{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T23:02:11Z","timestamp":1785538931575,"version":"3.56.0"},"reference-count":47,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2022,6,15]],"date-time":"2022-06-15T00:00:00Z","timestamp":1655251200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100010428","name":"Innovation and Technology Fund","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100010428","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>We aim to develop a multi-task three-dimensional (3D) deep learning (DL) model to detect glaucomatous optic neuropathy (GON) and myopic features (MF) simultaneously from spectral-domain optical coherence tomography (SDOCT) volumetric scans.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Each volumetric scan was labelled as GON according to the criteria of retinal nerve fibre layer (RNFL) thinning, with a structural defect that correlated in position with the visual field defect (i.e., reference standard). MF were graded by the SDOCT <jats:italic>en face<\/jats:italic> images, defined as presence of peripapillary atrophy (PPA), optic disc tilting, or fundus tessellation. The multi-task DL model was developed by ResNet with output of Yes\/No GON and Yes\/No MF. SDOCT scans were collected in a tertiary eye hospital (Hong Kong SAR, China) for training (80%), tuning (10%), and internal validation (10%). External testing was performed on five independent datasets from eye centres in Hong Kong, the United States, and Singapore, respectively. For GON detection, we compared the model to the average RNFL thickness measurement generated from the SDOCT device. To investigate whether MF can affect the model\u2019s performance on GON detection, we conducted subgroup analyses in groups stratified by Yes\/No MF. The area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and accuracy were reported.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>A total of 8,151 SDOCT volumetric scans from 3,609 eyes were collected. For detecting GON, in the internal validation, the proposed 3D model had significantly higher AUROC (0.949 vs. 0.913, <jats:italic>p<\/jats:italic> &amp;lt; 0.001) than average RNFL thickness in discriminating GON from normal. In the external testing, the two approaches had comparable performance. In the subgroup analysis, the multi-task DL model performed significantly better in the group of \u201cno MF\u201d (0.883 vs. 0.965, <jats:italic>p<\/jats:italic>-value &amp;lt; 0.001) in one external testing dataset, but no significant difference in internal validation and other external testing datasets. The multi-task DL model\u2019s performance to detect MF was also generalizable in all datasets, with the AUROC values ranging from 0.855 to 0.896.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>The proposed multi-task 3D DL model demonstrated high generalizability in all the datasets and the presence of MF did not affect the accuracy of GON detection generally.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fmed.2022.860574","type":"journal-article","created":{"date-parts":[[2022,6,15]],"date-time":"2022-06-15T04:47:00Z","timestamp":1655268420000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":21,"title":["Three-Dimensional Multi-Task Deep Learning Model to Detect Glaucomatous Optic Neuropathy and Myopic Features From Optical Coherence Tomography Scans: A Retrospective Multi-Centre Study"],"prefix":"10.3389","volume":"9","author":[{"given":"An Ran","family":"Ran","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xi","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Poemen P.","family":"Chan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Noel C.","family":"Chan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wilson","family":"Yip","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alvin L.","family":"Young","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mandy O. M.","family":"Wong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hon-Wah","family":"Yung","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Robert T.","family":"Chang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Suria S.","family":"Mannil","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yih Chung","family":"Tham","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ching-Yu","family":"Cheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hao","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fei","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiulan","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pheng-Ann","family":"Heng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Clement C.","family":"Tham","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Carol Y.","family":"Cheung","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2022,6,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"e339","DOI":"10.1016\/S2214-109X(13)70113-X","article-title":"Causes of vision loss worldwide, 1990-2010: a systematic analysis.","volume":"1","author":"Bourne","year":"2013","journal-title":"Lancet Glob Health."},{"key":"B2","doi-asserted-by":"publisher","first-page":"2081","DOI":"10.1016\/j.ophtha.2014.05.013","article-title":"Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis.","volume":"121","author":"Tham","year":"2014","journal-title":"Ophthalmology."},{"key":"B3","doi-asserted-by":"publisher","first-page":"1263.e1","DOI":"10.1016\/j.ophtha.2009.04.013","article-title":"Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: a variability and diagnostic performance study.","volume":"116","author":"Leung","year":"2009","journal-title":"Ophthalmology."},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0199134","article-title":"Diagnostic accuracy of macular ganglion cell-inner plexiform layer thickness for glaucoma detection in a population-based study: comparison with optic nerve head imaging parameters.","volume":"13","author":"Koh","year":"2018","journal-title":"PLoS One."},{"key":"B5","doi-asserted-by":"publisher","first-page":"849","DOI":"10.1016\/j.ophtha.2013.10.044","article-title":"Diagnostic performance of optical coherence tomography ganglion cell\u2013inner plexiform layer thickness measurements in early glaucoma.","volume":"121","author":"Mwanza","year":"2014","journal-title":"Ophthalmology."},{"key":"B6","doi-asserted-by":"crossref","first-page":"2294","DOI":"10.1016\/j.ophtha.2009.06.012","article-title":"Sensitivity and specificity of time-domain versus spectral-domain optical coherence tomography in diagnosing early to moderate glaucoma.","volume":"116","author":"Chang","year":"2009","journal-title":"Ophthalmology."},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1167\/tvst.6.4.9","article-title":"Associations between optic nerve head-related anatomical parameters and refractive error over the full range of glaucoma severity.","volume":"6","author":"Baniasadi","year":"2017","journal-title":"Transl Vis Sci Technol."},{"key":"B8","doi-asserted-by":"publisher","first-page":"1873","DOI":"10.1016\/j.ophtha.2015.05.031","article-title":"Fundus tessellation: prevalence and associated factors: the Beijing Eye Study 2011.","volume":"122","author":"Yan","year":"2015","journal-title":"Ophthalmology."},{"key":"B9","doi-asserted-by":"crossref","first-page":"260","DOI":"10.1097\/IJG.0b013e31820719e1","article-title":"Myopic optic disc tilt and the characteristics of peripapillary retinal nerve fiber layer thickness measured by spectral-domain optical coherence tomography.","volume":"21","author":"Hwang","year":"2012","journal-title":"J Glaucoma."},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0047237","article-title":"Parapapillary atrophy: histological gamma zone and delta zone.","volume":"7","author":"Jonas","year":"2012","journal-title":"PLoS One."},{"key":"B11","doi-asserted-by":"publisher","first-page":"646","DOI":"10.1016\/j.ajo.2011.04.002","article-title":"Diagnostic classification of retinal nerve fiber layer measurement in myopic eyes: a comparison between time-domain and spectral-domain optical coherence tomography.","volume":"152","author":"Qiu","year":"2011","journal-title":"Am J Ophthalmol."},{"key":"B12","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning.","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature."},{"key":"B13","first-page":"715","article-title":"Glaucoma detection based on deep convolutional neural network.","volume":"2015","author":"Xiangyu","year":"2015","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc."},{"key":"B14","doi-asserted-by":"publisher","first-page":"1086","DOI":"10.1097\/IJG.0000000000000765","article-title":"Hybrid deep learning on single wide-field optical coherence tomography scans accurately classifies glaucoma suspects.","volume":"26","author":"Muhammad","year":"2017","journal-title":"J Glaucoma."},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-35044-9","article-title":"Performance of deep learning architectures and transfer learning for detecting glaucomatous optic neuropathy in fundus photographs.","volume":"8","author":"Christopher","year":"2018","journal-title":"Sci Rep."},{"key":"B16","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1016\/j.ophtha.2018.12.033","article-title":"From machine to machine: an OCT-trained deep learning algorithm for objective quantification of glaucomatous damage in fundus photographs.","volume":"126","author":"Medeiros","year":"2018","journal-title":"Ophthalmology."},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0219126","article-title":"A feature agnostic approach for glaucoma detection in OCT volumes.","volume":"14","author":"Maetschke","year":"2019","journal-title":"PLoS One."},{"key":"B18","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1016\/j.ajo.2019.01.011","article-title":"A deep learning algorithm to quantify neuroretinal rim loss from optic disc photographs.","volume":"201","author":"Thompson","year":"2019","journal-title":"Am J Ophthalmol."},{"key":"B19","doi-asserted-by":"publisher","first-page":"1199","DOI":"10.1016\/j.ophtha.2018.01.023","article-title":"Efficacy of a deep learning system for detecting glaucomatous optic neuropathy based on color fundus photographs.","volume":"125","author":"Li","year":"2018","journal-title":"Ophthalmology."},{"key":"B20","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-018-33013-w","article-title":"Development of a deep residual learning algorithm to screen for glaucoma from fundus photography.","volume":"8","author":"Shibata","year":"2018","journal-title":"Sci Rep."},{"key":"B21","doi-asserted-by":"publisher","first-page":"e172","DOI":"10.1016\/S2589-7500(19)30085-8","article-title":"Detection of glaucomatous optic neuropathy with spectral-domain optical coherence tomography: a retrospective training and validation deep-learning analysis.","volume":"1","author":"Ran","year":"2019","journal-title":"Lancet Digit Health."},{"key":"B22","doi-asserted-by":"publisher","first-page":"e273","DOI":"10.1111\/aos.12324","article-title":"Peripapillary ring: histology and correlations.","volume":"92","author":"Jonas","year":"2014","journal-title":"Acta Ophthalmol."},{"key":"B23","doi-asserted-by":"publisher","first-page":"830","DOI":"10.1167\/iovs.12-11158","article-title":"Association between myopia and glaucoma in the United States Population.","volume":"54","author":"Qiu","year":"2013","journal-title":"Invest Ophth Vis Sci."},{"key":"B24","article-title":"Multi-task learning with deep neural networks: a survey.","author":"Crawshaw","year":"2020","journal-title":"arXiv."},{"key":"B25","doi-asserted-by":"publisher","first-page":"6545","DOI":"10.1109\/EMBC.2019.8856981","article-title":"Fetal ultrasound image segmentation for measuring biometric parameters using multi-task deep learning.","volume":"2019","author":"Sobhaninia","year":"2019","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc."},{"key":"B26","doi-asserted-by":"crossref","DOI":"10.1016\/j.compbiomed.2020.104037","article-title":"Multi-task deep learning based CT imaging analysis for COVID-19 pneumonia: classification and segmentation.","volume":"126","author":"Amyar","year":"2020","journal-title":"Comput Biol Med."},{"key":"B27","doi-asserted-by":"publisher","first-page":"2912","DOI":"10.1109\/JBHI.2020.2973614","article-title":"An end-to-end multi-task deep learning framework for skin lesion analysis.","volume":"24","author":"Song","year":"2020","journal-title":"IEEE J Biomed Health."},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101695","article-title":"Towards multi-center glaucoma OCT image screening with semi-supervised joint structure and function multi-task learning.","volume":"63","author":"Wang","year":"2020","journal-title":"Med Image Anal."},{"key":"B29","doi-asserted-by":"publisher","first-page":"2078","DOI":"10.2337\/dc20-3064","article-title":"A multitask deep-learning system to classify diabetic macular edema for different optical coherence tomography devices: a multicenter analysis.","volume":"44","author":"Tang","year":"2021","journal-title":"Diabetes Care."},{"key":"B30","doi-asserted-by":"publisher","first-page":"184","DOI":"10.1097\/IAE.0000000000003287","article-title":"A multitask deep-learning system for assessment of diabetic macular ischemia on optical coherence tomography angiography images.","volume":"42","author":"Yang","year":"2022","journal-title":"Retina."},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1117\/1.NPh.6.4.041110","article-title":"Artificial intelligence deep learning algorithm for discriminating ungradable optical coherence tomography three-dimensional volumetric optic disc scans.","volume":"6","author":"Ran","year":"2019","journal-title":"Neurophotonics."},{"key":"B32","doi-asserted-by":"publisher","first-page":"498","DOI":"10.1016\/s0002-9394(98)00272-4","article-title":"The effectiveness of intraocular pressure reduction in the treatment of normal-tension glaucoma. Collaborative Normal-Tension Glaucoma Study Group","volume":"126","year":"1998","journal-title":"Am J Ophthalmol"},{"key":"B33","doi-asserted-by":"publisher","first-page":"531","DOI":"10.1016\/j.ophtha.2010.07.013","article-title":"Analysis of peripapillary atrophy using spectral domain optical coherence tomography.","volume":"118","author":"Manjunath","year":"2011","journal-title":"Ophthalmology."},{"key":"B34","doi-asserted-by":"publisher","first-page":"247","DOI":"10.1016\/j.ajo.2004.08.076","article-title":"Optic disk ovality as an index of tilt and its relationship to myopia and perimetry.","volume":"139","author":"Tay","year":"2005","journal-title":"Am J Ophthalmol."},{"key":"B35","doi-asserted-by":"crossref","first-page":"2921","DOI":"10.1109\/CVPR.2016.319","article-title":"Learning Deep Features for Discriminative Localization.","author":"Zhou","year":"2016","journal-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)."},{"key":"B36","doi-asserted-by":"crossref","first-page":"312","DOI":"10.1001\/archopthalmol.2011.1576","article-title":"Effect of race, age, and axial length on optic nerve head parameters and retinal nerve fiber layer thickness measured by Cirrus HD-OCT.","volume":"130","author":"Knight","year":"2012","journal-title":"Arch Ophthalmol."},{"key":"B37","doi-asserted-by":"crossref","first-page":"1347","DOI":"10.1016\/j.ophtha.2007.11.027","article-title":"Relationship between retinal nerve fiber layer measurement and signal strength in optical coherence tomography.","volume":"115","author":"Cheung","year":"2008","journal-title":"Ophthalmology."},{"key":"B38","doi-asserted-by":"publisher","first-page":"7194","DOI":"10.1167\/iovs.12-9726","article-title":"Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: interpreting the rnfl maps in healthy myopic eyes.","volume":"53","author":"Leung","year":"2012","journal-title":"Invest Ophth Vis Sci."},{"key":"B39","doi-asserted-by":"publisher","first-page":"1032","DOI":"10.1001\/jamaophthalmol.2016.2343","article-title":"Evaluation of a myopic normative database for analysis of retinal nerve fiber layer thickness.","volume":"134","author":"Biswas","year":"2016","journal-title":"JAMA Ophthalmol."},{"key":"B40","doi-asserted-by":"crossref","first-page":"96","DOI":"10.1097\/ICU.0b013e32835cef31","article-title":"Myopia and glaucoma: diagnostic and therapeutic challenges.","volume":"24","author":"Chang","year":"2013","journal-title":"Curr Opin Ophthalmol."},{"key":"B41","doi-asserted-by":"publisher","first-page":"9629","DOI":"10.1167\/iovs.11-7481","article-title":"Determinants of quantitative optic nerve measurements using spectral domain optical coherence tomography in a population-based sample of non-glaucomatous subjects.","volume":"52","author":"Cheung","year":"2011","journal-title":"Invest Ophth Vis Sci."},{"key":"B42","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1016\/j.ophtha.2006.06.050","article-title":"High myopia and glaucoma susceptibility the Beijing Eye Study.","volume":"114","author":"Xu","year":"2007","journal-title":"Ophthalmology."},{"key":"B43","doi-asserted-by":"publisher","first-page":"1989","DOI":"10.1016\/j.ophtha.2011.03.012","article-title":"Myopia as a risk factor for open-angle glaucoma: a systematic review and meta-analysis.","volume":"118","author":"Marcus","year":"2011","journal-title":"Ophthalmology."},{"key":"B44","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1016\/j.ophtha.2015.07.002","article-title":"The association of refractive error with glaucoma in a multiethnic population.","volume":"123","author":"Shen","year":"2016","journal-title":"Ophthalmology."},{"key":"B45","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1097\/ICU.0000000000000548","article-title":"Myopic optic disc changes and its role in glaucoma.","volume":"30","author":"Tan","year":"2019","journal-title":"Curr Opin Ophthalmol."},{"key":"B46","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1117\/1.JBO.22.12.121713","article-title":"Ametropia, retinal anatomy, and OCT abnormality patterns in glaucoma. 1. Impacts of refractive error and interartery angle.","volume":"22","author":"Elze","year":"2017","journal-title":"J Biomed Opt."},{"key":"B47","first-page":"2672","article-title":"Generative adversarial nets","author":"Goodfellow","year":"2014","journal-title":"Proceedings of the Advances in Neural Information Processing Systems (NIPS)"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2022.860574\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,6,15]],"date-time":"2022-06-15T04:47:21Z","timestamp":1655268441000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2022.860574\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,6,15]]},"references-count":47,"alternative-id":["10.3389\/fmed.2022.860574"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2022.860574","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,6,15]]},"article-number":"860574"}}