A foundation model for efficient and assumption-free characterization of brain microstructure from diffusion MRI

A foundation model for efficient and assumption-free characterization of brain microstructure from diffusion MRI | 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 A foundation model for efficient and assumption-free characterization of brain microstructure from diffusion MRI Weikang Gong, Ziyu Li, Yan Fu, Michiel Cottaar, Wenchuan Wu, Jianfeng Feng, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8877816/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Diffusion MRI (dMRI) is the primary tool for in vivo investigation of brain microstructure. However, dominant analytical methods rely on biophysical models with specific assumptions and large computational requirements, limiting their biological accuracy and generalisability. Here, we introduce SSDiff, a self-supervised transformer-based foundation model that summarises the entire dMRI scan into a compact set of biologically interpretable latent features, without a priori model assumptions. Trained on ~80,000 scans from diverse populations, protocols, and ages, these features capture individual variability in both local microstructural properties (reflecting aspects of conventional metrics like fractional anisotropy and neurite density) and tract-level architecture. Across three large population datasets (N=65,316), these features significantly improved phenome-wide prediction of over 3,000 non-imaging phenotypes compared to standard dMRI biomarkers, such as diffusion tensor imaging, NODDI and structural connectomes. Crucially, SSDiff demonstrated robust zero-shot generalizability, accurately predicting clinical phenotypes across 7 independent cohorts (N=3,671) with different diseases, including data acquired with simplified clinical-grade protocols. A genome-wide association study on SSDiff features in 53,276 participants revealed 363 independent novel genetic loci for brain microstructure. Mendelian randomization analyses further revealed causal relationships between these features and brain disorders; for instance, linking the microstructural properties of the cerebellum to Parkinson's disease and of the hippocampus to Alzheimer's disease. SSDiff is a powerful, biologically interpretable foundation model for dMRI analysis, enabling more accurate clinical prediction and revealing novel genetic insights into brain health and disease. Biological sciences/Neuroscience Biological sciences/Biological techniques/Imaging/Magnetic resonance imaging Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryTables.xlsx Supplementary Tables Cite Share Download PDF Status: Under Review 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-8877816","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":594474436,"identity":"2160693b-ca22-413c-8329-df2ed8b946ce","order_by":0,"name":"Weikang Gong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYFACxmY460EFAzOYJUGsFmaDM8RpgaoCAjYJorQYHG9uNvi5o9auX7r9WsXBNmt5cwbmg7d5GOzycGo5c7A5sffM8eSZc86U3TjYlm64s4Et2ZqHIbkYlxazG4nNB3jbjiUb3MhJu/2x7TDjhgM8ZtI8DAcSG3Bpuf+w+eBfqJaCg22H7Tcc4P+GX8sNxuZk3rYaO4Mb6ccYgFoSgbaw4dVifyax2Vi27UCC5IwcZokD59KTNxxmM7acY5CMU4tk+/HHkm/b6uz5JdIffjhQZm274XjzwxtvKuxwaoGCw0AFPAYQNjhqDPCrB4I6ewYG9gcElY2CUTAKRsHIBAAjPGGod4EJdgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-5387-4428","institution":"Fudan University","correspondingAuthor":true,"prefix":"","firstName":"Weikang","middleName":"","lastName":"Gong","suffix":""},{"id":594474437,"identity":"051de37e-da37-4f36-9bf4-e8624574300b","order_by":1,"name":"Ziyu Li","email":"","orcid":"","institution":"Centre for Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, Oxford University Centre for Integrative Neuroimaging, University of Oxford","correspondingAuthor":false,"prefix":"","firstName":"Ziyu","middleName":"","lastName":"Li","suffix":""},{"id":594474438,"identity":"1f800e54-79b5-4990-b0a7-8705dd603c29","order_by":2,"name":"Yan Fu","email":"","orcid":"","institution":"Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Yan","middleName":"","lastName":"Fu","suffix":""},{"id":594474439,"identity":"58924c4a-11e3-419f-9773-2328bed13a32","order_by":3,"name":"Michiel Cottaar","email":"","orcid":"https://orcid.org/0000-0003-4679-7724","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Michiel","middleName":"","lastName":"Cottaar","suffix":""},{"id":594474440,"identity":"e3d1d3f9-d886-4a48-916d-ad7a6432b3d6","order_by":4,"name":"Wenchuan Wu","email":"","orcid":"","institution":"Centre for Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, Oxford University Centre for Integrative Neuroimaging, University of Oxford","correspondingAuthor":false,"prefix":"","firstName":"Wenchuan","middleName":"","lastName":"Wu","suffix":""},{"id":594474441,"identity":"2023bcca-f1bd-418c-8897-1df294f9e894","order_by":5,"name":"Jianfeng Feng","email":"","orcid":"https://orcid.org/0000-0001-5987-2258","institution":"Fudan University","correspondingAuthor":false,"prefix":"","firstName":"Jianfeng","middleName":"","lastName":"Feng","suffix":""},{"id":594474442,"identity":"5949b973-68e5-4c55-ad13-5506fdb1867e","order_by":6,"name":"Saad Jbabdi","email":"","orcid":"","institution":"FMRIB, Wellcome Centre for Integrative Neuroimaging, University of Oxford","correspondingAuthor":false,"prefix":"","firstName":"Saad","middleName":"","lastName":"Jbabdi","suffix":""},{"id":594474443,"identity":"9c26e3fb-036b-4f40-90ea-3af2d1b426b7","order_by":7,"name":"Christian Beckmann","email":"","orcid":"","institution":"Donders Institute - Radboud University","correspondingAuthor":false,"prefix":"","firstName":"Christian","middleName":"","lastName":"Beckmann","suffix":""},{"id":594474444,"identity":"4253cebe-cffc-4a80-a047-0e98b77d4e91","order_by":8,"name":"Stephen Smith","email":"","orcid":"","institution":"University of Oxford","correspondingAuthor":false,"prefix":"","firstName":"Stephen","middleName":"","lastName":"Smith","suffix":""}],"badges":[],"createdAt":"2026-02-14 07:10:38","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8877816/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8877816/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103505060,"identity":"11c9e0cb-69a4-44d8-a7b7-4d96adf7eeab","added_by":"auto","created_at":"2026-02-26 13:22:45","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1629276,"visible":true,"origin":"","legend":"","description":"","filename":"SSDiffclean.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8877816/v1_covered_007a4f60-b4a1-420d-a925-8d2fe292abf1.pdf"},{"id":103193395,"identity":"a80a374d-a784-4dd6-8546-2e1329be0557","added_by":"auto","created_at":"2026-02-23 02:59:42","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":8452484,"visible":true,"origin":"","legend":"Supplementary Tables","description":"","filename":"SupplementaryTables.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8877816/v1/4e8c8bde8d334af87d3e5863.xlsx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"A foundation model for efficient and assumption-free characterization of brain microstructure from diffusion MRI","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8877816/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8877816/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Diffusion MRI (dMRI) is the primary tool for in vivo investigation of brain microstructure. However, dominant analytical methods rely on biophysical models with specific assumptions and large computational requirements, limiting their biological accuracy and generalisability. Here, we introduce SSDiff, a self-supervised transformer-based foundation model that summarises the entire dMRI scan into a compact set of biologically interpretable latent features, without a priori model assumptions. Trained on ~80,000 scans from diverse populations, protocols, and ages, these features capture individual variability in both local microstructural properties (reflecting aspects of conventional metrics like fractional anisotropy and neurite density) and tract-level architecture. Across three large population datasets (N=65,316), these features significantly improved phenome-wide prediction of over 3,000 non-imaging phenotypes compared to standard dMRI biomarkers, such as diffusion tensor imaging, NODDI and structural connectomes. Crucially, SSDiff demonstrated robust zero-shot generalizability, accurately predicting clinical phenotypes across 7 independent cohorts (N=3,671) with different diseases, including data acquired with simplified clinical-grade protocols. A genome-wide association study on SSDiff features in 53,276 participants revealed 363 independent novel genetic loci for brain microstructure. Mendelian randomization analyses further revealed causal relationships between these features and brain disorders; for instance, linking the microstructural properties of the cerebellum to Parkinson's disease and of the hippocampus to Alzheimer's disease. SSDiff is a powerful, biologically interpretable foundation model for dMRI analysis, enabling more accurate clinical prediction and revealing novel genetic insights into brain health and disease.","manuscriptTitle":"A foundation model for efficient and assumption-free characterization of brain microstructure from diffusion MRI","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-23 02:59:37","doi":"10.21203/rs.3.rs-8877816/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-neuroscience","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"neuro","sideBox":"Learn more about [Nature Neuroscience](http://www.nature.com/neuro/)","snPcode":"","submissionUrl":"","title":"Nature Neuroscience","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"71667d0a-732b-411e-8d37-fadf9f76f13b","owner":[],"postedDate":"February 23rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":63264147,"name":"Biological sciences/Neuroscience"},{"id":63264148,"name":"Biological sciences/Biological techniques/Imaging/Magnetic resonance imaging"}],"tags":[],"updatedAt":"2026-04-22T16:05:15+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-23 02:59:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8877816","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8877816","identity":"rs-8877816","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}