Detection of Hard Exudates in Retinal Fundus Images Using Deep Learning

Abstract

Diabetic Retinopathy (DR) is a retinal disorder that affects the people having diabetes mellitus for a long time (20 years). DR is one of the main reasons for the preventable blindness all over the world. If not detected early, the patient may progress to severe stages of irreversible blindness. Lack of Ophthalmologists poses a serious problem for the growing diabetes patients. It is advised to develop an automated DR screening system to assist the Ophthalmologist in decision making. Hard exudates develop when DR is present. It is important to detect hard exudates in order to detect DR in an early stage. Research has been done to detect hard exudates using regular image processing techniques and Machine Learning techniques. Here, a deep learning algorithm has been presented in this paper that detects hard exudates in fundus images of the retina.