Attention U-Net for Glaucoma Identification Using Fundus Image Segmentation

Glaucoma is a fatal and worldwide ocular disease that can result in irreversible blindness to the optic nerve fibers of the eye. After cataracts, glaucoma is a main reason for blindness. Optic Disc (OD) and Optic Cup (OC) are important for fundus image segmentation. This study proposes attention U-Net models with three Convolutional Neural Networks (CNNs) architectures, namely Inception-v3, Visual Geometry Group 19 (VGG19), Residual Neural Network 50 (ResNet50) to segment fundus images. Several data augmentation techniques were used to avoid overfitting and achieve high accuracy. The attention U-Net with ResNet50 as the encoder backbone showed the highest accuracy of 99.53% in segmenting OD using the RIM-ONE dataset among the considered configurations.