DM-Net: a physics-model-independent direct mapping approach for calibration-free multi-coil MRI

DM-Net: a physics-model-independent direct mapping approach for calibration-free multi-coil 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 DM-Net: a physics-model-independent direct mapping approach for calibration-free multi-coil MRI Yan Wu, Cagan Alkan, Julio Oscanoa, Aiqi Sun, Kawin Setsompop, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7174070/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 Deep learning-based multi-coil magnetic resonance image reconstruction has been actively investigated and increasingly applied in clinical settings. However, most models are physics-model-based approaches, which either require precalculation of coil sensitivity or fail to achieve optimal performance without taking coil sensitivity into consideration. Inspired by ESPIRiT, we propose a physics-model-independent direct mapping approach, namely DM-Net, which aims for optimal reconstruction without explicitly using coil sensitivity. A densely connected convolutional network is employed, where coil sensitivity and channel correlation are intrinsically exploited. For comparison, other reconstruction models that mimic SENSE and GRAPPA are implemented. DMwS-Net is a direct mapping approach that incorporates precalculated coil sensitivity maps as additional input; and CCR-Nets provide coil-by-coil reconstruction, where individual coil images are jointly predicted and combined with or without coil sensitivity. The proposed models are trained and tested on 5440 images from 17 subjects acquired with 3DFT (Fourier transform). DM-Net achieves superior performance, suggesting the feasibility of excluding precalculated coil sensitivity from model input. Moreover, DM-Net can be applied on k-space data without a fully sampled calibration region, which may improve image quality by sampling high frequency regions more densely. This work expands the scope of DL-based image reconstruction and provides evidence that a direct mapping framework without using coil sensitivity may lead to reliable, simplified and faster reconstruction. Physical sciences/Engineering Physical sciences/Mathematics and computing Physical sciences/Physics Full Text Additional Declarations Competing interest reported. Two patents were filed. (1) Deep learning-based physics-model-independent direct mapping and coil-by-coil reconstruction techniques by Shreyas Vasanawala, Yan Wu, and John Pauly. (2) Deep Learning-based Calibrationless MRI by Yan Wu, John Pauly, Shreyas Vasanawala, and Kawin Setsompop. All the remaining authors declare no conflict of interest. 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-7174070","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":505653072,"identity":"a5c38d6e-7382-4675-a7bc-add3d62151a2","order_by":0,"name":"Yan Wu","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Yan","middleName":"","lastName":"Wu","suffix":""},{"id":505653073,"identity":"fa90c04e-5b37-4385-866b-e345e36c83af","order_by":1,"name":"Cagan Alkan","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Cagan","middleName":"","lastName":"Alkan","suffix":""},{"id":505653074,"identity":"dcbb57cc-30d3-4dd5-896d-c962db3f66c8","order_by":2,"name":"Julio Oscanoa","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Julio","middleName":"","lastName":"Oscanoa","suffix":""},{"id":505653075,"identity":"cacdf062-d1d5-42b5-a6ae-368b235beb59","order_by":3,"name":"Aiqi Sun","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Aiqi","middleName":"","lastName":"Sun","suffix":""},{"id":505653076,"identity":"2a3c01ec-4123-4366-aac9-ea4cce64be82","order_by":4,"name":"Kawin Setsompop","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Kawin","middleName":"","lastName":"Setsompop","suffix":""},{"id":505653077,"identity":"9da3e6f3-bd4e-4cbf-a69c-e50880bfb644","order_by":5,"name":"Ali Syed","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Ali","middleName":"","lastName":"Syed","suffix":""},{"id":505653078,"identity":"c0665975-a156-4a33-8cfc-68f7b7e8a497","order_by":6,"name":"Yajun Ma","email":"","orcid":"","institution":"University of California, San Diego","correspondingAuthor":false,"prefix":"","firstName":"Yajun","middleName":"","lastName":"Ma","suffix":""},{"id":505653079,"identity":"63da9e28-bcfd-4c05-b965-8cb099740d64","order_by":7,"name":"Congyu Liao","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Congyu","middleName":"","lastName":"Liao","suffix":""},{"id":505653080,"identity":"c0375cc7-b36d-40e4-bfe7-bd2fc4c19bbd","order_by":8,"name":"Marcus Alley","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Marcus","middleName":"","lastName":"Alley","suffix":""},{"id":505653081,"identity":"c4198dda-c80e-449d-8ba0-02f6b295377f","order_by":9,"name":"Fan Zhang","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Fan","middleName":"","lastName":"Zhang","suffix":""},{"id":505653082,"identity":"12d3e697-aaf0-460b-9200-dba0abb013ff","order_by":10,"name":"John Pauly","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"","lastName":"Pauly","suffix":""},{"id":505653083,"identity":"422442df-04b0-4f65-a8fb-1ffdeb555fe6","order_by":11,"name":"Shreyas Vasanawala","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA90lEQVRIie3LvYrCQBSG4RMCYzNx2hNSeAuzCGlEvBVF0CZCKrEQsZpKYutlrI31yIA20RvIFisBbSwCARFk/dnFSkzQbot54XCa7wHQ6f5zRTDl72e34wDkBUKA1OGG7OEbhPLXCF8uN6nfVyVSCNM06X8hc9QnJF2VTUKv7EwW6kPQzgzlYot20PKNyTqb2EMPHEoiQ0BnBpKoAQ8pNy2RQ8a7+ETPUU2wfZzIs8I/8pNDGNZdxxJRQ6AHOBd3YuSSvVuxgktT4NbFVaDQHhF/Plq3Mwlh7Tiih1Z1zJpx0jsoZNScfh+7lUzyPPnmXqfT6XQPXQGtXFcge1+3XAAAAABJRU5ErkJggg==","orcid":"","institution":"Stanford University","correspondingAuthor":true,"prefix":"","firstName":"Shreyas","middleName":"","lastName":"Vasanawala","suffix":""}],"badges":[],"createdAt":"2025-07-21 06:53:50","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7174070/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7174070/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95799333,"identity":"283f49cd-6e25-4acc-bf40-ec6059be9d03","added_by":"auto","created_at":"2025-11-13 08:19:37","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1511667,"visible":true,"origin":"","legend":"","description":"","filename":"multicoilimagereconstruction.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7174070/v1_covered_b56390c5-7591-4c29-9e74-40c77f520b46.pdf"}],"financialInterests":"Competing interest reported. Two patents were filed.\n(1) Deep learning-based physics-model-independent direct mapping and coil-by-coil reconstruction techniques by Shreyas Vasanawala, Yan Wu, and John Pauly.\n(2) Deep Learning-based Calibrationless MRI by Yan Wu, John Pauly, Shreyas Vasanawala, and Kawin Setsompop.\nAll the remaining authors declare no conflict of interest.","formattedTitle":"DM-Net: a physics-model-independent direct mapping approach for calibration-free multi-coil MRI","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-7174070/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7174070/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDeep learning-based multi-coil magnetic resonance image reconstruction has been actively investigated and increasingly applied in clinical settings. However, most models are physics-model-based approaches, which either require precalculation of coil sensitivity or fail to achieve optimal performance without taking coil sensitivity into consideration.\u003c/p\u003e\u003cp\u003eInspired by ESPIRiT, we propose a physics-model-independent direct mapping approach, namely DM-Net, which aims for optimal reconstruction without explicitly using coil sensitivity. A densely connected convolutional network is employed, where coil sensitivity and channel correlation are intrinsically exploited.\u003c/p\u003e\u003cp\u003eFor comparison, other reconstruction models that mimic SENSE and GRAPPA are implemented. DMwS-Net is a direct mapping approach that incorporates precalculated coil sensitivity maps as additional input; and CCR-Nets provide coil-by-coil reconstruction, where individual coil images are jointly predicted and combined with or without coil sensitivity.\u003c/p\u003e\u003cp\u003eThe proposed models are trained and tested on 5440 images from 17 subjects acquired with 3DFT (Fourier transform). DM-Net achieves superior performance, suggesting the feasibility of excluding precalculated coil sensitivity from model input. Moreover, DM-Net can be applied on k-space data without a fully sampled calibration region, which may improve image quality by sampling high frequency regions more densely.\u003c/p\u003e\u003cp\u003eThis work expands the scope of DL-based image reconstruction and provides evidence that a direct mapping framework without using coil sensitivity may lead to reliable, simplified and faster reconstruction.\u003c/p\u003e","manuscriptTitle":"DM-Net: a physics-model-independent direct mapping approach for calibration-free multi-coil MRI","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-01 17:47:32","doi":"10.21203/rs.3.rs-7174070/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":"456a428e-7cd9-4fc8-84ab-e7f6b96c1ae4","owner":[],"postedDate":"September 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":53703333,"name":"Physical sciences/Engineering"},{"id":53703334,"name":"Physical sciences/Mathematics and computing"},{"id":53703335,"name":"Physical sciences/Physics"}],"tags":[],"updatedAt":"2025-11-11T18:23:30+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-01 17:47:32","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7174070","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7174070","identity":"rs-7174070","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}