High level design to diagnosis glaucoma using CNN

Abstract

To minimize long-term structural damage and irreversible visual loss, glaucoma must be diagnosed early. Glaucoma is diagnosed by assessing visual activity as well as monitoring structural damage to the optic nerve. Transparent laser ophthalmoscopy scanning, and laser polarimetry are examples of computer-based equipment that can aid in the analysis of visual acuity and structure. Most persons with glaucoma do not experience any early signs or pain. Glaucoma must be diagnosed and treated by an eye doctor on a regular basis. Traditional glaucoma detection procedures include an eye doctor analyzing photographs and looking for irregularities. Because the image contains noise and other elements that make appropriate analysis difficult, this method takes a long time and is not always precise. In addition, a machine that has been educated for analysis becomes more accurate than a human analyst. The proposed method will automate this procedure by employing deep learning techniques in image processing to diagnose glaucoma without the need for human intervention.