README.md

metadata

title: RespireX Backend
emoji: 🫁
colorFrom: blue
colorTo: red
sdk: docker
app_port: 7860
pinned: false
tags:
  - django
  - tensorflow
  - medical-ai
  - backend

🫁 RespireX - AI-Powered Disease Detection API

This is the Backend API for RespireX, an AI-powered medical diagnostic application designed to detect respiratory diseases (like Tuberculosis and Pneumonia) and skin conditions using Deep Learning.

It serves as the inference engine for the RespireX Frontend (deployed on Vercel).

🚀 Tech Stack

• Framework: Django (Python) & Django REST Framework
• ML Engine: TensorFlow / Keras
• Model Architecture: EfficientNetB0
• Database: PostgreSQL (Supabase)
• Deployment: Docker on Hugging Face Spaces

🔗 API Endpoints

The backend exposes the following REST API endpoints:

Method	Endpoint	Description
GET	/	Health check / Welcome message
POST	/api/auth/	User authentication (Supabase Integration)
POST	/api/predict/	Upload X-Ray/Image for disease prediction
GET	/api/history/	Fetch past test results

🛠️ Environment Variables

To run this project, the following environment variables must be set in the Settings > Variables and Secrets tab of your Space:

• SECRET_KEY: Django secret key.
• DEBUG: False (for production).
• DATABASE_URL: Connection string for Supabase PostgreSQL (Transaction Pooler).
• SUPABASE_URL: Your Supabase Project URL.
• SUPABASE_KEY: Your Supabase Anon Key.
• AWS_ACCESS_KEY_ID: (Optional) For email services.
• AWS_SECRET_ACCESS_KEY: (Optional) For email services.

🐳 Local Development (Docker)

To run this backend locally using Docker:

1. Build the image:

   docker build -t respirex-backend .
   

2. Run the container:

   docker run -p 7860:7860 --env-file .env respirex-backend
   

📦 Model Information

• Model File: efficientnet_b0.h5
• Input Shape: (224, 224, 3)
• Performance: Optimized for high sensitivity in chest X-ray analysis.

---

Built with ❤️ by the RespireX Team