Date Submitted: Nov 10, 2025. Open Peer Review Period: Nov 10, 2025 - Jan 5, 2026.

Date Submitted: Oct 22, 2025. Open Peer Review Period: Nov 10, 2025 - Jan 5, 2026.

Date Submitted: Nov 9, 2025. Open Peer Review Period: Nov 9, 2025 - Jan 4, 2026.

Date Submitted: Nov 9, 2025. Open Peer Review Period: Nov 10, 2025 - Jan 5, 2026.

Date Submitted: Nov 6, 2025. Open Peer Review Period: Nov 6, 2025 - Jan 1, 2026.

Date Submitted: Nov 9, 2025. Open Peer Review Period: Nov 10, 2025 - Jan 5, 2026.

Date Submitted: Nov 9, 2025. Open Peer Review Period: Nov 10, 2025 - Jan 5, 2026.

Background: This paper provides a comprehensive overview of the quantitative component of the Pacific Island Families #Study: Ala mo Tupulaga Pasifika Aotearoa (PIF: ATP; Life Course Trajectories for Young Pacific in Aotearoa), the latest follow-up of the longitudinal PIF birth cohort #Study, which employs a mixed-methods approach. Objective: This paper provides a comprehensive overview of the quantitative component of the Pacific Island Families #Study: Ala mo Tupulaga Pasifika Aotearoa (Life Course Trajectories for Young Pacific in Aotearoa), the latest follow-up of the longitudinal PIF birth cohort #Study, which employs a mixed-methods approach. Methods: The PIF #Study is a multidisciplinary longitudinal #Study that tracks the health and development of 1,398 Pacific children born in 2000 at Middlemore Hospital, South Auckland, Aotearoa | New Zealand. Data collection occurred at ten time points from infancy through young adulthood, with this assessment phase at ages 25–26 years, which aims to reach at least 750 participants. The assessments will take place at participants’ homes or at Auckland University of Technology, South Campus for those residing in Auckland. Data collection will be conducted across multiple sites, including Auckland, Wellington, Hamilton, and Whangārei in ANZ, as well as Brisbane, Sydney, and Melbourne in Australia. Physical measurements: weight, height, waist and hip circumferences, grip strength, body fat mass and muscle mass, blood pressure and pulse, glucose and lipid screening, and skin carotenoid concentration will be undertaken. Additionally, self-reported data will be collected on psychological wellbeing (e.g., depression, anxiety, and family functioning), nutritional and metabolic wellbeing (e.g., food intake and physical activity), and economic wellbeing (e.g., educational attainment, employment status, and job occupation and industry). Results: Data collection is scheduled to commence in June 2025 and conclude by December 2026. The first set of results and analysis is expected to be published from December 2027 onward. Reporting of all results will comply with the STrengthening the Reporting of OBservational studies in Epidemiology guidelines. Conclusions: This paper presents the #Protocol for the 25–26-year follow-up of the first Pacific longitudinal cohort #Study, which will comprehensively examine psychological, nutritional, metabolic, and economic well-being of Pacific young adults. With 25 years of longitudinal data and extensive expertise in life course #Research, this #Protocol outlines the design, methodology and scope of quantitative component of the PIF: ATP #Study. This phase is uniquely positioned to address key issues identified by Pacific communities and generate evidence to inform meaningful interventions and guide policy development while providing robust, contemporary, high-quality empirical evidence.

Background: Postoperative atrial fibrillation (AF) after cardiac surgery is common and is associated with substantial clinical and economic repercussions. However, existing strategies for preventing postoperative AF remain suboptimal, limiting proactive management. Advances in artificial intelligence (AI) may improve the prediction of postoperative AF. Studies have shown that deep learning applied to electrocardiograms (ECGs) can detect subtle patterns in non-AF ECGs associated with a history of (or impending) AF (referred to as the AI-ECG-AF model). Objective: We aimed to determine whether the AI-ECG-AF model can serve as an independent risk factor for postoperative AF after cardiac surgery, compare its predictive performance with existing postoperative AF prediction tools, and assess its additive value. Methods: This single-center retrospective cohort study included 2266 patients (5402 standard 12-lead ECGs) who underwent cardiac surgery at a tertiary hospital in South Korea between December 2018 and December 2023. The AI-ECG-AF model was trained on 4.05 million non-AF standard 12-lead ECGs (1.13 million patients) using a 1-dimensional EfficientNet-B0 architecture and achieved an area under the receiver operating characteristic curve (AUROC) of 0.901 (95% confidence interval: 0.900–0.902) in its held-out test set. Postoperative AF was defined as AF documented by ECG within 30 days after surgery. Using multivariable logistic regression, we assessed the association between the AI-ECG-AF model score and postoperative AF, adjusting for conventional clinical variables. We also investigated the additive or synergistic predictive value of the AI-ECG-AF model score when combined with an existing postoperative AF tool (the POAF score) or other risk factors, based on the AUROC. Results: After adjusting for other clinical variables, a 10% absolute increase in the AI-ECG-AF model score was associated with a 1.197 to 1.209-fold increase in the odds of developing postoperative AF. The AI-ECG-AF model score significantly enhanced postoperative AF prediction: the AUROC of the existing POAF score was 0.643; adding the AI-ECG-AF model score increased it to 0.680 (p < 0.001), and combining the AI-ECG-AF model score with other risk factors raised it to 0.710 (p < 0.001). Conclusions: The AI-ECG-AF model serves as a novel, robust, and independent risk factor for postoperative AF following cardiac surgery and provides additive or synergistic predictive value when integrated with existing postoperative AF prediction tools or other risk factors. Its incorporation can help identify high-risk patients, enabling targeted prophylaxis and closer monitoring during the perioperative period.

Date Submitted: Nov 10, 2025. Open Peer Review Period: Nov 10, 2025 - Oct 26, 2026.

Background: Electronic Gambling Machines and online gambling are the reputedly most damaging gambling type from a public health perspective. Pop-up messages are often used as a responsible gambling (RG) measure to prevent harm for these screen-based types of gambling. Despite some evidence of effectiveness in the literature for these messages, limitations persist, among which low ecological validity is of particular concern. Objective: This #Study aims to test (1) the potential of pop-up messages as a prevention measure in a gambling setting, and (2) if this potential is moderated by characteristics of people exposed to the messages. Secondary objectives also tackle some fundamental assumptions of gambling studies conducted in a laboratory setting. Methods: This is a two-arm stratified block randomised controlled #Study. Eighty participants are recruited under the false pretense of evaluating the realism of a gambling session in a laboratory replicating a bar. Duplicity is also used to make participants believe that they are risking their own money during the experimentation (i.e. winnings and losses are real). Participants are randomised to one of the two arms in a 1:1 ratio: (1) Experimental group (regular gambling session with prevention pop-up messages presented on a fixed schedule; (2) Active control group (regular gambling session). Outcomes measures include behaviours, cognitive and emotional responses to the pop-up-messages. Believability of the gambling session’s realism is also evaluated. Results: Ethical approval was obtained from the Comités d’éthique de la recherche avec des êtres humains de l’Université Laval (CÉRUL; reference no 2020-076 A-1 / 27-05-2024). Recruitment began in February 2024 and was extended to conclude in December 2025. #Study completion is expected in February 2026. No results are currently available. Conclusions: This #Study will provide new insights on the efficacy of pop-up messages as a prevention measure for gambling. Clinical Trial: ClinicalTrials.gov NCT06341504; https://clinicaltrials.gov/#Study/NCT06341504

Background: The diagnosis of sleep disorders presents a challenging landscape, characterized by the complex nature of their assessment and the often divergent views between objective clinical assessment and subjective #patient experience. This study explores the interplay between these perspectives, focusing on the variability of individual perceptions of sleep quality and latency. Objective: Our primary goal is to investigate the alignment, or lack thereof, between subjective experiences and objective measures in the assessment of sleep disorders. Methods: To study this, we developed an aspect-based sentiment analysis method: Using large language models (Falcon 40b and Mixtral 8X7B), we are identifying entity groups of three aspects related to sleep behavior (day sleepiness, sleep quality, fatigue). From phrases referring to these aspects, we are assigning sentiment values between 0 and 1 using a BERT-BiLSTM-based approach (accuracy 78%) and a fine-tuned GPT-2 sentiment classifier (accuracy 87%). Results: The results show that our approach is able to handle the specialized language occurring in the sleep disorder domain and identify the sentiment and opinion in clinical records. Conclusions: Our method has potential in uncovering critical insights into #patient self-perception versus clinical evaluations. Clinical Trial: The secondary usage of Berner Sleep Data Base (BSDB) from Inselspital, University Hospital Bern, was approved by the local ethics committee (KEK-Nr. 2022-00415)

Background: Atrial fibrillation (AF) burden is associated with #cardiovascular events such as stroke and #heart failure. Recent advancements in photoplethysmography (PPG) technology have provided new insights into non-invasive and convenient AF burden detection. Objective: This study aims to establish an AF burden model based on smartwatch-monitored PPG technology to track the progression of AF. Methods: This prospective pilot study (January 2024 to January 2025) at Chinese PLA General Hospital enrolled patients with paroxysmal AF. Participants underwent simultaneous rhythm monitoring using smartwatch PPG and 24-hour Holter ECG (the gold standard). Five PPG-derived AF burden metrics were defined: M1: AF episode duration/total monitoring time. M2: AF episode frequency / total measurements. M3: AF episode density. M4: AF episode Variability. M5: Proportion of rapid ventricular rate AF episodes (>120 bpm). Smartwatch PPG signals were collected once per minute. Sensitivity, specificity, accuracy, precision and F1-score were used to evaluate the PPG algorithm’s AF detection capability by comparison with the gold standard (24-hour Holter monitoring). Mean absolute error (MAE) and Spearman's rank correlation coefficient (rs) were used to assess the correlation between the PPG-based AF burden metrics and the gold standard. Results: A total of 145 participants with paroxysmal AF (66% male, mean age: 63.28±14.23 years) were included. Compared to the gold standard, the PPG-based AF burden model demonstrated a sensitivity of 91.5% (95%CI: 87.9%-95.1%), specificity of 97.2% (95%CI: 95.9%-98.5%), precision of 92.9% (95%CI: 88.6%-97.3%), accuracy of 93.3% (95%CI: 88.2%-98.5%) and F1-score of 90.5% (95%CI: 86.3%-94.7%). The AF burden model exhibited strong discriminatory power in the test cohort (AUC: 89.5%, 95% CI: 89.4%–89.7%). M1: MAE for the model of AF episode duration as a proportion of total monitoring time was 0.0400 (P=0.0076), with a correlation coefficient (rs) of 0.8788 (P

Background: Diagnostic errors and administrative burdens, including medical coding, remain major challenges in healthcare. Large language models (LLMs) have the potential to alleviate these problems, but their adoption has been limited by concerns regarding reliability, transparency, and clinical safety. Objective: This study introduces and evaluates two LLM-based frameworks, implemented within the Rhazes Clinician platform, designed to address these challenges: Generation-Assisted Retrieval-Augmented Generation (GARAG) for automated evidence-based treatment planning and Generation-Assisted Vector Search (GAVS) for automated medical coding. Methods: GARAG was evaluated on 21 clinical test cases created by medically qualified authors. Each case was executed three times independently, and outputs were assessed using four criteria: correctness of references, absence of duplication, adherence to formatting, and clinical appropriateness of the generated management plan. GAVS was evaluated on 958 randomly selected admissions from the MIMIC-IV database, in which billed ICD-10 codes served as ground truth. Two approaches were compared: a direct GPT-4.1 baseline prompted to predict ICD-10 codes without constraints, and GAVS, in which GPT-4.1 generated diagnostic entities that were each mapped onto the top 10 matching ICD-10 codes through vector search. Results: Across the 63 outputs, 62 (98.4%) satisfied all evaluation criteria, with the only exception being a minor ordering inconsistency in one repetition of case 14. For GAVS, the 958 admissions contained 8,576 assigned ICD-10 subcategory codes (1,610 unique). The vanilla LLM produced 131,329 candidate codes, whereas GAVS produced 136,920. At the subcategory level, the vanilla LLM achieved 17.95% average recall (15.86% weighted), while GAVS achieved 20.63% (18.62% weighted), a statistically significant improvement (p < .001). At the category level, performance converged (32.60% vs 32.58% average weighted recall; p = 0.986). Conclusions: GARAG demonstrated a workflow that grounds management plans in diagnosis-specific, peer-reviewed guideline evidence, preserving fine-grained clinical detail during retrieval. GAVS significantly improved fine-grained diagnostic coding recall compared with a direct LLM baseline. Together, these frameworks illustrate how LLM-based methods can enhance clinical decision support and medical coding. Both were subsequently integrated into Rhazes Clinician, a clinician-facing web application that orchestrates LLM agents to call specialized tools, providing a single interface for physician use. Further independent validation and large-scale studies are required to confirm generalizability and assess their impact on patient outcomes.

Background: Although rates of tobacco use in California have declined overall, adults in the San Joaquin Valley (SJV), particularly Hispanic/Latinos (“Latinos”), have disproportionately high rates of tobacco use, tobacco-related illness, and mortality. Residents of the SJV also have limited access to cessation support services, and need accessible, non-clinical alternatives. Given high smartphone use rates among Latinos and residents of rural communities, digital health tools may present an accessible approach to expand cessation support. Objective: This study explored tobacco use behaviors, cessation experiences, and views about digital cessation tools for tobacco cessation among SJV residents. The secondary objective was to assess the appeal, #usability, and necessary adaptations of two existing digital smoking cessation tools—a smoking cessation app and a social media-based intervention. Methods: Through a SJV-based academic-community partnership, we recruited 29 predominantly Latino adults who reported current smoking. We conducted four focus groups (two English, two Spanish) to explore tobacco use and cessation experiences, and preferences for smoking cessation tools. Nine participants subsequently completed in-depth interviews where they viewed videos describing two digital smoking cessation tools — a cessation app and a social media intervention — to assess their appeal and #usability. Results: Most participants were motivated to quit despite experiencing barriers, emphasizing the need for culturally tailored digital cessation tools to enhance engagement. They preferred interventions that integrated culturally relevant content reflecting lived experiences, featured language-concordant communications, and provided social supports, such as chat rooms for peer connection. While participants appreciated the app’s private interface and comprehensive curriculum, the social-media based program was favored for its engaging design, despite privacy concerns. Preferences for specific interventions varied by age and digital literacy. Material rewards increased appeal to use both digital health tools to quit smoking. Conclusions: This sample of predominantly Latino adults from the SJV expressed favorable interest in digital cessation support, yet existing tools require adaptation to improve cultural relevance, accessibility, and #usability. Participants emphasized language-concordant services, representation from people with lived experience, and community-building features. While digital interventions were well received, privacy concerns and digital literacy barriers must be addressed to enhance engagement.

Background: Maintaining mobility is important for older adults to retain independence and reduce fall risk. Wearable technology like fitness trackers and smartwatches can track physical activity. Unfortunately, gait recognition algorithms are often calibrated using younger adults and are not accurate for older adults, especially when using mobility aids. Objective: Our goal was to develop a gait recognition algorithm capable of detecting the walking patterns of older adults that is robust to using mobility aids. Wrist-worn wearable devices were used to maximize the ubiquity of the approach. Methods: We collected walking and other daily activity data on 9 independent older adults to develop a gait recognition algorithm. Four participants used mobility aids (2 cane users, 2 rollator users). We calibrated a heuristic-based “one-size-fits-most” algorithm leveraging the harmonic patterns associated with walking to recognize the walking patterns of our cohort. This algorithm is computationally lightweight and relies only on accelerometer data. We used hyperparameter tuning using a Parzen tree estimator to find the optimal parameters in a leave-one-subject-out fashion. Results: The calibration process was required for this algorithm to detect walking. The signal amplitude threshold lowered from 0.3g to 0.2g to detect the more subtle walking patterns of older adults. The walking frequency range widened from [1.4Hz, 2.3Hz] to [0.8Hz, 2.6Hz], showing that older adults walk more slowly. The ratio for superharmonics increased from 1.4 to 38. Analyzing the false positive rate for the other daily activity classes implies that these superharmonics are artifacts of back-and-forth arm motions that characterize walking in our collected data. Additionally, we report the performance metrics of sensitivity, specificity, and F1-score to evaluate our algorithm. Sensitivity increased tenfold from 0.08 to 0.80. F1-score increased from 0.12 to 0.68. Specificity decreased from 0.99 to 0.77 due to false positives for the activities of brushing teeth and washing hands. Conclusions: This experiment successfully recognized the walking patterns of older adults with or without mobility aids. The performance metrics show that this algorithm has promise for being used to monitor physical activity. This approach is computationally lightweight and explainable. Our calibration approach can be adopted to tune to new populations and has a low barrier to entry due to the sole reliance on accelerometer data which is a standard sensor in wearable devices. The most noteworthy parameter adjustment is the ratio for superharmonics. Low values cause the algorithm not to detect walking in our older adult data. We validated the algorithm on two rollator users. A larger study with more participants using mobility aids is necessary to conduct a deeper analysis on what parameters work best for this population. Future work includes validating the algorithm’s ability to estimate step counts and measure physical activity in real-world settings.

As artificial intelligence (#AI) (AI) transforms healthcare, aligning implementation with evolving management strategies is critical. However, limited research explores the link between the specific nature of AI regulation in healthcare and managing its deployment. FDA and EC regulatory frameworks typically focus on pre-market approval and validation yet largely fail to address the need for continuous monitoring and re-validation of AI models post-marketing. As AI models are exposed to new data in clinical settings, their performance may degrade or alter over time, necessitating ongoing oversight.This often means that healthcare providers must step into the regulatory uncertainty zone to develop local protocols for quality assurance and recalibration. This study was conducted to explore how the specific nature of guidelines for AI in healthcare creates an experimental space where healthcare managers and expert-users (radiologists and other physicians) engage in configuring a usable framework for AI implementation during an innovative, early adoption phase.

Background: Despite the widespread prevalence of social media, there remains a paucity of evidence regarding the use of social media among older adults and its association with mental health. Objective: This study aimed to outline the social media use status among retired elderly, and explore the association between social media use, including time spent and addiction, with mental health status. Methods: A cross-sectional survey was conducted in Shanghai, China, during 2024. A total of 15,986 retired participants were recruited via universities for the aged and primary healthcare institutions. Short version of anxiety (GAD-2) and depression (PHQ-2) scales were used to minimize the required time to complete the questionnaire for the elderly. Logistic regressions were used to examine the impact of social media use on mental health after controlling for covariates. Subgroup analysis were performed considering gender, age, marital status, urbanicity, and socio-economic status. Results: The participants had an average age of 68.49 (SD 7.6) years, with the majority being married and living with spouse (13854/15986, 86.66%) and around half being male (8155/15986, 51.01%). Our research indicated that over 98% of retired elderly individuals had used social media, with WeChat, Douyin, and Kuaishou being the most common platforms. Among them, 52.30% spend 2-3 hours/day, 32.29% spend >4 hours/day, and 20.35% were addicted to social media. Older adults with longer daily social media use time (≥6 hours) exhibited higher rates of anxiety (OR 1.44, 95% CI 1.20-1.72; P

Background: Out-of-hospital cardiac arrests (OHCAs) predominantly occur in residential settings, often witnessed by children who could act as first responders. The World Health Organization (WHO) supports the Kids Save Lives (KSL) initiative, recommending basic life support (BLS) training for children aged ≥11 years. However, disparities in BLS education persist globally, particularly in low-resource regions where socioeconomic barriers, such as school type, malnutrition, and limited infrastructure hinder implementation. Younger children (aged

Background: The integration of AI into clinical workflows is advancing even before full compliance with the MyHealth@EU framework is achieved. While AI-based Clinical Decision Support Systems (CDSS) are automatically classified as high-risk under the EU AI Act, cross-border health data exchange must also satisfy MyHealth@EU interoperability requirements. This creates a dual-compliance challenge: vertical safety and ethics controls mandated by the AI Act, and horizontal semantic-transport requirements enforced through OpenNCP gateways, many of which are still maturing toward production readiness. Objective: This paper provides a practical, phase-oriented tutorial that enables developers and providers to embed AI Act safeguards before approaching MyHealth@EU interoperability tests. The goal is to show how AI-specific metadata can be included in HL7 CDA and FHIR messages without disrupting standard structures, ensuring both compliance and trustworthiness in AI-assisted clinical decisions. Regulatory foundations: We systematically analysed Regulation (EU) 2024/1689 (AI Act) and the MyHealth@EU/OpenNCP technical specifications, extracting a harmonised set of overlapping obligations. AI Act provisions on transparency, provenance, and robustness are mapped directly onto MyHealth@EU workflows, identifying the points where outgoing messages must record AI involvement, log provenance, and trigger validation. Walkthrough: To operationalise this mapping, we propose a minimal extension set, covering AI contribution status, rationale, risk classification, and Annex IV documentation links, together with a phase-based compliance checklist that aligns AI Act controls with MyHealth@EU conformance steps. Illustrative example: A simulated International Patient Summary (IPS) transmission demonstrates how CDA/FHIR extensions can annotate AI involvement, how OpenNCP processes such enriched payloads, and how clinicians in another Member State view the result with backward compatibility preserved. Discussion: We expand on security considerations (e.g., OWASP GenAI risks such as prompt injection and adversarial inputs), continuous post-market risk assessment, monitoring, and alignment with MyHealth@EU’s incident aggregation system. Limitations reflect the immaturity of current infrastructures and regulations, with real-world validation pending the rollout of key dependencies. Conclusions: AI-enabled clinical software succeeds only when AI Act safeguards and MyHealth@EU interoperability rules are engineered together from “day zero.” This tutorial provides developers with a forward-looking blueprint that reduces duplication of effort, streamlines conformance testing, and embeds compliance early. While the concept is still in its early phases of practice, it represents a necessary and worthwhile direction for ensuring that future AI-enabled clinical systems can meet both EU regulatory requirements from day one.

Citing findings published in the Journal of Medical Internet Research, researchers said the AI model "clearly distinguished" between normal, pre- ...

A systematic review published in the Journal of Medical Internet Research found that mobile health interventions significantly improve glycemic ...

... Journal of Medical Internet Research [link here]. “The person appeared to be wearing a wig during this second encounter.” When the participant was ...

Date Submitted: Nov 8, 2025. Open Peer Review Period: Nov 10, 2025 - Jan 5, 2026.

Date Submitted: Nov 5, 2025. Open Peer Review Period: Nov 10, 2025 - Jan 5, 2026.