{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,27]],"date-time":"2026-01-27T09:15:45Z","timestamp":1769505345501,"version":"3.49.0"},"reference-count":38,"publisher":"Optica Publishing Group","issue":"9","license":[{"start":{"date-parts":[[2024,8,2]],"date-time":"2024-08-02T00:00:00Z","timestamp":1722556800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/doi.org\/10.1364\/OA_License_v2#VOR-OA"},{"start":{"date-parts":[[2024,8,2]],"date-time":"2024-08-02T00:00:00Z","timestamp":1722556800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/opg.optica.org\/policies\/opg-tdm-policy.json"}],"funder":[{"DOI":"10.13039\/100010132","name":"\u00d6sterreichische Nationalstiftung f\u00fcr Forschung, Technologie und Entwicklung","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100010132","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100006012","name":"Christian Doppler Forschungsgesellschaft","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100006012","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["opg.optica.org"],"crossmark-restriction":false},"short-container-title":["Biomed. Opt. Express"],"published-print":{"date-parts":[[2024,9,1]]},"abstract":"<jats:p>Deep learning algorithms have allowed the automation of segmentation for many biomarkers in retinal OCTs, enabling comprehensive clinical research and precise patient monitoring. These segmentation algorithms predominantly rely on supervised training and specialised segmentation networks, such as U-Nets. However, they require segmentation annotations, which are challenging to collect and require specialized expertise. In this paper, we explore leveraging 3D self-supervised learning based on image restoration techniques, that allow to pretrain 3D networks with the aim of improving segmentation performance. We test two methods, based on image restoration and denoising. After pretraining on a large 3D OCT dataset, we evaluate our weights by fine-tuning them on two challenging fluid segmentation datasets utilising different amount of training data. The chosen methods are easy to set up while providing large improvements for fluid segmentation, enabling the reduction of the amount of required annotation or an increase in the performance. Overall, the best results were obtained for denoising-based SSL methods, with higher results on both fluid segmentation datasets as well as faster pretraining durations.<\/jats:p>","DOI":"10.1364\/boe.524603","type":"journal-article","created":{"date-parts":[[2024,7,25]],"date-time":"2024-07-25T15:16:40Z","timestamp":1721920600000},"page":"5025","update-policy":"https:\/\/doi.org\/10.1364\/crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Pretraining of 3D image segmentation models for retinal OCT using denoising-based self-supervised learning"],"prefix":"10.1364","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3705-3347","authenticated-orcid":true,"given":"Antoine","family":"Rivail","sequence":"first","affiliation":[{"id":[{"id":"https:\/\/ror.org\/05n3x4p02","id-type":"ROR","asserted-by":"publisher"}]}]},{"given":"Teresa","family":"Ara\u00fajo","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/05n3x4p02","id-type":"ROR","asserted-by":"publisher"}]}]},{"given":"Ursula","family":"Schmidt-Erfurth","sequence":"additional","affiliation":[{"name":"Medical University of Vienna"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9168-0894","authenticated-orcid":true,"given":"Hrvoje","family":"Bogunovi\u0107","sequence":"additional","affiliation":[{"id":[{"id":"https:\/\/ror.org\/05n3x4p02","id-type":"ROR","asserted-by":"publisher"}]},{"name":"Medical University of Vienna"}]}],"member":"285","published-online":{"date-parts":[[2024,8,2]]},"reference":[{"key":"boe-15-9-5025-R1","doi-asserted-by":"publisher","first-page":"759","DOI":"10.1097\/00006982-200212000-00012","volume":"22","author":"Goebel","year":"2002","journal-title":"Retina"},{"key":"boe-15-9-5025-R2","doi-asserted-by":"publisher","first-page":"566","DOI":"10.1016\/j.ajo.2007.01.028","volume":"143","author":"Fung","year":"2007","journal-title":"Am. J. Ophthalmol."},{"key":"boe-15-9-5025-R3","doi-asserted-by":"publisher","first-page":"389","DOI":"10.1016\/j.survophthal.2012.01.006","volume":"57","author":"Keane","year":"2012","journal-title":"Surv. Ophthalmol."},{"key":"boe-15-9-5025-R4","doi-asserted-by":"publisher","first-page":"3627","DOI":"10.1364\/BOE.8.003627","volume":"8","author":"Roy","year":"2017","journal-title":"Biomed. Opt. Express"},{"key":"boe-15-9-5025-R5","doi-asserted-by":"publisher","first-page":"100972","DOI":"10.1016\/j.preteyeres.2021.100972","volume":"86","author":"Schmidt-Erfurth","year":"2022","journal-title":"Prog. Retinal Eye Res."},{"key":"boe-15-9-5025-R6","doi-asserted-by":"publisher","first-page":"2119","DOI":"10.1038\/s41433-021-01487-0","volume":"35","author":"Kodjikian","year":"2021","journal-title":"Eye"},{"key":"boe-15-9-5025-R7","doi-asserted-by":"publisher","first-page":"464","DOI":"10.1007\/s10278-017-0038-7","volume":"31","author":"Khalid","year":"2018","journal-title":"J. Digit. Imaging"},{"key":"boe-15-9-5025-R8","doi-asserted-by":"publisher","first-page":"4502","DOI":"10.1002\/mp.13728","volume":"46","author":"Yu","year":"2019","journal-title":"Med. Phys."},{"key":"boe-15-9-5025-R9","doi-asserted-by":"publisher","first-page":"3793","DOI":"10.1038\/s41433-023-02615-8","volume":"37","author":"Pawloff","year":"2023","journal-title":"Eye"},{"key":"boe-15-9-5025-R10","doi-asserted-by":"publisher","first-page":"204","DOI":"10.1016\/j.visres.2017.03.008","volume":"139","author":"Gerendas","year":"2017","journal-title":"Vision Res."},{"key":"boe-15-9-5025-R11","doi-asserted-by":"publisher","author":"Ronneberger","year":"2015","DOI":"10.48550\/arXiv.1505.04597"},{"key":"boe-15-9-5025-R12","doi-asserted-by":"publisher","first-page":"1346","DOI":"10.1038\/s41551-022-00914-1","volume":"6","author":"Krishnan","year":"2022","journal-title":"Nat. Biomed. Eng."},{"key":"boe-15-9-5025-R13","doi-asserted-by":"publisher","author":"Chen","year":"2020","DOI":"10.48550\/arXiv.2002.05709"},{"key":"boe-15-9-5025-R14","doi-asserted-by":"publisher","author":"He","year":"2020","DOI":"10.48550\/arXiv.1911.05722"},{"key":"boe-15-9-5025-R15","doi-asserted-by":"publisher","author":"Bardes","year":"2022","DOI":"10.48550\/arXiv.2105.04906"},{"key":"boe-15-9-5025-R16","doi-asserted-by":"publisher","author":"Gidaris","year":"2018","DOI":"10.48550\/arXiv.1803.07728"},{"key":"boe-15-9-5025-R17","doi-asserted-by":"publisher","author":"Zhang","year":"2016","DOI":"10.48550\/arXiv.1603.08511"},{"key":"boe-15-9-5025-R18","doi-asserted-by":"publisher","author":"Dosovitskiy","year":"2015","DOI":"10.48550\/arXiv.1406.6909"},{"key":"boe-15-9-5025-R21","doi-asserted-by":"publisher","author":"Ho","year":"2020","DOI":"10.48550\/arXiv.2006.11239"},{"key":"boe-15-9-5025-R22","doi-asserted-by":"publisher","author":"Asiedu","year":"2022","DOI":"10.48550\/arXiv.2205.11423"},{"key":"boe-15-9-5025-R23","doi-asserted-by":"publisher","author":"Zhuang","year":"2019","DOI":"10.48550\/arXiv.1910.02241"},{"key":"boe-15-9-5025-R24","doi-asserted-by":"publisher","author":"He","year":"2021","DOI":"10.48550\/arXiv.2111.06377"},{"key":"boe-15-9-5025-R25","doi-asserted-by":"publisher","first-page":"156","DOI":"10.1038\/s41586-023-06555-x","volume":"622","author":"Zhou","year":"2023","journal-title":"Nature"},{"key":"boe-15-9-5025-R27","first-page":"625","article-title":"TINC: Temporally informed non-contrastive learning for disease progression modeling in retinal OCT volumes","author":"EmreWang","year":"2022"},{"key":"boe-15-9-5025-R28","doi-asserted-by":"publisher","first-page":"101539","DOI":"10.1016\/j.media.2019.101539","volume":"58","author":"Chen","year":"2019","journal-title":"Med. Image Anal."},{"key":"boe-15-9-5025-R29","doi-asserted-by":"publisher","first-page":"101840","DOI":"10.1016\/j.media.2020.101840","volume":"67","author":"Zhou","year":"2021","journal-title":"Med. Image Anal."},{"key":"boe-15-9-5025-R30","doi-asserted-by":"publisher","first-page":"1837","DOI":"10.1109\/TMI.2022.3150682","volume":"41","author":"Ouyang","year":"2022","journal-title":"IEEE Trans. Med. Imaging"},{"key":"boe-15-9-5025-R31","doi-asserted-by":"publisher","author":"Tang","year":"2022","DOI":"10.48550\/arXiv.2111.14791"},{"key":"boe-15-9-5025-R32","doi-asserted-by":"publisher","author":"Kay","year":"2017","DOI":"10.48550\/arXiv.1705.06950"},{"key":"boe-15-9-5025-R33","doi-asserted-by":"publisher","first-page":"989","DOI":"10.1109\/TBME.2017.2734058","volume":"65","author":"Rashno","year":"2018","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"boe-15-9-5025-R34","doi-asserted-by":"publisher","first-page":"1436","DOI":"10.1109\/TMI.2009.2016958","volume":"28","author":"Garvin","year":"2009","journal-title":"IEEE Trans. Med. Imaging"},{"key":"boe-15-9-5025-R35","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1109\/TPAMI.2006.19","volume":"28","author":"Li","year":"2006","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"boe-15-9-5025-R36","doi-asserted-by":"publisher","first-page":"169","DOI":"10.1109\/RBME.2010.2084567","volume":"3","author":"Abrmoff","year":"2010","journal-title":"IEEE Rev. Biomed. Eng."},{"key":"boe-15-9-5025-R37","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1109\/JBHI.2022.3217962","volume":"27","author":"Fazekas","year":"2023","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"boe-15-9-5025-R38","doi-asserted-by":"publisher","author":"\u00c7i\u00e7ek","year":"2016","DOI":"10.48550\/arXiv.1606.06650"},{"key":"boe-15-9-5025-R39","doi-asserted-by":"publisher","author":"Feichtenhofer","year":"2019","DOI":"10.48550\/arXiv.1812.03982"},{"key":"boe-15-9-5025-R40","doi-asserted-by":"publisher","author":"Romano","year":"2017","DOI":"10.48550\/arXiv.1611.02862"},{"key":"boe-15-9-5025-R41","doi-asserted-by":"publisher","author":"Isensee","year":"2024","DOI":"10.48550\/arXiv.2404.09556"}],"container-title":["Biomedical Optics Express"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/opg.optica.org\/viewmedia.cfm?URI=boe-15-9-5025&seq=0","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,6]],"date-time":"2024-08-06T19:31:33Z","timestamp":1722972693000},"score":1,"resource":{"primary":{"URL":"https:\/\/opg.optica.org\/abstract.cfm?URI=boe-15-9-5025"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,8,2]]},"references-count":38,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2024]]},"published-print":{"date-parts":[[2024]]}},"URL":"https:\/\/doi.org\/10.1364\/boe.524603","relation":{},"ISSN":["2156-7085","2156-7085"],"issn-type":[{"value":"2156-7085","type":"print"},{"value":"2156-7085","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,8,2]]},"assertion":[{"value":"Optica Publishing Group","name":"publisher","label":"This article is maintained by"},{"value":"https:\/\/doi.org\/10.1364\/BOE.524603","name":"articlelink","label":"Crossref DOI link to publisher maintained version"},{"value":"research-article","name":"content_type","label":"Article type"},{"value":"Screened by Similarity Check","name":"cross_check","label":"Similarity check"},{"value":"Yes","order":0,"name":"peer_reviewed","label":"Peer reviewed","group":{"name":"peer_review","label":"Peer review"}},{"value":"Single blind","order":1,"name":"review_process","label":"Review process","group":{"name":"peer_review","label":"Peer review"}},{"value":"25 March 2024","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"19 July 2024","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2 August 2024","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}},{"value":"\u00a9 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement","name":"copyright","label":"Copyright"}]}}