Towards Precision Pneumonia Diagnosis: A Hybrid Deep Learning Approach Integrating ResNet50 and Vision Transformer

Towards Precision Pneumonia Diagnosis: A Hybrid Deep Learning Approach Integrating ResNet50 and Vision Transformer | 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 Towards Precision Pneumonia Diagnosis: A Hybrid Deep Learning Approach Integrating ResNet50 and Vision Transformer Riduana Adneen Adrita, Md Mahenur Islam, Mohammad Khaled Sohel This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6853898/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 Pneumonia remains one of the leading killers in the world, and thus its early diagnosis is critical to successful treatment. Chest X-ray is a valuable diagnostic imaging modality, but its manual interpretation by radiologists is both time-consuming and prone to mistakes. This paper proposes a hybrid deep learning model for classifying pneumonia vs normal lung using chest X-ray images. The model here integrates the feature extraction capability of ResNet50 with Vision Transformer (ViT) self-attention for better classification accuracy. By utilizing the strengths of these architectures combined, the model achieves satisfactory accuracy in classifying pneumonia-affected and normal lung images. The approach is evaluated on an open chest X-ray dataset and records 96.56% validation accuracy and 86.54% test accuracy. This hybrid model has tremendous potential in automating pneumonia identification, saving diagnostic time, and helping medical practitioners achieve better patient outcomes. The model can be optimized further and the dataset increased for even greater performance in future studies. ResNet50 Vision Transformer (ViT) hybrid model medical image classification attention mechanism computer-aided diagnosis (CAD) convolutional neural network (CNN) radiology automation Full Text Additional Declarations No competing interests reported. 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-6853898","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":477237087,"identity":"d2a23fd7-8f60-4190-87c1-e81bb3dddbaf","order_by":0,"name":"Riduana Adneen Adrita","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDUlEQVRIie3RsUoDMRjA8S8E4pL21oR67/AdBxaxnK9yJdBRhEKXDgaEdNK5Q8VXOBdxjGRwkTpLB697CwdOogXv6qh37eiQ/xIC+SUkAfD5/mMUsBqSU01dXmCPmGrKAVKwzUQhsEE0PR/sQeCHkAw4dnjhiN5FuhN6Lz4eEtotbczxJbwWKcnXBs7aNeTQsZG8elbtYw3pcoaL2IiURjcGhrKGCMqPRMtQCm+5xRUu+iVhnZaBftZA5MZcEG2JFhznFTn43EXKPR3JLAXJ0W5Poc2EDU9C86SkZiyaoooNX17K2VzU3yVwd69rM04CCN7z4isJbyfqsViNerUv9lfbrxH7r/f5fD7fr74BD0FYqOs3G9sAAAAASUVORK5CYII=","orcid":"","institution":"Daffodil International University","correspondingAuthor":true,"prefix":"","firstName":"Riduana","middleName":"Adneen","lastName":"Adrita","suffix":""},{"id":477237088,"identity":"a128a477-63f5-4356-a08a-1b9894c27547","order_by":1,"name":"Md Mahenur Islam","email":"","orcid":"","institution":"University of Dhaka","correspondingAuthor":false,"prefix":"","firstName":"Md","middleName":"Mahenur","lastName":"Islam","suffix":""},{"id":477237092,"identity":"cde11c04-a09a-42a6-948a-59ec0bcc3ae0","order_by":2,"name":"Mohammad Khaled Sohel","email":"","orcid":"","institution":"Daffodil International University","correspondingAuthor":false,"prefix":"","firstName":"Mohammad","middleName":"Khaled","lastName":"Sohel","suffix":""}],"badges":[],"createdAt":"2025-06-09 11:23:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6853898/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6853898/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90538645,"identity":"9b8ac200-f305-45e0-ac99-2d566cb5d310","added_by":"auto","created_at":"2025-09-03 21:47:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":708051,"visible":true,"origin":"","legend":"","description":"","filename":"pnemuniaclassificationfinalversion.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6853898/v1_covered_3901ba08-a211-4c40-b838-8f56558a4faf.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Towards Precision Pneumonia Diagnosis: A Hybrid Deep Learning Approach Integrating ResNet50 and Vision Transformer","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":"ResNet50, Vision Transformer (ViT), hybrid model, medical image classification, attention mechanism, computer-aided diagnosis (CAD), convolutional neural network (CNN), radiology automation","lastPublishedDoi":"10.21203/rs.3.rs-6853898/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6853898/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Pneumonia remains one of the leading killers in the world, and thus its early diagnosis is critical to successful treatment. Chest X-ray is a valuable diagnostic imaging modality, but its manual interpretation by radiologists is both time-consuming and prone to mistakes. This paper proposes a hybrid deep learning model for classifying pneumonia vs normal lung using chest X-ray images. The model here integrates the feature extraction capability of ResNet50 with Vision Transformer (ViT) self-attention for better classification accuracy. By utilizing the strengths of these architectures combined, the model achieves satisfactory accuracy in classifying pneumonia-affected and normal lung images. The approach is evaluated on an open chest X-ray dataset and records 96.56% validation accuracy and 86.54% test accuracy. This hybrid model has tremendous potential in automating pneumonia identification, saving diagnostic time, and helping medical practitioners achieve better patient outcomes. The model can be optimized further and the dataset increased for even greater performance in future studies.","manuscriptTitle":"Towards Precision Pneumonia Diagnosis: A Hybrid Deep Learning Approach Integrating ResNet50 and Vision Transformer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-15 07:04:40","doi":"10.21203/rs.3.rs-6853898/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":"694e525d-0dfa-4147-b517-cfbd9ce1640f","owner":[],"postedDate":"July 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-09-03T21:39:30+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-15 07:04:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6853898","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6853898","identity":"rs-6853898","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}