An automatic glaucoma screening algorithm using cup-to-disc ratio and ISNT rule with support vector machine

Abstract

This study introduces a glaucoma screening system which is suitable for both normal eyesight and myopia. For myopia eye, only one clinical indicator is not sufficient. Therefore, cup-to-disc ratio and rim width based on ISNT rule are used as indicators in this screening system. First, image segmentation is performed to obtain optic cup, optic disc and rim width. These information will be used as features in glaucoma classification step. The support vector machine (SVM) is selected as a classifier. According to the results, our proposed method is an efficient system for screening glaucoma.