Automatic Computation of CDR Using Fuzzy Clustering Techniques

Abstract

Eye disease identification techniques are highly important in the field of ophthalmology. A vertical Cup-to-Disc Ratio which is the ratio of the vertical diameter of the optic cup to that of the optic disc, of the fundus eye image is one of the important signs of glaucoma. This paper presents an automated method for the extraction of optic disc and optic cup using Fuzzy C Means clustering technique. The validity of this new method has been tested on 454 colour fundus images from three different publicly available databases DRION, DIARATDB0 and DIARETDB1 and, images from an ophthalmologist. The average success rate of optic disc and optic cup segmentation is 94.26percentage. The scatter plot depicts high positive correlation between clinical CDR and the CDR obtained using the new method. The result of the system seems to be promising and useful for clinical work.