Optic Disc Segmentation in Diabetic Retinopathy using Image Processing

Abstract

Diabetic Retinopathy and Glaucoma are optic diseases that involve optic disk identification, which is a crucial phase in the current diagnostic tools that can be computerized. When these diseases are identified early by any screening applications, measures may be taken to avoid blindness. Early indicators of the numerous illness such as Macula Edema, Diabetic Retinopathy and Glaucoma are the changes in the anatomy structures in the retina of the human eye which also has the inclusion of the retinal vasculature. Of these, the Optic Disc is the most crucial feature, as its visible factors are essential for the identification of glaucoma and other disease-related assessments called Diabetic Retinopathy. In this paper, we present methods to detect the likelihood of Diabetic Retinopathy being present from fundus images. This technique starts with pre-processing on the optic retinal image to concentrate on the main area of the disease that we need to identify. Afterwards we apply Image processing algorithms to detect the optic disk. Detecting the optic disc is vital because it is the origin of all the nerves and detecting the position and radius of optic disc can be used as the reference for approximating fovea i.e. a pit like area responsible for vision. Size and shape of optic disc is responsible for diagnosing the disease. Therefore, this paper addresses the analysis of different techniques to detect the optic disc.