In this study, FAF images of patients with retinitis pigmentosa (RP), Best disease (BD), Stargardt disease (STGD), as well as a healthy comparable group were used to train a multilayer deep convolutional neural network (CNN) to differentiate FAF images between each type of IRD and normal FAF. The CNN was trained and validated with 389 FAF images. Established augmentation techniques were used. An Adam optimizer was used for training. For subsequent testing, the built classifiers were then tested with 94 untrained FAF images. Results. For the inherited retinal disease classifiers, global accuracy was 0.95. The precision-recall area under the curve (PRC-AUC) averaged 0.988 for BD, 0.999 for RP, 0.996 for STGD, and 0.989 for healthy controls. Conclusions. This study describes the use of a deep learning-based algorithm to automatically detect and classify inherited retinal disease in FAF. Hereby, the created classifiers showed excellent results. With further developments, this model may be a diagnostic tool and may give relevant information for future therapeutic approaches.

View Full Text

Authors:

Alexandra Miere 1,2, Thomas Le Meur3, Karen Bitton1, Carlotta Pallone1, Oudy Semoun1, Vittorio Capuano1, Donato Colantuono1, Kawther Taibouni2, Yasmina Chenoune2,4, Polina Astroz1, Sylvain Berlemont3, Eric Petit2 and Eric Souied1

1 Department of Ophthalmology, Centre Hospitalier Intercommunal de Créteil, 94010 Créteil, France

2 Laboratory of Images, Signals and Intelligent Systems (LISSI, (EA N° 3956), University Paris-Est Créteil, 94400 Vitry sur Seine, France

3 Keen Eye Technologies SAS, 75012 Paris, France

4 ESME Sudria, 69002 Lyon, France

J. Clin. Med. 2020, 9, 3303