Illustrates the complete workflow, from fundus image upload and validation to AI analysis and final screening results
Displays detailed retinal analysis, DR grading, DME risk assessment, Grad-CAM visualization, and clinical recommendations.
NetraX introduces an AI-powered diabetic retinopathy screening platform for early detection and smarter referrals.

Inspiration

Diabetic retinopathy is a leading cause of preventable vision loss. Limited access to ophthalmologists and early screening facilities, especially in rural areas, inspired us to build Netrax.

What it does

Netrax is an AI-powered diabetic retinopathy screening solution that analyzes retinal fundus images to identify potential signs of the disease. It aims to support early detection, risk assessment, and timely medical referrals.

How we built it

We explored deep learning and computer vision techniques for retinal image analysis, using Python, CNN-based approaches, and publicly available retinal image datasets such as APTOS 2019 and IDRiD. Our approach focuses on image preprocessing, model training, classification, and performance evaluation.

Challenges & Learnings

We explored challenges involving image quality, class imbalance, preprocessing, and model performance. This project strengthened our understanding of medical image analysis, deep learning, and AI-assisted healthcare.

Impact

Our vision is to make preliminary diabetic retinopathy screening more accessible, affordable, and scalable, particularly for underserved rural communities. Netrax is designed to support healthcare professionals, not replace clinical diagnosis.