Diabetes Retinopathy Detection : A Survey

Abstract

Diabetic Retinopathy (DR) is a disease that may cause vision impairment. The early detection of this disease is important. This work surveys the different detection and feature selection techniques involved in the detection of this disease. This can be done by studying the lesions found in the human retina using fundus images. This work also reviews the various feature extraction techniques such as Support Vector Machines, Neural Networks, and Convolutional Networks. Deep learning algorithms are discussed and the different ways of implementation of the automated system are discussed.