Self-Supervised Model-Informed Deep Learning for Low-SNR SS-OCT Domain Transformation

Self-Supervised Model-Informed Deep Learning for Low-SNR SS-OCT Domain Transformation | 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 Self-Supervised Model-Informed Deep Learning for Low-SNR SS-OCT Domain Transformation Sajed Rakhshani, Mahnoosh Tajmirriahi, Farnaz Sedighin, Hossein Rabbani, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5730705/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 22 May, 2025 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract This article introduces a novel deep-learning based framework, Super-resolution/Denosing network (SDNet), for simultaneous denoising and super-resolution of swept-source optical coherence tomography (SS-OCT) images. The novelty of this work lies in the hybrid integration of data-driven deep-learning with a model-informed noise representation, specifically designed to address the very low signal-to-noise ratio (SNR) and low-resolution challenges in SS-OCT imaging. SDNet introduces a two-step training process, leveraging noise-free OCT references to simulate low-SNR conditions. In the first step, the network learns to enhance noisy images by combining denoising and super-resolution within noise-corrupted reference domain. To refine its performance, the second step incorporates Principle Component Analysis (PCA) as self-supervised denoising strategy, eliminating the need for ground-truth noisy image data. This unique approach enhances SDNet’s adaptability and clinical relevance. A key advantage of SDNet is its ability to balance contrast-texture by adjusting the weights of the two training steps, offering clinicians flexibility for specific dagnostic needs. Experimental results across diverse datasets demonstrate that SDNet surpasses traditional model-based and data-driven methods in computational efficiency, noise reduction, and structural fidelity. The framework excels in improving both image quality and diagnostic accuracy. Additionally, SDNet shows promising adaptability for analyzing low-resolution, low-SNR OCT images, such as those from patients with diabetic macular edema (DME). This study establishes SDNet as a robust, efficient, and clinically adaptable solution for OCT image enhancement addressing critical limitations in contemporary imaging workflows. Health sciences/Health care/Medical imaging Physical sciences/Optics and photonics Physical sciences/Engineering/Biomedical engineering Denoising Low-SNR OCT Model-aware Deep Learning Self-Supervised Learning Super-Resolution Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 22 May, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 21 Feb, 2025 Reviews received at journal 19 Feb, 2025 Reviews received at journal 06 Feb, 2025 Reviewers agreed at journal 29 Jan, 2025 Reviewers agreed at journal 22 Jan, 2025 Reviewers invited by journal 22 Jan, 2025 Editor assigned by journal 17 Jan, 2025 Editor invited by journal 07 Jan, 2025 Submission checks completed at journal 06 Jan, 2025 First submitted to journal 29 Dec, 2024 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-5730705","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":398241098,"identity":"4f24df0b-4786-43c6-b670-be8cb81040c3","order_by":0,"name":"Sajed Rakhshani","email":"","orcid":"","institution":"Isfahan University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Sajed","middleName":"","lastName":"Rakhshani","suffix":""},{"id":398241099,"identity":"5ac550ce-95e7-46d1-9927-80d719f4ec32","order_by":1,"name":"Mahnoosh Tajmirriahi","email":"","orcid":"","institution":"Isfahan University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Mahnoosh","middleName":"","lastName":"Tajmirriahi","suffix":""},{"id":398241100,"identity":"74bc2189-063d-4949-bd5c-a0ba98422ae3","order_by":2,"name":"Farnaz Sedighin","email":"","orcid":"","institution":"Isfahan University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Farnaz","middleName":"","lastName":"Sedighin","suffix":""},{"id":398241101,"identity":"8c1f0579-e1ba-46fc-8889-40fa1575231b","order_by":3,"name":"Hossein Rabbani","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1ElEQVRIiWNgGAWjYFACHjYgIcFgwN7YQKoWnoOkaWFgMJBIIFKDefvZYw9+7rBgMJd83PjgQwWDnHn/AvxaZM7kpRv2npFgsJyd2Gw44wyDscyNB/i1SDDkmEnwtgH9cjuxTZq3jSFxhsQBAlr435hJ/gVpuXmw/ffff8RokcgxkwbbcoOxjZmxAaiFv4GQlnfpxrJtEjwGZxKbJXuOSRhLSODXAXRY7rGHb9vq5AyOH3/44UeNjZwEPwGHwQAPzAggSiBOCxIg1pZRMApGwSgYMQAA4Xg+hcgpD9cAAAAASUVORK5CYII=","orcid":"","institution":"Isfahan University of Medical Sciences","correspondingAuthor":true,"prefix":"","firstName":"Hossein","middleName":"","lastName":"Rabbani","suffix":""},{"id":398241102,"identity":"5e11c3dc-1262-43c1-aa9e-b1732d3ad67b","order_by":4,"name":"Amirali Arbab","email":"","orcid":"","institution":"Isfahan University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Amirali","middleName":"","lastName":"Arbab","suffix":""},{"id":398241103,"identity":"89b620a8-fac5-4d58-ac5a-d4001dfc9597","order_by":5,"name":"Aref Habibi","email":"","orcid":"","institution":"Isfahan University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Aref","middleName":"","lastName":"Habibi","suffix":""},{"id":398241104,"identity":"ad4ad943-1a7d-4454-8825-dc657b677bf4","order_by":6,"name":"Mohsen Pourazizi","email":"","orcid":"","institution":"Isfahan Eye Research Center, Isfahan University of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Mohsen","middleName":"","lastName":"Pourazizi","suffix":""}],"badges":[],"createdAt":"2024-12-29 15:38:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5730705/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5730705/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-02375-3","type":"published","date":"2025-05-22T15:57:01+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":83459953,"identity":"4d2c8a51-3750-41d2-a395-3ffc20e2b9ad","added_by":"auto","created_at":"2025-05-26 16:05:47","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1704344,"visible":true,"origin":"","legend":"","description":"","filename":"SelfSupervisedModelInformedDeepLearningforLowSNRSSOCTDomainTransformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5730705/v1_covered_45872dc2-9bdd-44f5-967a-35344d868b5e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Self-Supervised Model-Informed Deep Learning for Low-SNR SS-OCT Domain Transformation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"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":"Denoising, Low-SNR OCT, Model-aware Deep Learning, Self-Supervised Learning, Super-Resolution","lastPublishedDoi":"10.21203/rs.3.rs-5730705/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5730705/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis article introduces a novel deep-learning based framework, Super-resolution/Denosing network (SDNet), for simultaneous denoising and super-resolution of swept-source optical coherence tomography (SS-OCT) images. The novelty of this work lies in the hybrid integration of data-driven deep-learning with a model-informed noise representation, specifically designed to address the very low signal-to-noise ratio (SNR) and low-resolution challenges in SS-OCT imaging. SDNet introduces a two-step training process, leveraging noise-free OCT references to simulate low-SNR conditions. In the first step, the network learns to enhance noisy images by combining denoising and super-resolution within noise-corrupted reference domain. To refine its performance, the second step incorporates Principle Component Analysis (PCA) as self-supervised denoising strategy, eliminating the need for ground-truth noisy image data. This unique approach enhances SDNet\u0026rsquo;s adaptability and clinical relevance. A key advantage of SDNet is its ability to balance contrast-texture by adjusting the weights of the two training steps, offering clinicians flexibility for specific dagnostic needs. Experimental results across diverse datasets demonstrate that SDNet surpasses traditional model-based and data-driven methods in computational efficiency, noise reduction, and structural fidelity. The framework excels in improving both image quality and diagnostic accuracy. Additionally, SDNet shows promising adaptability for analyzing low-resolution, low-SNR OCT images, such as those from patients with diabetic macular edema (DME). This study establishes SDNet as a robust, efficient, and clinically adaptable solution for OCT image enhancement addressing critical limitations in contemporary imaging workflows.\u003c/p\u003e","manuscriptTitle":"Self-Supervised Model-Informed Deep Learning for Low-SNR SS-OCT Domain Transformation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-08 03:23:03","doi":"10.21203/rs.3.rs-5730705/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-02-21T06:48:18+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-19T12:04:55+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-07T03:01:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"123064889323700294424313381491991740879","date":"2025-01-29T13:08:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"186964230844280679174508116515198144059","date":"2025-01-23T01:36:20+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-01-22T17:10:41+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-01-17T14:55:33+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-01-08T03:41:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-01-06T11:12:18+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-12-29T15:25:13+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":"479e4d4b-c063-4ec5-bdb9-c873d6ce8b02","owner":[],"postedDate":"January 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":42421441,"name":"Health sciences/Health care/Medical imaging"},{"id":42421442,"name":"Physical sciences/Optics and photonics"},{"id":42421443,"name":"Physical sciences/Engineering/Biomedical engineering"}],"tags":[],"updatedAt":"2025-05-26T15:59:30+00:00","versionOfRecord":{"articleIdentity":"rs-5730705","link":"https://doi.org/10.1038/s41598-025-02375-3","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-05-22 15:57:01","publishedOnDateReadable":"May 22nd, 2025"},"versionCreatedAt":"2025-01-08 03:23:03","video":"","vorDoi":"10.1038/s41598-025-02375-3","vorDoiUrl":"https://doi.org/10.1038/s41598-025-02375-3","workflowStages":[]},"version":"v1","identity":"rs-5730705","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5730705","identity":"rs-5730705","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}