Pneumonia Detection API

This project provides a REST API for predicting pneumonia in chest X-ray images using a pre-trained model built with FastAI.

Table of Contents

• Project Overview
• Setup and Installation
• Usage
• API Endpoints
• Model Information
• Dependencies
• License

Project Overview

This project uses FastAI and Django to create an API that accepts chest X-ray images and returns a prediction of whether the image indicates pneumonia. The model is exported from FastAI and loaded into a Django REST Framework backend to serve predictions.

Setup and Installation

Follow these steps to set up the project locally.

1. Clone the repository

git clone https://github.com/yourusername/pneumonia-detection-api.git
cd pneumonia-detection-api

2. Set up a virtual environment

python -m venv venv
source venv/bin/activate  # For Windows use `venv\Scripts\activate`

3. Install dependencies

pip install -r requirements.txt

4. Add your pre-trained model

Place your exported FastAI model (e.g., exported_model.pkl) in the models/ directory.

5. Run migrations

python manage.py migrate

6. Start the server

python manage.py runserver

The API will be available at http://127.0.0.1:8000.

Usage

You can use the API by sending a POST request to the /predict/ endpoint with an image file of a chest X-ray.

Example request:

curl -X POST -F "image=@path/to/your/xray_image.jpg" http://127.0.0.1:8000/api/predict/

The response will contain the predicted label (Pneumonia or Normal) and the confidence score.

API Endpoints

/api/predict/

• Method: POST
• Description: Accepts an image of a chest X-ray and returns the prediction (Pneumonia/Normal) with confidence score.
• Request:
  • image: The chest X-ray image to analyze (JPEG/PNG format).
• Response:

  {
    "prediction": "Pneumonia",
    "confidence": 0.95
  }

Model Information

This project uses a pre-trained model saved as exported_model.pkl using FastAI. The model was trained to detect pneumonia from chest X-ray images.

How to train the model

To train a similar model, follow these steps:

1. Collect and preprocess the chest X-ray dataset.
2. Use FastAI to build and train the model.
3. Export the model using the learn.export() method.

Dependencies

• Python 3.7+
• Django 3.0+
• FastAI
• PyTorch
• Django REST Framework

Install dependencies with:

pip install -r requirements.txt

License

This project is licensed under the MIT License - see the LICENSE file for details.