Automated Grading of Age-Related Macular Degeneration From Color Fundus Images Using Deep Convolutional Neural Networks

Abstract

IMPORTANCE Age-related macular degeneration (AMD) affects millions of people throughout the world. The intermediate stage may go undetected, as it typically is asymptomatic. However, the preferred practice patterns for AMD recommend identifying individuals with this stage of the disease to educate how to monitor for the early detection of the choroidal neovascular stage before substantial vision loss has occurred and to consider dietary supplements that might reduce the risk of the disease progressing from the intermediate to the advanced stage. Identification, though, can be time-intensive and requires expertly trained individuals. OBJECTIVE To develop methods for automatically detecting AMD from fundus images using a novel application of deep learning methods to the automated assessment of these images and to leverage artificial intelligence advances. DESIGN, SETTING, AND PARTICIPANTSDeep convolutional neural networks that are explicitly trained for performing automated AMD grading were compared with an alternate deep learning method that used transfer learning and universal features and with a trained clinical grader. Age-related macular degeneration automated detection was applied to a 2-class classification problem in which the task was to distinguish the disease-free/early stages from the referable intermediate/advanced stages. Using several experiments that entailed different data partitioning, the performance of the machine algorithms and human graders in evaluating more than 130 000 images that were deidentified with respect to age, sex, and race/ethnicity from 4613 patients against a gold standard included in the National Institutes of Health Age-Related Eye Disease Study data set was evaluated.MAIN OUTCOMES AND MEASURES Accuracy, receiver operating characteristics and area under the curve, and κ score. RESULTSThe deep convolutional neural network method yielded accuracy that ranged between 88.4% (SD, 0.5%) and 91.6% (SD, 0.1%), the area under the receiver operating characteristic curve was between 0.94 and 0.96, and κ (SD) between 0.764 (0.010) and 0.829 (0.003), which indicated a substantial agreement with the gold standard Age-Related Eye Disease Study data set.CONCLUSIONS AND RELEVANCE Applying a deep learning-based automated assessment of AMD from fundus images can produce results that are similar to human performance levels. This study demonstrates that automated algorithms could play a role that is independent of expert human graders in the current management of AMD and could address the costs of screening or monitoring, access to health care, and the assessment of novel treatments that address the development or progression of AMD.

Current state and future prospects of artificial intelligence in ophthalmology: a review

“…Their model was able to do that, with a root mean squared error of 12.9 letters. 52 They trained their programs with different MLCs with over 130 000 colour fundus images from 4613 patients. 50 These results should be considered within the context of the lower intersession reliability of visual acuity measurement in AMD.…”

Development and validation of a deep‐learning algorithm for the detection of neovascular age‐related macular degeneration from colour fundus photographs

“…Over the past decade, automated techniques for the assessment of AMD, via feature extraction from small retinal image datasets (<1000), have been reported with variable accuracy. More recent reports provide novel data on the accuracy of deep‐learning systems for the detection of AMD . The majority of these studies have utilized the Age‐Related Eye Disease Study (AREDS) participants to develop and test the accuracy of their DLAs .…”

“…Although they used late AMD as a classifier, this included geographic atrophy which our DLA does not classify as referable AMD. A binary classification was used to determine referable AMD (≥intermediate AMD) in the investigations by Ting et al (2017) and Burlina et al (2017) who both reported AUC's between 0.93 and 0.96. Whilst these findings are similar to the current study, they are not directly comparable as only those classified as neovascular AMD we considered referable.…”

“…1 There are two types of advanced AMD, atrophic ("dry") and neovascular ("wet"), with the latter having been responsible for an estimated 90% of severe vision loss cases. 2 Effective therapeutic options are currently only available for neovascular AMD, with anti-vascular endothelial growth factor (VEGF) agents effective in reducing vision loss [3][4][5] and, in many cases, restoring vision. 6,7 Detection of early onset neovascular AMD and timely treatment is essential for the protection of visual function.…”

All eyes are on AI

“…Most recently, a deep-learning algorithm classified age-related macular degeneration and diabetic retinopathy from optical coherence tomography images of the retina 1 . Both of these eye conditions and others (such as glaucoma and macular oedema), have also been automatically assessed by deep learning trained on fundus images [2][3][4] (a retinal fundus image is a photograph of the internal surface at the back of the eye). In all these tests, the algorithms, which had been trained on hundreds of thousands of medically labelled images, performed on par with teams of human ophthalmologists and better than many individual experts when validated with thousands or hundreds of thousands of images.…”