{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T06:28:16Z","timestamp":1776320896055,"version":"3.50.1"},"reference-count":61,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,6,13]],"date-time":"2023-06-13T00:00:00Z","timestamp":1686614400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,6,13]],"date-time":"2023-06-13T00:00:00Z","timestamp":1686614400000},"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>A plethora of classification models for the detection of glaucoma from fundus images have been proposed in recent years. Often trained with data from a single glaucoma clinic, they report impressive performance on internal test sets, but tend to struggle in generalizing to external sets. This performance drop can be attributed to data shifts in glaucoma prevalence, fundus camera, and the definition of glaucoma ground truth. In this study, we confirm that a previously described regression network for glaucoma referral (G-RISK) obtains excellent results in a variety of challenging settings. Thirteen different data sources of labeled fundus images were utilized. The data sources include two large population cohorts (Australian Blue Mountains Eye Study, BMES and German Gutenberg Health Study, GHS) and 11 publicly available datasets (AIROGS, ORIGA, REFUGE1, LAG, ODIR, REFUGE2, GAMMA, RIM-ONEr3, RIM-ONE DL, ACRIMA, PAPILA). To minimize data shifts in input data, a standardized image processing strategy was developed to obtain 30\u00b0 disc-centered images from the original data. A total of 149,455 images were included for model testing. Area under the receiver operating characteristic curve (AUC) for BMES and GHS population cohorts were at 0.976 [95% CI: 0.967\u20130.986] and 0.984 [95% CI: 0.980\u20130.991] on participant level, respectively. At a fixed specificity of 95%, sensitivities were at 87.3% and 90.3%, respectively, surpassing the minimum criteria of 85% sensitivity recommended by Prevent Blindness America. AUC values on the eleven publicly available data sets ranged from 0.854 to 0.988. These results confirm the excellent generalizability of a glaucoma risk regression model trained with homogeneous data from a single tertiary referral center. Further validation using prospective cohort studies is warranted.<\/jats:p>","DOI":"10.1038\/s41746-023-00857-0","type":"journal-article","created":{"date-parts":[[2023,6,13]],"date-time":"2023-06-13T19:02:06Z","timestamp":1686682926000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":73,"title":["A generalizable deep learning regression model for automated glaucoma screening from fundus images"],"prefix":"10.1038","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3214-4247","authenticated-orcid":false,"given":"Ruben","family":"Hemelings","sequence":"first","affiliation":[]},{"given":"Bart","family":"Elen","sequence":"additional","affiliation":[]},{"given":"Alexander K.","family":"Schuster","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2640-181X","authenticated-orcid":false,"given":"Matthew B.","family":"Blaschko","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7816-816X","authenticated-orcid":false,"given":"Jo\u00e3o","family":"Barbosa-Breda","sequence":"additional","affiliation":[]},{"given":"Pekko","family":"Hujanen","sequence":"additional","affiliation":[]},{"given":"Annika","family":"Junglas","sequence":"additional","affiliation":[]},{"given":"Stefan","family":"Nickels","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6931-5797","authenticated-orcid":false,"given":"Andrew","family":"White","sequence":"additional","affiliation":[]},{"given":"Norbert","family":"Pfeiffer","sequence":"additional","affiliation":[]},{"given":"Paul","family":"Mitchell","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5197-8215","authenticated-orcid":false,"given":"Patrick","family":"De Boever","sequence":"additional","affiliation":[]},{"given":"Anja","family":"Tuulonen","sequence":"additional","affiliation":[]},{"given":"Ingeborg","family":"Stalmans","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,6,13]]},"reference":[{"key":"857_CR1","doi-asserted-by":"publisher","first-page":"2081","DOI":"10.1016\/j.ophtha.2014.05.013","volume":"121","author":"Y-C 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, 2081\u20132090 (2014).","journal-title":"Ophthalmology"},{"key":"857_CR2","doi-asserted-by":"publisher","first-page":"1661","DOI":"10.1016\/S0161-6420(96)30449-1","volume":"103","author":"P Mitchell","year":"1996","unstructured":"Mitchell, P., Smith, W., Attebo, K. & Healey, P. R. Prevalence of open-angle glaucoma in australia: the blue mountains eye study. Ophthalmology 103, 1661\u20131669 (1996).","journal-title":"Ophthalmology"},{"key":"857_CR3","doi-asserted-by":"publisher","first-page":"511","DOI":"10.1016\/j.ajo.2007.06.029","volume":"144","author":"F Topouzis","year":"2007","unstructured":"Topouzis, F. et al. Prevalence of open-angle glaucoma in Greece: the Thessaloniki Eye Study. Am. J. Ophthalmol. 144, 511\u2013519 (2007).","journal-title":"Am. J. Ophthalmol."},{"key":"857_CR4","doi-asserted-by":"publisher","first-page":"651","DOI":"10.1001\/jamaophthalmol.2013.1686","volume":"131","author":"DL Budenz","year":"2013","unstructured":"Budenz, D. L. et al. Prevalence of glaucoma in an urban West African population: The tema eye survey. JAMA Ophthalmol. 131, 651\u2013658 (2013).","journal-title":"JAMA Ophthalmol."},{"key":"857_CR5","doi-asserted-by":"publisher","first-page":"8250","DOI":"10.1167\/iovs.11-7472","volume":"52","author":"YB Liang","year":"2011","unstructured":"Liang, Y. B. et al. Prevalence of primary open angle glaucoma in a rural adult Chinese population: the Handan eye study. Invest. Ophthalmol. Vis. Sci. 52, 8250\u20138257 (2011).","journal-title":"Invest. Ophthalmol. Vis. Sci."},{"key":"857_CR6","first-page":"1","volume":"11","author":"JM Burr","year":"2007","unstructured":"Burr, J. M. et al. The clinical effectiveness and cost-effectiveness of screening for open angle glaucoma: a systematic review and economic evaluation. Health Technol. Assess. Winch. Engl. 11, 1\u2013190 (2007).","journal-title":"Health Technol. Assess. Winch. Engl."},{"key":"857_CR7","doi-asserted-by":"publisher","first-page":"200","DOI":"10.1111\/aos.13912","volume":"97","author":"E Karvonen","year":"2019","unstructured":"Karvonen, E. et al. Prevalence of glaucoma in the Northern Finland Birth Cohort Eye Study. Acta Ophthalmol. (Copenh.) 97, 200\u2013207 (2019).","journal-title":"Acta Ophthalmol. (Copenh.)"},{"key":"857_CR8","doi-asserted-by":"publisher","first-page":"1399","DOI":"10.1136\/bjophthalmol-2019-314795","volume":"104","author":"E Karvonen","year":"2020","unstructured":"Karvonen, E. et al. Diagnostic performance of modern imaging instruments in glaucoma screening. Br. J. Ophthalmol. 104, 1399\u20131405 (2020).","journal-title":"Br. J. Ophthalmol."},{"key":"857_CR9","doi-asserted-by":"publisher","first-page":"1024","DOI":"10.1016\/j.ophtha.2010.10.016","volume":"118","author":"C Kim","year":"2011","unstructured":"Kim, C., Seong, G. J., Lee, N. & Song, K., Namil Study Group, Korean Glaucoma Society. Prevalence of primary open-angle glaucoma in central South Korea the Namil study. Ophthalmology 118, 1024\u20131030 (2011).","journal-title":"Ophthalmology"},{"key":"857_CR10","doi-asserted-by":"publisher","first-page":"851","DOI":"10.1097\/IJG.0000000000001899","volume":"30","author":"P Founti","year":"2021","unstructured":"Founti, P. et al. Twelve-year incidence of open-angle glaucoma: The Thessaloniki eye study. J. Glaucoma 30, 851\u2013858 (2021).","journal-title":"J. Glaucoma"},{"key":"857_CR11","doi-asserted-by":"publisher","first-page":"1187","DOI":"10.1001\/archopht.118.9.1187","volume":"118","author":"JL Keltner","year":"2000","unstructured":"Keltner, J. L. et al. Confirmation of visual field abnormalities in the ocular hypertension treatment study. ocular hypertension treatment study group. Arch. Ophthalmol. Chic. Ill. 1960 118, 1187\u20131194 (2000).","journal-title":"Arch. Ophthalmol. Chic. Ill. 1960"},{"key":"857_CR12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.preteyeres.2018.07.004","volume":"67","author":"U Schmidt-Erfurth","year":"2018","unstructured":"Schmidt-Erfurth, U., Sadeghipour, A., Gerendas, B. S., Waldstein, S. M. & Bogunovi\u0107, H. Artificial intelligence in retina. Prog. Retin. Eye Res. 67, 1\u201329 (2018).","journal-title":"Prog. Retin. Eye Res."},{"key":"857_CR13","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1016\/j.ophtha.2018.12.033","volume":"126","author":"FA Medeiros","year":"2019","unstructured":"Medeiros, F. A., Jammal, A. A. & Thompson, A. C. From machine to machine: An OCT-trained deep learning algorithm for objective quantification of glaucomatous damage in fundus photographs. Ophthalmology 126, 513\u2013521 (2019).","journal-title":"Ophthalmology"},{"key":"857_CR14","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-021-99605-1","volume":"11","author":"R Hemelings","year":"2021","unstructured":"Hemelings, R. et al. Deep learning on fundus images detects glaucoma beyond the optic disc. Sci. Rep. 11, 20313 (2021).","journal-title":"Sci. Rep."},{"key":"857_CR15","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, 1199\u20131206 (2018).","journal-title":"Ophthalmology"},{"key":"857_CR16","doi-asserted-by":"crossref","unstructured":"Li, L., Xu, M., Wang, X., Jiang, L. & Liu, H. Attention based glaucoma detection: A large-scale database and CNN model. Preprint at http:\/\/arxiv.org\/abs\/1903.10831 (2019).","DOI":"10.1109\/CVPR.2019.01082"},{"key":"857_CR17","doi-asserted-by":"publisher","first-page":"101570","DOI":"10.1016\/j.media.2019.101570","volume":"59","author":"JI Orlando","year":"2020","unstructured":"Orlando, J. I. et al. REFUGE challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs. Med. Image Anal. 59, 101570 (2020).","journal-title":"Med. Image Anal."},{"key":"857_CR18","doi-asserted-by":"crossref","unstructured":"Beede, E. et al. A Human-Centered Evaluation of a Deep Learning System Deployed in Clinics for the Detection of Diabetic Retinopathy. in Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems 1\u201312 (Association for Computing Machinery, 2020).","DOI":"10.1145\/3313831.3376718"},{"key":"857_CR19","doi-asserted-by":"crossref","unstructured":"Qui\u00f1onero-Candela, J., Sugiyama, M., Schwaighofer, A. & Lawrence, N. D. When Training and Test Sets Are Different: Characterizing Learning Transfer. in Dataset Shift in Machine Learning 3\u201328 (MIT Press, 2009).","DOI":"10.7551\/mitpress\/9780262170055.001.0001"},{"key":"857_CR20","doi-asserted-by":"publisher","first-page":"346","DOI":"10.1364\/BOE.379978","volume":"11","author":"D Romo-Bucheli","year":"2019","unstructured":"Romo-Bucheli, D. et al. Reducing image variability across OCT devices with unsupervised unpaired learning for improved segmentation of retina. Biomed. Opt. Expr. 11, 346\u2013363 (2019).","journal-title":"Biomed. Opt. Expr."},{"key":"857_CR21","doi-asserted-by":"publisher","first-page":"101654","DOI":"10.1016\/j.media.2020.101654","volume":"61","author":"Y Shen","year":"2020","unstructured":"Shen, Y. et al. Domain-invariant interpretable fundus image quality assessment. Med. Image Anal. 61, 101654 (2020).","journal-title":"Med. Image Anal."},{"key":"857_CR22","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1016\/j.ajo.2019.01.011","volume":"201","author":"AC Thompson","year":"2019","unstructured":"Thompson, A. C., Jammal, A. A. & Medeiros, F. A. A deep learning algorithm to quantify neuroretinal rim loss from optic disc photographs. Am. J. Ophthalmol. 201, 9\u201318 (2019).","journal-title":"Am. J. Ophthalmol."},{"key":"857_CR23","doi-asserted-by":"publisher","first-page":"1695","DOI":"10.1007\/s00417-018-4011-z","volume":"256","author":"R H\u00f6hn","year":"2018","unstructured":"H\u00f6hn, R. et al. Prevalence of glaucoma in Germany: results from the Gutenberg Health Study. Graefes Arch. Clin. Exp. Ophthalmol. Albrecht Von. Graefes Arch. Klin. Exp. Ophthalmol. 256, 1695\u20131702 (2018).","journal-title":"Graefes Arch. Clin. Exp. Ophthalmol. Albrecht Von. Graefes Arch. Klin. Exp. Ophthalmol."},{"key":"857_CR24","doi-asserted-by":"publisher","first-page":"699","DOI":"10.1016\/S0002-9394(14)72723-0","volume":"120","author":"WE Sponsel","year":"1995","unstructured":"Sponsel, W. E. et al. Prevent Blindness America visual field screening study. The Prevent Blindness America Glaucoma Advisory Committee. Am. J. Ophthalmol. 120, 699\u2013708 (1995).","journal-title":"Am. J. Ophthalmol."},{"key":"857_CR25","doi-asserted-by":"publisher","first-page":"1667","DOI":"10.1016\/j.ophtha.2010.07.001","volume":"117","author":"PR Healey","year":"2010","unstructured":"Healey, P. R., Lee, A. J., Aung, T., Wong, T. Y. & Mitchell, P. Diagnostic accuracy of the Heidelberg retina tomograph for glaucoma: A population-based assessment. Ophthalmology 117, 1667\u20131673 (2010).","journal-title":"Ophthalmology"},{"key":"857_CR26","doi-asserted-by":"publisher","first-page":"1353","DOI":"10.1001\/jamaophthalmol.2019.3501","volume":"137","author":"H Liu","year":"2019","unstructured":"Liu, H. et al. Development and validation of a deep learning system to detect glaucomatous optic neuropathy using fundus photographs. JAMA Ophthalmol. 137, 1353\u20131360 (2019).","journal-title":"JAMA Ophthalmol."},{"key":"857_CR27","doi-asserted-by":"publisher","first-page":"1217","DOI":"10.1016\/j.ajhg.2021.05.004","volume":"108","author":"B Alipanahi","year":"2021","unstructured":"Alipanahi, B. et al. Large-scale machine-learning-based phenotyping significantly improves genomic discovery for optic nerve head morphology. Am. J. Hum. Genet. 108, 1217\u20131230 (2021).","journal-title":"Am. J. Hum. Genet."},{"key":"857_CR28","first-page":"1006","volume":"62","author":"R Fan","year":"2021","unstructured":"Fan, R. et al. Deep learning for detecting glaucoma in the Ocular Hypertension Treatment Study: Implications for clinical trial endpoints. Invest. Ophthalmol. Vis. Sci. 62, 1006 (2021).","journal-title":"Invest. Ophthalmol. Vis. Sci."},{"key":"857_CR29","doi-asserted-by":"publisher","first-page":"701","DOI":"10.1001\/archopht.120.6.701","volume":"120","author":"MA Kass","year":"2002","unstructured":"Kass, M. A. et al. The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma. Arch. Ophthalmol. Chic. Ill. 1960 120, 701\u2013713 (2002).","journal-title":"Arch. Ophthalmol. Chic. Ill. 1960"},{"key":"857_CR30","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1167\/tvst.9.2.27","volume":"9","author":"M Christopher","year":"2020","unstructured":"Christopher, M. et al. Effects of study population, labeling and training on glaucoma detection using deep learning algorithms. Transl. Vis. Sci. Technol. 9, 27 (2020).","journal-title":"Transl. Vis. Sci. Technol."},{"key":"857_CR31","unstructured":"M\u00fcller, R., Kornblith, S. & Hinton, G. When does label smoothing help? In: Advances in Neural Information Processing Systems, Vol. 32 (eds. Wallach, H. et al.) (Curran Associates, Inc., 2019)."},{"key":"857_CR32","doi-asserted-by":"publisher","first-page":"102038","DOI":"10.1016\/j.media.2021.102038","volume":"71","author":"C Gros","year":"2021","unstructured":"Gros, C., Lemay, A. & Cohen-Adad, J. SoftSeg: Advantages of soft versus binary training for image segmentation. Med. Image Anal. 71, 102038 (2021).","journal-title":"Med. Image Anal."},{"key":"857_CR33","first-page":"4539","volume":"61","author":"T Estrela","year":"2020","unstructured":"Estrela, T. et al. From machine to the real world: assessing the accuracy of a machine-to-machine (M2M) deep learning model to detect glaucoma during a population-based screening effort in Brazil. Invest. Ophthalmol. Vis. Sci. 61, 4539 (2020).","journal-title":"Invest. Ophthalmol. Vis. Sci."},{"key":"857_CR34","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 (2020).","journal-title":"Am. J. Ophthalmol."},{"key":"857_CR35","doi-asserted-by":"publisher","first-page":"1473","DOI":"10.1038\/eye.2012.204","volume":"26","author":"J Lee","year":"2012","unstructured":"Lee, J., Kim, J. & Kee, C. Characteristics of patients with a localized retinal nerve fiber layer defect and normal optic disc appearance. Eye 26, 1473\u20131478 (2012).","journal-title":"Eye"},{"key":"857_CR36","doi-asserted-by":"publisher","first-page":"759","DOI":"10.1007\/BF00184280","volume":"232","author":"JB Jonas","year":"1994","unstructured":"Jonas, J. B. & Schiro, D. Localized retinal nerve fiber layer defects in nonglaucomatous optic nerve atrophy. Graefes Arch. Clin. Exp. Ophthalmol. Albrecht Von. Graefes Arch. Klin. Exp. Ophthalmol. 232, 759\u2013760 (1994).","journal-title":"Graefes Arch. Clin. Exp. Ophthalmol. Albrecht Von. Graefes Arch. Klin. Exp. Ophthalmol."},{"key":"857_CR37","doi-asserted-by":"publisher","first-page":"1147","DOI":"10.1016\/S0161-6420(93)31513-7","volume":"100","author":"E Chihara","year":"1993","unstructured":"Chihara, E., Matsuoka, T., Ogura, Y. & Matsumura, M. Retinal nerve fiber layer defect as an early manifestation of diabetic retinopathy. Ophthalmology 100, 1147\u20131151 (1993).","journal-title":"Ophthalmology"},{"key":"857_CR38","doi-asserted-by":"publisher","first-page":"1651","DOI":"10.1161\/STROKEAHA.113.004629","volume":"45","author":"D Wang","year":"2014","unstructured":"Wang, D. et al. Localized retinal nerve fiber layer defects and stroke. Stroke 45, 1651\u20131656 (2014).","journal-title":"Stroke"},{"key":"857_CR39","doi-asserted-by":"publisher","unstructured":"Ronneberger, O., Fischer, P. & Brox, T. U-Net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention \u2013 MICCAI 2015 (eds. Navab, N., Hornegger, J., Wells, W. M. & Frangi, A. F.) 234\u2013241 (Springer International Publishing, 2015). https:\/\/doi.org\/10.1007\/978-3-319-24574-4_28.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"857_CR40","doi-asserted-by":"publisher","first-page":"1597","DOI":"10.1109\/TMI.2018.2791488","volume":"37","author":"H Fu","year":"2018","unstructured":"Fu, H. et al. Joint optic disc and cup segmentation based on multi-label deep network and polar transformation. IEEE Trans. Med. Imaging 37, 1597\u20131605 (2018).","journal-title":"IEEE Trans. Med. Imaging"},{"key":"857_CR41","unstructured":"D\u2019Amour, A. et al. Underspecification presents challenges for credibility in modern machine learning. Preprint at http:\/\/arxiv.org\/abs\/2011.03395 (2020)."},{"key":"857_CR42","unstructured":"Guo, C., Pleiss, G., Sun, Y. & Weinberger, K. Q. On calibration of modern neural networks. Preprint at http:\/\/arxiv.org\/abs\/1706.04599 (2017)."},{"key":"857_CR43","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. Preprint at http:\/\/arxiv.org\/abs\/1512.03385 (2015).","DOI":"10.1109\/CVPR.2016.90"},{"key":"857_CR44","doi-asserted-by":"publisher","first-page":"238","DOI":"10.1136\/bjo.86.2.238","volume":"86","author":"PJ Foster","year":"2002","unstructured":"Foster, P. J., Buhrmann, R., Quigley, H. A. & Johnson, G. J. The definition and classification of glaucoma in prevalence surveys. Br. J. Ophthalmol. 86, 238\u2013242 (2002).","journal-title":"Br. J. Ophthalmol."},{"key":"857_CR45","doi-asserted-by":"publisher","first-page":"509","DOI":"10.1177\/193229680900300315","volume":"3","author":"J Cuadros","year":"2009","unstructured":"Cuadros, J. & Bresnick, G. EyePACS: An adaptable telemedicine system for diabetic retinopathy screening. J. Diabetes Sci. Technol. Online 3, 509\u2013516 (2009).","journal-title":"J. Diabetes Sci. Technol. Online"},{"key":"857_CR46","first-page":"1019","volume":"62","author":"HG Lemij","year":"2021","unstructured":"Lemij, H. G., Kliffen, H. & Vermeer, K. Building a labeled dataset for training an Artificial Intelligence (AI) algorithm for glaucoma screening. Invest. Ophthalmol. Vis. Sci. 62, 1019\u20131019 (2021).","journal-title":"Invest. Ophthalmol. Vis. Sci."},{"key":"857_CR47","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.5793241","author":"C de Vente","year":"2021","unstructured":"de Vente, C. et al. Rotterdam EyePACS AIROGS train. set. https:\/\/doi.org\/10.5281\/zenodo.5793241 (2021).","journal-title":"Rotterdam EyePACS AIROGS train. set."},{"key":"857_CR48","first-page":"3065","volume":"2010","author":"Z Zhang","year":"2010","unstructured":"Zhang, Z. et al. ORIGA(-light): an online retinal fundus image database for glaucoma analysis and research. Conf. Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Conf. 2010, 3065\u20133068 (2010).","journal-title":"Conf. Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. IEEE Eng. Med. Biol. Soc. Annu. Conf."},{"key":"857_CR49","unstructured":"Fang, H. et al. REFUGE2 Challenge: Treasure for Multi-Domain Learning in Glaucoma Assessment. (2022)."},{"key":"857_CR50","unstructured":"ODIR-2019. Peking university international competition on ocular disease intelligent recognition (ODIR-2019). (2019)."},{"key":"857_CR51","unstructured":"Wu, J. et al. GAMMA Challenge:Glaucoma grAding from Multi-Modality imAges. Preprint at http:\/\/arxiv.org\/abs\/2202.06511 (2022)."},{"key":"857_CR52","doi-asserted-by":"publisher","unstructured":"Fumero, F., Alayon, S., Sanchez, J. L., Sigut, J. & Gonzalez-Hernandez, M. RIM-ONE: An open retinal image database for optic nerve evaluation. in 2011 24th International Symposium on Computer-Based Medical Systems (CBMS) 1\u20136 (2011). https:\/\/doi.org\/10.1109\/CBMS.2011.5999143.","DOI":"10.1109\/CBMS.2011.5999143"},{"key":"857_CR53","doi-asserted-by":"publisher","first-page":"161","DOI":"10.5566\/ias.2346","volume":"39","author":"FJF Batista","year":"2020","unstructured":"Batista, F. J. F. et al. RIM-ONE DL: A unified retinal image database for assessing glaucoma using deep learning. Image Anal. Stereol. 39, 161\u2013167 (2020).","journal-title":"Image Anal. Stereol."},{"key":"857_CR54","doi-asserted-by":"publisher","DOI":"10.1186\/s12938-019-0649-y","volume":"18","author":"A Diaz-Pinto","year":"2019","unstructured":"Diaz-Pinto, A. et al. CNNs for automatic glaucoma assessment using fundus images: an extensive validation. Biomed. Eng. OnLine 18, 29 (2019).","journal-title":"Biomed. Eng. OnLine"},{"key":"857_CR55","doi-asserted-by":"publisher","first-page":"179","DOI":"10.1109\/TIT.1962.1057692","volume":"8","author":"M-K Hu","year":"1962","unstructured":"Hu, M.-K. Visual pattern recognition by moment invariants. IRE Trans. Inf. Theory 8, 179\u2013187 (1962).","journal-title":"IRE Trans. Inf. Theory"},{"key":"857_CR56","doi-asserted-by":"publisher","first-page":"101636","DOI":"10.1016\/j.compmedimag.2019.05.004","volume":"76","author":"R Hemelings","year":"2019","unstructured":"Hemelings, R. et al. Artery\u2013vein segmentation in fundus images using a fully convolutional network. Comput. Med. Imaging Graph. 76, 101636 (2019).","journal-title":"Comput. Med. Imaging Graph."},{"key":"857_CR57","doi-asserted-by":"publisher","first-page":"1389","DOI":"10.1109\/LSP.2014.2337313","volume":"21","author":"X Sun","year":"2014","unstructured":"Sun, X. & Xu, W. Fast implementation of DeLong\u2019s algorithm for comparing the areas under correlated receiver operating characteristic curves. IEEE Signal Process. Lett. 21, 1389\u20131393 (2014).","journal-title":"IEEE Signal Process. Lett."},{"key":"857_CR58","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1038\/s41592-019-0686-2","volume":"17","author":"P Virtanen","year":"2020","unstructured":"Virtanen, P. et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat. Methods 17, 261\u2013272 (2020).","journal-title":"Nat. Methods"},{"key":"857_CR59","doi-asserted-by":"publisher","unstructured":"Niculescu-Mizil, A. & Caruana, R. Predicting good probabilities with supervised learning. in Proceedings of the 22nd international conference on Machine learning 625\u2013632 (Association for Computing Machinery, 2005). https:\/\/doi.org\/10.1145\/1102351.1102430.","DOI":"10.1145\/1102351.1102430"},{"key":"857_CR60","first-page":"1","volume":"20","author":"M Alber","year":"2019","unstructured":"Alber, M. et al. iNNvestigate neural networks! J. Mach. Learn. Res. 20, 1\u20138 (2019).","journal-title":"J. Mach. Learn. Res."},{"key":"857_CR61","unstructured":"Sun, Y. et al. Test-time training with self-supervision for generalization under distribution shifts. in Proceedings of the 37th International Conference on Machine Learning 9229\u20139248 (PMLR, 2020)."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00857-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00857-0","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00857-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,6,13]],"date-time":"2023-06-13T20:35:56Z","timestamp":1686688556000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00857-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,13]]},"references-count":61,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["857"],"URL":"https:\/\/doi.org\/10.1038\/s41746-023-00857-0","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,6,13]]},"assertion":[{"value":"14 August 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 June 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 June 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"No outside entities have been involved in the study design, in the collection, analysis and interpretation of data, in the writing of the manuscript, nor in the decision to submit the manuscript for publication. I.S. is co-founder, shareholder, and consultant of Mona.health, a KU Leuven \/ VITO spin-off to which the described model was transferred. The study design was conceptualized in light of the PhD thesis of R.H., prior to the model transfer. Under their terms of employment at KU Leuven, R.H. and M.B.B. are entitled to stock options in Mona.health. R.H. has received consultancy fees from Mona.health.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"112"}}