A review of the utilization of deep learning technology in diagnosing one such cause, Glaucoma which has non-curable blindness. Deep learning models, especially convolutional neural networks (CNNs) for automatic detection of glaucomatous changes in ocular images. The paper is meant to share a reflection on applicability, efficiency and reliability of these models for clinical use. It also covers challenges on data quality, model interpretability and integration with healthcare workflows. The aim of this review was to present the latest hitherto progress and prospective approaches that is being used in diagnosing glaucoma employing deep learning.