Diabetic Retinopathy Disease Detection Using TensorFlow

Diabetic Retinopathy Disease Detection Using TensorFlow

Mr. Omkar V. Khute, Om Takate, Shriyash Sonawane, Chetan Ahire, Abhay Uchit

1.    Mr. Omkar V. Khute, lecturer Information technology, Mahavir Polytechnic, Nashik

2. Mr. Om Takate, Student, Information technology, Mahavir Polytechnic, Nashik

3. Mr. Shriyash Sonawane, Student, Information technology, Mahavir Polytechnic, Nashik

4. Mr. Chetan Ahire, Student, Information technology, Mahavir Polytechnic, Nashik

5. Mr. Abhay Uchit, Student, Information technology, Mahavir Polytechnic, Nashik

---------------------------------------------------------------------***---------------------------------------------------------------------

     Abstract –

Diabetic Retinopathy (DR) is a serious eye condition caused by prolonged diabetes, leading to vision impairment and blindness if untreated. Early detection is critical for preventing severe damage. Traditional diagnostic methods require expert ophthalmologists, making the process time-consuming and expensive.

This project aims to develop an automated DR detection system using deep learning, specifically Convolutional Neural Networks (CNNs) in TensorFlow. The system processes retinal images, extracts features, and classifies them into different DR severity levels. The model will be trained on public datasets (APTOS 2019, Messidor, IDRiD) to ensure high accuracy and generalization.

The outcome of this research will provide an AI-based screening tool that assists ophthalmologists, improves early detection, and reduces diagnosis time.

Key Words: Diabetic Retinopathy, Deep Learning, TensorFlow, CNN, Fundus Imaging, AI in Healthcare