CerebroNet: A systematically derived explainable brain tumor classifier for resource-constrained MRI diagnostics

CerebroNet: A systematically derived explainable brain tumor classifier for resource-constrained MRI diagnostics | 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 CerebroNet: A systematically derived explainable brain tumor classifier for resource-constrained MRI diagnostics Umar Hasan, Muhammad Ali Nayeem, Riasat Khan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8315329/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 While Artificial Intelligence (AI) has demonstrated potential in assisting MRI-based brain tumor diagnosis, its real-world utility appears to be limited by high computational demands and demographic biases inherent in Western-centric datasets. This study proposes a systematic optimization framework to develop a resource-efficient, explainable classifier tailored to an underrepresented South Asian cohort. We initially benchmarked 21 deep learning architectures to isolate the experimentally optimal teacher model. Subsequently, "CerebroNet" was derived via architectural ablation and refined through a holistic pipeline integrating knowledge distillation and unstructured pruning. The resulting model contains merely 0.637 million parameters, representing a 17-fold reduction relative to state-of-the-art benchmarks. Despite this substantial compression, CerebroNet seems to retain over 96% of the teacher's diagnostic fidelity, attaining 95.79% accuracy on augmented data and 96.21% on raw clinical scans. Validation on an external dataset yielded 91% accuracy, suggesting a degree of robust generalization. Furthermore, explainable AI (Layer-CAM) analysis indicated that the student model likely preserves the relevant visual reasoning of the teacher. By balancing computational efficiency with demographic inclusivity, this work attempts to offer a reproducible strategy for implementing reliable AI diagnostics in resource-constrained healthcare environments. Biological sciences/Computational biology and bioinformatics 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: 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-8315329","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":610483181,"identity":"5a114799-9d4f-4b27-bb91-1954a227dc3c","order_by":0,"name":"Umar Hasan","email":"","orcid":"","institution":"North South University","correspondingAuthor":false,"prefix":"","firstName":"Umar","middleName":"","lastName":"Hasan","suffix":""},{"id":610483186,"identity":"87031cf4-828b-4c6c-abe4-badee5c7dc45","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":""},{"id":610483187,"identity":"71a54dab-70bb-46b3-b7d6-94d16f9ef56e","order_by":2,"name":"Riasat Khan","email":"","orcid":"","institution":"North South University","correspondingAuthor":false,"prefix":"","firstName":"Riasat","middleName":"","lastName":"Khan","suffix":""}],"badges":[],"createdAt":"2025-12-09 08:53:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8315329/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8315329/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108805376,"identity":"fb2c0d30-9a0f-4e8c-ad1d-d10f1517258f","added_by":"auto","created_at":"2026-05-08 15:25:46","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2738260,"visible":true,"origin":"","legend":"","description":"","filename":"SourceSciReps.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8315329/v1_covered_e28beebd-adaa-46b4-a723-7ce57e75414a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"CerebroNet: A systematically derived explainable brain tumor classifier for resource-constrained MRI diagnostics","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"","lastPublishedDoi":"10.21203/rs.3.rs-8315329/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8315329/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"While Artificial Intelligence (AI) has demonstrated potential in assisting MRI-based brain tumor diagnosis, its real-world utility appears to be limited by high computational demands and demographic biases inherent in Western-centric datasets. This study proposes a systematic optimization framework to develop a resource-efficient, explainable classifier tailored to an underrepresented South Asian cohort. We initially benchmarked 21 deep learning architectures to isolate the experimentally optimal teacher model. Subsequently, \"CerebroNet\" was derived via architectural ablation and refined through a holistic pipeline integrating knowledge distillation and unstructured pruning. The resulting model contains merely 0.637 million parameters, representing a 17-fold reduction relative to state-of-the-art benchmarks. Despite this substantial compression, CerebroNet seems to retain over 96% of the teacher's diagnostic fidelity, attaining 95.79% accuracy on augmented data and 96.21% on raw clinical scans. Validation on an external dataset yielded 91% accuracy, suggesting a degree of robust generalization. Furthermore, explainable AI (Layer-CAM) analysis indicated that the student model likely preserves the relevant visual reasoning of the teacher. By balancing computational efficiency with demographic inclusivity, this work attempts to offer a reproducible strategy for implementing reliable AI diagnostics in resource-constrained healthcare environments.","manuscriptTitle":"CerebroNet: A systematically derived explainable brain tumor classifier for resource-constrained MRI diagnostics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-26 18:20:56","doi":"10.21203/rs.3.rs-8315329/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":"2210fca7-ac09-4876-be5b-618fab491456","owner":[],"postedDate":"March 26th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Rejected","date":"2026-05-06T15:40:46+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-04T12:51:36+00:00","index":391,"fulltext":""},{"type":"reviewerAgreed","content":"63185252287905404448755833743219347387","date":"2026-05-02T11:46:48+00:00","index":389,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":64949871,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":64949872,"name":"Health sciences/Health care"},{"id":64949873,"name":"Physical sciences/Mathematics and computing"},{"id":64949874,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-05-06T15:55:44+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-26 18:20:56","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8315329","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8315329","identity":"rs-8315329","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}