This project focused on developing an AI-based system for the detection of Diabetic Retinopathy using deep learning techniques. A comparative analysis was conducted between EfficientNet and Swin Transformer models, evaluating their performance on retinal image datasets. The study also implemented Grad-CAM visualizations to provide explainability and insight into model predictions. While unpublished, the project showcased advanced model evaluation and real-world application of deep learning in medical imaging.