{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T01:59:54Z","timestamp":1781661594203,"version":"3.54.5"},"reference-count":41,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T00:00:00Z","timestamp":1763424000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Cardiovascular Diseases (CVD) are the leading cause of death globally. Non-invasive, cost-effective imaging techniques play a crucial role in early detection and prevention of CVD. Optical Coherence Tomography (OCT) has gained recognition as a noninvasive method of detecting microvascular alterations that might enable earlier identification and targeting of at-risk patients. In this study, we investigated the potential of OCT as an additional imaging technique to predict future CVD events.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We analyzed retinal OCT data from the UK Biobank. The dataset included 612 patients who suffered a Myocardial Infarction (MI) or stroke within five years of imaging and 2,234 controls without CVD (total: 2,846 participants). A self-supervised deep learning approach based on Variational Autoencoders (VAE) was used to extract low-dimensional latent representations from high-dimensional 3D OCT images, capturing structural and morphological features of retinal and choroidal layers. These latent features, along with clinical data, were used to train a Random Forest (RF) classifier to differentiate between patients at risk of future CVD events (MI or stroke) and healthy controls.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Our model achieved an AUC of 0.75, sensitivity of 0.70, specificity of 0.70, and accuracy of 0.70. The choroidal layer in OCT images was identified as a key predictor of future CVD events, revealed through a novel model explainability approach.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Our findings demonstrate the potential of retinal OCT imaging, when combined with advanced deep learning methods, as a predictive tool for identifying individuals at increased risk of CVD events.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2025.1624550","type":"journal-article","created":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T06:23:23Z","timestamp":1763447003000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Predicting cardiovascular disease risk using retinal optical coherence tomography imaging"],"prefix":"10.3389","volume":"8","author":[{"given":"Cynthia","family":"Maldonado-Garcia","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rodrigo","family":"Bonazzola","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Enzo","family":"Ferrante","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thomas H.","family":"Julian","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Panagiotis I.","family":"Sergouniotis","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nishant","family":"Ravikumar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alejandro F.","family":"Frangi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,11,18]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"e1560","DOI":"10.1002\/widm.1560","article-title":"Artificial intelligence in assessing cardiovascular diseases and risk factors via retinal fundus images: A review of the last decade. Wiley Interdiscipl. Rev","volume":"14","author":"Abdollahi","year":"2024","journal-title":"Data Mining Knowl. Discov"},{"key":"B2","first-page":"1430","article-title":"Advancements in AI-enhanced OCT imaging for early disease detection and prevention in aging populations","volume":"8","author":"Arefin","year":"2025","journal-title":"Int. J. Multidiscipl. Res. Sci. Eng. Technol."},{"key":"B3","doi-asserted-by":"publisher","first-page":"19004705","DOI":"10.2139\/ssrn.3437793","article-title":"Generalisability of results from uk biobank: comparison with a pooling of 18 cohort studies","volume":"368","author":"Batty","year":"2019","journal-title":"medRxiv"},{"key":"B4","doi-asserted-by":"publisher","first-page":"m131","DOI":"10.1136\/bmj.m131","article-title":"Comparison of risk factor associations in uk biobank against representative, general population based studies with conventional response rates: prospective cohort study and individual participant meta-analysis","volume":"368","author":"Batty","year":"2020","journal-title":"BMJ"},{"key":"B5","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn"},{"key":"B6","doi-asserted-by":"publisher","first-page":"4314","DOI":"10.1038\/s41467-023-39902-7","article-title":"Detecting shortcut learning for fair medical Ai using shortcut testing","volume":"14","author":"Brown","year":"2022","journal-title":"Nat. Commun"},{"key":"B7","doi-asserted-by":"publisher","first-page":"7","DOI":"10.1186\/s12916-023-02728-7","article-title":"Retinal nerve fiber layer thinning as a novel fingerprint for cardiovascular events: results from the prospective cohorts in UK and China","volume":"21","author":"Chen","year":"2023","journal-title":"BMC Med"},{"key":"B8","doi-asserted-by":"publisher","first-page":"498","DOI":"10.1038\/s41551-020-00626-4","article-title":"A deep-learning system for the assessment of cardiovascular disease risk via the measurement of retinal-vessel calibre","volume":"5","author":"Cheung","year":"2020","journal-title":"Nat. Biomed. Eng"},{"key":"B9","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1038\/s41569-024-01060-8","article-title":"Standardization and clinical applications of retinal imaging biomarkers for cardiovascular disease: a roadmap from an NHLBI workshop","volume":"22","author":"Chew","year":"2025","journal-title":"Nat. Rev. Cardiol"},{"key":"B10","doi-asserted-by":"publisher","first-page":"334","DOI":"10.1136\/bjophthalmol-2013-303904","article-title":"Direct comparison of spectral-domain and swept-source oct in the measurement of choroidal thickness in normal eyes","volume":"98","author":"Copete","year":"2013","journal-title":"Br, J, Ophthalmol"},{"key":"B11","doi-asserted-by":"publisher","first-page":"55","DOI":"10.1038\/s42256-021-00427-7","article-title":"Predicting myocardial infarction through retinal scans and minimal personal information","volume":"4","author":"Diaz-Pinto","year":"2022","journal-title":"Nat. Mach. Intellig"},{"key":"B12","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1016\/B978-0-12-824349-7.00015-3","article-title":"\u201cAutoencoders and variational autoencoders in medical image analysis,\u201d","volume-title":"Biomedical Image Synthesis and Simulation","author":"Ehrhardt","year":"2022"},{"key":"B13","doi-asserted-by":"publisher","first-page":"323","DOI":"10.1016\/j.kint.2020.01.039","article-title":"The eye, the kidney, and cardiovascular disease: old concepts, better tools, and new horizons","volume":"98","author":"Farrah","year":"2020","journal-title":"Kidney Int"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1818","DOI":"10.1038\/s41433-021-01437-w","article-title":"Biomechanical properties of retina and choroid: a comprehensive review of techniques and translational relevance","volume":"35","author":"Ferrara","year":"2021","journal-title":"Eye"},{"key":"B15","doi-asserted-by":"publisher","first-page":"1026","DOI":"10.1093\/aje\/kwx246","article-title":"Comparison of sociodemographic and health-related characteristics of uk biobank participants with those of the general population","volume":"186","author":"Fry","year":"2017","journal-title":"Am. J. Epidemiol"},{"key":"B16","doi-asserted-by":"publisher","first-page":"27089","DOI":"10.1038\/s41598-024-78587-w","article-title":"Artificial intelligence-based prediction of neurocardiovascular risk score from retinal swept-source optical coherence tomography-angiography","volume":"14","author":"Germanese","year":"2024","journal-title":"Sci. Rep"},{"key":"B17","doi-asserted-by":"publisher","first-page":"114","DOI":"10.4103\/ijdt.ijdt_8_23","article-title":"Using machine learning to assess cardiovascular risk in type 2 diabetes mellitus patients without established cardiovascular disease: the mark-2 analysis","volume":"1","author":"Ghosal","year":"2022","journal-title":"Int. J. Diabet. Technol"},{"key":"B18","doi-asserted-by":"publisher","first-page":"j2099","DOI":"10.1136\/bmj.j2099","article-title":"Development and validation of qrisk3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study","volume":"357","author":"Hippisley-Cox","year":"2017","journal-title":"BMJ"},{"key":"B19","first-page":"1936","article-title":"\u201cLearning dynamics in linear vae: posterior collapse threshold, superfluous latent space pitfalls, and speedup with KL annealing,\u201d","volume-title":"International Conference on Artificial Intelligence and Statistics","author":"Ichikawa","year":"2024"},{"key":"B20","doi-asserted-by":"publisher","first-page":"e0164095","DOI":"10.1371\/journal.pone.0164095","article-title":"Optical coherence tomography in the uk biobank study \u2013 rapid automated analysis of retinal thickness for large population-based studies","volume":"11","author":"Keane","year":"2016","journal-title":"PLoS ONE"},{"key":"B21","doi-asserted-by":"publisher","first-page":"11178","DOI":"10.1038\/s41598-022-15374-5","article-title":"RF-CNN-F: random forest with convolutional neural network features for coronary artery disease diagnosis based on cardiac magnetic resonance","volume":"12","author":"Khozeimeh","year":"2022","journal-title":"Sci. Rep"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1312.6114","article-title":"Auto-encoding variational bayes","author":"Kingma","year":"2013","journal-title":"CoRR"},{"key":"B23","doi-asserted-by":"publisher","first-page":"9315","DOI":"10.1038\/s41598-022-13312-z","article-title":"Structural retinal changes in cerebral small vessel disease","volume":"12","author":"Langner","year":"2022","journal-title":"Sci. Rep"},{"key":"B24","doi-asserted-by":"publisher","first-page":"2139","DOI":"10.12688\/f1000research.21679.1","article-title":"R package \u201cQRISK3\u201d: an unofficial research purposed implementation of ClinRisk's QRISK3 algorithm into r","volume":"8","author":"Li","year":"2019","journal-title":"F1000Research"},{"key":"B25","volume-title":"Deaths from Cardiovascular Disease Surged 60% Globally over the last 30 Years: Report","author":"Lindstrom","year":"2023"},{"key":"B26","doi-asserted-by":"publisher","first-page":"e352","DOI":"10.1016\/S2666-7568(21)00088-X","article-title":"Effect of competing mortality risks on predictive performance of the QRISK3 cardiovascular risk prediction tool in older people and those with comorbidity: external validation population cohort study","volume":"2","author":"Livingstone","year":"2021","journal-title":"Lancet. Healthy Longevity"},{"key":"B27","article-title":"\u201cAn iterative image registration technique with an application to stereo vision,\u201d","volume-title":"International Joint Conference on Artificial Intelligence","author":"Lucas","year":"1981"},{"key":"B28","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-12053-4_58","article-title":"\u201cPredicting myocardial infarction using retinal oct imaging,\u201d","volume-title":"Annual Conference on Medical Image Understanding and Analysis","author":"Maldonado-Garcia","year":"2022"},{"key":"B29","doi-asserted-by":"publisher","first-page":"2016","DOI":"10.3390\/diagnostics12082016","article-title":"Retinal and choroidal thinning\u2014a predictor of coronary artery occlusion?","volume":"12","author":"Matulevici\u016b\u0117","year":"2022","journal-title":"Diagnostics"},{"key":"B30","doi-asserted-by":"publisher","first-page":"afac065","DOI":"10.1093\/ageing\/afac065","article-title":"Retinal photograph-based deep learning predicts biological age, and stratifies morbidity and mortality risk","volume":"51","author":"Nusinovici","year":"2022","journal-title":"Age Ageing"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1292","DOI":"10.1016\/j.ophtha.2012.11.021","article-title":"Transneuronal retrograde degeneration of the retinal ganglion cells in patients with cerebral infarction","volume":"120","author":"Park","year":"2013","journal-title":"Ophthalmology"},{"key":"B32","doi-asserted-by":"publisher","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":"B33","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2411.07348","article-title":"Exploring variational autoencoders for medical image generation: a comprehensive study","author":"Rais","year":"2024","journal-title":"arXiv"},{"key":"B34","doi-asserted-by":"publisher","first-page":"2659","DOI":"10.3390\/jcm10122659","article-title":"The impact of chronic heart failure on retinal vessel density assessed by optical coherence tomography angiography in children with dilated cardiomyopathy","volume":"10","author":"Rakusiewicz","year":"2021","journal-title":"J. Clin. Med"},{"key":"B35","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.pcad.2024.10.011","article-title":"Optical coherence tomography angiography in cardiovascular disease","volume":"87","author":"Runsewe","year":"2024","journal-title":"Prog. Cardiovasc. Dis"},{"key":"B36","doi-asserted-by":"publisher","first-page":"186","DOI":"10.1136\/bjo.2004.059824","article-title":"A new quality assessment parameter for optical coherence tomography","volume":"90","author":"Stein","year":"2006","journal-title":"Br. J. Ophthalmol"},{"key":"B37","doi-asserted-by":"publisher","first-page":"440","DOI":"10.1097\/ICU.0000000000000886","article-title":"Artificial intelligence in retinal imaging for cardiovascular disease prediction: current trends and future directions","volume":"33","author":"Wong","year":"2022","journal-title":"Curr. Opin. Ophthalmol"},{"key":"B38","doi-asserted-by":"publisher","first-page":"5245","DOI":"10.1038\/s41598-020-62133-5","article-title":"Study of cardiovascular disease prediction model based on random forest in eastern china","volume":"10","author":"Yang","year":"2020","journal-title":"Sci. Rep"},{"key":"B39","doi-asserted-by":"publisher","first-page":"473","DOI":"10.1016\/j.survophthal.2019.12.007","article-title":"Choroidal thickness in patients with cardiovascular disease: a review","volume":"65","author":"Yeung","year":"2020","journal-title":"Surv. Ophthalmol"},{"key":"B40","doi-asserted-by":"publisher","first-page":"4247","DOI":"10.2147\/DMSO.S438111","article-title":"Diabetic retinopathy and cardiovascular disease: a literature review","volume":"16","author":"Yu","year":"2023","journal-title":"Diabet. Metabol. Syndrome Obesity"},{"key":"B41","doi-asserted-by":"publisher","first-page":"156","DOI":"10.1038\/s41586-023-06555-x","article-title":"A foundation model for generalizable disease detection from retinal images","volume":"622","author":"Zhou","year":"2023","journal-title":"Nature"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1624550\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,18]],"date-time":"2025-11-18T06:23:25Z","timestamp":1763447005000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1624550\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,18]]},"references-count":41,"alternative-id":["10.3389\/frai.2025.1624550"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1624550","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,18]]},"article-number":"1624550"}}