Skip to content

Latest commit

 

History

History
47 lines (37 loc) · 3.9 KB

README.md

File metadata and controls

47 lines (37 loc) · 3.9 KB

McMedHacks project by CaffineOverflow

Project overview

The coronavirus pandemic in 2020 , also known as COVID-19, continues to infect more than 500,000 people daily even after the widespread vaccination against it. The common method used to detect COVID-19 is PCR tests, however some facilities might not have enough equipment and trained professionals for the amount of tests needed. Therefore our team proposes to detect COVID-19 through chest X-rays (CXR), which can minimize transportation of patients with portable radiography units which are easy to disinfect. Our project consisted of two parts.

Our project consisted of two parts.

  • First, using a Convolutional Neural Networks (CNN), our project aims at determining if X-ray images represent healthy patients or COVID-19 patients and propose this innovative method of COVID-19 detection.
  • Second, we used Gradient-weighted Class Activation Mapping (Grad-CAM) to determine which area of the CXR is essential for analyses.

By the end of this project, we were able to determine whether a patient has COVID-19 with an accuracy of (data), found the key area of analysis for the result which is (data), and have made a website for better user experience. Covid dectection website
The full submission on the dev post can be accessed through here

Report

A short report on this project can be found here

Presentation

A 5min presentation is prepared. The recording of the presentation can be accessed from here
Access the slides from here

Code

The CNN model and visualization code is developed in jupiter notebook. Access the code from here

About us

We are a 5 member team participating the 2021 McGill McMedHacks from various backgrounds

Name Github Short Description
Martin Gallois MartinGALLOIS McGill undergrad in Physics (BS Pysics)
Karim Jabbour KarimJabbour Concordia University U2 Bachelors in Computer Science (BS Computer Science)
Yinan Wang Yinan2 McGill Master-thesis in Computer vision and image processing (MSc Electrical Engineering)
Yichen Wu yichenemma McGill U2 undergrad in Software Engineering (BS Software Engineering)
Jessica Zhang SiqiZhang6 University of Toronto U1 undergrad in Computer Science (BS Computer Science)

Overview Tables

Summary

Name Role Contributions
Martin Gallois Software Developer
  • powepoint
  • research
  • report
  • presentation
Karim Jabbour Software Developer
  • CNN model
  • report
  • presentation
Yinan Wang Software Developer
  • visualization
  • Grad-CAM
  • report
  • presentation
Yichen Wu Software Developer
  • website
  • report
  • presentation
Jessica Zhang Documentation Lead
  • documentation
  • powepoint
  • research
  • report
  • presentation