RetinOpTIC-Automatic Evaluation of Diabetic Retinopathy

Bruno Laÿ (1) , Ronan Danno (1) , Gwenolé Quellec (2) , Etienne Decencière (3) , Ali Erginay (4) , Pascale Massin (5) , Alexandre Le Guilcher , Mathieu Lamard (6) , Beatrice Cochener (7) , Robin Alais (3)

1 ADCIS - Société ADCIS
2 INSERM - Institut National de la Santé et de la Recherche Médicale
3 CMM - Centre de Morphologie Mathématique
4 Service d'Ophtalmologie - Service d'Ophtalmologie, Hôpital Lariboisière, AP-HP
5 Centre Hospitalier Lariboisière
6 LaTIM - Laboratoire de Traitement de l'Information Medicale
7 UBO - Université de Brest

Purpose: The RetinOpTIC project performs mass screening of color fundus images and assesses image quality and Diabetic Retinopathy (DR) grade. Algorithm performance is evaluated on the Messidor-2 image database. Methods: Based on artificial intelligence (AI) solutions, referable DR is detected using convolutional neural networks (CNNs). The solution includes first the automatic assessment of the quality of the photography, and then the DR grade

Domaines: Traitement des images [eess.IV]; Traitement d'image; Réseau de neurones [cs.NE]

Bruno Laÿ, Ronan Danno, Gwenolé Quellec, Etienne Decencière, Ali Erginay, et al.. RetinOpTIC-Automatic Evaluation of Diabetic Retinopathy. ARVO, 2019, Vancouver, Canada. ⟨hal-02428873⟩