Adaptive Transition Metaheuristic for Swin Transformer Hyperparameter Optimization in Brain Tumor MRI Classification

Adaptive Transition Metaheuristic for Swin Transformer Hyperparameter Optimization in Brain Tumor MRI Classification | 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 Adaptive Transition Metaheuristic for Swin Transformer Hyperparameter Optimization in Brain Tumor MRI Classification Abdullahi Ahmed Abdirahman, Abdirahman Osman Hashi, Ubaid Mohamed Dahir, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9136325/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Accurate brain tumor classification from magnetic resonance imaging (MRI) remains a significant challenge in medical image analysis due to the high sensitivity of deep learning models to hyperparameter configurations and the limitations of conventional metaheuristic optimization methods that rely on fixed thresholds to balance exploration and exploitation. These rigid mechanisms often result in premature convergence and inefficient search behavior. To address this limitation, this paper proposes A²-HHO, a hybrid optimization framework that integrates the Aquila Optimizer and Harris Hawks Optimization with a Population Diversity Index (PDI) driven adaptive phase transition mechanism. The proposed framework dynamically regulates switching between global exploration, local exploitation, and stagnation-triggered re-diversification based on real-time population diversity, enabling a more effective search process. The framework is coupled with a Swin Transformer backbone for hierarchical spatial feature extraction and explores a nine-dimensional hyperparameter search space to optimize model performance. Experimental evaluation on three publicly available MRI brain tumor datasets demonstrates that the proposed approach achieves classification accuracies of 93.47%, 94.82%, and 95.63%, outperforming established optimization methods including PSO, GA, WOA, and fixed-threshold AO-HHO with statistically significant improvements (p < 0.05), while completing the optimization process approximately 3.36× faster than PSO, highlighting its effectiveness and computational efficiency for brain tumor classification. Biological sciences/Computational biology and bioinformatics Physical sciences/Engineering Physical sciences/Mathematics and computing Brain tumor classification Swin Transformer Metaheuristic optimization Aquila Optimizer Harris Hawks Optimization Hyperparameter optimization Population Diversity Index Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 22 Apr, 2026 Reviews received at journal 20 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers invited by journal 20 Apr, 2026 Editor assigned by journal 29 Mar, 2026 Editor invited by journal 27 Mar, 2026 Submission checks completed at journal 25 Mar, 2026 First submitted to journal 25 Mar, 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-9136325","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":630057421,"identity":"4237f1ef-2cc2-4e8e-aa97-e34e40db40af","order_by":0,"name":"Abdullahi Ahmed Abdirahman","email":"","orcid":"","institution":"SIMAD University","correspondingAuthor":false,"prefix":"","firstName":"Abdullahi","middleName":"Ahmed","lastName":"Abdirahman","suffix":""},{"id":630057422,"identity":"62f95e34-b80b-485d-b6e5-0ef74a9aea2c","order_by":1,"name":"Abdirahman Osman Hashi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYHACAyCSkLM/3v/xAZDHw0ekFgtjhjMHjA1AWtiI08JQkdhwI8FMAsQlqMXg+OGNnwsKJBgbGxLSKr/m2MmwMTA/fHQDn5YzacXSMwwkmJkZDhy7LbstGegwNmPjHDxazA7kGEjzGEiwsTE2tt2W3MYM1MLDJo1Xy/k3xr+BWnh4mJnZiiW31ROh5UaOGcgWCaA9bIwftx0mrMX+xrMya6AWAwOgPdKM247zsDET8Itkf/Lm2zx/6uo3yL9h/PhzW7U9P3vzw8f4tKAAZh4wSaxyEGD8QYrqUTAKRsEoGDEAAFWfQLeAuKlYAAAAAElFTkSuQmCC","orcid":"","institution":"SIMAD University","correspondingAuthor":true,"prefix":"","firstName":"Abdirahman","middleName":"Osman","lastName":"Hashi","suffix":""},{"id":630057423,"identity":"f0de4bf7-878a-4e64-9cad-b9bf59d49601","order_by":2,"name":"Ubaid Mohamed Dahir","email":"","orcid":"","institution":"SIMAD University","correspondingAuthor":false,"prefix":"","firstName":"Ubaid","middleName":"Mohamed","lastName":"Dahir","suffix":""},{"id":630057425,"identity":"05df0a6c-4455-4e2f-804d-9260df1d4638","order_by":3,"name":"Mohamed Abdirahman Elmi","email":"","orcid":"","institution":"SIMAD University","correspondingAuthor":false,"prefix":"","firstName":"Mohamed","middleName":"Abdirahman","lastName":"Elmi","suffix":""}],"badges":[],"createdAt":"2026-03-16 10:10:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9136325/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9136325/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108181277,"identity":"23b406d5-2a95-4397-8226-483c2013c887","added_by":"auto","created_at":"2026-04-30 08:58:30","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":993287,"visible":true,"origin":"","legend":"","description":"","filename":"Paper.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9136325/v1_covered_c86437d4-f4d9-4092-b13a-b392e08765bb.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Adaptive Transition Metaheuristic for Swin Transformer Hyperparameter Optimization in Brain Tumor MRI Classification","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":"Brain tumor classification, Swin Transformer, Metaheuristic optimization, Aquila Optimizer, Harris Hawks Optimization, Hyperparameter optimization, Population Diversity Index","lastPublishedDoi":"10.21203/rs.3.rs-9136325/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9136325/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAccurate brain tumor classification from magnetic resonance imaging (MRI) remains a significant challenge in medical image analysis due to the high sensitivity of deep learning models to hyperparameter configurations and the limitations of conventional metaheuristic optimization methods that rely on fixed thresholds to balance exploration and exploitation. These rigid mechanisms often result in premature convergence and inefficient search behavior. To address this limitation, this paper proposes A\u0026sup2;-HHO, a hybrid optimization framework that integrates the Aquila Optimizer and Harris Hawks Optimization with a Population Diversity Index (PDI) driven adaptive phase transition mechanism. The proposed framework dynamically regulates switching between global exploration, local exploitation, and stagnation-triggered re-diversification based on real-time population diversity, enabling a more effective search process. The framework is coupled with a Swin Transformer backbone for hierarchical spatial feature extraction and explores a nine-dimensional hyperparameter search space to optimize model performance. Experimental evaluation on three publicly available MRI brain tumor datasets demonstrates that the proposed approach achieves classification accuracies of 93.47%, 94.82%, and 95.63%, outperforming established optimization methods including PSO, GA, WOA, and fixed-threshold AO-HHO with statistically significant improvements (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), while completing the optimization process approximately 3.36\u0026times; faster than PSO, highlighting its effectiveness and computational efficiency for brain tumor classification.\u003c/p\u003e","manuscriptTitle":"Adaptive Transition Metaheuristic for Swin Transformer Hyperparameter Optimization in Brain Tumor MRI Classification","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-28 21:20:54","doi":"10.21203/rs.3.rs-9136325/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"110310948630331522492279444517967957168","date":"2026-04-22T06:09:50+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-20T06:15:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"140634757857035062769297040570313406861","date":"2026-04-20T06:10:29+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-20T06:01:42+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-29T15:28:52+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-27T12:12:48+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-25T18:58:49+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-03-25T17:57:04+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":"f0bc3869-eee8-40f5-99fd-2226b53df95e","owner":[],"postedDate":"April 28th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67055202,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":67055203,"name":"Physical sciences/Engineering"},{"id":67055204,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-04-28T21:20:55+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-28 21:20:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9136325","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9136325","identity":"rs-9136325","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}