SEARCH · Search PubMed

Results for “Machine learning algorithms”

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.

193 records · Page 7Linked to original sources

PubMed2026

Teaching Acute Coronary Syndrome High-Risk ECG Interpretation and Clinical Decision-Making Through FOAMed Videos and Podcast Versus Print-Based Materials Among Emergency Care Providers: Randomized Controlled Mixed Methods Trial.

BACKGROUND: Accurate interpretation of high-risk acute coronary syndrome (ACS) electrocardiograms (ECGs) is essential for early diagnosis and timely reperfusion, yet substantial deficits persist across health care professions. Digital self-learning formats such as FOAMed (Free Open Access Medical Education) are widely used, but their effectiveness has rarely been evaluated for complex, high-risk ACS ECG patterns. Existing ECG education studies often focus on students or single professional groups and established ST-segment elevation myocardial infarction (STEMI) criteria, leaving newer guideline-recognized STEMI equivalents, selected emerging occlusion myocardial infarction (OMI)-related patterns, and interprofessional emergency care underrepresented. OBJECTIVE: This study aimed to compare the effectiveness of FOAMed podcast and videos versus traditional print-based materials for teaching high-risk ACS ECG patterns and related clinical decision-making in emergency providers. METHODS: We conducted a prospective, interprofessional, controlled mixed methods trial across 5 training sites in Germany. Paramedics, prehospital emergency physicians, and emergency department clinicians received either a FOAMed multimedia module or print-based materials through concealed allocation; deviations from the intended 1:1 ratio resulted from participant no-shows. The intervention consisted of a 30-minute supervised self-learning session. In total, 103 participants were allocated to FOAMed (n=45) or print-based materials (n=58). Two coprimary outcomes were assessed: ECG interpretation accuracy and text-based ACS clinical decision-making. Secondary outcomes included subjective confidence, learning experience, and exploratory qualitative free-text responses. Outcome assessment was automated and blinded; mixed ANOVA was the primary analysis. The study was not prospectively registered because it assessed educational outcomes in health care professionals rather than patient health outcomes. RESULTS: All 103 participants completed the study. Both groups improved, with greater gains in the FOAMed group: ECG interpretation increased from 55% to 65.5% and text-based ACS clinical decision-making from 45% to 68%, versus 57% to 60% and from 47% to 63%, respectively, in the print-based group. Effect sizes were η²=0.055 for ECG interpretation and η²=0.044 for clinical decision-making. Exploratory subgroup analyses provided no evidence of differential effects across age, gender, or professional background and were likely underpowered. Qualitative responses (46 and 37 entries) provided contextual insights into perceived clarity, engagement, and practical relevance supporting the quantitative findings. CONCLUSIONS: This study is innovative in directly comparing a curated FOAMed multimedia module with selected print-based materials in an interprofessional emergency care population. It differs from existing research by focusing on subtle, emerging ischemic patterns and evaluating realistic, time-limited self-learning formats. The findings provide evidence that curated FOAMed resources can produce greater short-term improvements in ECG interpretation and text-based ACS clinical decision-making than traditional print-based materials in this setting. Although implications for clinical performance remain hypothetical, concise, high-quality digital modules may represent a practical supplement to structured continuing education in emergency care.

Humans↗

PubMed2026

Artificial intelligence for dental caries detection: An umbrella review.

Artificial intelligence (AI) has been proposed as a tool to improve dental caries detection across imaging modalities; however, its clinical value remains uncertain. This umbrella review aimed to synthesize and critically appraise systematic reviews evaluating AI for caries detection and diagnosis. An umbrella review was conducted following PRIOR guidance (PROSPERO CRD420261340728). Searches were performed in MEDLINE, Embase, Scopus, Web of Science, and Google Scholar up to 15 March 2026. Methodological quality was assessed using AMSTAR 2, and overlap of primary studies was quantified using the corrected covered area (CCA). Seventeen systematic reviews were included, of which five reported diagnostic test accuracy meta-analyses using bivariate or HSROC models. Across these meta-analyses, pooled sensitivity ranged from 0.76 to 0.94 and specificity from 0.85 to 0.91. Most systems were based on deep learning models applied to bitewing radiographs and intraoral photographs. However, substantial heterogeneity was observed in imaging modalities, lesion thresholds, analytical tasks, and evaluation metrics. In addition, a high degree of overlap across reviews and recurrent methodological limitations, including reliance on retrospective datasets, limited external validation, and inconsistent reporting, substantially weaken the reliability of the evidence. Although AI models demonstrate high diagnostic performance under experimental conditions, current evidence does not support their use as stand-alone diagnostic tools. Their clinical applicability remains limited, and implementation should be restricted to decision-support contexts until robust prospective validation demonstrates meaningful impact on clinical decision-making and patient outcomes.

Dental Caries↗

PubMed2026

Multimodal alignment improves generalizability of genomic biomarker prediction in computational pathology.

Computational pathology models that use digitized histopathology whole-slide images have the potential to become a cost-effective and scalable alternative to molecular assays for the prediction of genomic biomarkers, a key task in precision oncology. However, as new genomic biomarkers are discovered or quantified, large, labeled datasets must be prospectively collected to train new models. To address this challenge, we developed multimodal alignment for biomarker learning and generalization (MARBLE), a multimodal contrastive pretraining strategy that integrates structured biomarker knowledge into representation learning of histopathology images. MARBLE aligns histopathology-derived representations with representations of genomic biomarkers generated by a large language model (LLM) and a protein language model (PLM). This biologically informed alignment enables data-efficient generalization to novel, out-of-distribution biomarkers. Using the MSK-IMPACT cohort of over 40,000 patients across multiple biomarker panel versions, we design experiments grounded in real-world data to demonstrate the value of our proposed approach.

CP: computational biology↗

PubMed2026

One Year After a Cyberattack: Lessons Learned and Dosimetric Analysis of Contingency Radiotherapy Plans.

PURPOSE: Cyberattacks on health care institutions pose significant risks to patient care, particularly in radiotherapy departments, which are heavily reliant on digital systems. This study examines the impact of a ransomware attack on our hospital and evaluates the effectiveness of the contingency measures implemented to resume radiotherapy treatments. METHODS AND MATERIALS: Following the cyberattack, our radiotherapy department faced a complete shutdown. After an initial estimate considering a shutdown of several weeks, a contingency plan was executed, including manual patient data retrieval and collaboration with a backup hospital. Contingency plans were prepared and delivered within hours, despite a partial lack of information. These plans allowed some patients to restart treatment 3 days after the attack. A dosimetric analysis was performed for the contingency plans, including various pathologies, mainly glioblastoma, head and neck cancers, and lung cancer. We compared the original and contingency plans in terms of dose coverage to the clinical target volume, biological effective dose, and their clinical impact as assessed at the 1‑year follow‑up after the cyberattack. RESULTS: Treatments resumed within 12 days at our hospital. Patients with glioblastoma showed good target coverage because of generous margins, resulting in favorable outcomes. In head and neck cases, the lack of detailed imaging led to significant target volume misses, suggesting that more conservative initial treatments could have been beneficial. Lung cases demonstrated accurate peripheral lesion targeting but faced challenges in central lesions because of the absence of positron emission tomography information. In most cases, the approach of using a contingency plan, even with limited information, led to a higher biological effective dose than would have been achieved if treatment had been stopped until full recovery at our hospital. CONCLUSIONS: The study highlights the critical importance of robust contingency planning in radiotherapy departments, emphasizing the need for backup systems and tailored approaches based on tumor location and available diagnostic information. These lessons emphasize that preparedness for digital disruptions should not focus exclusively on information and technology infrastructure.

Humans↗

PubMed2026

Structured robotic colorectal training in a non-tertiary NHS hospital: a 502-case consecutive cohort implementation study.

Robotic-assisted colorectal surgery has expanded rapidly across NHS practice in the UK. Structured unit-wide training pathways are essential for safe technology adoption, yet published outcome data from non-tertiary hospitals remain limited. This study describes the implementation and feasibility of a unit-wide robotic colorectal program at a high-volume non-tertiary hospital, reporting outcomes across 502 consecutive resections performed by eight consultant surgeons and presenting these in the context of nationally published benchmarks. A retrospective cohort study of 502 consecutive robotic colorectal resections performed at York Teaching Hospital between May 2022 and December 2025. Eight consultant surgeons (A-H) participated in a structured four-phase training pathway incorporating simulation training, proctored cases, complexity-based case progression, and formal credentialing. Primary outcomes were 30-day mortality, unplanned return to theatre (RTT), and anastomotic leak (AL). Anastomotic leak was calculated using only patients who underwent anastomosis as the denominator. Procedure-stratified and individual surgeon outcomes with 95% confidence intervals were reported. Risk-adjusted cumulative sum (RA-CUSUM) analysis was performed to evaluate learning curves. Outcomes are presented descriptively alongside nationally published reference data; no formal statistical comparison against national benchmarks was performed. 502 robotic colorectal resections were performed. Mean patient age was 70.0 ± 11.3 years; 58.4% were male. Median ASA grade was III. The indication was malignancy in 89.2% of cases. Length of stay was non-normally distributed and is therefore reported using median and interquartile range in the revised analysis. Key outcomes: - 30-day mortality: 1.0% (5/502; 95% CI 0.4-2.3%) - Unplanned return to theatre (RTT): 5.2% (26/502; 95% CI 3.6-7.5%) - Anastomotic leak (AL): 3.3% (15/450; 95% CI 2.0-5.5%; denominator = patients with anastomosis) - 30-day unplanned readmission: 5.0% (25/502; 95% CI 3.4-7.2%) - Conversion to open surgery: 3.6% (18/502; 95% CI 2.3-5.6%) - Lymph node yield ≥12: 91.3% of cancer resections - R0 resection rate: 95.1% of cancer resections All primary outcomes fell within or below the published reference ranges used for descriptive context. RA-CUSUM trajectories were heterogeneous: no surgeon crossed the predefined upper control limit, but several curves showed later upward movement. Accordingly, the analysis is interpreted as safety surveillance rather than evidence of uniform performance improvement. RA-CUSUM monitoring showed that no surgeon crossed the predefined upper control limit; however, heterogeneous trajectories precluded a claim of uniform performance improvement.

Humans↗

PubMed2026

Testing How Mindfulness Skills Change for Novice Meditators Using Headspace: Examining Trait Mindfulness and Perceived Stress as Moderators.

Mindfulness-based interventions are found to effectively reduce stress and improve mental health outcomes. Yet, it is not always clear how the mindfulness skills of attention and acceptance develop throughout the intervention. This knowledge gap is especially pertinent for novice meditators learning these skills for the first time, including whether some individuals are more prone to learning them. Using a randomized waitlist-controlled trial, we tested the effect of the app Headspace on changes in attention and acceptance over 8 weeks among participants new to mindfulness meditation. Further, we tested the moderating effects of trait mindfulness and perceived stress. Non-faculty university employees were randomized to a Headspace or waitlist control condition. Trait mindfulness and perceived stress were measured at baseline. Ecological momentary assessment survey data for attention and acceptance were collected five times a day in 4-day bursts at baseline and 2, 5, and 8 weeks post-randomisation. Attention and acceptance were significantly higher at Week 8 compared to baseline for the Headspace group, but not the control group. For the Headspace group, both skills showed significant change by Week 2. Trait mindfulness moderated this effect with those who were lower in trait mindfulness displaying greater increases in attention, but not acceptance. Perceived stress also moderated this effect with those who were lower in perceived stress displaying greater increases in attention and acceptance. Our discussion draws attention to implications for matching intervention content to individual needs to ensure participants reporting different levels of characteristics benefit from mindfulness training.

Humans↗

PubMed2026

Efficacy of a self-guided online resilience intervention for improving mental health among university students: A randomized controlled trial.

Epidemiological data indicates that university students are an at-risk population for the development of mental disorders. Online interventions have been proposed as promising tools for reducing barriers to treatment and establishing easily accessible health-care services promoting mental health and resilience. This study investigated the efficacy of a Learning Management System (LMS)-based self-guided online resilience intervention. 216 university students took part in a randomized controlled trial with an intervention and a waitlist control group and three measurement points (pre, post and follow-up). We conducted per-protocol (PP) and intention-to-treat (ITT) analyses with mental distress as primary outcome, and self-reported resilience and resilience factors as secondary outcomes. Further, attitudes towards online interventions, adherence, satisfaction and possible negative effects were explored. Satisfaction with the intervention was high and PP analyses (n = 150) revealed significant improvements in mental distress and self-compassion at post-test and acceptance at follow-up. No favourable effects were found for self-reported resilience and resilience factors. Adherence was low and ITT analyses revealed no significant effects. Overall, the study provides preliminary evidence for the LMS-based self-guided online intervention as a potentially valuable tool for university mental health services under optimal adherence conditions. Further research into determinants of adherence is needed to improve intervention reach.

Humans↗

PubMed2026

The effects of visuomotor training and tDCS stimulation on visuomotor integration and visual processing: an electrophysiological approach.

BACKGROUND: Visuomotor integration coordinates visual and motor cortical activity to produce goal-directed responses and can be indexed by Rolandic Mu-rhythm suppression and visual evoked potential (VEP) P100 parameters. Perceptual-motor training improves visuomotor performance, and transcranial direct current stimulation (tDCS) over primary motor cortex (M1) has been reported to enhance motor learning when paired with training. This study examined whether anodal M1 tDCS augments the effects of Senaptec visuomotor training in healthy adults. METHODS: Sixty participants were randomized to active anodal tDCS (five 10-minute sessions, 1 mA; n = 31) or sham (n = 29) over M1 immediately before each Senaptec training session; 53 completed all sessions and post-testing. Outcomes were Mu-suppression ratios, VEP P100 latency and amplitude, and Senaptec measures of visual sensitivity and visuomotor control. RESULTS: Active tDCS produced no augmentation of any outcome, with no significant group × time interaction for any measure, consistent across composite and task-level analyses. Training alone produced no change in Mu suppression or visuomotor control. By contrast, both groups showed significant training-related gains in visual sensitivity, including near-far quickness and stereopsis, accompanied by shorter P100 latencies and larger amplitudes, indicating more efficient early visual processing. CONCLUSIONS: A clear dissociation emerged: training produced robust improvements in early visual processing, whereas neither tDCS nor training altered sensorimotor (Mu) or visuomotor-control measures. The tDCS results should be interpreted cautiously given the modest dose and limited power to detect small effects, rather than as evidence of inefficacy. Tablet-based perceptual training enhanced visual processing independent of neuromodulation.

Humans↗

PubMed2026

Effects of anodal transcranial direct current stimulation over the right primary motor cortex on a sequential motor finger tapping task in developmental stuttering.

INTRODUCTION: This study investigates the impact of anodal transcranial direct current stimulation (tDCS) on non-speech sequential motor practice in adults who stutter (AWS), compared to non-stuttering controls (ANS). Recent research has explored the effects of tDCS on speech fluency in stuttering. However, its effect on non-speech motor tasks has not yet been studied. METHODS: 20 AWS and 30 ANS right-handed participants were randomly assigned to anodal or sham tDCS conditions, performing a sequential finger tapping task. We targeted over the right primary motor cortex, stimulating at 2 mA for 20 min. Sequence duration and reaction time were analyzed. RESULTS: AWS analysis revealed that the anodal condition had significantly slower reaction times in the second half of the task compared to sham. For sequence durations, AWS in the anodal condition had slower overall sequence durations than the sham condition. However, there were no block-by-block differences in sequence duration. When comparing AWS and ANS, no significant differences were observed for sequence duration. However, there were significant differences in reaction time between AWS and ANS, specifically in earlier blocks. Additionally, there was no significant Group × Condition interaction. DISCUSSION: The findings suggest that anodal stimulation impeded finger sequencing in AWS, showing overall slower sequence durations and a diminishing effect on reaction times in the second half of the experiment, suggesting anodal tDCS may interact uniquely with the neural mechanisms in stuttering. Future studies should explore the effects of anodal tDCS on non-speech motor tasks to gain a broader understanding of its impact on motor control and motor learning.

Humans↗

PubMed2026

Mining Stored-Specimen Studies for Information about Cancer Natural History.

The advent of new multicancer early detection tests and publication of early diagnostic results have generated expectations of clinical benefit from multicancer screening. The clinical benefit of a cancer screening test depends critically on disease natural history, which is typically learned from prospective screening studies. Retrospective studies of stored blood specimens are important in learning about a test's preclinical diagnostic performance but have rarely been used to infer natural history. The extent to which these studies might be harnessed to also learn natural history is discussed in the context of an article in this issue that infers the combined natural history of a range of cancers targeted by a multicancer early detection test using a case-control subsample of specimens from a large cohort study. The critical question concerns the identifiability of key transition rates in multistate models of natural history alongside state-specific sensitivities. The article suggests that these parameters are estimable within a Bayesian framework that leverages prior information about test sensitivity from diagnostic studies. We offer a heuristic discussion of identifiability in this setting and encourage formal study to determine the extent to which models with varying degrees of complexity may be learned from stored-specimen studies. See related article by Dai et al., p. 1535.

Humans↗

PubMed2026

Using the OPTIMAL Theory to Optimize Aerodynamics in Respiratory Training for Healthy Adults and Individuals With Parkinson's Disease.

BACKGROUND: The OPTIMAL (Optimizing Performance Through Intrinsic Motivation and Attention for Learning) theory is a motor learning framework proposing that optimizing intrinsic motivation enhances motor performance and learning. The theory identifies three key components-Enhanced Expectancies (EE), Autonomy Support (AS) and External Focus of Attention (EF)-which facilitate more efficient, goal-directed movement. These components have been shown to improve motor outcomes in limb-based tasks; however, their application to respiratory training, particularly in clinical contexts such as voice and swallowing therapy in patients with Parkinson's disease (pwPD), has not yet been systematically explored. AIMS: This study aimed to investigate whether implementing OPTIMAL theory strategies during a respiratory muscle strength training (RMST) task improves immediate respiratory motor performance in healthy adults and pwPD. Additionally, we aimed to examine the effects of these strategies on motivation and cognitive engagement. METHODS: This quasi-randomized, single-session trial included 47 participants: Healthy CONTROL (n = 17), Healthy OPTIMAL (n = 16) and PD OPTIMAL (n = 14). Healthy participants were quasi-randomly assigned to either intervention or control conditions, whereas pwPD completed the intervention only. All participants completed a single respiratory session that included baseline, practice and retention phases. Outcome measures included peak expiratory flow, cough peak expiratory flow, cognitive engagement (EEG-based Cognitive Engagement Index) and self-administered motivation questionnaire. OUTCOMES AND RESULTS: Exhalation force improved from baseline to retention in the Healthy OPTIMAL group (baseline: M = 296 L/min; retention: M = 338 L/min; p < 0.001) and the PD OPTIMAL group (baseline: M = 315 L/min; retention: M = 370 L/min; p < 0.0001), but not in the Healthy CONTROL group (p > 0.05). No significant changes in cough strength were observed in any group. No correlations were found between cognitive engagement and exhalation force or motivation scores. However, motivation increased more in the Healthy OPTIMAL group (Questionnaire 1: M = 57.2; Questionnaire 2: M = 60.7) and the PD OPTIMAL group (Questionnaire 1: M = 60.1; Questionnaire 2: M = 62.8) than in the Healthy CONTROL group (Questionnaire 1: M = 61.1; Questionnaire 2: M = 62.5). CONCLUSIONS AND IMPLICATIONS: Implementing the OPTIMAL theory enhances immediate respiratory motor performance in both healthy participants and pwPD. OPTIMAL theory has clinical value in voice and swallowing therapy, although further research is needed to establish long-term efficacy and clinical impact. WHAT THIS PAPER ADDS: What is already known on the subject Motivation is a critical factor in rehabilitation. The OPTIMAL theory has been shown to improve both motivation and motor performance in limb-based tasks. Its impact on respiratory training, however, has not been previously examined. What this paper adds to the existing knowledge This study shows that applying OPTIMAL strategies during a respiratory muscle strength training task significantly improved peak expiratory flow in both healthy adults and people with Parkinson's disease. What are the potential or clinical implications of this work? Integrating the OPTIMAL theory principles into respiratory therapy may enhance motor outcomes, supporting voice, swallowing and cough rehabilitation.

Humans↗

PubMed2026

What Constitutes Effective Support and Provision Within Day Service Centres for People With Intellectual Disabilities? A Systematic Review of Qualitative Research.

BACKGROUND: This review aimed to investigate the effectiveness and quality of support and provision within day service centres for people with intellectual disabilities. METHOD: The International Bibliography of the Social Sciences, Scopus and PsycInfo databases were searched in August 2024, and the results were reported according to the PRISMA guidelines. Peer-reviewed, English-language, qualitative studies that investigated the effectiveness of day service provision for people with intellectual disabilities in non-residential settings were considered for review. Methodological quality of the included studies was assessed using the JBI Critical Appraisal Tool for qualitative research. Qualitative themes were identified through thematic analysis and synthesised using the ConQual approach. RESULTS: Fourteen studies were included and four key themes emerged: 'perceptions of service quality'; 'community-orientation, integration, and empowerment'; 'challenging behaviours and safety'; and 'staff-centred factors and job satisfaction'. Confidence in the evidence was 'very low' for 3/4 themes, while there was 'moderate' confidence in the evidence related to the theme 'perceptions of service quality'. CONCLUSIONS: Day service centres for people with intellectual disabilities may enhance their effectiveness and quality of provision by concentrating on promoting communication, engagement, relationships, social networks and community integration. Addressing the methodological shortcomings and incomplete reporting of related research in future would contribute to improvements in overall confidence in the evidence base. This can then be better used to inform and further enhance day service provision for people with intellectual disabilities.

Humans↗

PubMed2026

Bioprospecting microbial genomes to expand the biocatalytic toolbox of rubber oxygenases.

A set of rubber oxygenases was discovered through phylogenetic analysis and AI-based structural modeling of complexes of the putative enzymes with a substrate mimicking cis-1,4-polyisoprene. Sixteen candidate proteins were selected from thermophilic microorganisms, all sequence-related to the Latex clearing protein from Streptomyces sp. K30 (LcpK30). Sequence truncation and solubility tags were then evaluated to enhance protein expression, with the SUMO tag proving to be the most effective. Including LcpK30, nine heme-containing oxygenases were successfully expressed in E. coli NEB 10-beta cells, purified (35-157 mg L-1 yield) and characterized. Steady-state kinetics revealed significant rubber latex-degrading properties for six of them, with the truncated SUMO-fused LcpK30 (SUMO-LcpK30T) showing activity in agreement with literature. Notably, the catalytic efficiencies of all the expressed homologs lay within one order of magnitude and the oxygenase from Thermomonospora echinospora was found to be particularly promising in terms of activity, especially at high latex concentrations (more than 1% w/v). The analysis of reaction mixtures by both HPLC and HPLC-MS confirmed the oxidation of cis-1,4-polyisoprene to form the expected isoprenoid oligomers (n = 2-12), whose distribution was consistent with the usual endo-type cleavage pattern in all but one case. This bioprospecting effort afforded a platform of new rubber-degrading enzymes with diverse efficiencies and product profiles, capable of adapting to targeted applications.

Oxygenases↗

PubMed2026

AI echo INSIGHT study: A prospective blinded randomized trial of artificial intelligence echocardiogram interpretation.

BACKGROUND: Transthoracic echocardiography (TTE) is the most commonly performed cardiac imaging modality with over 30 million studies annually. Demand for timely expert interpretation continues to outpace capacity, creating diagnostic delays and inter-observer variability that impact patient care. Recent research has suggested computer vision artificial intelligence (AI) models can generate accurate preliminary comprehensive TTE reports, however, prospective evaluation is needed to determine whether AI-assisted TTE interpretation can improve clinician efficiency while preserving diagnostic accuracy. METHODS: AI ECHO INSIGHT is a prospective randomized blinded clinical trial conducted at Kaiser Permanente Northern California that will evaluate 1200 historical TTE studies (1000 consecutive unselected studies plus 200 with moderate or greater valvular disease) interpreted using three workflows: (1) AI-generated preliminary report finalized by a blinded cardiologist (AI-assisted); (2) cardiologist-generated preliminary report finalized by a blinded cardiologist (cardiologist-assisted); and (3) sonographer-generated preliminary report finalized by a blinded cardiologist (sonographer-assisted). The primary outcome is the rate of substantial change between preliminary and final reports, comparing the AI-assisted workflow to the pooled cardiologist-assisted and sonographer-assisted workflows. Secondary outcomes include cardiologist interpretation time for report finalization, superiority testing for diagnostic accuracy, and reporting consistency. CONCLUSION: AI ECHO INSIGHT is a prospective randomized blinded clinical trial evaluating the clinical impact of AI-assisted TTE interpretation on diagnostic accuracy, cardiologist efficiency, and reporting consistency in real-world echocardiography workflows. TRIAL REGISTRATION: ClinicalTrials.gov registration number NCT07229300.

Humans↗

PubMed2026

Comparison of summative assessments between simulated electronic health records versus traditional paper-based patient cases: A non-inferiority randomized controlled trial.

INTRODUCTION: Electronic health records are fundamental to contemporary pharmacy practice, yet evidence supporting their use in pharmacy education is lacking. This single-center, non-inferiority randomized controlled trial with blinded outcome assessment evaluated whether delivering patient cases via a simulated academic EHR (aEHR) was non-inferior to a traditional paper-based format in student exam performance. METHODS: 53 third-year PharmD students at the University of British Columbia were randomized 1:1 to complete a mock summative examination using either the aEHR or paper-based case delivery, stratified by self-reported EHR comfort level. The primary outcome was mean written exam score (%). Non-inferiority was pre-specified at a margin of 14%. Adjusted linear regression was used for the primary analysis, with a multiple imputation sensitivity analysis. Student perceptions were explored through post-exam focus groups analyzed using inductive thematic analysis. RESULTS: 42 students (21 per group) completed the exam and were included in the primary analysis. Mean scores were 66% (SD 11) in the aEHR group and 68% (SD 10) in the paper group. The adjusted mean difference (paper minus aEHR) was -2.2% (95% CI -9.2% to +4.8%), satisfying non-inferiority but not superiority. Sensitivity analysis (n = 53) yielded consistent results (-2.3%; 95% CI -7.1% to +4.1%). Focus groups revealed initial student anxiety with the aEHR but recognized its alignment with clinical practice. DISCUSSION: These findings support the feasibility of integrating simulated EHRs into summative pharmacy assessments without compromising performance. CONCLUSION: Simulated EHRs are a non-inferior assessment medium compared with paper-based formats and represent a viable step toward technology-driven pharmacy practice environments.

Humans↗

PubMed2026

Pedagogical Efficacy of LLM-Generated Synthetic Data Versus Real-World Clinical Records: A Randomized Controlled Non-Inferiority Trial.

BACKGROUND: Expert-reviewed clinical cases generated by large language models (LLMs) may supplement case resources in medical education, but their short-term educational performance relative to real-case-derived teaching materials remains uncertain. We compared immediate post-training test performance after teaching with the two types of case materials and assessed non-inferiority against a prespecified margin. METHODS: We conducted a prospective, parallel-group, randomized non-inferiority trial. Through the Wenjuanxing online platform, participants were randomized 1:1 to learn with either real-case-derived teaching cases compiled by clinicians and reviewed by experts or AI-generated clinical cases produced by Gemini 3.0 Pro from fully de-identified matched real cases and reviewed by three senior general surgery specialists with full-professor rank. The primary outcome was the total score on an independent 10-item immediate post-training test (0-10 points), with a prespecified non-inferiority margin of -0.5 points. Secondary outcomes included the training-phase performance score, learning efficiency index, single-item mental effort rating, case realism, and case-source judgment. RESULTS: A total of 403 participants were randomized, of whom 386 were included in the modified intention-to-treat analysis: 192 in the real-case group and 194 in the AI-generated case group. The mean post-training test score was 4.95 (SD, 3.35) in the real-case group and 4.61 (SD, 3.35) in the AI-generated case group. The mean difference (AI-generated minus real-case group) was -0.335 points (95% CI, -1.006 to 0.337). Because the lower bound of the confidence interval was below the prespecified non-inferiority margin of -0.5 points, non-inferiority was not demonstrated (one-sided P = 0.314). No significant between-group differences were observed in the training-phase performance score, learning efficiency index, or single-item mental effort rating. AI-generated cases received lower realism ratings for Level 3 cases. The proportion of participants with at least one high-confidence completely incorrect response was 1.6% in the real-case group and 2.1% in the AI-generated case group. CONCLUSIONS: In this short-term, text-based online case-learning setting, no statistically significant between-group difference was observed in immediate post-training test performance; however, non-inferiority of AI-generated clinical cases relative to real-case-derived teaching materials was not demonstrated.

Humans↗

PubMed2026

A comparative systematic review of pharmacist education systems and pharmacy service quality in ASEAN-5: Indonesia, Malaysia, Thailand, the Philippines, and Singapore.

BACKGROUND: The global transition toward patient-centered pharmaceutical care has exposed structural disparities in ASEAN pharmacy workforce training and deployment. This review examines four research questions: how pharmacy education systems and accreditation standards differ across Indonesia, Malaysia, Thailand, the Philippines, and Singapore (collectively, the ASEAN-5); the extent to which pre-registration education influences clinical service scope and professional confidence; how education reform and regulatory change have shaped pharmacist clinical roles; and what barriers and enablers exist for regional qualification harmonization. METHODS: A systematic literature review following PRISMA 2020 was conducted. Searches of PubMed/MEDLINE and Scopus, supplemented by grey literature, were completed in May 2026. Of 78 unique records screened, 46 studies published between 2005 and 2026 met inclusion criteria. Quality appraisal used an adapted Mixed Methods Appraisal Tool; synthesis employed narrative thematic analysis. RESULTS: The five countries represent four structurally distinct pharmacy education architectures: Thailand's standardized six-year Doctor of Pharmacy with dual specialization tracks; four-year Bachelor of Pharmacy programmes in Malaysia and the Philippines with institutional variation; Indonesia's clinically underdeveloped system despite rapid expansion; and Singapore's four-year Bachelor of Pharmacy followed by a nationally mandated one-year pre-registration pathway. Evidence links deeper clinical training to broader practice scope, higher confidence, and improved patient outcomes. Reform produced uneven results: Thailand's PharmD transition improved clinical recognition but exposed deployment paradoxes; Singapore achieved the strongest training-to-practice alignment; Indonesia's health insurance reforms were not absorbed by an underprepared workforce; the Philippines lacks a national competency framework. No binding mutual recognition arrangement was identified; divergent qualification structures, incompatible accreditation systems, and an asymmetric evidence base remain the primary barriers. DISCUSSION: These findings indicate that clinical service scope is bounded less by national policy ambition than by the depth and clinical orientation of the pre-registration education that precedes it, and that credentialing reforms which outpace a health system's capacity to absorb new clinical roles, or the reverse, do not by themselves translate into expanded practice. CONCLUSIONS: Pharmacy education across the ASEAN-5 remains nationally distinct and clinically uneven. Clinical service scope is directly bounded by pre-registration education quality. No country has fully closed the education-practice gap. Regional harmonization requires national-level educational reform as a prerequisite.

Humans↗

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↗