xGNN4MI: Explainability of Graph Neural Networks in 12-lead Electrocardiography for Cardiovascular Disease Classification

xGNN4MI: Explainability of Graph Neural Networks in 12-lead Electrocardiography for Cardiovascular Disease Classification | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article xGNN4MI: Explainability of Graph Neural Networks in 12-lead Electrocardiography for Cardiovascular Disease Classification Miriam Cindy Maurer, Philip Hempel, Kristin Elisabeth Steinhaus, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7721630/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 06 Feb, 2026 Read the published version in npj Digital Medicine → Version 1 posted 13 You are reading this latest preprint version Abstract The clinical deployment of artificial intelligence (AI) solutions for assessing cardiovascular disease (CVD) risk in 12-lead electrocardiography (ECG) is hindered by limitations in interpretability and explainability. To address this, we present xGNN4MI, an open-source framework for graph neural networks (GNNs) in ECG modeling for interpretable CVD prediction. Our framework facilitates modeling clinically relevant spatial relationships between ECG leads and their temporal dynamics. We also developed specific explainable AI (XAI) and visualization tools to identify ECG leads crucial to the model's decision-making process, enabling a systematic comparison with established clinical knowledge. We evaluated xGNN4MI on two challenging tasks: diagnostic superclass classification and localization of myocardial infarction. Our findings show that the interpretable ECG-GNN models demonstrate good performance across the tasks. XAI analysis revealed clinically meaningful training effects, such as differentiating between anteroseptal and inferior myocardial infarction. Our work demonstrates the potential of ECG-GNNs for providing trustworthy and interpretable AI-based CVD diagnosis. Health sciences/Cardiology Biological sciences/Computational biology and bioinformatics AI-ECG Explainable AI Graph Neural Networks Deep Learning Myocardial Infarction Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 06 Feb, 2026 Read the published version in npj Digital Medicine → Version 1 posted Editorial decision: Revision requested 01 Nov, 2025 Reviews received at journal 30 Oct, 2025 Reviews received at journal 20 Oct, 2025 Reviews received at journal 15 Oct, 2025 Reviewers agreed at journal 10 Oct, 2025 Reviewers agreed at journal 09 Oct, 2025 Reviewers agreed at journal 09 Oct, 2025 Reviewers agreed at journal 09 Oct, 2025 Reviewers agreed at journal 09 Oct, 2025 Reviewers invited by journal 09 Oct, 2025 Editor assigned by journal 03 Oct, 2025 Submission checks completed at journal 03 Oct, 2025 First submitted to journal 26 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7721630","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":532081724,"identity":"cf329390-faad-4abd-9cf3-022637699c58","order_by":0,"name":"Miriam Cindy Maurer","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABdklEQVRIie2RT0sCQRTAnw2MHWbb64b/PkEwsWCIoV/FRWgvKwmBeJASgvVinrdvYZeNbisD22VRuileEkE6FGxEUpTSuNauJN6D9ncZ5r35vfeGBxAS8hdBABY/CEDBD/EAXqY8sP9yo4ILqwqVfisBvkKoX9lTguIeew1hZL1rufhBVKNb/Ro7FmPOa9utgHIdBfTgVj9Pd2KNu3EZssmlkmZR2rkwiyTTfKRIs1nmslUyB0YXlJtzwBnD4YPFnRPZAFX+UTBYgmkR2te4ghmljmAOBR2UNhOnsqBzRdKOYgSYUveVzsxX5ozmHTIZzjwFsCzMPUX94MpZoDC/S0nnXQjBw8hSQWOhvlBUG3GlEAzGEvwv1JmUWamlUsnB6UGzK8mLLhHDlnd1SUMxQtX97y49Gz0/mbk8vS1ejbVplopNNOm/VQ4T7Z6FXLeWFFOGOnoh1WxqbS/b1Fq5edvAUrApur5JiN6vhZC7MRUSEhLyj/gCIqyJIYFd6+kAAAAASUVORK5CYII=","orcid":"","institution":"University Medical Center Göttingen","correspondingAuthor":true,"prefix":"","firstName":"Miriam","middleName":"Cindy","lastName":"Maurer","suffix":""},{"id":532081725,"identity":"e980a705-b159-4da5-9be7-340a1209006c","order_by":1,"name":"Philip Hempel","email":"","orcid":"","institution":"University Medical Center Göttingen","correspondingAuthor":false,"prefix":"","firstName":"Philip","middleName":"","lastName":"Hempel","suffix":""},{"id":532081726,"identity":"ada10418-37f8-4953-9c8c-7f1f850433d0","order_by":2,"name":"Kristin Elisabeth Steinhaus","email":"","orcid":"","institution":"University Medical Center Göttingen","correspondingAuthor":false,"prefix":"","firstName":"Kristin","middleName":"Elisabeth","lastName":"Steinhaus","suffix":""},{"id":532081728,"identity":"b5570b0e-cbe7-4fe1-9723-7bcc88df9e3d","order_by":3,"name":"Hryhorii Chereda","email":"","orcid":"","institution":"Friedrich-Alexander University Erlangen-Nürnberg","correspondingAuthor":false,"prefix":"","firstName":"Hryhorii","middleName":"","lastName":"Chereda","suffix":""},{"id":532081731,"identity":"c3988b34-a005-47e7-95e7-5026014dfea7","order_by":4,"name":"Marcus Vollmer","email":"","orcid":"","institution":"University Medicine Greifswald","correspondingAuthor":false,"prefix":"","firstName":"Marcus","middleName":"","lastName":"Vollmer","suffix":""},{"id":532081733,"identity":"3dc51d6b-b16d-4460-9cb9-38cccefc37a2","order_by":5,"name":"Dagmar Krefting","email":"","orcid":"","institution":"University Medical Center Göttingen","correspondingAuthor":false,"prefix":"","firstName":"Dagmar","middleName":"","lastName":"Krefting","suffix":""},{"id":532081735,"identity":"8f83b43c-e2c4-4893-b6b5-c10187b5e4fb","order_by":6,"name":"Nicolai Spicher","email":"","orcid":"","institution":"Technical University of Denmark","correspondingAuthor":false,"prefix":"","firstName":"Nicolai","middleName":"","lastName":"Spicher","suffix":""},{"id":532081737,"identity":"c3b1c28e-6864-48f2-9a76-f7dcca9428fb","order_by":7,"name":"Anne-Christin Hauschild","email":"","orcid":"","institution":"Justus-Liebig University Giessen","correspondingAuthor":false,"prefix":"","firstName":"Anne-Christin","middleName":"","lastName":"Hauschild","suffix":""}],"badges":[],"createdAt":"2025-09-26 12:08:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7721630/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7721630/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41746-026-02367-1","type":"published","date":"2026-02-06T15:58:52+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":94100992,"identity":"fbff2175-5b9c-487f-850d-494d3586c047","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9431,"visible":true,"origin":"","legend":"","description":"","filename":"121c308fb2734d46bc42d3b28fd82168.json","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/e3567e4804cb04f2217901ec.json"},{"id":94102206,"identity":"864abf88-3a99-4c06-8fd2-12dd8a2c9b00","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"xml","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":233236,"visible":true,"origin":"","legend":"","description":"","filename":"121c308fb2734d46bc42d3b28fd821681enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/df304efc137becf6380f9659.xml"},{"id":94103241,"identity":"ff04b01f-868e-4a72-b67f-8a8e0144c156","added_by":"auto","created_at":"2025-10-22 11:41:52","extension":"svg","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47707,"visible":true,"origin":"","legend":"","description":"","filename":"AMIMEANSHIP.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/5aaf279f03d0b1570235562c.svg"},{"id":94100991,"identity":"9c290d6b-d93a-40de-a136-3910b2250447","added_by":"auto","created_at":"2025-10-22 11:17:51","extension":"svg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50001,"visible":true,"origin":"","legend":"","description":"","filename":"ASMIMEAN.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/1c3034fc5415bd661bc503b6.svg"},{"id":94102231,"identity":"54cda3ce-d16b-454b-b355-79e297dcbf91","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18816448,"visible":true,"origin":"","legend":"","description":"","filename":"ECGworkflow.png","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/d036811b89de8fea4a35faff.png"},{"id":94101026,"identity":"a92c0b2b-418f-460b-836b-8aa9210e251c","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25537872,"visible":true,"origin":"","legend":"","description":"","filename":"GNN4MInpj.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/33afd48b165ddca1bdc3d48c.pdf"},{"id":94102534,"identity":"f731c0a0-bcca-4826-a7a8-7ea980c4a06d","added_by":"auto","created_at":"2025-10-22 11:33:52","extension":"svg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48921,"visible":true,"origin":"","legend":"","description":"","filename":"IMIMEAN.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/e4dfc5069d2b72a24512c8bb.svg"},{"id":94103242,"identity":"476da422-4dbc-4754-9ed1-17fd055d4ded","added_by":"auto","created_at":"2025-10-22 11:41:52","extension":"svg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48291,"visible":true,"origin":"","legend":"","description":"","filename":"IMIMEANSHIP.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/6603b67f46578a1a3a369903.svg"},{"id":94101000,"identity":"f1724197-2f4f-459c-99f9-c6c88c374d48","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"svg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60452,"visible":true,"origin":"","legend":"","description":"","filename":"NORMMEAN.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/6b78af993159589712b81b13.svg"},{"id":94100995,"identity":"a00f7f95-6033-494b-99fa-540597489ee9","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"svg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48950,"visible":true,"origin":"","legend":"","description":"","filename":"NORMMEANSHIP.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/93fbea31af82f6e450e83f47.svg"},{"id":94102536,"identity":"6c225fdb-5f1b-43f7-94e9-759490c0c4d1","added_by":"auto","created_at":"2025-10-22 11:33:52","extension":"svg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49330,"visible":true,"origin":"","legend":"","description":"","filename":"XAIplotCDMEAN.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/47639a9acf0a71aa59362f9b.svg"},{"id":94102219,"identity":"3d3c7a7b-0a4b-4581-b4f3-89e33e63d5a6","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"svg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49592,"visible":true,"origin":"","legend":"","description":"","filename":"XAIplotHYPMEAN.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/ca5b72bba28dbb8306047169.svg"},{"id":94100998,"identity":"7dd12bd1-af85-46c0-a156-60eacf012463","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"svg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48892,"visible":true,"origin":"","legend":"","description":"","filename":"XAIplotMIMEAN.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/5eca035123e7fb4fb63ebf90.svg"},{"id":94101002,"identity":"b30afbbb-95b3-4f77-963e-660383d9d7fc","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"svg","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49905,"visible":true,"origin":"","legend":"","description":"","filename":"XAIplotNORMMEAN.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/1c1af961b517e52e807bd338.svg"},{"id":94102538,"identity":"1c930f35-afe0-4336-aced-56d52fc8a234","added_by":"auto","created_at":"2025-10-22 11:33:52","extension":"svg","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49813,"visible":true,"origin":"","legend":"","description":"","filename":"XAIplotSTTCMEAN.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/c2e6950566a8237daa4455d7.svg"},{"id":94102225,"identity":"83f4cf61-88ac-4a9f-bf3c-f6aa08761fd9","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"svg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38151,"visible":true,"origin":"","legend":"","description":"","filename":"confusionmatrix.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/58ee32b28fcc7ca0d7a3d335.svg"},{"id":94101003,"identity":"6bb84e09-95a9-4a93-8782-bf99ebfdd461","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"svg","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30433,"visible":true,"origin":"","legend":"","description":"","filename":"confusionmatrixSHIPres4.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/6bca09032ee6b89ef10dcdf7.svg"},{"id":94102221,"identity":"7988d4f5-b79c-4d03-a5ee-f4e70dcf8ccb","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"svg","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31003,"visible":true,"origin":"","legend":"","description":"","filename":"confusionmatrixmiless.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/a0f7771c88b865b2487464df.svg"},{"id":94103243,"identity":"a2986111-71a7-4b60-8d6a-1916a0abb814","added_by":"auto","created_at":"2025-10-22 11:41:52","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64948,"visible":true,"origin":"","legend":"","description":"","filename":"example.png","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/875d7d3b94d9b0d404eb11b1.png"},{"id":94102229,"identity":"80eaee3f-f47c-43fe-8175-22944dd34748","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"svg","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79860,"visible":true,"origin":"","legend":"","description":"","filename":"example.svg","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/6e7d0c59ca1a1628379aed24.svg"},{"id":94103418,"identity":"21c35a4f-d72e-444d-afa1-07bec750bc7a","added_by":"auto","created_at":"2025-10-22 11:49:52","extension":"pdf","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":263844,"visible":true,"origin":"","legend":"","description":"","filename":"npjCoverLettersig.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/110ebb14b5f51167c018bda4.pdf"},{"id":94101017,"identity":"ff65cf6e-9516-48f0-adc9-5874a3b17d1c","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"bst","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":146013,"visible":true,"origin":"","legend":"","description":"","filename":"snapacite.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/a417dd1268e3e1ab3800c2a1.bst"},{"id":94101005,"identity":"6d260002-5258-4e40-8495-276f1891d92c","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"bst","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29828,"visible":true,"origin":"","legend":"","description":"","filename":"snaps.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/686ab2f958da1353b0cd634e.bst"},{"id":94102222,"identity":"40311de6-152b-4eb6-9205-97952b0f8013","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"bst","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35515,"visible":true,"origin":"","legend":"","description":"","filename":"snbasic.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/09418a307bff932b4fbfab9f.bst"},{"id":94102227,"identity":"04ea4e82-e98d-4d13-ba80-920ebafba8a9","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"bst","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33968,"visible":true,"origin":"","legend":"","description":"","filename":"snchicago.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/267036c1191c1daaf5b0c44f.bst"},{"id":94101010,"identity":"38762bfb-7463-4ea5-89b1-600788c1d285","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"cls","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":55857,"visible":true,"origin":"","legend":"","description":"","filename":"snjnl.cls","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/3164406abf259fb50436337b.cls"},{"id":94101011,"identity":"e58f48b8-9790-40b0-9d17-b2f44606ee4f","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"bst","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64023,"visible":true,"origin":"","legend":"","description":"","filename":"snmathphysay.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/84f771bb5c1bd0df98967301.bst"},{"id":94102542,"identity":"cc462d35-a7bd-4302-9891-9d760711957f","added_by":"auto","created_at":"2025-10-22 11:33:52","extension":"bst","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64166,"visible":true,"origin":"","legend":"","description":"","filename":"snmathphysnum.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/e5eb5fc1247aad27fce7dff2.bst"},{"id":94101014,"identity":"a14e58dd-c138-4c46-834d-48df8d5aee04","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"bst","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37333,"visible":true,"origin":"","legend":"","description":"","filename":"snnature.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/a980f25cf9863687334bb979.bst"},{"id":94101019,"identity":"7b60d3d1-8b74-4492-a72c-9900fa0e02e3","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"bst","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39951,"visible":true,"origin":"","legend":"","description":"","filename":"snvancouveray.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/0c946a527e3609c59f291832.bst"},{"id":94101022,"identity":"a7e90005-ecfc-409c-a983-87015b47a27c","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"bst","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40758,"visible":true,"origin":"","legend":"","description":"","filename":"snvancouvernum.bst","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/01425eb17fa615a5e4aee4d1.bst"},{"id":94102541,"identity":"ba1d40bd-63d3-4c83-ac26-a8e10ce18be3","added_by":"auto","created_at":"2025-10-22 11:33:52","extension":"pdf","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17552,"visible":true,"origin":"","legend":"","description":"","filename":"technicalcheckamendment.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/3234aca2d3cbdef4dafc7eea.pdf"},{"id":94102232,"identity":"2280c5a0-5c8a-4ea3-9e33-54741c25c106","added_by":"auto","created_at":"2025-10-22 11:25:52","extension":"pdf","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":418495,"visible":true,"origin":"","legend":"","description":"","filename":"usermanual.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/78a3968ee2f8f0453e3af014.pdf"},{"id":94101020,"identity":"f4488585-dc3d-4c9c-8b50-4e2d43f7a7d5","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"png","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":57133,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineexample.png","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/b1e2941e4735e0aef4c4eb5b.png"},{"id":94101023,"identity":"f1f21c25-4d43-4c48-8abd-758031d31463","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"xml","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":243730,"visible":true,"origin":"","legend":"","description":"","filename":"121c308fb2734d46bc42d3b28fd821681structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/4d2ed9524d722c5d713b206c.xml"},{"id":94101024,"identity":"6431635e-df7b-4c02-955e-a4222f90ba96","added_by":"auto","created_at":"2025-10-22 11:17:52","extension":"html","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":263929,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1/0ccc54a6c499fa4544b6aa5f.html"},{"id":102234101,"identity":"c72d0b97-3b2c-47c2-a346-566560b0954a","added_by":"auto","created_at":"2026-02-09 16:06:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8536142,"visible":true,"origin":"","legend":"","description":"","filename":"GNN4MInpj.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7721630/v1_covered_f421bb86-335f-4390-921c-f0f5a82f3b5f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"xGNN4MI: Explainability of Graph Neural Networks in 12-lead Electrocardiography for Cardiovascular Disease Classification","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"AI-ECG, Explainable AI, Graph Neural Networks, Deep Learning, Myocardial Infarction","lastPublishedDoi":"10.21203/rs.3.rs-7721630/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7721630/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe clinical deployment of artificial intelligence (AI) solutions for assessing cardiovascular disease (CVD) risk in 12-lead electrocardiography (ECG) is hindered by limitations in interpretability and explainability. To address this, we present xGNN4MI, an open-source framework for graph neural networks (GNNs) in ECG modeling for interpretable CVD prediction. Our framework facilitates modeling clinically relevant spatial relationships between ECG leads and their temporal dynamics. We also developed specific explainable AI (XAI) and visualization tools to identify ECG leads crucial to the model's decision-making process, enabling a systematic comparison with established clinical knowledge. We evaluated xGNN4MI on two challenging tasks: diagnostic superclass classification and localization of myocardial infarction. Our findings show that the interpretable ECG-GNN models demonstrate good performance across the tasks. XAI analysis revealed clinically meaningful training effects, such as differentiating between anteroseptal and inferior myocardial infarction. Our work demonstrates the potential of ECG-GNNs for providing trustworthy and interpretable AI-based CVD diagnosis.\u003c/p\u003e","manuscriptTitle":"xGNN4MI: Explainability of Graph Neural Networks in 12-lead Electrocardiography for Cardiovascular Disease Classification","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-22 11:17:47","doi":"10.21203/rs.3.rs-7721630/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-01T09:23:26+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-30T14:02:10+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-20T04:59:03+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-15T08:57:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"143864942457461043344062817055328525255","date":"2025-10-10T08:13:42+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"48207533929008021183415333530511717295","date":"2025-10-09T23:48:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"111996357766612564826163636503746203142","date":"2025-10-09T10:32:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"205600674636354795885759282467176343376","date":"2025-10-09T10:18:40+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"124304551563163752514426807140696168462","date":"2025-10-09T10:17:52+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-09T10:04:51+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-03T19:38:45+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-03T16:07:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Digital Medicine","date":"2025-09-26T11:54:47+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f60c82b5-7bac-45db-9c79-b76731954a5f","owner":[],"postedDate":"October 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":56559666,"name":"Health sciences/Cardiology"},{"id":56559667,"name":"Biological sciences/Computational biology and bioinformatics"}],"tags":[],"updatedAt":"2026-02-09T16:02:21+00:00","versionOfRecord":{"articleIdentity":"rs-7721630","link":"https://doi.org/10.1038/s41746-026-02367-1","journal":{"identity":"npj-digital-medicine","isVorOnly":false,"title":"npj Digital Medicine"},"publishedOn":"2026-02-06 15:58:52","publishedOnDateReadable":"February 6th, 2026"},"versionCreatedAt":"2025-10-22 11:17:47","video":"","vorDoi":"10.1038/s41746-026-02367-1","vorDoiUrl":"https://doi.org/10.1038/s41746-026-02367-1","workflowStages":[]},"version":"v1","identity":"rs-7721630","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7721630","identity":"rs-7721630","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}