{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T17:33:38Z","timestamp":1776965618014,"version":"3.51.4"},"reference-count":47,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2019,9,20]],"date-time":"2019-09-20T00:00:00Z","timestamp":1568937600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2019,9,20]],"date-time":"2019-09-20T00:00:00Z","timestamp":1568937600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>The global burden of diabetic retinopathy (DR) continues to worsen and DR remains a leading cause of vision loss worldwide. Here, we describe an algorithm to predict DR progression by means of deep learning (DL), using as input color fundus photographs (CFPs) acquired at a single visit from a patient with DR. The proposed DL models were designed to predict future DR progression, defined as 2-step worsening on the Early Treatment Diabetic Retinopathy Diabetic Retinopathy Severity Scale, and were trained against DR severity scores assessed after 6, 12, and 24 months from the baseline visit by masked, well-trained, human reading center graders. The performance of one of these models (prediction at month 12) resulted in an area under the curve equal to 0.79. Interestingly, our results highlight the importance of the predictive signal located in the peripheral retinal fields, not routinely collected for DR assessments, and the importance of microvascular abnormalities. Our findings show the feasibility of predicting future DR progression by leveraging CFPs of a patient acquired at a single visit. Upon further development on larger and more diverse datasets, such an algorithm could enable early diagnosis and referral to a retina specialist for more frequent monitoring and even consideration of early intervention. Moreover, it could also improve patient recruitment for clinical trials targeting DR.<\/jats:p>","DOI":"10.1038\/s41746-019-0172-3","type":"journal-article","created":{"date-parts":[[2019,9,20]],"date-time":"2019-09-20T10:15:48Z","timestamp":1568974548000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":255,"title":["Deep learning algorithm predicts diabetic retinopathy progression in individual patients"],"prefix":"10.1038","volume":"2","author":[{"given":"Filippo","family":"Arcadu","sequence":"first","affiliation":[]},{"given":"Fethallah","family":"Benmansour","sequence":"additional","affiliation":[]},{"given":"Andreas","family":"Maunz","sequence":"additional","affiliation":[]},{"given":"Jeff","family":"Willis","sequence":"additional","affiliation":[]},{"given":"Zdenka","family":"Haskova","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0203-0129","authenticated-orcid":false,"given":"Marco","family":"Prunotto","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2019,9,20]]},"reference":[{"key":"172_CR1","unstructured":"International Diabetes Federation. IDF diabetes atlas: 8th edn. http:\/\/www.diabetesatlas.org\/ (2017)."},{"key":"172_CR2","doi-asserted-by":"publisher","first-page":"S84","DOI":"10.2337\/diacare.27.2007.S84","volume":"27","author":"DS Fong","year":"2004","unstructured":"Fong, D. S. et al. American Diabetes Association. Retinopathy in diabetes. Diabetes Care 27, S84\u2013S87 (2004).","journal-title":"Diabetes Care"},{"key":"172_CR3","unstructured":"Brar, M. & Ho, A. C. Diabetic eye disease: a multidisciplinary discussion on managing diabetic eye disease. EyeNet Magazine Suppl. 1\u20134 (2016)."},{"key":"172_CR4","doi-asserted-by":"publisher","DOI":"10.1136\/bmjdrc-2016-000333","volume":"5","author":"AP Murchison","year":"2017","unstructured":"Murchison, A. P. et al. Nonadherence to eye care in people with diabetes. BMJ Open Diab. Res. Care 5, e000333 (2017).","journal-title":"BMJ Open Diab. Res. Care"},{"key":"172_CR5","doi-asserted-by":"publisher","first-page":"649","DOI":"10.1016\/j.ophtha.2010.08.003","volume":"118","author":"K Mazhar","year":"2011","unstructured":"Mazhar, K. et al. Los Angeles Latino Eye Study Group Severity of diabetic retinopathy and health-related quality of life: the Los Angeles Latino Eye Study. Ophthalmology 118, 649\u2013655 (2011).","journal-title":"Ophthalmology"},{"key":"172_CR6","unstructured":"National Center for Chronic Disease Prevention Health Promotion; Division of Diabetes Translation. National Diabetes Statistics Report. https:\/\/stacks.cdc.gov\/view\/cdc\/23442 (2014)."},{"key":"172_CR7","doi-asserted-by":"publisher","first-page":"926","DOI":"10.1001\/jamaophthalmol.2017.2553","volume":"135","author":"JR Willis","year":"2017","unstructured":"Willis, J. R. et al. Vision-related functional burden of diabetic retinopathy across severity levels in the United States. JAMA Ophthalmol. 135, 926\u2013932 (2017).","journal-title":"JAMA Ophthalmol."},{"key":"172_CR8","unstructured":"Krizhevsky, A., Sutskever, I. & Hinton, G. E. Imagenet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst. https:\/\/papers.nips.cc\/paper\/4824-imagenet-classification-with-deep-convolutional-neural-networks (2012)."},{"key":"172_CR9","doi-asserted-by":"publisher","first-page":"719","DOI":"10.1038\/s41551-018-0305-z","volume":"2","author":"K-H Yu","year":"2018","unstructured":"Yu, K.-H., Beam, A. L. & Kohane, I. S. Artificial intelligence in healthcare. Nat. Biomed. Eng. 2, 719\u2013731 (2018).","journal-title":"Nat. Biomed. Eng."},{"key":"172_CR10","unstructured":"Marr, B. First FDA approval for clinical cloud-based deep learning in healthcare. Forbes. https:\/\/www.forbes.com\/sites\/bernardmarr\/2017\/01\/20\/first-fda-approval-for-clinical-cloud-based-deep-learning-in-healthcare\/#112785c9161c (2017)."},{"key":"172_CR11","doi-asserted-by":"publisher","first-page":"6654","DOI":"10.1118\/1.4967345","volume":"43","author":"RK Samala","year":"2016","unstructured":"Samala, R. K. et al. Mass detection in digital breast tomosynthesis: deep convolutional neural network with transfer learning from mammography. Med. Phys. 43, 6654\u20136666 (2016).","journal-title":"Med. Phys."},{"key":"172_CR12","doi-asserted-by":"publisher","first-page":"574","DOI":"10.1148\/radiol.2017162326","volume":"284","author":"P Lakhani","year":"2017","unstructured":"Lakhani, P. & Sundaram, B. Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology 284, 574\u2013582 (2017).","journal-title":"Radiology"},{"key":"172_CR13","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1038\/nature21056","volume":"542","author":"A Esteva","year":"2017","unstructured":"Esteva, A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 542, 115\u2013118 (2017).","journal-title":"Nature"},{"key":"172_CR14","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","volume":"42","author":"G Litjens","year":"2017","unstructured":"Litjens, G. et al. A survey on deep learning in medical image analysis. Med. Image Anal. 42, 60\u201388 (2017).","journal-title":"Med. Image Anal."},{"key":"172_CR15","doi-asserted-by":"publisher","first-page":"2402","DOI":"10.1001\/jama.2016.17216","volume":"316","author":"V Gulshan","year":"2016","unstructured":"Gulshan, V. et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA 316, 2402\u20132410 (2016).","journal-title":"JAMA"},{"key":"172_CR16","doi-asserted-by":"publisher","first-page":"2211","DOI":"10.1001\/jama.2017.18152","volume":"318","author":"DSW Ting","year":"2017","unstructured":"Ting, D. S. W. et al. Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes. JAMA 318, 2211\u20132223 (2017).","journal-title":"JAMA"},{"key":"172_CR17","unstructured":"U.S. Food and Drug Administration. FDA permits marketing of artificial intelligence-based device to detect certain diabetes-related eye problems. https:\/\/www.fda.gov\/newsevents\/newsroom\/pressannouncements\/ucm604357.htm (2018)."},{"key":"172_CR18","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":"172_CR19","doi-asserted-by":"publisher","first-page":"852","DOI":"10.1167\/iovs.18-25634","volume":"60","author":"F Arcadu","year":"2019","unstructured":"Arcadu, F. et al. Deep learning predicts OCT measures of diabetic macular thickening from color fundus photographs. Invest. Ophthalmol. Vis. Sci. 60, 852\u2013857 (2019).","journal-title":"Invest. Ophthalmol. Vis. Sci."},{"key":"172_CR20","doi-asserted-by":"publisher","first-page":"823","DOI":"10.1016\/S0161-6420(13)38014-2","volume":"98","author":"Early Treatment Diabetic Retinopathy Study Research Group.","year":"1991","unstructured":"Early Treatment Diabetic Retinopathy Study Research Group. Grading diabetic retinopathy from stereoscopic color fundus photographs\u2014an extension of the modified Airlie House classification: ETDRS report number 12. Ophthalmology 98, 823\u2013833 (1991).","journal-title":"Ophthalmology"},{"key":"172_CR21","unstructured":"Lundberg, S. M. & Lee, S. I. A unified approach to interpreting model predictions. Adv. Neural Inf. Process. Syst. https:\/\/papers.nips.cc\/paper\/7062-a-unified-approach-to-interpreting-model-predictions (2017)."},{"key":"172_CR22","unstructured":"Simonyan, K., Vedaldi, A. & Zisserman, A. Deep inside convolutional networks: visualising image classification models and saliency maps. https:\/\/arxiv.org\/abs\/1312.6034 (2014)."},{"key":"172_CR23","unstructured":"Zeiler M.D., Fergus R. in Computer Vision \u2013 ECCV 2014. ECCV 2014. Lecture Notes in Computer Science, vol 8689 (eds Fleet, D. et al.) (Springer, Cham, 2014)."},{"key":"172_CR24","unstructured":"Springenberg, J. T., Dosovitskiy, A., Brox, T. & Riedmiller, M. Striving for simplicity: the all convolutional net. https:\/\/arxiv.org\/abs\/1412.6806 (2015)."},{"key":"172_CR25","doi-asserted-by":"publisher","first-page":"2660","DOI":"10.1109\/TNNLS.2016.2599820","volume":"28","author":"W Samek","year":"2017","unstructured":"Samek, W., Binder, A., Montavon, G., Lapuschkin, S. & M\u00fcller, K. R. Evaluating the visualization of what a deep neural network has learned. IEEE Trans. Neural Netw. Learn. Syst. 28, 2660\u20132673 (2017).","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"172_CR26","doi-asserted-by":"publisher","first-page":"222","DOI":"10.1136\/bjophthalmol-2018-311887","volume":"103","author":"RKR Pappuru","year":"2019","unstructured":"Pappuru, R. K. R., Ribeiro, L., Lobo, C., Alves, D. & Cunha-Vaz, J. Microaneurysm turnover is a predictor of diabetic retinopathy progression. Br. J. Ophthalmol. 103, 222\u2013226 (2019).","journal-title":"Br. J. Ophthalmol."},{"key":"172_CR27","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1016\/j.dss.2017.05.012","volume":"101","author":"S Piri","year":"2017","unstructured":"Piri, S., Delen, D., Liu, T. & Zolbanin, H. M. A data analytics approach to building a clinical decision support system for diabetic retinopathy: developing and deploying a model ensemble. Dec. Support Syst. 101, 12\u201327 (2017).","journal-title":"Dec. Support Syst."},{"key":"172_CR28","doi-asserted-by":"publisher","first-page":"970","DOI":"10.1016\/j.ophtha.2017.02.012","volume":"124","author":"SP Silva","year":"2017","unstructured":"Silva, S. P. et al. Hemorrhage and\/or microaneurysm severity and count in ultrawide field images and Early Treatment Diabetic Retinopathy Study photography. Ophthalmology 124, 970\u2013976 (2017).","journal-title":"Ophthalmology"},{"key":"172_CR29","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1016\/j.visres.2017.02.009","volume":"139","author":"K Ghasemi Falavarjani","year":"2017","unstructured":"Ghasemi Falavarjani, K., Tsui, I. & Sadda, S. R. Ultra-wide-field imaging in diabetic retinopathy. Vis. Res. 139, 187\u2013190 (2017).","journal-title":"Vis. Res."},{"key":"172_CR30","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1001\/jamaophthalmol.2018.4982","volume":"137","author":"LP Aiello","year":"2019","unstructured":"Aiello, L. P. et al. Comparison of Early Treatment Diabetic Retinopathy Study standard 7-field imaging with ultrawide-field imaging for determining severity of diabetic retinopathy. JAMA Ophthalmol. 137, 65\u201373 (2019).","journal-title":"JAMA Ophthalmol."},{"key":"172_CR31","doi-asserted-by":"publisher","first-page":"2013","DOI":"10.1016\/j.ophtha.2013.02.034","volume":"120","author":"DM Brown","year":"2013","unstructured":"Brown, D. M. et al. Long-term outcomes of ranibizumab therapy for diabetic macular edema: the 36-month results from two phase III trials: RISE and RIDE. Ophthalmology 120, 2013\u20132022 (2013).","journal-title":"Ophthalmology"},{"key":"172_CR32","doi-asserted-by":"publisher","first-page":"909","DOI":"10.2337\/diacare.17.8.909","volume":"17","author":"JC Javitt","year":"1994","unstructured":"Javitt, J. C. et al. Preventive eye care in people with diabetes is cost saving to the federal government. Implic. Health Care Reform. Diabetes Care 17, 909\u2013917 (1994).","journal-title":"Implic. Health Care Reform. Diabetes Care"},{"key":"172_CR33","doi-asserted-by":"publisher","first-page":"1145","DOI":"10.1001\/archophthalmol.2012.1043","volume":"130","author":"MS Ip","year":"2012","unstructured":"Ip, M. S., Domalpally, A., Hopkins, J. J., Wong, P. & Ehrlich, J. S. Long-term effects of ranibizumab on diabetic retinopathy severity and progression. Arch. Ophthalmol. 130, 1145\u20131152 (2012).","journal-title":"Arch. Ophthalmol."},{"key":"172_CR34","doi-asserted-by":"publisher","first-page":"998","DOI":"10.1016\/S0161-6420(98)96025-0","volume":"105","author":"SE Moss","year":"1998","unstructured":"Moss, S. E., Klein, R. & Klein, B. E. The 14-year incidence of visual loss in a diabetic population. Ophthalmology 105, 998\u20131003 (1998).","journal-title":"Ophthalmology"},{"key":"172_CR35","doi-asserted-by":"publisher","first-page":"2013","DOI":"10.1016\/j.ophtha.2013.02.034","volume":"120","author":"DM Brown","year":"2013","unstructured":"Brown, D. M. et al. Long-term outcomes of ranibizumab therapy for diabetic macular edema: the 36-month results from two phase III trials: RISE and RIDE. Ophthalmology 120, 2013\u20132022 (2013).","journal-title":"Ophthalmology"},{"key":"172_CR36","doi-asserted-by":"crossref","unstructured":"Royle, P. et al. Pan-retinal photocoagulation and other forms of laser treatment and drug therapies for non-proliferative diabetic retinopathy: systematic review and economic evaluation. Health Technol. Assess. 19, 1\u2013247 (2015).","DOI":"10.3310\/hta19510"},{"key":"172_CR37","doi-asserted-by":"publisher","first-page":"547","DOI":"10.1001\/archopht.119.4.547","volume":"119","author":"R Klein","year":"2001","unstructured":"Klein, R., Klein, B. E. K. & Moss, A. E. How many steps of progression of diabetic retinopathy are meaningful? Arch. Ophthalmol. 119, 547\u2013553 (2001).","journal-title":"Arch. Ophthalmol."},{"key":"172_CR38","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J. & Wojna, Z. Rethinking the inception architecture for computer vision. https:\/\/arxiv.org\/abs\/1512.00567 (2015).","DOI":"10.1109\/CVPR.2016.308"},{"key":"172_CR39","unstructured":"Yosinski, J., Clune, J., Bengio, Y. & Lipson, H. How transferable are features in deep neural networks? Adv. Neural Inf. Process. Syst. https:\/\/papers.nips.cc\/paper\/5347-how-transferable-are-features-in-deep-neural-networks (2014)."},{"key":"172_CR40","doi-asserted-by":"crossref","unstructured":"Deng, J. et al. ImageNet: a large-scale hierarchical image database. CVPR. http:\/\/www.image-net.org\/papers\/imagenet_cvpr09.pdf (2009).","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"172_CR41","unstructured":"Kaggle. Diabetic retinopathy detection. https:\/\/www.kaggle.com\/c\/diabetic-retinopathy-detection (2017)."},{"key":"172_CR42","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","volume":"45","author":"L Breiman","year":"2001","unstructured":"Breiman, L. Random forests. J. Mach. Learn. 45, 5\u201332 (2001).","journal-title":"J. Mach. Learn."},{"key":"172_CR43","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1002\/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3","volume":"3","author":"WJ Youden","year":"1950","unstructured":"Youden, W. J. Index for rating diagnostic tests. Cancer 3, 32\u201335 (1950).","journal-title":"Cancer"},{"key":"172_CR44","volume-title":"Classification and Regression Trees.","author":"L Breiman","year":"1984","unstructured":"Breiman, L., Friedman, J., Stone, C. J. & Olshen, R. A. Classification and Regression Trees. (CRC Press, Boca Raton, 1984)."},{"key":"172_CR45","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2105-8-25","volume":"8","author":"C Strobl","year":"2007","unstructured":"Strobl, C., Boulesteix, A. L., Zeileis, A. & Hothorn, T. Bias in random forest variable importance measures: illustrations, sources and a solution. BMC Bioinforma. 8, 25 (2007).","journal-title":"BMC Bioinforma."},{"key":"172_CR46","unstructured":"Springenberg, J. T., Dosovitskiy, A., Brox, T. & Riedmiller, M. Striving for simplicity: the all convolutional net. https:\/\/arxiv.org\/abs\/1412.6806 (2015)."},{"key":"172_CR47","unstructured":"Simonyan, K., Vedaldi, A. & Zisserman, A. Deep inside convolutional networks: visualising image classification models and saliency maps. https:\/\/arxiv.org\/abs\/1312.6034 (2014)."}],"updated-by":[{"DOI":"10.1038\/s41746-020-00365-5","type":"correction","label":"Correction","source":"publisher","updated":{"date-parts":[[2020,12,8]],"date-time":"2020-12-08T00:00:00Z","timestamp":1607385600000}}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-019-0172-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-019-0172-3","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-019-0172-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,17]],"date-time":"2022-12-17T18:33:33Z","timestamp":1671302013000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-019-0172-3"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,9,20]]},"references-count":47,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2019,12]]}},"alternative-id":["172"],"URL":"https:\/\/doi.org\/10.1038\/s41746-019-0172-3","relation":{"correction":[{"id-type":"doi","id":"10.1038\/s41746-020-00365-5","asserted-by":"object"}]},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,9,20]]},"assertion":[{"value":"15 April 2019","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 September 2019","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"20 September 2019","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 November 2020","order":4,"name":"change_date","label":"Change Date","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Update","order":5,"name":"change_type","label":"Change Type","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"A Correction to this paper has been published:","order":6,"name":"change_details","label":"Change Details","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"https:\/\/doi.org\/10.1038\/s41746-020-00365-5","URL":"https:\/\/doi.org\/10.1038\/s41746-020-00365-5","order":7,"name":"change_details","label":"Change Details","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"F.A., F.B., M.P., and A.M. are employees and shareholders of Roche, Inc. J.W. and Z.H. are employees and shareholders of Genentech, Inc.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"92"}}