Glaucoma Grading Using Fundus Images

Glaucoma is a chronic, progressive eye disease caused by gradual damage to the optic nerve and is considered the major cause of irreversible visual damage. Because it is impossible to reverse the loss of vision caused by the disease, early detection is essential that interventions can be carried out in the early stages of the disease to stop its progression. Fundus imaging is one of the main methods used to diagnose the disease, making it possible to assess the cup-to-disc ratio by a specialist. In this work, we propose a method based on deep learning, which uses fundus images to help detect the disease in its early stages. In this way, the proposed method can have clinical use and be used to develop tools for classifying more serious disease cases. As a best result, the proposed method achieved a kappa value of 0.83.