COVID-19 X-ray Classifier

Disclaimer: This app is built only for educational purposes and not suitable to use in the real world whatsoever!

Project Overview

A web app to run an image classifier to distinct between COVID-19 Xrays versus normal[1]. The model is based on the Convolutional Neural Network built with Keras with Tensorflow backend[2]. Datasets: [3-4]

Run with Docker 🐳

# 1. Clone this repo 
$ cd COVID19-Classification

# 2. Build Docker image
$ docker-compose build

# 3. Run!
$ docker-compose up

Open http://localhost and wait till the webpage is loaded.

Run locally

• Clone this repo
• Install requirements
• gunicorn -k gevent -w 5 app:app -b 0.0.0.0:5000
• Go to http://localhost:5000
• Done! 🎉

👇 Screenshot:

Tools and Modules

Python – web development:

Flask

Python – CNN:

keras | tensorflow | scikit-image | pandas | numpy | h5py

Web Development:

HTML | CSS | JS

References:

1. Deploy Keras Model with Flask as Web App in 10 Minutes
2. Detecting COVID-19 in X-ray images with Keras, TensorFlow, and Deep Learning
3. Open database of COVID-19 cases with chest X-ray or CT images
4. X-ray images for healthy individuals