Glaucoma is a very common eye disease (over 2.2million americans have it). It is widely regarded that the vertical cup to disc ratio of the eye, i.e. the ratio between the length of the optic disc and the length of the optic cup, is a very good indicator of whether a patient has glaucoma or not.

The goal of this project is to use image segmentation to identify the cup and the disc from a fundus image (an image of the back of the eye), and then measure the ratio between their lengths, thus automatically obtaining a key indicator for whether the patient has glaucoma or not. This automated approach can be potentially useful in massive screening for patients.