A deep-learning system predicts glaucoma incidence and progression using retinal photographs

Abstract

Background Deep learning has been widely used for glaucoma diagnosis. However, there is no clinically validated algorithm for glaucoma incidence and progression prediction. This study aims to develop a clinically feasible deep-learning system for predicting and stratifying the risk of glaucoma onset and progression based on color fundus photographs (CFPs), with clinical validation of performance in external population cohorts. Methods We established data sets of CFPs and visual fields collected from longitudinal cohorts. The mean follow-up duration was 3 to 5 years across the data sets. Artificial intelligence (AI) models were developed to predict future glaucoma incidence and progression based on the CFPs of 17,497 eyes in 9346 patients. The area under the receiver operating characteristic (AUROC) curve, sensitivity, and specificity of the AI models were calculated with reference to the labels provided by experienced ophthalmologists. Incidence and progression of glaucoma were determined based on longitudinal CFP images or visual fields, respectively. Results The AI model to predict glaucoma incidence achieved an AUROC of 0.90 (0.81–0.99) in the validation set and demonstrated good generalizability, with AUROCs of 0.89 (0.83–0.95) and 0.88 (0.79–0.97) in external test sets 1 and 2, respectively. The AI model to predict glaucoma progression achieved an AUROC of 0.91 (0.88–0.94) in the validation set, and also demonstrated outstanding predictive performance with AUROCs of 0.87 (0.81–0.92) and 0.88 (0.83–0.94) in external test sets 1 and 2, respectively. Conclusion Our study demonstrates the feasibility of deep-learning algorithms in the early detection and prediction of glaucoma progression. FUNDING National Natural Science Foundation of China (NSFC); the High-level Hospital Construction Project, Zhongshan Ophthalmic Center, Sun Yat-sen University; the Science and Technology Program of Guangzhou, China (2021), the Science and Technology Development Fund (FDCT) of Macau, and FDCT-NSFC.

Review of Visualization Approaches in Deep Learning Models of Glaucoma

“…For example, a “heatmap” (saliency map or activation map) can depict regions of an image that highly influence a model’s decision-making process 7,17 . Nearly half (8/18, 44%) of the review articles in this study introduced a visualization approach that allowed superimposition on a single fundus photograph 15,22,23,25,27,31 . These results are not surprising given the widespread use of fundus photography in ophthalmology due to its availability, affordability, and ease of use 41,42 .…”

“…7,17 Nearly half (8/18, 44%) of the review articles in this study introduced a visualization approach that allowed superimposition on a single fundus photograph. 15,22,23,25,27,31 These results are not surprising given the widespread use of fundus photography in ophthalmology due to its availability, affordability, and ease of use. 41,42 Other tools used by researchers included photography encompassing the optic nerve head with the surrounding fundus (Fig.…”

“…1B). 25,27,31 Some studies used optical coherence tomography (OCT), which is useful in 3-dimensional visualization, to build heatmaps that evaluate the performance of DL models (Fig. 1C).…”

Clinical Applications of Artificial Intelligence in Glaucoma

“…[73] Some of the recent deep CNN models applied to fundus photographs have reached AUROCs up to 0.99 for glaucoma diagnosis. [74][75][76]78] Other deep learning models have obtained AUC up to about 0.97 for glaucoma screening and AUC up to 0.94 for glaucoma referral. [36,79] A recent meta-analysis paper analyzed the accuracy of seventeen deep learning-based studies that utilized 30 different patient cohorts and reported an AUC of 0.93 (95% CI 0.92-0.94) for diagnosing glaucoma based on color fundus photographs.…”

Assistive applications of artificial intelligence in ophthalmology

“…Their proposed model outperformed linear models with an AUROC of 0.86, showing improved ability to predict glaucomatous progression over 3 years. Prediction of glaucomatous progression can also be evaluated with color fundus photos using the AI model developed by Fei et al [22], which achieved an AUROC >0.87 in multiple test sets. Over the past 5 years, research and integration of AI in ROP diagnosis and prognostication have increased.…”