Learning Lead-Invariant ECG Representations via Cross-View Contrastive Learning

Learning Lead-Invariant ECG Representations via Cross-View Contrastive Learning | 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 Learning Lead-Invariant ECG Representations via Cross-View Contrastive Learning Faraaz Akhtar, Rocio Mexia Diaz This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8639727/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Electrocardiograms (ECGs) are inherently multi-view signals, with cardiac electrical activity observed through multiple leads that differ in anatomical orientation and sensitivity to underlying cardiac sources. Although contrastive learning has been previously applied to ECG signals, its effects on the geometry of learned representations remain poorly understood. Here, we study how crossview contrastive learning between single-lead and full 12-lead ECGs reshapes representation structure. Using the PTB-XL dataset, we train a self-supervised model to align the embeddings of randomly sampled single-lead views and the corresponding 12-lead signals. We find that contrastive learning induces near complete linear alignment across leads, a structure absent in raw waveform space, and that embeddings of full 12-lead ECGs are also largely linearly recoverable from single-lead embeddings. We further show that this geometric alignment is accompanied by substantially improved diagnostic decodability across leads using linear probes, while preserving physiologically structured lead differences. Together, these results provide a mechanistic explanation for the robustness of contrastively learned ECG representations. Health sciences/Cardiology Biological sciences/Computational biology and bioinformatics Physical sciences/Engineering Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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-8639727","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":576853281,"identity":"eda231fa-5b8f-4218-92fa-ce04837f77fa","order_by":0,"name":"Faraaz Akhtar","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYBAC9gYwZUOCFp4DYCpNgoGBmTQth0nRwn724Ycff87Xmc/IP/i4co8dA397AwEtPOnGkj08tyVkbiQzG555lswgceYAfi32DGkMEjwStyUkJJLZJBsOMDMYSCQQsIX/GfPPPwbnQFrYfzYcqCdCi0QamzRPwgGwLYwNBw4To+UZm7XMgWTJGTyPjYEOO85D0C88/GnMN9/8seOXYE98+LHhQLUcwRBDAAGIe3iIVQ8E/ATcMwpGwSgYBSMXAAAetzqrQ98g0QAAAABJRU5ErkJggg==","orcid":"","institution":"Terra API","correspondingAuthor":true,"prefix":"","firstName":"Faraaz","middleName":"","lastName":"Akhtar","suffix":""},{"id":576853282,"identity":"05ec7771-0e66-4922-8aac-ec07fc0f3871","order_by":1,"name":"Rocio Mexia Diaz","email":"","orcid":"","institution":"Terra API","correspondingAuthor":false,"prefix":"","firstName":"Rocio","middleName":"Mexia","lastName":"Diaz","suffix":""}],"badges":[],"createdAt":"2026-01-19 12:57:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8639727/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8639727/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100663518,"identity":"c82e0369-435d-4e30-b6a0-3129a60c50f1","added_by":"auto","created_at":"2026-01-20 09:07:42","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4062,"visible":true,"origin":"","legend":"","description":"","filename":"546590f3ac754ee19d2c678107379ee1.json","url":"https://assets-eu.researchsquare.com/files/rs-8639727/v1/0d8e2c89f5dbce44975fa23b.json"},{"id":100666515,"identity":"f985cfc8-8e63-4a71-98e5-010b3dc13b21","added_by":"auto","created_at":"2026-01-20 09:39:11","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1483595,"visible":true,"origin":"","legend":"","description":"","filename":"LearningLeadInvariantECGRepresentationsviaCrossViewContrastiveLearning.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8639727/v1_covered_de805cfb-fafc-425b-a959-a7e4cd808663.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Learning Lead-Invariant ECG Representations via Cross-View Contrastive Learning","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8639727/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8639727/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Electrocardiograms (ECGs) are inherently multi-view signals, with cardiac electrical activity observed through multiple leads that differ in anatomical orientation and sensitivity to underlying cardiac sources. Although contrastive learning has been previously applied to ECG signals, its effects on the geometry of learned representations remain poorly understood. Here, we study how crossview contrastive learning between single-lead and full 12-lead ECGs reshapes representation structure. Using the PTB-XL dataset, we train a self-supervised model to align the embeddings of randomly sampled single-lead views and the corresponding 12-lead signals. We find that contrastive learning induces near complete linear alignment across leads, a structure absent in raw waveform space,\nand that embeddings of full 12-lead ECGs are also largely linearly recoverable from single-lead embeddings. We further show that this geometric alignment is accompanied by substantially improved diagnostic decodability across leads using linear probes, while preserving physiologically structured lead differences. Together, these results provide a mechanistic explanation for the robustness of contrastively learned ECG representations.","manuscriptTitle":"Learning Lead-Invariant ECG Representations via Cross-View Contrastive Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-20 08:18:57","doi":"10.21203/rs.3.rs-8639727/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"fd7b7df0-5249-40ad-a939-da80a3b399ed","owner":[],"postedDate":"January 20th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":61371874,"name":"Health sciences/Cardiology"},{"id":61371875,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":61371876,"name":"Physical sciences/Engineering"}],"tags":[],"updatedAt":"2026-03-02T19:39:47+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-20 08:18:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8639727","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8639727","identity":"rs-8639727","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}