README.md

metadata

title: DiabeticRetionPathyDetection
emoji: 🚀
colorFrom: red
colorTo: red
sdk: docker
app_port: 8501
tags:
  - streamlit
pinned: false
short_description: Streamlit template space

Diabetic Retinopathy Detection

A Streamlit-based web application for detecting diabetic retinopathy from eye fundus images using a deep learning model from Hugging Face.

🔍 Overview

This application uses a pre-trained deep learning model to classify eye fundus images into different stages of diabetic retinopathy. The model is hosted on Hugging Face and integrated into a user-friendly Streamlit interface.

🚀 Getting Started

Using Docker (Recommended)

1. Build the Docker image:

docker build -t diabetic-retinopathy .

2. Run the container:

docker run -p 8501:8501 diabetic-retinopathy

3. Access the app at http://localhost:8501

Without Docker

1. Install dependencies:

pip install -r requirements.txt

2. Download the model:

python download_model.py

3. Run the Streamlit app:

streamlit run src/streamlit_app.py

🧠 Model Information

• Model Name: diabetic-eye
• Repository: Asmaa111/diabetic-eye
• Framework: PyTorch
• Input: Eye fundus images (JPEG/PNG)
• Output: Classification into retinopathy stages

📂 Project Structure

DiabeticRetionPathyDetection/
├── .streamlit/            # Streamlit configuration
├── src/
│   └── streamlit_app.py   # Main application code
├── Dockerfile             # Docker configuration
├── download_model.py      # Model download script
├── requirements.txt       # Python dependencies
└── README.md              # Project documentation

🌐 Live Demo

Try the live version hosted on Hugging Face Spaces:
DiabeticRetionPathyDetection Demo

📚 Resources

• Streamlit Documentation
• Hugging Face Transformers
• Diabetic Retinopathy Detection Research