SEARCH · Search PubMed

Search Search PubMed

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources

PubMed2026

The Role of Artificial Intelligence Combined With Digital Cholangioscopy for Indeterminant and Malignant Biliary Strictures: A Systematic Review and Meta-analysis.

BACKGROUND: Current endoscopic retrograde cholangiopancreatography (ERCP) and cholangioscopic-based diagnostic sampling for indeterminant biliary strictures remain suboptimal. Artificial intelligence (AI)-based algorithms by means of computer vision in machine learning have been applied to cholangioscopy in an effort to improve diagnostic yield. The aim of this study was to perform a systematic review and meta-analysis to evaluate the diagnostic performance of AI-based diagnostic performance of AI-associated cholangioscopic diagnosis of indeterminant or malignant biliary strictures. METHODS: Individualized searches were developed in accordance with PRISMA and MOOSE guidelines, and meta-analysis according to Cochrane Diagnostic Test Accuracy working group methodology. A bivariate model was used to compute pooled sensitivity and specificity, likelihood ratio, diagnostic odds ratio, and summary receiver operating characteristics curve (SROC). RESULTS: Five studies (n=675 lesions; a total of 2,685,674 cholangioscopic images) were included. All but one study analyzed a deep learning AI-based system using a convoluted neural network (CNN) with an average image processing speed of 30 to 60 frames per second. The pooled sensitivity and specificity were 95% (95% CI: 85-98) and 88% (95% CI: 76-94), with a diagnostic accuracy (SROC) of 97% (95% CI: 95-98). Sensitivity analysis of CNN studies (4 studies, 538 patients) demonstrated a pooled sensitivity, specificity, and accuracy (SROC) of 95% (95% CI: 82-99), 88% (95% CI: 72-95), and 97% (95% CI: 95-98), respectively. CONCLUSIONS: Artificial intelligence-based machine learning of cholangioscopy images appears to be a promising modality for the diagnosis of indeterminant and malignant biliary strictures.

Humans↗

PubMed2026

The role of artificial intelligence in the diagnosis and prognosis of traumatic brain injury based on brain CT scans: a systematic review.

Traumatic brain injury (TBI) is a leading cause of emergency department visits and a major contributor to injury-related mortality and long-term neurological disability. Non-contrast computed tomography (CT) is the gold-standard imaging modality for the rapid diagnosis of TBI. Clinical outcomes depend strongly on early detection and prompt acute management. Artificial intelligence (AI)-based models may support faster automated identification of traumatic findings and early prediction of patient prognosis. A systematic literature search was conducted in PubMed/MEDLINE, Scopus, IEEE Xplore, ACM Digital Library, and the Cochrane Library in accordance with PRISMA 2020 guidelines to evaluate AI-based models for automated detection of TBI-related findings on CT and for prediction of clinical outcomes. Risk of bias and applicability were assessed using QUADAS-2 for diagnostic accuracy studies and PROBAST + AI for prediction model studies. Twenty-two studies were included. Sixteen studies evaluated diagnostic tasks and 10 evaluated prognostic outcomes, with four studies contributing to both categories. Diagnostic performance was generally high, with many studies reporting AUC values approaching or exceeding 0.90, particularly for larger lesion volumes.Prognostic performance was more variable, with moderate to high discrimination and substantial heterogeneity. Only 9 studies incorporated independent external validation, and performance was frequently lower in external cohorts. All prognostic model studies were judged to be at high overall risk of bias using PROBAST + AI, and most diagnostic accuracy studies also demonstrated high or unclear risk of bias in at least one QUADAS-2 domain, most frequently in patient selection. AI-based models applied to brain CT demonstrate strong technical performance for both diagnostic and prognostic tasks in TBI. However, most studies relied on retrospective designs and lacked independent external validation which limits models generalizability and raises concern for potential overfitting. Prospective, multicenter studies with standardized methodologies and rigorous external validation are required before widespread clinical implementation.

Humans↗

PubMed2026

Performance of AI-Based Screening Tools for Obstructive Sleep Apnea Across Apnea-Hypopnea Index Thresholds: Systematic Review and Meta-Analysis.

BACKGROUND: Obstructive sleep apnea (OSA) is highly prevalent but remains substantially underdiagnosed. Polysomnography (PSG) is the reference standard, but its cost and limited availability constrain large-scale case identification. AI-based screening tools may support risk stratification and referral prioritization, but their diagnostic accuracy across apnea-hypopnea index (AHI) thresholds remains uncertain. OBJECTIVE: This review aimed to systematically evaluate the diagnostic accuracy of AI-based OSA screening tools at AHI thresholds of ≥5, ≥15, and ≥30 events/hour, with emphasis on models using non-PSG-derived inputs. METHODS: PubMed, Embase, Scopus, and Web of Science were searched for studies published from January 1, 2016, to May 3, 2026. Eligible studies included adults evaluated for suspected OSA or recruited from population-based cohorts, assessed AI-based models intended or interpretable for OSA screening, risk prediction, or screening-oriented severity classification, used PSG as the reference standard, and reported sufficient data to construct or reconstruct 2×2 contingency tables. Diagnostic accuracy was synthesized separately by AHI threshold and input source using bivariate random-effects models, with 95% CIs and prediction intervals (PIs). Risk of bias and certainty of evidence were assessed using QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2) and GRADE (Grading of Recommendations Assessment, Development, and Evaluation), respectively. RESULTS: A total of 60 studies were included, of which 47 contributed data to the meta-analysis. At AHI thresholds of ≥5, ≥15, and ≥30 events/hour, pooled sensitivities were 0.94 (95% CI 0.92-0.96; 95% PI 0.71-0.99), 0.87 (95% CI 0.84-0.89; 95% PI 0.66-0.96), and 0.83 (95% CI 0.79-0.87; 95% PI 0.61-0.94), respectively; the corresponding specificities were 0.77 (95% CI 0.69-0.84; 95% PI 0.30-0.96), 0.81 (95% CI 0.75-0.85; 95% PI 0.39-0.96), and 0.91 (95% CI 0.87-0.94; 95% PI 0.55-0.99), respectively. The corresponding areas under the summary receiver operating characteristic curves were 0.943, 0.907, and 0.920. For non-PSG-derived tools, sensitivities were 0.92, 0.85, and 0.81, and specificities were 0.70, 0.74, and 0.85 at the 3 thresholds, respectively. For PSG-derived models, sensitivities were 0.96, 0.90, and 0.85, and specificities were 0.82, 0.88, and 0.96, respectively. Exploratory subgroup analyses suggested performance variation across selected study and model characteristics, including region, algorithmic framework, data source, and validation method. CONCLUSIONS: AI-based tools showed generally favorable screening performance for OSA across clinically relevant AHI thresholds, although wide PIs suggest variable performance across future comparable populations and settings. By synthesizing diagnostic accuracy across 3 AHI thresholds and distinguishing non-PSG-derived from PSG-derived models, this review extends previous broad or modality-specific reviews and offers a clinically interpretable, pathway-specific basis for linking model performance to intended use. The findings may clarify potential roles for non-PSG-derived tools in front-end screening and referral prioritization and for PSG-derived models in reduced-channel assessment and sleep-laboratory workflow support. Given substantial heterogeneity, limited external validation, and low or very low certainty of evidence, prospective validation is needed before routine implementation.

Humans↗

PubMed2026

Comparative Efficacy of Different AI Systems for Polyp Detection by Size During Colonoscopy: Systematic Review and Network Meta-Analysis.

BACKGROUND: Colorectal cancer remains a leading cause of death despite being largely preventable through polypectomy. AI systems designed to enhance polyp detection during colonoscopy have shown promise, but the extent to which they improve detection of different-sized polyps remains unclear. OBJECTIVE: This study compared the size-stratified efficacy of AI-assisted colonoscopy vs standard colonoscopy using the Hartung-Knapp-Sidik-Jonkman (HKSJ) method, and generated exploratory rankings while acknowledging all cross-platform comparisons are indirect. METHODS: This systematic review and network meta-analysis (NMA) searched PubMed, Embase, Cochrane CENTRAL, and Web of Science from inception to July 25, 2026, supplemented by citation searching. We included randomized controlled trials (RCTs) comparing AI-assisted vs standard colonoscopy in adults (≥18 years of age), reporting mean polyp detection counts stratified by size (≤5 mm, 6-9 mm, and ≥10 mm). Two reviewers screened studies, extracted data, and assessed risk of bias using the Cochrane Risk of Bias 2.0. We conducted frequentist NMA using the HKSJ method with restricted maximum likelihood estimation, calculated 95% prediction intervals (PIs), and assessed heterogeneity using I2 and τ2. Certainty of evidence was rated using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) framework. RESULTS: A total of 13 RCTs (4156 participants) compared 8 AI systems to standard colonoscopy, forming a network without direct AI comparisons. For diminutive polyps (≤5 mm), AI showed a modest advantage (standardized mean difference [SMD] 0.21, 95% CI 0.07 to 0.35, 95% PI -1.12 to 1.54), but substantial heterogeneity (I2=86.6%) and wide PI crossing the null indicated high uncertainty. EndoScreener showed the most consistent evidence (SMD 0.36, 95% CI 0.18-0.54). For small and large polyps, effects were minimal (SMD 0.02, 95% CI -0.02 to 0.06, 95% PI -0.03 to 0.07; SMD 0.01, 95% CI 0.00-0.02, 95% PI -0.01 to 0.03). GRADE certainty was very low for diminutive polyps and low for small and large polyps. Sensitivity analysis excluding Tianjin YuJin did not materially change findings. CONCLUSIONS: AI may modestly enhance diminutive polyp detection, but effects on small and large polyps are minimal, with no platform superiority. Given very low to low certainty, findings are hypothesis-generating. This exploratory NMA provides size-stratified comparisons that can inform future head-to-head trial design. Unlike prior reviews aggregating all polyp sizes, we show the overall AI benefit is driven by diminutive polyp detection, providing a framework for targeted deployment-prioritizing AI for diminutive polyp screening, with limited value for larger lesions. Head-to-head trials are urgently needed. TRIAL REGISTRATION: PROSPERO International Prospective Register of Systematic Reviews CRD420251266932; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251266932.

Colonoscopy↗

PubMed2026

Plant cis-regulatory grammar: Decoding the multidimensional code of transcriptional regulation for programmable crop engineering.

Cis-regulatory elements (CREs) orchestrate the spatiotemporal precision of gene expression that underlies plant development, adaptation, and domestication. Decoding the cis-regulatory grammar of plant genomes remains a central challenge in modern biology, with profound implications for programmable crop engineering. Here, recent conceptual and technological advances are synthesized to reshape our understanding of plant CREs. This review first argues that CRE function is not only an intrinsic property of DNA sequence alone but also emerges from a multidimensional context, including chromatin accessibility, histone modifications, three-dimensional genome topology, and cell type-specific regulatory landscapes. Furthermore, the convergence of single-cell epigenomics, high-throughput functional assays, and CRISPR-based dissection has begun to unravel this contextual grammar, revealing the computational principles governing transcriptional regulation. Critically, we propose that artificial intelligence (AI) platforms are catalyzing an ongoing transition from descriptive discovery to predictive engineering, wherein these platforms outperform natural evolution in designing synthetic CREs. Finally, a roadmap is outlined toward a plant regulatory grammar foundation model, which will enable truly predictive engineering of gene expression when fine-tuned for specific tasks. Collectively, the integration of single-cell resolution maps, precise genome editing, AI-driven design, and regulatory-compliant delivery systems promises to transform our ability to reprogram plant gene regulation for next-generation agriculture, bridging the gap between foundational regulatory biology and tangible crop improvement.

artificial intelligence↗

PubMed2026

Applications of artificial intelligence in robot-assisted surgery: a systematic review.

To characterize applications of artificial intelligence (AI) in robot-assisted surgery, summarize technical and clinical performance, and assess the quality of the available evidence. PubMed, Web of Science Core Collection, and Scopus were searched for English-language journal articles published from 1 January 2020 through 31 October 2025. Randomized, observational, model-development, validation, and feasibility studies evaluating AI in robot-assisted surgery or closely related image-guided minimally invasive workflows were eligible. Two reviewers independently performed study selection, data extraction, and risk-of-bias assessment. Owing to heterogeneity in surgical procedures, AI tasks, analytical units, validation strategies, and outcomes, findings were synthesized descriptively without statistical pooling. The review was registered in the International Prospective Register of Systematic Reviews (CRD420251175699). Seventeen studies were included: seven clinical prediction or decision-support studies, eight intraoperative recognition, segmentation, or image-guided studies, and two training or workflow studies. Five prediction studies reported area-under-the-curve values of 0.74-0.95. Technical studies reported F1 or Dice scores of 0.525-0.995 and task-specific accuracies of 0.840-0.998. Two randomized studies suggested benefits for personalized suturing feedback and automated camera control, but neither established improved patient outcomes. Only one study had low overall risk of bias; the remaining studies were at high or unclear risk or raised some concerns. AI applications in robot-assisted surgery show promise for prediction, intraoperative perception, training, and workflow support. Evidence primarily demonstrates technical feasibility rather than established clinical effectiveness. Independent multicenter validation and prospective evaluation of patient, educational, and workflow outcomes are required before widespread implementation.

Robotic Surgical Procedures↗

PubMed2026

Artificial Intelligence for Diagnosis, Risk Stratification, and Prognosis of Neuroblastoma - A Systematic Review and Meta-Analysis.

PURPOSE: To synthesizes evidence on artificial intelligence (AI) performance in neuroblastoma (NB) diagnosis, risk stratification, prognosis, and genomic characterization. MATERIALS AND METHODS: A systematic review and meta-analysis was conducted following PRISMA 2020 guidelines (PROSPERO: CRD42024539475) across five databases. Meta-analyses used random-effects models with logit-transformed Area Under the Curve (AUCs) and cluster-robust standard errors. AI models were classified as Machine Learning Models (MLM) or Hybrid Nomograms (HN) based on their construction methodology. RESULTS: Of 3,742 articles identified, 53 were included. MLMs demonstrated higher point estimates than radiologists in differential diagnosis (AUC: 0.87 vs. 0.83), though this difference was not statistically significant and carried substantial uncertainty. HNs achieved stronger performance in risk stratification (AUC: 0.87). AI-derived nomograms (AUC: 0.9) and gene signatures (AUC: 0.8) outperformed conventional prognostic markers descriptively. Chemotherapy response prediction remained below clinical utility thresholds across all model types. Only 33.9% of models reported calibration and 24.5% underwent external validation. CONCLUSIONS: AI demonstrates proof-of-concept across multiple NB clinical domains. However, clinical adoption remains premature given persistent gaps in external validation, calibration, dataset size, and pediatric-specific model development. Future studies should test these models prospectively in multicenter pediatric cohorts, ideally through COG or SIOPEN, using shared definitions for diagnosis, risk group, treatment response, and survival outcomes.

Humans↗

PubMed2026

Diagnostic Performance of Machine Learning for Systemic Lupus Erythematosus: Systematic Review and Meta-Analysis.

BACKGROUND: Early and accurate diagnosis of systemic lupus erythematosus (SLE) and its organ involvement is essential. Previous reviews of machine learning (ML) in SLE combined heterogeneous tasks and validation strategies and may have overinterpreted model performance. OBJECTIVE: This study evaluated the diagnostic performance of ML and deep learning (DL) models for 3 clinically distinct SLE-related tasks: SLE classification or diagnosis, lupus nephritis (LN) diagnosis, and neuropsychiatric systemic lupus erythematosus (NPSLE) discrimination. We also assessed methodological quality and certainty of evidence. METHODS: PubMed, Embase, Cochrane Library, Web of Science, and IEEE Xplore were searched from January 2014 to April 2026. Eligible peer-reviewed diagnostic accuracy studies developed or validated ML or DL models for 1 of the 3 prespecified tasks, used an accepted reference standard, and provided data for a 2×2 contingency table. Bivariate random-effects meta-analyses with the Hartung-Knapp-Sidik-Jonkman adjustment were used to pool sensitivity and specificity. We reported 95% prediction intervals (PIs), assessed risk of bias using the Quality Assessment of Diagnostic Accuracy Studies for Artificial Intelligence tool (QUADAS-AI; Viknesh Sounderajah [Imperial College London]), and evaluated certainty of evidence using the Grading of Recommendations Assessment, Development, and Evaluation framework for diagnostic test accuracy. RESULTS: Twenty-nine studies were included: 17 for SLE classification, 5 for LN diagnosis, and 7 for NPSLE discrimination. In the primary task-stratified analysis, pooled sensitivity was 0.91 (95% CI 0.86-0.94; 95% PI 0.56-0.99), and pooled specificity was 0.94 (95% CI 0.91-0.96; 95% PI 0.69-0.99), with low heterogeneity (I²=23.9% and 22.9%, respectively). DL models showed a sensitivity of 0.93 and specificity of 0.95, compared with 0.88 and 0.94 for traditional ML models. Certainty of evidence was high for most analyses but low for LN diagnosis because of inconsistency and imprecision. All studies were retrospective, and only 9 of 29 (31%) performed independent external validation. Overall risk of bias was high or unclear in 22 of 29 (75.9%) studies. No study reported model calibration, decision-curve analysis, or net clinical benefit. CONCLUSIONS: ML models showed promising diagnostic accuracy across 3 distinct SLE-related tasks, but wide PIs, limited external validation, and pervasive risk of bias restrict conclusions about real-world generalizability. Prospective multicenter studies with standardized tasks and reference standards, independent external validation, and formal assessment of calibration and clinical utility are required before clinical implementation.

Humans↗

PubMed2026

Evaluation of a cornea-specialized large language model for diagnostic and management accuracy in complex corneal cases.

PURPOSE: To evaluate whether a cornea-specialized large language model (LLM) enhanced with retrieval-augmented generation (RAG) improves clinicians' diagnostic and management accuracy in complex corneal cases compared to a general-purpose GPT-4o model and unaided clinician performance. METHODS: This prospective, randomized, masked evaluation study involved three cornea trainees who each independently reviewed 39 real-world corneal cases under three experimental conditions: unaided, GPT-4o-assisted, and assisted by a cornea-specialized GPT-4o model. The cornea-specialized model was constructed by embedding over 200 publicly available Wikipedia articles into GPT-4o's RAG framework. Participants provided open-ended diagnoses and selected the next-step management options (multiple choice). They were allowed up to three GPT-4o queries per case, and the AI-assisted arms were randomized to minimize bias. Accuracy for both tasks was compared against expert reference standards using McNemar's test. RESULTS: Diagnostic accuracy was 48.7%, 20.5%, and 38.5% unaided, improving to 69.2%, 46.2%, and 59.0% with general GPT-4o (p<0.04). The cornea-specialized GPT-4o further improved accuracy to 71.8%, 48.7%, and 74.4%, with improvements over unaided performance for all clinicians (p<0.01). For next-step decisions, unaided accuracy was 76.9%, 87.2%, and 59.0%. With the specialized model, Ophthalmologist 3 improved to 71.8% (p<0.05), Ophthalmologist 1 remained high at 82.1%, and Ophthalmologist 2 declined to 64.1% (p<0.05). CONCLUSIONS: A cornea-specialized LLM enhanced with RAG improved diagnostic accuracy in complex corneal cases, particularly among clinicians with lower baseline performance. Effects on management accuracy were inconsistent. Future studies should explore the use of open-ended management tasks and examine whether smaller, curated retrieval corpora yield better model performance.

Humans↗

PubMed2026

Assessing AI literacy and attitudes among medical students: implications for integration into healthcare practice.

PURPOSE: This study aims to assess AI literacy and attitudes among medical students and explore their implications for integrating AI into healthcare practice. DESIGN/METHODOLOGY/APPROACH: A quantitative research design was employed to comprehensively evaluate AI literacy and attitudes among 374 Lusaka Apex Medical University medical students. Data were collected from April 3, 2024, to April 30, 2024, using a closed-ended questionnaire. The questionnaire covered various aspects of AI literacy, perceived benefits of AI in healthcare, strategies for staying informed about AI, relevant AI applications for future practice, concerns related to AI algorithm training and AI-based chatbots in healthcare. FINDINGS: The study revealed varying levels of AI literacy among medical students with a basic understanding of AI principles. Perceptions regarding AI's role in healthcare varied, with recognition of key benefits such as improved diagnosis accuracy and enhanced treatment planning. Students relied predominantly on online resources to stay informed about AI. Concerns included bias reinforcement, data privacy and over-reliance on technology. ORIGINALITY/VALUE: This study contributes original insights into medical students' AI literacy and attitudes, highlighting the need for targeted educational interventions and ethical considerations in AI integration within medical education and practice.

Students, Medical↗

PubMed2026

Factors influencing the enhancement of the new iron triangle in healthcare organisations.

PURPOSE: A new paradigm, "healthcare's new iron triangle," has been developed to emphasise the technological perspective of healthcare delivery, focusing on automation, value and empathy. The study aims to build a conceptual model and to identify factors for the enhancement of the new iron triangle in healthcare organisations. DESIGN/METHODOLOGY/APPROACH: The healthcare organisation is the primary focus point of the current study. To determine the factors, a survey of the literature and healthcare experts' opinions was conducted. The healthcare professionals validated the identified factors. Data for this study were gathered using a closed-ended questionnaire and scheduled interviews. The study employed "Total Interpretive Structural Modeling methodology and Matriced' Impacts Croise´s Multiplication Appliqué´ a UN Classement/Cross-Impact Matrix Multiplication Applied to a Classification (MICMAC) analysis" to address the "why" and "how" the factors interact and prioritise the identified factors. FINDINGS: The study found that organisational structure (F8), artificial intelligence (F1), innovation (F2) and human resources (F5) are the driving or key factors of the study. RESEARCH LIMITATIONS/IMPLICATIONS: The study primarily focused on identifying factors for the enhancement of a new iron triangle in healthcare organisations. The scope could eventually be expanded to explore more areas. PRACTICAL IMPLICATIONS: Academics and other stakeholders will have a better understanding of the key drivers for the enhancement of the new iron triangle in healthcare organisations. ORIGINALITY/VALUE: In this study, total interpretive structural modeling and cross-impact MICMAC analysis are proposed as an innovative approach to address the new iron triangle in healthcare organisations.

Humans↗

PubMed2026

The future of precision oncology and artificial intelligence in Belgium: scenarios and policy responses.

PURPOSE: Precision medicine, also known as personalized medicine, enables the provision of tailored health services to patients. In the prevention, early detection, and treatment of cancers, precision medicine is highly promising, given the increasing use of genomic profiling for diagnosis and adapting therapies in several tumor types. Artificial Intelligence (AI) can support this process by analyzing vast amounts of relevant data. However, high-quality data and financial investments in the health system are essential for the implementation of precision medicine and AI solutions in routine cancer care. DESIGN/METHODOLOGY/APPROACH: Building on the quantitative outcomes of a foresight exercise published in another study, this article collects qualitative data to gain more detailed insights into the future of precision oncology in Belgium and discusses the role of AI in this field. It reports the results of a series of expert workshops, focusing on four hypothetical future scenarios that are centered around technological and economic issues that must be overcome for the widespread use of precision oncology in Belgium. FINDINGS: The study concludes that all four scenarios discussed in the workshops would require supportive policy measures in Belgium, which should go beyond mere technological and economic considerations, such as involving patient associations and the public in policy design or creating multi-disciplinary expert groups for precision medicine. ORIGINALITY/VALUE: To the best of our knowledge, this is the first study to employ foresight methodology to illustrate possible future scenarios, scrutinize feasible approaches for implementing precision oncology in Belgium, and discuss the use of AI in this context.

Belgium↗

PubMed2026

From fear to empowerment: the impact of employees AI awareness on workplace well-being - a new insight from the JD-R model.

PURPOSE: The primary purpose of the study was to explore the impact of health workers' awareness of artificial intelligence (AI) on their workplace well-being, addressing a critical gap in the literature. By examining this relationship through the lens of the Job demands-resources (JD-R) model, the study aimed to provide insights into how health workers' perceptions of AI integration in their jobs and careers could influence their informal learning behaviour and, consequently, their overall well-being in the workplace. The study's findings could inform strategies for supporting healthcare workers during technological transformations. DESIGN/METHODOLOGY/APPROACH: The study employed a quantitative research design using a survey methodology to collect data from 420 health workers across 10 hospitals in Ghana that have adopted AI technologies. The study was analysed using OLS and structural equation modelling. FINDINGS: The study findings revealed that health workers' AI awareness positively impacts their informal learning behaviour at the workplace. Again, informal learning behaviour positively impacts health workers' workplace well-being. Moreover, informal learning behaviour mediates the relationship between health workers' AI awareness and workplace wellbeing. Furthermore, employee learning orientation was found to strengthen the effect of AI awareness on informal learning behaviour. RESEARCH LIMITATIONS/IMPLICATIONS: While the study provides valuable insights, it is important to acknowledge its limitations. The study was conducted in a specific context (Ghanaian hospitals adopting AI), which may limit the generalizability of the findings to other healthcare settings or industries. Self-reported data from the questionnaires may be subject to response biases, and the study did not account for potential confounding factors that could influence the relationships between the variables. PRACTICAL IMPLICATIONS: The study offers practical implications for healthcare organizations navigating the digital transformation era. By understanding the positive impact of health workers' AI awareness on their informal learning behaviour and well-being, organizations can prioritize initiatives that foster a learning-oriented culture and provide opportunities for informal learning. This could include implementing mentorship programs, encouraging knowledge-sharing among employees and offering training and development resources to help workers adapt to AI-driven changes. Additionally, the findings highlight the importance of promoting employee learning orientation, which can enhance the effectiveness of such initiatives. ORIGINALITY/VALUE: The study contributes to the existing literature by addressing a relatively unexplored area - the impact of AI awareness on healthcare workers' well-being. While previous research has focused on the potential job displacement effects of AI, this study takes a unique perspective by examining how health workers' perceptions of AI integration can shape their informal learning behaviour and, subsequently, their workplace well-being. By drawing on the JD-R model and incorporating employee learning orientation as a moderator, the study offers a novel theoretical framework for understanding the implications of AI adoption in healthcare organizations.

Humans↗

PubMed2026

Improving insurance deduction identification: a hybrid artificial intelligence model using machine learning and expert systems.

PURPOSE: Financial challenges in healthcare systems worldwide, especially in low- and middle-income countries like Iran, have increased hospitals' reliance on insurance reimbursements. Unrecognized insurance deductions often cause severe financial shortages, making efficient deduction management crucial. This study aimed to design a hybrid intelligent system for identifying and predicting insurance deductions by combining machine learning and expert system frameworks. DESIGN/METHODOLOGY/APPROACH: A mixed-methods design was applied in four stages. First, a scoping review identified the causes and patterns of insurance deductions. Second, interviews with 15 insurance experts produced a validated checklist and a dataset from inpatient billing records. Third, using the CRISP-DM methodology, machine learning algorithms were developed and tested in SPSS Modeler alongside a fuzzy expert system developed in MATLAB. Finally, the model was validated using the holdout method. FINDINGS: Four categories of deduction drivers were identified: service provision, registration errors, document submission issues, and revenue conversion processes. The CHAID decision tree outperformed other algorithms with a 99% precision rate and the lowest Mean Absolute Error (9.43). A brief assessment of potential overfitting was conducted to ensure that the CHAID model's high accuracy was interpreted cautiously and supported by the validation results. The fuzzy expert system with validated rules was adaptable for deduction classification, especially for cases unsuitable for quantitative modeling. ORIGINALITY/VALUE: The hybrid model improves detection and prevention of deductions, offering actionable insights for hospital administrators, insurers, and policymakers. Its implementation can enhance hospital information systems, streamline claims processing, and optimize revenue management amid financial constraints.

Machine Learning↗

PubMed2026

User Engagement and Feature Preferences in an AI-Powered mHealth Intervention for Diabetes Prevention: Secondary Analysis of a Randomized Controlled Trial.

BACKGROUND: Prediabetes is highly prevalent and increasing globally, yet lifestyle interventions remain underused. AI-driven mobile health (mHealth) tools can help scale diabetes prevention efforts, but the key factors driving their success are not well understood. OBJECTIVE: This post hoc secondary analysis of a randomized controlled trial (RCT) aimed to characterize the most valued features and the role of user engagement in outcomes of a fully automated mHealth intervention for diabetes prevention. METHODS: Data from 151 participants with prediabetes and overweight or obesity who were assigned to an AI-based diabetes prevention program (Sweetch) in a parent RCT (NCT05056376) were analyzed. Engagement (defined as the total number of days the app was used) was categorized into tertiles (low, medium, and high). Baseline characteristics were compared across engagement groups using ANOVA, Kruskal-Wallis, and chi-square tests, and regression models assessed the association between engagement and achievement of diabetes risk reduction outcomes (≥5% weight loss, ≥4% weight loss with ≥150 min/week of physical activity, or ≥0.2 percentage point reduction in hemoglobin A1c [HbA1c] at 12 months). Perceived usefulness of intervention features was surveyed at 12 months. RESULTS: Median engagement was 98 (IQR 34-232) days. Older age (P<.001) and lower baseline BMI (P=.04) were significantly associated with higher engagement. Compared with low engagement, high engagement was associated with greater odds of achieving the composite diabetes risk reduction outcome (odds ratio [OR] 2.59, 95% CI 1.11-6.01; P=.03), ≥5% weight loss (OR 3.31, 95% CI 1.16-9.42; P=.03), and ≥0.2 percentage point reduction in HbA1c (OR 3.57, 95% CI 1.19-10.75; P=.02). Participants most frequently rated weight tracking, physical activity tracking, and the digital body weight scale as the features that were most helpful for achieving their health goals. CONCLUSIONS: Higher engagement with an AI-driven intervention requiring no human intervention was associated with improved diabetes risk reduction. Contrary to concerns about lower digital literacy, older adults engaged with the intervention more than younger adults. Features related to weight and physical activity tracking were most valued by patients in the program. TRIAL REGISTRATION: ClinicalTrials.gov NCT05056376; https://clinicaltrials.gov/study/NCT05056376.

Humans↗

PubMed2026

How Following Medical Artificial Intelligence Advice Can Mitigate Malpractice Liability: Cross-National Insights from a Randomized Trial.

Artificial intelligence (AI) increasingly influences clinical decision-making, yet its recommendations may diverge from standard care. Although malpractice concerns are thought to discourage physicians from following AI advice, experimental evidence from the United States suggests the opposite: lay jurors are more likely to hold physicians liable when they reject AI recommendations. Whether this pattern extends to systems in which court-appointed experts, not lay jurors, determine liability remains unknown. Methods: To examine how physicians and laypeople in expert-based and lay-juror legal systems evaluate physicians' acceptance or rejection of AI recommendations, particularly when those recommendations deviate from standard care, we designed a randomized vignette study: a 2 × 2 factorial design varying the AI recommendation (standard vs. nonstandard care) and a fictional physician's decision (accept vs. reject). The study was conducted online in 2023 among nationally representative samples of U.S. and German adults and from 2023 to 2024 among German physicians. In total, 387 German physicians, 2291 U.S. adults, and 2283 German adults participated; those not completing the survey or failing attention checks were excluded per preregistered criteria. Participants were randomly assigned to 1 of 4 vignettes, varying the AI recommendation (standard vs. nonstandard care) and physician's decision (accept vs. reject). The reasonableness of the fictional physician's decision was measured, rated by participants on a Likert scale. Results: Analysis, following preregistered exclusion criteria, included 248 German physicians, 1202 U.S. adults, and 1358 German adults. Physicians accepting standard-care AI recommendations were rated more reasonable than those rejecting them (U.S. laypeople: t = 5.36; 95% CI, 0.45-0.97; P < 0.001; German physicians: t = 2.47; 95% CI, 0.14-1.30; P = 0.02; German laypeople: t = 4.14; 95% CI, 0.27-0.76; P < 0.001). Ratings of physicians accepting versus rejecting AI nonstandard-care recommendations were statistically equivalent. Equivalence was tested at an α-value of 0.05 using a two 1-sided tests procedure, reported with 90% CIs per standard convention (U.S. laypeople: t = -4.90; 90% CI, -0.1 to 0.36; P < 0.001; German physicians: t = -1.76; 90% CI, -0.12 to 0.67; P = 0.04; German laypeople: t = 5.35; 90% CI, -0.35 to 0.06; P < 0.001). Conclusion: Across the United States and Germany, samples representative of lay jurors and court-appointed experts viewed accepting standard-care AI advice as more reasonable, whereas accepting or rejecting nonstandard-care AI advice was judged similarly. Contrary to predictions, malpractice liability regimes do not necessarily pose a barrier to AI use in precision medicine.

Artificial Intelligence↗

PubMed2026

An AI-assisted Clinical Decision Support System for Green Classification of Cystocele on Dynamic Transperineal Ultrasound.

Green classification of cystocele on dynamic transperineal ultrasound (TPUS) remains operator-dependent because it requires manual frame selection and landmark-based assessment of the Valsalva maneuver. We developed a workflow-oriented AI-assisted clinical decision support system for automated urethrovesical junction localization and dynamic Green classification and prospectively evaluated its standalone and reader-support performance. This diagnostic accuracy and reader study included 881 patients from a tertiary referral hospital, comprising a retrospective development cohort (n = 688) and an independent prospective test cohort (n = 193). A nested subset of 67 prospective patients was used for a reader study involving two junior and two intermediate radiologists under unaided and AI-assisted conditions. In the complete prospective test cohort, Green-AttGRU achieved a macro-averaged AUC of 0.939 (95% CI, 0.897-0.971) and an overall accuracy of 0.902 (95% CI, 0.860-0.943). In the reader study, overall accuracy increased from 0.761 to 0.821 without AI to 0.851-0.881 with AI, while macro-F1 increased from 0.660 to 0.777 to 0.820-0.860. Overall inter-reader agreement increased from a Fleiss' κ of 0.453 to 0.786, and pooled median interpretation time decreased from 26.7 s to 9.9 s. These findings support the preliminary feasibility of the system as a workflow-oriented decision-support tool for dynamic TPUS interpretation.

Humans↗

PubMed2026

Automated CEAP Classification of Venous Duplex Reports Using Multimodal Artificial Intelligence.

OBJECTIVE: To develop and internally validate a prototype multimodal artificial intelligence system for automated CEAP (Clinical, Etiological, Anatomical and Pathophysiological) classification of venous duplex ultrasound (VDUS) reports, integrating natural language processing of free-text components with computer vision analysis of hand-drawn anatomical diagrams. METHODS: Single centre retrospective observational study using routinely collected clinical data. One thousand consecutive venous duplex ultrasound reports from Cambridge University Hospitals NHS Foundation Trust, UK (July 2024 - May 2025) were labelled according to the CEAP classification, excluding the Etiological component, which could not be reliably determined from duplex reports alone. Transfer learning was applied using ClinicalBERT for text and MobileNetV3 for diagrammatic data. Clinical classes were predicted from request line text. Text- and image-based pathophysiological models were developed for four anatomical territories (Great Saphenous Vein, Small Saphenous Vein, Deep system, Perforators), combined using late fusion with probability averaging. RESULTS: The clinical CEAP model achieved accuracy of 0.91, macro-F1 of 0.82, and macro-AUC of 0.98. Pathophysiological prediction varied, with text models broadly outperforming image models. Fusion yielded heterogeneous benefits, improving SSV performance but reducing Deep system accuracy. The performance of the final pathophysiological CEAP fusion models varied across anatomical territories: accuracy ranged from 0.70-0.92 and macro-AUC from 0.80-0.92. CONCLUSION: This study demonstrates the feasibility of automated CEAP classification from VDUS reports. Despite class imbalance affecting minority class predictions, the strong discriminatory performance validates this multimodal ML model for extracting clinically meaningful information from real-world data. This approach offers potential, pending external validation, to streamline vascular services through automated triage and guideline-compliant decision making.

Artificial intelligence↗