Artificial Intelligence to Identify Retinal Fundus Images, Quality Validation, Laterality Evaluation, Macular Degeneration, and Suspected Glaucoma

• Published in: Clinical ophthalmology (Auckland, N.Z.) Vol. 14; pp. 419 - 429
• Main Authors: Zapata, Miguel Angel, Royo-Fibla, Didac, Font, Octavi, Vela, Jose Ignacio, Marcantonio, Ivanna, Moya-Sanchez, Eduardo Ulises, Sanchez-Perez, Abraham, Garcia-Gasulla, Dario, Cortes, Ulises, Ayguade, Eduard, Labarta, Jesus
• Format: Journal Article
• Language: English
• Published: New Zealand Dove Medical Press Limited 01.02.2020; Taylor & Francis Ltd; Dove; Dove Medical Press
• Subjects: Accuracy; Age; Algorithms; Artificial intelligence; artificial intelligence retinal diseases screening retinal fundus image; Cameras; Classification; Data mining; Datasets; Deep learning; Diabetes; Diabetic retinopathy; Diseases; Glaucoma; Macular degeneration; Medical imaging equipment; Medical personnel; Operating systems (Software); Ophthalmology; Optics; Original Research; Photography; Physiological aspects; Quality; Retinal diseases; Telemedicine; Tomography; Spain; retinal diseases; screening; retinal fundus image; artificial intelligence
• ISSN: 1177-5483; 1177-5467; 1177-5483
• DOI: 10.2147/OPTH.S235751

To assess the performance of deep learning algorithms for different tasks in retinal fundus images: (1) detection of retinal fundus images versus optical coherence tomography (OCT) or other images, (2) evaluation of good quality retinal fundus images, (3) distinction between right eye (OD) and left eye (OS) retinal fundus images,(4) detection of age-related macular degeneration (AMD) and (5) detection of referable glaucomatous optic neuropathy (GON). Five algorithms were designed. Retrospective study from a database of 306,302 images, Optretina's tagged dataset. Three different ophthalmologists, all retinal specialists, classified all images. The dataset was split per patient in a training (80%) and testing (20%) splits. Three different CNN architectures were employed, two of which were custom designed to minimize the number of parameters with minimal impact on its accuracy. Main outcome measure was area under the curve (AUC) with accuracy, sensitivity and specificity. Determination of retinal fundus image had AUC of 0.979 with an accuracy of 96% (sensitivity 97.7%, specificity 92.4%). Determination of good quality retinal fundus image had AUC of 0.947, accuracy 91.8% (sensitivity 96.9%, specificity 81.8%). Algorithm for OD/OS had AUC 0.989, accuracy 97.4%. AMD had AUC of 0.936, accuracy 86.3% (sensitivity 90.2% specificity 82.5%), GON had AUC of 0.863, accuracy 80.2% (sensitivity 76.8%, specificity 83.8%). Deep learning algorithms can differentiate a retinal fundus image from other images. Algorithms can evaluate the quality of an image, discriminate between right or left eye and detect the presence of AMD and GON with a high level of accuracy, sensitivity and specificity.