retinal-fundus

Production-grade AI system for automated Diabetic Retinopathy detection from retinal fundus images. Features EfficientNet-B4 classification (5 DR grades), Grad-CAM explainability, FastAPI backend, Flutter mobile app with offline ONNX inference, PDF clinical reports, and Docker deployment

Research repository for "Age Prediction and Categorization from Retinal Fundus Images Using Residual Neural Networks"

Advancing Ophthalmic Diagnostics: A Hybrid CNN-Transformer Ensemble with Deep Feature Fusion. 10-class retinal fundus diagnosis fusing EfficientNet-B0, VGG16, and ViT-Base with Grad-CAM explainability (88.99% Acc, 0.8718 MCC).

Research repository for "Exploring the Impact of Alternative Color Spaces in Deep Retinal Age Prediction from Fundus Images"

Research repository for "Decoding Chronological Age from the Retinal Fundus Images: A Deep Learning-based Analysis with Vision Transformers"