codeamt / FastEfficientCovidDeployment

A COVID-19 Chest XRay classification model deployed as a Streamlit app

Fast Efficient Covidnet

Streamlit inference service deployment submodule for Udacity's Machine Learning Engineer Nanodegree program.

DISCLAIMER: THIS TOOL SHOULD NOT BE USED FOR MEDICAL DIAGNOSIS/REPLACE CONSULTING FROM A MEDICAL EXPERT AND SHOULD SERVE EDUCATIONAL PURPOSES ONLY.

Related Repos/Files:

• All repos
• Data Engineering
• Model Training
• Technical Report

About

This is a Chest X-Ray (CXR) classification API. Building on previous work of [1], the CovNet model for this ML project utilizes a pre-trained EfficientNet-b1 to extract features and a fine-tuned Fast.ai classifier to differentiate between infection classes (Normal, Viral Pneumonia, or COVID-19) with 95% test accuracy.

Build Instructions

Locally

Clone this repo:

git clone https://github.com/codeamt/mle-capstone-deployment FastEfficientCovidnet 
cd FastEfficientCovidnet

Install packages:

cd src
pip3 -r install requirements.txt 

or Dockerized:

docker build -f Dockerfile -t app:latest .

Running

Locally:

From the src of the repo:

streamlit run app.py

with Docker:

docker run -p 8501:8501 app:latest

References

1 COVID-Net: A Tailored Deep Convolutional Neural Network Design for Detection of COVID-19 Cases from Chest X-Ray Image. L. Wang and A. Wong., 2020.