Just a question, I am using your method to detect glaucoma, which needs calculating CDR value

[alikeavo]

Can you guide me on how to calculate the Cup to Disc Ratio for any dataset like the RIM-ONE v3, in this paper-you are cropping the OD to segment the OC, is it necessary to crop it? and how do we calculate the CDR values as input size is different?

[seva100]

Hi, the network was trained on the images cropped by OD, but you can try retraining it on the original images. CDR is the ratio between the height of the optic cup and the optic disc, so I guess it should not change with the input size.