{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T10:47:41Z","timestamp":1781866061557,"version":"3.54.5"},"reference-count":13,"publisher":"Public Library of Science (PLoS)","issue":"8","license":[{"start":{"date-parts":[[2025,8,21]],"date-time":"2025-08-21T00:00:00Z","timestamp":1755734400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n<jats:title>Background<\/jats:title>\n<jats:p>Retinal fundus images provide a noninvasive window into systemic health, offering opportunities for early detection of metabolic disorders such as metabolic syndrome (METS).<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec002\">\n<jats:title>Objective<\/jats:title>\n<jats:p>This study aimed to develop a deep learning model to predict METS from fundus images obtained during routine health checkups, leveraging a multi-task learning approach.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec003\">\n<jats:title>Methods<\/jats:title>\n<jats:p>We retrospectively analyzed 5,000 fundus images from Japanese health checkup participants.<\/jats:p>\n<jats:p>Convolutional neural network (CNN) models were trained to classify METS status, incorporating fundus-specific data augmentation strategies and auxiliary regression tasks targeting clinical parameters such as abdominal circumference (AC). Model performance was evaluated using validation accuracy, test accuracy, and the area under the receiver operating characteristic curve (AUC).<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec004\">\n<jats:title>Results<\/jats:title>\n<jats:p>Models employing fundus-specific augmentation demonstrated more stable convergence and superior validation accuracy compared to general-purpose augmentation. Incorporating AC as an auxiliary task further enhanced performance across architectures. The final ensemble model with test-time augmentation achieved a test accuracy of 0.696 and an AUC of 0.73178.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec005\">\n<jats:title>Conclusion<\/jats:title>\n<jats:p>Combining multi-task learning, fundus-specific data augmentation, and ensemble prediction substantially improves deep learning-based METS classification from fundus images. This approach may offer a practical, noninvasive screening tool for metabolic syndrome in general health checkup settings.<\/jats:p>\n<\/jats:sec>","DOI":"10.1371\/journal.pone.0325337","type":"journal-article","created":{"date-parts":[[2025,8,21]],"date-time":"2025-08-21T17:28:26Z","timestamp":1755797306000},"page":"e0325337","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["Multi-task deep learning for predicting metabolic syndrome from retinal fundus images in a Japanese health checkup dataset"],"prefix":"10.1371","volume":"20","author":[{"given":"Tohru","family":"Itoh","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2394-9561","authenticated-orcid":true,"given":"Koichi","family":"Nishitsuka","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yasufumi","family":"Fukuma","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Satoshi","family":"Wada","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2025,8,21]]},"reference":[{"issue":"17","key":"pone.0325337.ref001","doi-asserted-by":"crossref","first-page":"2735","DOI":"10.1161\/CIRCULATIONAHA.105.169404","article-title":"Diagnosis and management of the metabolic syndrome: an American Heart Association\/National Heart, Lung, and Blood Institute Scientific Statement","volume":"112","author":"SM Grundy","year":"2005","journal-title":"Circulation"},{"issue":"9491","key":"pone.0325337.ref002","doi-asserted-by":"crossref","first-page":"1059","DOI":"10.1016\/S0140-6736(05)67402-8","article-title":"The metabolic syndrome--a new worldwide definition","volume":"366","author":"KGMM Alberti","year":"2005","journal-title":"Lancet"},{"issue":"2","key":"pone.0325337.ref003","doi-asserted-by":"crossref","first-page":"161","DOI":"10.1136\/bjo.2007.127449","article-title":"The metabolic syndrome and retinal microvascular signs in a Japanese population: the Funagata study","volume":"92","author":"R Kawasaki","year":"2008","journal-title":"Br J Ophthalmol"},{"key":"pone.0325337.ref004","article-title":"An overview of multi-task learning in deep neural networks","author":"S Ruder","year":"2017","journal-title":"arXiv"},{"key":"pone.0325337.ref005","first-page":"3730","article-title":"Deep learning face attributes in the wild.","author":"Z Liu","year":"2015","journal-title":"Proceedings of the IEEE International Conference on Computer Vision (ICCV). Santiago, Chile"},{"issue":"3","key":"pone.0325337.ref006","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":"R Poplin","year":"2018","journal-title":"Nat Biomed Eng"},{"issue":"1","key":"pone.0325337.ref007","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1038\/s41746-025-01588-0","article-title":"Vision transformer based interpretable metabolic syndrome classification using retinal Images","volume":"8","author":"TK Lee","year":"2025","journal-title":"NPJ Digit Med"},{"issue":"4","key":"pone.0325337.ref008","first-page":"794","article-title":"Definition and the diagnostic standard for metabolic syndrome","volume":"94","author":"Committee to Evaluate Diagnostic Standards for Metabolic Syndrome","year":"2005","journal-title":"Nihon Naika Gakkai Zasshi"},{"issue":"6","key":"pone.0325337.ref009","doi-asserted-by":"crossref","first-page":"499","DOI":"10.1007\/s10384-022-00941-0","article-title":"Japan Ocular Imaging Registry: a national ophthalmology real-world database","volume":"66","author":"M Miyake","year":"2022","journal-title":"Jpn J Ophthalmol"},{"issue":"6","key":"pone.0325337.ref010","article-title":"A deep learning algorithm to detect chronic kidney disease from retinal photographs in community-based populations","volume":"2","author":"C Sabanayagam","year":"2020","journal-title":"Lancet Digit Health"},{"issue":"1","key":"pone.0325337.ref011","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1038\/s41551-019-0487-z","article-title":"Detection of anaemia from retinal fundus images via deep learning","volume":"4","author":"A Mitani","year":"2020","journal-title":"Nat Biomed Eng"},{"issue":"1","key":"pone.0325337.ref012","doi-asserted-by":"crossref","first-page":"19369","DOI":"10.1038\/s41598-020-76513-4","article-title":"Prediction of age and brachial-ankle pulse-wave velocity using ultra-wide-field pseudo-color images by deep learning","volume":"10","author":"D Nagasato","year":"2020","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0325337.ref013","doi-asserted-by":"crossref","first-page":"28","DOI":"10.1186\/s12877-023-04593-8","article-title":"Ocular biomarkers of cognitive decline based on deep-learning retinal vessel segmentation","volume":"24","author":"R Li","year":"2024","journal-title":"BMC Geriatr"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0325337","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,21]],"date-time":"2025-08-21T17:28:29Z","timestamp":1755797309000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0325337"}},"subtitle":[],"editor":[{"given":"Pan","family":"Li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,8,21]]},"references-count":13,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2025,8,21]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0325337","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8,21]]}}}