Comparative Analysis of Deep Machine Learning Models for Identification of Glaucoma from Fundus Images

Glaucoma has recently become a prominent cause of blindness. In most situations, sickness goes unnoticed. To address the challenges, we are promoting an AI method to solving the problems that individuals confront during glaucoma detection and treatment. And our paper will boost the speed of future developments. A few artificial intelligence algorithms have emerged as viable tools for improving clinical activities and workflows in ophthalmology, as well as finding and recognizing patterns related to the necessary disorders such as age-related macula degeneration, glaucoma, and diabetic retinopathy. A huge amount of high-quality, labeled data is essential for the creation of deep learning (DL)-based models for disease diagnosis. However, research on the usage of neural networks for image grouping in glaucoma, a progressive optic nerve condition, has been limited to and not reported on optic nerve to head OCT images, in which nerve fiber layer decreasing examination results are frequently performed in the glaucoma result examination. In this paper, we investigate neural networks’ utility in distinguishing between circumpapillary OCT scans for eyes with and without glaucoma, as well as their ability to train deep learning-based neural networks for glaucoma diagnosis.