{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T20:06:02Z","timestamp":1772136362030,"version":"3.50.1"},"reference-count":28,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,1,10]],"date-time":"2026-01-10T00:00:00Z","timestamp":1768003200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,1,23]],"date-time":"2026-01-23T00:00:00Z","timestamp":1769126400000},"content-version":"vor","delay-in-days":13,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100007185","name":"University of California, Los Angeles","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100007185","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>To develop a self-supervised learning (SSL) model that classifies optic disc phenotypes in primary open angle glaucoma (POAG) and explores novel phenotypic patterns with optic disc photographs (ODPs). We collected 850 ODPs from patients with POAG and applied data augmentation to address class imbalances, yielding 10,493 images. Using the DINO Vision Transformer as the backbone model, we trained an SSL model to extract 2048-dimensional latent features. These features were used for both supervised classification of six known phenotypes and unsupervised clustering. Classification performance was evaluated with Random Forest and XGBoost models. UMAP (Uniform Manifold Approximation and Projection)\u00a0was used for dimensionality reduction and feature visualization, and attention maps were generated for model interpretability. The DINO-EYE model features enabled phenotype classification with 91% accuracy with Random Forest and 92.1% after merging clinically similar phenotypes. Unsupervised clustering revealed coherent groupings, particularly for concentric thinning and extensive Peripapillary Atrophy (PPA), though no new phenotypes were unanimously confirmed by clinicians. The proposed model outperformed the RETFound SSL model in phenotype classification and demonstrated interpretable attention regions consistent with expert criteria. Our DINO-EYE effectively extracts clinically meaningful features from fundus images and enables accurate classification of optic disc phenotypes in POAG. It surpasses existing SSL models in performance and interpretability, offering promise for real-world glaucoma decision support and individualized care planning.<\/jats:p>","DOI":"10.1038\/s41598-025-33140-1","type":"journal-article","created":{"date-parts":[[2026,1,10]],"date-time":"2026-01-10T11:03:59Z","timestamp":1768043039000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["DINO-EYE: self-supervised learning for identification of different optic disc phenotypes in primary open angle glaucoma"],"prefix":"10.1038","volume":"16","author":[{"given":"Lourdes","family":"Grassi","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhe","family":"Fei","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Esteban","family":"Morales","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Joseph","family":"Caprioli","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2026,1,10]]},"reference":[{"issue":"3","key":"33140_CR1","doi-asserted-by":"publisher","first-page":"262","DOI":"10.1136\/BJO.2005.081224","volume":"90","author":"H Quigley","year":"2006","unstructured":"Quigley, H. & Broman, A. T. The number of people with glaucoma worldwide in 2010 and 2020. Br. J. Ophthalmol. 90(3), 262\u2013267. https:\/\/doi.org\/10.1136\/BJO.2005.081224 (2006).","journal-title":"Br. J. Ophthalmol."},{"issue":"11","key":"33140_CR2","doi-asserted-by":"publisher","first-page":"2081","DOI":"10.1016\/J.OPHTHA.2014.05.013","volume":"121","author":"YC Tham","year":"2014","unstructured":"Tham, Y. C. et al. Global prevalence of glaucoma and projections of glaucoma burden through 2040: A systematic review and Meta-Analysis. Ophthalmology 121(11), 2081\u20132090. https:\/\/doi.org\/10.1016\/J.OPHTHA.2014.05.013 (2014).","journal-title":"Ophthalmology"},{"issue":"4","key":"33140_CR3","doi-asserted-by":"publisher","first-page":"640","DOI":"10.1016\/S0161-6420(96)30640-4","volume":"103","author":"MT Nicolela","year":"1996","unstructured":"Nicolela, M. T. & Drance, S. M. Various glaucomatous optic nerve appearances: clinical correlations. Ophthalmology 103(4), 640\u2013649. https:\/\/doi.org\/10.1016\/S0161-6420(96)30640-4 (1996).","journal-title":"Ophthalmology"},{"issue":"18","key":"33140_CR4","doi-asserted-by":"publisher","first-page":"3839","DOI":"10.1038\/s41433-023-02627-4","volume":"37","author":"L Grassi","year":"2023","unstructured":"Grassi, L. et al. Phenotypic expressions of the optic disc in primary open-angle glaucoma. Eye (Basingstoke). 37(18), 3839\u20133846. https:\/\/doi.org\/10.1038\/s41433-023-02627-4 (2023).","journal-title":"Eye (Basingstoke)"},{"issue":"4","key":"33140_CR5","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1016\/S0002-9394(99)80185-8","volume":"125","author":"S Ugurlu","year":"1998","unstructured":"Ugurlu, S., Weitzman, M., Nduaguba, C. & Caprioli, J. Acquired pit of the optic nerve: a risk factor for progression of glaucoma. Am. J. Ophthalmol. 125(4), 457\u2013464. https:\/\/doi.org\/10.1016\/S0002-9394(99)80185-8 (1998).","journal-title":"Am. J. Ophthalmol."},{"issue":"3","key":"33140_CR6","doi-asserted-by":"publisher","first-page":"383","DOI":"10.1016\/j.ophtha.2020.07.045","volume":"128","author":"FA Medeiros","year":"2021","unstructured":"Medeiros, F. A., Jammal, A. A. & Mariottoni, E. B. Detection of progressive glaucomatous optic nerve damage on fundus photographs with deep learning. Ophthalmology 128(3), 383\u2013392. https:\/\/doi.org\/10.1016\/j.ophtha.2020.07.045 (2021).","journal-title":"Ophthalmology"},{"issue":"1","key":"33140_CR7","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1186\/S12938-023-01187-8","volume":"22","author":"X Huang","year":"2023","unstructured":"Huang, X. et al. Artificial intelligence in glaucoma: opportunities, challenges, and future directions. Biomed. Eng. Online. 22(1), 126. https:\/\/doi.org\/10.1186\/S12938-023-01187-8 (2023).","journal-title":"Biomed. Eng. Online"},{"issue":"9","key":"33140_CR8","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1167\/TVST.13.9.9","volume":"13","author":"SW Jin","year":"2024","unstructured":"Jin, S. W., Bouris, E., Morales, E. & Caprioli, J. Long-Term rate of optic disc rim loss in glaucoma patients measured from optic disc photographs with a deep neural network. Transl Vis. Sci. Technol. 13(9), 9\u20139. https:\/\/doi.org\/10.1167\/TVST.13.9.9 (2024).","journal-title":"Transl Vis. Sci. Technol."},{"key":"33140_CR9","doi-asserted-by":"publisher","unstructured":"Bouris, E. et al. A neural network for automated image quality assessment of optic disc photographs. J. Clin. Med. 12(3). https:\/\/doi.org\/10.3390\/JCM12031217 (2023).","DOI":"10.3390\/JCM12031217"},{"issue":"2","key":"33140_CR10","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1136\/BJOPHTHALMOL-2018-313173","volume":"103","author":"DSW Ting","year":"2019","unstructured":"Ting, D. S. W. et al. Artificial intelligence and deep learning in ophthalmology. Br. J. Ophthalmol. 103(2), 167\u2013175. https:\/\/doi.org\/10.1136\/BJOPHTHALMOL-2018-313173 (2019).","journal-title":"Br. J. Ophthalmol."},{"key":"33140_CR11","doi-asserted-by":"publisher","unstructured":"Sergouniotis, P. I. et al. Autoencoder-based phenotyping of ophthalmic images highlights genetic loci influencing retinal morphology and provides informative biomarkers. Peng H, ed. Bioinformatics. ;41(1):btae732. https:\/\/doi.org\/10.1093\/BIOINFORMATICS\/BTAE732 (2024).","DOI":"10.1093\/BIOINFORMATICS\/BTAE732"},{"key":"33140_CR12","doi-asserted-by":"publisher","first-page":"123","DOI":"10.1016\/j.ajo.2019.11.006","volume":"211","author":"AA Jammal","year":"2020","unstructured":"Jammal, A. A. et al. Human versus machine: comparing a deep learning algorithm to human gradings for detecting glaucoma on fundus photographs. Am. J. Ophthalmol. 211, 123\u2013131. https:\/\/doi.org\/10.1016\/j.ajo.2019.11.006 (2020).","journal-title":"Am. J. Ophthalmol."},{"issue":"8","key":"33140_CR13","doi-asserted-by":"publisher","first-page":"1199","DOI":"10.1016\/j.ophtha.2018.01.023","volume":"125","author":"Z Li","year":"2018","unstructured":"Li, Z. et al. Efficacy of a deep learning system for detecting glaucomatous optic neuropathy based on color fundus photographs. Ophthalmology 125(8), 1199\u20131206. https:\/\/doi.org\/10.1016\/j.ophtha.2018.01.023 (2018).","journal-title":"Ophthalmology"},{"key":"33140_CR14","doi-asserted-by":"publisher","unstructured":"Rasheed, H. A. et al. RimNet: A deep neural network pipeline for automated identification of the optic disc rim. Ophthalmol. Sci. 3(1). https:\/\/doi.org\/10.1016\/J.XOPS.2022.100244 (2022).","DOI":"10.1016\/J.XOPS.2022.100244"},{"issue":"12","key":"33140_CR15","doi-asserted-by":"publisher","first-page":"1627","DOI":"10.1016\/j.ophtha.2019.07.024","volume":"126","author":"S Phene","year":"2019","unstructured":"Phene, S. et al. Deep learning and glaucoma specialists: the relative importance of optic disc features to predict glaucoma referral in fundus photographs. Ophthalmology 126(12), 1627\u20131639. https:\/\/doi.org\/10.1016\/j.ophtha.2019.07.024 (2019).","journal-title":"Ophthalmology"},{"issue":"3","key":"33140_CR16","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1001\/JAMAOPHTHALMOL.2013.1926","volume":"131","author":"JY You","year":"2013","unstructured":"You, J. Y. et al. Focal lamina cribrosa defects associated with glaucomatous rim thinning and acquired pits. JAMA Ophthalmol. 131(3), 314\u2013320. https:\/\/doi.org\/10.1001\/JAMAOPHTHALMOL.2013.1926 (2013).","journal-title":"JAMA Ophthalmol."},{"issue":"1","key":"33140_CR17","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1055\/s-2008-1050223","volume":"193","author":"JB Jonas","year":"1988","unstructured":"Jonas, J. B., Gusek, C. G. & Naumann, G. O. H. The parapapillary region in normal and glaucomatous eyes. I. Planimetric values of 312 glaucomatous and 125 normal eyes. Klin. Monbl Augenheilkd. 193(1), 52\u201361. https:\/\/doi.org\/10.1055\/s-2008-1050223 (1988).","journal-title":"Klin. Monbl Augenheilkd"},{"key":"33140_CR18","doi-asserted-by":"publisher","unstructured":"Yamamoto, T., Akahoshi, H. & Kitazawa, S. Emergence of human-like attention and distinct head clusters in self-supervised vision transformers: A comparative eye-tracking study. Neural Netw. 189 https:\/\/doi.org\/10.1016\/j.neunet.2025.107595 (2025).","DOI":"10.1016\/j.neunet.2025.107595"},{"key":"33140_CR19","unstructured":"Grill, J. B. et al. Bootstrap your own latent: A new approach to self-supervised Learning. Adv Neural Inf Process Syst. 2020;2020-December. Accessed June 1, https:\/\/arxiv.org\/pdf\/2006.07733 (2025)."},{"issue":"9","key":"33140_CR20","doi-asserted-by":"publisher","first-page":"2679","DOI":"10.1007\/S11517-025-03353-7","volume":"63","author":"SY Ryu","year":"2025","unstructured":"Ryu, S. Y., Choi, J. Y. & Yoo, T. K. Automated detection of retinal artery occlusion in fundus photography via self-supervised deep learning and multimodal interpretability using a multimodal AI chatbot. Med. Biol. Eng. Comput. 63(9), 2679\u20132691. https:\/\/doi.org\/10.1007\/S11517-025-03353-7 (2025).","journal-title":"Med. Biol. Eng. Comput."},{"key":"33140_CR21","doi-asserted-by":"publisher","unstructured":"Zhou, Y. et al. A foundation model for generalizable disease detection from retinal images. Nature 2023 622:7981. 622(7981):156\u2013163. https:\/\/doi.org\/10.1038\/s41586-023-06555-x (2023).","DOI":"10.1038\/s41586-023-06555-x"},{"key":"33140_CR22","doi-asserted-by":"publisher","unstructured":"Chuter, B. et al. Evaluating a foundation artificial intelligence model for glaucoma detection using color fundus photographs. Ophthalmol. Sci. 5 (1). https:\/\/doi.org\/10.1016\/J.XOPS.2024.100623 (2024).","DOI":"10.1016\/J.XOPS.2024.100623"},{"key":"33140_CR23","doi-asserted-by":"publisher","unstructured":"Kazeminasab, S. et al. An artificial intelligence method for phenotyping of OCT-Derived thickness maps using unsupervised and Self-supervised deep Learning. Journal of imaging informatics in medicine. Published Online May. 20 https:\/\/doi.org\/10.1007\/S10278-025-01539-X (2025).","DOI":"10.1007\/S10278-025-01539-X"},{"issue":"6","key":"33140_CR24","doi-asserted-by":"publisher","first-page":"892","DOI":"10.1038\/S41591-020-0867-7","volume":"26","author":"J Yim","year":"2020","unstructured":"Yim, J. et al. Predicting conversion to wet age-related macular degeneration using deep learning. Nat. Med. 26 (6), 892\u2013899. https:\/\/doi.org\/10.1038\/S41591-020-0867-7 (2020).","journal-title":"Nat. Med."},{"issue":"1","key":"33140_CR25","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/S41746-023-00811-0","volume":"6","author":"SC Huang","year":"2023","unstructured":"Huang, S. C. et al. Self-supervised learning for medical image classification: a systematic review and implementation guidelines. NPJ Digit. Med. 6(1), 1\u201316. https:\/\/doi.org\/10.1038\/S41746-023-00811-0 (2023).","journal-title":"NPJ Digit. Med."},{"key":"33140_CR26","doi-asserted-by":"publisher","unstructured":"Yi, J., Zheng, Y., Ding, D., Zhao, J. & Yang, G. PaRCL: Pathology-aware Representation Contrastive Learning for Glaucoma Classification on Fundus Images. Proceedings \u2013\u20092024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM. Published online 2024:3912\u20133917. Published online 2024:3912\u20133917. https:\/\/doi.org\/10.1109\/BIBM62325.2024.10822360 (2024).","DOI":"10.1109\/BIBM62325.2024.10822360"},{"key":"33140_CR27","doi-asserted-by":"publisher","unstructured":"Deng, J., Zhang, F., Li, S. & Bao, J. Towards Semi-Supervised Segmentation of Retinal Fundus Images via Self-Training. 3rd International Conference on Pattern Recognition and Machine Learning, PRML 2022. Published online 2022:167\u2013172. Published online 2022:167\u2013172. https:\/\/doi.org\/10.1109\/PRML56267.2022.9882204 (2022).","DOI":"10.1109\/PRML56267.2022.9882204"},{"issue":"12","key":"33140_CR28","doi-asserted-by":"publisher","first-page":"3764","DOI":"10.1109\/TMI.2023.3307689","volume":"42","author":"R Fan","year":"2023","unstructured":"Fan, R. et al. One-Vote veto: Semi-Supervised learning for Low-Shot glaucoma diagnosis. IEEE Trans. Med. Imaging. 42(12), 3764\u20133778. https:\/\/doi.org\/10.1109\/TMI.2023.3307689 (2023).","journal-title":"IEEE Trans. Med. Imaging"}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-025-33140-1","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-025-33140-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-025-33140-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,24]],"date-time":"2026-01-24T02:04:26Z","timestamp":1769220266000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-025-33140-1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,10]]},"references-count":28,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["33140"],"URL":"https:\/\/doi.org\/10.1038\/s41598-025-33140-1","relation":{"has-preprint":[{"id-type":"doi","id":"10.21203\/rs.3.rs-7237712\/v1","asserted-by":"object"}]},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,10]]},"assertion":[{"value":"29 July 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 December 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"10 January 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests. Lourdes Grassi, None; Zhe Fei, None; Esteban Morales, None; Joseph Caprioli, None.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"3206"}}