[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-463211-105":59,"doc-detail-463211-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","machine-learning-based-risk-stratification-for-sudden-cardiac-death-using-clinical-and-device-derived-data","Machine Learning-Based Risk Stratification for Sudden Cardiac Death Using Clinical and Device-Derived Data","","Sudden cardiac death (SCD) remains a major clinical challenge, with implantable cardioverter-defibrillators (ICDs) serving as the primary preventive intervention. Current patient selection relies on limited and imperfect risk markers. This study evaluates machine learning models to improve SCD risk prediction using tabular clinical data and features derived from sensing devices, including ECGs and ICDs. Models were trained on demographic, clinical, laboratory, and device-derived variables, optimized for F2-score to prioritize high-risk detection, and interpreted with post hoc SHAP analysis; random forest achieved the highest F2-score (0.74) with 97.30% recall.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":35,"@type":76,"position":81},"https://docshare.wps.com/document/healthcare/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/machine-learning-based-risk-stratification-for-sudden-cardiac-death-using-clinical-and-device-derived-data/463211/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/machine-learning-based-risk-stratification-for-sudden-cardiac-death-using-clinical-and-device-derived-data/463211.png","ImageObject",300,407,{"name":92,"@type":93},"Rainbow Cat","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-09-30",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",11,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What is the main problem addressed in the study?","Question",{"text":113,"@type":114},"Sudden cardiac death remains a major clinical challenge, and selecting patients for ICD implantation is currently based on limited risk markers.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"Which data sources are used to build the machine learning risk models?",{"text":118,"@type":114},"The models use tabular clinical inputs including demographic, clinical, laboratory, and device-derived variables from medical sensing such as ECGs and ICDs.",{"name":120,"@type":111,"acceptedAnswer":121},"How is model performance optimized and interpreted?",{"text":122,"@type":114},"Models are optimized for the F2-score to prioritize detection of high-risk patients, and interpretability is provided using post hoc SHAP value analysis to identify predictors.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},463211,1790812125,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":34,"category_name":35,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":145,"read_time":146,"preview_page_count":4},962090769181,"https://ap-avatar.wpscdn.com/davatar_6f874abed73319feea01a86fa6f0fab8","Article  \nMachine Learning-Based Risk Stratification for Sudden Cardiac Death Using Clinical and Device-Derived Data  \nHana Ivandic 1, Branimir Pervan 1, Mislav Puljevic 2,3, Vedran Velagic 2,3 and Alan Jovic 1, *  \nAcademic Editor: Biswanath Samanta  \nReceived: 19 November 2025  \nRevised: 15 December 2025  \nAccepted: 19 December 2025  \nPublished: 22 December 2025  \nCopyright: © 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.  \n1 University of Zagreb Faculty of Electrical Engineering and Computing, Unska 3, HR-10000 Zagreb, Croatia; [hana.ivandic@fer.unizg.hr](hana.ivandic@fer.unizg.hr) (H.I.); [branimir.pervan@fer.unizg.hr](branimir.pervan@fer.unizg.hr) (B.P.)  \n2 University Hospital Centre Zagreb, Ulica Mije Kišpati´ca 12, HR-10000 Zagreb, Croatia; [mislav.puljevic@kbc-zagreb.hr](mislav.puljevic@kbc-zagreb.hr) (M.P.); [vedran.velagic@kbc-zagreb.hr](vedran.velagic@kbc-zagreb.hr) (V.V.)  \n3 University of Zagreb School of Medicine, Šalata 2, HR-10000 Zagreb, Croatia  \n* Correspondence: [alan.jovic@fer.unizg.hr](alan.jovic@fer.unizg.hr)  \nAbstract  \nSudden cardiac death (SCD) remains a major clinical challenge, with implantable cardioverter-defibrillators (ICDs) serving as the primary preventive intervention. Current patient selection guidelines rely on limited and imperfect risk markers. This study explores the potential of machine learning (ML) models to improve SCD risk prediction using tabular clinical data that include features derived from medical sensing devices such as electrocardiograms (ECGs) and ICDs. Several ML models, including tree-based models, Naive Bayes (NB), logistic regression (LR), and voting classifiers (VC), were trained on demographic, clinical, laboratory, and device-derived variables from patients who underwent ICD implantation at a Croatian tertiary center. The target variable was the activation of the ICD device (appropriate or inappropriate/missed), serving as a surrogate for high-risk SCD detection. Models were optimized for the F2-score to prioritize high-risk patient detection, and interpretability was achieved with post hoc SHAP value analysis, which confirmed known and revealed additional potential SCD predictors. The random forest (RF) model achieved the highest F2-score (F2-score 0.74, AUC-ROC 0.73), demonstrating a recall of 97.30% and meeting the primary objective of high true positive detection, while the VC classifier achieved the highest overall discrimination (F2-score 0.71, AUC-ROC 0.76) . The predictive performance of multiple ML models, particularly the high recall they achieved, demonstrates the promising potential of ML to refine ICD patient selection.  \nKeywords: machine learning; SHAP analysis; sudden cardiac death; implantable cardioverter defibrillator  \n1. Introduction  \nSudden cardiac death (SCD) is an unexpected death caused by a very fast and dangerous heart rhythm that stops the heart from pumping blood. Most often, this rhythm starts in the ventricles (ventricular tachycardia or ventricular fibrillation) and can lead to collapse and death within minutes if an electrical shock is not delivered in time [1] . Because such a condition usually happens suddenly and often in patients who could otherwise be perceived as stable, SCD is a major clinical problem. It remains a leading cause of cardiovascular mortality, particularly among individuals with left ventricular dysfunction. To address and reduce this risk, contemporary cardiology mostly uses implantable cardioverter–defibrillators (ICDs) . ICDs are relatively small, implantable devices that continuously monitor the heart rhythm and can deliver an internal electric shock  \nwhen a life-threatening arrhythmia appears. Large clinical trials have shown that ICDs can improve survival in patients after myocardial infarction or with heart failure and reduced left ventricular ejection fraction","cbCais5vLLk1qx5V","https://ap.wps.com/l/cbCais5vLLk1qx5V","pdf",2345109,25,"English","# Introduction\n## ICD-based prevention and current selection criteria\n## Limitations of LVEF-based risk stratification\n## Motivation for individualized risk stratification with ML","[{\"question\":\"What is the main problem addressed in the study?\",\"answer\":\"Sudden cardiac death remains a major clinical challenge, and selecting patients for ICD implantation is currently based on limited risk markers.\"},{\"question\":\"Which data sources are used to build the machine learning risk models?\",\"answer\":\"The models use tabular clinical inputs including demographic, clinical, laboratory, and device-derived variables from medical sensing such as ECGs and ICDs.\"},{\"question\":\"How is model performance optimized and interpreted?\",\"answer\":\"Models are optimized for the F2-score to prioritize detection of high-risk patients, and interpretability is provided using post hoc SHAP value analysis to identify predictors.\"}]","Machine Learning-Based Risk Stratification for Sudden Cardiac Death Using Clinical and Device-Derived Data | PDF",1790766248,63]