• Dibarun Roy Choudhury
• Poonam

Glaucoma is an irreversible neurological disorder that causes intraocular hypertension by increasing aqueous humor and restricting the drainage pathway between the iris and cornea. Damage to the optic nerve head, which transmits visual information from the eyes to the brain, causes vision loss and, eventually, blindness. The term "thief of sight" refers to the difficulty in detecting glaucoma in its early stages. Regular examinations are strongly advised to distinguish it from neurological disorders. Glaucoma diagnosis takes time, money, and is dependent on the availability of resources (trained ophthalmologists and expensive instruments), not to mention the risk of human error. The primary goal of the proposed project is to create a deep learning model for glaucoma diagnosis and automatic severity classification. Our research paper presents a novel glaucoma screening method using deep learning and image segmentation. We combined the architectures of several state-of-the-art convolutional neural networks (CNNs), including ResNet50, VGG16, Xception, ResNet101, Inception, MobileNet, and EfficientNetB7. By analyzing fundus images, we achieved high accuracy in glaucoma detection and severity grading. Our model shows promise as a cost-effective and efficient tool for glaucoma diagnosis, offering potential benefits for early intervention and improved patient care.

Glaucoma; Convolutional Neural Network; Ocular Illness; Deep Learning.

"DEEP LEARNING FOR GLAUCOMA DIAGNOSIS AND SEVERITY GRADING: A COST-EFFECTIVE APPROACH", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 8, page no.e768-e776, August-2023, Available :http://www.jetir.org/papers/JETIR2308488.pdf

"DEEP LEARNING FOR GLAUCOMA DIAGNOSIS AND SEVERITY GRADING: A COST-EFFECTIVE APPROACH", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 8, page no. ppe768-e776, August-2023, Available at : http://www.jetir.org/papers/JETIR2308488.pdf