Integrating Metaheuristics Methods To Detect Real-Life Glaucoma Problem Using Machine Learning Techniques

Abstract

In glaucoma, the fluid pressure inside the eye rises, causing damage to the retinal nerve fibers, which is a common complication. Once damage to nerve fibers has occurred, it is impossible to regain vision. Glaucoma causes an increase in fluid pressure inside the eye, which damages the retinal nerve fibers, which is a common consequence. Once nerve fibers have been damaged, it is impossible to restore vision. Image processing, analysis, and computer vision techniques are becoming increasingly relevant in medical research as they become more significant in modern ophthalmology. There is no denying that ophthalmology is an interdisciplinary discipline, both in academic study and clinical practice today. Imaging technologies in ophthalmology increase diagnostic and observational capabilities. Fundus photography is the imaging approach that provides the most thorough fundus examination with the least amount of patient engagement and the most simple and inexpensive equipment. In detection of Glaucoma fundus images are vital. Optic disk in fundus images can indicate numerous eye diseases, particularly in cases of glaucoma, and can be used to measure aberrant characteristics. This article describes the categorization and detection of Glaucoma illness using Image Processing and Feature Selection. In this system, fundus photos are used as input. Using the CLAHE method, pictures are preprocessed to increase their quality. Following that, pictures are segmented using the K Means method. This segmentation aids in locating the region of interest in the input image. The Relied algorithm is then used to choose features. It aids in the improvement of categorization accuracy. The SVM-RBF, BPNN, and Nave Bayes algorithms are used for classification. SVM RBF is having better accuracy for classification of Glaucoma disease.