Spatially-Conditioned Variational Autoencoder with Latent Manifold Harvesting for High-Fidelity 12-Lead ECG Synthesis

Spatially-Conditioned Variational Autoencoder with Latent Manifold Harvesting for High-Fidelity 12-Lead ECG Synthesis | 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 Research Article Spatially-Conditioned Variational Autoencoder with Latent Manifold Harvesting for High-Fidelity 12-Lead ECG Synthesis Ashutosh Mishra, Nicolas Stein, Louis Peter, Michael Guckert This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9222289/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 Generative modeling of physiological signals, particularly 12-lead electrocardiograms (ECG), faces significant challenges in preserving local morphological fidelity while adhering to strict global temporal constraints, such as R-peak positioning. Traditional conditional Variational Autoencoders (cVAEs) often suffer from posterior collapse or temporal misalignment when conditioning information is restricted to the network bottleneck. In this paper, we propose a Spatially-Conditioned VAE (SC-VAE) that injects temporal constraints (with Gaussian smoothing to reduce conditioning shock) at multiple resolutions within the decoder architecture. Furthermore, we introduce a Latent Manifold Harvesting strategy that iteratively curates a “gold-standard” latent bank during training, enabling robust post-hoc sampling via jittered replay. Our methodology significantly outperforms Diffusion Probabilistic Models in inference speed while resolving the topological drift commonly observed in standard VAEs. Artificial Intelligence and Machine Learning Biomedical Engineering ECG synthesis Medical data augmentation Latent manifold harvesting Generative models physiological signals Full Text Additional Declarations The authors declare no competing interests. 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-9222289","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":611971274,"identity":"23e7b28f-0187-46b6-a3ec-859dfbc0a34d","order_by":0,"name":"Ashutosh Mishra","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAApklEQVRIiWNgGAWjYFAC5gMHEgyg7ATitLAlALUYkKSFB6TcgKAyBOCfdubjgQcFf+QZ2A8/YHi4gwgtErdzN4AcZtjAk2bAkHiGGGugWhgbGHIYGBLbiNAhfzvnAUiLfQP/GyK1GNzOYQBpSWyQINYWw9tpBkAtxsltEs8MDhClRe528uOPP/7I2fbzJz98+JMYLXDABsQHSNEwCkbBKBgFowAPAAAlODZmkBmULgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0003-3524-8834","institution":"Indian Institue of Information Technology Allahabad, India","correspondingAuthor":true,"prefix":"","firstName":"Ashutosh","middleName":"","lastName":"Mishra","suffix":""},{"id":611971275,"identity":"454635e8-4a18-4467-ac73-1aea660edfba","order_by":1,"name":"Nicolas Stein","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIiWNgGAWjYDACZiB+AGUfYKgAk2xQcTxaEkAMNpDiMyAtzAS0MCBpYWBsI0KLOTvzwwcJNXUM8vObDx4unLctj+8A/7FHNxis5XBpsWxmMzZIOHaYweAYW8LhmdtuF0seYGY3zmFIN8alxeAwD5tEAtsBBgM2HoPDvNtuJ24AOkw6h+FwYgNeLf+ADmvj/3CYdw5CSz1eLYltQM8e42E4zNuA0JKA22FAvyT2HeYxOJYG1H7sduLMw8xm0jkG6YY4bTl/+OGDD9/q5OSbDz/+zFNzO7HveOMz6ZwKa3lctsAAD4IJjhEDQhpGwSgYBaNgFOADADB1VCzk3CIHAAAAAElFTkSuQmCC","orcid":"","institution":"Technische Hochschule Mittelhessen University of Applied Sciences,","correspondingAuthor":true,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"Stein","suffix":""},{"id":611971276,"identity":"5e605414-0cb7-4177-91cb-49cadc0dd067","order_by":2,"name":"Louis Peter","email":"","orcid":"","institution":"Technische Hochschule Mittelhessen University of Applied Sciences,","correspondingAuthor":false,"prefix":"","firstName":"Louis","middleName":"","lastName":"Peter","suffix":""},{"id":611971277,"identity":"4097d66d-d2c0-4026-8b6a-f1e667795033","order_by":3,"name":"Michael Guckert","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABE0lEQVRIiWNgGAWjYDACZgY2IGkBYQKBHBAbMDA2AKkDeLVIwLQYGIO1HMSnhQFNS2IDIS267czPHvyokGCQj0h+bPBxz5/0tbObNzB/3LGNge94A1YtZofZzA17zkgwGN5IM06c8cwgd9udYwUMB8/cZpA8g90as8M8bBK8bUAtM3KYD/McAGq5kQN0WNttBoMbCTi1SP79B9Xy54BBuhlcy/0HOLVI8zYA/SKRw5zMcMAgAaHlBnbvA/1iJi1zTILHgOeZsWHPAWPDbTfSCg6cbbvNI3kGh8POH34m+abGRk6+PfmxxI8DcvJmN5I3Pqhsuy3Hdxy792GAxwBZHsTmwaseBOQbCCoZBaNgFIyCkQoAZHtjsodbe70AAAAASUVORK5CYII=","orcid":"","institution":"Technische Hochschule Mittelhessen University of Applied Sciences,","correspondingAuthor":true,"prefix":"","firstName":"Michael","middleName":"","lastName":"Guckert","suffix":""}],"badges":[],"createdAt":"2026-03-25 11:23:08","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9222289/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9222289/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105464322,"identity":"9ea84d13-78d9-493e-b507-5c2055c5b23d","added_by":"auto","created_at":"2026-03-26 10:28:14","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2556329,"visible":true,"origin":"","legend":"","description":"","filename":"SCVAE7.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9222289/v1_covered_e1bebb28-01a2-4308-a5c8-04ccf5c9d272.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eSpatially-Conditioned Variational Autoencoder with Latent Manifold Harvesting for High-Fidelity 12-Lead ECG Synthesis\u003c/strong\u003e\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Indian Institute of Information Technology","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"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":"ECG synthesis, Medical data augmentation, Latent manifold harvesting, Generative models, physiological signals","lastPublishedDoi":"10.21203/rs.3.rs-9222289/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9222289/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGenerative modeling of physiological signals, particularly 12-lead electrocardiograms (ECG), faces significant challenges in preserving local morphological fidelity while adhering to strict global temporal constraints, such as R-peak positioning. Traditional conditional Variational Autoencoders (cVAEs) often suffer from posterior collapse or temporal misalignment when conditioning information is restricted to the network bottleneck. In this paper, we propose a Spatially-Conditioned VAE (SC-VAE) that injects temporal constraints (with Gaussian smoothing to reduce conditioning shock) at multiple resolutions within the decoder architecture. Furthermore, we introduce a Latent Manifold Harvesting strategy that iteratively curates a “gold-standard” latent bank during training, enabling robust post-hoc sampling via jittered replay. Our methodology significantly outperforms Diffusion Probabilistic Models in inference speed while resolving the topological drift commonly observed in standard VAEs.\u003c/p\u003e","manuscriptTitle":"Spatially-Conditioned Variational Autoencoder with Latent Manifold Harvesting for High-Fidelity 12-Lead ECG Synthesis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-26 10:27:10","doi":"10.21203/rs.3.rs-9222289/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":"66db2479-6529-4b1e-ab4d-d3b249f23610","owner":[],"postedDate":"March 26th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":65113600,"name":"Artificial Intelligence and Machine Learning"},{"id":65113601,"name":"Biomedical Engineering"}],"tags":[],"updatedAt":"2026-03-26T10:27:10+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-26 10:27:10","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9222289","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9222289","identity":"rs-9222289","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}