JMIR Medical Informatics最新文献

{"title":"Real-World Use of Controlled Terminologies, Ontologies, and Vocabularies for Evidence Generation Across a Large International Observational Network: Challenges and Lessons Learned From a Mixed Method Study.","authors":"Anna Ostropolets, Vlad Korsik, Tatsiana Skuhareuskaya, Aleh Zhuk, Maryia Khitrun, Alexander Davydov, Dmitry Dymshyts, Christian Reich, George Hripcsak, Patrick Ryan","doi":"10.2196/92727","DOIUrl":"10.2196/92727","url":null,"abstract":"<p><strong>Background: </strong>Large-scale international real-world evidence generation benefits from terminology harmonization. Despite widespread adoption of standardized vocabularies, their effective use and long-term sustainability at scale remain poorly understood.</p><p><strong>Objective: </strong>This study aimed to examine real-world terminology and code use in data across a federated network of observational data sources.</p><p><strong>Methods: </strong>We conducted a 2-part survey of researchers and data owners within the Observational Health Data Sciences and Informatics community on their terminology use, challenges, and needs, accompanied by the analysis of code use across a subset of real-world data sources.</p><p><strong>Results: </strong>The survey covered 144 institutions across the United States, the United Kingdom, Europe, Asia, and Africa. Data on terminology use covered 60 sources, including 22 data sources that provided detailed code-level use information. We observed significant variations in terminology use, with 61 out of 89 terminologies used in the data present in less than 10% of the data sources. Code use even after data harmonization was also highly variable: less than 1% (95/742,337) of codes were found in all data sources. Mapping and hierarchy completeness, terminology coverage, versioning, and terminology changes were among the most common challenges. We outlined several of our subsequent process improvements: community contribution and stewardship pipelines, metadata for relationships, and informatics tools for assessment of the impact of terminology change.</p><p><strong>Conclusions: </strong>Terminology and coding inconsistencies across observational data sources require a standardized terminology system. Such a system is complex and time-consuming and needs community contribution and informatics solutions for harmonization to be scalable and sustainable. Even with a common reference standard, high heterogeneity of terminology and code use across different observational data sources remains.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e92727"},"PeriodicalIF":5.0,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13577442/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148978759","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":"Automated Renal Tumor Segmentation in Computed Tomography Images Using a Global Attention-Based DeepLabV3+ Model: Algorithm Development and Validation.","authors":"Yueyan Zhao, Jianqiang Liu, Lingyu Shao, Lin Li, Zhaoqing Liu, Yujie Liu, Jiaxin Wen, Xinyao Hao, Shuyan Li, Jianhong Zhao, Boming Song","doi":"10.2196/78523","DOIUrl":"10.2196/78523","url":null,"abstract":"<p><strong>Background: </strong>The rising global incidence of renal tumors necessitates precise diagnostic interventions. Accurate segmentation of computed tomography (CT) scans is essential for nephron-sparing surgery and radiotherapy. However, conventional manual delineation is labor-intensive and prone to significant interobserver variability due to tumor morphological heterogeneity. There is an urgent clinical demand for robust, automated segmentation solutions.</p><p><strong>Objective: </strong>This study aims to develop and validate GAM-DeepLabV3+, an automated framework designed to address boundary ambiguity and high false-positive rates in complex renal imaging scenarios.</p><p><strong>Methods: </strong>We propose an optimized encoder-decoder architecture specifically tailored for renal mass detection. The framework incorporates three key innovations: (1) a lightweight MobileNetV2 backbone to minimize computational overhead for clinical deployment; (2) an Atrous Spatial Pyramid Pooling (ASPP) module to capture multiscale contextual information; and (3) a Global Attention Mechanism (GAM) in the decoder to enhance channel-spatial interactions, thereby refining boundary delineation by suppressing background noise. The model was rigorously evaluated on a private clinical dataset (n=218) and the KiTS19 benchmark (n=210).</p><p><strong>Results: </strong>GAM-DeepLabV3+ consistently outperformed state-of-the-art baselines. On the private dataset, the model achieved a mean Dice similarity coefficient (DSC) of 0.939 (SD 0.008), significantly surpassing feature pyramid network (FPN; mean 0.893, SD 0.008; <i>P</i><.001) and no-new-Net (nnU-Net; 2D, custom; mean 0.908, SD 0.013; <i>P</i><.001). It also achieved a mean 95% Hausdorff distance (HD95) of 1.485 (SD 0.522) pixels. On the KiTS19 dataset, it maintained a mean robust DSC of 0.928 (SD 0.006). To facilitate clinical translation, a demonstration-only online platform was developed.</p><p><strong>Conclusions: </strong>The GAM-DeepLabV3+ framework provides an accurate, efficient, and fully automated solution for renal tumor segmentation. By overcoming boundary ambiguity and optimizing feature fusion, this approach shows potential as a decision-support aid, pending future validation with 3D reconstruction.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e78523"},"PeriodicalIF":5.0,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13579559/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148978753","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":"Locally Deployed Large Language Models for AI-Assisted Outpatient Prescription Review: Crossover Study.","authors":"Zhengyue Liu, Yi Ding, Jingxia Chen, Ziqiang Yan, Xuxi Cheng, Wei Zhou, Peng Fu, Zhuo Wang","doi":"10.2196/97520","DOIUrl":"10.2196/97520","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Pharmacists' prescription review is key to medication safety, but rising numbers of outpatient prescriptions and expanding formularies leave less time per case, increasing error risk. Large language models (LLMs) show promise, yet two barriers hinder routine use: first, most systems rely on cloud-based commercial models, risking data breaches by transmitting protected patient information externally, and second, retrieval-augmented generation (RAG) pipelines used to reduce hallucinations depend on complex text vectorization and vector databases, which hospitals with limited IT resources struggle to build and maintain.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to evaluate the feasibility and usefulness of a locally deployed, knowledge-augmented LLM as a decision-support tool for pharmacist-led outpatient prescription review.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;The open-source Qwen3-14B model was deployed on a hospital intranet server using the Ollama framework. A structured knowledge base derived from drug package inserts served as the principal reference and supported lightweight knowledge augmentation through exact-match injection. A 2-period crossover design was used: 2 pharmacists independently reviewed 213 outpatient prescriptions under both unaided and AI-assisted conditions, yielding paired unaided and collaborative results for each prescription. Plain AI review and knowledge-augmented AI review were also run on the same 213-prescription test set to quantify hallucination rates, and stand-alone AI performance was assessed against the reference standard. The reference standard was established by independent consensus between two supervising pharmacists, with disagreements adjudicated by a deputy chief pharmacist. Accuracy, sensitivity, and specificity were compared between conditions using paired McNemar tests, and review time was compared using the Wilcoxon signed-rank test.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Overall review accuracy was 97.2% (207/213, 95% CI 94.0%-98.7%) in the human-AI collaborative condition vs 82.6% (176/213, 95% CI 77.0%-87.1%) in the pharmacist-alone condition (P&lt;.001). Sensitivity was 98% (61/62) vs 55% (34/62), and the false-negative rate fell from 45% to 2% (P&lt;.001). Specificity did not differ significantly (146/151, 96.7% vs 142/151, 94%; P=.29). Knowledge augmentation reduced the model hallucination rate from 19.7% (42/213 prescriptions) to 4.7% (10/213), an absolute reduction of 15.0 percentage points (relative reduction 76.2%; P&lt;.001). Collaborative review reduced the mean per-prescription review time from 2.33 (SD 0.97) to 1.12 (SD 0.49) minutes (approximately 51.9% reduction; Wilcoxon signed-rank Z=-12.65; P&lt;.001; r=0.87).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;A locally deployed, knowledge-augmented LLM used as a pharmacist-supervised prescreening tool was associated with substantially higher accuracy and sensitivity in outpatient retrospective prescription review, while keeping","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e97520"},"PeriodicalIF":5.0,"publicationDate":"2026-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13620432/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148972499","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-Based Prediction of Culture-Confirmed Neonatal Sepsis in a Tertiary Neonatal Intensive Care Unit: Retrospective Cohort Study.","authors":"Eman Badran, Oraib Al-Smadi, Loiy T Algazo, Alaa T Al Ghazo, Arwa Al Anber, Areej Sharaqa, Lena Abu-Argoub, Shatha Al Jaberi, Abdulrahman Alhanbali, Taimein Yacoub, Shahd Rihan, Hala Al-Jaberi","doi":"10.2196/88732","DOIUrl":"10.2196/88732","url":null,"abstract":"<p><strong>Background: </strong>Neonatal sepsis remains a major cause of neonatal morbidity and mortality in low- and middle-income countries (LMICs). Early diagnosis is challenging because of nonspecific clinical manifestations and delays in laboratory confirmation. Machine learning (ML) approaches using structured electronic health record (EHR) data may improve early risk stratification in neonatal intensive care units (NICUs).</p><p><strong>Objective: </strong>This study aimed to evaluate ML models for predicting culture-confirmed neonatal sepsis among neonates admitted to a tertiary NICU in Jordan, with the objective of addressing diagnostic gaps in resource-limited settings. Specifically, we aimed to identify key predictors through feature importance analysis, evaluate model performance with class-imbalanced data, and propose strategies to improve interpretability and generalizability in LMICs.</p><p><strong>Methods: </strong>A retrospective cohort study was conducted using structured EHRs of 3274 neonates admitted to a tertiary NICU in Jordan between 2018 and 2024. Neonates who underwent blood culture testing were included. The dataset was divided into training (n=2619, 80%) and testing (n=655, 20%) subsets using stratified sampling. Three ML models-Extreme Gradient Boosting (XGBoost), decision trees, and neural networks-were trained using clinical, laboratory, and demographic variables. Class imbalance was addressed using the synthetic minority oversampling technique (SMOTE) applied to the training dataset. Model performance was evaluated using accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC).</p><p><strong>Results: </strong>Among 3274 neonates included in the study, the XGBoost model demonstrated the best predictive performance on the independent test set (n=655, 20%), achieving an accuracy of 94% (616/655 correct predictions, 95% CI 92% to 96%), sensitivity of 98% (95/97 sepsis cases correctly identified, 95% CI 96% to 99%), and an AUC of 0.98 (95% CI 0.97 to 0.99). Decision trees provided interpretable classification rules with moderate performance, whereas neural networks showed lower discriminative ability, with an AUC of 0.81 (95% CI 0.78 to 0.84). Important predictive features included C-reactive protein, platelet count, and gestational age.</p><p><strong>Conclusions: </strong>XGBoost demonstrated strong predictive performance in this retrospective cohort, supporting its potential as a foundation for future prospective clinical decision support tools. External validation and prospective studies are required before clinical implementation.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e88732"},"PeriodicalIF":5.0,"publicationDate":"2026-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13573726/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148972448","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":"Reinforcement Learning-Based Temporal Knowledge Graph Reasoning for Predicting Chronic Gastritis Diagnosis and Treatment: Development and Validation Study.","authors":"Xiaolong Qu, Zhou Sun, Yuhang Wang, Haiyu Liu, Lei Yao, Dongmei Li, Guanli Song, Runshun Zhang, Xiaoping Zhang","doi":"10.2196/83544","DOIUrl":"10.2196/83544","url":null,"abstract":"<p><strong>Background: </strong>The clinical progression of chronic gastritis involves intricate temporal dependencies, which makes it difficult to capture both the dynamic trajectory of the disease and the underlying relationships among medical events using conventional methods.</p><p><strong>Objective: </strong>This study aims to dynamically predict chronic gastritis diagnosis. We propose RL4TKGR, a reinforcement learning-based temporal knowledge graph (TKG) reasoning model, to predict a chronic gastritis diagnosis.</p><p><strong>Methods: </strong>RL4TKGR incorporates a reinforcement learning policy network that uses a dual-path encoding module to model historical and nonhistorical diagnostic information separately. RL4TKGR further integrates a dual-channel reward function with a dynamic weight allocation mechanism, which adaptively balances the two information sources. This design addresses the strategic bias problem and enables interpretable reasoning. The chronic gastritis temporal knowledge graph (CG-TKG) was constructed from chronic gastritis diagnosis and treatment records of 17,906 patients comprising 38,360 visits from March 2009 to October 2022.</p><p><strong>Results: </strong>Experiments on the self-constructed CG-TKG resulted in RL4TKGR achieving the highest mean reciprocal rank (MRR) on CG-TKG-4 (51.66, SD 0.76), CG-TKG-8 (53.18, SD 0.79), and CG-TKG-12 (56.95, SD 0.84). After adding recent TKG reasoning baselines, adaptive path-memory network (DaeMon) achieved the strongest baseline MRR for all 3 subsets, with MRR values of 43.26 (SD 0.95), 51.76 (SD 1.12), and 54.87 (SD 1.17), respectively. Compared with DaeMon, the overall paired <i>t</i> tests across MRR and Hits@1/3/10 yielded 𝑃<.001 for all 3 subsets. Ablation experiments and case analyses further supported the contribution of the historical modeling components and the practical utility of the model for predicting disease subtypes and therapeutic medications.</p><p><strong>Conclusions: </strong>This study improves chronic gastritis diagnosis and treatment prediction by integrating a history-aware dual-path policy network with a dynamic event balance factor in reinforcement learning-based TKG reasoning.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e83544"},"PeriodicalIF":5.0,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13578680/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148978755","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":"Privacy Leakage in Federated Learning in Radiology Reports: Comparative Evaluation of Tokenizer and Batch-Size Privacy Risks.","authors":"Santhosh Parampottupadam, Andrés Martínez Mora, Dimitrios Bounias, Sinem Sav, Klaus Maier-Hein, Ralf Floca","doi":"10.2196/88390","DOIUrl":"10.2196/88390","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Federated learning (FL) enables multi-institutional model training on clinical text without sharing raw data; however, gradient inversion methods can reconstruct sensitive information from shared model updates. The extent of such privacy leakage in FL applied to radiology reports, and the role of tokenizer design, remains unclear.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to quantify gradient-based reconstruction of radiology report text in an FL setting and to compare privacy risk across 3 transformer tokenization strategies in a controlled, tokenizer-aware evaluation.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Six FL clients trained a GPT-2-style transformer (sequence length 32) on 2 public clinical-text corpora comprising 368,751 diagnostic reports, 98,206 discharge summaries, and 1500 MIMIC-CXR (Medical Information Mart for Intensive Care Chest X-Ray) radiology reports. Models were trained using 3 tokenizers (GPT-2, RadBERT, and LLaMA-2) with batch sizes of 64, 128, and 256. An active malicious-server threat model was assumed, and analytic gradient inversion was applied to recover text. Reconstruction fidelity was measured over 5 runs using exact sentence accuracy, sentence-level bilingual evaluation understudy (S-BLEU), and recall-oriented understudy for gisting evaluation (ROUGE-L).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Exact sentence reconstruction ranged from 27% to 75% across tokenizers, datasets, and batch sizes. At batch size 64 on the discharge dataset, accuracy was 64.7% (GPT-2), 70% (RadBERT), and 67.5% (LLaMA-2), decreasing to 27.3%, 28.5%, and 27.5% at batch size 256. S-BLEU declined with increasing batch size (eg, discharge reports from 0.69 to 0.31). Reconstruction fidelity did not differ significantly across tokenizers (all but one of 27 comparisons nonsignificant; none significant after Holm correction), and approximately 75% of clinical concepts were represented in the reconstructed text (a corpus-level upper bound), regardless of tokenizer. Batch size was the dominant factor governing leakage.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;Under a worst-case malicious server that tampers with the shared model and observes unprotected per-client gradients (no secure aggregation or differential privacy), substantial portions of radiology-report text can be reconstructed, with up to approximately 75% of reconstructed 32-token sequences and 75% of clinical concepts (not direct patient identifiers) recovered from FL gradients. In a controlled ablation holding model architecture fixed, tokenizer choice, including domain-specific tokenizers, did not significantly affect leakage under the evaluated conditions, whereas batch size was the primary determinant, and no tokenizer significantly reduced the risk. Tokenizer selection should therefore not be treated as a privacy safeguard in this setting. Safeguards such as secure aggregation and differential privacy should therefore be evaluated as candidate protections for FL d","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e88390"},"PeriodicalIF":5.0,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13568693/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148946385","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":"Quantifying the Impact of Anonymization-Induced Clinical Data Quality Loss: Methodological Quantitative Case Study Using Primary Diagnosis Codes and Hospital Length of Stay.","authors":"Gaetan Kamdje Wabo, Piotr Pawel Sokolowski, Mahboubeh Jannesari Ladani, Michael Hagmann, Thomas Ganslandt, Fabian Siegel","doi":"10.2196/97661","DOIUrl":"10.2196/97661","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;The secondary use of electronic health record data requires robust privacy protection. k-Anonymity is widely used to enable data sharing by ensuring that each quasi-identifier combination occurs in at least k records; yet, its analytical impact on clinically meaningful structures remains insufficiently characterized, particularly for the combination of record suppression and microaggregation that arises when a numeric attribute lacks a natural generalization hierarchy. A further gap is that anonymization tools report internal information-loss values but do not signal the downstream distributional and inferential distortions these transformations introduce.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study evaluated the analytical footprint of k-anonymity at k=5, 10, and 15 on 2 core data elements in retrospective hospital research: primary &lt;i&gt;International Classification of Diseases, 10th Revision, German Modification&lt;/i&gt; (ICD-10-GM) diagnosis codes, and hospital length of stay (LOS). It aimed to determine and quantify whether anonymization introduces meaningful distortions not captured by the anonymization tool itself, and whether diagnosis-specific LOS patterns remain reproducible after anonymization.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We analyzed 719,387 inpatient encounters from University Hospital Mannheim from 2010 to 2024. Anonymization was performed with the ARX tool. It used record suppression and microaggregation. Distributional distortion was assessed with the Kolmogorov-Smirnov &lt;i&gt;D&lt;/i&gt; statistic, quantile shifts, IQR changes, and tail changes. Categorical fidelity was assessed with the Jaccard coefficient and Cramer &lt;i&gt;V&lt;/i&gt;. Inferential reproducibility was assessed with a 3-level linear mixed model. The model included random intercepts for the three-character diagnosis codes from the &lt;i&gt;International Classification of Diseases&lt;/i&gt; (&lt;i&gt;ICD-3&lt;/i&gt;) and patients. We compared the intraclass correlation coefficient and diagnosis-level effect concordance. Concordance was quantified using Spearman ρ and Lin concordance correlation coefficient, both with 95% CIs. A composite traffic-light verdict summarized the results.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;ARX masked quasi-identifier cells rather than deleting rows; the proportion of encounters with a masked cell rose from 0.77% (k=5) to 2.62% (k=15), distributed almost uniformly across admission years. Kolmogorov-Smirnov &lt;i&gt;D&lt;/i&gt; was stable at 0.147. Median LOS shifted by 1 day, and the SD declined by about 6.5 days, while the diagnosis-level mean changed considerably (absolute mean shift of -1.81 days). Jaccard overlap fell from 0.624 to 0.421. The diagnosis intraclass correlation coefficient rose from 0.294 to 0.837, reflecting variance compression rather than improved signal. Best linear unbiased prediction rank concordance (Spearman ρ 0.964-0.970) and aggregate magnitude agreement (Lin concordance correlation coefficient 0.959-0.966) were high; yet, about 7% of low-si","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e97661"},"PeriodicalIF":5.0,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13566071/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148946311","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":"Open-Source R Shiny Dashboard (Global Burden of Disease Analysis Suite) for Multimethod Burden of Disease Research: Development and Evaluation Study.","authors":"Belal Mohamed Hamed, Mohammed Tarek Hasan, Omar Ahmed Abdelwahab, Moaz Elsayed Abouelmagd, Ramez M Odat, Yasar Sattar, Ramy Mohamed Ghazy","doi":"10.2196/98718","DOIUrl":"10.2196/98718","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;The Global Burden of Disease (GBD) study provides the most comprehensive global burden estimates available, but translating GBD data downloads into structured analyses requires an integrated, multimethod workflow spanning trend estimation, changepoint detection, demographic decomposition, frontier benchmarking, and health inequality measurement. Existing tools cover portions of this workflow, but none integrate it end-to-end in a code-free interface, creating a substantial technical barrier, particularly for investigators in resource-limited settings.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to develop and release the GBD Analysis Suite, an open-source R Shiny dashboard that consolidates 11 commonly used analytical modules for GBD-style burden-of-disease research into a single, code-free, web interface supporting user-uploaded data, and to describe its underlying architecture, analytical methodology, and intended applications.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;The application was built using R 4.5.3 and the Shiny framework, integrating established statistical packages, including forecast, segmented, openxlsx, ggplot2, plotly, and others, into modules covering data import, choropleth mapping, estimated annual percentage change, joinpoint regression, Kitagawa and Das Gupta decomposition, multimodel time-series forecasting, descriptive visualization, sociodemographic index-based frontier analysis, descriptive age-period-cohort visualization, slope and relative indices of inequality, and batch export. The application accepts Global Health Data Exchange (GHDx)-formatted CSV or Excel files, automatically detects standard variable names, and enforces internal consistency through built-in validation checks. Kitagawa decomposition uses the symmetric mid-point formulation to ensure exact additivity. The three ensemble forecast models (autoregressive integrated moving average [ARIMA], exponential smoothing state-space [ETS], and a feed-forward neural network [NNETAR]) apply Box-Cox log transformation, with an explicit nonnegativity floor applied to the point forecast and prediction intervals of every model.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The application is publicly deployed and can be accessed through any modern web browser without requiring local installation. On the bundled 196-country, 34-year demonstration dataset, exporting the full set of outputs completes in approximately 12 seconds at a peak memory footprint of approximately 600 MB. A feature comparison against 5 existing tools (GBD Compare, GBD Foresight, NCI Joinpoint, BAPC, and NordPred) across 14 capability dimensions confirms that the GBD Analysis Suite is the only reviewed platform combining user-uploaded data input with code-free access to all 6 core analytical methods of modern GBD papers. Source code is publicly available under the MIT license.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;The GBD Analysis Suite lowers the technical barrier to reproducible, mu","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e98718"},"PeriodicalIF":5.0,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13615381/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148946376","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":"Survivor Treatment Selection Bias in Evaluations of Virtual Transition of Care.","authors":"Hongmei Yang, Christopher D Russell, Amy Arnder Greene, Angela D Humphrey","doi":"10.2196/97119","DOIUrl":"10.2196/97119","url":null,"abstract":"","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e97119"},"PeriodicalIF":5.0,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13613042/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148931986","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":"Global Research Trends and Insights on AI Usage in Tuberculosis: Bibliometric Analysis.","authors":"Jiawen He, Shuzhen Feng, Yufang Chen, Zhijuan Lai, Beihua Lyu, Yingying Li","doi":"10.2196/93885","DOIUrl":"10.2196/93885","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Tuberculosis (TB) remains a major global health challenge despite prevention efforts. AI offers promising approaches to long-standing challenges in TB diagnosis, drug resistance detection, and case management; however, a systematic mapping of global research priorities and translational gaps in AI applications to TB is currently lacking.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to analyze development patterns, collaboration networks, and knowledge structure of AI applications in TB management through bibliometric analysis of Web of Science publications, identifying research frontiers and translational gaps.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;This study conducted a bibliometric analysis of Web of Science Core Collection publications (2000-2025; last searched April 4, 2025). Original articles and reviews containing both AI- and TB-related terms were screened by 2 independent reviewers following PRISMA 2020 principles. Coauthorship, cocitation, and keyword cooccurrence networks were constructed with VOSviewer (Nees Jan van Eck and Ludo Waltman, Leiden University), CiteSpace (Dr. Chaomei Chen, Drexel University), and Bibliometrix R package (Massimo Aria and Corrado Cuccurullo, University of Naples Federico II), and results were synthesized through descriptive statistics and scientific knowledge mapping.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Analysis of 1300 articles shows substantial growth in AI applications for TB management during 2000-2025, with post-2018 publication growth of 29.7% and citation growth of 54.2% annually. The United States dominated output (346 publications; 12,143 citations), whereas high-burden countries remained underrepresented relative to disease burden, revealing a systematic research-demand inversion. Of the 15 most cited papers, 4 (26.7%) focused on imaging analysis, 2 (13.3%) on drug resistance prediction, and 3 (20.0%) on drug discovery, together comprising 60% (9/15) of the highly cited literature; the remainder addressed foundational biology, decentralized-learning methodology, immunoinformatics, and review literature rather than a specific TB application domain. Across this literature, technical validation-stage work markedly outweighed papers reporting clinical application, underscoring a technology-over-translation imbalance; key gaps persist in cross-regional data sharing, atypical lesion recognition, and socioeconomic factor integration. These findings indicate that clinicians and program managers in high-burden settings should prioritize locally validated, implementation-ready AI tools over algorithmic refinements, and that funders should direct resources toward prospective clinical trials and equity-focused deployment frameworks.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;AI-TB research has expanded rapidly, yet clinical translation remains disproportionately limited relative to algorithmic innovation. To help bridge this gap, frontline clinicians in high-burden settings could prioritiz","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"14 ","pages":"e93885"},"PeriodicalIF":5.0,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13561195/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148931984","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}