Contour detection and deep convolutional neural networks for glaucoma detection

A persistent eye condition called glaucoma causes vision loss. Early illness detection is crucial since there is no treatment for it. Current glaucoma screening methods that measure intraocular pressure are insufficiently sensitive. The solution to this problem, according to recent research, aims to make use of a technique known as deep learning in order to detect and predict glaucoma before the onset of symptoms. When using deep learning principles to segment the optic cup, the learning models that have been taught are combined with the U-Net architecture in order to produce the required results. However, in the existing method parameters of the deep learning are not optimized and it leads to inaccurate results.This study suggested a crucial phase in the process of glaucoma identification using fundus imaging that is thought of as a two-way technique to prevent this issue. First one is preprocessing and the other is the detection of the abnormalities of Glaucoma Retinal. In this work preprocessing is done by using median filter-based noise removal and Contour detection using active contour algorithm.Region of Interest Extraction is done by using Multi- Threshold Segmentation. A Deep Convolutional Neural Network (DCNN) was used to classify the photos in order to find glaucoma. The DCNN’s parameters are improved via Particle Swarm Optimization. Experimental findings demonstrate the suggested model’s efficacy in terms of precision, accuracy, recall, and f-measure.