A Systematic Survey on the Glaucoma Eye Disease Detection using Enhanced DL and Modified Classification Techniques

This survey study presents a concise literature review on glaucoma eye disease detection using enhanced deep learning and modified classification techniques. The selected works highlight the evolution of automated glaucoma diagnosis from traditional methods to advanced deep learning models utilizing fundus images, optical coherence tomography (OCT), and hybrid approaches. The reviewed studies demonstrate significant improvements in detection accuracy, early-stage diagnosis, segmentation of optic disc and cup regions, and model interpretability. Key advantages include high diagnostic performance, availability of public datasets for benchmarking, and the development of lightweight and real-time applicable models. However, common limitations such as lack of dataset diversity, high dependency on costly imaging devices, limited generalization across populations, and challenges in explainability are still evident. Overall, the survey emphasizes the need for robust, costeffective, and interpretable deep learning frameworks to achieve reliable glaucoma detection in real-world clinical environments.