Evaluation of Cup to Disc Ratio for Glaucoma detection through optic cup and disc segmentation using Fuzzy C-means clustering and comparing with adaptive thresholding method

The objective of the study is to develop an automated approach for detecting Glaucoma from fundus images by using Fuzzy c-means clustering (FCM) and comparing with adaptive thresholding. Materials and Methods: With a sample size of 40 fundus images, the FCM clustering and adaptive thresholding is used to locate the optic disc in fundus images. The sample size is calculated with pretest power of 80%, confidence interval of 95% and an alpha of 0.05. Results: The accuracy obtained is 87.63 % and 83.6288% using FCM and adaptive thresholding respectively. There is a significant difference between the two groups, since p<0.05 (0.0001, two-tailed) as per SPSS Statistical analysis. Conclusion: According to this study, the accuracy of the suggested FCM is significantly better than adaptive thresholding.