A Deep Learning Algorithm to Quantify Neuroretinal Rim Loss From Optic Disc Photographs

Abstract

Purpose: To train a deep learning (DL) algorithm that quantifies glaucomatous neuroretinal damage on fundus photographs using the minimum rim width relative to Bruch's membrane opening (BMO-MRW) from spectral domain-optical coherence tomography (SDOCT). Design: Cross-sectional study Methods: 9,282 pairs of optic disc photographs and SDOCT optic nerve head scans from 927 eyes of 490 subjects were randomly divided into the validation plus training (80%) and test sets (20%). A DL convolutional neural network was trained to predict the SDOCT BMO-MRW global and sector values when evaluating optic disc photographs. The predictions of the DL network were compared to the actual SDOCT measurements. The area under the receiver operating curve (AUC) was used to evaluate the ability of the network to discriminate glaucomatous visual field loss from normal eyes. Results: The DL predictions of global BMO-MRW from all optic disc photos in the test set (mean ± standard deviation [SD]: 228.8±63.1μm) were highly correlated with the observed values from SDOCT (mean ± SD: 226.0±73.8μm) (Pearson's r=0.88; R 2 =77%; P<0.001), with mean absolute error of the predictions of 27.8μm. The AUCs for discriminating glaucomatous from healthy eyes with the DL predictions and actual SDOCT global BMO-MRW measurements were 0.945 (95% CI:0.874-0.980) and 0.933 (95% CI:0.856-0.975), respectively (P=0.587). Conclusions: A DL network can be trained to quantify the amount of neuroretinal damage on optic disc photographs using SDOCT BMO-MRW as a reference. This algorithm showed high accuracy for glaucoma detection, and may potentially eliminate the need for human gradings of disc photos.

Artificial intelligence deep learning algorithm for discriminating ungradable optical coherence tomography three-dimensional volumetric optic disc scans

“…At present, multiple DLSs have been developed based on OCT in ophthalmology, such as referable retina diseases detection, 27,28 glaucoma quantification and classification, [29][30][31] and antivascular endothelial growth factor treatment. 32 These studies perfectly underscored the promise of DL to lower the cost of disease interpretation from OCT images.…”

Artificial Intelligence Mapping of Structure to Function in Glaucoma

“…Recently, neural networks and other artificial intelligence (AI) algorithms have been shown to successfully model complex, nonlinear relationships in data from diverse medical fields. [8][9][10][11][12] In particular, convolutional neural networks (CNNs) are able to take advantage of spatial information to identify underlying relationships that may not be easily discerned by conventional methods. A few studies have attempted to use AI algorithms to predict visual field results from SDOCT measurements, with good results.…”

Macular Ganglion Cell-Inner Plexiform Layer Thickness Prediction from Red-free Fundus Photography using Hybrid Deep Learning Model

“…There are several studies that demonstrated the potential of deep learning models for glaucoma diagnosis using SD-OCT data 19,20 . Going one step further, research has shown the possibility of deep learning-based quantitative prediction (RNFL thickness or BMO-MRW) beyond the limits of simple classification of disease status 14,15 . Thus prompted, we undertook a new challenge, which was to use the HDLM algorithm to convert qualitative data (RNFLP) to quantitative data (mGCIPL thickness).…”

“…photographs by deep learning models 14,15 . They showed the potential of deep learning models to provide quantitative information on the extent of neural damage from qualitative data (optic disc photographs).…”