Skip to main content

  • Deep Learning for Identifying Eyes at Risk for Glaucomatous Optic Neuropathy

    By Lynda Seminara
    Selected By: Stephen D. McLeod, MD
    Add to My Bookmarks

    Journal Highlights

    Ophthalmology, December 2019

    Download PDF

    Phene et al. developed an algorithm based on deep learning and tested its effectiveness for photographic features of the optic nerve head (OHN) that would prompt referral for further eval­uation of glaucomatous optic neuropa­thy (referable GON). They found that a deep learning algorithm trained solely on fundus images has greater sensitivity than eye care providers for detecting referable GON; specificity was compa­rable for the two methods of detection.

    The fundus images used in this research were obtained from screen­ing programs, published studies, and a glaucoma clinic. The algorithm was trained using 86,618 images that also were graded by eye care providers for glaucomatous ONH features and refer­able GON. Of the 43 graders, 14 were fellowship-trained glaucoma specialists, 26 were comprehensive ophthalmolo­gists, and three were optometrists.

    The algorithm was validated using three datasets: 1) Dataset A included 1,205 images (one per patient; 18.1% referable) adjudicated by panels of glaucoma specialists; 2) dataset B consisted of 9,642 images (one per patient; 9.2% referable) from a diabetic teleretinal screening program; and 3) dataset C comprised 346 images (one per patient; 81.7% referable) from a glaucoma clinic. Outcome measures were area under the receiver operating characteristic curve (AUC), sensitivity, and specificity for referable GON and glaucomatous ONH features.

    The algorithm’s AUC for referable GON was 0.945 in dataset A (95% confidence interval [CI], 0.929-0.960), 0.855 in dataset B (95% CI, 0.841-0.870), and 0.881 in dataset C (95% CI, 0.838-0.918). AUCs for glaucoma­tous ONH features ranged from 0.661 to 0.973. The sensitivity of the algorithm was sig­nificantly higher than that of seven of 10 graders not involved in determining the reference stan­dard, including two of three glau­coma specialists. The specificity of the algorithm exceeded that of three graders (including one glaucoma specialist) and was comparable to that of other graders. The algorithm per­formed favorably across independent datasets. According to specialists and the algorithm, crucial features of re­ferable GON were vertical cup-to-disc ratio ≥0.7, notching of the neuroretinal rim, abnormality of the retinal nerve fiber layer, and baring of the circumlin­ear vessels.

    The authors suggested that algo­rithms such as this one may improve the effectiveness of glaucoma screening in settings without clinicians who can interpret ONH features.

    The original article can be found here.