Detection of retinal whitening, cotton wool spots and retinal Hemorrhages for diagnosis of Malarial Retinopathy

Malarial Retinopathy is an eye disease which affects retina of a patient suffering from malaria. The changes in vision are only observable at advanced stages of malaria. In this paper, we present a novel method for automated diagnosis of malarial retinopathy by detecting retinal whitening, cotton wool spots and Hemorrhages cases. The proposed system extracts a number of features to differentiate normal images from the images containing retinal abnormalities. Then it applies feature selection technique to extract best discriminating features followed by support vector machine (SVM) based classification. The evaluation of proposed system has been done on locally gathered images from malarial patients and results show the significance of proposed system.