{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T05:39:18Z","timestamp":1777354758272,"version":"3.51.4"},"reference-count":43,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T00:00:00Z","timestamp":1767830400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T00:00:00Z","timestamp":1770595200000},"content-version":"vor","delay-in-days":32,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"cience and Technology Agency (JST) Moonshot R&D grant","award":["JPMJMS2025"],"award-info":[{"award-number":["JPMJMS2025"]}]},{"name":"JSPS Core-to-Core Program","award":["JPJSCCA20200006"],"award-info":[{"award-number":["JPJSCCA20200006"]}]},{"DOI":"10.13039\/501100007664","name":"Manpei Suzuki Diabetes Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100007664","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Retinitis pigmentosa (RP) is an inherited progressive retinal degeneration that shows symptoms of night blindness, visual field loss, declining of vision and eventually, blindness. Currently, gene therapy and retinal prosthesis are available, but the indication for these treatments is limited. In this study, we report on the development of a diagnostic and prognostic model for RP based on large-scale deep learning (DL) models pre-trained with fundus images. The EfficientNetB4 model performed best in diagnosing RP with an AUC of 0.94. The diagnosis of RP with this model is superior in cases with good vision. For visual prognosis, we applied machine learning survival analysis to DL-derived image features and clinical metadata, using a strict patient-level split to avoid data leakage. The hybrid model combining imaging and clinical data outperformed models based on either modality alone, especially in female patients. Time-dependent AUC analysis showed that prognostic performance was highest between 500 and 1400 days after examination. SHAP-based interpretability analysis revealed that the features contributing to RP diagnosis and those associated with prognosis were distinct. While our findings demonstrate the added value of fundus images in visual outcome prediction, further validation using external and multi-center datasets is necessary for clinical translation.<\/jats:p>","DOI":"10.1038\/s41746-025-02311-9","type":"journal-article","created":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T22:44:04Z","timestamp":1767912244000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Leveraging large scale deep learning models for diagnosis and visual outcome prediction in retinitis pigmentosa"],"prefix":"10.1038","volume":"9","author":[{"given":"Tatsuya","family":"Nagai","sequence":"first","affiliation":[]},{"given":"Koya","family":"Homma","sequence":"additional","affiliation":[]},{"given":"Yuto","family":"Kawamata","sequence":"additional","affiliation":[]},{"given":"Masahito","family":"Yoshihara","sequence":"additional","affiliation":[]},{"given":"Eiryo","family":"Kawakami","sequence":"additional","affiliation":[]},{"given":"Takayuki","family":"Baba","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2026,1,8]]},"reference":[{"key":"2311_CR1","doi-asserted-by":"publisher","first-page":"1795","DOI":"10.1016\/S0140-6736(06)69740-7","volume":"368","author":"DT Hartong","year":"2006","unstructured":"Hartong, D. T., Berson, E. L. & Dryja, T. P. Retinitis pigmentosa. Lancet 368, 1795\u20131809 (2006).","journal-title":"Lancet"},{"key":"2311_CR2","doi-asserted-by":"publisher","first-page":"849","DOI":"10.1016\/S0140-6736(17)31868-8","volume":"390","author":"S Russell","year":"2017","unstructured":"Russell, S. et al. Efficacy and safety of voretigene neparvovec (AAV2-hRPE65v2) in patients with RPE65-mediated inherited retinal dystrophy: a randomised, controlled, open-label, phase 3 trial. Lancet 390, 849\u2013860 (2017).","journal-title":"Lancet"},{"key":"2311_CR3","doi-asserted-by":"publisher","first-page":"2248","DOI":"10.1016\/j.ophtha.2016.06.049","volume":"123","author":"L da Cruz","year":"2016","unstructured":"da Cruz, L. et al. Five-year safety and performance results from the argus II retinal prosthesis system clinical trial. Ophthalmology 123, 2248\u20132254 (2016).","journal-title":"Ophthalmology"},{"key":"2311_CR4","doi-asserted-by":"publisher","first-page":"268","DOI":"10.1016\/j.oret.2021.12.010","volume":"6","author":"S Nakamura","year":"2022","unstructured":"Nakamura, S. et al. Long-term outcomes of cataract surgery in patients with retinitis pigmentosa. Ophthalmol. Retin. 6, 268\u2013272 (2022).","journal-title":"Ophthalmol. Retin."},{"key":"2311_CR5","doi-asserted-by":"publisher","first-page":"850","DOI":"10.1016\/j.ajo.2005.12.030","volume":"141","author":"S Grover","year":"2006","unstructured":"Grover, S., Apushkin, M. A. & Fishman, G. A. Topical dorzolamide for the treatment of cystoid macular edema in patients with retinitis pigmentosa. Am. J. Ophthalmol. 141, 850\u2013858 (2006).","journal-title":"Am. J. Ophthalmol."},{"key":"2311_CR6","doi-asserted-by":"publisher","DOI":"10.1186\/s12886-019-1169-z","volume":"19","author":"G Garip","year":"2019","unstructured":"Garip, G. & Kamal, A. Systematic review and meta-synthesis of coping with retinitis pigmentosa: implications for improving quality of life. BMC Ophthalmol. 19, 181 (2019).","journal-title":"BMC Ophthalmol."},{"key":"2311_CR7","doi-asserted-by":"publisher","first-page":"967","DOI":"10.1038\/s42256-025-01040-8","volume":"7","author":"N Pontikos","year":"2025","unstructured":"Pontikos, N. et al. Next-generation phenotyping of inherited retinal diseases from multimodal imaging with Eye2Gene. Nat. Mach. Intell. 7, 967\u2013978 (2025).","journal-title":"Nat. Mach. Intell."},{"key":"2311_CR8","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1038\/s41746-017-0012-2","volume":"1","author":"AL Fogel","year":"2018","unstructured":"Fogel, A. L. & Kvedar, J. C. Artificial intelligence powers digital medicine. npj Digit. Med. 1, 5 (2018).","journal-title":"npj Digit. Med."},{"key":"2311_CR9","first-page":"23","volume":"283","author":"D Muller","year":"2021","unstructured":"Muller, D., Soto-Rey, I. & Kramer, F. Multi-disease detection in retinal imaging based on ensembling heterogeneous deep learning models. Stud. Health Technol. Inform. 283, 23\u201331 (2021).","journal-title":"Stud. Health Technol. Inform."},{"key":"2311_CR10","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1038\/s41746-018-0040-6","volume":"1","author":"MD Abramoff","year":"2018","unstructured":"Abramoff, M. D., Lavin, P. T., Birch, M., Shah, N. & Folk, J. C. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. npj Digit. Med. 1, 39 (2018).","journal-title":"npj Digit. Med."},{"key":"2311_CR11","doi-asserted-by":"publisher","first-page":"e000569","DOI":"10.1136\/bmjophth-2020-000569","volume":"5","author":"J Bridge","year":"2020","unstructured":"Bridge, J., Harding, S. & Zheng, Y. Development and validation of a novel prognostic model for predicting AMD progression using longitudinal fundus images. BMJ Open Ophthalmol. 5, e000569 (2020).","journal-title":"BMJ Open Ophthalmol."},{"key":"2311_CR12","doi-asserted-by":"publisher","first-page":"e6900","DOI":"10.7717\/peerj.6900","volume":"7","author":"H Masumoto","year":"2019","unstructured":"Masumoto, H. et al. Accuracy of a deep convolutional neural network in detection of retinitis pigmentosa on ultrawide-field images. PeerJ 7, e6900 (2019).","journal-title":"PeerJ"},{"key":"2311_CR13","doi-asserted-by":"publisher","first-page":"948","DOI":"10.1007\/s10278-021-00479-6","volume":"34","author":"TC Chen","year":"2021","unstructured":"Chen, T. C. et al. Artificial Intelligence-Assisted Early Detection of Retinitis Pigmentosa - the Most Common Inherited Retinal Degeneration. J. Digit. Imaging 34, 948\u2013958 (2021).","journal-title":"J. Digit. Imaging"},{"key":"2311_CR14","doi-asserted-by":"publisher","first-page":"1272","DOI":"10.1136\/bjophthalmol-2020-318544","volume":"105","author":"Y Fujinami-Yokokawa","year":"2021","unstructured":"Fujinami-Yokokawa, Y. et al. Prediction of causative genes in inherited retinal disorder from fundus photography and autofluorescence imaging using deep learning techniques. Br. J. Ophthalmol. 105, 1272\u20131279 (2021).","journal-title":"Br. J. Ophthalmol."},{"key":"2311_CR15","doi-asserted-by":"publisher","unstructured":"Peng, Y. et al. Predicting risk of late age-related macular degeneration using deep learning. npj Digital Medicine 3, https:\/\/doi.org\/10.1038\/s41746-020-00317-z (2020).","DOI":"10.1038\/s41746-020-00317-z"},{"key":"2311_CR16","doi-asserted-by":"publisher","unstructured":"Holste, G. et al. Harnessing the power of longitudinal medical imaging for eye disease prognosis using Transformer-based sequence modeling. npj Digit. Med. 7, https:\/\/doi.org\/10.1038\/s41746-024-01207-4 (2024).","DOI":"10.1038\/s41746-024-01207-4"},{"key":"2311_CR17","doi-asserted-by":"publisher","unstructured":"Tan, M. & Le, Q. V. EfficientNet: rethinking model scaling for convolutional neural networks. ArXiv https:\/\/doi.org\/10.48550\/arXiv.1905.11946 Focus to learn more (2019).","DOI":"10.48550\/arXiv.1905.11946"},{"key":"2311_CR18","doi-asserted-by":"publisher","DOI":"10.1136\/bjophthalmol-2020-316290","author":"B Li","year":"2021","unstructured":"Li, B. et al. Development and evaluation of a deep learning model for the detection of multiple fundus diseases based on colour fundus photography. Br. J. Ophthalmol. https:\/\/doi.org\/10.1136\/bjophthalmol-2020-316290 (2021).","journal-title":"Br. J. Ophthalmol."},{"key":"2311_CR19","doi-asserted-by":"publisher","unstructured":"Moon, S. et al. Prediction of anti-vascular endothelial growth factor agent-specific treatment outcomes in neovascular age-related macular degeneration using a generative adversarial network. Sci. Rep. 13, https:\/\/doi.org\/10.1038\/s41598-023-32398-7 (2023).","DOI":"10.1038\/s41598-023-32398-7"},{"key":"2311_CR20","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1016\/S0161-6420(00)00413-9","volume":"108","author":"JP Szlyk","year":"2001","unstructured":"Szlyk, J. P. et al. Perceived and actual performance of daily tasks: relationship to visual function tests in individuals with retinitis pigmentosa. Ophthalmology 108, 65\u201375 (2001).","journal-title":"Ophthalmology"},{"key":"2311_CR21","doi-asserted-by":"publisher","first-page":"363","DOI":"10.1007\/BF00157066","volume":"70","author":"UL Nilsson","year":"1988","unstructured":"Nilsson, U. L. Visual rehabilitation of patients with advanced stages of glaucoma, optic atrophy, myopia or retinitis pigmentosa. Doc. Ophthalmol. 70, 363\u2013383 (1988).","journal-title":"Doc. Ophthalmol."},{"key":"2311_CR22","doi-asserted-by":"publisher","first-page":"e25165","DOI":"10.2196\/25165","volume":"9","author":"BK Betzler","year":"2021","unstructured":"Betzler, B. K. et al. Gender prediction for a multiethnic population via deep learning across different retinal fundus photograph fields: retrospective cross-sectional study. JMIR Med. Inform. 9, e25165 (2021).","journal-title":"JMIR Med. Inform."},{"key":"2311_CR23","doi-asserted-by":"publisher","first-page":"158","DOI":"10.1038\/s41551-018-0195-0","volume":"2","author":"R Poplin","year":"2018","unstructured":"Poplin, R. et al. Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning. Nat. Biomed. Eng. 2, 158\u2013164 (2018).","journal-title":"Nat. Biomed. Eng."},{"key":"2311_CR24","doi-asserted-by":"publisher","first-page":"9432","DOI":"10.1038\/s41598-020-65794-4","volume":"10","author":"N Gerrits","year":"2020","unstructured":"Gerrits, N. et al. Age and sex affect deep learning prediction of cardiometabolic risk factors from retinal images. Sci. Rep. 10, 9432 (2020).","journal-title":"Sci. Rep."},{"key":"2311_CR25","doi-asserted-by":"publisher","first-page":"1484","DOI":"10.1136\/bjo-2021-320897","volume":"107","author":"TYA Liu","year":"2023","unstructured":"Liu, T. Y. A. et al. Prediction of visual impairment in retinitis pigmentosa using deep learning and multimodal fundus images. Br. J. Ophthalmol. 107, 1484\u20131489 (2023).","journal-title":"Br. J. Ophthalmol."},{"key":"2311_CR26","doi-asserted-by":"publisher","first-page":"e071043","DOI":"10.1136\/bmjopen-2022-071043","volume":"13","author":"Q Nguyen","year":"2023","unstructured":"Nguyen, Q. et al. Can artificial intelligence accelerate the diagnosis of inherited retinal diseases? Protocol for a data-only retrospective cohort study (Eye2Gene). BMJ Open 13, e071043 (2023).","journal-title":"BMJ Open"},{"key":"2311_CR27","doi-asserted-by":"publisher","first-page":"100652","DOI":"10.1016\/j.xops.2024.100652","volume":"5","author":"WA Woof","year":"2025","unstructured":"Woof, W. A. et al. Quantification of fundus autofluorescence features in a molecularly characterized cohort of >3500 patients with inherited retinal disease from the United Kingdom. Ophthalmol. Sci. 5, 100652 (2025).","journal-title":"Ophthalmol. Sci."},{"key":"2311_CR28","unstructured":"Dosovitskiy, A. et al. An image is worth 16x16 words: transformers for image recognition at scale. ArXiv https:\/\/arxiv.org\/abs\/2010.11929 (2020)."},{"key":"2311_CR29","doi-asserted-by":"publisher","first-page":"110979","DOI":"10.1016\/j.dib.2024.110979","volume":"57","author":"S Sharmin","year":"2024","unstructured":"Sharmin, S. et al. A dataset of color fundus images for the detection and classification of eye diseases. Data Brief. 57, 110979 (2024).","journal-title":"Data Brief."},{"key":"2311_CR30","doi-asserted-by":"publisher","first-page":"29","DOI":"10.3390\/data8020029","volume":"8","author":"S Panchal","year":"2023","unstructured":"Panchal, S. et al. Retinal fundus multi-disease image dataset (RFMiD) 2.0: a dataset of frequently and rarely identified diseases. Data 8, 29 (2023).","journal-title":"Data"},{"key":"2311_CR31","doi-asserted-by":"publisher","first-page":"1236","DOI":"10.1167\/iovs.05-0981","volume":"47","author":"K Schulze-Bonsel","year":"2006","unstructured":"Schulze-Bonsel, K., Feltgen, N., Burau, H., Hansen, L. & Bach, M. Visual acuities \u201chand motion\u201d and \u201ccounting fingers\u201d can be quantified with the freiburg visual acuity test. Invest. Ophthalmol. Vis. Sci. 47, 1236\u20131240 (2006).","journal-title":"Invest. Ophthalmol. Vis. Sci."},{"key":"2311_CR32","unstructured":"He, K., Zhang, X., Ren, S. & Sun, J. in IEEE conference on computer vision and pattern recognition. (IEEE, 2016)."},{"key":"2311_CR33","unstructured":"Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J. & Wojna, Z. in IEEE conference on computer vision and pattern recognition."},{"key":"2311_CR34","unstructured":"Huang, G., Liu, Z., Van Der Maaten, L. & Weinberger, K. Q. in IEEE conference on computer vision and pattern recognition."},{"key":"2311_CR35","unstructured":"Tan, M. & Le, Q. in Proceedings of the 36th International Conference on Machine Learning Vol. 97 (eds Chaudhuri Kamalika & Salakhutdinov Ruslan) 6105--6114 (PMLR, 2019)."},{"key":"2311_CR36","doi-asserted-by":"publisher","first-page":"14","DOI":"10.3390\/data6020014","volume":"6","author":"S Pachade","year":"2021","unstructured":"Pachade, S. et al. Retinal fundus multi-disease image dataset (RFMiD): a dataset for multi-disease detection research. Data 6, 14 (2021).","journal-title":"Data"},{"key":"2311_CR37","doi-asserted-by":"publisher","first-page":"841","DOI":"10.1214\/08-AOAS169","volume":"2","author":"H Ishwaran","year":"2008","unstructured":"Ishwaran, H., Kogalur, U. B., Blackstone, E. H. & Lauer, M. S. Random survival forests. Ann. Appl. Stat. 2, 841\u2013860 (2008).","journal-title":"Ann. Appl. Stat."},{"key":"2311_CR38","first-page":"6","volume":"21","author":"P Sebastian","year":"2020","unstructured":"Sebastian, P. scikit-survival: a library for time-to-event analysis built on top of SCIKIT-learn. J. Mach. Learn. Res. 21, 6 (2020).","journal-title":"J. Mach. Learn. Res."},{"key":"2311_CR39","unstructured":"Lundberg, S., Lee, S. A unified approach to interpreting model predictions. ArXiv https:\/\/arxiv.org\/abs\/1705.07874 (2017)."},{"key":"2311_CR40","doi-asserted-by":"publisher","unstructured":"Krzyzi\u0144ski, M., Spytek, M., Baniecki, H. & Biecek, P. SurvSHAP(t): time-dependent explanations of machine learning survival models. Knowledge-Based Systems 262, https:\/\/doi.org\/10.1016\/j.knosys.2022.110234 (2023).","DOI":"10.1016\/j.knosys.2022.110234"},{"key":"2311_CR41","doi-asserted-by":"publisher","first-page":"305","DOI":"10.1001\/jamaophthalmol.2022.6393","volume":"141","author":"D Nagasato","year":"2023","unstructured":"Nagasato, D. et al. Estimation of visual function using deep learning from ultra-widefield fundus images of eyes with retinitis pigmentosa. JAMA Ophthalmol. 141, 305 (2023).","journal-title":"JAMA Ophthalmol."},{"key":"2311_CR42","doi-asserted-by":"crossref","unstructured":"Sumaroka, A. et al. Treatment potential for macular cone vision in leber congenital amaurosis Due toCEP290orNPHP5Mutations: predictions from artificial intelligence. Investig. Ophthalmol. Vis. Sci. 60, 2551 (2019).","DOI":"10.1167\/iovs.19-27156"},{"key":"2311_CR43","doi-asserted-by":"publisher","first-page":"446","DOI":"10.1111\/ceo.14234","volume":"51","author":"SH Yassin","year":"2023","unstructured":"Yassin, S. H. et al. Efficacy and accuracy of artificial intelligence to overlay multimodal images from different optical instruments in patients with retinitis pigmentosa. Clin. Exp. Ophthalmol. 51, 446\u2013452 (2023).","journal-title":"Clin. Exp. Ophthalmol."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-02311-9","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-02311-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-02311-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T11:21:33Z","timestamp":1770636093000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-02311-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,8]]},"references-count":43,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["2311"],"URL":"https:\/\/doi.org\/10.1038\/s41746-025-02311-9","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,8]]},"assertion":[{"value":"26 November 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"21 December 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 January 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Author E.K. acts as a paid consultant for THE PHAGE Inc. but declares no competing interests related to this study. The other authors declare no financial or non-financial competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"137"}}