Diagnosis of glaucoma using Cup to Disc Ratio in stratus OCT retinal images

Abstract

Glaucoma is one of the second most important cause of blindness after cataracts. Detection of glaucoma is essential to prevent visual damage. In India, Glaucoma is the third leading cause of blindness. This paper discusses an algorithm developed for detection and diagnosis of glaucoma. The algorithm calculates the Cup to Disc Ratio (CDR) which is obtained through image processing of the retinal images obtained from stratus OCT from Sudhalkar Eye Hospital, Vadodara. Image processing has been carried out on 120 retinal images obtained from Stratus OCT out which 79 are glaucomatous eyes and 41 are normal eyes. OTSU histogram and watershed algorithm have been used for image segmentation. Image analysis has been carried out through image segmentation, which is the process of dividing an image into regions or object. Morphological operations have been performed for the enhancement of the images to extract optic cup and optic disc region from the eye. The accuracy of the algorithm is 94 %. MATLAB software has been used for image processing on the retinal images.