Explainable Deep Learning for Cardiac MRI: Multi-Stage Segmentation, Cascade Classification, and Visual Interpretation

Explainable Deep Learning for Cardiac MRI: Multi-Stage Segmentation, Cascade Classification, and Visual Interpretation | 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 Explainable Deep Learning for Cardiac MRI: Multi-Stage Segmentation, Cascade Classification, and Visual Interpretation Vitalii Slobodzian, Barmak Oleksandr, Pavlo Radiuk, Liliana Klymenko, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5930463/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 Cardiac MRI images are vital in diagnosing a range of heart diseases, yet standard solutions frequently struggle with inadequate region delineation, confusion among similar pathologies, and opaque decision-making processes. In this work, we aim to resolve these problems by introducing dedicated methods for careful region extraction, specialized classification, and metric-based interpretation. Our approach notably improves segmentation, achieving Dice coefficients of 0.974 for the left ventricle and 0.947 for the right ventricle—outperforming prior baselines. Classification results reach a 97% overall accuracy, substantially higher than reference architectures that only attained 72–84%. Furthermore, clinical relevance is enhanced through a structured output that pinpoints key anatomical and functional indicators. These findings suggest a reliable pipeline that refines MRI analysis and facilitates healthcare professionals making more informed decisions in real-world medical settings. Full Text Additional Declarations No competing interests reported. Supplementary Files SlobodzianManuscriptSupplementaryv4.pdf 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-5930463","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":627543562,"identity":"14b9a0cd-6420-4d3a-9b5f-15c712e55e14","order_by":0,"name":"Vitalii Slobodzian","email":"","orcid":"","institution":"Khmelnytskyi National University","correspondingAuthor":false,"prefix":"","firstName":"Vitalii","middleName":"","lastName":"Slobodzian","suffix":""},{"id":627543563,"identity":"572ae1ef-1ff8-489a-b7aa-5c1b4f66f515","order_by":1,"name":"Barmak Oleksandr","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIiWNgGAWjYBACAzB5gEFOQgIiwNhArBZj0rUkziBaizn72acbfpyxSZ85uzvtMQ+DjeyGAzyGH/BpsexJN7vZcyMtd7bM2e3GPAxpxkAtxhJ4HXYgje0Gz4fDufMkcrdJ8zAcTgRqMcCv5fwztpt/PhxOl4No+Q/SYvwDr5YbaWy3eW4cTpCGaDkA0mKG1xbLGc/YbsucSTOcOSN3m+Qcg2TjmYfZyizwaTHnT2O7+eaYjbzEjdxtEm8q7GT7jjdvvoFPC7o7gZiZBPWjYBSMglEwCrADAA0ZTOh2x90VAAAAAElFTkSuQmCC","orcid":"","institution":"Khmelnytskyi National University","correspondingAuthor":true,"prefix":"","firstName":"Barmak","middleName":"","lastName":"Oleksandr","suffix":""},{"id":627543565,"identity":"df744db9-15f3-44b3-b006-02ee3cc1836d","order_by":2,"name":"Pavlo Radiuk","email":"","orcid":"","institution":"Khmelnytskyi National University","correspondingAuthor":false,"prefix":"","firstName":"Pavlo","middleName":"","lastName":"Radiuk","suffix":""},{"id":627543568,"identity":"f323e648-5fb3-4a6d-9da3-b682f7331204","order_by":3,"name":"Liliana Klymenko","email":"","orcid":"","institution":"Shupyk National Healthcare University of Ukraine","correspondingAuthor":false,"prefix":"","firstName":"Liliana","middleName":"","lastName":"Klymenko","suffix":""},{"id":627543570,"identity":"70a52d35-fc62-4bc4-91a2-e26c833d1fae","order_by":4,"name":"Iurii Krak","email":"","orcid":"","institution":"Taras Shevchenko National University of Kyiv","correspondingAuthor":false,"prefix":"","firstName":"Iurii","middleName":"","lastName":"Krak","suffix":""}],"badges":[],"createdAt":"2025-01-30 13:08:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5930463/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5930463/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107705728,"identity":"cc7889c1-c445-41bb-bfb3-ed1796b63f9b","added_by":"auto","created_at":"2026-04-24 09:14:50","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2699410,"visible":true,"origin":"","legend":"","description":"","filename":"SlobodzianManuscriptv4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5930463/v1_covered_5c88b0b0-49df-45fb-a36e-b23bbed59539.pdf"},{"id":107598425,"identity":"b871e396-d712-49c2-9173-7cf1aee16ecb","added_by":"auto","created_at":"2026-04-23 05:57:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":8820933,"visible":true,"origin":"","legend":"","description":"","filename":"SlobodzianManuscriptSupplementaryv4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5930463/v1/1ceca66f2592d0d1506795e7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Explainable Deep Learning for Cardiac MRI: Multi-Stage Segmentation, Cascade Classification, and Visual Interpretation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"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-5930463/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5930463/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Cardiac MRI images are vital in diagnosing a range of heart diseases, yet standard solutions frequently struggle with inadequate region delineation, confusion among similar pathologies, and opaque decision-making processes. In this work, we aim to resolve these problems by introducing dedicated methods for careful region extraction, specialized classification, and metric-based interpretation. Our approach notably improves segmentation, achieving Dice coefficients of 0.974 for the left ventricle and 0.947 for the right ventricle—outperforming prior baselines. Classification results reach a 97% overall accuracy, substantially higher than reference architectures that only attained 72–84%. Furthermore, clinical relevance is enhanced through a structured output that pinpoints key anatomical and functional indicators. These findings suggest a reliable pipeline that refines MRI analysis and facilitates healthcare professionals making more informed decisions in real-world medical settings.","manuscriptTitle":"Explainable Deep Learning for Cardiac MRI: Multi-Stage Segmentation, Cascade Classification, and Visual Interpretation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-23 05:57:02","doi":"10.21203/rs.3.rs-5930463/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":"f57de125-f750-4a2a-85ca-9bf197e3278f","owner":[],"postedDate":"April 23rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-23T05:57:02+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-23 05:57:02","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5930463","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5930463","identity":"rs-5930463","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}