Glaucoma Detection and Segmentation as Computer Aided Design: A review and study

Glaucoma is the leading cause of irreversible blindness in India, little is known about the prevention of glaucoma, early detection and preventive treatment is the best way to keep one falling prey to the disease. In this paper, we present a review of the latest techniques developed to detect and segment glaucoma in an eye. Glaucoma is hard to diagnose with a single test, opthalmologists perform various tests to classify a patient as glaucomatous or not. Consequently, early detection can prove to be helpful for the patients as well as the experts who are studying the field. The techniques we discuss explores the intriguing properties of retinal images. Most of the papers reviewed are focussed on extracting cup-to-disc ratio(CDR), CDR is commonly used to analyse the damage to optic nerves, larger CDR indicates higher risk of glaucoma. These methods are inspired form deep learning, feature engineering methods like thresholding, pixel-classification.