Date Submitted: Oct 7, 2025. Open Peer Review Period: Oct 8, 2025 - Dec 3, 2025.

Date Submitted: Oct 7, 2025. Open Peer Review Period: Oct 8, 2025 - Dec 3, 2025.

Background: The rapid advancement and widespread adoption of wearable devices provide opportunities to collect longitudinal objective activity and health data and integrate the information directly in...

Date Submitted: Oct 7, 2025. Open Peer Review Period: Oct 7, 2025 - Dec 2, 2025.

Date Submitted: Oct 7, 2025. Open Peer Review Period: Oct 8, 2025 - Dec 3, 2025.

Date Submitted: Oct 7, 2025. Open Peer Review Period: Oct 7, 2025 - Dec 2, 2025.

Background: Depression and anxiety are prevalent among patients with metastatic breast #Cancer (MBC), but there are few evidence-based psychological interventions specifically designed for this population. Objective: To assess the #feasibility, acceptability, and clinical impact of a multi-component positive psychological intervention (PPI), enhanced with an ecological momentary intervention (EMI) for symptom management, for patients with MBC. Methods: We recruited patients with MBC from an NCI-designated comprehensive #Cancer center. Participants completed 5 weekly, virtual individual sessions with a study counselor focused on positive emotion regulation skills. Participants also reported physical and psychological symptoms daily between sessions via SMS messaging. Clinically elevated symptoms triggered a personalized coaching text message tailored to the symptom(s) reported and the skill(s) learned that week. Primary outcomes were intervention #feasibility and acceptability. We also examined pre-to-post-intervention changes in depression, anxiety, and positive affect. Finally, a subset of participants completed qualitative exit interviews focusing on their experience in the study; interview data was analyzed using rapid qualitative analysis (RQA). Results: We approached 20 MBC patients, established contact with 15 (75%), consented 10 (67%), and retained 9 (90%) through the end of the study. Participants were 55 years old on average (range: 35-75) and identified as non-Hispanic White (50%), non-Hispanic Black (40%), or Hispanic/Latina (10%). Participants attended 92% of intervention sessions (M=50 minutes, range: 36-71 minutes). On average, they completed 85% of daily symptom assessments (range: 46%-100%) and received 23 coaching messages (range: 13-31). Participants reported high perceived intervention #feasibility (M=4.81/5), acceptability (M=4.78/5), and appropriateness for patients with MBC (M=4.83/5), above our a priori cut-off of ≥4. All participants (100%) recommended the intervention for other patients with MBC. We observed pre-to-post intervention decreases in depression (d=-0.32) and anxiety (d=-0.27), and increases in positive affect (d=0.30). RQA results demonstrate participants’ positive experiences with the intervention, as well as suggestions for improvement. Conclusions: This pilot study supports the #feasibility of enrolling and retaining racially and ethnically diverse patients with MBC to this trial, the acceptability of the PPI enhanced with EMI, and preliminary intervention impacts on depression, anxiety, and positive affect. A large-scale randomized controlled trial is needed to assess long-term intervention efficacy for outcomes of interest.

Background: Negative symptoms of schizophrenia, such as lack of motivation, pleasure, social interest, and expression, are key contributors to functional impairments in people with schizophrenia. While psychosocial interventions have demonstrated efficacy, no Food and Drug Administration–approved pharmacotherapies exist specifically to target these symptoms. Evidence-based #Digital therapeutics (DTx) may offer novel, scalable treatment options to augment existing treatments. Objective: This article describes the #Study design and methods of a phase 3, multicenter, double-blind, randomized controlled #Study (CONVOKE). It aims to evaluate the effectiveness and safety of CT-155/BI 3972080 (CT-155), a #Smartphone #mHealth-based DTx, as an adjunct to standard-of-care antipsychotic medication in adults with experiential negative symptoms of schizophrenia. Methods: Eligible participants were 18 years or older with a primary diagnosis of schizophrenia receiving stable antipsychotic medication for ≥12 weeks, scored ≥2 on average in at least 2 Clinical Assessment Interview for Negative Symptoms Motivation and Pleasure subscale (CAINS-MAP) domains, and were #Smartphone #mHealth owners. Participants were randomized 1:1 to CT-155 (intervention arm) or a #Digital control #App (control arm). CT-155 integrates aspects of multiple evidence-based psychosocial therapeutic techniques, incorporating principles of in-person psychotherapy aimed at targeting negative symptoms. Development of CT-155 was informed by patients during early clinical learning studies using earlier versions of the #App. The #Digital control included elements of the disease educational components of CT-155 and daily #Digital check-ins. Participants were blinded to their assigned intervention. A blind-to-hypothesis was used so participants appropriately engaged with both apps. Accordingly, participants were informed that they would receive one of 2 interventions under investigation. Investigators, designated site personnel, and central raters were blinded throughout the #Study. The #Study comprised a 2-week screening period, 16-week active period, and a 4-week follow-up period. Change in experiential negative symptoms from baseline to Week 16 (primary end point) was assessed using CAINS-MAP (centrally rated). Other #Study end points included Clinical Assessment Interview for Negative Symptoms Expressivity, Positive and Negative Syndrome Scale, Personal and Social Performance Scale, Dysfunctional Attitudes Scale, Clinical Global Impressions-Severity, and Patient Global Impression of Improvement Scale. Frequency and severity of adverse events were also assessed, as well as engagement and adherence to either #App. Results: #Study enrollment began in March 2023 and was completed in January 2025. Overall, 457 participants were enrolled across 66 clinical #Study sites in the United States. Conclusions: We summarize an innovative trial design for CONVOKE, a phase 3 randomized controlled #Study aimed at assessing the effectiveness, engagement, and safety of CT-155 as an adjunct to standard-of-care for people with negative symptoms of schizophrenia. CONVOKE is the largest-to-date and most robust clinical trial evaluating the effectiveness and safety of a DTx in schizophrenia. The #Study #Protocol included a centrally rated primary end point (CAINS-MAP), blind-to-hypothesis, with an appropriately designed #Digital control. Clinical Trial: ClinicalTrials.gov NCT05838625; https://www.clinicaltrials.gov/#Study/NCT05838625

Background: Drug repositioning is a pivotal strategy in pharmaceutical #research, offering accelerated and cost-effective therapeutic discovery. However, bio#medical information relevant to drug repositioning is often complex, dispersed, and underutilized due to limitations in traditional extraction methods, such as reliance on annotated data and poor generalizability. Large Language Models (LLMs) show promise but face challenges like hallucinations and interpretability issues. Objective: This study proposes Long Chain-of-Thought for Drug Repositioning Knowledge Extraction (LCoDR-KE), a lightweight and domain-specific framework to enhance LLMs’ accuracy and adaptability in extracting structured bio#medical knowledge for drug repositioning. Methods: A domain-specific schema defined 11 entities (e.g., drug, disease) and 18 relationships (e.g., treats, is biomarker of). Following the established schema architecture, we constructed automatic annotation based on 10,000 PubMed abstracts via chain-of-thought prompt engineering. 1000 expert-validated abstracts were curated into DrugReC, a high-quality specialized corpus, while the remaining entries were allocated for model training purposes. Then, the proposed LCoDR-KE framework combined supervised fine-tuning of the Qwen2.5-7B-Instruct model with reinforcement learning and dual-reward mechanisms. Performance was evaluated against state-of-the-art models (e.g., CRF, BERT, BioBERT, Qwen2.5, DeepSeek-R1, OpenBioLLM-70B and model variants) using precision, recall, and F1-score. Additionally, the convergence of the training method was assessed by analyzing performance progression across iteration steps. Results: LCoDR-KE achieved an entity F1 of 81.46% (e.g., drug: 95.83%, disease: 90.52%) and triplet F1 of 69.04%, outperforming traditional models and rivaling larger LLMs (DeepSeek-R1: entity F1=84.64%, triplet F1=69.02%). Ablation studies confirmed the contributions of SFT (8.61% and 20.70% F1 drop if removed) and reinforcement learning (6.09% and 14.09% F1 drop if removed). The training process demonstrated stable convergence, validated through iterative performance monitoring. Qualitative analysis of the model’s CoT outputs showed that LCoDR-KE performed structured and schema-aware reasoning by validating entity types, rejecting incompatible relations, enforcing constraints, and generating compliant JSON. Error analysis revealed four main types of mistakes and challenges for further improvement. Conclusions: LCoDR-KE enhances LLMs’ domain-specific adaptability for drug repositioning by offering an open-source drug repositioning corpus (DrugReC) and a LCoT-farmwork based on lightweight LLM model. This framework supports drug discovery and knowledge reasoning while providing scalable, interpretable solutions applicable to broader bio#medical knowledge extraction tasks.

Background: #Obesity and related disorders are rising globally, especially in China, where they are linked to chronic diseases like #Diabetes and cardiovascular issues. As intermittent fasting gains popularity for weight management, the use of IF apps has increased, yet their quality varies significantly. A systematic evaluation of these apps is essential to assess their effectiveness and reliability. Objective: This study aimed to conduct a comprehensive evaluation of intermittent fasting apps available in the Chinese #Mobile #App market. We concentrated on evaluating their features, quality, and overall user experience to help users avoid low-quality options and direct #App developers to enhance their offers. Methods: A systematic search was performed across five major #App stores in China, including the Apple #App Store, Huawei AppGallery, Oppo Software Store, Vivo #App Store, and Xiaomi Market. “Fasting” in Chinese and English was used as keywords to identify relevant apps, which were then screened based on inclusion and exclusion criteria. The evaluation was conducted using the User Version of the #Mobile Application Rating Scale (uMARS). The uMARS assessment examined four key subscales: engagement, functionality, aesthetics, and information. Each #App was independently evaluated by two raters who underwent uniform training to ensure consistency in scoring. Results: A total of 34 apps were assessed for the study. These apps mostly contain features such as fasting timer (100%), recording weight (97%), fasting reminder (85%), and recording water intake (85%). The results showed that the overall average uMARS score across the apps was 4.35 (SD 0.51), with the highest score reaching 4.97 and the lowest at 2.95. Notably, the functionality subscale had the highest mean score of 4.65 (SD 0.35), while the aesthetic subscale showed the greatest range of scores, from 2.17 to 5.00. The overall uMARS score was significantly positively correlated with the subscale scores (r=0.790-0.955, P

Background: Uptake of the COVID-19 vaccine has been low in the United States despite ongoing public health recommendations. This has been linked to many factors, including pandemic fatigue; reduced risk perception; dis- and misinformation; and, more recently, symptoms of depression and anxiety. Novel communication and messaging strategies are one potential approach to promote vaccine uptake. Objective: This randomized controlled trial aimed to fill research gaps by testing the effect of 2 communication-based approaches―the use of a short attitudinal inoculation message and cognitive behavioral therapy (CBT) kernel messaging―compared to standard public health messaging on vaccine uptake in a cohort of adult US residents. Methods: We completed a 3-arm, parallel-group, assessor-blinded stratified randomized trial between April 15, 2024, and May 2, 2024. Individuals were eligible if they were aged ≥18 years and (1) had received at least one dose of the COVID-19 vaccine but (2) had not received COVID-19 vaccine doses since September 11, 2023, and (3) had not been infected with SARS-CoV-2 in the previous 3 months. We purposively sampled eligible individuals with and without symptoms of anxiety and depression. Participants were randomly allocated to the (1) attitudinal inoculation intervention, (2) CBT kernel intervention, or (3) standard public health messaging intervention. Results: By the 4-week follow-up, COVID-19 vaccination uptake was low overall (17/1403, 1.2%, 95% CI 0.6%-1.8%) and did not significantly differ by study arm―1.5% (7/469) in the CBT kernel arm (95% CI 0.4%-2.8%), 0.9% (4/466) in the inoculation arm (95% CI 0%-1.8%), and 1.3% (6/468) in the standard arm (95% CI 0.3%-2.4%). Compared to the standard arm, the CBT kernel intervention yielded a risk difference (RD) of 0.3% (95% CI −1.3% to 1.8%) and a risk ratio (RR) of 1.21 (95% CI 0.41-3.59); the inoculation intervention yielded an RD of −0.4% (95% CI −1.8% to 1%) and an RR of 0.69 (95% CI 0.19-2.44). Reported SARS-CoV-2 infections and vaccine uptake did not differ by anxiety or depression symptoms. At baseline, approximately one-third of participants (466/1403, 33.21%) reported high willingness to receive another COVID-19 vaccine dose, with no significant differences across arms. At the 4-week follow-up, willingness remained similar across groups (CBT kernel vs standard arm: RD=−0.3%, 95% CI −6.3% to 5.8%, and RR=0.99, 95% CI 0.79-1.25; inoculation vs standard arm: RD=7%, 95% CI 0.8%-13.3%, and RR=1.23, 95% CI 0.98-1.53). Willingness did not differ by mental health status. Conclusions: Successful efforts to increase uptake of the COVID-19 vaccine via theory-enhanced messaging remain elusive. Findings underscore the challenges of shifting behavior through messaging alone in a context of declining public trust and a diminished sense of urgency years after the onset of the COVID-19 pandemic. Ongoing research is needed to better understand and address informational and behavioral barriers to vaccination. Trial Registration: ClinicalTrials.gov NCT06119854; https://clinicaltrials.gov/study/NCT06119854

Background: Hand hygiene is a critical strategy for preventing health care–associated infections (HAIs) and reducing health care costs. However, adherence remains low, particularly among health care assistants (HCAs) and informal caregivers (ICs), who often lack formal training. Virtual reality (VR) delivers standardized, immersive practice with active learning and real-time feedback. It has shown favorable effects on skill execution and acceptability in training paramedics and caregivers. To our knowledge, VR has not been systematically applied to train World Health Organization (WHO)–aligned hand hygiene techniques. Given its portability and suitability for brief, repeatable drills, VR is a plausible solution to upskill HCAs and ICs in both hospital and home-care settings. Objective: This study aims to assess the immediate training effectiveness and implementation feasibility of a brief VR-based hand hygiene program for HCAs and ICs in Colombia. We quantified pre-post changes in correct execution (primary outcome), timing, errors, and knowledge. Success was defined a priori as achieving ≥75% correct execution after training, consistent with adherence levels associated with HAI reductions when embedded in WHO-aligned bundles in prior studies. Methods: In this quasi-experimental, one-group pretest-posttest study, 215 participants (94 HCAs, 121 ICs) completed up to three 15-minute VR training sessions with real-time feedback on hand hygiene technique following the WHO recommendations for hand hygiene. Data were collected at baseline (pre) and immediately after the VR intervention (post). Variables assessed included correct execution (primary; binary), error counts, timing adequacy, knowledge assessment, and acceptability. Results: Correct hand hygiene performance increased from 26.6% to 97.9% among HCAs (95% CI 92.6-99.4; P

Background: Cardiovascular disease (CVD) remains the leading cause of death worldwide, yet many web-based sources on cardiovascular (CV) health are inaccessible. Large language models (LLMs) are increasingly used for health-related inquiries and offer an opportunity to produce accessible and scalable CV health information. However, because these models are trained on heterogeneous data, including unverified user-generated content, the quality and reliability of food and nutrition information on CVD prevention remain uncertain. Recent studies have examined LLM use in various health care applications, but their effectiveness for providing nutrition information remains understudied. Although retrieval-augmented generation (RAG) frameworks have been shown to enhance LLM consistency and accuracy, their use in delivering nutrition information for CVD prevention requires further evaluation. Objective: To evaluate the effectiveness of off-the-shelf and RAG-enhanced LLMs in delivering guideline-adherent nutrition information for CVD prevention, we assessed 3 off-the-shelf models (ChatGPT-4o, Perplexity, and Llama 3-70B) and a Llama 3-70B+RAG model. Methods: We curated 30 nutrition questions that comprehensively addressed CVD prevention. These were approved by a registered dietitian providing preventive cardiology services at an academic medical center and were posed 3 times to each model. We developed a 15,074-word knowledge bank incorporating the American Heart Association’s 2021 dietary guidelines and related website content to enhance Meta’s Llama 3-70B model using RAG. The model received this and a few-shot prompt as context, included citations in a Context Source section, and used vector similarity to align responses with guideline content, with the temperature parameter set to 0.5 to enhance consistency. Model responses were evaluated by 3 expert reviewers against benchmark CV guidelines for appropriateness, reliability, readability, harm, and guideline adherence. Mean scores were compared using ANOVA, with statistical significance set at P70%; P

Background: Early-stage diagnosis of laryngeal #Cancer can significantly enhance patient survival rates and quality of life. However, the scarcity of specialists in low-resource settings constrains the timely review of flexible nasopharyngoscopy (FNS) videos, essential for accurately triaging at-risk patients. Objective: We introduce a preliminary AI-based screening framework to address this challenge in triaging at-risk patients in low-resource settings. The formative research tackles multiple specific challenges common in high-dimensional FNS data: first, the selection of clear, non-blurry images; second, the localization within frames that show an anatomical landmark of interest; and, lastly, the classification of patients into referral grades based on the recorded FNS frames. Methods: The system includes an image quality model (IQM) to identify high-quality endoscopic images fed into a disease classification model (DCM) trained on efficient GhostNet modules. To validate our approach, we curated a real-world dataset from 132 patients at a leading academic tertiary care center. Results: Based on this dataset, we demonstrated that the IQM quality frame selection achieved 0.895 for the area under the receiver operating characteristic curve (AUROC) and 0.878 for the area under the precision-recall curve (AUPRC). Given the high-quality images from the IQM, the DCM improved its performance by 38% in AUROC (from 0.600 to 0.833) and 8% in AUPRC (from 0.840 to 0.912). In addition, the AI model achieved a 2.5 times faster inference time than the ResNet50. Conclusions: This study demonstrated the #feasibility of an AI-based screening framework designed for low-resource settings, demonstrating its capability to screen patients accurately and efficiently. This approach promises substantial benefits for healthcare accessibility and patient outcomes in regions with limited specialist care.

Background: Heart valve surgery is associated with a high risk of perioperative complications. However, current approaches for predicting perioperative complications are all based on preoperative or intraoperative factors, without taking into account the fact that perioperative complications are multifactorial, dynamic, heterogeneous, and interdependent. Objective: We aimed at constructing and quantifying the association network among multiple perioperative complications to elucidate the possible evolution trajectories. Methods: This study utilized the data from China Cardiac Surgery Registry (CCSR), in which 37285 patients were included in the analysis. Bayesian network was used to analyze the associations among 12 complications. Score-based hill-climbing algorithms were used to build the structure and the association between them was quantified using conditional probabilities. Results: We obtained the network of valve surgery complications. 13 nodes represented complications or death, and 34 arcs with arrows represented the directly dependent relationship between them. We identified clusters of complications that were logically related and not related, and quantified the associations. The correlation coefficient between complications increases with the severity of the complications, ranging from 0.01 to 0.41. Meanwhile, the probability of death when multiple complications occurred was calculated. Even mild complications, when progressing to MODS, result in a mortality rate of over 90%. Conclusions: Our network facilitates the identification of associations among specific complications, which help to develop targeted measures to halt the cascade of complications in patients undergoing the valve surgery.

Purpose: As ChatGPT has become more powerful, its application has begun to permeate the field of medical education #mededu. This article employs bibliometric analysis to examine the evolution of research concerning the use of ChatGPT in medical education #mededu and to identify areas for further development in this domain. Methods: As of July 14, 2024, all articles regarding the use of ChatGPT in medical education #mededu within the Web of Science Core Collection were screened by two reviewers. VOSViewer was utilized to illustrate the relationships among countries, authors, and keywords, whereas Citespace was employed to detect references exhibiting significant bursts of activity. Results: The study analyzed a total of 243 publications on the use of ChatGPT in medical education #mededu. There were 1,226 different authors from 60 countries. The top contributors to this field were the United States, China, and India. The journal with the highest number of publications was JMIR medical education #mededu, while the Cures Journal of Medical Science had the highest H-index. Conclusions: This study conducted a bibliometric analysis of ChatGPT research in medical education #mededu from 2023 to 2024, identifying the countries, authors, journals, and publications involved in this field. The findings provide a comprehensive overview of the research conducted on the use of ChatGPT in medical education #mededu.

Background: Freezing of gait (FOG) is a common and debilitating symptom of parkinsonism. Although visual cues have proven efficacy in alleviating FOG, most current visual cues are fixed in place, restricting their use to controlled environments such as clinics or homes. Mobile open-loop cueing devices have been developed to address this limitation; however, they typically require manual activation to deliver the visual cues, which can be particularly challenging for patients with attention or cognitive impairments, leading to equivocal results in improving gait performance. Objective: The aim of the #Study is to assess the efficacy of an ankle bracelet laser, a new mobile visual cue designed for practical use, in improving gait performance in patients with parkinsonism and FOG. Methods: A randomized controlled 2-period crossover trial was conducted from June 15, 2020, to October 1, 2020, at Ramathibodi Hospital. In total, 10 patients with parkinsonism and FOG were enrolled in 2 conditions: walking with laser-off first and walking with laser-on first. Gait speed, the timed up and go test, stride length, and the locomotor #Rehabilitation index were assessed twice in each trial with a 10-minute washout period. Results: The results showed favorable improvement in all parameters. Gait speed and stride length improved by 0.07 m/s (95% CI 0.04-0.09 m/s; P

Background: Australian Aboriginal children experience dental decay at more than twice the rate of non-Aboriginal children. The Select Committee into the Provision of and Access to Dental Services in Australia noted that the rate of potentially preventable hospitalizations was the highest among children aged between 5 and 9 years and was higher among Indigenous Australians and those living in remote locations. The application of a silver fluoride (AgF) solution to decayed surfaces has been shown to be effective in stopping the decay process and reducing the occurrence of new decay but has been tested to a limited extent in the Australian context. Objective: This #Study aims to evaluate the feasibility of using the skills of an Aboriginal health practitioner to undertake the application of AgF to carious primary molars to arrest the caries progression and prevent the occurrence of new caries among young Aboriginal children in remote communities. Methods: This #Study is a cluster-#RCT #ClinicalTrial with communities randomized and stratified based on caries level and water fluoridation status. The trial will recruit 640 children (aged between 6 months and 7 years) from 30 communities. Informed consent will be obtained. At baseline, each child in the intervention group will be examined by a calibrated examiner and subsequently by an oral health practitioner who will prescribe to an Aboriginal health practitioner the teeth to be treated with AgF. A formulation with 38% AgF will be applied for 1 minute (0.004 mL per tooth). The control group will be provided with standard minimally invasive care. Participants will be followed annually for 2 years to assess caries arrest and prevention by blinded calibrated examiners. Child oral health–related quality of life and dental anxiety will be elicited through validated questionnaires. Tests of proportions will be used to evaluate the proportion of lesions arrested and the proportion of surfaces at risk that decayed over the follow-up. Multiple logistic regression with appropriate control for clustering of teeth and communities will be used to evaluate caries arrest, controlling for potential confounding factors. Results: Community engagement has commenced, and data collection #Protocols have been prepared. Staff specific to the #Study (eg, Aboriginal health practitioners or workers) are in the process of recruitment. Participant recruitment will commence in March 2026 and conclude in December 2026. #Study outcomes will be reported at 12- and 24-month follow-ups. Conclusions: This #Study will test the effectiveness and feasibility of a non-oral health professional applying AgF to achieve caries arrest and prevention and validate clinical findings against #Digital imagery acquired on site. This pragmatic #Study will inform the development of suitable and accessible models of care for dental service provision in rural and remote communities in Australia. Trial Registration: Australian New Zealand Clinical Trials Registry (ANZCTR) ACTRN12624000457549p; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=387518&isReview=true