查看文章

Retinal disease classification is an important challenge in computer aided diagnosis (CAD) for medical applications. Eye diseases can cause different symptoms from mild vision problems to complete blindness if it is not timely treated. The early diagnosis is crucial to prevent blindness. In this work, we use deep Convolutional Neural Networks (CNN) on a 4-class classification problem to automatically detect choroidal neovascularization (CNV), diabetic macular edema (DME), drusen, and normal cases using Optical Coherence Tomography (OCT) images. The obtained results achieve state-of-the-art performance and show that the proposed network leads to higher classification rates with an accuracy of 98.46%, and an Area Under Curve (AUC) of 0.998. An explainability algorithm was also developed and shows the efficiency of the proposed approach in detecting retinal disease signs.