Detection of ARMD and Grading of Maculopathy Severity Levels from Retinal Fundus Images

Abstract

Age Related Macular Degeneration (ARMD) is a retinal disorder usually found in old people, this affects the central vision, but not the peripheral vision. The objective of this work is to develop an automated system for the classification of ARMD using digital fundus images and maculopathy severity levels, so that an intelligent computer aided system can be developed for the diagnosis of ARMD and maculopathy. Diagnosis and treatment of ARMD and maculopathy in the earlier stages helps to cure the disease and According to American Society of Retinal Specialists, about fifteen million people around the globe are suffering will save many from vision loss. ARMD may result from the ageing and thinning of macular tissues, depositing of pigment in the macula, growth of abnormal blood vessels or a combination of these processes. So, it is possible to detect and diagnosis the disease by analyzing such abnormalities present in the retinal images. The abnormalities should have significant difference with the neighboring tissues and by properly segmenting such regions, as a result macular degeneration is predicted whether it is present or not. Here the abnormalities are segmented from the fundus images after multi stage segmentation processes. Threshold based segmentation is used along with Canny edge detection algorithm to efficiently identify the abnormality region. Threshold based binary classification of the analyzed image is evaluated by finding the extent of abnormalities present in the image.

Diabetic maculopathy is the major cause of irreversible vision loss due to retinopathy and is found in 10% of the world diabetic population. Compulsory mass screening will help to identify the maculopathy at early stage and reduce the risk of

severe vision loss. In this paper, we present a computer based system for automatic detection and grading of diabetic maculopathy severity level without manual intervention. The optic disc is detected automatically and its location and diameter is used to detect fovea and to mark the macular region respectively. Next, hard exudates are detected using clustering and mathematical morphological techniques. Based on the location of exudates in marked macular region the severity level of maculopathy is classified into mild, moderate and severe.