{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T11:16:49Z","timestamp":1776943009232,"version":"3.51.4"},"reference-count":33,"publisher":"BMJ","issue":"10","license":[{"start":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T00:00:00Z","timestamp":1761696000000},"content-version":"unspecified","delay-in-days":28,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Frost Trust Charity","award":["SBalal"],"award-info":[{"award-number":["SBalal"]}]},{"DOI":"10.13039\/100010399","name":"European Society of Cataract and Refractive Surgeons","doi-asserted-by":"publisher","award":["OCT-KC"],"award-info":[{"award-number":["OCT-KC"]}],"id":[{"id":"10.13039\/100010399","id-type":"DOI","asserted-by":"publisher"}]},{"name":"University College London Research Scholarship Award","award":["2562232"],"award-info":[{"award-number":["2562232"]}]},{"name":"National Institute for Health Research (NIHR) Doctoral Fellowship Award","award":["NIHR305630"],"award-info":[{"award-number":["NIHR305630"]}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Open"],"accepted":{"date-parts":[[2025,9,30]]},"published-print":{"date-parts":[[2025,10]]},"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To assess the capability of a convolutional neural network trained by transfer learning on anterior segment optical coherence tomography (AS-OCT) images, Placido-disk corneal topography images and external photographs to predict age and biological sex.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Design<\/jats:title>\n                    <jats:p>Development of a deep learning model trained on retrospectively collected data using transfer learning.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Setting<\/jats:title>\n                    <jats:p>A multicentre secondary care public health trust based in London.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Participants<\/jats:title>\n                    <jats:p>We included 557,468 scans from 40,592 eyes of 20,542 patients. Data were extracted from all patients who underwent MS-39 imaging within our trust from October 2020 to March 2023.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Primary and secondary outcome measures<\/jats:title>\n                    <jats:p>Primary outcome measures for biological sex classification included accuracy, precision, recall, F1-score and area under the receiver operating curve (ROC-AUC). Primary outcome measures for age prediction were Pearson correlation coefficients (r), coefficients of determination (R\u00b2) and the mean absolute error (MAE) to evaluate the predictive performance. The secondary outcome was to visualise and interpret the model\u2019s decision-making process through the construction of saliency maps.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>For age prediction, the MAEs for the Placido, AS-OCT and external photograph models were 5.2, 5.1 and 6.2 years, respectively. For gender classification, the same models achieved ROC-AUCs of 0.88, 0.73 and 0.81, respectively. No difference in performance was found in the analysis of corneas with pathological topography. The saliency maps highlighted the peri-limbal cornea for age prediction and the central cornea for gender discrimination.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Our study demonstrates that deep learning models can extract age and gender information from anterior segment images. These findings support the concept that the anterior segment, like the retina, encodes important biological information. Future research should explore whether these models can predict specific systemic conditions.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjopen-2025-107196","type":"journal-article","created":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T17:27:54Z","timestamp":1761758874000},"page":"e107196","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":1,"title":["Investigating the capability of deep learning models to predict age and biological sex from anterior segment ophthalmic imaging: a multi-centre retrospective study"],"prefix":"10.1136","volume":"15","author":[{"given":"Shafi","family":"Balal","sequence":"first","affiliation":[{"name":"University College London Institute of Ophthalmology, London, UK"},{"name":"Moorfields Eye Hospital NHS Foundation Trust, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6658-8503","authenticated-orcid":false,"given":"Laurence","family":"Cox","sequence":"additional","affiliation":[{"name":"Moorfields Eye Hospital NHS Foundation Trust, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ajmal","family":"Khan","sequence":"additional","affiliation":[{"name":"Anglia Ruskin University, Chelmsford, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lynn","family":"Kandakji","sequence":"additional","affiliation":[{"name":"University College London, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Marcello","family":"Leucci","sequence":"additional","affiliation":[{"name":"Moorfields Eye Hospital NHS Foundation Trust, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Pearse A","family":"Keane","sequence":"additional","affiliation":[{"name":"University College London Institute of Ophthalmology, London, UK"},{"name":"Moorfields Eye Hospital NHS Foundation Trust, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Daniel","family":"Gore","sequence":"additional","affiliation":[{"name":"Moorfields Eye Hospital NHS Foundation Trust, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1782-4711","authenticated-orcid":false,"given":"Nikolas","family":"Pontikos","sequence":"additional","affiliation":[{"name":"University College London Institute of Ophthalmology, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bruce","family":"Allan","sequence":"additional","affiliation":[{"name":"University College London Institute of Ophthalmology, London, UK"},{"name":"Moorfields Eye Hospital NHS Foundation Trust, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"239","published-online":{"date-parts":[[2025,10,29]]},"reference":[{"key":"2025120415091096000_15.10.e107196.1","doi-asserted-by":"publisher","DOI":"10.2337\/ds17-0010"},{"key":"2025120415091096000_15.10.e107196.2","doi-asserted-by":"crossref","first-page":"1703","DOI":"10.1007\/s11357-023-00920-4","article-title":"Deep neural network-estimated age using optical coherence tomography predicts mortality","volume":"46","author":"Chen","year":"2024","journal-title":"Geroscience"},{"key":"2025120415091096000_15.10.e107196.3","first-page":"2042","article-title":"Prediction of gender from macular optical coherence tomography using deep learning","volume":"61","author":"Chueh","year":"2020","journal-title":"Invest Ophthalmol Vis Sci"},{"key":"2025120415091096000_15.10.e107196.4","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-021-89743-x"},{"key":"2025120415091096000_15.10.e107196.5","first-page":"1105","article-title":"Age prediction from retinal fundus images and segmented vessel images using deep learning","volume":"64","author":"Wang","year":"2023","journal-title":"Invest Ophthalmol Vis Sci"},{"key":"2025120415091096000_15.10.e107196.6","doi-asserted-by":"crossref","first-page":"158","DOI":"10.1038\/s41551-018-0195-0","article-title":"Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning","volume":"2","author":"Poplin","year":"2018","journal-title":"Nat Biomed Eng"},{"key":"2025120415091096000_15.10.e107196.7","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1016\/j.inffus.2023.03.007","article-title":"Deep learning for brain age estimation: A systematic review","volume":"96","author":"Tanveer","year":"2023","journal-title":"Information Fusion"},{"key":"2025120415091096000_15.10.e107196.8","doi-asserted-by":"publisher","DOI":"10.1136\/bjophthalmol-2021-319807"},{"key":"2025120415091096000_15.10.e107196.9","doi-asserted-by":"crossref","DOI":"10.3390\/jcm12010152","article-title":"Application of Deep Learning to Retinal-Image-Based Oculomics for Evaluation of Systemic Health: A Review","volume":"12","author":"Wu","year":"2022","journal-title":"J Clin Med"},{"key":"2025120415091096000_15.10.e107196.10","doi-asserted-by":"crossref","first-page":"e257","DOI":"10.1016\/S2589-7500(23)00022-5","article-title":"A deep learning model for novel systemic biomarkers in photographs of the external eye: a retrospective study","volume":"5","author":"Babenko","year":"2023","journal-title":"Lancet Digit Health"},{"key":"2025120415091096000_15.10.e107196.11","doi-asserted-by":"crossref","first-page":"1370","DOI":"10.1038\/s41551-022-00867-5","article-title":"Detection of signs of disease in external photographs of the eyes via deep learning","volume":"6","author":"Babenko","year":"2022","journal-title":"Nat Biomed Eng"},{"key":"2025120415091096000_15.10.e107196.12","doi-asserted-by":"crossref","first-page":"4796","DOI":"10.1167\/iovs.15-16483","article-title":"Three-Dimensional Morphometric Analysis of the Iris by Swept-Source Anterior Segment Optical Coherence Tomography in a Caucasian Population","volume":"56","author":"Invernizzi","year":"2015","journal-title":"Invest Ophthalmol Vis Sci"},{"key":"2025120415091096000_15.10.e107196.13","doi-asserted-by":"crossref","first-page":"12","DOI":"10.1167\/tvst.10.1.12","article-title":"Predicting Age From Optical Coherence Tomography Scans With Deep Learning","volume":"10","author":"Shigueoka","year":"2021","journal-title":"Trans Vis Sci Tech"},{"key":"2025120415091096000_15.10.e107196.14","doi-asserted-by":"crossref","first-page":"3637","DOI":"10.1038\/s41598-024-54251-1","article-title":"Deep learning predicts prevalent and incident Parkinson\u2019s disease from UK Biobank fundus imaging","volume":"14","author":"Tran","year":"2024","journal-title":"Sci Rep"},{"key":"2025120415091096000_15.10.e107196.15","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1167\/tvst.11.11.7","article-title":"Demographics Prediction and Heatmap Generation From OCT Images of Anterior Segment of the Eye: A Vision Transformer Model Study","volume":"11","author":"Lee","year":"2022","journal-title":"Trans Vis Sci Tech"},{"key":"2025120415091096000_15.10.e107196.16","doi-asserted-by":"crossref","DOI":"10.1016\/j.xops.2023.100388","article-title":"A Datasheet for the INSIGHT University Hospitals Birmingham Retinal Vein Occlusion Data Set","volume":"3","author":"Bilton","year":"2023","journal-title":"Ophthalmol Sci"},{"key":"2025120415091096000_15.10.e107196.17","doi-asserted-by":"crossref","unstructured":"He K , Zhang X , Ren S , eds. Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition; 2016","DOI":"10.1109\/CVPR.2016.90"},{"key":"2025120415091096000_15.10.e107196.18","unstructured":"Dosovitskiy A , Beyer L , Kolesnikov A , et al . An image is worth 16x16 words: Transformers for image recognition at scale. arXiv 2020. doi:10.48550\/arXiv.2010.11929"},{"key":"2025120415091096000_15.10.e107196.19","doi-asserted-by":"crossref","unstructured":"Selvaraju RR , Cogswell M , Das A , eds. Grad-cam: visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE international conference on computer vision; 2017","DOI":"10.1109\/ICCV.2017.74"},{"key":"2025120415091096000_15.10.e107196.20","doi-asserted-by":"publisher","DOI":"10.1016\/S2589-7500(20)30216-8"},{"key":"2025120415091096000_15.10.e107196.21","doi-asserted-by":"publisher","DOI":"10.2147\/OPTH.S219912"},{"key":"2025120415091096000_15.10.e107196.22","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1186\/s12886-016-0219-z","article-title":"Fourier domain optical coherence tomography to assess the iridocorneal angle and correlation study in a large Caucasian population","volume":"16","author":"Fern\u00e1ndez-Vigo","year":"2016","journal-title":"BMC Ophthalmol"},{"key":"2025120415091096000_15.10.e107196.23","doi-asserted-by":"crossref","first-page":"100013","DOI":"10.1016\/j.ajoint.2024.100013","article-title":"Glaucoma in women: What do we know so far - A systematic review","volume":"1","author":"Rizk","year":"2024","journal-title":"AJO International"},{"key":"2025120415091096000_15.10.e107196.24","doi-asserted-by":"publisher","DOI":"10.1016\/S2589-7500(22)00169-8"},{"key":"2025120415091096000_15.10.e107196.25","doi-asserted-by":"publisher","DOI":"10.1136\/bjophthalmol-2020-317659"},{"key":"2025120415091096000_15.10.e107196.26","doi-asserted-by":"publisher","DOI":"10.1016\/j.ophtha.2017.02.008"},{"key":"2025120415091096000_15.10.e107196.27","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0233166","article-title":"Prediction of hypertension, hyperglycemia and dyslipidemia from retinal fundus photographs via deep learning: A cross-sectional study of chronic diseases in central China","volume":"15","author":"Zhang","year":"2020","journal-title":"PLoS ONE"},{"key":"2025120415091096000_15.10.e107196.28","doi-asserted-by":"crossref","first-page":"2109","DOI":"10.1038\/s41433-022-02298-7","article-title":"UK Biobank retinal imaging grading: methodology, baseline characteristics and findings for common ocular diseases","volume":"37","author":"Warwick","year":"2023","journal-title":"Eye (Lond)"},{"key":"2025120415091096000_15.10.e107196.29","doi-asserted-by":"crossref","DOI":"10.1136\/bmjopen-2021-058552","article-title":"AlzEye: longitudinal record-level linkage of ophthalmic imaging and hospital admissions of 353 157 patients in London, UK","volume":"12","author":"Wagner","year":"2022","journal-title":"BMJ Open"},{"key":"2025120415091096000_15.10.e107196.30","doi-asserted-by":"publisher","DOI":"10.1007\/s12325-021-01922-4"},{"key":"2025120415091096000_15.10.e107196.31","doi-asserted-by":"publisher","DOI":"10.2147\/CIA.S61790"},{"key":"2025120415091096000_15.10.e107196.32","doi-asserted-by":"crossref","DOI":"10.1186\/s12886-022-02578-6","article-title":"Prevalence and associated factors of corneal arcus in the geriatric population; Tehran geriatric eye study","volume":"22","author":"Hashemi","year":"2022","journal-title":"BMC Ophthalmol"},{"key":"2025120415091096000_15.10.e107196.33","doi-asserted-by":"publisher","DOI":"10.1167\/iovs.11-8404"}],"container-title":["BMJ Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjopen-2025-107196","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,4]],"date-time":"2025-12-04T23:16:33Z","timestamp":1764890193000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmjopen.bmj.com\/lookup\/doi\/10.1136\/bmjopen-2025-107196"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10]]},"references-count":33,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2025,10,29]]},"published-print":{"date-parts":[[2025,10]]}},"alternative-id":["10.1136\/bmjopen-2025-107196"],"URL":"https:\/\/doi.org\/10.1136\/bmjopen-2025-107196","relation":{},"ISSN":["2044-6055","2044-6055"],"issn-type":[{"value":"2044-6055","type":"print"},{"value":"2044-6055","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10]]}}}