Automated identification of exudates for detection of macular edema

Macular edema is an advance stage of diabetic retinopathy which affects central vision of diabetes patients. The main cause of edema is the appearance of exudates near or on macular region in human retina. An automated system for early detection of macular edema should identify all possible exudates present on the surface of retina. In this paper, we present a method for the identification of exudates in colored retinal images which will help in building a computer aided diagnostic system for macular edema. The proposed system consists of three stages i.e. candidate exudate detection, feature extraction and classification. We use filter bank for candidate exudate detection, basic properties of exudates for feature extraction and Gaussian mixture model for classification. This paper presents the performance of our system on three retinal image databases and comparative results with existing methods.