, evidence-based approach to preventing, managing, and potentially reversing chronic diseases by targeting modifiable lifestyle factors such as nutrition, physical activity, sleep, stress, substance use, and social connectivity. However, real-world implementation of LM is often hindered by patient adherence issues, limited clinical time, and the need for ongoing personalized support. Artificial intelligence (AI), with its capabilities in data processing, pattern recognition, and predictive modeling, presents a unique opportunity to overcome these barriers and enhance the reach and precision of LM interventions. This narrative review explores AI's integration into LM's core domains. In nutrition, AI facilitates real-time dietary assessment and personalized recommendations through image recognition and machine learning. In physical activity and fitness, AI-powered wearable devices deliver tailored feedback, support virtual coaching, and predict injury risk. AI applications in sleep medicine allow for continuous, non-invasive monitoring and the early detection of sleep disorders. AI-driven cognitive behavioral therapy chatbots and biosensor-based stress prediction tools provide scalable, cost-effective support for mental health and stress management. Moreover, AI is pivotal in chronic disease prevention by integrating lifestyle data with electronic health records to forecast disease trajectories and optimize interventions. Despite these advances, several challenges remain. Data privacy concerns, algorithmic bias, regulatory ambiguities, and varying user trust and engagement levels must be addressed to ensure equitable and ethical implementation. AI's integration with digital twin technology and precision LM represents the next frontier in personalized health. As LM continues to evolve, AI will be indispensable in driving a more proactive, participatory, and person-centered model of care that meets the complex demands of chronic disease management in the 21st century.","author":[{"family":"Saeed","given":"Diana"},{"family":"Nashwan","given":"Abdulqadir"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.85580","URL":"https://doi.org/10.7759/cureus.85580","source":"openalex"},{"id":"oa:W4406548567","type":"article-journal","title":"The ethics of artificial intelligence use in university libraries in Zimbabwe","abstract":"Introduction: The emergence of artificial intelligence (AI) has revolutionised higher education teaching and learning. AI has the power to analyse large amounts of data and make intelligent predictions thus changing the whole teaching and learning processes. However, such a rise has led to institutions questioning the morality of these applications. The changes have left librarians and educators worried about the major ethical questions surrounding privacy, equality of information, protection of intellectual property, cheating, misinformation and job security. Libraries have always been concerned about ethics and many go out of their way to make sure communities are educated about the ethical question. However, the emergence of artificial intelligence has caught them unaware. Methods: This research investigates the preparedness of higher education librarians to support the ethical use of information within the higher and tertiary education fraternity. A qualitative approach was used for this study. Interviews were done with thirty purposively selected librarians and academics from universities in Zimbabwe. Results: Findings indicated that many university libraries in Zimbabwe are still at the adoption stage of artificial intelligence. It was also found that institutions and libraries are not yet prepared for AI use and are still crafting policies on the use of AI. Discussion: Libraries seem prepared to adopt AI. They are also prepared to offer training on how to protect intellectual property but have serious challenges in issues of transparency, data security, plagiarism detection and concerns about job losses. However, with no major ethical policies having been crafted on AI use, it becomes challenging for libraries to full adopt its usage.","author":[{"family":"Tsekea","given":"Stephen"},{"family":"Mandoga","given":"Edward"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/frma.2024.1522423","URL":"https://doi.org/10.3389/frma.2024.1522423","source":"openalex"},{"id":"oa:W7160824505","type":"article-journal","title":"Evaluation of Artificial Intelligence and Blockchain Integration Utilizing the DEM ATEL Method to Improve Privacy and Transparent in the Financial Sector","abstract":"d IoT security perspective. It makes use of three essential Blockchain features— transparency, immutability, and decentralization— to build environment that are reliable and impenetrable. This application is realized through the utilization of features such as AI-driven fraud detection, Blockchain security, data privacy, the reliability of Smart Contracts, transaction speed, and system scalability. The result is, Blockchain-IoT Security Perspective, the first rank is System Scalability, the lowest rank is AI-based Fraud Detection, Blockchain Security is the fourth rank, Data Privacy is the fifth rank, Smart Contract Reliability is the third rank, and Transaction Speed is the first rank.","author":[{"family":"Dommeti","given":"Rajendar"}],"issued":{"date-parts":[[2026]]},"DOI":"10.55124/ijbs.v1i1.103","URL":"https://doi.org/10.55124/ijbs.v1i1.103","source":"openalex"},{"id":"oa:W4408143893","type":"article-journal","title":"Artificial intelligence machines as relational nonhuman actors in entrepreneurial teams","abstract":"We theoretically examine how artificial intelligence (AI) machines interact with and shape strategic decision-making within entrepreneurial teams. First, we explain how AI machines lead and assist teams to engage in cognitively complex decisions and explore new ways to combine resources. Second, we unpack this strategic decision-making reconfiguration process and examine how AI machines allow entrepreneurial teams to better manage diversity and use it as a substitute for external knowledge. Overall, we propose the idea that AI technologies cannot be conceptualized as merely additional resources but as relational nonhuman actors in entrepreneurial teams. Then, we encourage scholars to reflect on human-AI interactions in entrepreneurial teams and outline an agenda for future research.","author":[{"family":"Murtinu","given":"Samuele"},{"family":"Massis","given":"Alfredo"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/00472778.2025.2461031","URL":"https://doi.org/10.1080/00472778.2025.2461031","source":"openalex"},{"id":"oa:W4410293199","type":"article-journal","title":"Artificial intelligence and free will: generative agents utilizing large language models have functional free will","abstract":"Abstract Combining large language models (LLMs) with memory, planning, and execution units has made possible almost human-like agentic behavior, where the artificial intelligence creates goals for itself, breaks them into concrete plans, and refines the tactics based on sensory feedback. Do such generative LLM agents possess free will? Free will requires that an entity exhibits intentional agency, has genuine alternatives, and can control its actions. Building on Dennett’s intentional stance and List’s theory of free will, I will focus on functional free will, where we observe an entity to determine whether we need to postulate free will to understand and predict its behavior. Focusing on two running examples, the recently developed Voyager, an LLM-powered Minecraft agent, and the fictitious Spitenik, an assassin drone, I will argue that the best (and only viable) way of explaining both of their behavior involves postulating that they have goals, face alternatives, and that their intentions guide their behavior. While this does not entail that they have consciousness or that they possess physical free will, where their intentions alter physical causal chains, we must nevertheless conclude that they are agents whose behavior cannot be understood without postulating that they possess functional free will.","author":[{"family":"Martela","given":"Frank"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1007/s43681-025-00740-6","URL":"https://doi.org/10.1007/s43681-025-00740-6","source":"openalex"},{"id":"oa:W4415824542","type":"article-journal","title":"Artificial Intelligence in Clinical and Translational Science : From Bench Insights to Bedside Impact","abstract":"The healthcare and life sciences sectors are experiencing a transformative moment that cannot be overlooked. Our biological understanding, technology, and data are coalescing to leverage unprecedented opportunities for innovation [1, 2]. At the center of this transformation lies Artificial Intelligence (AI) and Machine Learning (ML), which have advanced from speculation to working technologies that can make actual differences in patient care and drug development [3, 4]. Only a few years ago, AI was framed in the context of its potential in clinical pharmacology, drug discovery, and development. Following the 2024 Nobel Prize in Chemistry, which was awarded for the AI-based prediction of protein structure, it became increasingly difficult to ignore the scientific merit of the technology [5]. We are now experiencing implementations that are changing how we approach these disciplines [6-8]. We have transcended previous discussions about whether AI will help and are asking more nuanced questions about how we deploy these technologies in a responsible manner, such that they deliver reliable and reproducible results, and produce meaningful value in clinical and translational research. The AI-themed issue in Clinical and Translational Science addresses exactly these questions by collating different viewpoints from around the field and providing substantive evidence to demonstrate the breadth of the current AI applications that are transforming the practice of clinical and translational sciences (Table 1). From the earliest phases of drug discovery through design and optimization of clinical trials, from developing personalized treatment approaches to monitoring drug safety postmarket approval, and to collecting real-world evidence—these contributions illustrate the current state of the art in utilizing AI in drug discovery and development and also characterize our current capabilities, while providing a vision for future innovation within clinical and translational sciences. In this piece, we aggregate contributions from over 30 manuscripts in this special issue, organized into three general categories (Figure 1, Table S1): Discovery and Preclinical Innovation, Clinical Development and Precision Medicine, and Postmarketing, Safety, and Real-World Implementation. We also discuss the cross-cutting aspects of regulation, ethics, and operations that may ensure AI achieves its potential. As readers progress through this editorial, they will see that the contributions in this issue do more than showcase technical innovation. They reflect a growing maturity in how the clinical pharmacology and translational science communities approach AI, which is not being expressed as a ‘one-size-fits-all’ solution but rather as a toolkit, where the impact of AI will depend on thoughtful integration, appropriate validation, and a deeper understanding of both its strengths and limitations. At the beginning of the drug development continuum—where drug developers identify targets, discover biomarkers, and create preclinical models—AI's intuitive ability to assess and map the meaning within massive, multidimensional datasets is, more often than not, beginning to be leveraged. At the earliest stages of discovery science, AI's transformative property is not necessarily its computational speed; rather, it is AI's ability to spot deep and subtle patterns that statistical approaches have most often failed to recognize, especially nonlinear relationships. This special issue includes many contributions that demonstrate how AI can now enable broader hypothesis generation, more rapid compound optimization, and more biologically founded predictions. One area that is rapidly progressing is in silico pharmacokinetic (PK) and pharmacodynamic (PD) modeling. Walter et al., make a systematic examination of ML-based approaches to empirical, compartmental, and physiologically based pharmacokinetic (PBPK) models for predicting plasma PK profiles in rats [9]. In a study with ov","author":[{"family":"Shahin","given":"Mohamed"},{"family":"Liu","given":"Qi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/cts.70383","URL":"https://doi.org/10.1111/cts.70383","source":"openalex"},{"id":"oa:W4407675019","type":"article-journal","title":"Exploring the Acceptability of Artificial Intelligence in Human Resources Management: Insights From Swiss Organizations","abstract":"ABSTRACT This study looks at perceptions of artificial intelligence (AI) systems in human resources (HR) management within Swiss organizations. Based on a survey experiment provided to 324 private and public HR professionals, it explores how UTAUT's predictors—performance expectancy, effort expectancy, social influence and facilitating conditions—as well as top management support, the Private/Public dimension and control variables—age, gender, time with organization and hierarchical position—influence their acceptability of four different type of AI HR tools. To do this, this article is based on a multiple regression method. Its main findings are that, irrespective of the type of tool, performance expectancy, effort expectancy and social influence positively influence the acceptability of the HR AI tools studied, whereas working in a public organization has systematically a negative influence. This makes a significant contribution to the literature by offering valuable insights into how these factors collectively shape the willingness of HR professionals to embrace AI technologies in their practices. It also offers an overview of the levers that organizations aiming to adopt these AI tools could act upon.","author":[{"family":"Revillod","given":"Guillaume"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/sres.3140","URL":"https://doi.org/10.1002/sres.3140","source":"openalex"},{"id":"oa:W4413055396","type":"article-journal","title":"Unleashing the Future of Endodontics: Exploring the Potential Role of Explainable Artificial Intelligence in Risk Stratification and Decision‐Making in Endodontics","abstract":"The incorporation of artificial intelligence (AI) into endodontics has predominantly focused on augmenting diagnostic accuracy, particularly in pinpointing periradicular radiolucencies and delineating the complex architecture of root canal systems via radiographic imaging and cone beam computed tomography (CBCT). These advancements represent a pivotal development within dental practice [1]; however, they merely scratch the surface of AI's broader potential to revolutionise the field of dentistry. The decision-making process in endodontic clinical practice is multifaceted, involving a myriad of factors that extend beyond diagnostic capabilities. Clinicians are tasked with integrating a thorough understanding of the patient's dental and medical history, recognising the distinct anatomical variations presented by each tooth, evaluating the implications of previous dental interventions, and formulating coherent treatment plans based on the preferences of the patient. In this nuanced landscape, the advent of Explainable Artificial Intelligence (XAI) emerges as a crucial development, fostering transparency and trust in AI-assisted workflows and paving the way for its broader acceptance and implementation in clinical settings. Despite the impressive predictive performance of numerous deep learning models, a central characteristic of these systems is their ‘black box’ nature. They generate outputs without elucidating the underlying mechanisms driving these conclusions. This lack of transparency presents significant ethical and clinical dilemmas for healthcare professionals. Clinicians may be reluctant to adopt AI technologies that fail to provide comprehensible justifications for their recommendations, particularly as societal expectations shift towards greater accountability and clarity in AI-generated medical decisions [2]. Furthermore, the expansion of regulatory frameworks aimed at enforcing ethical standards in healthcare underscores the imperative for transparency. This shift reinforces the notion that AI tools must not only achieve high accuracy but also be interpretable and justifiable to both healthcare practitioners and patients. Explainable AI (XAI) refers to a suite of machine learning techniques designed to illuminate the decision-making processes behind AI models. Unlike traditional models that may output a single prediction, XAI systems are engineered to provide insights into which input variables significantly influenced their conclusions [3]. In the context of endodontics, for instance, an XAI framework could identify that a recommendation for root canal retreatment is heavily influenced by a cluster of factors, including lesion size, the existence of missed root canals, the failure of coronal restorations, the specific type of tooth being treated, and the overall medical status of the patient. This level of transparency not only enhances clinical decision-making but also fosters improved communication between clinicians and patients, ultimately leading to more informed consent and shared decision-making. XAI is currently being utilised effectively across various medical specialties, such as oncology, cardiology, and critical care. In these areas, it plays a crucial role by improving risk stratification—helping healthcare professionals identify patients at higher risk for adverse outcomes—and enhancing decision support systems, which aid clinicians in making more informed choices tailored to individual patient needs. While the integration of XAI in dentistry, particularly in endodontics, is still largely aspirational, its application is increasingly feasible, which may hold the promise of transforming treatment planning, diagnosis, and patient management in dental practices, allowing for more precise interventions and improved patient outcomes. Oncology: XAI has facilitated the stratification of cancer patients by integrating multiple data points, such as tumour biology, treatment responses, and survival probabilities","author":[{"family":"Turky","given":"Mohammed"},{"family":"Dummer","given":"PMH"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/aej.70010","URL":"https://doi.org/10.1111/aej.70010","source":"openalex"},{"id":"oa:W7117630788","type":"article-journal","title":"Cognitive offloading and the reshaping of human thought: The subtle influence of Artificial Intelligence","abstract":"The integration of artificial intelligence (AI) into daily cognitive tasks has transformed traditional cognitive offloading—the delegation of mental processes to external tools—into a dynamic partnership with intelligent systems. This article examines the dual role of AI-driven cognitive offloading, exploring its potential to enhance efficiency and creativity while posing risks to memory consolidation, critical thinking, and intellectual autonomy. Grounded in extended mind theory (Clark & Chalmers, 1998) and empirical studies like the Google Effect (Sparrow et al., 2011), the analysis reveals how AI's generative capabilities (e.g., ChatGPT, Midjourney) shift offloading from passive storage to delegated thinking, where users adopt AI outputs with minimal scrutiny, fostering automation bias (Logg et al., 2019). In educational contexts, AI tools risk undermining deep learning by reducing retrieval practice and encouraging superficial engagement, as illustrated by a case study of Hatt University, where AI-assisted essays distorted grading systems and eroded student critical analysis. Similarly, in healthcare and finance, overreliance on AI recommendations may compromise professional judgment. To mitigate these risks, the article proposes strategies such as metacognitive training, explainable AI design (Sundar, 2020), and curriculum reforms prioritizing active engagement with AI outputs. Ethical and policy interventions are urged to address epistemic opacity, intellectual property, and the cultural redefinition of authorship. The study underscores the need for balanced AI integration—one that harnesses its benefits while safeguarding human cognitive autonomy.","author":[{"family":"Hooper","given":"Vincent"}],"issued":{"date-parts":[[2025]]},"DOI":"10.31207/colloquia.v12i1.185","URL":"https://doi.org/10.31207/colloquia.v12i1.185","source":"openalex"},{"id":"oa:W4406580523","type":"article-journal","title":"Evaluation of Medical Diagnosis Capabilities of Three Artificial Intelligence Models – ChatGPT-3.5, Google Gemini, Microsoft Copilot: Sustainable Development Goals (SDGs)","abstract":"Objectives: This study aims to assess and compare the diagnostic accuracy of three artificial intelligence (AI) models—ChatGPT-3.5, Microsoft Copilot, and Google Gemini—through their performance on clinical vignettes. Theoretical Framework: Building on prior research into the application of AI in healthcare, particularly in diagnostic support, this study examines the potential of AI models to aid clinicians by providing accurate medical diagnoses, thus supporting decision-making in clinical contexts. Methodology: A meta-analysis was conducted, followed by a comparative analysis using 34 clinical vignettes from Texas Tech University Health Sciences Center. Each AI model’s responses were evaluated for accuracy in diagnosing medical cases, and statistical significance was tested using the chi-square test. Results and Discussion: ChatGPT-3.5 achieved the highest diagnostic accuracy (70.59%), outperforming Google Gemini (61.76%) and Microsoft Copilot (35.29%). ChatGPT-3.5 provided concise answers, while Google Gemini and Microsoft Copilot included disclaimers and additional recommendations. Chi-square analysis confirmed significant differences in performance, highlighting variations in diagnostic capabilities across models. Research Implications: These findings underscore the importance of model selection when integrating AI into clinical workflows. AI models show promise in diagnostics but vary in approach and accuracy, warranting further refinement. Originality/Value: This study is among the first to compare the diagnostic accuracy of ChatGPT-3.5, Google Gemini, and Microsoft Copilot, contributing valuable insights into AI’s application in healthcare diagnostics and supporting evidence for its potential role in enhancing patient care.","author":[{"family":"Eneva","given":"Yordanka"},{"family":"Doğan","given":"Bora"}],"issued":{"date-parts":[[2025]]},"DOI":"10.47172/2965-730x.sdgsreview.v5.n02.pe03545","URL":"https://doi.org/10.47172/2965-730x.sdgsreview.v5.n02.pe03545","source":"openalex"},{"id":"oa:W4409247146","type":"article-journal","title":"Artificial Intelligence and Venous Thromboembolism: A Narrative Review of Applications, Benefits, and Limitations","abstract":"Background: Venous thromboembolism (VTE), including deep vein thrombosis and pulmonary embolism, remains a leading cause of cardiovascular morbidity and mortality. Artificial intelligence (AI) holds promise for potential improvement of risk stratification, diagnosis, and management of VTE. Summary: This narrative review explores the applications, benefits, and limitations of AI in VTE management. AI models were shown to outperform conventional methods in identifying high-risk candidates for VTE prophylaxis treatments in several postsurgical settings. It has also been demonstrated to be efficient in the early detection of VTE events, particularly through point-of-care AI-guided sonography and computer tomography image processing. Data biases, model transparency, and the need for regulatory frameworks remain significant limitations in the full integration of AI into clinical practice. Key Messages: AI has the potential to improve VTE care by enhancing risk stratification and diagnosis. The integration of AI-driven models into clinical workflows has the potential to reduce costs, streamline diagnostic processes, and ensure effective management of VTE. Safe and effective integration of AI into VTE care requires addressing its limitations, such as interpretability, privacy, and algorithmic bias. .","author":[{"family":"Mudrik","given":"Aya"},{"family":"Efros","given":"Orly"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1159/000545760","URL":"https://doi.org/10.1159/000545760","source":"openalex"},{"id":"oa:W4406074275","type":"article-journal","title":"AI-driven multi-omics integration for multi-scale predictive modeling of genotype-environment-phenotype relationships","abstract":"Despite the wealth of single-cell multi-omics data, it remains challenging to predict the consequences of novel genetic and chemical perturbations in the human body. It requires knowledge of molecular interactions at all biological levels, encompassing disease models and humans. Current machine learning methods primarily establish statistical correlations between genotypes and phenotypes but struggle to identify physiologically significant causal factors, limiting their predictive power. Key challenges in predictive modeling include scarcity of labeled data, generalization across different domains, and disentangling causation from correlation. In light of recent advances in multi-omics data integration, we propose a new artificial intelligence (AI)-powered biology-inspired multi-scale modeling framework to tackle these issues. This framework will integrate multi-omics data across biological levels, organism hierarchies, and species to predict genotype-environment-phenotype relationships under various conditions. AI models inspired by biology may identify novel molecular targets, biomarkers, pharmaceutical agents, and personalized medicines for presently unmet medical needs.","author":[{"family":"Wu","given":"You"},{"family":"Xie","given":"Lei"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.csbj.2024.12.030","URL":"https://doi.org/10.1016/j.csbj.2024.12.030","source":"openalex"},{"id":"oa:W4417151913","type":"article-journal","title":"The Application of Artificial Intelligence (AI) in Regenerative Medicine: Current Insights and Challenges","abstract":"Artificial intelligence (AI) is rapidly emerging as a transformative tool capable of addressing critical challenges and improving outcomes in tissue engineering and regenerative medicine. This paper demonstrates how machine learning and data fusion predict stem cell activity and potency, improve cellular characterization, and optimize therapeutic design. It also highlights important uses of AI in tissue engineering and cell-based therapeutics. By enabling accurate, non-invasive, and quantitative examination of living cells, AI also advances microscopy and imaging, facilitating better decision-making and real-time monitoring. Using search criteria including artificial intelligence, machine learning, deep learning, regenerative medicine, stem cells, and tissue engineering, the review was carried out using PubMed, Scopus, Web of Science, and Google Scholar. A total of 71 articles were screened; 8 non-peer-reviewed sources, 5 conference abstracts, and 4 duplicates were excluded. The final dataset included 7 clinical studies, 6 preclinical investigations, 18 original research articles, and 23 review papers. AI techniques, datasets, performance indicators, and regeneration results were compiled in the extracted data. To summarize, AI speeds up the development of tissue engineering, minimizes trial-and-error experimentation, lowers research expenses, forecasts tissue interactions, and enhances scaffold and biomaterial design. Consequently, AI integration enhances stem cell-based treatments and regenerative approaches, underscoring the necessity of interdisciplinary cooperation and ongoing technical development.","author":[{"family":"Abuarqoub","given":"Duaa"},{"family":"Mutahar","given":"Mahdi"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/biomedinformatics5040069","URL":"https://doi.org/10.3390/biomedinformatics5040069","source":"openalex"},{"id":"oa:W4408592630","type":"article-journal","title":"On the ethical and moral dimensions of using artificial intelligence for evidence synthesis","abstract":"Artificial intelligence (AI) is increasingly being used in the field of medicine and healthcare. However, there are no articles specifically examining ethical and moral dimensions of AI use for evidence synthesis. This article attempts to fills this gap. In doing so, I deploy in written form, what in Bengali philosophy and culture, is the Adda (আড্ডা) approach, a form of oral exchange, which involves deep but conversational style discussion. Adda developed as a form of intellectual resistance against the cultural hegemony of British Imperialism and entails asking provocative question to encourage critical discourse.The raison d'être for using AI is that it would enhance efficiency in the conduct of evidence synthesis, thus leading to greater evidence uptake. I question whether assuming so without any empirical evidence is ethical. I then examine the challenges posed by the lack of moral agency of AI; the issue of bias and discrimination being amplified through AI driven evidence synthesis; ethical and moral dimensions of epistemic (knowledge-related) uncertainty on AI; impact of knowledge systems (training of future scientists, and epistemic conformity), and the need for looking at ethical and moral dimensions beyond technical evaluation of AI models. I then discuss ethical and moral responsibilities of government, multi-laterals, research institutions and funders in regulating and having an oversight role in development, validation, and conduct of evidence synthesis. I argue that industry self-regulation for responsible use of AI is unlikely to address ethical and moral concerns, and that there is a need to develop legal frameworks, ethics codes, and of bringing such work within the ambit of institutional ethics committees to enable appreciation of the complexities around use of AI for evidence synthesis, mitigate against moral hazards, and ensure that evidence synthesis leads to improvement of health of individuals, nations and societies.","author":[{"family":"Bhaumik","given":"Soumyadeep"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1371/journal.pgph.0004348","URL":"https://doi.org/10.1371/journal.pgph.0004348","source":"openalex"},{"id":"oa:W7118159180","type":"article-journal","title":"Artificial Intelligence in Oculoplastic Surgery: A Systematic Review","abstract":"Artificial Intelligence (AI) has become an integral component of modern ophthalmology, with oculoplastic surgery representing a rapidly evolving subspecialty that stands to benefit from advances in automation and deep learning.Despite promising innovations, a comprehensive understanding of AI's role in oculoplastic diagnosis and management remains limited.This systematic review, conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, aims to evaluate the current landscape, clinical performance, and translational potential of AI applications in oculoplastic diseases published between 2000 and 2025.A structured search of PubMed, Scopus, and Embase identified 25 peer-reviewed studies involving AI-driven image analysis, disease classification, surgical planning, and prognostic modelling across eyelid, lacrimal, orbital, and periocular disorders.Studies were assessed for model performance, clinical utility, and methodological rigor.The including studies demonstrated that AI algorithms achieved diagnostic accuracies exceeding 90% in detecting periocular malignancies, outperforming or complementing traditional clinician-based assessment.Machine learning models also facilitated surgical planning and postoperative outcome prediction, contributing to enhance clinical workflow efficiency and reduced inter-observer variability.Nevertheless, limitations related to dataset heterogeneity, small sample sizes, and limited external validation constrain generalizability.AI holds significant promise in advancing precision and efficiency in oculoplastic care.Future research should prioritize multicentric validation, explainable AI frameworks, and integration with robotic-assisted surgery to enable safe and ethical clinical translation, ultimately bridging the gap between technological innovation and patient-centered ophthalmic practice.","author":[{"family":"Yadav","given":"Niraj"},{"family":"Plastic","given":"Ophthalmic"}],"issued":{"date-parts":[[2026]]},"DOI":"10.52338/joed.2025.5236","URL":"https://doi.org/10.52338/joed.2025.5236","source":"openalex"},{"id":"oa:W4411969645","type":"article-journal","title":"Generative artificial intelligence and the risk of technodigital colonialism","abstract":"The use of Generative Artificial Intelligence has raised concerns related to plagiarism in scientific contexts. However, bad academic writing is far from being the main ethical challenge related to digital transformations in knowledge production. Additionally, science is not the only trust discourse affected, as journalism and law are deeply impacted in its social roles by the dissemination of artificially generated discourses. Power and knowledge are increasingly imbricated in digital society in a global context where colonial hierarchization, dehumanization and exploitation strategies are still in place. In response to the insufficiency of high-level moral principles before the ethical and Human Rights challenges brought by GenAI applications, this paper offers an alternative theoretical approach to digital ethics presented in the “decolonizing ethical thinking” section. The aim is to focus on the role that the new epistemic dynamics play to the risk of technodigital colonialism. Decoloniality readings should account for why the benefits and risks are not universally distributed and therefore may help ethical responses be more attentive to the connections between knowledge and power.","author":[{"family":"Cambraia","given":"Leonardo"},{"family":"Pyrrho","given":"Monique"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fpos.2025.1628139","URL":"https://doi.org/10.3389/fpos.2025.1628139","source":"openalex"},{"id":"oa:W4409067430","type":"article-journal","title":"Artificial intelligence in neurology, ethics, recent guideline, and law-an Indian perspective","abstract":"Artificial Intelligence in Neurology, Ethics, Recent Guideline and Law- An Indian PerspectiveIntroductionArtificial intelligence (AI), a boon, kept healthcare professionals safe during the COVID-19 era in the form of “Tommy,” the robot nurse in Italy or “Mitra” in India. But continuous involvement of AI in healthcare also brings various challenges, i.e., quality and ownership of the data, belief issues, and ethical challenges. Therefore, in this article, we want to discuss the global policies regarding AI from different countries including India. We also present the summary of the new ethical guideline for AI applications in healthcare by the Indian Council of Medical Research (ICMR) and the Digital Personal Data Protection Act 2023 (DPDP Act),the first law for personal data protection in India, released recently in India. Current AI applications in Neurology in IndiaAI can be utilized in different parts of neurology from diagnosis of seizures, developmental anomalies, i.e., Down syndrome, cerebral palsy, etc., and neurodegenerative diseases (e.g., Alzheimer’s disease) to the rehabilitation of stroke patients1. Recently in India, Aster CMI hospital has developed a screening tool to diagnose carpal tunnel syndrome2. Scientists from Kyoto University, Japan, and IIT Roorkee, India, have created an AI-based model to predict the glioma grading from brain MRI which will help to treat the tumors3. A USA-based company named Intel and the University of California have partnered with 29 international institutes including Tata Memorial Hospital, Mumbai to create an AI model for early detection of brain tumor4. Medtronic India has partnered with Qure.ai to develop an AI-based tool for early detection and management of stroke patients5.Ethical Laws and Guidelines Around the WorldUSAAround 120 bills are being contemplated by US Congress regarding AI but significantly less number of bills are related to ethics in healthcare. The Health Insurance Portability and Accountability Act of 1996 (HIPAA) exists in USA for personal data security of the citizens. However HIPPA can not shield the citizen from the “Black box” problem of AI. American Medical Association’s “Augmented Intelligence in Medicine” (2018) and “Payment and Coverage of AI” (2019), “Blueprint for an AI Bill of Rights” (2022) by the White House, and “Artificial Intelligence Risk Management Framework 1.0 (AI RMF)” (2023) by National Institute of Standards and Technology (NIST), exist regarding AI6. But these policies can not protect the citizens completely. Food and Drug Administration (FDA) has several guiding principles regarding AI-enabled medical devices (e.g. DermaSensor and Paige Prostate to detect skin cancer and carcinoma prostate respectively), i.e., “Good Machine Learning Practice for Medical Device Development: Guiding Principles” (2021), “Transparency for Machine Learning-Enabled Medical Devices: Guiding Principles”(2021), “Transparency for Machine Learning-Enabled Medical Devices: Guiding Principles” (2024)7. But till date no such guideline exists for non-device AI i.e., AI associated with the clinical decision support system.GermanySimilar to HIPPA in the USA, the General Data Protection Regulation (GDPR) is present in the European Union (EU) for the personal data security of citizens. Similarly, as a member of the EU, Bundesdatenschutzgesetz (BDSG) is present in Germany. General Product Safety Regulation (GPSR) 2023/988 involves safety rules for products. The New Product Liability Directive includes compensation for AI software which will be implemented in December 2026. Though the terminology “AI” is included as a reference in the German Works Constitution Act 2021, Germany does not have a separate AI regulation and will probably follow the EU AI Act. Among all the regulations, the EU AI Act (2024) is the most significant legislation as it is the first legal framework for AI. This act divides AI into several risk categories, ie., unacceptable (e.g. Biometric identific","author":[{"family":"Kundu","given":"Tithishri"},{"family":"Bardhan","given":"Mainak"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3389/fneur.2025.1515041","URL":"https://doi.org/10.3389/fneur.2025.1515041","source":"openalex"},{"id":"oa:W4411756025","type":"article-journal","title":"Artificial Intelligence in Obstetrics and Gynaecology: Advancing Precision and Personalised Care","abstract":"Artificial intelligence (AI) is rapidly transforming the landscape of obstetrics and gynaecology, offering unprecedented capabilities in diagnostics, monitoring, and personalised treatment. This review highlights the integration of AI in various domains, including obstetric imaging, fetal monitoring, gynaecologic oncology, fertility treatment, and robotic surgery. AI-powered tools are shown to enhance precision by improving diagnostic accuracy, reducing human error, and supporting clinical decision-making. Importantly, the article explores the global implications of AI adoption, including applications in low-resource settings, and emphasises the need for ethical considerations, data inclusiveness, and clinician trust. Overall, this comprehensive review demonstrates how AI is enabling more individualised and effective care in women's health.","author":[{"family":"Aftab","given":"Nida"}],"issued":{"date-parts":[[2025]]},"DOI":"10.7759/cureus.86929","URL":"https://doi.org/10.7759/cureus.86929","source":"openalex"},{"id":"oa:W4407608219","type":"article-journal","title":"The Role of AI in Reshaping Medical Education: Opportunities and Challenges","abstract":"Artificial intelligence (AI) is redefining medical education, bringing new dimensions of personalized learning, enhanced visualization and simulation-based clinical training to the forefront. Additionally, AI-powered simulations offer realistic, immersive training opportunities, preparing students for complex clinical situations and fostering interprofessional collaboration skills essential for modern healthcare. However, the integration of AI into medical education presents challenges, particularly around ethical considerations, skill atrophy due to overreliance and the exacerbation of the digital divide among educational institutions. Addressing these challenges demands a balanced approach that includes blended learning models, digital literacy and faculty development to ensure AI serves as a supplement to, rather than a replacement for, core medical competencies. As medical education evolves alongside AI, institutions must prioritize strategies that preserve human-centred skills while advancing technological innovation to prepare future healthcare professionals for an AI-enhanced landscape.","author":[{"family":"Ali","given":"Majid"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/tct.70040","URL":"https://doi.org/10.1111/tct.70040","source":"openalex"},{"id":"oa:W4414685754","type":"article-journal","title":"Toward a new era of healthcare services: the role of artificial intelligence in shaping tomorrow’s landscape","abstract":"The rapid advancement of artificial intelligence (AI) has revolutionized various sectors, including healthcare, by enhancing diagnostic precision, operational efficiency and personalized care. Despite its vast potential, integrating AI into healthcare remains a complex challenge. Addressing this requires a comprehensive understanding of the research landscape to guide future innovations. This study analyzes the intellectual structure and emerging trends in AI applications within healthcare through co-citation and co-word analysis. A total of 24,973 publications from the Scopus database, spanning 2000–2024, were examined to observe growth patterns, with 4,043 journal articles analyzed in depth. The findings identify five thematic clusters: technology and implementation, health and disease, treatment and care, algorithms and techniques, and social and ethical aspects. These clusters reflect the evolving focus of AI research in healthcare. The insights highlight AI’s growing impact on healthcare outcomes and its potential to address patient needs, while drawing attention to ethical and operational challenges. This study provides a robust framework for understanding recent trends in AI applications in healthcare, offering strategic guidance for academics, practitioners and policymakers to foster equitable AI integration into global healthcare systems.","author":[{"family":"Hamsal","given":"Mohammad"},{"family":"Binsar","given":"Faisal"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1080/23311975.2025.2566441","URL":"https://doi.org/10.1080/23311975.2025.2566441","source":"openalex"},{"id":"oa:W4413422755","type":"article-journal","title":"Performance of Advanced Artificial Intelligence Models in Pulp Therapy for Immature Permanent Teeth: A Comparison of ChatGPT-4 Omni, DeepSeek, and Gemini Advanced in Accuracy, Completeness, Response Time, and Readability","abstract":"INTRODUCTION: This study aims to evaluate and compare the performance of three advanced chatbots-ChatGPT-4 Omni (ChatGPT-4o), DeepSeek, and Gemini Advanced-on answering questions related to pulp therapies for immature permanent teeth. The primary outcomes assessed were accuracy, completeness, and readability, while secondary outcomes focused on response time and potential correlations between these parameters. METHODS: A total of 21 questions were developed based on clinical resources provided by the American Association of Endodontists, including position statements, clinical considerations, and treatment options guides, and assessed by three experienced pediatric dentists and three endodontists. Accuracy and completeness scores, as well as response times, were recorded, and readability was evaluated using Flesch Kincaid Reading Ease Score, Flesch Kincaid Grade Level, Gunning Fog Score, SMOG Index, and Coleman Liau Index. RESULTS: Results revealed significant differences in accuracy (P < .05) and completeness (P < .05) scores among the chatbots, with ChatGPT-4o and DeepSeek outperforming Gemini Advanced in both categories. Significant differences in response times were also observed, with Gemini Advanced providing the quickest responses (P < .001). Additionally, correlations were found between accuracy and completeness scores (ρ: .719, P < .001), while response time showed a positive correlation with completeness (ρ: .144, P < .05). No significant correlation was found between accuracy and readability (P > .05). CONCLUSIONS: ChatGPT-4o and DeepSeek demonstrated superior performance in terms of accuracy and completeness when compared to Gemini Advanced. Regarding readability, DeepSeek scored the highest, while ChatGPT-4o showed the lowest. These findings highlight the importance of considering both the quality and readability of artificial intelligence-driven responses, in addition to response time, in clinical applications.","author":[{"family":"Sezer","given":"Berkant"},{"family":"Aydoğdu","given":"Tuğba"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1016/j.joen.2025.08.011","URL":"https://doi.org/10.1016/j.joen.2025.08.011","source":"openalex"},{"id":"oa:W4414036954","type":"article-journal","title":"Smart technology framework for medical waste optimization by integrating wireless tracking with artificial intelligence classification","abstract":"In this study, we developed and validated an integrated Radio Frequency Identification-Artificial Intelligence (RFID-AI) framework to optimize medical waste management in resource-constrained healthcare settings. The system combines: (1) UHF RFID-enabled smart bins with real-time mass and environmental monitoring, (2) a fine-tuned ResNet-50 computer vision model achieving 93.1% (±2.1%) waste classification accuracy, and (3) genetic algorithm-based route optimization reducing collection distances by 18%. Implemented across four Jordanian hospitals (92-203 beds) for six months, the system demonstrated significant improvements across key metrics: Operational efficiency (30.1% reduction in collection time, P < 0.01; 81% fewer hazardous mixing incidents), staff safety (40.2% reduction in sharps injuries through AI monitoring), environmental impact (15.2% lower particulate emissions via optimized incineration scheduling), and cost-effectiveness (23.2% operational cost reduction with 14-month ROI). The framework's modular design successfully addressed institution-specific challenges, including 78.3% storage overcapacity at Princess Basma Hospital and 39.1% improper sharps disposal at Ibn Al-Nafis Hospital, while maintaining 90.3% compliance with WHO 2022 guidelines. Technical innovations included moisture-resistant RFID tags, maintaining 98.3% read accuracy in high-humidity environments and Arabic-language AI interfaces that reduced training time by 42%. These results provide empirical evidence for the viability of smart waste systems in LMICs, offering a replicable model that balances technological sophistication with practical implementation constraints. We established a new benchmark for intelligent medical waste management systems under resource limitations.","author":[{"family":"Bdour","given":"Ahmed"},{"family":"Kharabsheh","given":"Raha"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3934/environsci.2025035","URL":"https://doi.org/10.3934/environsci.2025035","source":"openalex"},{"id":"oa:W4410870469","type":"article-journal","title":"Leashes, not guardrails: A management‐based approach to artificial intelligence risk regulation","abstract":"Calls to regulate artificial intelligence (AI) have sought to establish guardrails to protect the public against AI going awry. Although physical guardrails can lower risks on roadways by serving as fixed, immovable protective barriers, the regulatory equivalent in the digital age of AI is unrealistic and even unwise. AI is too heterogeneous and dynamic to circumscribe fixed paths along which it must operate-and, in any event, the benefits of the technology proceeding along novel pathways would be limited if rigid, prescriptive regulatory barriers were imposed. But this does not mean that AI should be left unregulated, as the harms from irresponsible and ill-managed development and use of AI can be serious. Instead of \"guardrails,\" though, policymakers should impose \"leashes.\" Regulatory leashes imposed on digital technologies are flexible and adaptable-just as physical leashes used when walking a dog through a neighborhood allow for a range of movement and exploration. But just as a physical leash only protects others when a human retains a firm grip on the handle, the kind of leashes that should be deployed for AI will also demand human oversight. In the regulatory context, a flexible regulatory strategy known in other contexts as management-based regulation will be an appropriate model for AI risk governance. In this article, we explain why regulating AI by management-based regulation-a leash approach-will work better than a prescriptive or guardrail regulatory approach. We discuss how some early regulatory efforts include management-based elements. We also elucidate some of the questions that lie ahead in implementing a management-based approach to AI risk regulation. Our aim is to facilitate future research and decision-making that can improve the efficacy of AI regulation by leashes, not guardrails.","author":[{"family":"Coglianese","given":"Cary"},{"family":"Crum","given":"Colton"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1111/risa.70020","URL":"https://doi.org/10.1111/risa.70020","source":"openalex"},{"id":"oa:W4413828726","type":"article-journal","title":"Assessment of university students’ earthquake coping strategies using artificial intelligence methods","abstract":"Earthquakes are one of the most destructive natural disasters that pose a serious threat to human life and infrastructure worldwide. The aim of this study is to evaluate the coping strategies of adult individuals in Turkey regarding earthquake stress using artificial intelligence-based methods. The data was collected from 858 university students living in Turkey during January, February, and March 2024. A dataset was created using the 'Coping Scale for Earthquake Stress.' Prediction models were established using artificial intelligence algorithms such as Logistic Regression (LR), Bagging, and Random Forest (RF) based on information from 24 variables. The cross-validation method was applied during model training. The Logistic Regression algorithm achieved the highest accuracy rate of 98.60%, while the Bagging algorithm demonstrated the lowest performance with an accuracy rate of 79.95%. The Random Forest algorithm showed moderate performance with an accuracy rate of 85.89%. The findings provide important insights into the coping strategies of the community regarding earthquake stress. This study is expected to contribute significantly to areas such as disaster management, psychology, public health, and community resilience.","author":[{"family":"Sulak","given":"Süleyman"},{"family":"Köklü","given":"Niğmet"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-17555-4","URL":"https://doi.org/10.1038/s41598-025-17555-4","source":"openalex"},{"id":"oa:W4413796682","type":"article-journal","title":"Artificial Intelligence in Education (AIED): Towards More Effective Regulation","abstract":"Abstract This paper critically assesses the effectiveness of the EU AI Act in regulating artificial intelligence in higher education (AIED), with a focus on how it interacts with existing education regulation. It examines the growing use of high-risk AI systems – such as those used in admissions, assessment, academic progression, and exam proctoring – and identifies key regulatory frictions that arise when AI regulation and education regulation pursue overlapping but potentially conflicting aims. Central to this analysis is the concept of human oversight: while the AI Act frames oversight as a safeguard for accountability and fundamental rights, education regulation emphasises the professional autonomy of teachers and their role in maintaining pedagogical integrity. Yet, the regulatory role of teachers in AI-mediated environments remains unclear. Applying Mousmouti’s effectiveness test, the paper evaluates the AI Act along four dimensions – purpose, coherence, results, and structural integration with the broader legal framework – and argues that legal effectiveness in this context requires a more precise alignment between AI and education regulation.","author":[{"family":"Colonna","given":"Liane"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1017/err.2025.10039","URL":"https://doi.org/10.1017/err.2025.10039","source":"openalex"},{"id":"oa:W4406549764","type":"article-journal","title":"Leveraging Artificial Intelligence in Business Intelligence Systems for Predictive Analytics","abstract":"Artificial Intelligence (AI) and Business Intelligence (BI) are rapidly emerging as the next big things for organizations to analyze data and gain insights. As this article will go on to examine, the concept of using AI for BI is one that has significant implications about the possible integration of AI into various Business Intelligence systems examined in this article will focus on the application of AI for BI in the use of predicting analytics. When integrating Machine learning, natural language processing, and intelligent automation, these AI-Advanced BI systems assist organizations to go beyond data reporting or simple descriptive analytics and gain an insight to use BI systems to discover and pre-empt issues, besides noticing them using proactive decision making. In discussing the elements of AI-embedded BI systems, this article analyzes how organizations across industries use real-time intelligence and predictive models as indispensable resources for the generation of competitive edge. Some of the advantages highlighted includes improved accuracy for predictions, efficiency of cost on data handling, scalability on large data and the shorter delays on decision making. However, alongside these benefits, the article also addresses key challenges, such as data privacy concerns, biases in AI algorithms, and the complexities of integrating AI into legacy BI platforms. These limitations are critical considerations for organizations seeking to implement AI-driven BI systems effectively. Furthermore, this work discusses the issues relating to the implementation of AI for BI, for example, the integration of AI into existing BI platforms, data quality issues, ethical issues, and the skill gaps in specialized AI talents. The article also discusses new developments in AI integration to BI systems including the growing incorporation of deep learning techniques, automation of decision making and BI democratization for small businesses. They suggest that BI must evolve new business strategies to be effective and meet the information demands needed for corporate competitiveness in today’s data-centric economy. The convergence of advanced analytics and operational decision making makes AI driven BI system the tool with tremendous potential to become the lingua franca of business strategy and growth.","author":[{"family":"Ebule","given":"Amejuma"}],"issued":{"date-parts":[[2025]]},"DOI":"10.18535/ijsrm/v13i01.ec02","URL":"https://doi.org/10.18535/ijsrm/v13i01.ec02","source":"openalex"},{"id":"oa:W4411823317","type":"article-journal","title":"Integrating artificial intelligence into Ayurveda: Pathways, potentials, and challenges","abstract":"Introduction Ayurveda, the ancient system of holistic medicine, is grounded in experiential knowledge, personalized treatment approaches, and a comprehensive view of health and disease. Its time-tested principles continue to serve as a foundation for preventive, promotive, and curative healthcare practices. In recent years, the rapid advancement of artificial intelligence (AI) has opened new avenues for all sectors of healthcare including Traditional Medicine(TM) and Ayurveda. The World Health Organization (WHO), in partnership with the International Telecommunication Union (ITU), launched the Focus Group on AI for Health (FG-AI4H) in 2018, which later led to the formation of the Global Initiative on AI for Health (GI-AI4H) in July 2023, along with ITU and WIPO, as a long-term structure to advance AI in healthcare. WHO has since published several key documents: Ethics and Governance of Artificial Intelligence for Health: WHO Guidance Executive Summary (28 June 2021), Global Strategy on Digital Health 2020–2025 (18 August 2021), Generating Evidence for Artificial Intelligence-Based Medical Devices: A Framework for Training (17 November 2021), Regulatory Considerations on Artificial Intelligence for Health (19 October 2023), and Ethics and Governance of Artificial Intelligence for Health: Guidance on Large Multi-Modal Models (25 March 2025), highlighting its ongoing commitment to ethical, strategic, and regulatory aspects of AI in health.[1] The Ministry of Ayush is actively advancing AI integration in Indian Systems of Medicine through curriculum development, AI-powered e-learning, Ayush Grid applications, and global expert collaborations—highlighted by the Global Experts Convene on AI in Traditional Medicine held at AIIA, New Delhi (12–13 Sept 2024).[2] AI technologies such as machine learning, natural language processing (NLP), and data integration tools offer promising applications in clinical decision-making, educational innovations, pharmacovigilance, and the digitization of classical Ayurvedic knowledge systems.[3,4] This appraisal aims to explore the current and potential applications of AI in Ayurveda, identify key opportunities and challenges in this integrative journey, and propose strategic directions for responsible and epistemologically aligned adoption of AI within the Ayurvedic framework. Scope and Applications of Artificial Intelligence in Ayurveda Artificial intelligence in traditional medicine: Real-world applications and opportunities Globally, AI is being increasingly integrated into traditional medicine systems to support evidence generation, drug discovery, and safety monitoring. For example, in China, AI has been effectively applied in traditional Chinese medicine (TCM) through the development of machine learning models to classify syndromes, predict therapeutic outcomes, and mine historical clinical records for drug discovery.[5] A notable instance includes the use of AI-driven knowledge graphs and text-mining algorithms to extract pharmacological insights from TCM classical texts, supporting reverse pharmacology and identifying novel therapeutic leads.[5,6] Adverse drug reaction and pharmacovigilance AI plays a crucial role in enhancing drug safety by analyzing large datasets to detect patterns of adverse drug reactions (ADRs) that may not surface during clinical trials. By predicting potential risks early, AI supports safer, faster, and more informed decision-making in drug development and post-marketing surveillance. Researchers opined that AI-based tools for ADR detection from Ayurvedic hospital databases and public health forums, showcasing AI’s role in proactive safety monitoring.[7] These examples illustrate how AI applications such as text mining, clinical data analytics, and predictive modeling can significantly advance traditional medicine by enhancing therapeutic safety, optimizing drug development, and bridging classical knowledge with contemporary biomedical research. Digitization and semant","author":[{"family":"Acharya","given":"Rabinarayan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.4103/jdras.jdras_176_25","URL":"https://doi.org/10.4103/jdras.jdras_176_25","source":"openalex"},{"id":"oa:W4413913331","type":"article-journal","title":"Deep computer vision with artificial intelligence based sign language recognition to assist hearing and speech-impaired individuals","abstract":"Sign language (SL) is a non-verbal language applied by deaf and hard-of-hearing individuals for daily communication between them. Studies in SL recognition (SLR) have recently become essential developments. The current successes present the base for upcoming applications to assist the combination of deaf and hard-of-hearing people. SLR could help break down the obstacles for SL users in the community. In general, glove-based and vision-based techniques are the dual major types measured for SLR methods. Several investigators presented various techniques with significant development by deep learning (DL) models in computer vision (CV) and became performed to SLR. This study presents a novel Harris Hawk Optimization-Based Deep Learning Model for Sign Language Recognition (HHODLM-SLR) technique. The HHODLM-SLR technique mainly concentrates on the advanced automatic detection and classification of SL for hearing and speech-impaired individuals. Initially, the image pre-processing stage applies bilateral filtering (BF) to eliminate noise in an input image dataset. Furthermore, the ResNet-152 model is employed for the feature extraction process. The bidirectional long short-term memory (Bi-LSTM) model is used for SLR. Finally, the Harris hawk optimization (HHO) approach optimally adjusts the Bi-LSTM approach's hyperparameter values, resulting in more excellent classification performance. The efficiency of the HHODLM-SLR methodology is validated under the SL dataset. The experimental analysis of the HHODLM-SLR methodology portrayed a superior accuracy value of 98.95% over existing techniques.","author":[{"family":"Almjally","given":"Abrar"},{"family":"Almukadi","given":"Wafa"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1038/s41598-025-09106-8","URL":"https://doi.org/10.1038/s41598-025-09106-8","source":"openalex"},{"id":"oa:W4409646233","type":"article-journal","title":"Research on the Transformation of Enterprise Marketing Strategy Driven by Artificial Intelligence","abstract":"This study explores the transformation of enterprise marketing strategies driven by artificial intelligence (AI) through two case studies: Beijing Nanshe Technology Co., Ltd. and Juran Design Home. Both companies, operating in different sectors, have embraced AI technologies to optimize their marketing practices, enhance customer engagement, and improve operational efficiency. Nanshe Tech, a leader in smart office devices, integrated AI to develop predictive models for customer targeting and lifecycle management, leading to improved conversion rates and a shift towards a subscription-based business model. Juran Design Home, a digital platform for the home furnishing industry, utilized AI-powered design assistants and VR/AR features to enhance the customer experience, resulting in higher customer satisfaction and increased sales. The study highlights the significant impact of AI on marketing practices, demonstrating how AI-driven tools can revolutionize customer segmentation, resource optimization, and marketing outcomes. The findings provide valuable insights for businesses considering AI adoption in their marketing strategies, emphasizing the need for a data-driven approach and cross-functional collaboration to successfully integrate AI technologies.","author":[{"family":"Xie","given":"Lijuan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.55014/pij.v8i2.802","URL":"https://doi.org/10.55014/pij.v8i2.802","source":"openalex"},{"id":"oa:W4413907480","type":"article-journal","title":"Artificial Intelligence in the Diagnosis of Pediatric Rare Diseases: From Real-World Data Toward a Personalized Medicine Approach","abstract":"Background: Artificial intelligence (AI) is increasingly applied in the diagnosis of pediatric rare diseases, enhancing the speed, accuracy, and accessibility of genetic interpretation. These advances support the ongoing shift toward personalized medicine in clinical genetics. Objective: This review examines current applications of AI in pediatric rare disease diagnostics, with a particular focus on real-world data integration and implications for individualized care. Methods: A narrative review was conducted covering AI tools for variant prioritization, phenotype–genotype correlations, large language models (LLMs), and ethical considerations. The literature was identified through PubMed, Scopus, and Web of Science up to July 2025, with priority given to studies published in the last seven years. Results: AI platforms provide support for genomic interpretation, particularly within structured diagnostic workflows. Tools integrating Human Phenotype Ontology (HPO)-based inputs and LLMs facilitate phenotype matching and enable reverse phenotyping. The use of real-world data enhances the applicability of AI in complex and heterogeneous clinical scenarios. However, major challenges persist, including data standardization, model interpretability, workflow integration, and algorithmic bias. Conclusions: AI has the potential to advance earlier and more personalized diagnostics for children with rare diseases. Achieving this requires multidisciplinary collaboration and careful attention to clinical, technical, and ethical considerations.","author":[{"family":"Ilić","given":"Nikola"},{"family":"Sarajlija","given":"Adrijan"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/jpm15090407","URL":"https://doi.org/10.3390/jpm15090407","source":"openalex"},{"id":"oa:W4407640086","type":"article-journal","title":"Correlation Between Artificial Intelligence Literacy and Artificial Intelligence Anxiety in Audiology Students","abstract":"ABS TRACT Objective: Artificial intelligence (AI) is becoming increasingly important in health sciences.Therefore, it is essential to assess health science students' knowledge, literacy, awareness, attitudes, and anxiety regarding AI.Although AI literacy and AI anxiety have been studied separately, the relationship between them has not been adequately explored.This study aims to evaluate the AI literacy and AI anxiety levels of undergraduate audiology students and examine the relationship between these 2 variables.Material and Methods: A correlational study was conducted with 231 undergraduate audiology students.Data were collected using the Artificial Intelligence Literacy Scale (AILS) and Artificial Intelligence Anxiety Scale (AIAS).Results: The mean AILS score was 39.49±5.50, and the mean AIAS score was 46.13±12.31.Significant correlations were observed among various subscales (p<0.05) and between the total and subscale scores (p<0.05) of AILS and AIAS in audiology students.However, no significant relationship was found between the total scores of AIAS and AILS.Conclusion: The findings suggest that higher scores in the awareness and usage subscales of AI literacy are associated with lower AI anxiety.Therefore, integrating AI training into the audiology curriculum may enhance AI literacy and help reduce AI-related anxiety among students.","author":[{"family":"Kuntman","given":"Berna"},{"family":"Polat","given":"Anı"}],"issued":{"date-parts":[[2026]]},"DOI":"10.24179/kbbbbc.2026-117553","URL":"https://doi.org/10.24179/kbbbbc.2026-117553","source":"openalex"},{"id":"oa:W7133191197","type":"article-journal","title":"Artificial Intelligence-Enhanced Flexible Sensors for Human Motion and Posture Sensing","abstract":"In the era of Industry 4.0, artificial intelligence technology is experiencing rapid development, and the integration of artificial intelligence (AI) with flexible sensors has emerged as a transformative approach for human motion and posture sensing. This paper explores the advancements in AI-enhanced flexible sensors, focusing on the application of flexible sensors on various parts of the human body. Flexible sensors, due to their conformability and sensitivity, are ideal for capturing the dynamic and subtle movements of the human body. AI algorithms, particularly machine learning and deep learning techniques are employed to process the complex data streams from these sensors, enabling the accurate recognition and prediction of various human postures and motions. The combination of these technologies overcomes the limitations of traditional sensing systems, offering higher precision, adaptability, and real-time feedback. It can be applied to healthcare for rehabilitation monitoring, sports for performance enhancement, and human-computer interaction for intuitive control. This review also discusses the challenges such as sensor reliability, data privacy, and power management. The future outlook emphasizes more sophisticated AI models and deeper technology integration, promising a seamless integration into everyday life for enhanced human-machine interaction and health monitoring.","author":[{"family":"Jiang","given":"Yiru"},{"family":"He","given":"Tianyiyi"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/s26051562","URL":"https://doi.org/10.3390/s26051562","source":"openalex"},{"id":"oa:W4415882333","type":"article-journal","title":"Artificial intelligence and the impact of the EU AI Act in business organizations","abstract":"Abstract Artificial intelligence (AI) is transforming industries worldwide, and the e‐commerce sector is at the forefront of leveraging its capabilities to drive innovation and efficiency. The paper explores the integration of artificial intelligence in e‐commerce, focusing on the ethical and regulatory implications introduced by the EU AI Act. This legislative framework aims to ensure the responsible deployment of AI by classifying AI systems into risk categories and imposing compliance requirements. It also underscores both the opportunities and challenges that AI presents to businesses, particularly in enhancing consumer experiences through automation and data‐driven decision‐making processes. The paper provides a comprehensive review of the AI landscape in Europe, analyzing the impact of the EU AI Act, particularly on small and medium‐sized enterprises and startups. Through a mixed‐methods approach, the study investigates how regulatory compliance may influence business innovation, market competitiveness, and consumer trust. The recommendations proposed aim to develop a trustworthy AI ecosystem that could stimulate long‐term growth and enhance the global positioning of small European businesses.","author":[{"family":"Selgas-Cors","given":"Marc"},{"family":"Thiébaut","given":"Renata"}],"issued":{"date-parts":[[2025]]},"DOI":"10.1002/aaai.70039","URL":"https://doi.org/10.1002/aaai.70039","source":"openalex"},{"id":"oa:W4408123538","type":"article-journal","title":"Patho-Net: enhancing breast cancer classification using deep learning and explainable artificial intelligence","abstract":"Breast cancer is a disorder affecting women globally, and hence an early and precise classification is the best possible treatment to increase the survival rate. However, the breast cancer classification faced difficulties in scalability, fixed-size input images, and overfitting on limited datasets. To tackle these issues, this work proposes a Patho-Net model for breast cancer classification that overcomes the problems of scalability in color normalization, integrates the Gated Recurrent Unit (GRU) network with the U-Net architecture to process images without the need for resizing and computational efficiency, and addresses the overfitting problems. The proposed model collects and normalizes histopathology images using automated reference image selection with the Reinhard method for color standardization. Also, the Enhanced Adaptive Non-Local Means (EANLM) filtering is utilized for noise removal to preserve image features. These preprocessed images undergo semantic segmentation to isolate specific parts of an image, followed by feature extraction using an Improved Gray Level Co-occurrence Matrix (I-GLCM) to reveal fine patterns and textures in images. These features serve as input into the classification U-Net model integrated with GRU networks to improve the model performance. Finally, the classification result is expanded, and XAI is used for clear visual explanations of the model's predictions. The proposed Patho-Net model, which uses the 100X BreakHis dataset, achieves an accuracy of 98.90% in the classification of breast cancer.","author":[{"family":"Manojee","given":"Kalappanaickenpatty"}],"issued":{"date-parts":[[2025]]},"DOI":"10.62347/xkfn1793","URL":"https://doi.org/10.62347/xkfn1793","source":"openalex"},{"id":"oa:W7128777056","type":"article-journal","title":"Generative Artificial Intelligence In Health Informatics Education: A Comprehensive Bibliometric Assessment Of Cognitive Outcome Research (2019–2025)","abstract":"This study provides a comprehensive bibliometric assessment of generative artificial intelligence (GenAI) research in health informatics education, with particular emphasis on cognitive outcomes. A total of 264 PubMed-indexed publications (2019–2025) were analyzed using Bibliometrix (R) and VOSviewer to evaluate annual scientific output, core journals, authorship patterns, institutional and country productivity, conceptual trends, and collaboration networks. Results show an exceptional rise in publication volume beginning in 2023, coinciding with the widespread introduction of large language models such as ChatGPT. The intellectual structure of the field is dominated by themes related to artificial intelligence, machine learning, large language models, and digital health applications. Two major clusters of application were identified: clinical and patient-centered communication, and educational processes involving competence development and assessment. Ethical themes, including bias and transparency, emerged rapidly in 2024–2025. Research output is highly concentrated in the United States and China, whereas collaboration patterns remain fragmented with multiple small author clusters. Most studies relied on cross-sectional designs, with limited experimental or longitudinal evaluation of learning outcomes. The findings highlight a methodological gap between technical GenAI development and established educational or cognitive frameworks. The study recommends integrating cognitive theory, improving methodological rigor, and expanding interdisciplinary and international collaboration. This bibliometric mapping offers an evidence-oriented foundation for guiding future work on GenAI-enhanced teaching, learning, curriculum design, and evaluation in health informatics education.","author":[{"family":"Alhur","given":"Anas"},{"family":"Al-Kahtani","given":"Nouf"}],"issued":{"date-parts":[[2026]]},"DOI":"10.29284/r2man269","URL":"https://doi.org/10.29284/r2man269","source":"openalex"},{"id":"oa:W7129064965","type":"article-journal","title":"Artificial Intelligence in Child and Adolescent Mental Health: Prevention, Diagnosis, and Treatment in Hybrid Human–AI Care Models","abstract":"Mental health disorders among children and adolescents have become increasinglycommon and burdensome, with conditions such as anxiety, depression, suicidality,and trauma-related disorders contributing significantly to disability and death. Whiletimely identification and intervention are vital, progress is often limited by thescarcity of trained providers, ongoing stigma, and dependence on subjectiveevaluation methods. Against this backdrop, artificial intelligence (AI) is beingexplored to improve mental healthcare through enhanced early detection,monitoring, individualized interventions, and clinical decision support. Thisnarrative review synthesizes research and systematic reviews from 2015 to 2025,sourced from Google Scholar, Web of Science, PubMed Central, PsycINFO, ScienceDirect, and EBSCO. Articles included focused on AI applications in children andadolescents’ mental health, highlighting advances in machine learning, naturallanguage processing, multimodal data integration, and digital cognitive-behavioraltherapy. Evidence suggests that AI can analyze behavioral, physiological, andlinguistic data to predict mental health risks, detect emerging symptoms, and deliverpersonalized interventions within a hybrid human–AI care model, where AIcomplements clinician expertise to improve access, engagement, and treatmentoutcomes. However, challenges persist, including algorithmic bias, limited modelinterpretability, data quality, privacy concerns, and integration into clinicalworkflows. Ethical and practical governance are essential to ensure that AI supports,rather than replaces, human-centered care. Future priorities include expandingresearch on underrepresented populations and conditions, developing explainableand equitable models, validating tools in real-world settings, and building large,FAIR-compliant datasets. Responsible, human-centered integration of AI has thepotential to improve early intervention, personalize treatment, and enhance equitableaccess to mental healthcare for young people globally.","author":[{"family":"Ui","given":"Nnubia"},{"family":"Ej","given":"Nwauzoije"}],"issued":{"date-parts":[[2026]]},"DOI":"10.66043/jfsr.v4i2.148","URL":"https://doi.org/10.66043/jfsr.v4i2.148","source":"openalex"},{"id":"oa:W7142248183","type":"article-journal","title":"Assessing complexity of educational texts of Russian as a foreign language: Prospects and challenges of using artificial intelligence","abstract":"The growing interest in Russian education, culture, and science results in the pressing demand for tools to select educational texts for Russian as a foreign language. The study is aimed at working out the algorithm and instruments for assessing the lexical complexity of text in Russian as a foreign language on CEFR with the help of LLM. The study is based on the material of a training sample, including standardized lexical minima in Russian as a foreign language and 232 texts ranked in difficulty by experts, and a test sample with 14 texts for listening in Russian as a foreign language. The methods of computational linguistics (Python script process_word_lists, LLM), expert assessment and metrics for statistical evaluation of the quality of classification models were used in the work. The study describes the successfully used large language models to assess the complexity of Russian-language texts on the RuLingva platform. The results of the study include the created linguistic profiles and the identified abilities of the large GLM 4.6 and Grok 4 fast language models to assess the complexity of educational texts in Russian as a foreign language (A1-C1). The proposed algorithm ranks texts by complexity with a high degree of accuracy, develops test tasks and selects texts for textbooks on Russian as a foreign language. The results obtained can be used by teachers in Russian as a foreign language, testologists, and linguists for preparing teaching materials, glossaries, and test assignments. The prospect of the work is to improve the developed algorithm by expanding the corpus and applying classification models for texts of different genres.","author":[{"family":"Solnyshkina","given":"Marina"},{"family":"Andreeva","given":"Mariia"}],"issued":{"date-parts":[[2026]]},"DOI":"10.22363/2618-8163-2026-24-1-120-137","URL":"https://doi.org/10.22363/2618-8163-2026-24-1-120-137","source":"openalex"},{"id":"oa:W7134903860","type":"article-journal","title":"Overview of allergic disease: Anaphylaxis – WAO White Book on Allergy 2026 – 2.11","abstract":"Anaphylaxis is a rapid-onset, potentially fatal systemic hypersensitivity reaction marked by airway, breathing, or circulatory compromise, which may occur even without skin symptoms. It is driven primarily by mast-cell activation-usually through IgE-mediated pathways-leading to the release of mediators such as tryptase and histamine. Basophils can also contribute. Although traditionally considered rare, anaphylaxis has a lifetime prevalence of 0.3-5.1%, with incidence rising globally in both adults and children, likely due to increased allergic diseases. Fatalities remain relatively low. Common triggers vary by age and geography: foods such as milk, peanuts, and tree nuts dominate in children; medications-especially beta-lactam antibiotics and non-steroidal anti-inflammatory drugs (NSAIDs)-are more common in adults; and insect venoms are major causes in some regions. Diagnosis is clinical, based on rapid multisystem involvement, despite sometimes isolated cardiovascular or non-inhaled respiratory symptoms may occur. First measures should be to remove the allergen if possible and position the patient appropriately. Immediate intramuscular epinephrine injected into the lateral thigh is the first-line treatment, with repeat dosing as needed. Intranasal epinephrine is emerging as a needle-free alternative. Adjunctive measures include oxygen, intravenous fluids, and bronchodilators for bronchospasm. Antihistamines and corticosteroids may alleviate some symptoms but do not reverse anaphylaxis and must not delay epinephrine administration. Long-term management focuses on identifying triggers through detailed history, IgE testing, skin testing, or challenge procedures; however, up to 10% of cases remain idiopathic. Patients require action plans and access to self-administered epinephrine, though availability is limited in many regions. Key unmet needs include global device access, harmonized definitions and severity scoring, improved education, digital health integration, and deeper investigation into mechanisms, biomarkers, genetics, and cofactors to optimize diagnosis, prevention, and care.","author":[{"family":"Cardona","given":"Victória"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1016/j.waojou.2026.101338","URL":"https://doi.org/10.1016/j.waojou.2026.101338","source":"openalex"},{"id":"oa:W7154025012","type":"article-journal","title":"Responsibility Definition and Risk Management in the Clinical Application of Medical Artificial Intelligence:A Review Based on Four-Level Classification","abstract":"The rapid penetration of medical artificial intelligence (MAI) into clinical diagnosis and treatment scenarios has reshaped the traditional medical service model. However, issues such as ambiguous responsibility definition and lagging risk management have severely constrained its safe and compliant development. This article uses the four-level AI classification system (tool-type, advisor-type, collaborative-type, autonomous-type) as the analytical framework to systematically review the responsibility definition logic and full-lifecycle risk management mechanisms for MAI clinical applications. The study finds that a multi-stakeholder responsibility system covering manufacturers, medical institutions, doctors, and regulatory authorities has been formed, along with a “prevention-control-remediation” risk management closed loop. However, deficiencies still exist in empirical validation, standard unification, adaptation to special scenarios, and coordination between technology and institutions. Future research should focus on directions such as dynamic responsibility quantification and matching, practical design of insurance pools, and development of lightweight management tools for primary care, providing theoretical support and practical references for the systematic governance of MAI clinical applications.","author":[{"family":"Yin","given":"Shihan"}],"issued":{"date-parts":[[2026]]},"DOI":"10.70267/cai.26v3n2.2834","URL":"https://doi.org/10.70267/cai.26v3n2.2834","source":"openalex"},{"id":"oa:W7154614260","type":"article-journal","title":"A Decade of Artificial Intelligence in Stroke Care (2015–2025): Trends, Clinical Translation, and the Precision Medicine Frontier—A Narrative Review","abstract":"Background/Objectives: Stroke generates 157 million disability-adjusted life-years (DALYs) annually, making it the leading neurological cause of global disease burden. Artificial intelligence (AI) and machine learning (ML) have emerged as transformative technologies across the stroke care continuum. This narrative review maps the trajectory of AI in stroke medicine over the decade from 2015 to 2025. Methods: We conducted a narrative review with a structured, pre-specified search strategy across eight pre-specified thematic clusters using PubMed/MEDLINE (January 2015–December 2025), identifying 8549 records and including 1335 studies after screening. Inclusion criteria encompassed primary research articles, systematic reviews, meta-analyses, and RCTs reporting quantitative performance metrics or clinical outcome data for AI/ML in stroke. Results: Stroke imaging AI is the most commercially mature domain, with over 30 FDA-cleared tools. Automated ASPECTS scoring reduced radiologist reading time by 74.8% (AUC 84.97%; 95% CI: 83.1–86.8%). The only triage AI RCT demonstrated an 11.2 min reduction in door-to-groin time without significant improvement in 90-day functional independence (OR 1.3, 95% CI 0.42–4.0). Brain–computer interface rehabilitation showed significant upper limb recovery in a 17-center RCT (FMA-UE mean difference +3.35 points, 95% CI 1.05–5.65; p = 0.0045). AF detection AI is FDA-cleared and RCT-validated. LLMs and federated learning are pre-regulatory but growing exponentially. Conclusions: AI in stroke has achieved diagnostic maturity but therapeutic immaturity. Bridging algorithmic performance to patient outcomes, addressing equity gaps, and building the economic evidence base for scalable deployment are the defining challenges of the next decade.","author":[{"family":"Urfy","given":"Mian"},{"family":"Mir","given":"Mariam"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/jpm16040218","URL":"https://doi.org/10.3390/jpm16040218","source":"openalex"},{"id":"oa:W7160510139","type":"article-journal","title":"Rationalization of reproduction - path towards dehumanization of humanity: Political, legal, and ethical aspects of using artificial intelligence in embryo selection","abstract":"Artificial Intelligence is becoming an essential part of human life, and the creation of life is no exception. While using AI in biomedically assisted reproduction (BMAR) gives new hope to couples struggling with infertility, it also raises a difficult question: where are the ethical limits of letting a machine interfere with human conception? This paper argues that the growing “rationalization” of the reproductive process through AI selection is not just a technical upgrade, but a deeply political and complex issue. In this paper, we analyze the legal and ethical risks of this trend, specifically focusing on how the process of conception is becoming “dehumanized”. A major concern is that we are relying too much on algorithms that nobody truly understands. This “black box” nature of AI can easily undermine the autonomy of parents and the validity of their informed consent in one of the most private moments of their lives. The philosophical danger of treating human embryos as mere objects for selection, which directly threatens their inherent dignity, is also examined. With the goal of pointing out gaps in our current laws, the paper looks closely at the regulations in the Republic of Serbia. We find that existing legal solutions are not fully prepared for these new challenges. Finally, we propose new regulatory steps that would put the dignity of the embryo first. Through a prism of ethics and a warning against the loss of human empathy, this paper concludes that an uncritical use of AI in reproduction carries a serious risk of turning the act of creating a human being into a cold, mechanical procedure.","author":[{"family":"Stjepanović","given":"Bogdana"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5937/spm96-64787","URL":"https://doi.org/10.5937/spm96-64787","source":"openalex"},{"id":"oa:W7124545010","type":"article-journal","title":"Can artificial intelligence debunk health misinformation more effectively than humans? A three‐dimensional persuasion analysis","abstract":"Abstract Health misinformation presents significant challenges to public well‐being, making effective debunking strategies crucial. While artificial intelligence (AI) shows potential in generating debunking texts, its persuasiveness compared to human‐generated content remains underexplored. Drawing on Aristotle's three modes of persuasion, this study investigated the persuasive effectiveness of AI versus human‐generated health debunking texts through three complementary studies. Our findings reveal a novel pattern: AI‐generated texts significantly outperformed human texts in pathos (emotional appeal) and logos (logical argument) but underperformed in ethos (credibility), with all three dimensions serving as significant mediators of persuasiveness. More importantly, we demonstrate that source labeling effects are not uniform. While “AI‐written” labels reduced perceived persuasiveness for both AI and human texts, this algorithmic aversion was attenuated when argument quality (logos) was made salient. These findings advance persuasion theory by revealing that classical rhetoric operates differently for AI versus human sources and that algorithmic aversion is context‐dependent rather than universal. The results offer both theoretical insights into human‐AI communication and practical guidance for deploying AI in health misinformation mitigation.","author":[{"family":"Ji","given":"Xinyu"},{"family":"Zhang","given":"Xing"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1002/asi.70049","URL":"https://doi.org/10.1002/asi.70049","source":"openalex"},{"id":"oa:W7169884846","type":"article-journal","title":"Research and Innovation in Case Management: A Decade in Review (2016–2026)","abstract":"ABSTRACT: Research and innovation are foundational to the transformation of professional case management and utilization review (CM/UR). Over the past decade, CM/UR has evolved significantly to address care complexity, workforce challenges, regulatory change, and administrative burden. Innovative workforce models, expanded community coordination, and advances in analytics, automation, artificial intelligence, and interoperability have improved efficiency, compliance, continuity of care, and patient outcomes while supporting value-based care. As regulatory requirements continue to evolve, sustained research and agile innovation remain essential to optimizing CM/UR delivery while maintaining a strong person-centered focus.","author":[{"family":"Provo","given":"Heaven"}],"issued":{"date-parts":[[2026]]},"DOI":"10.1097/ncm.0000000000000890","URL":"https://doi.org/10.1097/ncm.0000000000000890","source":"openalex"},{"id":"oa:W7159795434","type":"article-journal","title":"The Impact of Generative Artificial Intelligence Use on Perceived English Learning Achievement: The Roles of Use Behavior and Task–Technology Fit","abstract":"The rapid advancement of generative artificial intelligence (GAI) has intensified interest in its potential to support English learning in higher education. However, the mechanisms through which students' perceptions and motivations translate into learning achievement remain unclear. Drawing on the Unified Theory of Acceptance and Use of Technology (UTAUT) and Task-Technology Fit (TTF) theory, this study investigates how undergraduate students' use of GAI relates to perceived English learning achievement and under what conditions these associations are amplified. Using covariance-based structural equation modeling (CB-SEM), data from 537 undergraduate students across five public universities in China were analyzed. The findings indicate that performance expectancy, effort expectancy, facilitating conditions, perceived competitiveness, and artificial intelligence self-efficacy significantly predict GAI use. In turn, use behavior mediates their relationships with perceived English learning achievement. Task-Technology Fit further moderates the link between use behavior and learning achievement, with stronger associations observed when GAI functionalities are perceived as closely aligned with task requirements. These results highlight the importance of use behavior and task alignment in explaining how GAI is associated with students' perceived English learning achievement and extend technology acceptance research within AI-supported language learning contexts.","author":[{"family":"Wang","given":"Zhongrui"},{"family":"Guo","given":"Shibao"}],"issued":{"date-parts":[[2026]]},"DOI":"10.3390/bs16050643","URL":"https://doi.org/10.3390/bs16050643","source":"openalex"},{"id":"oa:W7164344883","type":"article-journal","title":"Nanomedicine in 2026: Illustrative Quantitative Analyses of EPR Heterogeneity, Clinical Trial Attrition, and Emerging Horizons for Active Nanotherapeutics","abstract":"2026 is a turning point for nanomedicine, marking the field's transition from decades of preclinical promise toward tangible clinical impact. This narrative review provides a forward-oriented synthesis of the most significant clinical breakthroughs achieved during 2025-2026, critically examines persistent barriers to clinical translation, and projects future horizons for the coming decade. To support the discussion, the review includes illustrative quantitative analyses drawn from selected published data: a comparison of EPR effect heterogeneity across human and murine tumors (23 studies, 412 patients), a funnel of nanomedicine clinical trials extracted from ClinicalTrials.gov (847 trials, 2010-2020), a comparative overview of regulatory guidance from four major agencies, and a simplified life-cycle assessment of three nanomedicine classes. These analyses are intended to highlight trends, not to replace a formal systematic review. We identify four important clinical advances: first Phase II data for hafnium oxide nanoparticle radioenhancers in inoperable lung cancer; logic-gated STING-agonistic nanoparticles for metastasis-specific immunotherapy; ultrasmall silica nanoparticles that remodel suppressive tumor microenvironments independent of a drug cargo; and CNM-Au8 gold nanocrystals advancing toward regulatory submission for amyotrophic lateral sclerosis. Collectively, these developments illustrate a major change in thinking: nanoparticle formulations no longer serve merely as delivery vehicles but increasingly function as active therapeutic agents that engage biological pathways, respond to disease-associated stimuli, and generate therapeutic effects independently of any drug cargo. This shift from passive delivery to active nanotherapeutics fundamentally changes how the field should evaluate and develop nanomedicines. Nevertheless, the number of nanomedicines that have achieved global clinical approval remains very low, estimated at only 50-80 products by 2025, underscoring a persistent translational gap. We analyze principal obstacles to clinical success, including the limited predictive validity of the enhanced permeability and retention (EPR) effect in humans, batch-to-batch manufacturing variability, safety concerns arising from bio-corona formation and organ accumulation, and the absence of harmonized regulatory frameworks. Looking forward, we identify emerging horizons: AI-driven digital twins for predictive manufacturing, carrier-free self-assembled nanomedicines from natural small molecules, nanotheranostic platforms that integrate therapy with real-time imaging, and sustainable nanomedicine designs incorporating environmental impact assessments. By bridging clinical reality with future potential, this review aims to inform researchers, clinicians, and regulatory stakeholders navigating the rapidly evolving landscape of nanomedicine.","author":[{"family":"Fayez","given":"Sayed"}],"issued":{"date-parts":[[2026]]},"DOI":"10.2147/ijn.s618407","URL":"https://doi.org/10.2147/ijn.s618407","source":"openalex"},{"id":"oa:W7196941818","type":"article-journal","title":"Challenges in the diagnosis of malignant skin neoplasms using artificial intelligence software and approaches to their resolution on the example of Derma Onko Melanoma Check software","abstract":"Background. Artificial intelligence (AI)-based computer vision software is becoming an increasingly important tool in dermato-oncological practice. Deep learning models have demonstrated the ability to classify skin neoplasms with accuracy comparable to that of experienced dermatologists. However, their implementation in real clinical practice has revealed several significant challenges. Objective: To analyze and systematize the major challenges arising in the AI-assisted diagnosis of malignant skin neoplasms and to propose solutions to each identified challenge using the Derma Onko Melanoma Check software as an example. Material and methods. The clinical experience of applying AI software for the diagnosis of malignant skin neoplasms was analyzed, and six major categories of challenges were identified. Specific solutions to each of these challenges are proposed on the example of the Derma Onko Melanoma Check clinical decision support system (a multimodel AI-based computer vision system) for the preliminary diagnosis and triage of patients with skin neoplasms. Results. Six major challenges were identified and systematized: (1) imbalanced training datasets, with an overrepresentation of dermoscopic images and an insufficient number of clinical photographs; (2) poor quality of images acquired by physicians (e.g., overexposure, blur, and underexposure), limiting diagnostic reliability; (3) submission of images in which the lesion occupies an insufficient area of the frame for reliable analysis; (4) the single-model architecture of most AI systems and the resulting lack of adequate oncological alertness; (5) the absence of explainable artificial intelligence (XAI) in AI-generated reports; and (6) the inability of physicians to modify or refine the AI-generated conclusion using information obtained from the patient’s medical history and clinical examination. For each challenge, a specific solution implemented in Derma Onko Melanoma Check is proposed, including automated image quality assessment, lesion size verification, a multimodel ensemble of neural networks, XAI-based visualization of high-risk regions, interactive entry of clinical history data, and support for multimodal analysis. Conclusion. The implementation of AI systems incorporating the proposed solutions into routine clinical practice, particularly in primary care settings involving general practitioners, may significantly improve the early detection of malignant skin neoplasms by raising oncological vigilance for suspicious lesions and increasing the clinical interpretability and reliability of AI-generated diagnostic conclusions.","author":[{"family":"Korabelnikov","given":"DI"},{"family":"Lamotkin","given":"AI"}],"issued":{"date-parts":[[2026]]},"DOI":"10.17749/2070-4909/farmakoekonomika.2026.403","URL":"https://doi.org/10.17749/2070-4909/farmakoekonomika.2026.403","source":"openalex"},{"id":"oa:W7133527232","type":"article-journal","title":"An Anthropological Understanding of Artificial Intelligence Transformations in Civic and Domestic Life, Labor, and Higher Education Through the Cybernetic Organism (Cyborg) Concept","abstract":"Este ensayo utiliza la antropología cyborg para modelar las interacciones entre los humanos y la inteligencia artificial (ia), y destaca las fortalezas y limitaciones de la iaen relación con el aprendizaje y la innovación. Basados en el énfasis de Haraway en las desigualdades, reconocemos que la iano constituye una forma separada ni superior de inteligencia, sino una herramienta que amplifica la capacidad de acción humana. Los ejemplos contemporáneos de implementación de IA en ámbitos cívicos, domésticos, laborales y educativos evidencian su potencial para reproducir sesgos y formas de opresión, en tanto se trata de una construcción humana que refleja —y a la vez influye en— los valores socioculturales y éticos de sus creadores y usuarios. Estos casos ofrecen insumos para plantear consideraciones prácticas y éticas sobre el desarrollo y uso futuros de la ia, entre ellas la necesidad de fomentar colaboraciones estructuradas entre humanos y sistemas algorítmicos, promover la alfabetización humano-ia, garantizar el respeto por la dignidad y los derechos humanos, y abordar los sesgos y desigualdades inherentes.","author":[{"family":"Wells","given":"Joshua"},{"family":"Vanderveen","given":"James"}],"issued":{"date-parts":[[2026]]},"DOI":"10.15446/mag.v40n1.124922","URL":"https://doi.org/10.15446/mag.v40n1.124922","source":"openalex"},{"id":"oa:W4410071166","type":"article-journal","title":"Forecasting Cancer Incidence in Canada by Age, Sex, and Region Until 2026 Using Machine Learning Techniques","abstract":"This study analyzes cancer trends in Canada using machine learning techniques to extract insights from extensive cancer data sourced from the Canadian Cancer Society and Statistics Canada. It aims to enhance the understanding of cancer epidemiology and inform better prevention, diagnosis, and treatment strategies. Data preprocessing addressed issues like missing values and normalization, ensuring reliability. The findings indicate a steady increase in new cancer cases, with estimates reaching 248,700 in 2026, up from 244,000 in 2022. Male incidence rates are projected to rise slightly to 602.3 per 100,000, while female rates may decline to 530.6. Regions such as Alberta, British Columbia, Ontario, and Quebec show rising incidence rates, contrasted by declines in Newfoundland and Labrador, Nunavut, and Yukon. Notably, this research reveals significant increases in cancer cases among individuals aged 60 and older, particularly those 70+. The hybrid ARIMA-LSTM model demonstrated superior forecasting accuracy compared with the other selected models. These findings offer valuable insights for health policymakers and highlight the potential of machine learning in public health forecasting, providing a framework for future research in other disease areas.","author":[{"family":"Kaviani","given":"Ehsan"},{"family":"Passi","given":"Kalpdrum"}],"issued":{"date-parts":[[2025]]},"DOI":"10.3390/a18050265","URL":"https://doi.org/10.3390/a18050265","source":"openalex"},{"id":"oa:W7168284949","type":"article-journal","title":"Application of Data-Based Artificial Intelligence in the Aviation Industry: A Conceptual-Analytic Review of Machine Learning and Deep Learning Methods","abstract":"Abstract: The aviation industry, as a complex and safety-sensitive technical-operational system, faces a huge volume of heterogeneous data, including flight time series, aircraft and engine health data, spatial and temporal air traffic data, meteorological data, textual safety and repair reports, and visual inspection data. This article provides a conceptual-analytical review that aims to explain the logic of \"data-driven artificial intelligence\" in aviation and describe how machine learning and deep learning methods can be purposefully utilized to produce \"operational knowledge\" and \"actionable decisions.\" The present review approach, rather than simply comparing algorithms, focuses on the “problem-data-model-output” mapping framework and suggests that the choice of analytical method should be a function of the type of operational problem (prediction, anomaly detection, classification, image analysis, and sequential decision making), the nature of the available data, and the requirements of industrial deployment. The results of the review indicate that classical machine learning methods have greater advantages in more structured problems requiring interpretability, and deep learning models have greater advantages in large/complex or unstructured data (images, text, long time series). However, successful transition from a research environment to an operational environment faces challenges such as lack of labeled data, class imbalance and rare events, changing data scope, need for interpretability, and regulatory constraints. Finally, the paper suggests future directions in the form of multi-source learning, robust and adaptive learning, secure reinforcement learning, human-in-the-loop, and certification frameworks for sustainable deployment of AI in aviation.","author":[{"family":"Narimanidehnavi","given":"Mortza"}],"issued":{"date-parts":[[2026]]},"DOI":"10.65278/ijtaci.2026.3","URL":"https://doi.org/10.65278/ijtaci.2026.3","source":"openalex"},{"id":"doi:10.5281/zenodo.22062863","type":"article-journal","title":"AI Companion Mortality Database: Documented Deaths Associated with Conversational AI Systems (2023–2026)","abstract":"Superseded by version 3.5.3 (10.5281/zenodo.22063180). In this version, the derived files platform-analysis.csv and timeline.json were generated from v3.5.0 and do not reflect v3.5.2 changes (ChatGPT third-party fatalities 15 → 16; Margaux Whittemore outcome and legal status; Kim Seoul case name). mortality-data.json is correct. Use 3.5.3 or later. A public, source-verified register of deaths in which AI chatbot interaction was alleged as a contributing factor, March 2023 through August 2026. Version 3.5.2 documents 35 fatalities across 24 incidents (occurrences of harm) on 8 tracked platforms: 18 AI users who died and 17 third-party victims killed by AI users. Every incident is supported by court filings, official government statements, or multiple independent news sources; the database makes no independent claim of causation. Scope. This database documents deaths associated with conversational AI systems — chatbots and LLM-based assistants with which a person interacted through natural language. It does not cover autonomous vehicles, clinical or diagnostic machine learning, industrial or robotic automation, autonomous or AI-assisted weapons, content-recommendation systems, or any other application of artificial intelligence. Those are distinct phenomena with distinct evidentiary standards. \"Companion\" in the database's name reflects its origin in companion-chatbot cases; the scope has since expanded to general-purpose assistants, and the name is retained for continuity of citation. The canonical record is mortality-data.json (incident records with sources, legal status, mechanism classification, and key factors; platform records; regulatory responses; derived statistics). Derived exports platform-analysis.csv and timeline.json are regenerated from it. methodology.md defines the incident definition, the three causal pathways (relational, cognitive, instrumental), the adjudicated-status rule for describing killings, the rule for counting third-party victims, and the verification tiers; verification-standards.md details the evidence tiers and sub-labels. Compiled and maintained by one independent researcher with AI assistance (disclosed in the record and on the site; the assisting system's developer is among the platforms tracked). Live site: aimortality.org. Source and history: github.com/aimortality/ai-companion-mortality-database. Correspondence: contact@aimortality.org.","author":[{"family":"Karman","given":"Hunter"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22062863","URL":"https://doi.org/10.5281/zenodo.22062863","source":"datacite"},{"id":"doi:10.5281/zenodo.22063180","type":"article-journal","title":"AI Companion Mortality Database: Documented Deaths Associated with Conversational AI Systems (2023–2026)","abstract":"A public, source-verified register of deaths in which AI chatbot interaction was alleged as a contributing factor, March 2023 through August 2026. Version 3.5.3 documents 35 fatalities across 24 incidents (occurrences of harm) on 8 tracked platforms: 18 AI users who died and 17 third-party victims killed by AI users. Every incident is supported by court filings, official government statements, or multiple independent news sources; the database makes no independent claim of causation. Scope. This database documents deaths associated with conversational AI systems — chatbots and LLM-based assistants with which a person interacted through natural language. It does not cover autonomous vehicles, clinical or diagnostic machine learning, industrial or robotic automation, autonomous or AI-assisted weapons, content-recommendation systems, or any other application of artificial intelligence. Those are distinct phenomena with distinct evidentiary standards. \"Companion\" in the database's name reflects its origin in companion-chatbot cases; the scope has since expanded to general-purpose assistants, and the name is retained for continuity of citation. The canonical record is mortality-data.json (incident records with sources, legal status, mechanism classification, and key factors; platform records; regulatory responses; derived statistics). Derived exports platform-analysis.csv and timeline.json are regenerated from it. methodology.md defines the incident definition, the three causal pathways (relational, cognitive, instrumental), the adjudicated-status rule for describing killings, the rule for counting third-party victims, and the verification tiers; verification-standards.md details the evidence tiers and sub-labels. Compiled and maintained by one independent researcher with AI assistance (disclosed in the record and on the site; the assisting system's developer is among the platforms tracked). Live site: aimortality.org. Source and history: github.com/aimortality/ai-companion-mortality-database. Correspondence: contact@aimortality.org.","author":[{"family":"Karman","given":"Hunter"}],"issued":{"date-parts":[[2026]]},"DOI":"10.5281/zenodo.22063180","URL":"https://doi.org/10.5281/zenodo.22063180","source":"datacite"}]