Glaucoma detection by using Pearson-R correlation filter

Glaucoma is the most common cause of vision loss and is apparently becoming more important. In this paper, the research is focused on development of novel automated classification system for Glaucoma, based on image features from eye fundus photographs. A study done already has revealed that the optic cup-to-disc ratio, Neuro-retinal rim thickness and Neuro-retinal rim area in eye fundus image are the key parameters used to assess the progression of the disease. These aspects have been used by us for the detection of possible Glaucoma. Pearson-R coefficients corresponding to the eye fundus image are used as features. Segmentation algorithm is used to segment optic cup and disc and their respective vertical diameters are calculated to determine cup-to-disc ratio. Neuro-retinal rim thickness and rim area are measured using segmented portions of optic cup and disc. Methodology developed is found out to be very accurate for classification of Glaucoma. These novel techniques resulted in an overall efficiency of 97%.