JMIR Medical Informatics最新文献

{"title":"Cardiorenal Inter-organ Assessment: A Novel Clustering Method Using Dynamic Time Warping on ECG.","authors":"Sally Zhao, Zhan Ye, Bhavna Adhin, Matti Vuori, Jari Laukkanen, Sudeshna Fisch","doi":"10.2196/73353","DOIUrl":"https://doi.org/10.2196/73353","url":null,"abstract":"<p><strong>Background: </strong>The heart and kidneys have vital functions in the human body that reciprocally influence each physiologically. Pathological changes in one organ can damage the other. Epidemiologic studies show that greater than 50% of patients with heart failure (HF) have preserved ejection fraction (HFpEF). Additionally, one in six patients identified as having chronic kidney disease (CKD) also has HF. Thus, it is important to be able to predict and identify the cardiorenal relationship between HFpEF and CKD.</p><p><strong>Objective: </strong>Creating an ECG-enabled model that stratifies HFpEF suspected patients would help identify CKD enriched HFpEF clusters and phenogroups. Simultaneously, a minimal set of significant ECG features derived from the stratification model would aid precision medicine and practical diagnoses due to being more accessible and widely readable than a large set of clinical inputs. Furthermore, validation of the existing cardiorenal relationship using this ECG-enabled model may lead to better biological understanding.</p><p><strong>Methods: </strong>Using unsupervised clustering on all extractable ECG features from FinnGen, patients with an indication of HFpEF (filtered by LVEF ≥ 50% and NT-proBNP > 450 pg/mL) were categorized into different phenogroups and analyzed for CKD risk. After isolating significant predictive ECG features, unsupervised clustering and risk analysis were performed again to demonstrate the efficacy of using a minimal set of features for phenogrouping. These clusters were then compared to clusters formed using Dynamic Time Warping (DTW) on raw ECG time series electrical signals. Afterwards, these clusters were analyzed for CKD enrichment.</p><p><strong>Results: </strong>PR interval and QRS duration stood out as significant features and were used as the minimal feature set. After generating and comparing clusters (K-means with all extracted ECG features, K-means with minimal feature set, and DTW with full Lead II ECG waveform), the DTW generated clusters were most stable. ANOVA analysis also showed that several HFpEF clusters exhibited a deviation of CKD risk from baseline, allowing for further trajectory analysis. Specifically, the creatinine levels (a proxy for CKD) of several DTW created clusters showed significant differences from average. Based off Jaccard score, the DTW clusters also showed the greatest alignment to baseline comparison clusters created by clustering on creatinine. In comparison, the other two sets of clusters (created by all extracted ECG features and the minimal set) performed similarly.</p><p><strong>Conclusions: </strong>This project validates both the known cardiorenal relationship between HFpEF and CKD and the importance of the PR interval and QRS duration. After exploring the use of ECG data for patient clustering and stratification, DTW clustering with Lead II waveforms resulted in the most clinically meaningful clusters in the context of HFpEF and CKD. T","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":" ","pages":""},"PeriodicalIF":3.1,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144310904","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction: Toward a Domain-Overarching Metadata Schema for Making Health Research Studies FAIR (Findable, Accessible, Interoperable, and Reusable): Development of the NFDI4Health Metadata Schema.","authors":"Haitham Abaza, Aliaksandra Shutsko, Sophie A I Klopfenstein, Carina N Vorisek, Carsten Oliver Schmidt, Claudia Brünings-Kuppe, Vera Clemens, Johannes Darms, Sabine Hanß, Timm Intemann, Franziska Jannasch, Elisa Kasbohm, Birte Lindstädt, Matthias Löbe, Katharina Nimptsch, Ute Nöthlings, Marisabel Gonzalez Ocanto, Tracy Bonsu Osei, Ines Perrar, Manuela Peters, Tobias Pischon, Ulrich Sax, Matthias B Schulze, Florian Schwarz, Carolina Schwedhelm, Sylvia Thun, Dagmar Waltemath, Atinkut A Zeleke, Wolfgang Müller, Martin Golebiewski","doi":"10.2196/78151","DOIUrl":"10.2196/78151","url":null,"abstract":"<p><p>[This corrects the article DOI: 10.2196/63906.].</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"13 ","pages":"e78151"},"PeriodicalIF":3.1,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12177417/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144227816","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":"Service Users' Perspectives on an Integrated Electronic Care Record in Mental Health Care: Qualitative Vignette and Interview Study.","authors":"Timothy Kariotis, Megan Prictor, Kathleen Gray, Shanton Chang","doi":"10.2196/64162","DOIUrl":"10.2196/64162","url":null,"abstract":"<p><strong>Background: </strong>There have been suggestions that electronic health records (EHRs) should be expanded beyond clinical mental health care services to a broader array of care services that support mental health service users, which we call an integrated electronic care record (IECR). Previous research has considered service users' general views on information being stored and shared via an EHR. However, little consideration has been given to service users' attitudes toward how EHRs should be used in the provision of care or the concept of an IECR.</p><p><strong>Objective: </strong>This study aimed to understand mental health care service users' perspectives on an IECR and how it should be used in practice when receiving care.</p><p><strong>Methods: </strong>Ten people with lived experience of accessing multiple services in Australia's mental health care system were provided with 2 vignettes that depicted fictional service users making decisions about an IECR. Participants were asked to respond to several scenarios that the fictional service users might experience in their journey through the mental health care system with an IECR. Participants provided written responses and took part in a semistructured interview to discuss their responses. An interpretative phenomenological analysis was undertaken, which led to 5 major themes and 15 subthemes being developed.</p><p><strong>Results: </strong>Service users wanted an IECR that they had control over, supported them as equal partners in their care, and contributed toward more collaborative and proactive mental health care. However, participants were concerned that care professionals' perspectives would be privileged in the IECR and overshadow service users' needs. Participants also had concerns that stigmatizing and discriminatory information documented in their IECR would negatively impact their interactions with the mental health care system and their access to care. Participants saw value in an IECR bringing together information to support collaborative and proactive care. However, participants thought that the benefits of the IECR had to be balanced with potential risks to their privacy. Participants thought that the IECR should contain only information relevant to their care and should be shared only with relevant care professionals. There were concerns that service users might lack the skills, resources, and information required to manage their IECR.</p><p><strong>Conclusions: </strong>An IECR has the potential to fill the gaps in an increasingly complex and fragmented mental health care system but risks entrenching service users' experiences of stigma and discrimination unless service users are meaningfully involved in their IECR.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"13 ","pages":"e64162"},"PeriodicalIF":3.1,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12174869/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144217649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Developing an Intelligent Mobile Clinic-A Medical Vehicle for Improve Access to Healthcare in Remote Areas: Evidence From China.","authors":"Xinlei Chen, Xufang Huang, Yimiao Xu, Jiabin Xu, Yanan Wang, Xinyue Ren, Xuebo Zhu, Xiaoge Xie, Yeqin Yang","doi":"10.2196/59103","DOIUrl":"10.2196/59103","url":null,"abstract":"<p><strong>Unlabelled: </strong>Lishui, a mountainous city in Zhejiang Province, China, is characterized by extensive mountainous terrain and a dispersed population. To address this issue, Lishui has introduced the intelligent mobile clinic service. This model leverages 5G technology and integrates the benefits of mobile clinics and telehealth, tailored to the region's geography and demographic characteristics. The intelligent mobile clinic uses real-time data analysis to deliver medical services effectively to remote mountainous areas. Medical personnel from the intelligent mobile clinic visit villages to provide in-person professional health services and facilitate residents' access to higher-level hospital resources through telehealth. This model has received widespread praise and recognition from residents and has yielded significant outcomes. During 2018-2023, a total of 25,000 visits have been made, benefiting 648,000 individuals. The intelligent mobile clinic provides a valuable reference for enhancing health care access in similar regions.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"13 ","pages":"e59103"},"PeriodicalIF":3.1,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12148251/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144210379","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":"Enhancing Antidiabetic Drug Selection Using Transformers: Machine-Learning Model Development.","authors":"Hisashi Kurasawa, Kayo Waki, Tomohisa Seki, Eri Nakahara, Akinori Fujino, Nagisa Shiomi, Hiroshi Nakashima, Kazuhiko Ohe","doi":"10.2196/67748","DOIUrl":"10.2196/67748","url":null,"abstract":"<p><strong>Background: </strong>Diabetes affects millions worldwide. Primary care physicians provide a significant portion of care, and they often struggle with selecting appropriate medications.</p><p><strong>Objective: </strong>This study aimed to develop a model that accurately predicts what drug an endocrinologist would prescribe based on the current measurements. The goal was to create a system that would assist nonspecialists in choosing medications, thereby potentially improving diabetes treatment outcomes. Based on the performance of previous studies, we set a performance target of achieving a receiver operating characteristic area under the curve (ROC-AUC) above 0.95.</p><p><strong>Methods: </strong>A transformer-based encoder-decoder model predicts whether 44 types of diabetes drugs will be prescribed. The model uses sequences of age, sex, history for 12 laboratory tests, and prescribed drug history as inputs. We assessed the model using the electronic health records from 7034 patients with diabetes seeing endocrinologists between 2012 and 2022 at the University of Tokyo Hospital. We assessed model performance trained on data subsets spanning different time periods (2, 5, and 10 years) using micro- and macro-averaged ROC-AUC on a hold-out test set comprising data solely from 2022. The model's performance was compared against LightGBM.</p><p><strong>Results: </strong>The model trained on data from the past 5 years (2017-2021) yielded the best predictive performance, achieving a microaverage (95% CI) ROC-AUC of 0.993 (0.992-0.994) and a macroaverage (95% CI) ROC-AUC of 0.988 (0.980-0.993). The model achieved an ROC-AUC above 0.95 for 43 out of 44 drugs. These results surpassed the predefined performance target and outperformed both previous studies and the LightGBM model's microaverage ROC-AUC of 0.988 (0.985-0.990) in terms of prediction accuracy. Furthermore, training the model with short-term data from the past 5 years yielded high accuracy compared to using data from the past 10 years, suggesting that learning from more recent prescribing patterns might be advantageous.</p><p><strong>Conclusions: </strong>The proposed model demonstrates the feasibility of accurately predicting the next prescribed drugs. This model, trained from the past prescriptions of endocrinologists, has the potential to provide information that can assist nonspecialists in making diabetes-treatment decisions. Future studies will focus on incorporating important factors such as prescription contraindications and constraints to enhance safety, as well as leveraging large-scale clinical data across multiple hospitals to improve the generalizability of the model.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"13 ","pages":"e67748"},"PeriodicalIF":3.1,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12148250/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144210380","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":"Intelligent Prediction Platform for Sepsis Risk Based on Real-Time Dynamic Temporal Features: Design Study.","authors":"Mingwei Zhang, Ming Zhong, Yunzhang Cheng, Tianyi Zhang","doi":"10.2196/74940","DOIUrl":"10.2196/74940","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;The development of sepsis in the intensive care unit (ICU) is rapid, the golden rescue time is short, and the effective way to reduce mortality is rapid diagnosis and early warning. Therefore, real-time prediction models play a key role in the clinical diagnosis and management of sepsis. However, the existing sepsis prediction models based on artificial intelligence still have limitations, such as poor real-time performance and insufficient interpretation.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;Our objective is to develop a real-time sepsis prediction model that integrates high timeliness and clinical interpretability. The model is designed to dynamically predict the risk of sepsis in ICU patients and establish a practical, tailored sepsis prediction platform.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Within a retrospective analysis framework, the model comprises a real-time prediction module and an interpretability module. The real-time prediction module leverages 3-hour dynamic temporal features derived from 8 noninvasive, real-time physiological indicators: heart rate, respiratory rate, blood oxygen saturation, mean arterial pressure, systolic blood pressure, diastolic blood pressure, body temperature, and blood glucose. Three linear parameters (mean, SD, and endpoint value) were calculated to construct the prediction model using multiple ML algorithms. The interpretability module uses the TreeSHAP (Tree-Based Shapley Additive Explanations) method to enhance model transparency through both individual prediction and global explanations. Further, it added a link between the output interpretation of the explainable artificial intelligence method and its potential physiological or pathophysiological significance, including the relationship among the output interpretation and the patient's hemodynamics, thermoregulatory response, and the balance between oxygen delivery and oxygen consumption. Finally, a web-based platform was developed to integrate prediction and interpretability functions.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The sepsis prediction model demonstrated robust performance in the test cohort (224 patients), achieving an accuracy of 0.7 (95% CI 0.68-0.71), precision of 0.69 (95% CI 0.68-0.71), F&lt;sub&gt;1&lt;/sub&gt;-score of 0.69 (95% CI 0.67-0.70), and area under the receiver operating characteristic curve of 0.76 (95% CI 0.74-0.77). The TreeSHAP method effectively visualized feature contributions, enabling clinicians to interpret the model's prediction logic and identify anomalies. The link between the output interpretation of the model and its potential physiological or pathophysiological significance improved the interpretability and credibility of the explainable artificial intelligence method. The web-based platform significantly enhanced clinical utility by providing real-time risk assessment, statistical summaries, trend analysis, and actionable insights.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;This platform provides real-time dy","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"13 ","pages":"e74940"},"PeriodicalIF":3.1,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12166316/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144188578","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 Novel Framework to Assess Clinical Information in Digital Health Technologies: Cross-Sectional Survey Study.","authors":"Kayode Philip Fadahunsi, Petra A Wark, Nikolaos Mastellos, Ana Luisa Neves, Joseph Gallagher, Azeem Majeed, Josip Car","doi":"10.2196/58125","DOIUrl":"10.2196/58125","url":null,"abstract":"<p><strong>Background: </strong>Digital health is a critical driver of quality, safety, and efficiency in health care. However, poor quality of clinical information in digital health technologies (DHTs) can compromise the quality and safety of care. The Clinical Information Quality (CLIQ) framework was developed, based on a systemic review of literature and an international eDelphi study, as a tool to assess the quality of clinical information in DHTs.</p><p><strong>Objectives: </strong>The aim of this study is to assess the applicability, internal consistency, and construct validity of the CLIQ framework.</p><p><strong>Methods: </strong>This study was conducted as a cross-sectional survey of health care professionals across the United Kingdom who regularly use SystmOne electronic health records. Participants were invited through emails and social media platforms. The CLIQ questionnaire was administered as a web-based survey. Spearman correlation coefficients were computed to investigate the linear relationship between the dimensions in the CLIQ framework. The Cronbach α coefficients were computed to assess the internal consistency of the global scale (ie, CLIQ framework) and the subscales (ie, the informativeness, availability, and usability categories). Confirmatory factor analysis was used to assess the extent to which the survey data supported the construct validity of the CLIQ framework.</p><p><strong>Results: </strong>A total of 109 health care professionals completed the survey, of which two-thirds (67, 61.5%) were doctors and a quarter (26, 23.9%) were nurses or advanced nurse practitioners. Overall, the CLIQ dimensions had good quality scores except for portability, which had a modest score. The inter-item correlations were all positive and not likely due to chance. The Cronbach α coefficient for the overall CLIQ framework was 0.89 (95% CI 0.85-0.92). The confirmatory factor analysis provided a modest support for the construct validity of the CLIQ framework with the comparative fit index of 0.86 and standardized root mean square residual of 0.08.</p><p><strong>Conclusions: </strong>The CLIQ framework demonstrated a high reliability and a modest construct validity. The CLIQ framework offers a pragmatic approach to assessing the quality of clinical information in DHTs and could be applied as part of information quality assurance systems in health care settings to improve quality of health information.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"13 ","pages":"e58125"},"PeriodicalIF":3.1,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12143851/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144188576","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":"Experience of Cardiovascular and Cerebrovascular Disease Surgery Patients: Sentiment Analysis Using the Korean Bidirectional Encoder Representations from Transformers (KoBERT) Model.","authors":"Hocheol Lee, Yu Seong Hwang, Ye Jun Kim, Yukyung Park, Heui Sug Jo","doi":"10.2196/65127","DOIUrl":"10.2196/65127","url":null,"abstract":"<p><strong>Background: </strong>Cardiovascular and cerebrovascular diseases significantly contribute to global mortality and disability. The shift to outpatient postoperative care, accelerated by the COVID-19 pandemic, emphasizes the need for effective management of postoperative outcomes. The high rates of cardiovascular and cerebrovascular diseases in Korea necessitate focused transitional care during patient discharge periods. However, limited research exists on the postoperative experiences of discharged patients, underscoring the necessity of establishing evidence-based services to optimize transitional care.</p><p><strong>Objective: </strong>The objective of this paper was to analyze the emotional experiences of patients who underwent cardiovascular and cerebrovascular surgeries using data from Naver, a major South Korean web portal.</p><p><strong>Methods: </strong>Posts were collected using specific keywords and processed with the Korean Bidirectional Encoder Representations from Transformers (KoBERT) model based on Transformer, which classified sentiments into positive, neutral, and negative categories. Model performance was validated according to precision, recall, F1-score, and support. Sentiment analysis was conducted within the Transitional Care Model (TCM) framework, divided into 5 domains: health status, care resources, care demand, interaction, and mental state.</p><p><strong>Results: </strong>The KoBERT model demonstrated high classification performance, achieving a precision of 96%, recall of 94%, and an F1-score of 94%. Sentiment analysis revealed that compared with cardiovascular surgery patients, cerebrovascular surgery patients experienced higher negative emotions regarding health status, whereas cardiovascular surgery patients expressed more negative sentiments in care demands.</p><p><strong>Conclusions: </strong>Different patient groups experience distinct emotional and practical challenges postdischarge. Particularly, keywords within the TCM framework highlight that cerebrovascular surgery patients require robust rehabilitation and caregiver support, whereas cardiovascular surgery patients need better cost management. These findings underscore the importance of personalized transitional care strategies tailored for cardiovascular and cerebrovascular diseases. The insights derived from this study can guide health care policymakers in designing more targeted and patient-centered interventions to improve postdischarge care and patient-centered transitional care, ensuring continuous and effective postoperative management.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"13 ","pages":"e65127"},"PeriodicalIF":3.1,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12143735/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144188577","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 and Deep Learning Techniques for Prediction and Diagnosis of Leptospirosis: Systematic Literature Review.","authors":"Suhila Sawesi, Arya Jadhav, Bushra Rashrash","doi":"10.2196/67859","DOIUrl":"10.2196/67859","url":null,"abstract":"<p><strong>Background: </strong>Leptospirosis, a zoonotic disease caused by Leptospira bacteria, continues to pose significant public health risks, particularly in tropical and subtropical regions.</p><p><strong>Objective: </strong>This systematic review aimed to evaluate the application of machine learning (ML) and deep learning (DL) techniques in predicting and diagnosing leptospirosis, focusing on the most used algorithms, validation methods, data types, and performance metrics.</p><p><strong>Methods: </strong>Using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (CHARMS), and Prediction model Risk of Bias Assessment Tool (PROBAST) tools, we conducted a comprehensive review of studies applying ML and DL models for leptospirosis detection and prediction, examining algorithm performance, data sources, and validation approaches.</p><p><strong>Results: </strong>Out of a total of 374 articles screened, 17 studies were included in the qualitative synthesis, representing approximately 4.5% of the initial pool. The review identified frequent use of algorithms such as support vector machines, artificial neural networks, decision trees, and convolutional neural networks (CNNs). Among the included studies, 88% (15/17) used traditional ML methods, and 24% (4/17) used DL techniques. Several models demonstrated high predictive performance, with reported accuracy rates ranging from 80% to 98%, notably with the U-Net CNN achieving 98.02% accuracy. However, public datasets were underused, with only 35% (6/17) of studies incorporating publicly available data sources; the majority (65%, 11/17) relied primarily on private datasets from hospitals, clinical records, or regional surveillance systems.</p><p><strong>Conclusions: </strong>ML and DL techniques demonstrate potential for improving leptospirosis prediction and diagnosis, but future research should focus on using larger, more diverse datasets, adopting transfer learning strategies, and integrating advanced ensemble and validation techniques to strengthen model accuracy and generalization.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":"13 ","pages":"e67859"},"PeriodicalIF":3.1,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12140502/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144181334","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 for the Prediction of Acute Kidney Injury in Critically Ill Patients With Coronary Heart Disease: Algorithm Development and Validation.","authors":"Yike Li, Mingyang Xiao, Yaqian Li, Lulu Lv, Shanshan Zhang, Yuhui Liu, Juan Zhang","doi":"10.2196/72349","DOIUrl":"10.2196/72349","url":null,"abstract":"<p><strong>Background: </strong>Acute kidney injury (AKI) frequently occurs in critically ill patients with coronary heart disease (CHD), and its development markedly elevates mortality rates and prolongs hospitalization duration. Early AKI prediction is crucial for timely intervention and amelioration of patient outcomes.</p><p><strong>Objective: </strong>This study aimed to develop and verify a clinical prediction model for the occurrence of AKI upon admission in the critically ill population with CHD through machine learning (ML).</p><p><strong>Methods: </strong>Data from the MIMIC-IV (Medical Information Mart for Intensive Care IV) version 2.2 database were gathered and included information about critically ill individuals with CHD in the intensive care unit (ICU). The dataset was randomized into a training set (70%) and a testing set (30%). Least absolute shrinkage and selection operator (LASSO) regression was used for feature variable selection. ML models, including logistic regression (LR), decision tree (DT), naive Bayes (NB), random forest (RF), extreme gradient boosting (XGBoost), and support vector machine (SVM), were constructed using 13 variables in the training set. The 6 models were compared in the testing set to identify the best-performing model. Subsequently, the model was assessed using calibration curve analysis and decision curve analysis (DCA). External validation was conducted using data from the Second Affiliated Hospital of Zhengzhou University. Ultimately, the predictive model was interpreted via Shapley Additive Explanation (SHAP) values.</p><p><strong>Results: </strong>In total, 2711 patients with CHD admitted to the ICU were selected, with 1809 (66.7%) having AKI. XGBoost exhibited the best performance regarding discrimination (area under the receiver operating characteristic curve [AUROC]=0.765, 95% CI 0.731-0.800), accuracy (0.725), and sensitivity (0.759). External validation using a cohort of 226 patients confirmed the strong generalizability of the XGBoost model (AUROC=0.835, 95% CI 0.782-0.887). Feature importance analyses derived from SHAP values, DT, RF, and XGBoost consistently identified 5 key predictors associated with the development of AKI: mechanical ventilation, use of antiplatelet agents, age, N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels, and acute physiology score III (APSIII).</p><p><strong>Conclusions: </strong>ML models can serve as reliable tools for forecasting AKI in the critically ill population with CHD. The XGBoost model is highly accurate and may aid doctors in identifying high-risk individuals for early intervention to lower mortality.</p>","PeriodicalId":56334,"journal":{"name":"JMIR Medical Informatics","volume":" ","pages":"e72349"},"PeriodicalIF":3.1,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12159552/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144095943","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}