Prediction of Glaucoma Using Deep Learning Based Approaches.

Glaucoma is an eye related condition, which mainly occurs due to the damage to the optic nerve, that connects the eye to our brain. Unfortunately, the damage due to glaucoma is irreversible and it is usually with high pressure in the eye. There are different types of glaucoma. Some are primary there is no other eye condition involved and it can be open-angle or closed-angle. Some are secondary glaucoma because of other eye conditions like inflammation in the eye or previous surgeries etc. But in general, most of the glaucoma patients have no symptoms. Glaucoma detection become a very popular field of study and research. Different approaches are being used by different researchers in Glaucoma classification and detection. In this chapter, we have taken a handwritten dataset from Kaggle which consists of the glaucomatous eye and healthy normal eye. We have applied VGG19, Xception which is a CNN architecture to achieve the model’s performance in terms of sensitivity, specificity, accuracy and F1 score. This research outcome will help in the early diagnosis of Glaucoma disease in healthcare industry.