Enhanced Glaucoma Severity Detection in Rental Fundus Images using Improvised PCOA (iPCOA) and Convolution Neural Network

Glaucoma, a severe eye disorder that damages optic nerve fibres and astrocytes irreversibly. Early identification of Glaucoma is essential to prevent vision loss. The proposed work presents an Improvised Pulikulam Cattle Optimization Algorithm (iPCOA) designed to improvise the detection of glaucoma severity in retinal fundus images. The iPCOA optimizes Convolutional Neural Network (CNN) parameters to precisely classify images based on optic disc size and shape to distinguish between healthy and glaucomatous conditions. The proposed work is evaluated using MATLAB with diverse datasets and it achieves an accuracy above 95%, surpassing benchmarks in accuracy, sensitivity and specificity metrics which are commonly used for glaucoma diagnosis. This research introduces an innovative tool for advancing glaucoma diagnostics with promising significant contributions to ophthalmic research and clinical practice. Recent advancements in medical imaging and machine learning techniques have emphasized the importance of non-invasive approaches for early glaucoma detection, reducing the burden on healthcare providers and improving patient outcomes. The iPCOA framework not only enhances diagnostic accuracy but also offers a scalable solution adaptable to various healthcare settings, facilitating broader access to reliable glaucoma screening. By integrating these enhanced algorithmic improvements with robust validation on real-world datasets, this study establishes a foundation for future developments in computer-aided diagnostics for ocular diseases.