DIGITAL HEALTH最新文献

{"title":"Hybrid deep learning for IoT-based health monitoring with physiological event extraction.","authors":"Sivanagaraju Vallabhuni, Kumar Debasis","doi":"10.1177/20552076251337848","DOIUrl":"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}

{"title":"A comprehensive digital medicine platform for hypertension and diabetes care in primary care: A real-world feasibility test.","authors":"Ye Seul Bae, Seoyeon Park, Changbo Noh, Boram Choi, Seung Ah Yi, So Eun Kim, Yoon Ji Kim, Jae-Heon Kang","doi":"10.1177/20552076251344375","DOIUrl":"10.1177/20552076251344375","url":null,"abstract":"<p><strong>Background: </strong>Hypertension and diabetes are typically managed in primary care settings, where continuous and tailored care based on continuity in daily life is essential to improve health outcomes.</p><p><strong>Objective: </strong>To address the shortage of human and material resources in primary care and to enhance health outcomes, we developed and implemented a comprehensive digital medicine platform in a real-world setting.</p><p><strong>Methods: </strong>A patient application (app) and a web app for healthcare providers were developed to enable bidirectional communication between patients and healthcare providers. Primary physicians and care coordinators utilized this platform to deliver personalized care management. To manage multiple patients with limited staff, automated message generation logic was used.</p><p><strong>Results: </strong>Using the app, patients could self-measure blood pressure and blood glucose level, receive feedback from healthcare providers, and obtain personalized medication management, disease education, and lifestyle guidance regarding smoking, alcohol consumption, and exercise. Healthcare providers can view the data generated by the patient app, in real time, on the web app, and immediately send messages when an action is required. An evaluation of effectiveness was conducted with 502 patients in the intervention group and 502 patients in the control group over a 24-week intervention. In the intervention group, systolic blood pressure decreased by 3.8% (<i>P</i> < 0.001), diastolic blood pressure by 3.4% (<i>P</i> < 0.001), body mass index by 1.6% (<i>P</i> < 0.001), and waist circumference by 1.5% (<i>P</i> < 0.001). HDL cholesterol increased by 2.4% (<i>P</i> < 0.05), and triglycerides decreased by 5.4% (<i>P</i> < 0.05).</p><p><strong>Conclusion: </strong>This study demonstrated that the adoption of a digital medicine platform is effective and essential for continuous patient management in primary care. Information and communications technology-based tools and applications are becoming increasingly important in healthcare, and our study has provided valuable insights into the management of chronic diseases.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251344375"},"PeriodicalIF":2.9,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12103660/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144144281","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":"Undergraduate nursing students' preferences for virtual reality simulations in nursing skills training: A discrete choice experiment.","authors":"Yanya Chen, Yan He, Xiaochun Zou, Hongya Cai, Hei Hang Edmund Yiu, Wai-Kit Ming","doi":"10.1177/20552076251339009","DOIUrl":"10.1177/20552076251339009","url":null,"abstract":"<p><strong>Objective: </strong>Nursing students face increasing challenges, making it crucial to explore innovative teaching methods such as virtual reality simulations to enhance skills. This study aimed to assess their preferences for virtual reality simulation in skills training using a discrete choice experiment.</p><p><strong>Method: </strong>A discrete choice experiment with six attributes (types of virtual reality, interaction, learning contents, collaboration, frequency, and costs for each additional training module) was used. A mixed logit model and latent class analysis were adopted to analyze data using Sawtooth Software and STATA BE 18.</p><p><strong>Result: </strong>A total of 518 undergraduate nursing students completed this study. They identified costs for each additional training module, interaction, and types of virtual reality as the three most important attributes. The combination that was most preferred was immersive virtual reality simulation, high-level interaction types, advanced learning contents, mixed learning modules, lower prices, and a practice frequency of once every two weeks. Two classes of students were identified: Class 1 valued the cost for each additional training module, while Class 2 preferred a better immersion experience.</p><p><strong>Conclusion: </strong>Students' preference for virtual reality simulation depended on some factors, including types of virtual reality, interaction, learning contents, and costs for each additional training module. Nursing educators should take students' preferences into account to ensure that their preferences and needs are addressed as fully as possible.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251339009"},"PeriodicalIF":2.9,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12102573/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144144400","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":"Assessing the usability and performance of digital healthcare systems in Nigerian teaching hospitals: Challenges and future directions.","authors":"Uchechukwu Solomon Onyeabor, Okechukwu Onwuasoigwe, Wilfred Okwudili Okenwa, Thorsten Schaaf, Niels Pinkwart, Felix Balzer","doi":"10.1177/20552076251341089","DOIUrl":"10.1177/20552076251341089","url":null,"abstract":"<p><strong>Background: </strong>Several developing countries, including Nigeria, are still in the nascent stages of adopting digital healthcare support solutions to enhance clinical service delivery. Consequently, there currently exists a scarcity of research and gaps in the literature regarding the efficacy, effectiveness, overall performance, and possibly success factors associated with these systems, or conversely, identify the design and implementation deficiencies, as well as the use-based challenges present in them. This is the gap this research seeks to address. The findings from these evaluations are anticipated to inform improvements to the existing systems and guide future implementations.</p><p><strong>Method: </strong>The research was conducted at three referral and university teaching hospitals in southern Nigeria. It involved an extensive period of on-site observations and clinician engagements. A validated 5-point Likert scale questionnaire was designed to capture the peculiarities of the prevailing contexts across these hospitals. The survey targeted 150 clinicians, and responses were analyzed using SPSS, while visual representations were created in MS Excel.</p><p><strong>Result: </strong>Findings showed that 79.4% of clinicians identified feature gaps and expressed the need for additional functionalities. However, 71.9% acknowledged that their systems had interfaces facilitating electronic requests to service units like radiology and pharmacy. Despite this, some clinicians faced challenges due to missing features, which prevented them from fully achieving their clinical goals. Furthermore, 80.2% reported experiencing instances where the electronic health record (EHR) systems were slow, unresponsive, or caused prolonged interruptions that hindered workflow efficiency.</p><p><strong>Conclusion: </strong>The findings, particularly the 79.4% of clinicians desiring additional features and the 80.2% experiencing system slowdowns, highlight the urgent need for digital healthcare policies in developing nations to prioritize user-centered design protocols during systems implementation in order to better align EHR systems with clinical workflows and reduce clinician burnout. It would as a result be pertinent to engage the clinicians in any future design or redesign process and also provide targeted trainings which will ensure EHR systems better support healthcare providers in delivering quality patient care.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251341089"},"PeriodicalIF":2.9,"publicationDate":"2025-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12099148/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144144305","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":"Testing certain and uncertain incentives on study retention in a longitudinal social media survey among young adults: An embedded recruitment trial.","authors":"Jennifer Cantrell, Megumi Ichimiya, Paul Mowery, Alexander P D'Esterre, Jeffrey Bingenheimer, Shreya Tulsiani, Elizabeth C Hair, Jennifer M Kreslake, Madeline Martin, Raquel Gerard, William Douglas Evans","doi":"10.1177/20552076251336522","DOIUrl":"10.1177/20552076251336522","url":null,"abstract":"<p><strong>Introduction: </strong>Incentives can be effective in survey research but evidence is limited on how incentive type impacts survey retention in longitudinal social media-based surveys. This study examined how certain and uncertain incentives affect study retention among US young adults recruited online and whether incentive effects vary by sociodemographic factors.</p><p><strong>Methods: </strong>Participants were randomized in a 1:1:1 ratio to a three-arm parallel trial (<i>n</i> = 1615) with (1) a lottery for a $200 gift card (uncertain), (2) a cash equivalent (CE) of a $5 gift card per survey (certain); or (3) a combination of both options (combined), and were surveyed at baseline, 30 days, and 60 days. This study focused on survey retention at 30 days (among baseline completers, <i>n</i> = 1491) and 60 days (among 30-day completers, <i>n</i> = 1018). Participants were not blinded to their condition but were blinded to other conditions and researchers were blinded until data collection was complete. Logistic regressions examined survey retention as a function of incentive condition and sociodemographics, with additional analyses of interaction effects. We report average marginal effects (AMEs) with significance defined as <i>p</i> < 0.05.</p><p><strong>Results: </strong>The certain CE was effective for survey retention versus the lottery at 30-day follow-up only (43.8% [lottery] vs. 77.7% [CE], AME: 0.346, <i>p</i> < 0.000); there were no differences between CE versus lottery at 60-day follow-up (76.1% [lottery] and 81.3% [CE], AME: 0.054, <i>p</i> = 0.192). The combined incentive demonstrated significantly higher retention at both follow-ups versus the lottery but no significant advantage over the CE. Incentive effectiveness showed minimal variation across sociodemographic factors.</p><p><strong>Discussion: </strong>This study is among the few to experimentally test incentives for retention in online social-media based research. A certain CE was most effective for short-term web survey retention among young adults compared with a lottery. Findings suggest that small guaranteed rewards may better motivate study retention than uncertain larger amounts.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251336522"},"PeriodicalIF":2.9,"publicationDate":"2025-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12099130/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144144396","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":"A web-based tool for predicting gastric ulcers in Chinese elderly adults based on machine learning algorithms and noninvasive predictors: A national cross-sectional and cohort study.","authors":"Xingjian Xiao, Xiaohan Yi, Zumin Shi, Zongyuan Ge, Hualing Song, Hailei Zhao, Tiantian Liang, Xinming Yang, Suxian Liu, Bo Sun, Xianglong Xu","doi":"10.1177/20552076251336951","DOIUrl":"10.1177/20552076251336951","url":null,"abstract":"<p><strong>Background: </strong>As the Chinese population continues to age, the prevalence of gastric ulcers, a common nutrition and diet-related disorder, is rising among the elderly. Gastric ulcers pose a significant public health challenge in China, yet there is limited research to predict gastric ulcers accurately.</p><p><strong>Objective: </strong>Our study aims to employ machine learning algorithms to predict the occurrence of gastric ulcers and develop an online tool to assess the risk of gastric ulcers for elderly individuals, both currently and in the future, while identifying important predictors.</p><p><strong>Method: </strong>We used baseline data from the Chinese Longitudinal Healthy Longevity Survey in 2011 and 2014, with a follow-up endpoint of 2018. We employed nine machine learning algorithms to construct predictive models for gastric ulcers over the next seven years (2011-2018, with 1482 samples) and the next three years (2014-2018, with 2659 samples). Additionally, we utilized cross-sectional data from 2018 (with 13,775 samples) to construct a predictive model for current gastric ulcers.</p><p><strong>Results: </strong>Noninvasive predictors such as demographic, behavioral, nutritional, and physical examination factors were utilized to predict the current and future occurrence of gastric ulcers. In our study, Support Vector Machine (SVM), Random Forest (RF), and Light Gradient Boosting Machine (LGBM) achieved an accuracy of 0.97 for predicting gastric ulcers over seven years; Logistic Regression, Adaptive Boosting, SVM, RF, Gradient Boosting Machine, LGBM, and K-Nearest Neighbors reached 0.98 for three-year predictions; and SVM, Extreme Gradient Boosting, RF, and LGBM attained 0.95 for current gastric ulcer prediction.</p><p><strong>Conclusions: </strong>We developed MyGutRisk, built on optimal machine learning models, relatively accurately predicts gastric ulcer risk in elderly adults using noninvasive factors like diet and lifestyle. It supports self-assessment via a public link and clinical screening in community health settings to guide preventive measures. However, as a prototype, it requires further validation to ensure accuracy and generalizability across diverse populations and real-world applications.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251336951"},"PeriodicalIF":2.9,"publicationDate":"2025-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12099082/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144144284","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":"Scoping review of digital health technologies and interventions that target lifestyle behavior change in Singapore.","authors":"Nikita Kanumoory Mandyam, Jacqueline Lau, Camille Keck, Elya Chen, Alexander Wenjun Yip","doi":"10.1177/20552076251337032","DOIUrl":"10.1177/20552076251337032","url":null,"abstract":"<p><strong>Objective: </strong>A rise in non-communicable diseases, driven by poor lifestyle behaviors, demands a shift from conventional, reactive, and episodic care to next-gen, proactive, and real-time lifestyle management. Digital health technologies (DHTs) and digital health interventions (DHIs) are poised to shepherd this transformation. Given the proliferation of digital health applications, a scoping review was conducted to map the DHTs and DHIs targeting lifestyle behavior change in Singapore.</p><p><strong>Methods: </strong>A systematic search of PubMed, Scopus, and clinical trial registries and a manual search of gray literature and mobile app stores were conducted to identify patient-facing IoT (internet of things) technologies (wearables, apps, bots, websites) that target physical activity, diet, tobacco/alcohol use, sleep quality, and/or mindfulness in Singapore from 2013 to 2023.</p><p><strong>Results: </strong>Forty-six DHTs and thirty-five DHIs were identified. Apps were the most common while websites and bots were the least common. Most DHTs and DHIs played monitoring or preventative behavior change functions. Most applications targeted multiple behaviors, with physical activity being the most common and tobacco/alcohol use being the least common. Behavioral change strategies included feedback and monitoring (80%), goals and planning (70%), associations (62%), personalization (59%), gamification (54%), rewards (33%), human coaching (30%), and just-in-time adaptations (4%).</p><p><strong>Conclusion: </strong>This review identified gaps in digital health applications that address addictive behaviors and the elderly. Future efforts should prioritize adaptive, personalized technologies based on user-centric designs and robust behavioral frameworks to enhance long-term behavioral change. Insights from Singapore's experience can guide the global development of more effective health-improving digital applications.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251337032"},"PeriodicalIF":2.9,"publicationDate":"2025-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12099102/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144144312","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":"Design thinking in cancer care: A systematic literature review.","authors":"Muhammad Mohsin Memon, Noel Carroll, Katie Crowley","doi":"10.1177/20552076241313279","DOIUrl":"10.1177/20552076241313279","url":null,"abstract":"<p><strong>Objective: </strong>Cancer can have a profound impact on the life of the patient, presenting challenges such as dealing with complex healthcare models and psychological burden. Implementing design thinking (DT) in cancer care can improve the quality of life for patients. Although DT has been used in healthcare, there is limited research highlighting use of DT in cancer care. The objective of this review is to explore the applications of DT within a cancer care context.</p><p><strong>Methods: </strong>We systematically searched databases (PubMed Central, Scopus, and Medline) for relevant papers published between January 2018 and March 2023. Articles were identified using keywords: 'cancer', 'cancer care', 'oncology', 'design thinking', and 'design science'. Studies meeting our inclusion criteria were included and data was collected on the focus of study (i.e., design thinking and cancer care), target condition, target intervention and objective of the study. Thematic analysis was performed to identify recurring themes across studies. Articles were evaluated by the lead author and cross verified by the other two authors to reduce the risk of bias.</p><p><strong>Results: </strong>Twenty studies were included out of the 160 articles identified whereby 11 focus on cancer care (5 patient-facing, 5 community-facing, and 1 provider-facing studies) and 9 on design thinking (5 patient-facing, 1 community-facing, and 3 provider-facing studies). Overall, seven themes were identified with several subthemes.</p><p><strong>Conclusions: </strong>Our findings indicate that design thinking has been successfully applied to improve patient experiences in cancer care. By involving various stakeholders, including patients, healthcare providers, and communities, design thinking helps better understand real world problems. However, there is a gap in design thinking research concerning the long-term evaluation and scalability of design thinking-based interventions. Additionally, our findings suggest that mixed methods approach for future studies would support to establish more empirical evidence in this domain.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076241313279"},"PeriodicalIF":2.9,"publicationDate":"2025-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12099083/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144144306","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":"SSC-Net: A multi-task joint learning network for tongue image segmentation and multi-label classification.","authors":"Xiaopeng Sha, Zheng Guan, Ying Wang, Jinglu Han, Yi Wang, Zhaojun Chen","doi":"10.1177/20552076251343696","DOIUrl":"10.1177/20552076251343696","url":null,"abstract":"<p><strong>Background: </strong>Traditional Chinese medicine (TCM) tongue diagnosis, through the comprehensive observation of tongue's diverse characteristics, allows an understanding of the state of the body's viscera as well as Qi and blood levels. Automatic tongue image recognition methods could support TCM practitioners by providing auxiliary diagnostic suggestions. However, most learning-based methods often address a narrow scope of the tongue's attributes, failing to fully exploit the information contained within the tongue images.</p><p><strong>Objective: </strong>To classify multifaceted tongue characteristics, and fully utilize the latent correlation information between tongue segmentation and classification tasks, we proposed a multi-task joint learning network for simultaneous tongue body segmentation and multi-label Classification, named SSC-Net.</p><p><strong>Methods: </strong>Firstly, the shared feature encoder extracts features for both segmentation and classification tasks, where the segmentation result is utilized to mask redundant features that may impede classification accuracy. Subsequently, the ROI extraction module locates and extracts the tongue body region, and the feature fusion module combines tongue body features from bottom to top. Finally, a fine-grained classification module is employed for multi-label classification on multiple tongue characteristics.</p><p><strong>Results: </strong>To evaluate the performance of the SSC-Net, we collected a tongue image dataset, BUCM, and conducted extensive experiments on it. The experimental results show that the proposed method when segmenting and classifying simultaneously, achieved 0.9943 DSC for the segmentation task, 92.02 mAP, and 0.851 overall F1-score for the classification task.</p><p><strong>Conclusion: </strong>The proposed method can effectively classify multiple tongue characteristics with the support of the multi-task learning strategy and the integration of a fine-grained classification module. Code is available here.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251343696"},"PeriodicalIF":2.9,"publicationDate":"2025-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12099091/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144144325","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":"Ethical and privacy challenges of integrating generative AI into EHR systems in Tanzania: A scoping review with a policy perspective.","authors":"Augustino Mwogosi","doi":"10.1177/20552076251344385","DOIUrl":"10.1177/20552076251344385","url":null,"abstract":"<p><strong>Objectives: </strong>This study examines the ethical and privacy challenges of integrating generative artificial intelligence (AI) into electronic health record (EHR) systems, focusing on Tanzania's healthcare context. It critically analyses the extent to which Tanzania's Policy Framework for Artificial Intelligence in the Health Sector (2022) addresses these challenges and proposes regulatory and practical safeguards for responsible generative AI deployment.</p><p><strong>Methods: </strong>A systematic scoping review was conducted using PubMed, IEEE Xplore, Scopus and Google Scholar to identify relevant studies published between 2014 and 2024. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines informed the search and selection process. Fourteen studies met the inclusion criteria and were thematically analysed to identify key ethical and privacy concerns of generative AI in healthcare. Moreover, a policy analysis of Tanzania's AI framework was conducted to assess its alignment with global best practices and regulatory preparedness.</p><p><strong>Results: </strong>The review identified six key ethical and privacy challenges associated with generative AI in EHR systems: data privacy and security risks, algorithmic bias and fairness concerns, transparency and accountability issues, consent and autonomy challenges, human oversight gaps and risks of data re-identification. The policy analysis revealed that while Tanzania's AI framework aligns with national health priorities and promotes capacity building and ethical governance, it lacks generative AI-specific guidelines, regulatory clarity and resource mobilisation strategies necessary for healthcare settings.</p><p><strong>Conclusion: </strong>Integrating generative AI into Tanzania's EHR systems presents transformative opportunities and significant ethical and privacy risks. Tanzania's policy framework should incorporate AI-specific ethical guidelines, operationalise regulatory mechanisms, foster stakeholder engagement through participatory co-design and strengthen infrastructural investments. These measures will promote ethical integrity, enhance patient trust and position Tanzania as a regional leader in responsible AI use in healthcare.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251344385"},"PeriodicalIF":2.9,"publicationDate":"2025-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12093014/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144121405","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}