Do Hybrid CNN–Transformer Architectures Really Generalize? A Systematic Review for Medical Imaging

Do Hybrid CNN–Transformer Architectures Really Generalize? A Systematic Review for Medical Imaging | 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 Systematic Review Do Hybrid CNN–Transformer Architectures Really Generalize? A Systematic Review for Medical Imaging Roaa Ehab, Shimaa El-Bana, Ahmad Al-Kabbany This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9216007/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 This article presents a systematic review of hybrid CNN–Transformer architectures, examining whether and how their structural design supports generalization across diverse medical imaging scenarios. While convolutional neural networks offer strong spatial inductive biases and data efficiency, and Vision Transformers provide superior global context modeling, neither paradigm alone adequately addresses the full complexity of clinical imaging tasks. Hybrid architectures, by integrating both components, offer a compelling middle ground that is particularly valuable in medical imaging, where models must remain reliable across heterogeneous acquisition protocols, scanner variability, and diverse patient populations. At the same time, medical imaging poses unique generalization challenges — including cross-organ transfer, multi-modal fusion, and cross-dataset robustness — that expose the limitations of architectures optimized narrowly for benchmark performance. Following PRISMA guidelines, we systematically queried major academic databases, screened the resulting literature, and synthesized a representative body of peer-reviewed studies spanning a range of imaging modalities, anatomical targets, and learning paradigms. Our analysis covers the architectural taxonomy of hybrid designs, their learning and optimization strategies, and the evaluation practices adopted across the reviewed literature. The findings reveal that while hybrid models consistently demonstrate competitive performance, critical limitations persist: high computational overhead, insufficient external validation, and a heavy reliance on fully supervised learning constrain their real-world applicability. We conclude with a set of forward-looking recommendations emphasizing efficiency-aware design, standardized cross-domain evaluation, and the broader adoption of self-supervised learning strategies to advance the clinical translation of hybrid architectures. Medical Informatics Biomedical Engineering hybrid CNN-Transformer medical image analysis transfer learning domain generalization cross-organ segmentation multi-modal fusion vision transformer out-of-distribution generalization systematic review deep learning 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-9216007","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":611610273,"identity":"8cd5b43d-63e1-4cd2-9023-00112ab3d625","order_by":0,"name":"Roaa Ehab","email":"","orcid":"","institution":"Cairo University","correspondingAuthor":false,"prefix":"","firstName":"Roaa","middleName":"","lastName":"Ehab","suffix":""},{"id":611610274,"identity":"fdd8b6a4-4357-426e-b2d7-fd56e07a5de9","order_by":1,"name":"Shimaa El-Bana","email":"","orcid":"https://orcid.org/0000-0001-6533-3146","institution":"Alexandria Higher Institute of Engineering and Technology","correspondingAuthor":false,"prefix":"","firstName":"Shimaa","middleName":"","lastName":"El-Bana","suffix":""},{"id":611610275,"identity":"cca7399e-7bd1-4687-a92e-67187deab553","order_by":2,"name":"Ahmad Al-Kabbany","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3ElEQVRIiWNgGAWjYDACdjYGZsYGEKOxgUgtzDAtPAdJ1iKRQKQO/ma2xM+FO+zyDG4+bnzwoYJBzrx/AX4tEofZDkvPPJNcbHA7sdlwxhkGY5kbDwhYc5i9QZq3jTlxw+3ENiCDIXGGxAH8OuQPszf/5m2rT9xw82D777//iNBicJjtGNDww4kbbjC2gcIhcQZ/A34thofZ0qxnnjmeOPNMYrNkzzEJYwkJAl6RO95mfLtwR3Vi3/HjDz/8qLGRk+An4DA4UIAolCAhguQbYCyibRkFo2AUjIKRAgB5iki4qo9YOAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-3941-0061","institution":"Arab Academy for Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Ahmad","middleName":"","lastName":"Al-Kabbany","suffix":""}],"badges":[],"createdAt":"2026-03-24 20:57:22","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-9216007/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9216007/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105566773,"identity":"16c034f9-7f98-4836-9ced-b4e6cb195897","added_by":"auto","created_at":"2026-03-27 12:57:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1227808,"visible":true,"origin":"","legend":"","description":"","filename":"main.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9216007/v1_covered_d9a8aca6-37e4-4f19-9028-c87acceeccbf.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eDo Hybrid CNN–Transformer Architectures Really Generalize? A Systematic Review for Medical Imaging\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Arab Academy for Science, Technology and Maritime Transport","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":"hybrid CNN-Transformer, medical image analysis, transfer learning, domain generalization, cross-organ segmentation, multi-modal fusion, vision transformer, out-of-distribution generalization, systematic review, deep learning","lastPublishedDoi":"10.21203/rs.3.rs-9216007/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9216007/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis article presents a systematic review of hybrid CNN\u0026ndash;Transformer architectures, examining whether and how their structural design supports generalization across diverse medical imaging scenarios. While convolutional neural networks offer strong spatial inductive biases and data efficiency, and Vision Transformers provide superior global context modeling, neither paradigm alone adequately addresses the full complexity of clinical imaging tasks. Hybrid architectures, by integrating both components, offer a compelling middle ground that is particularly valuable in medical imaging, where models must remain reliable across heterogeneous acquisition protocols, scanner variability, and diverse patient populations. At the same time, medical imaging poses unique generalization challenges \u0026mdash; including cross-organ transfer, multi-modal fusion, and cross-dataset robustness \u0026mdash; that expose the limitations of architectures optimized narrowly for benchmark performance. Following PRISMA guidelines, we systematically queried major academic databases, screened the resulting literature, and synthesized a representative body of peer-reviewed studies spanning a range of imaging modalities, anatomical targets, and learning paradigms. Our analysis covers the architectural taxonomy of hybrid designs, their learning and optimization strategies, and the evaluation practices adopted across the reviewed literature. The findings reveal that while hybrid models consistently demonstrate competitive performance, critical limitations persist: high computational overhead, insufficient external validation, and a heavy reliance on fully supervised learning constrain their real-world applicability. We conclude with a set of forward-looking recommendations emphasizing efficiency-aware design, standardized cross-domain evaluation, and the broader adoption of self-supervised learning strategies to advance the clinical translation of hybrid architectures.\u003c/p\u003e","manuscriptTitle":"Do Hybrid CNN–Transformer Architectures Really Generalize? A Systematic Review for Medical Imaging","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-26 05:36:27","doi":"10.21203/rs.3.rs-9216007/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":"be6ccb3e-7ced-4093-8154-a4c6ad3ffa2f","owner":[],"postedDate":"March 26th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":65077665,"name":"Medical Informatics"},{"id":65077666,"name":"Biomedical Engineering"}],"tags":[],"updatedAt":"2026-03-26T05:36:27+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-26 05:36:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9216007","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9216007","identity":"rs-9216007","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}