Alexandra Miere1, Vittorio Capuano2, Arthur Kessler3, Olivia Zambrowski2, Camille Jung4, Donato Colantuono2, Carlotta Pallone2, Oudy Semoun2, Eric Petit5, Eric Souied2. 1. Department of Ophthalmology, Centre Hospitalier Intercommunal de Créteil, Créteil, France; Laboratory of Images, Signals and Intelligent Systems (LISSI), (EA N° 3956), University Paris-Est, Créteil, France. Electronic address: alexandra.miere@chicreteil.fr. 2. Department of Ophthalmology, Centre Hospitalier Intercommunal de Créteil, Créteil, France. 3. EPISEN - ISBS, University Paris-Est, Créteil, France. 4. Clinical Research Center, Centre Hospitalier Intercommunal de Créteil, Créteil, France. 5. Laboratory of Images, Signals and Intelligent Systems (LISSI), (EA N° 3956), University Paris-Est, Créteil, France.
Abstract
PURPOSE: To automatically classify retinal atrophy according to its etiology, using fundus autofluorescence (FAF) images, using a deep learning model. METHODS: In this study, FAF images of patients with advanced dry age-related macular degeneration (AMD), also called geographic atrophy (GA), and genetically confirmed inherited retinal diseases (IRDs) in late atrophic stages [Stargardt disease (STGD1) and Pseudo-Stargardt Pattern Dystrophy (PSPD)] were included. The FAF images were used to train a multi-layer deep convolutional neural network (CNN) to differentiate on FAF between atrophy in the context of AMD (GA) and atrophy secondary to IRDs. Three-hundred fourteen FAF images were included, of which 110 images were of GA eyes and 204 were eyes with genetically confirmed STGD1 or PSPD. In the first approach, the CNN was trained and validated with 251 FAF images. Established augmentation techniques were used and an Adam optimizer was used for training. For the subsequent testing, the built classifiers were then tested with 63 untrained FAF images. The visualization method was integrated gradient visualization. In the second approach, 10-fold cross-validation was used to determine the model's performance. RESULTS: In the first approach, the best performance of the model was obtained using 10 epochs, with an accuracy of 0.92 and an area under the curve for Receiver Operating Characteristic (AUC-ROC) of 0.981. Mean accuracy was 87.30 ± 2.96. In the second approach, a mean accuracy of 0.79 ± 0.06 was obtained. CONCLUSION: This study describes the use of a deep learning-based algorithm to automatically classify atrophy on FAF imaging according to its etiology. Accurate differential diagnosis between GA and late-onset IRDs masquerading as GA on FAF can be performed with good accuracy and AUC-ROC values.
PURPOSE: To automatically classify retinal atrophy according to its etiology, using fundus autofluorescence (FAF) images, using a deep learning model. METHODS: In this study, FAF images of patients with advanced dry age-related macular degeneration (AMD), also called geographic atrophy (GA), and genetically confirmed inherited retinal diseases (IRDs) in late atrophic stages [Stargardt disease (STGD1) and Pseudo-Stargardt Pattern Dystrophy (PSPD)] were included. The FAF images were used to train a multi-layer deep convolutional neural network (CNN) to differentiate on FAF between atrophy in the context of AMD (GA) and atrophy secondary to IRDs. Three-hundred fourteen FAF images were included, of which 110 images were of GA eyes and 204 were eyes with genetically confirmed STGD1 or PSPD. In the first approach, the CNN was trained and validated with 251 FAF images. Established augmentation techniques were used and an Adam optimizer was used for training. For the subsequent testing, the built classifiers were then tested with 63 untrained FAF images. The visualization method was integrated gradient visualization. In the second approach, 10-fold cross-validation was used to determine the model's performance. RESULTS: In the first approach, the best performance of the model was obtained using 10 epochs, with an accuracy of 0.92 and an area under the curve for Receiver Operating Characteristic (AUC-ROC) of 0.981. Mean accuracy was 87.30 ± 2.96. In the second approach, a mean accuracy of 0.79 ± 0.06 was obtained. CONCLUSION: This study describes the use of a deep learning-based algorithm to automatically classify atrophy on FAF imaging according to its etiology. Accurate differential diagnosis between GA and late-onset IRDs masquerading as GA on FAF can be performed with good accuracy and AUC-ROC values.
Authors: Alexandra Miere; Olivia Zambrowski; Arthur Kessler; Carl-Joe Mehanna; Carlotta Pallone; Daniel Seknazi; Paul Denys; Francesca Amoroso; Eric Petit; Eric H Souied Journal: J Clin Med Date: 2021-12-08 Impact factor: 4.241