Date of Publication :5th May 2025

Abstract: One of the main causes of vision loss in the elderly is age-related macular degeneration (AMD), a degenerative retinal disease. This work puts forward a deep learning-based method using EfficientNet-B5 for automatic AMD categorization from retinal fundus images. The model uses transfer learning and fine-tuning to achieve accuracy while keeping computation speed. An ablation study was done to see how tweaking different layers affects results, in addition to being benchmarked against standard diagnostic methods. Tests show better classification results; this method can become a scalable and trustworthy solution for early AMD identification, helping ophthalmologists with on time diagnosis and treatment planning.

Reference :