Automated Detection of Exudates in Retinal Images

Abstract

Diabetic Retinopathy (DR) is a major cause of blindness which is caused by damage to the small blood vessels of the retina. Early detection of Diabetic Retinopathy is important to prevent vision loss. Main two stages of diabetic retinopathy are non-proliferative diabetic retinopathy (NPDR) and proliferative diabetic retinopathy (PDR). Microaneurysms, hemorrhages, cotton-wool spots and exudates are symptoms of non proliferative diabetic retinopathy which is the early stage of diabetic retinopathy. Of these symptoms exudates detection is more important for the early detection of diabetic retinopathy. In this paper, we are proposing a method for the automated detection of exudates by using retinal images. We show that different morphological operations could be incorporated in order to detect exudates quite accurately. In the process of exudates detection optic disk detection is an important part. Otherwise, optic disk could be wrongly identified as exudates. We have used different morphological operations for the exudates detection from the retinal images and the developed algorithm has a sensitivity and specificity 94.59% and 88.46% respectively. This is considerably a higher success rate with the existing systems.