A new method for detection of optical disc and macula for diabetic retinopathy

Abstract

In recent years, the rapid growth of the silent killer, Diabetes, is a major reason of concern for health and also started a buzz among our society. On the same time, World Health Organization (WHO) is a dedicated health organization from United Nations quotes, “By 2030 diabetes will be one of the seventh leading causes of death, whereas, the global prevalence of diabetes was estimated to be about 9% among adults aged above 18 (measured in 2014)”. Though, diabetes is not completely curable but early stage detection can helps in controlling the degree of severity of this silent killer. One of the controlling factors, which could help in early detection, is the retina of the eye, a light sensitive metabolically active region on which noninvasive observation can possible. As eye is a metabolically active organ, it gets affected at the earliest stage for the people suffering from Diabetes. Thereafter, analyzing the symptoms of retina, like, miroaneurysms, exudates, and edemas can predict the degree of severity. Since, a plan is required towards the studying these symptoms on digital image, the fundus image of the retina, and the detection of attributes, like, miroaneurysms, exudates and edemas. Before carrying out the analysis, it is very essential to properly locate the optic disc and macula which are geometrically important for the detection of Diabetic Retinopathy on digital images. In this paper, we propose intensity based preprocessing algorithm which is suitable for the detection of optic disc (OD) and Macula before the analysis of Diabetic Retinopathy.