Assessment of the Cup-to-Disc ratio method for Glaucoma detection

Abstract

Glaucoma is a leading cause of blindness and early detection of glaucoma is very important to its treatment. Several methods are being used to detect glaucoma. One of the common methods used in the detection of glaucoma is the optic cup-to-disc ratio which is commonly called CDR. The CDR is obtained from retinal fundus images which are evaluated to detect the presence or absence of glaucoma. This work assesses the performance of the CDR on two publicly available databases: The RIMONE v3 and the DRISHTI-GS. Though a CDR threshold of 0.5 is mostly used in detecting glaucoma, our results show that a threshold of 0.6 performs better. Furthermore, our results show that the CDR method, because of its use of threshold values, is not adequate in detecting early stage glaucoma. To address this, we propose an algorithm that includes a classifier with a high accuracy and better performance in detecting early to moderate glaucoma.