An Explainable Ensemble Based Approach to Diabetic Retinopathy Grading

An Explainable Ensemble Based Approach to Diabetic Retinopathy Grading | 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 An Explainable Ensemble Based Approach to Diabetic Retinopathy Grading Javier Civit-Masot, Francisco Luna-Perejon, Luis Muñoz-Saavedra, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6878828/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 04 Feb, 2026 Read the published version in Medical & Biological Engineering & Computing → Version 1 posted You are reading this latest preprint version Abstract Diabetic retinopathy is a dangerous eye anomaly that may cause vision loss and blindness in people with diabetes (more than 500 million adults in 2021). This study presents a novel approach to diabetic retinopathy grading using explainable deep learning techniques to create a diagnosis aid tool that provides a fully detailed final report with alternative explanations for the result. This is clear novelty over existing solutions. The proposed approach is based on an ensemble of two very efficient deep learning networks: efficientNetV2 and ConvNeXt. This deep learning model is evaluated using a public available dataset and compared with the results obtained from previous works. Also, explainable deep learning techniques are applied to present the final report. The architecture is capable of achieving state-of-the-art performance for both the two-class problem and the five-class international clinical diabetes retinopathy (ICDR) classification problem. This work achieves a 96.7% accuracy and an AUC over 96% for all classes. The ensemble provides explainability records which are significantly improved when compared with those obtained from a single network. The ensemble architecture provides good quality explanations to the ophthalmologist with several configurable superimposed heat maps and two probabilityordered diagnostic suggestions, including a quality factor indicating the estimation of the probability ratio between the alternatives. This makes it a valuable diagnostic assistance tool for ophthalmologists and other healthcare professionals. This study introduces a novel approach to diabetic retinopathy (DR) grading, utilizing explainable deep learning techniquesto create a diagnostic aid tool. The system provides a detailed final report with alternative explanations, which is a clear advancement over existing solutions. Our method is based on an ensemble of two highly efficient deep learning networks: EfficientNetV2 and ConvNeXt. This ensemble approach achieves a 96.7% accuracy and an AUC over 96% for the ICDR classification problem, outperforming single-network models in both accuracy and explainability. The system offers ophthalmologists configurable heat maps and probability-ordered diagnostic suggestions, making it a valuable diagnostic tool. Artificial Intelligence and Machine Learning Biomedical Engineering Diabetic Retinopathy explainable AI Deep Learning Explainable ensemble Medical Imaging Medical Diagnosis Tool Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Published Journal Publication published 04 Feb, 2026 Read the published version in Medical & Biological Engineering & Computing → 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-6878828","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":470279769,"identity":"0f8298b1-5c40-4911-aa27-517b0293a9cf","order_by":0,"name":"Javier Civit-Masot","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAqklEQVRIiWNgGAWjYFCCAwwMCQwMcqRrMSbdrsQGopXqNh5+JvGg4l76hvMHGB/+IEaL2YFjZhIJZ4pzN9xIYDbmIU7LAWODxLYEoBYGNmmiHGZ24Phng8R/CekG5w+w/yTSYWcMHyQ2JCQYHEhgYyDSYWcKHyQcSzCceSOxWZo4LTeObzj4oyZBnu/84YMfiXIYg8QBGIuxgSgNDAz8xCocBaNgFIyCkQsA5eU5kuyy/g4AAAAASUVORK5CYII=","orcid":"","institution":"Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla*Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla","correspondingAuthor":true,"prefix":"","firstName":"Javier","middleName":"","lastName":"Civit-Masot","suffix":""},{"id":470279770,"identity":"a4741f7b-a4a6-4f03-b5ae-8f081afac8bb","order_by":1,"name":"Francisco Luna-Perejon","email":"","orcid":"","institution":"Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla*Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla","correspondingAuthor":false,"prefix":"","firstName":"Francisco","middleName":"","lastName":"Luna-Perejon","suffix":""},{"id":470279771,"identity":"37627e3e-963d-46d5-8520-c455c9dafd56","order_by":2,"name":"Luis Muñoz-Saavedra","email":"","orcid":"","institution":"Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla*Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla","correspondingAuthor":false,"prefix":"","firstName":"Luis","middleName":"","lastName":"Muñoz-Saavedra","suffix":""},{"id":470279772,"identity":"9a83d60d-8b94-41fa-a61a-7c88fdca984a","order_by":3,"name":"José María Rodríguez Corral","email":"","orcid":"","institution":"Applied Robotics Research Group, School of Engineering, Universidad de Cádiz","correspondingAuthor":false,"prefix":"","firstName":"José","middleName":"María Rodríguez","lastName":"Corral","suffix":""},{"id":470279773,"identity":"0ac3fbe8-a64a-4a48-86de-3060961642b5","order_by":4,"name":"Manuel Domínguez-Morales","email":"","orcid":"","institution":"Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla*Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla","correspondingAuthor":false,"prefix":"","firstName":"Manuel","middleName":"","lastName":"Domínguez-Morales","suffix":""},{"id":470279774,"identity":"8628c93b-0dd8-4c2e-9c00-b7f569894339","order_by":5,"name":"Anton Civit","email":"","orcid":"","institution":"Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla*Robotics and Computer Technology Lab, ETSII, Universidad de Sevilla","correspondingAuthor":false,"prefix":"","firstName":"Anton","middleName":"","lastName":"Civit","suffix":""}],"badges":[],"createdAt":"2025-06-12 09:34:30","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6878828/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6878828/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11517-026-03514-2","type":"published","date":"2026-02-05T00:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":102243795,"identity":"d1fa094a-55b9-4f8a-b7ea-f1a4d53c95aa","added_by":"auto","created_at":"2026-02-09 17:29:20","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1497204,"visible":true,"origin":"","legend":"","description":"","filename":"MBECRetinopata.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6878828/v1_covered_4d83b65d-e56a-40fe-a235-c45adce2bf03.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eAn Explainable Ensemble Based Approach to Diabetic Retinopathy Grading\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"University of Seville","isAcceptedByJournal":true,"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":"Diabetic Retinopathy, explainable AI, Deep Learning, Explainable ensemble, Medical Imaging, Medical Diagnosis Tool","lastPublishedDoi":"10.21203/rs.3.rs-6878828/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6878828/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDiabetic retinopathy is a dangerous eye anomaly that may cause vision loss and blindness in people with diabetes (more than 500 million adults in 2021). This study presents a novel approach to diabetic retinopathy grading using explainable deep learning techniques to create a diagnosis aid tool that provides a fully detailed final report with alternative explanations for the result. This is clear novelty over existing solutions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe proposed approach is based on an ensemble of two very efficient deep learning networks: efficientNetV2 and ConvNeXt. This deep learning model is evaluated using a public available dataset and compared with the results obtained from previous works. Also, explainable deep learning techniques are applied to present the final report.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe architecture is capable of achieving state-of-the-art performance for both the two-class problem and the five-class international clinical diabetes retinopathy (ICDR) classification problem. This work achieves a 96.7% accuracy and an AUC over 96% for all classes. The ensemble provides explainability records which are significantly improved when compared with those obtained from a single network. The ensemble architecture provides good quality explanations to the ophthalmologist with several configurable superimposed heat maps and two probabilityordered diagnostic suggestions, including a quality factor indicating the estimation of the probability ratio between the alternatives. This makes it a valuable diagnostic assistance tool for ophthalmologists and other healthcare professionals. This study introduces a novel approach to diabetic retinopathy (DR) grading, utilizing explainable deep learning techniquesto create a diagnostic aid tool. The system provides a detailed final report with alternative explanations, which is a clear advancement over existing solutions. Our method is based on an ensemble of two highly efficient deep learning networks: EfficientNetV2 and ConvNeXt. This ensemble approach achieves a 96.7% accuracy and an AUC over 96% for the ICDR classification problem, outperforming single-network models in both accuracy and explainability. The system offers ophthalmologists configurable heat maps and probability-ordered diagnostic suggestions, making it a valuable diagnostic tool.\u003c/p\u003e","manuscriptTitle":"An Explainable Ensemble Based Approach to Diabetic Retinopathy Grading","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-13 02:26:46","doi":"10.21203/rs.3.rs-6878828/v1","editorialEvents":[{"type":"communityComments","content":1}],"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":"650d9c51-b414-4956-b327-3bea15e1f834","owner":[],"postedDate":"June 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":49937729,"name":"Artificial Intelligence and Machine Learning"},{"id":49937730,"name":"Biomedical Engineering"}],"tags":[],"updatedAt":"2026-02-09T17:29:14+00:00","versionOfRecord":{"articleIdentity":"rs-6878828","link":"https://doi.org/10.1007/s11517-026-03514-2","journal":{"identity":"medical-and-biological-engineering-and-computing","isVorOnly":false,"title":"Medical \u0026 Biological Engineering \u0026 Computing"},"publishedOn":"2026-02-05 00:00:00","publishedOnDateReadable":"February 5th, 2026"},"versionCreatedAt":"2025-06-13 02:26:46","video":"","vorDoi":"10.1007/s11517-026-03514-2","vorDoiUrl":"https://doi.org/10.1007/s11517-026-03514-2","workflowStages":[]},"version":"v1","identity":"rs-6878828","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6878828","identity":"rs-6878828","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}