Detection and Classification of Age Related Macular Degeneration in Retinal Images using Support Vector Machine Classifier

Nowadays retinal diseases are quite common in elderly people. In some cases retinal diseases can even leads to permanent blindness, one such retinal disease is Macular degeneration. Macula, it is central part of retina responsible for the sharp central vision needed for detailed activities that requires central vision such as reading, writing, driving. But progressive destruction of delicate cells of macula leads to a condition called Age related macular degeneration (ARMD or AMD). Fatty deposits called as drusen will accumulate in macula region which progressively destructs the macula and in the worst case it even leads to permanent blindness. In this paper an algorithm for semi-automated detection of AMD using image processing techniques using fundus images have been proposed. The proposed method includes pre-processing, segmentation and feature extraction techniques. Feature extraction includes Gray Level Co-occurrence matrix (GLCM), which is used to extract the features which are required for the classification process. The classification results are evaluated with the use of accuracy, sensitivity and specificity with the help of Support Vector Machine (SVM) classifier. Proposed method obtains accuracy 94.12%, sensitivity 91.66% and specificity 100%.