Detection of Glaucoma Using Deep Learning Techniques: Literature Survey

Abstract

Glaucoma has emerged as a major cause of vision loss. Glaucoma may be diagnosed by an expert ophthalmologist by assessing the optic nerve head. This procedure is time-consuming and demanding, and it requires a lot of effort. Glaucoma may still be avoided during this first identification stage even if this illness hasn't been well studied. As a result, routine glaucoma screening is both necessary and highly recommended. When it comes to glaucoma detection, machine learning approaches may help. With the help of an Alexnet model trained with an SVM classifier, we have developed an automated glaucoma testing framework. Three publicly accessible datasets were utilised in this study: HRF, Origa, and Drishti GS1. The suggested model was able to accurately classify images 91.21% of the time. This research found that utilising a pre-trained CNN and SVM for disease diagnosis was more accurate than using only CNN or SVM.