GLAUCOMA DETECTION USING DEEP LEARNING AND STREAMLIT WITH SEGMENTATION

Glaucoma is a condition affecting the vision of the human eye. This illness is considered an irreversible sickness that causes visual degeneration. So far, many deep learning (DL) models have been built to accurately find glaucoma. So, this research proposes an architecture for effective glaucoma diagnosis based on deep learning and the use of a convolutional neural network (CNN). The CNN can differentiate between glaucoma and non-glaucoma patterns. The CNN generates a hierarchical structure of pictures for distinction. The proposed work may be assessed using six tiers. The dropout method is also employed to provide satisfactory performance in glaucoma detection. The datasets used in the experiments are the SCES and ORIGA.