Neural Steered Mixture of Experts for Medical Image Denoising, and Super-Resolution

Neural Steered Mixture of Experts for Medical Image Denoising, and Super-Resolution | 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 Research Article Neural Steered Mixture of Experts for Medical Image Denoising, and Super-Resolution Aytac Ozkan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7667284/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract In medical imaging (MI) analysis, achieving high-fidelity spatial resolution remains challenging due to extended acquisition durations and limited frame rates, constraining the quality and diagnostic value of clinical data. Super-resolution (SR) methodologies reconstruct high-resolution (HR) representations from low-resolution (LR) input, mitigating hardware and temporal constraints while enhancing interpretability and diagnostic reliability. Current single-image super-resolution (SISR) paradigms often rely on oversimplified noise assumptions, modeled as Additive White Gaussian Noise (AWGN), which fail to capture the complex and modality-dependent noise distributions inherent to clinical imaging scenarios. These limitations require more sophisticated SR frameworks capable of accurately representing non-stationary degradations and ensuring robust performance across imaging modalities. We introduce a neural parametric Steered Mixture of Experts (N-SMoE) framework that leverages a generative adversarial-based training paradigm and a Stochastic Degradation Model (SDM), applying diverse perturbations to downsampled inputs to approximate clinical conditions. This framework combines a novel encoder network with an implicit probabilistic SMoE decoder. The encoder utilizes a Laplacian resizer with bandpass filtering to capture local spatial information and employs multi-head attention to preserve high-frequency (HF) structural patterns while estimating latent representations of the input image. The imsplicit probabilistic gating mechanisms of the SMoE decoder, using two-dimensional edge-aware kernels, represent the signal of interest with continuous transitions, making this autoregressive approach more robust and effective for SR and denoising. The proposed N-SMoE framework not only provides interpretability for the learned representations but also demonstrates state-of-the-art (SOTA) performance in restoration tasks across multiple medical imaging datasets, achieving improved fidelity and perceptual metrics. Mixture of Experts Noise reduction Super-resolution Deep learning Full Text Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 16 Apr, 2026 Reviewers invited by journal 01 Mar, 2026 Editor assigned by journal 22 Sep, 2025 First submitted to journal 21 Sep, 2025 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-7667284","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":598988539,"identity":"f30d98fb-1146-499f-99ae-101be089a752","order_by":0,"name":"Aytac Ozkan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEUlEQVRIiWNgGAWjYFADHiD+AMQGDAyMBxKARANOpcwILYwzEsBaGIjXwswD08KARwv/jPyDH34w2MjJ9xx+9tj2x2EGc/beBwce7mCQ7cehReJGMrNkD0OaMWNvm7lxTsJhBsue4wYHEs8wGM/EZc2NZAZpBobDic38DGbSQC31G26kMRxIbGNI3HAAuw55oC2/gVrq2/jZv0lbAG0xuP8MomU/Di0GN5LZQLYk8PD2mEkzgLTcYIPagsNdhmcem1n2GKQZzuA5UybZk5YO9AvYYRLGM3DYInc88fGNHxU28vI96dskfthYA0PsGOPDn202sv24vC+QwACJCzQggUM9EPDjsH4UjIJRMApGARwAAB2fWYClHO/UAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-8188-7824","institution":"Technical University of Berlin: Technische Universitat Berlin","correspondingAuthor":true,"prefix":"","firstName":"Aytac","middleName":"","lastName":"Ozkan","suffix":""}],"badges":[],"createdAt":"2025-09-21 11:02:00","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7667284/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7667284/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104402707,"identity":"2cd73aed-c3cf-4849-ad29-ba5cb084e252","added_by":"auto","created_at":"2026-03-11 12:16:10","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":38727160,"visible":true,"origin":"","legend":"","description":"","filename":"JIVPD2500211.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7667284/v1_covered_e2004ffd-622d-4523-b535-4ada1f793845.pdf"}],"financialInterests":"","formattedTitle":"Neural Steered Mixture of Experts for Medical Image Denoising, and Super-Resolution","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":"eurasip-journal-on-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jivp","sideBox":"Learn more about [EURASIP Journal on Image and Video Processing](http://jivp-eurasipjournals.springeropen.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/jivp/default.aspx","title":"EURASIP Journal on Image and Video Processing","twitterHandle":"@SpringerEng","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Mixture of Experts, Noise reduction, Super-resolution, Deep learning","lastPublishedDoi":"10.21203/rs.3.rs-7667284/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7667284/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e In medical imaging (MI) analysis, achieving high-fidelity spatial resolution remains challenging due to extended acquisition durations and limited frame rates, constraining the quality and diagnostic value of clinical data. Super-resolution (SR) methodologies reconstruct high-resolution (HR) representations from low-resolution (LR) input, mitigating hardware and temporal constraints while enhancing interpretability and diagnostic reliability. Current single-image super-resolution (SISR) paradigms often rely on oversimplified noise assumptions, modeled as Additive White Gaussian Noise (AWGN), which fail to capture the complex and modality-dependent noise distributions inherent to clinical imaging scenarios. These limitations require more sophisticated SR frameworks capable of accurately representing non-stationary degradations and ensuring robust performance across imaging modalities. We introduce a neural parametric Steered Mixture of Experts (N-SMoE) framework that leverages a generative adversarial-based training paradigm and a Stochastic Degradation Model (SDM), applying diverse perturbations to downsampled inputs to approximate clinical conditions. This framework combines a novel encoder network with an implicit probabilistic SMoE decoder. The encoder utilizes a Laplacian resizer with bandpass filtering to capture local spatial information and employs multi-head attention to preserve high-frequency (HF) structural patterns while estimating latent representations of the input image. The imsplicit probabilistic gating mechanisms of the SMoE decoder, using two-dimensional edge-aware kernels, represent the signal of interest with continuous transitions, making this autoregressive approach more robust and effective for SR and denoising. The proposed N-SMoE framework not only provides interpretability for the learned representations but also demonstrates state-of-the-art (SOTA) performance in restoration tasks across multiple medical imaging datasets, achieving improved fidelity and perceptual metrics.\u003c/p\u003e","manuscriptTitle":"Neural Steered Mixture of Experts for Medical Image Denoising, and Super-Resolution","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-04 17:39:29","doi":"10.21203/rs.3.rs-7667284/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2026-04-16T06:55:34+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-02T04:08:05+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-22T23:21:11+00:00","index":"","fulltext":""},{"type":"submitted","content":"EURASIP Journal on Image and Video Processing","date":"2025-09-22T01:39:20+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"eurasip-journal-on-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jivp","sideBox":"Learn more about [EURASIP Journal on Image and Video Processing](http://jivp-eurasipjournals.springeropen.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/jivp/default.aspx","title":"EURASIP Journal on Image and Video Processing","twitterHandle":"@SpringerEng","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"51fdeebf-1549-4704-a079-9986fe9f3a27","owner":[],"postedDate":"March 4th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-04T17:39:30+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-04 17:39:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7667284","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7667284","identity":"rs-7667284","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}