{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,27]],"date-time":"2026-01-27T23:02:46Z","timestamp":1769554966350,"version":"3.49.0"},"reference-count":26,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2021,5,15]],"date-time":"2021-05-15T00:00:00Z","timestamp":1621036800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/501100005860","name":"Helsefonden","doi-asserted-by":"publisher","award":["20\u2010B\u20100099"],"award-info":[{"award-number":["20\u2010B\u20100099"]}],"id":[{"id":"10.13039\/501100005860","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Acta Ophthalmologica"],"published-print":{"date-parts":[[2022,2]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Purpose<\/jats:title><jats:p>To meet the demands imposed by the continuing growth of the Age\u2010related macular degeneration (AMD) patient population, automation of follow\u2010ups by detecting retinal oedema using deep learning might be a viable approach. However, preparing and labelling data for training is time consuming. In this study, we investigate the feasibility of training a convolutional neural network (CNN) to accurately detect retinal oedema on optical coherence tomography (OCT) images of AMD patients with labels derived directly from clinical treatment decisions, without extensive preprocessing or relabelling.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>A total of 50\u2009439 OCT images with associated treatment information were retrieved from databases at the Department of Ophthalmology, Rigshospitalet, Copenhagen, Denmark between 01.06.2007 and 01.06.2018. A CNN was trained on the retrieved data with the recorded treatment decisions as labels and validated on a subset of the data relabelled by three ophthalmologists to denote presence of oedema.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Moderate inter\u2010grader agreement on presence of oedema in the relabelled data was found (76.4%). Despite different training and validation labels, the CNN performed on par with inter\u2010grader agreement in detecting oedema on OCT images (AUC 0.97, accuracy 90.9%) and previously published models based on relabelled datasets.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>The level of performance shown by the current model might make it valuable in detecting disease activity in automated AMD patient follow\u2010up systems. Our approach demonstrates that high accuracy is not necessarily constrained by incongruent training and validation labels. These results might encourage the use of existing clinical databases for development of deep learning based algorithms without labour\u2010intensive preprocessing in the future.<\/jats:p><\/jats:sec>","DOI":"10.1111\/aos.14895","type":"journal-article","created":{"date-parts":[[2021,5,15]],"date-time":"2021-05-15T09:25:52Z","timestamp":1621070752000},"page":"103-110","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":10,"title":["Detection of oedema on optical coherence tomography images using deep learning model trained on noisy clinical data"],"prefix":"10.1111","volume":"100","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7201-655X","authenticated-orcid":false,"given":"Ivan","family":"Potapenko","sequence":"first","affiliation":[{"name":"Department of Ophthalmology Rigshospitalet Copenhagen Denmark"},{"name":"Faculty of Health and Medical Sciences University of Copenhagen Copenhagen Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mads","family":"Kristensen","sequence":"additional","affiliation":[{"name":"Enversion A\/S Aarhus Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bo","family":"Thiesson","sequence":"additional","affiliation":[{"name":"Enversion A\/S Aarhus Denmark"},{"name":"Department of Engineering Aarhus University Aarhus Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tomas","family":"Ilginis","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology Rigshospitalet Copenhagen Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6790-0199","authenticated-orcid":false,"given":"Torben","family":"Lykke S\u00f8rensen","sequence":"additional","affiliation":[{"name":"Faculty of Health and Medical Sciences University of Copenhagen Copenhagen Denmark"},{"name":"Department of Ophthalmology Zealand University Hospital Roskilde Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Javad","family":"Nouri Hajari","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology Rigshospitalet Copenhagen Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Josefine","family":"Fuchs","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology Rigshospitalet Copenhagen Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4318-2716","authenticated-orcid":false,"given":"Steffen","family":"Hamann","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology Rigshospitalet Copenhagen Denmark"},{"name":"Faculty of Health and Medical Sciences University of Copenhagen Copenhagen Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7712-9772","authenticated-orcid":false,"given":"Morten","family":"la Cour","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology Rigshospitalet Copenhagen Denmark"},{"name":"Faculty of Health and Medical Sciences University of Copenhagen Copenhagen Denmark"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2021,5,15]]},"reference":[{"key":"e_1_2_7_2_1","doi-asserted-by":"publisher","DOI":"10.1097\/ICU.0b013e32835f8bf8"},{"key":"e_1_2_7_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.retram.2020.01.002"},{"key":"e_1_2_7_4_1","doi-asserted-by":"publisher","DOI":"10.1097\/IAE.0000000000000729"},{"key":"e_1_2_7_5_1","doi-asserted-by":"publisher","DOI":"10.1111\/aos.14183"},{"key":"e_1_2_7_6_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41433-020-0799-y"},{"key":"e_1_2_7_7_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0107-6"},{"key":"e_1_2_7_8_1","unstructured":"GlorotX&BengioY(2010):Understanding the difficulty of training deep feedforward neural networks. Proc 13th Int Conf Artif Intell Stat PMLR 9:249\u2013256 2010."},{"key":"e_1_2_7_9_1","unstructured":"HeK ZhangX RenS&SunJ(2015):Delving deep into rectifiers: Surpassing human\u2010level performance on imagenet classification. Proc IEEE Int Conf Comput Vis. Institute of Electrical and Electronics Engineers Inc.1026\u20131034.https:\/\/arxiv.org\/abs\/1502.01852v1. (Accessed on 04 Nov 2020)."},{"key":"e_1_2_7_10_1","doi-asserted-by":"publisher","DOI":"10.1111\/opo.12675"},{"key":"e_1_2_7_11_1","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0126557","article-title":"Interobserver agreement in detecting spectral\u2010domain optical coherence tomography features of diabetic macular edema","volume":"10","author":"Heng LZ","year":"2015","journal-title":"PLoS One"},{"key":"e_1_2_7_12_1","unstructured":"HestnessJ NarangS ArdalaniNet\u00a0al. (2017):Deep learning scaling is predictable empirically.https:\/\/arxiv.org\/abs\/1705.10694. (Accessed on 04 Nov 2020)."},{"key":"e_1_2_7_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ophtha.2019.06.005"},{"key":"e_1_2_7_14_1","unstructured":"KingmaDP&BaJL(2015):Adam: A method for stochastic optimization. 3rd Int Conf Learn Represent ICLR 2015 \u2013 Conf. Track Proc. International Conference on Learning Representations ICLR.https:\/\/arxiv.org\/abs\/1412.6980v9. (Accessed on 04 Nov 2020)."},{"key":"e_1_2_7_15_1","doi-asserted-by":"publisher","DOI":"10.1167\/iovs.09-4635"},{"key":"e_1_2_7_16_1","doi-asserted-by":"publisher","DOI":"10.1111\/aos.12781"},{"key":"e_1_2_7_17_1","unstructured":"RolnickD VeitA BelongieS&ShavitN(2017):Deep learning is robust to massive label noise.https:\/\/arxiv.org\/abs\/1705.10694. (Accessed on 04 Nov 2020)."},{"key":"e_1_2_7_18_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1600-0420.2006.00736.x"},{"key":"e_1_2_7_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ophtha.2017.10.031"},{"key":"e_1_2_7_20_1","doi-asserted-by":"publisher","DOI":"10.1136\/bjophthalmol-2014-305702"},{"key":"e_1_2_7_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ophtha.2020.03.010"},{"key":"e_1_2_7_22_1","article-title":"Prevalence of neovascular age\u2010related macular degeneration and geographic atrophy in Denmark","volume":"64","author":"Sedeh FB","year":"2017","journal-title":"Dan Med J"},{"key":"e_1_2_7_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2019.101758"},{"key":"e_1_2_7_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.preteyeres.2019.04.003"},{"key":"e_1_2_7_25_1","doi-asserted-by":"publisher","DOI":"10.2337\/dc18-0148"},{"key":"e_1_2_7_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/S2214-109X(13)70145-1"},{"key":"e_1_2_7_27_1","unstructured":"ZhouB KhoslaA LapedrizaA OlivaA&TorralbaA(2015):Learning deep features for discriminative localization.https:\/\/arxiv.org\/abs\/1512.04150v1. (Accessed on 01 Nov 2020)."}],"container-title":["Acta Ophthalmologica"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/aos.14895","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/aos.14895","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/aos.14895","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,27]],"date-time":"2023-08-27T04:39:57Z","timestamp":1693111197000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/aos.14895"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,5,15]]},"references-count":26,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2022,2]]}},"alternative-id":["10.1111\/aos.14895"],"URL":"https:\/\/doi.org\/10.1111\/aos.14895","archive":["Portico"],"relation":{},"ISSN":["1755-375X","1755-3768"],"issn-type":[{"value":"1755-375X","type":"print"},{"value":"1755-3768","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,5,15]]},"assertion":[{"value":"2020-12-06","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-04-18","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-05-15","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}