EchoApex: A General-Purpose Vision Foundation Model for Echocardiography

EchoApex: A General-Purpose Vision Foundation Model for Echocardiography | 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 EchoApex: A General-Purpose Vision Foundation Model for Echocardiography Yue Zhang, Abdoul Amadou, Sebastien Piat, Paul Klein, Ingo Schmuecking, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5332987/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Quantitative evaluation of echocardiography is essential for precise assessment of cardiac condition and guiding treatment decisions. The diverse nature of echo images, including variations in probe types, manufacturers, and pathologies, poses challenges for developing artificial intelligent models that can generalize across different clinical practice. Here, we introduce EchoApex, a general-purpose vision foundation model designed for comprehensive echocardiography analysis. Pretrained on over 20 million images from 11 clinical centers, EchoApex utilizes self-supervised learning and task-specific decoders for diverse applications in echocardiography. It performs sequence view classification, interactive structure segmentation, left ventricle measurement, and ejection fraction estimation. In benchmark evaluations, EchoApex outperforms task-specific models, achieving a mean BACC of 0.976 in classification of 18 common views, DICE of 0.93 in chamber segmentation, MAE of 3.9mm in left ventricle linear measurement and a zero-shot performance improvement over specialist models in all evaluated datasets. For ejection fraction estimation, EchoApex achieves an MAE of 5.6% and an AUC of 0.927 for cardiomyopathy detection. Despite using less than 4% of trainable parameters with frozen encoders, EchoApex with adapter demonstrates strong performance with minimal degradation compared to fully finetuned models. This work establishes EchoApex as a scalable, general-purpose model for echocardiography, enabling efficient adaptation across various clinical tasks. Health sciences/Diseases/Cardiovascular diseases Biological sciences/Computational biology and bioinformatics/Computational platforms and environments Full Text Additional Declarations Yes there is potential Competing Interest. A.A.A., Y.Z., S.P., P.K., I.S., T.P. and P.S. are employees and equity holders of Siemens Healthineers. Cite Share Download PDF Status: Under Review 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-5332987","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":371321749,"identity":"76bfe7dd-fbfa-4289-bfd2-9e75173d6877","order_by":0,"name":"Yue Zhang","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-4017-3681","institution":"Siemens Healthineers","correspondingAuthor":true,"prefix":"","firstName":"Yue","middleName":"","lastName":"Zhang","suffix":""},{"id":371321750,"identity":"19152e96-e16b-4026-8850-6b6e5ed7a3ac","order_by":1,"name":"Abdoul Amadou","email":"","orcid":"","institution":"Siemens Healthineers","correspondingAuthor":false,"prefix":"","firstName":"Abdoul","middleName":"","lastName":"Amadou","suffix":""},{"id":371321751,"identity":"c0421c5e-393c-4af7-b4cf-52775be32b6b","order_by":2,"name":"Sebastien Piat","email":"","orcid":"","institution":"Siemens Healthineers","correspondingAuthor":false,"prefix":"","firstName":"Sebastien","middleName":"","lastName":"Piat","suffix":""},{"id":371321752,"identity":"43e92a88-3d89-44a5-b469-51156f67c5fa","order_by":3,"name":"Paul Klein","email":"","orcid":"","institution":"Siemens Healthineers","correspondingAuthor":false,"prefix":"","firstName":"Paul","middleName":"","lastName":"Klein","suffix":""},{"id":371321753,"identity":"d9618a53-483e-46f3-b697-67afbf54c82f","order_by":4,"name":"Ingo Schmuecking","email":"","orcid":"","institution":"Siemens Healthineers","correspondingAuthor":false,"prefix":"","firstName":"Ingo","middleName":"","lastName":"Schmuecking","suffix":""},{"id":371321754,"identity":"e6226248-8fbd-4bb1-ac85-7af31a98e0bb","order_by":5,"name":"Tiziano Passerini","email":"","orcid":"","institution":"Siemens Healthineers","correspondingAuthor":false,"prefix":"","firstName":"Tiziano","middleName":"","lastName":"Passerini","suffix":""},{"id":371321755,"identity":"75c27964-1206-4313-a59f-fefb6ba1c02c","order_by":6,"name":"Puneet Sharma","email":"","orcid":"","institution":"Siemens Healthineers","correspondingAuthor":false,"prefix":"","firstName":"Puneet","middleName":"","lastName":"Sharma","suffix":""}],"badges":[],"createdAt":"2024-10-25 14:15:51","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5332987/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5332987/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":70994358,"identity":"7b439c05-e313-48f4-b9d4-d1dc9ce90dcb","added_by":"auto","created_at":"2024-12-10 04:10:01","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4227779,"visible":true,"origin":"","legend":"Article File","description":"","filename":"EchoApexManuscriptNMEDA136741A.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5332987/v1_covered_5b761e0d-c8cd-49fa-a716-8243584db5bd.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nA.A.A., Y.Z., S.P., P.K., I.S., T.P. and P.S. are employees and equity holders of Siemens Healthineers.","formattedTitle":"EchoApex: A General-Purpose Vision Foundation Model for Echocardiography","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5332987/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5332987/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Quantitative evaluation of echocardiography is essential for precise assessment of cardiac condition and guiding treatment decisions. The diverse nature of echo images, including variations in probe types, manufacturers, and pathologies, poses challenges for developing artificial intelligent models that can generalize across different clinical practice. Here, we introduce EchoApex, a general-purpose vision foundation model designed for comprehensive echocardiography analysis. Pretrained on over 20 million images from 11 clinical centers, EchoApex utilizes self-supervised learning and task-specific decoders for diverse applications in echocardiography. It performs sequence view classification, interactive structure segmentation, left ventricle measurement, and ejection fraction estimation. In benchmark evaluations, EchoApex outperforms task-specific models, achieving a mean BACC of 0.976 in classification of 18 common views, DICE of 0.93 in chamber segmentation, MAE of 3.9mm in left ventricle linear measurement and a zero-shot performance improvement over specialist models in all evaluated datasets. For ejection fraction estimation, EchoApex achieves an MAE of 5.6% and an AUC of 0.927 for cardiomyopathy detection. Despite using less than 4% of trainable parameters with frozen encoders, EchoApex with adapter demonstrates strong performance with minimal degradation compared to fully finetuned models. This work establishes EchoApex as a scalable, general-purpose model for echocardiography, enabling efficient adaptation across various clinical tasks.","manuscriptTitle":"EchoApex: A General-Purpose Vision Foundation Model for Echocardiography","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-10 04:01:46","doi":"10.21203/rs.3.rs-5332987/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"169e1297-c898-46cf-93a8-2d649ec6d651","owner":[],"postedDate":"December 10th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":39513424,"name":"Health sciences/Diseases/Cardiovascular diseases"},{"id":39513425,"name":"Biological sciences/Computational biology and bioinformatics/Computational platforms and environments"}],"tags":[],"updatedAt":"2024-12-10T04:01:46+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-10 04:01:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5332987","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5332987","identity":"rs-5332987","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}