Hybrid Deep Ensemble Architecture for Robust Diabetic Retinopathy Classification: Leveraging Transfer Learning and CNN-Transformer Synergy

Hybrid Deep Ensemble Architecture for Robust Diabetic Retinopathy Classification: Leveraging Transfer Learning and CNN-Transformer Synergy | 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 Hybrid Deep Ensemble Architecture for Robust Diabetic Retinopathy Classification: Leveraging Transfer Learning and CNN-Transformer Synergy Youssef Maaod, Tamer Abuhamd, Shaker El-Sappagh This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8614100/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 19 You are reading this latest preprint version Abstract Diabetic Retinopathy (DR) is still one of the main reasons for vision loss worldwide, especially in places where people do not have easy access to regular eye checkups. Early and accurate disease detection is important to avoid permanent damage, but traditional methods are slow and sometimes inconsistent. This study proposes a deep learning framework that combines convolutional neural networks (CNNs), vision transformers, transfer learning, and ensemble techniques to improve DR detection. We used the APTOS 2019 dataset and tested the capabilities of 23 different pre-trained models. Then, we fine-tuned the top models and designed hybrid architectures by combining the best-performing CNNs and transformers in parallel and sequential ways to capture both image spatial features in short and long contexts. The best performance came from combining the top sequential hybrid models using the soft voting architecture, where we got an accuracy of 93.10%, ROC AUC of 99.22%, and F1-score of 93.07%. The optimized model showed that mixing different models and using ensemble methods can lead to better and more stable DR detection decisions. Our approach is a step toward building a reliable and automated system that could help doctors in real-world settings. Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Physical sciences/Engineering Health sciences/Health care Physical sciences/Mathematics and computing Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 27 Feb, 2026 Reviews received at journal 25 Feb, 2026 Reviews received at journal 20 Feb, 2026 Reviewers agreed at journal 19 Feb, 2026 Reviewers agreed at journal 19 Feb, 2026 Reviewers agreed at journal 18 Feb, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviewers agreed at journal 17 Feb, 2026 Reviewers invited by journal 17 Feb, 2026 Editor assigned by journal 12 Feb, 2026 Editor invited by journal 12 Feb, 2026 Submission checks completed at journal 09 Feb, 2026 First submitted to journal 09 Feb, 2026 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-8614100","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":593197400,"identity":"2924034c-34d8-4639-9773-f4cbf9c9bea8","order_by":0,"name":"Youssef Maaod","email":"","orcid":"","institution":"Galala University","correspondingAuthor":false,"prefix":"","firstName":"Youssef","middleName":"","lastName":"Maaod","suffix":""},{"id":593197401,"identity":"2804a62e-0737-440a-ab2c-5bf80761fbec","order_by":1,"name":"Tamer Abuhamd","email":"","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Tamer","middleName":"","lastName":"Abuhamd","suffix":""},{"id":593197402,"identity":"009f2f31-96e4-4716-8f68-aa6384df5736","order_by":2,"name":"Shaker El-Sappagh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABL0lEQVRIiWNgGAWjYDACCQZmMM0H5Sfwgyk2IrTA1CRINpCsxeAAAS3ys5sfG3zcwyDHxn/42eOCmto84/NrDBg+lB1m4OdfgFWLwZ1jxokznjEYs0mkmRvPOHa82OzGGwPGGecOM0jOeIBdi0SC8WGeAwyJbRIMZtI8bMcSt904Y8DM23aYweDGAewOm5H++fCfAwz1bfzHv0nz/DuWuHkGUMtfoBZ7HFoYbuQYJzMcYEhgY8gxk+Ztq0ncwN9jwMwIsoW/AbvDbuQUG/YckDBsk8gpN57ZdyBxxg22goM959J5JG7gCLEZ6ZslfhywkefnP77tccG3usT+/sMbH/wos5bj78fhMAiQABFswBg6DGQnMIDU8oAYhABISx0DAz/McDhjFIyCUTAKRjgAAMJeYp4sX5okAAAAAElFTkSuQmCC","orcid":"","institution":"Galala University","correspondingAuthor":true,"prefix":"","firstName":"Shaker","middleName":"","lastName":"El-Sappagh","suffix":""}],"badges":[],"createdAt":"2026-01-16 00:23:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8614100/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8614100/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103049746,"identity":"a5bb263f-35d1-490e-b334-62ef0d2cfb62","added_by":"auto","created_at":"2026-02-20 07:45:26","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":737260,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8614100/v1_covered_82f2e2c8-17f2-4d00-a605-b587cc7cb5f8.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Hybrid Deep Ensemble Architecture for Robust Diabetic Retinopathy Classification: Leveraging Transfer Learning and CNN-Transformer Synergy","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8614100/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8614100/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDiabetic Retinopathy (DR) is still one of the main reasons for vision loss worldwide, especially in places where people do not have easy access to regular eye checkups. Early and accurate disease detection is important to avoid permanent damage, but traditional methods are slow and sometimes inconsistent. This study proposes a deep learning framework that combines convolutional neural networks (CNNs), vision transformers, transfer learning, and ensemble techniques to improve DR detection. We used the APTOS 2019 dataset and tested the capabilities of 23 different pre-trained models. Then, we fine-tuned the top models and designed hybrid architectures by combining the best-performing CNNs and transformers in parallel and sequential ways to capture both image spatial features in short and long contexts. The best performance came from combining the top sequential hybrid models using the soft voting architecture, where we got an accuracy of 93.10%, ROC AUC of 99.22%, and F1-score of 93.07%. The optimized model showed that mixing different models and using ensemble methods can lead to better and more stable DR detection decisions. Our approach is a step toward building a reliable and automated system that could help doctors in real-world settings.\u003c/p\u003e","manuscriptTitle":"Hybrid Deep Ensemble Architecture for Robust Diabetic Retinopathy Classification: Leveraging Transfer Learning and CNN-Transformer Synergy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-19 06:04:13","doi":"10.21203/rs.3.rs-8614100/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-27T05:12:39+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-26T03:10:40+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-20T18:54:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"133261486618544943175812642263546111454","date":"2026-02-19T12:34:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"322910930992505167831349605938836574353","date":"2026-02-19T08:43:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"30729766406650661026560822959816453527","date":"2026-02-18T09:00:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"63985826608665842093906760977560614326","date":"2026-02-17T18:27:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"230993577463930506788714764989729862977","date":"2026-02-17T17:34:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"315754026585023418441105810358925453919","date":"2026-02-17T16:56:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"180956043258277026555674607157982137247","date":"2026-02-17T16:32:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"295232529936092804202326929474312654016","date":"2026-02-17T11:00:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"229153594477356879828410650705385716941","date":"2026-02-17T09:05:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"111502244286246315375182969287970804479","date":"2026-02-17T08:54:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"172823374876936756057833930246143064258","date":"2026-02-17T08:38:18+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-17T08:30:22+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-12T07:33:20+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-12T06:54:12+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-09T17:24:49+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-02-09T17:19:04+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"683b46f1-b002-4c8b-8dad-f48e71ebe2bc","owner":[],"postedDate":"February 19th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":63120929,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":63120930,"name":"Health sciences/Diseases"},{"id":63120931,"name":"Physical sciences/Engineering"},{"id":63120932,"name":"Health sciences/Health care"},{"id":63120933,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-04-08T00:08:16+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-19 06:04:13","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8614100","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8614100","identity":"rs-8614100","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}