DIGITAL HEALTH最新文献

{"title":"Performance evaluation of large language models for the national nursing examination in Japan.","authors":"Tomoki Kuribara, Kengo Hirayama, Kenji Hirata","doi":"10.1177/20552076251346571","DOIUrl":"https://doi.org/10.1177/20552076251346571","url":null,"abstract":"<p><strong>Objectives: </strong>Large language models (LLMs) are increasingly used in healthcare, with the potential for various applications. However, the performance of different LLMs on nursing license exams and their tendencies to make errors remain unclear. This study aimed to evaluate the accuracy of LLMs on basic nursing knowledge and identify trends in incorrect answers.</p><p><strong>Methods: </strong>The dataset consisted of 692 questions from the Japanese national nursing examinations over the past 3 years (2021-2023) that were structured with 240 multiple-choice questions per year and a total score of 300 points. The LLMs tested were ChatGPT-3.5, ChatGPT-4, and Microsoft Copilot. Questions were manually entered into each LLM, and their answers were collected. Accuracy rates were calculated to assess whether the LLMs could pass the exam, and deductive content analysis and Chi-squared tests were conducted to identify the tendency of incorrect answers.</p><p><strong>Results: </strong>For over 3 years, the mean total score and standard deviation (SD) using ChatGPT-3.5, ChatGPT-4, and Microsoft Copilot was 180.3 ± 22.2, 251.0 ± 13.1, and 256.7 ± 14.0, respectively. ChatGPT-4 and Microsoft Copilot showed sufficient accuracy rates to pass the examinations for all the years. All LLMs made more mistakes in the health support and social security system domains (<i>p</i> < 0.01).</p><p><strong>Conclusions: </strong>ChatGPT-4 and Microsoft Copilot may perform better than Chat GPT-3.5, and LLMs could incorrectly answer questions about laws and demographic data specific to a particular country.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251346571"},"PeriodicalIF":2.9,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12117227/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144175771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Qualitative study on the characteristics and dilemmas of eHealth literacy among family caregivers of breast cancer patients.","authors":"Chengcheng Guo, Yang Wang, Liting Dong, Juan Wu, Sijin Guo","doi":"10.1177/20552076251346240","DOIUrl":"https://doi.org/10.1177/20552076251346240","url":null,"abstract":"<p><strong>Objective: </strong>To explore the elements and dilemmas of eHealth literacy among family caregivers of breast cancer patients, providing a reference for improving their caregiving abilities.</p><p><strong>Methods: </strong>From September to October 2023, a phenomenological research method was adopted. Semi-structured interviews were conducted with 10 family caregivers of breast cancer patients in the Department of Breast and Thyroid Surgery of a tertiary grade A hospital in Xi'an. Braun's thematic analysis method was used for data analysis.</p><p><strong>Results: </strong>Two themes and six sub-themes were extracted. (1) Element characteristics: information acquisition ability, information discrimination ability, and information application ability. (2) Information dilemmas: information overload and chaos, difficulty in judging the authenticity and reliability of information, and obstacles in information application.</p><p><strong>Conclusion: </strong>There are many problems in the eHealth literacy of family caregivers of breast cancer patients. Medical staff should have provided guidance on information acquisition, conducted training on information discrimination and application, addressed information overload and chaos, strengthened information supervision, improved the proficiency of using e-communication tools, and provided more professional guidance to enhance their eHealth literacy and reduce their information-related distress.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251346240"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12117220/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144175774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Systematic review of digital health interventions to support self-management of low back pain in the workplace.","authors":"Minghao Chen, Valerie Sparkes, Liba Sheeran","doi":"10.1177/20552076251336281","DOIUrl":"https://doi.org/10.1177/20552076251336281","url":null,"abstract":"<p><strong>Background: </strong>Low back pain (LBP) is a prevalent condition in working populations, imposing significant individual, organisational, and societal burdens, including reduced quality of life, impaired work performance, and high healthcare costs. Digital health interventions (DHIs) offer scalable solutions for self-managing LBP in workplace settings, yet their tailoring, integration, and effectiveness remain unclear.</p><p><strong>Objective: </strong>This systematic review evaluates the effectiveness of DHIs in supporting LBP self-management in workplace environments. It examines intervention components, tailoring methods, integration with occupational health (OH) pathways, and their impact on clinical and work-related outcomes.</p><p><strong>Methods: </strong>A systematic search was conducted across PubMed, MEDLINE, EMBASE, CINAHL, the Cochrane Library, and Web of Science, following PRISMA guidelines. Randomised controlled trials evaluating DHIs for workplace LBP were included. Data extraction focused on intervention characteristics, tailoring approaches, and primary outcomes, including pain intensity, disability, and physical performance. The quality of evidence was assessed using the Cochrane risk-of-bias and GRADE frameworks.</p><p><strong>Results: </strong>Five studies were included, featuring DHIs delivered via web-based platforms or mobile applications. Interventions incorporated exercise, ergonomics education, and work activity modification. Only one study used a tailored approach based on theoretical frameworks and individualised work classifications. Moderate-quality evidence supported improvements in pain, disability, and physical performance, but effects on quality of life, psychosocial factors, and work outcomes were inconsistent. Integration with occupational health pathways was absent in all studies.</p><p><strong>Conclusions: </strong>The lack of tailoring and integration within occupational health systems limits the scalability and impact of DHIs for workplace LBP. Future research should focus on personalised, theory-driven interventions and systemic alignment with occupational health policies to enhance their feasibility, implementation, and long-term outcomes.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251336281"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12117239/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144175778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The mediation effect of social participation in the relationship between Internet use and health behaviors among middle-aged and older individuals.","authors":"Xinmei Yang, Yang Chen, Tuo Qiu, Xingyue Zhu","doi":"10.1177/20552076251346239","DOIUrl":"https://doi.org/10.1177/20552076251346239","url":null,"abstract":"<p><strong>Objectives: </strong>The increasing popularity and influence of the Internet in modern society have greatly impacted individuals' lives. This population-based study elucidated the dynamic linkage between Internet use (IU) and health behaviors (HBs), particularly emphasizing the intermediary function of social participation (SP) in middle-aged and older adults (MOA) in China.</p><p><strong>Methods: </strong>Data were obtained from the China Health and Retirement Longitudinal Study in 2020. This study employed binary logistic regression to investigate the influence of IU on the HBs of MOA in China. Additionally, binary logistic and multiple linear regressions were used to test whether SP regulates the relationship between IU and HBs (i.e. nonsmoking, non-drinking, physical activity, and physical examination). Furthermore, the Karlson-Holm-Breen method was employed to assess the mediating role of SP.</p><p><strong>Results: </strong>IU had a positive effect on nonsmoking (OR: 1.113, <i>p</i> < 0.05), physical activity (OR: 1.775, <i>p</i> < 0.001), and physical examination (OR: 1.226, <i>p</i> < 0.001). However, this study revealed that IU had a significant negative effect on non-drinking (OR: 0.775, <i>p</i> < 0.001). Moreover, the mediating effect analysis demonstrated that SP played a mediating role in the relationship between IU and the HBs.</p><p><strong>Conclusion: </strong>Active engagement in social activities is an effective method for enhancing the positive impact of IU on the adoption of physical activity and physical examinations. In order to meet individuals' needs, it is essential to design and promote Internet health services and social activities tailored to their age group and cultural background. Furthermore, greater efforts should be employed to implement public policy initiatives aimed at providing care for MOA.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251346239"},"PeriodicalIF":2.9,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12117234/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144175786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital natives: A systematic review of the digital health literacy and influencing factors among Chinese college students.","authors":"Jiajun Jiang, Mei Zhou, Zhihua Yin, Yuqing Cui, Sishuai Liu","doi":"10.1177/20552076251346006","DOIUrl":"https://doi.org/10.1177/20552076251346006","url":null,"abstract":"<p><strong>Objective: </strong>This study aimed to explore the current status of digital health literacy (DHL) among Chinese college students, and to summarize and analyze the related influencing factors.</p><p><strong>Methods: </strong>A systematic search was conducted in Web of Science, PubMed, EBSCO, Scopus, China National Knowledge Infrastructure, Wanfang Data, and VIP databases for relevant literature published between 1 January 2006 and 30 May 2024.</p><p><strong>Results: </strong>A total of 32 articles, with sample sizes ranging from 224 to 16,497 participants, were included in this review. The e-Health Literacy Scale was the most commonly used measurement for DHL among China college students. The level of DHL among Chinese college students was not high, which was associated with a variable of factors. Based on the social ecological model, the influencing factors at the individual level included age, gender, depression and anxiety status, self-rated health status, frequency of physical exercise, participation in health club activities, taking health courses, and attitude toward health; at the interpersonal level, the influencing factors included family income, parents' education level, family structure, and only child status; at the organizational level, influencing factors included major, grade, and interpersonal relationships; at the community level, influencing factors included hukou and place of origin.</p><p><strong>Conclusions: </strong>The findings of this review provide valuable insights for the development of targeted health education interventions and policies to enhance DHL among Chinese college students. Future research should prioritize nationwide studies to further investigate the factors influencing DHL among Chinese college students and to develop more scientifically rigorous and effective measurement tools for DHL.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251346006"},"PeriodicalIF":2.9,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12107012/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144163665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Patients prefer easy adverse event reporting: Observational study within clinical trial.","authors":"Lauri Lukka, Maria Vesterinen, Joonas J Juvonen, Satu Palva, J Matias Palva","doi":"10.1177/20552076251345894","DOIUrl":"https://doi.org/10.1177/20552076251345894","url":null,"abstract":"<p><strong>Background: </strong>Digital intervention safety is crucial for regulatory approval and clinical adoption. However, the evaluation and reporting of adverse events (AEs) in clinical trials are often insufficient. Digital qualitative self-reporting could enhance the detection of AEs, but patient preferences for using such channels remain understudied.</p><p><strong>Methods: </strong>This observational study was conducted in Finland between 2022 and 2024 within a randomized controlled trial evaluating the efficacy of <i>Meliora</i>, a game-based digital intervention for patients living with major depressive disorder. We assessed the preferences of 1001 patients for self-reporting AEs across four channels: a prompted, within-intervention questionnaire (CORTO: Contextual, One-item, Repeated, Timely, Open-ended), a Jira questionnaire, email, and phone.</p><p><strong>Results: </strong>148 (14.8%) patients reported AEs during the study. We found a significant imbalance between the channels: 11.3% (<i>n</i> = 113) of patients reported AEs using CORTO, 4.1% (<i>n</i> = 41) using email, 1.1% (<i>n</i> = 11) using Jira, and 0.4% (<i>n</i> = 4) using phone.</p><p><strong>Conclusions: </strong>These findings reveal that patients prefer low-effort methods for reporting AEs and are more likely to report AEs via a prompted, within-intervention questionnaire (CORTO) than through other methods. Integrating qualitative self-report channels into digital interventions may enhance AE detection rates, improve clinical trial safety monitoring, and support post-market surveillance.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251345894"},"PeriodicalIF":2.9,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12106987/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144163669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Developing a predictive model for septic shock risk in acute pancreatitis patients using interpretable machine learning algorithms.","authors":"Binglin Song, Ping Liu, Kangrui Fu, Chun Liu","doi":"10.1177/20552076251346361","DOIUrl":"https://doi.org/10.1177/20552076251346361","url":null,"abstract":"<p><strong>Background: </strong>Septic shock is a severe complication of acute pancreatitis (AP), often associated with poor prognosis. This study aims to analyze the clinical characteristics of patients with acute pancreatitis and develop an interpretable early prediction model for septic shock in these patients using machine learning (ML). The model is intended to assist emergency physicians in resource allocation and medical decision making.</p><p><strong>Methods: </strong>Data were collected from the MIMIC-IV 3.0 database. The dataset was divided into a training set and a test set in a 7:3 ratio. Feature selection was performed using LASSO (Least Absolute Shrinkage and Selection Operator) regression. Subsequently, 10 ML models were developed: Random Forest, Logistic Regression, Gradient Boosting Machine, Neural Network, Extreme Gradient Boosting (XGBoost), K-Nearest Neighbor, Adaptive Boosting, Light Gradient Boosting Machine, Category Boosting, and Support Vector Machine. To enhance and optimize model interpretability, Shapley Additive Explanations (SHAP) were employed.</p><p><strong>Results: </strong>A total of 1032 patients with AP were included in this study, from which 31 variables were selected for model development. By comparing the area under the receiver operating characteristic curve and decision curve analysis results between the training and test sets, the XGBoost model demonstrated a significant advantage over other models. SHAP analysis revealed that white blood cell count, total bilirubin (bilirubin total), and bicarbonate (HCO₃ ^-) levels were the three most critical risk factors for the development of septic shock in patients with AP.</p><p><strong>Conclusion: </strong>ML approaches exhibited promising performance in predicting septic shock in patients with AP. These models may aid in guiding treatment decisions for patients with AP in the emergency department.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251346361"},"PeriodicalIF":2.9,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12107010/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144163663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The relationship between nursing informatics competencies and clinical decision-making among registered nurses working at tertiary hospitals.","authors":"Mohammad Mousa Al-Qudah, Islam Ali Oweidat, Saleh Al Omar, Ghada Abu Shosha, Ala'a Dalky, Khalid Al-Mugheed, Sally Mohammed Farghaly Abdelaliem, Mohammed ALBashtawy","doi":"10.1177/20552076251346015","DOIUrl":"https://doi.org/10.1177/20552076251346015","url":null,"abstract":"<p><strong>Aim: </strong>This study aims to explore the relationship between nursing informatics competencies and clinical decision-making among nurses in Jordan.</p><p><strong>Design: </strong>A cross-sectional, descriptive-correlational design was used with a sample of 249 registered nurses from three tertiary governmental hospitals in Jordan, utilizing the Self-Assessment of Nursing Informatics Competencies Scale and the Clinical Decision-Making in Nursing Scale.</p><p><strong>Results: </strong>Nurses reported moderate-to-high informatics competencies (<i>M</i> = 3.17, SD = 0.76) and decision-making abilities (<i>M</i> = 3.30, SD = 0.94). Total nursing informatics competencies score showed a significant positive correlation with all clinical decision-making domains, with the strongest correlation observed for critical thinking (Pearson <i>r</i> = 0.57, <i>P</i> < 0.05), followed by clinical judgment (Pearson <i>r</i> = 0.52, <i>P</i> < 0.05).</p><p><strong>Conclusion: </strong>The findings underscore a significant link between nursing informatics competencies and clinical decision-making, with professional experience, system usage frequency, and informatics training serving as key predictors. These results highlight the importance of targeted interventions to enhance informatics skills and support effective clinical decisions.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251346015"},"PeriodicalIF":2.9,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12107008/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144163605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Delivering the Lee Silverman voice treatment-loud method in-site versus telerehabilitation in people with multiple sclerosis: Feasibility evidence of a non-inferiority pilot randomized controlled trial.","authors":"Chiara Vitali, Giulia Fusari, Diego Michael Cacciatore, Giulia Smecca, Cinzia Baldanzi, Alessio Carullo, Marco Rovaris, Davide Cattaneo, Francesca Baglio, Sara Isernia","doi":"10.1177/20552076251326222","DOIUrl":"https://doi.org/10.1177/20552076251326222","url":null,"abstract":"<p><strong>Objective: </strong>Telerehabilitation may overcome accessibility barriers related to the Lee Silverman Voice Treatment (LSVT)-Loud for dysphonia rehabilitation in multiple sclerosis (MS). The present study provides the feasibility evidence on patient-relevant structural and procedure effects of a pilot randomized controlled trial comparing LSVT-Loud telerehabilitation (Tele-LSVT-Loud) versus standard delivery.</p><p><strong>Methods: </strong>Twenty-one people with MS (six males) with dysphonia were 1:1 randomly allocated to 4 weeks of LSVT-Loud in-site or Tele-LSVT-Loud at home accessing a telemedicine platform. The feasibility of Tele-LSVT-Loud compared to LSVT-Loud was evaluated considering adherence rate, safety (adverse events), technology interaction (User Experience Questionnaire), intrinsic motivation to the treatment (Intrinsic Motivation Inventory), and perceived rehabilitation experience (individual qualitative interviews) during and after the intervention program.</p><p><strong>Results: </strong>Thirty-one percent of eligible subjects were unavailable to follow in-site treatment. Drops-outs were higher in the LSVT-Loud than Tele-LSVT-Loud group (4 versus 1). Also, the adherence rate of synchronous sessions was 68.75% in the LSVT-Loud compared to 87.5% in the Tele-LSVT-Loud group, related to greater difficulty in integrating the treatment into a daily routine, as mentioned in the qualitative interview. No relevant adverse events were observed in both groups. The user experience with technology in the Tele-LSVT-Loud group was positive. The interviews revealed a positive therapeutic alliance, regardless of the delivery path. Interestingly, only people in the Tele-LSVT-Loud group judged equivalent the therapist-user relationship in in-site and telerehabilitation settings.</p><p><strong>Conclusions: </strong>Telerehabilitation promotes the feasibility of LSVT-Loud. The modality of delivery is a relevant factor in determining eligibility and adherence to a voice rehabilitation program in MS.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251326222"},"PeriodicalIF":2.9,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12107005/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144163634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hybrid deep learning for IoT-based health monitoring with physiological event extraction.","authors":"Sivanagaraju Vallabhuni, Kumar Debasis","doi":"10.1177/20552076251337848","DOIUrl":"https://doi.org/10.1177/20552076251337848","url":null,"abstract":"<p><strong>Objective: </strong>Integrating IoT technologies into the healthcare system has significantly raised the prospects for patient monitoring and disease prediction. However, the present-day models have failed to effectively encompass spatial-temporal data samples.</p><p><strong>Methods: </strong>This paper presents a novel hybrid machine-learning model by amalgamating Convolutional Neural Networks (CNNs) with Long Short-Term Memory models (LSTMs) to boost prediction accuracy. Whereas the CNNs extract spatial features from medical images, the LSTMs model the temporal patterns of wearable sensor data. Such a configuration increases the prediction accuracy by 10% more than that achieved by the individual models. For better feature extraction, the proposed method implements Physiological Event Extraction (PEE), which is aimed at identifying important physiological events such as heart rate variability and respiratory changes from raw sensor data samples.</p><p><strong>Results: </strong>This method helps render the features interpretable, providing another 15% improvement in prediction performance. Anomaly detection employed ensemble techniques that combined the Isolation Forest and One-Class SVM, reducing false positives by 20%, thus outperforming conventional approaches. It further enhanced the True Positive Rate (TPR) by 25% through using an online learning algorithm with Incremental Gradient Descent with Momentums. Robust statistical methods based on M-estimator theory had been integrated for the treatment of outliers and missing data, which helped in reducing bias in estimation by 30% and increasing the False Positive Rate (FPR) by 12%.</p><p><strong>Conclusion: </strong>All these enhancements constitute a major step towards improving the IoT healthcare data processing chain, thereby providing a trusted and accurate system for real-time health monitoring and anomaly detection. In this regard, the research also paves the way for designing next-gen IoT healthcare analytics and their actual clinical applications.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251337848"},"PeriodicalIF":2.9,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12104608/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144152884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}