Distilling clinical sensitivity: Explainable and lightweight knee osteoarthritis diagnosis via decoupled knowledge distillation

Distilling clinical sensitivity: Explainable and lightweight knee osteoarthritis diagnosis via decoupled knowledge distillation | 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 Distilling clinical sensitivity: Explainable and lightweight knee osteoarthritis diagnosis via decoupled knowledge distillation Umar Hasan, Muhammad Ali Nayeem This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8648078/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 16 You are reading this latest preprint version Abstract Knee Osteoarthritis (KOA) is a leading cause of global disability, necessitating early and accurate diagnosis to prevent irreversible joint degeneration. While deep learning has automated radiographic assessment, current state-of-the-art models incur massive computational costs, hindering deployment in resource-constrained clinical settings. To bridge this gap, we propose a high-efficiency framework utilizing Decoupled Knowledge Distillation (DKD). We employ a Swin Transformer teacher to guide a lightweight GhostNetV2 student, decoupling the distillation loss into Target Class (TCKD) and Non-Target Class (NCKD) components. The method achieves a binary classification accuracy of 87.02% (95% CI: 0.870 ± 0.016) on the Osteoarthritis Initiative (OAI) dataset. While overall accuracy is statistically comparable to a non-distilled baseline (P > 0.05),the DKD framework yields a statistically significant improvement in diagnostic sensitivity (0.825 vs. 0.767; non-overlapping 95% CI), critically reducing false negatives. The model matches the heavy teacher’s performance while being approximately 6 times smaller and 5 times faster. Validated via LayerCAM, the model demonstrates precise localization of osteophytes consistent with radiological interpretations, presenting a scalable, clinically safe solution for automated KOA screening. Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Physical sciences/Engineering Health sciences/Health care Physical sciences/Mathematics and computing Health sciences/Medical research Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 30 Apr, 2026 Reviews received at journal 13 Apr, 2026 Reviews received at journal 05 Apr, 2026 Reviewers agreed at journal 22 Mar, 2026 Reviews received at journal 17 Mar, 2026 Reviewers agreed at journal 15 Mar, 2026 Reviewers agreed at journal 15 Mar, 2026 Reviewers agreed at journal 14 Mar, 2026 Reviews received at journal 08 Mar, 2026 Reviewers agreed at journal 24 Feb, 2026 Reviewers agreed at journal 16 Feb, 2026 Reviewers invited by journal 28 Jan, 2026 Editor invited by journal 23 Jan, 2026 Editor assigned by journal 21 Jan, 2026 Submission checks completed at journal 21 Jan, 2026 First submitted to journal 20 Jan, 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-8648078","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":581686754,"identity":"e67b53be-97a4-4386-a443-97d15ea40275","order_by":0,"name":"Umar Hasan","email":"","orcid":"","institution":"North South University","correspondingAuthor":false,"prefix":"","firstName":"Umar","middleName":"","lastName":"Hasan","suffix":""},{"id":581686755,"identity":"e26b475f-e1f2-4729-97ae-8368c53a22ca","order_by":1,"name":"Muhammad Ali Nayeem","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA00lEQVRIiWNgGAWjYDADfjDJRooWyQaStRgcIFaLbgPvwc88FTbRxsdzDBg+lB1mkG8/gF+L2QG+ZGmeM2m52868MWCcce4wg8GZBEJaeAykedsO5267kWPADGQwGDAQ1mL8m/ff/9zNM4Ba/gK1yPc/IKjFTJq34UDuBgmgFkagFoYbhGw5zGNmOedYcu6MM88KDvacS+cxuEHIluM9xjfe1Njl9rcnb3zwo8xaTr6fgC0MzAwMTDxgVgLDASDJQ0A9BDD+gGoZBaNgFIyCUYAVAADv9UQatpqZ6AAAAABJRU5ErkJggg==","orcid":"","institution":"Qassim University","correspondingAuthor":true,"prefix":"","firstName":"Muhammad","middleName":"Ali","lastName":"Nayeem","suffix":""}],"badges":[],"createdAt":"2026-01-20 10:48:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8648078/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8648078/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101751829,"identity":"c3a29e7f-52bb-43c6-854b-8bd752155418","added_by":"auto","created_at":"2026-02-03 10:23:51","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1506489,"visible":true,"origin":"","legend":"","description":"","filename":"Source.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8648078/v1_covered_9c80d2c5-812d-4ee2-8bd2-faf2f05098b8.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Distilling clinical sensitivity: Explainable and lightweight knee osteoarthritis diagnosis via decoupled knowledge distillation","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-8648078/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8648078/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Knee Osteoarthritis (KOA) is a leading cause of global disability, necessitating early and accurate diagnosis to prevent irreversible joint degeneration. While deep learning has automated radiographic assessment, current state-of-the-art models incur massive computational costs, hindering deployment in resource-constrained clinical settings. To bridge this gap, we propose a high-efficiency framework utilizing Decoupled Knowledge Distillation (DKD). We employ a Swin Transformer teacher to guide a lightweight GhostNetV2 student, decoupling the distillation loss into Target Class (TCKD) and Non-Target Class (NCKD) components. The method achieves a binary classification accuracy of 87.02% (95% CI: 0.870 ± 0.016) on the Osteoarthritis Initiative (OAI) dataset. While overall accuracy is statistically comparable to a non-distilled baseline (P \u003e 0.05),the DKD framework yields a statistically significant improvement in diagnostic sensitivity (0.825 vs. 0.767; non-overlapping 95% CI), critically reducing false negatives. The model matches the heavy teacher’s performance while being approximately 6 times smaller and 5 times faster. Validated via LayerCAM, the model demonstrates precise localization of osteophytes consistent with radiological interpretations, presenting a scalable, clinically safe solution for automated KOA screening.","manuscriptTitle":"Distilling clinical sensitivity: Explainable and lightweight knee osteoarthritis diagnosis via decoupled knowledge distillation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-30 05:46:28","doi":"10.21203/rs.3.rs-8648078/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-30T13:59:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-13T13:04:11+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-05T13:20:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"294331743115726414614190115351709792569","date":"2026-03-22T10:54:46+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-17T15:51:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"283060877197049460883004323007170054541","date":"2026-03-15T12:41:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"230955965650976327046926878290855351422","date":"2026-03-15T05:50:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"288646442133170382755346606227783182666","date":"2026-03-14T14:06:25+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-08T17:32:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"74123655311411063085899054427352074720","date":"2026-02-24T05:24:54+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"160156117975813592096741088877572136238","date":"2026-02-16T07:56:02+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-28T08:56:31+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-23T14:02:56+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-21T05:21:33+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-21T05:20:21+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-20T10:07:46+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":"2210fca7-ac09-4876-be5b-618fab491456","owner":[],"postedDate":"January 30th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-04-30T13:59:49+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":61881643,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":61881644,"name":"Health sciences/Diseases"},{"id":61881645,"name":"Physical sciences/Engineering"},{"id":61881646,"name":"Health sciences/Health care"},{"id":61881647,"name":"Physical sciences/Mathematics and computing"},{"id":61881648,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-05-15T04:53:26+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-30 05:46:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8648078","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8648078","identity":"rs-8648078","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}