Publication

Abstract : Glaucoma, a progressive eye disease known as the “Silent thief of sight,” poses a significant challenge in early detection and treatment. The conventional method of diagnosing glaucoma involves manual evaluation of retinal fundus images by trained ophthalmologists, which is not only time-consuming but also prone to human error (inter-observer variability) The aim of this study is to propose a novel approach for glaucoma detection utilizing a deep learning architecture called AlterNet-k. The key innovation lies in leveraging the AlterNet-k model, which is specifically designed to capture intricate patterns in medical images. By leveraging deep learning techniques, we pre-process raw retinal images and apply data augmentation to enhance model performance. The segmented AlterNet-k model effectively distinguishes between healthy and glaucomatous retinas, achieving a remarkable accuracy of 96.67%. This demonstrates the potential of deep learning models, such as AlterNet-k, in revolutionizing glaucoma diagnosis and improving patient outcomes by enabling early intervention strategies.