JMIR Medical Informatics最新文献

{"title":"Standardizing Disaster Health Data Visualization Using Fast Healthcare Interoperability Resources in Indonesia: Dashboard Development and Technical Evaluation.","authors":"Hiro Putra Faisal, Masaharu Nakayama","doi":"10.2196/98818","DOIUrl":"https://doi.org/10.2196/98818","url":null,"abstract":"<p><strong>Background: </strong>Health data management during disasters enables responders to assess the needs of survivors, efficiently allocate resources, and monitor survivors' health conditions. However, government agencies need to interpret health data in real time during disasters because the volume and complexity of such data can hinder timely decision-making.</p><p><strong>Objective: </strong>In this study, we aimed to develop a visual interface for Fast Healthcare Interoperability Resources (FHIR)-based data from various sources, using the daily report form of the World Health Organization (WHO) Emergency Medical Teams Minimum Data Set (EMT MDS) as a template and ensuring compliance with the SATUSEHAT platform.</p><p><strong>Methods: </strong>The proposed solution involves creating a dashboard application within a web-based system to visualize health data using FHIR, leveraging the WHO EMT MDS disaster profile. The interface includes Overview, Location, and Demography pages that display information in tables, graphs, and maps. This application was developed using JavaScript and React. The dataset consisted of 13,300 synthetic patient records generated using Synthea, subsequently modified with GPT-5, and stored on the Health Level 7 (HL7) Application Programming Interface (HAPI) FHIR server.</p><p><strong>Results: </strong>We developed a dashboard application based on the WHO EMT MDS Daily Reporting Form comprising 3 main pages: Overview, Demography, and Location. The Demography page contains several sections, including the total number of patients, health conditions, outcomes, relationships, and protection. The Location page lists all health care facilities, along with their corresponding locations, on a map. Each Location page displays detailed information regarding the organization and demographics of the respective location.</p><p><strong>Conclusions: </strong>The dashboard application demonstrates that health data collected in accordance with FHIR standards can be integrated and presented in a form suitable for disaster response. This dashboard has the potential to support disaster management agencies by enabling real-time, data-driven decision-making during disaster response.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e98818"},"PeriodicalIF":5.0,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13632908/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119165","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":"Development of an Interpretable Triage Tool for Colorectal Polyp Risk Stratification Within a Population-Based Screening Program: Machine Learning Approach.","authors":"Zhenmiao Ye, Jiajin Li, Yimin Xie, Qian Li, Yilun Huang, Guohua Zhang, Xue Yang","doi":"10.2196/89422","DOIUrl":"https://doi.org/10.2196/89422","url":null,"abstract":"<p><strong>Background: </strong>Colorectal polyps are a major source of precancerous lesions in colorectal cancer (CRC). In many population-based screening programs, a major challenge is the efficient triage of high-risk individuals for diagnostic colonoscopy amid limited endoscopic resources.</p><p><strong>Objective: </strong>To enrich current screening frameworks, we aimed to develop an accessible, noninvasive risk stratification tool to serve as a digital triage mechanism for colorectal polyps using machine learning (ML) and routinely collected data in China.</p><p><strong>Methods: </strong>We conducted a cross-sectional study in Wenzhou, China. A total of 4108 individuals (aged 50-74 y) who were referred for and accepted colonoscopy following an initial population-based risk assessment (questionnaire and fecal test) between May and November 2021 were included. The dataset was split into training and validation sets, and the synthetic minority oversampling technique (SMOTE) was applied only to the training dataset to address class imbalance. Twenty-one noninvasive predictors (lifestyle, dietary, clinical symptoms, and family history) were selected using the Boruta algorithm and least absolute shrinkage and selection operator (LASSO) regression. Nine ML models were evaluated, with the Shapley Additive Explanations (SHAP) method and local interpretable model-agnostic explanations (LIME) used for model interpretability and feature ranking.</p><p><strong>Results: </strong>Among the 9 ML algorithms evaluated, XGBoost (Extreme Gradient Boosting) achieved the highest area under the receiver operating characteristic curve of 0.672, while LightGBM (Light Gradient Boosting Machine) was identified as the optimal model for clinical triage due to its superior recall (0.6503), a key metric for minimizing missed lesions in community screenings. SHAP analysis identified current smoking status, sex, and family history of colorectal polyps as the most influential factors. Notably, the model captured significant nonlinear risk thresholds, such as an age of 50 years and a BMI of 25 kg/m², providing a more granular risk profile than traditional linear models.</p><p><strong>Conclusions: </strong>This study provides a scalable, interpretable triage tool to complement existing 2-step CRC screening protocols. By leveraging only noninvasive variables, the LightGBM model enables prioritized referral for colonoscopy, offering a resource-efficient strategy to optimize CRC prevention in resource-limited settings.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e89422"},"PeriodicalIF":5.0,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13626646/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119114","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":"Technology-Enhanced Health Care in Smart Homes: Scoping Review of Sensor Technologies, Clinical Applications, Integration Challenges, and Future Directions.","authors":"Yomna El-Saboni, Samira Awwal, Rakesh Mishra, Nityanand Sharma, Anshu Gaur, Leigh Fleming","doi":"10.2196/89189","DOIUrl":"https://doi.org/10.2196/89189","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Smart home technologies integrated with technology-enhanced health care (TEH) systems are transforming residential care by supporting independent living, continuous health monitoring, and remote clinical interventions. The Internet of Medical Things, wearable biosensors, and AI-driven analytics enable proactive health care delivery and personalized interventions, particularly for older adults and individuals with chronic conditions.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This review synthesizes literature on TEH integration within smart homes, examining global deployment patterns, technological maturity, biomedical sensor integration, machine learning applications, and health outcomes. It also identifies implementation challenges and disparities to improve digital health care strategies.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;A scoping review was conducted across PubMed, Scopus, Web of Science, ScienceDirect, and IEEE Xplore for peer-reviewed studies published between January 2005 and February 2025. Following screening of 6276 records, 169 studies were included, covering experimental, qualitative, and system design methodologies. Data were extracted on geographic deployment, sensor types, TEH architectures, machine learning algorithms, clinical outcomes, and adoption barriers.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;TEH adoption is concentrated in Europe, East and Southeast Asia, and higher-income countries, with potential emerging initiatives in West Asia in lower-income regions. Smart home maturity ranges from foundational systems with basic automation to connected ecosystems with centralized Internet of Things coordination, and intelligent systems with data-driven adaptive monitoring. The literature was synthesized across thematic domains, including sensor technologies, smart home infrastructure, predictive analytics, telehealth integration, and ethical and regulatory considerations, and interpreted through a 3-level maturity taxonomy of foundational, connected, and intelligent smart home systems. Integration of biomedical sensors can enable continuous monitoring of cardiovascular, respiratory, neurological, metabolic, and mobility parameters, while machine learning algorithms can support early disease detection, predictive health analytics, activity recognition, and personalized interventions. Evidence from current literature indicates remote monitoring improves early detection of health issues, chronic disease management, medication adherence, and psychological well-being. Several studies reported that remote monitoring systems improved early detection of health deterioration, chronic disease management, medication adherence, and patient well-being. Adoption barriers include interoperability challenges, data privacy, digital literacy gaps, social and economic disparities, and long-term sustainability concerns.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;Unlike previous work, this review emphasizes practical adoption barriers, interoperab","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e89189"},"PeriodicalIF":5.0,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13626297/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119142","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":"Machine Learning to Prioritize High-Severity Patient Safety Events for Institutional Investigation: Algorithm Development and Validation Study.","authors":"Hongbo Chen, Shehnaz Islam, Laura D Pozzobon, Lucas B Chartier, Eldan Cohen","doi":"10.2196/101393","DOIUrl":"10.2196/101393","url":null,"abstract":"<p><strong>Background: </strong>Patient safety events (PSEs) are preventable incidents that cause, or have the potential to cause, harm to patients during their medical journey. Although incident reporting systems capture large volumes of such events, only a small proportion undergo comprehensive investigation due to the resource-intensive nature of the review process. Patient safety specialists are tasked with triaging PSEs to prioritize high-severity cases that warrant timely institutional investigation; however, the rapidly increasing volume of events has made manual triage increasingly impractical.</p><p><strong>Objective: </strong>This study aimed to develop and validate ranking-based machine learning models for prioritizing high-severity PSEs and to compare their performance with conventional classification-based models and the reporter-assigned severity baseline.</p><p><strong>Methods: </strong>A total of 101,239 PSE reports were retrospectively extracted from a large Canadian academic health system between January 2017 and March 2025. Each report included a free-text incident description and a corresponding severity label denoting the level of severity, assigned through an institutional review process. Both feature-based and transformer-based models were developed using ranking and classification frameworks. Model performance was evaluated using ranking metrics, including average precision and normalized discounted cumulative gain (NDCG) at multiple cutoff thresholds (k=10, 20, 50, 100, 200), as well as Precision@k and Recall@k. The best-performing model was further benchmarked against the reporter-assigned severity baseline.</p><p><strong>Results: </strong>The ranking model, LLAMA3.1-RA, achieved the highest mean average precision of 0.94 (95% CI 0.90-0.98), representing a 22.1% relative improvement over the best classification model (LLAMA3.1-CL: mean 0.77, 95% CI 0.72-0.82; P<.001). LLAMA3.1-RA demonstrated superior performance across 10 of 16 evaluation metrics, with larger relative gains at broader evaluation cutoffs (eg, +12.9% in NDCG@200, +21.1% in Precision@200, and +20.3% in Recall@200). Compared with the reporter-assigned severity baseline, LLAMA3.1-RA achieved significantly higher performance across all 16 metrics, retrieving 83.0% of high-severity PSEs within the top 200 ranked events, compared with 41.0% retrieved under the reporter-assigned severity baseline.</p><p><strong>Conclusions: </strong>LLAMA3.1-RA demonstrated superior capability in prioritizing high-severity PSEs compared with both classification approaches and the reporter-assigned severity baseline. Integrating this model into triage workflows may offer a scalable and efficient solution for identifying and prioritizing high-severity safety events.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e101393"},"PeriodicalIF":5.0,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13662446/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149082897","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 Multi-Model, Pixel-Native Framework for Automated Computed Tomography Series Labeling and Characterization: Proof-of-Concept Study.","authors":"Yutong Wen, Anton Sheahan Quinsten, Cynthia Sabrina Schmidt, Christian Bojahr, Judith Kohnke, Kamyar Arzideh, Sina Warmer, Sebastian Blex, Ann-Christin Jacoby, Max Eberts, Hanna Lehmann, Olivia Barbara Pollok, Mathias Holtkamp, Luca Salhöfer, Lale Umutlu, Michael Forsting, Johannes Haubold, Felix Nensa, Katarzyna Borys, René Hosch","doi":"10.2196/93018","DOIUrl":"10.2196/93018","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Although Digital Imaging and Communications in Medicine (DICOM) metadata are widely used to manage medical imaging data and support clinical workflows, their suitability as a sole basis for automatic computed tomography (CT) series labeling and characterization is limited. DICOM metadata are frequently inconsistently populated, institution specific, use unregulated private tags, and have variable reliability even within standardized fields. Consequently, automated series selection for downstream AI applications often remains unreliable, necessitating manual curation within clinical workflows.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study presents Orchestrate, a modular AI framework for automated orchestration of CT imaging data. By integrating a hierarchy of deep learning models, Orchestrate enables pixel-level classification and routing of CT series and accurate metadata-independent identification of anatomical regions, contrast-enhanced series, and reconstruction kernels, supporting seamless downstream AI integration without manual curation.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Orchestrate combined 3 pretrained models for anatomical regions, landmarks, and body part classification and 4 newly developed You Only Look Once (YOLO) v8-based models to classify contrast enhancement, recognize reconstruction kernels, and infer laterality. Three datasets were used: an internal development dataset comprising 27,418 CT studies for individual model development, an internal framework evaluation dataset comprising 200 CT studies to assess the complete framework under a simulated real-world scenario, and an external dataset from The Cancer Imaging Archive comprising 100 CT studies. For the internal development data, reference standards were derived from complete and unambiguous DICOM metadata following institutional definitions. For the 2 evaluation datasets, 3 radiographers independently established reference standards as DICOM metadata were not assumed to be complete or consistent. Clinical utility was assessed through cohort selection tasks involving 3 predefined target cases, with 3 radiologists reviewing selection accuracy. The interrater agreement was assessed using the Fleiss κ. The model performance was evaluated using F&lt;sub&gt;1&lt;/sub&gt;-scores.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;DICOM metadata were incomplete in the internal framework evaluation dataset and external dataset, with missing rates of 42.9% (413/963) and 92.9% (105/113) for contrast enhancement, respectively, and a missing reconstruction kernel information rate of 0.4% (4/963) in the internal framework evaluation dataset. During model development, individual models achieved macro-F&lt;sub&gt;1&lt;/sub&gt;-scores ranging from 0.982 to 0.989. At the framework level, Orchestrate achieved high classification performance across internal (weighted F&lt;sub&gt;1&lt;/sub&gt;-score ranged from 0.920 to 1.000; macro-F&lt;sub&gt;1&lt;/sub&gt;-score ranged from 0.879 to 1.000) and external (weighted F&lt;sub&gt;1&lt;/sub&gt;-sco","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e93018"},"PeriodicalIF":5.0,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13662436/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149082912","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":"Implementation Readiness and Adoption of AI-Enabled Voice Electronic Medical Records in Resource-Constrained African Health Systems: Multisite Qualitative Study.","authors":"Melika Desalegn, Hocheol Lee, Dawit Wondifraw Tilahun, Alemu Sugi","doi":"10.2196/100805","DOIUrl":"10.2196/100805","url":null,"abstract":"<p><strong>Background: </strong>AI-enabled voice electronic medical records (EMRs) are increasingly promoted as tools to reduce clinician documentation burden; however, empirical evidence from multilingual, resource-constrained health systems in sub-Saharan Africa remains limited.</p><p><strong>Objective: </strong>This study examined health care professionals' perceptions, anticipated benefits, and concerns regarding AI voice-enabled EMRs in Ethiopia to inform context-sensitive implementation strategies.</p><p><strong>Methods: </strong>We conducted a qualitative, multisite study using semistructured written responses from 43 Ethiopian health care professionals recruited via purposive and maximum variation sampling between April 11, 2026, and April 21, 2026. Data were analyzed in Taguette using a hybrid deductive-inductive approach integrating 3 complementary frameworks: the Consolidated Framework for Implementation Research (CFIR), the technology acceptance model (TAM), and Normalization Process Theory (NPT). Analytical rigor was strengthened through independent dual coding, structured reconciliation, reflexive memos, and a version-controlled audit trail.</p><p><strong>Results: </strong>Four overarching themes were identified. First, participants anticipated clear clinical benefits, including reduced typing burden, improved documentation continuity, and enhanced patient interaction, yet expressed substantial concerns about automation errors, accent-related transcription failures, and persistent infrastructural instability. Second, usability barriers, including interface complexity, inadequate training, and digital anxiety, shaped technology acceptance across cadres. Third, participants anticipated shifts in workflow, task distribution, and clinical collaboration as documentation practices evolved. Finally, ethical and governance concerns, particularly regarding data confidentiality, unclear consent procedures, and fear of surveillance, emerged as major determinants of trust. Cross-framework synthesis revealed that adoption readiness was jointly shaped by organizational capacity, usability perceptions, emotional and cognitive responses, and evolving workflow expectations.</p><p><strong>Conclusions: </strong>Successful implementation of AI voice-enabled EMRs in Ethiopia requires coordinated investments in digital infrastructure, locally adapted language models, strengthened data governance, and iterative user onboarding. These findings underscore the urgency of context-sensitive and ethically grounded approaches when deploying speech-based AI in low-resource health systems.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e100805"},"PeriodicalIF":5.0,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13662444/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149082923","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":"Detecting Misspelled Drug Names Using Transformer-Based Language Models: Model Development and External Validation.","authors":"Jiayu Lu, Kevin W McConeghy, Andrew R Zullo, Jinying Chen","doi":"10.2196/91151","DOIUrl":"10.2196/91151","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Misspellings in medication names can compromise patient safety, reduce data utility, and impede large-scale data initiatives that integrate medication information from electronic health records (EHRs). Existing methods for detecting misspelled medical terms are mostly dictionary-based and can lead to high false-positive rates when correctly spelled but previously unseen (out-of-vocabulary) terms are encountered.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;We aimed to develop and validate domain-specific, transformer-based language models for detecting misspelled drug names, with an emphasis on performance for unseen terms.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Using RxNorm as a standardized drug vocabulary, we created an RxNorm-augmented training corpus and developed two BERT (Bidirectional Encoder Representations from Transformers)-based models-BERTDrug and CharBERTDrug-for misspelling detection. Specifically, we randomly split 69,824 RxNorm drug names into training, development, and test sets (3:1:1) and generated k misspellings per name using text-perturbation techniques (k optimized for training; fixed at 1 for development and test sets). The models were fine-tuned on the training and development sets and evaluated using the RxNorm test set and 3586 drug names from the Long-Term Care Data Cooperative (LTCDC) database (external validation). The RxNorm test set and out-of-vocabulary LTCDC dataset (1922 terms), neither overlapping with the RxNorm training data, were used to evaluate performance on unseen terms. SpellChecker served as a dictionary-based baseline, while fastText&lt;sub&gt;ML&lt;/sub&gt; and BioWordVec&lt;sub&gt;ML&lt;/sub&gt;, which used different subword embeddings as inputs for machine learning, served as additional baselines. Additionally, we compared model performance with GPT-4o, a generative large language model (LLM), using 2200 randomly sampled test terms.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;On the RxNorm test set, BERTDrug and CharBERTDrug outperformed the baseline models across most metrics. BERTDrug achieved the best overall performance (F&lt;sub&gt;1&lt;/sub&gt;-score=0.859; area under the receiver operating characteristic curve [ROC-AUC]=0.947), followed by CharBERTDrug (F&lt;sub&gt;1&lt;/sub&gt;-score=0.833; ROC-AUC=0.906). Both models also outperformed the baseline models on the out-of-vocabulary LTCDC dataset across most metrics, with CharBERTDrug performing best (F&lt;sub&gt;1&lt;/sub&gt;-score=0.696; ROC-AUC=0.788), followed by BERTDrug (F&lt;sub&gt;1&lt;/sub&gt;-score=0.669; ROC-AUC=0.786). In the secondary analysis, both models exceeded GPT-4o on most metrics (except Recall) for 2000 RxNorm terms. BERTDrug performed best (ROC-AUC=0.951; F&lt;sub&gt;1&lt;/sub&gt;-score=0.855), followed by CharBERTDrug (ROC-AUC=0.911; F&lt;sub&gt;1&lt;/sub&gt;-score=0.831) and GPT-4o (ROC-AUC=0.856; F&lt;sub&gt;1&lt;/sub&gt;-score=0.721). In contrast, among 200 randomly selected LTCDC terms, GPT-4o performed best on most metrics except precision and specificity.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;Domain-specific language m","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e91151"},"PeriodicalIF":5.0,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13662438/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149082862","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":"Health Care Access Barriers Among Reproductive-Age Women in East Africa: Development and Validation of Machine Learning Prediction Models Using DHS Data.","authors":"Jenberu Mekurianew Kelkay, Andualem Yalew Aschalew, Getachew Teshale, Melak Jejaw, Kaleb Assegid Demissie, Azmeraw Tadele, Misganaw Guadie Tiruneh, Tesfahun Zemene Tafere, Asebe Hagos, Nebebe Demis Baykemagn","doi":"10.2196/85695","DOIUrl":"10.2196/85695","url":null,"abstract":"<p><strong>Background: </strong>The World Health Organization advises that every nation should take responsibility for guaranteeing access to health care services as a basic human right. However, due to financial constraints and geographical hurdles, only around half of the population in Africa has access to contemporary health care services.</p><p><strong>Objective: </strong>This study aimed to predict barriers to health services and associated factors among reproductive-aged women in East Africa using machine learning algorithms and identify the best-performing predictive model.</p><p><strong>Methods: </strong>Analysis of secondary data from 6 East African countries using the Demographic and Health Surveys from 2016 to the recent 2023 was performed. A weighted total sample of 228,654 women of reproductive age was included in this study. Data were extracted and processed with Stata version 17. The dataset was then imported into a Jupyter notebook for further detailed analysis and visualization. A machine learning algorithm using different classification models was implemented. All analyses and calculations were performed in the Python 3 programming language in Jupyter Notebook using imblearn, scikit-learn, and Extreme Gradient Boosting (XGBoost) packages.</p><p><strong>Results: </strong>Among 228,654 reproductive-age women included in the study, the XGBoost classifier demonstrated the best predictive performance, with 94.46% accuracy, 94.62% precision, 93.73% recall, 94.17% <i>F</i>₁-score, and 98% area under the curve. Overall, 92.63% (211,794/228,654) of women experienced barriers to health care access. Predictors of barriers to health access were identified using an XGBoost model with the help of Shapley Additive Explanations. This study showed that maternal occupation, maternal age, low media exposure, parity, marital status, health insurance, and community literacy were the top predicting factors of barriers to health access.</p><p><strong>Conclusions: </strong>The XGBoost model demonstrated the best predictive performance among the evaluated algorithms. The findings indicate that a substantial proportion of reproductive-age women experience barriers to health care access, although no formal subnational \"extreme risk\" classification was conducted in this study. Enhancing comprehensive health education and reducing financial barriers through the expansion of health insurance coverage may help improve health care access, particularly among vulnerable populations such as rural women.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e85695"},"PeriodicalIF":5.0,"publicationDate":"2026-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13618409/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149076888","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":"Electronic Health Record Workflows in Acute Care Surgery: Ethnographic Study.","authors":"Alex H Lee, Devesh Narayanan, Kristan Staudenmayer, Aussama K Nassar, Syed Morad Hameed","doi":"10.2196/89846","DOIUrl":"10.2196/89846","url":null,"abstract":"<p><strong>Background: </strong>Systematic strategies to harness electronic health record (EHR) workflows, reduce redundancy, and support decision-making remain limited in acute care surgery (ACS). Understanding how EHR systems and workflows intersect with time-sensitive settings is critical to improving decision-making and outcomes in ACS.</p><p><strong>Objective: </strong>This study aimed to evaluate how ACS clinicians leverage the EHR for decision-making and to identify opportunities and challenges for EHR-enabled decision support.</p><p><strong>Methods: </strong>We conducted a qualitative ethnographic study over a 6-month period, combining in-depth interviews with 15 ACS surgeons and providers and 100 hours of \"paired fieldwork\" observations spanning the entire perioperative arc by a surgical provider and an organizational sociologist. Using constructivist grounded theory, we identified the enablers, challenges, and opportunities for EHR-enabled decision-making in ACS.</p><p><strong>Results: </strong>Surgeons fell into two groups: (1) those who accepted information overload as inherent to the EHR, relying on generic templates and standard attestations, and (2) others who viewed it as a problem to fix, actively correcting errors and composing individualized summaries. Ambiguity in billing requirements drove overdocumentation, resulting in \"note bloat\" that obscured high-yield information. EHR use during decision-making focused primarily on risk assessment, though navigation challenges hindered access to critical data. Forecasting key outcomes that alter management or facilitate shared decision-making was seen as valuable. Automated risk stratification, generated from live EHR data while minimizing alert fatigue, was seen as a potential solution.</p><p><strong>Conclusions: </strong>ACS clinicians use various tactics to navigate EHR challenges and focus on high-value tasks. Streamlined risk assessment using EHR data may strengthen decision-making in critical moments, but solutions must integrate seamlessly within existing workflows to provide rapid and accurate outputs that prioritize meaningful outcomes.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e89846"},"PeriodicalIF":5.0,"publicationDate":"2026-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13618404/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149076869","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":"SAFE_DTx: Safety-First Framework for AI-Driven Personalization in Digital Therapeutics.","authors":"Dohyoung Rim","doi":"10.2196/78202","DOIUrl":"10.2196/78202","url":null,"abstract":"<p><p>Digital therapeutics (DTx) are emerging as evidence-based software interventions, but current AI-driven personalization approaches lack dedicated safety-focused frameworks and face challenges due to scarce long-term outcome data and unpredictable model behaviors. We propose SAFE_DTx, a safety-first architectural framework for DTx that integrates well-established principles of predictive modeling and constrained decision-making to prioritize patient safety. SAFE_DTx's 2-module architecture comprises an AI feedback prediction module that forecasts short-term patient responses and a constrained planning module that selects the next intervention under explicit safety constraints. By decoupling these components and enforcing clear safety guardrails, the framework enables dynamic, real-time adaptation to individual patient feedback while staying within evidence-based safety limits. This modular design also enhances transparency in the decision-making process, and an in silico evaluation demonstrates its preliminary architectural feasibility, showing greater engagement and no safety violations compared to baseline strategies within the simulated environment. SAFE_DTx's safety-by-design architecture aligns with emerging regulatory emphasis on AI transparency and patient safety. It directly addresses key clinical challenges in AI-driven DTx personalization by ensuring that tailored interventions do not compromise patient safety.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e78202"},"PeriodicalIF":5.0,"publicationDate":"2026-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13653203/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149066616","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}