Pneumonia Detection Web App

A Flask web app that detects pneumonia from chest X-ray images using TensorFlow/Keras. Upload an image in the browser to receive a prediction and confidence score.

Features

• Upload X-ray images (PNG, JPG, JPEG, BMP, WEBP)
• Two model entry points (grayscale and RGB)
• Simple, responsive UI with clear results and warnings

Tech Stack

• Python
• Flask
• TensorFlow/Keras
• NumPy
• Pillow

Project Structure

app/
  factory.py
  templates/
    mainpage.html
  static/
  utils/
main.py
mainDense.py
model/
requirements.txt

Setup

1. Create and activate a virtual environment
   • Windows (PowerShell)

     python -m venv .venv
     .\.venv\Scripts\activate
     

   • macOS/Linux

     python -m venv .venv
     source .venv/bin/activate
     

2. Install dependencies

   pip install -r requirements.txt
   

Run

• Grayscale model (default)

  python main.py
  

• DenseNet RGB model

  python mainDense.py
  

Open http://127.0.0.1:5000 in your browser.

Configuration

The app is created via create_app in app/factory.py:

• model_path: path to a Keras .h5 model
• input_mode: grayscale or rgb
• compile_model: set to True if your model requires compilation

Prediction logic:

• A score is produced by the model
• Normal if score >= 0.5, otherwise Pneumonia

Models

• main.py uses model/pneumoniadetection2.h5 (grayscale)
• mainDense.py uses model/pneumoniadetection_densenet.h5 (RGB)

If the model files are not present, add them under the model/ directory. If the files are large (e.g., >100 MB), use Git LFS to store them.

Uploads

Uploaded files are saved to app/static/uploads. This folder is safe to clear.

Disclaimer

This project is for educational and demonstration purposes only. It is not a medical device and should not be used for clinical diagnosis. Always consult a qualified healthcare professional.

Tests

No automated tests yet.

License

Add a license before publishing (MIT, Apache-2.0, etc.).