Time-lapse Image Super-resolution Neural Network with Reliable Confidence Quantification for Optical Microscopy

Time-lapse Image Super-resolution Neural Network with Reliable Confidence Quantification for Optical Microscopy | 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 Time-lapse Image Super-resolution Neural Network with Reliable Confidence Quantification for Optical Microscopy Qionghai Dai, Chang Qiao, Shuran Liu, Yuwang Wang, Wencong Xu, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4618775/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 29 Jan, 2025 Read the published version in Nature Biotechnology → Version 1 posted You are reading this latest preprint version Abstract Single image super-resolution (SISR) neural networks for optical microscopy have shown great capability to directly transform a low-resolution (LR) image into its super-resolution (SR) counterpart, enabling low-cost long-term live-cell SR imaging. However, when processing time-lapse data, current SISR models failed to exploit the important temporal dependencies between neighbor frames, often resulting in temporally inconsistent outputs. Besides, SISR models are subject to inference uncertainty that is hard to accurately quantify, therefore it is difficult to determine to what extend can we trust the inferred SR images. Here, we first build a large-scale, high-quality fluorescence microscopy dataset for the time-lapse image super-resolution (TISR) task, and conducted a comprehensive evaluation on two essential components of TISR neural networks, i.e., propagation and alignment. Second, we devised a deformable phase-space alignment (DPA) based TISR neural network (DPA-TISR), which adaptively enhances the cross-frame alignment in the phase domain and outperforms existing state-of-the-art SISR and TISR models. Third, we combined the Bayesian training scheme and Monte Carlo dropout with DPA-TISR, developing Bayesian DPA-TISR, and designed an expected calibration error (ECE) minimization framework to obtain a well-calibrated confidence map along with each output SR image, which reliably implicates potential inference errors. We demonstrate the unique characteristics of Bayesian DPA-TISR underlie the ultralong-term live-cell SR imaging capability with high spatial fidelity, superb temporal consistency, and accurate confidence quantification on a wide variety of bioprocesses. Physical sciences/Optics and photonics/Optical techniques/Microscopy/Super-resolution microscopy Biological sciences/Biological techniques/Imaging/Fluorescence imaging Biological sciences/Biological techniques/Microscopy/Super-resolution microscopy Full Text Additional Declarations Yes there is potential Competing Interest. Q.D., D.L., C.Q. and S.L. have two pending patent application on the presented frameworks. Supplementary Files SuppVideo1FactinCCPs.mp4 Supplementary Video 1 SuppVideo2PKMOTFAM.mp4 Supplementary Video 2 SuppVideo3Tomm20Lyso.mp4 Supplementary Video 3 SuppVideo4Tomm20PMP.mp4 Supplementary Video 4 SuppVideo5Tomm20PMPCases.mp4 Supplementary Video 5 SuppMat.pdf Cite Share Download PDF Status: Published Journal Publication published 29 Jan, 2025 Read the published version in Nature Biotechnology → 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-4618775","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":319797177,"identity":"4686a62e-059a-4895-ba33-93ebf4e6bffd","order_by":0,"name":"Qionghai Dai","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6klEQVRIiWNgGAWjYDACCQYDhoQKBgY2MAKBA0RpOWNAqhbGNgMQk0gtfLebN354OO+PPZ90+7XHPDUMcnw3Ehg/F+DRInnnWLFE4jaDxDaZM+XGPMcYjCVvJDBLz8CjxeBGjgFISwKbRE6aNA8bQ+KGGwlszDz4tRj/SJxjYA/R8o+hnhgtZhKJDQaMbRLpx6R52xgSDAhpkbyRVmaRcMw4sU0ih01ybp+E4cwzD5ul8Wnhu5G8+eaPGjl7+RnpzyTefLOR5zuefPAzPi1IscADihwJIGZswKcBWQv7A/wqR8EoGAWjYMQCAFNISdcRyCx6AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-7043-3061","institution":"Tsinghua University","correspondingAuthor":true,"prefix":"","firstName":"Qionghai","middleName":"","lastName":"Dai","suffix":""},{"id":319797178,"identity":"869462e0-ac22-4c85-b844-3073354a2904","order_by":1,"name":"Chang Qiao","email":"","orcid":"https://orcid.org/0000-0002-6037-0842","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Chang","middleName":"","lastName":"Qiao","suffix":""},{"id":319797179,"identity":"555a34f9-3ea1-4f5b-9b25-199cb8f9dd05","order_by":2,"name":"Shuran Liu","email":"","orcid":"","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Shuran","middleName":"","lastName":"Liu","suffix":""},{"id":319797180,"identity":"2a9ab2fe-703c-4faa-9968-2e994e382277","order_by":3,"name":"Yuwang Wang","email":"","orcid":"https://orcid.org/0000-0002-6880-959X","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Yuwang","middleName":"","lastName":"Wang","suffix":""},{"id":319797181,"identity":"0ddf4a53-8037-439e-b4bc-eef44ac1ee2b","order_by":4,"name":"Wencong Xu","email":"","orcid":"","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Wencong","middleName":"","lastName":"Xu","suffix":""},{"id":319797182,"identity":"d3f34a4e-29de-40aa-8e67-97c19c1c6223","order_by":5,"name":"Xiaohan Geng","email":"","orcid":"","institution":"Institute of Biophysics, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xiaohan","middleName":"","lastName":"Geng","suffix":""},{"id":319797183,"identity":"ff3e0bf8-a71d-4a43-a01a-994bf1d2fa4e","order_by":6,"name":"Tao Jiang","email":"","orcid":"","institution":"Institute of Biophysics, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Tao","middleName":"","lastName":"Jiang","suffix":""},{"id":319797184,"identity":"d8ff6754-b174-493b-b673-dd019802989f","order_by":7,"name":"Jingyu Zhang","email":"","orcid":"","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Jingyu","middleName":"","lastName":"Zhang","suffix":""},{"id":319797185,"identity":"2712def4-f2a8-45b9-b78d-a0ed5b76b5af","order_by":8,"name":"Quan Meng","email":"","orcid":"","institution":"Institute of Biophysics, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Quan","middleName":"","lastName":"Meng","suffix":""},{"id":319797186,"identity":"60c35994-a7a3-4fbf-9e2c-5b433c1884ea","order_by":9,"name":"Hui Qiao","email":"","orcid":"https://orcid.org/0000-0002-4896-8657","institution":"Tsinghua University","correspondingAuthor":false,"prefix":"","firstName":"Hui","middleName":"","lastName":"Qiao","suffix":""},{"id":319797187,"identity":"c1c0a9cb-a7af-44f3-8243-35b41f12e0cd","order_by":10,"name":"Dong Li","email":"","orcid":"https://orcid.org/0000-0001-6787-5125","institution":"Institute of Biophysics","correspondingAuthor":false,"prefix":"","firstName":"Dong","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2024-06-21 17:40:03","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4618775/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4618775/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41587-025-02553-8","type":"published","date":"2025-01-29T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":75076949,"identity":"c23e7a5b-8407-41a1-bf00-9b1d02eec501","added_by":"auto","created_at":"2025-01-30 08:07:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":16860471,"visible":true,"origin":"","legend":"","description":"","filename":"MainText.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4618775/v1_covered_7a18cf6e-d057-4719-8c4e-7a74cd0f58d1.pdf"},{"id":59236705,"identity":"6ff6a731-e49d-4a56-bc3c-cc2e3e43be01","added_by":"auto","created_at":"2024-06-28 04:10:29","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":40040871,"visible":true,"origin":"","legend":"Supplementary Video 1","description":"","filename":"SuppVideo1FactinCCPs.mp4","url":"https://assets-eu.researchsquare.com/files/rs-4618775/v1/bdffda67be794abf9363bd09.mp4"},{"id":59236293,"identity":"adad7270-4ae1-4253-a8a6-b0b0d914abaa","added_by":"auto","created_at":"2024-06-28 04:02:29","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":24016979,"visible":true,"origin":"","legend":"Supplementary Video 2","description":"","filename":"SuppVideo2PKMOTFAM.mp4","url":"https://assets-eu.researchsquare.com/files/rs-4618775/v1/79ac7cbb95ecc36fbb9b3d2e.mp4"},{"id":59236295,"identity":"5ecc390d-710b-4b25-bb0d-c1dc7059a7ff","added_by":"auto","created_at":"2024-06-28 04:02:29","extension":"mp4","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":32008784,"visible":true,"origin":"","legend":"Supplementary Video 3","description":"","filename":"SuppVideo3Tomm20Lyso.mp4","url":"https://assets-eu.researchsquare.com/files/rs-4618775/v1/1c40a0baaac5b88db38aa918.mp4"},{"id":59236291,"identity":"0ed21b03-f927-43d4-a157-064320b60154","added_by":"auto","created_at":"2024-06-28 04:02:28","extension":"mp4","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":28759328,"visible":true,"origin":"","legend":"Supplementary Video 4","description":"","filename":"SuppVideo4Tomm20PMP.mp4","url":"https://assets-eu.researchsquare.com/files/rs-4618775/v1/aa2a9398ab1440bd9840b8d2.mp4"},{"id":59236292,"identity":"31bb9b69-7f2e-432b-ad6f-5ad28376ff15","added_by":"auto","created_at":"2024-06-28 04:02:28","extension":"mp4","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":22958374,"visible":true,"origin":"","legend":"Supplementary Video 5","description":"","filename":"SuppVideo5Tomm20PMPCases.mp4","url":"https://assets-eu.researchsquare.com/files/rs-4618775/v1/d4fb1cee6c46252ed19cd8c0.mp4"},{"id":59236290,"identity":"fd07ccd6-3923-4d09-adf8-4e6901eeab4e","added_by":"auto","created_at":"2024-06-28 04:02:28","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":6985637,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"SuppMat.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4618775/v1/41bcf355c43032bd80bd58ea.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nQ.D., D.L., C.Q. and S.L. have two pending patent application on the presented frameworks.","formattedTitle":"Time-lapse Image Super-resolution Neural Network with Reliable\r\nConfidence Quantification for Optical Microscopy","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"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-4618775/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4618775/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Single image super-resolution (SISR) neural networks for optical microscopy have shown great capability to directly transform a low-resolution (LR) image into its super-resolution (SR) counterpart, enabling low-cost long-term live-cell SR imaging. However, when processing time-lapse data, current SISR models failed to exploit the important temporal dependencies between neighbor frames, often resulting in temporally inconsistent outputs. Besides, SISR models are subject to inference uncertainty that is hard to accurately quantify, therefore it is difficult to determine to what extend can we trust the inferred SR images. Here, we first build a large-scale, high-quality fluorescence microscopy dataset for the time-lapse image super-resolution (TISR) task, and conducted a comprehensive evaluation on two essential components of TISR neural networks, i.e., propagation and alignment. Second, we devised a deformable phase-space alignment (DPA) based TISR neural network (DPA-TISR), which adaptively enhances the cross-frame alignment in the phase domain and outperforms existing state-of-the-art SISR and TISR models. Third, we combined the Bayesian training scheme and Monte Carlo dropout with DPA-TISR, developing Bayesian DPA-TISR, and designed an expected calibration error (ECE) minimization framework to obtain a well-calibrated confidence map along with each output SR image, which reliably implicates potential inference errors. We demonstrate the unique characteristics of Bayesian DPA-TISR underlie the ultralong-term live-cell SR imaging capability with high spatial fidelity, superb temporal consistency, and accurate confidence quantification on a wide variety of bioprocesses.","manuscriptTitle":"Time-lapse Image Super-resolution Neural Network with Reliable\nConfidence Quantification for Optical Microscopy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-06-28 04:02:23","doi":"10.21203/rs.3.rs-4618775/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-biotechnology","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nbt","sideBox":"Learn more about [Nature Biotechnology](http://www.nature.com/nbt/)","snPcode":"","submissionUrl":"","title":"Nature Biotechnology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"1e434f6c-5b3d-4511-b97b-55f29f5099e1","owner":[],"postedDate":"June 28th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":33814032,"name":"Physical sciences/Optics and photonics/Optical techniques/Microscopy/Super-resolution microscopy"},{"id":33814033,"name":"Biological sciences/Biological techniques/Imaging/Fluorescence imaging"},{"id":33814034,"name":"Biological sciences/Biological techniques/Microscopy/Super-resolution microscopy"}],"tags":[],"updatedAt":"2025-01-30T08:07:09+00:00","versionOfRecord":{"articleIdentity":"rs-4618775","link":"https://doi.org/10.1038/s41587-025-02553-8","journal":{"identity":"nature-biotechnology","isVorOnly":false,"title":"Nature Biotechnology"},"publishedOn":"2025-01-29 05:00:00","publishedOnDateReadable":"January 29th, 2025"},"versionCreatedAt":"2024-06-28 04:02:23","video":"","vorDoi":"10.1038/s41587-025-02553-8","vorDoiUrl":"https://doi.org/10.1038/s41587-025-02553-8","workflowStages":[]},"version":"v1","identity":"rs-4618775","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4618775","identity":"rs-4618775","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}