[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-290304-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-290304-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","usefulness-of-heat-map-explanations-for-deep-learning-based-electrocardiogram-analysis","Usefulness of Heat Map Explanations for Deep-Learning-Based Electrocardiogram Analysis","","Deep neural networks are powerful for analyzing high-dimensional medical data but their complexity makes predictions hard to understand and limits clinical trust. This study uses transfer learning to predict sex from electrocardiograms and applies Grad-CAM to generate heat maps for visual explanations. Feedback from medical doctors assesses whether the heat maps capture recognizable ECG features for discrimination. Results show the explanations provide no meaningful information for clinicians in this setting and are not useful in practice. Improved, data-tailored explanation methods are required before clinical deployment.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/usefulness-of-heat-map-explanations-for-deep-learning-based-electrocardiogram-analysis/290304/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/usefulness-of-heat-map-explanations-for-deep-learning-based-electrocardiogram-analysis/290304.png","ImageObject",300,407,{"name":42,"@type":43},"Olivia Brown","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-08","2026-09-17",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",10,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"How does the study generate heat map explanations for ECG predictions?","Question",{"text":63,"@type":64},"It trains a deep neural network with transfer learning to predict sex from electrocardiograms, then uses Grad-CAM to produce heat maps that reflect how the model supports its predictions.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What role did medical doctors play in evaluating the heat maps?",{"text":68,"@type":64},"Medical doctors provided feedback on whether the heat maps highlighted ECG features that could be recognized for distinguishing sex.",{"name":70,"@type":61,"acceptedAnswer":71},"What was the main conclusion about heat map usefulness in the clinical context?",{"text":72,"@type":64},"In this setting, the explainable AI heat maps did not provide meaningful information to medical doctors and were not useful in the clinic.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},290304,1790472330,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":55,"slug":125},"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":145,"preview_page_count":4},16904993612988,"https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd","diagnostics  \nArticle  \nUsefulness of Heat Map Explanations for Deep-Learning-Based Electrocardiogram Analysis  \nAndrea M. Storås 1,2, *, Ole Emil Andersen 3,4, Sam Lockhart 5, Roman Thielemann 6, Filip Gnesin 7, Vajira Thambawita 1, Steven A. Hicks 1, Jørgen K. Kanters 8, Inga Strümke 9, Pål Halvorsen 1,2 and Michael A. Riegler 1,10  \nCitation: Storås, A.M.; Andersen, O.E.; Lockhart, S.; Thielemann, R.; Gnesin, F.; Thambawita, V.; Hicks, S.A.; Kanters, J.K.; Strümke, I.; Halvorsen, P.; et al. Usefulness of Heat Map Explanations for Deep-Learning-Based Electrocardiogram Analysis. Diagnostics 2023, 13, 2345 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)diagnostics13142345  \nAcademic Editor: Shaker El-Sappagh  \nReceived: 7 June 2023  \nRevised: 6 July 2023  \nAccepted: 10 July 2023  \nPublished: 11 July 2023  \nCopyright: © 2023 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 ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) .  \n1 Department of Holistic Systems, Simula Metropolitan Center for Digital Engineering, 0167 Oslo, Norway  \n2 Department of Computer Science, Oslo Metropolitan University, 0130 Oslo, Norway  \n3 Department of Public Health, Aarhus University, 8000 Aarhus, Denmark  \n4 Steno Diabetes Center, Aarhus University, 8000 Aarhus, Denmark  \n5 Wellcome Trust-Medical Research Council Institute of Metabolic Science, University of Cambridge, Cambridge CB2 0QQ, UK  \n6 Novo Nordisk Foundation Center for Basic Metabolic Research, University of Copenhagen, 2200 Copenhagen, Denmark  \n7 Department of Cardiology, North Zealand Hospital, 3400 Hillerød, Denmark  \n8 Department of Biomedical Sciences, University of Copenhagen, 2200 Copenhagen, Denmark  \n9 Department of Computer Science, Norwegian University of Science and Technology, 7491 Trondheim, Norway  \n10 Department of Computer Science, UiT The Arctic University of Norway, 9037 Tromsø, Norway  \n* Correspondence: [anan@oslomet.no](anan@oslomet.no)  \nAbstract: Deep neural networks are complex machine learning models that have shown promising results in analyzing high-dimensional data such as those collected from medical examinations. Such models have the potential to provide fast and accurate medical diagnoses. However, the high complexity makes deep neural networks and their predictions difficult to understand. Providing model explanations can be a way of increasing the understanding of “blackbox” models and building trust. In this work, we applied transfer learning to develop a deep neural network to predict sex from electrocardiograms. Using the visual explanation method Grad-CAM, heat maps were generated from the model in order to understand how it makes predictions. To evaluate the usefulness of the heat maps and determine if the heat maps identified electrocardiogram features that could be recognized to discriminate sex, medical doctors provided feedback. Based on the feedback, we concluded that, in our setting, this mode of explainable artificial intelligence does not provide meaningful information to medical doctors and is not useful in the clinic. Our results indicate that improved explanation techniques that are tailored to medical data should be developed before deep neural networks can be applied in the clinic for diagnostic purposes.  \nKeywords: explainable artificial intelligence; electrocardiograms; heat maps  \n1. Introduction  \nElectrocardiograms (ECGs) are widely used in hospitals and clinics to evaluate the electrical currents in the heart and discover anomalies such as Myocardial Infarction (MI) . The method provides high-resolution time series data and is fast, inexpensive, and easily accessible, making ECGs an obvious target for Artificial Intelligence (AI) . Machine Learning (ML) is a subfield ","cbCaise1GHlP0lOI","https://ap.wps.com/l/cbCaise1GHlP0lOI","pdf",1554734,11,"English","# Abstract\n# 1. Introduction\n## ECGs and machine learning for diagnosis\n## Explainable AI and trust in clinical settings\n## Prior evidence on heat maps in medical domains","[{\"question\":\"How does the study generate heat map explanations for ECG predictions?\",\"answer\":\"It trains a deep neural network with transfer learning to predict sex from electrocardiograms, then uses Grad-CAM to produce heat maps that reflect how the model supports its predictions.\"},{\"question\":\"What role did medical doctors play in evaluating the heat maps?\",\"answer\":\"Medical doctors provided feedback on whether the heat maps highlighted ECG features that could be recognized for distinguishing sex.\"},{\"question\":\"What was the main conclusion about heat map usefulness in the clinical context?\",\"answer\":\"In this setting, the explainable AI heat maps did not provide meaningful information to medical doctors and were not useful in the clinic.\"}]","Usefulness of Heat Map Explanations for Deep-Learning-Based Electrocardiogram Analysis | PDF",1789641281,28]