/Pneumonia-detection-in-x-ray-chest

Primary LanguagePython

Convolution-Neural-Network-with-keras-and-django

✍️ Description

Pneumonia is an infection induced by diverse organisms. The seriousness of the disease depends on multiple factors:

  • Clinical symptoms like fatigue, chest pain, shortness of breath.
  • Patient background like a weakened immune system.

Doctors diagnose pneumonia by combining the results of clinical examination with the outcomes of the x-ray chest that should reveal areas of opacity. Our aim in this project is to create an application that uses an AI model, which takes an x-ray chest and classify it whether it concerns pneumonia or not. To do so, we will:

  1. first download our dataset, and preprocess it.
  2. Then, implement several architectures.
  3. Finally, we will choose the best according to metrics.

Project structure

Application

This folder contains the source code of the application named inTouch, which is a web application build with the Django framework. It receives as an input the x-ray chest and returns the diagnosis.

mini demo 👀 : https://cnn-django-intouch.herokuapp.com/

Models

This folder contains all the hyperparameters' configurations tested in this project.

Scripts

A comparative folder that allows to compare between the different hyperparameters and models to pick the best one.

⚡ Usage

You can use this project as a reference if you wish to work with binary image classification and Convolution Neural Network.