SUPER AGAINST COVID19

Classifying X-Ray images with Neural Networks to extract Covid-19 cases

Part of the https://www.hackthecrisis.at/ online hackathon Final Pitch: https://youtu.be/kgL3QGXr82g

Inspired by and original code from: https://www.pyimagesearch.com/2020/03/16/detecting-covid-19-in-x-ray-images-with-keras-tensorflow-and-deep-learning/

Setup Frontend

Avalible over the following link: https://0b01001101.github.io/covid19-classification/webapp/app.html

The Frontend is a static webpage, hostet on Github Pages. No real setup steps required, exept enable Githb Pages: https://guides.github.com/features/pages/

Illustraions from: https://undraw.co/illustrations - https://undraw.co/license Icons from: https://iconmonstr.com/ - https://iconmonstr.com/license/

Setup Backend

The Backend runs on Python Google Cloud Functions No image data or results are stored on the GCP servers

Some Limetations of Cloud Functions:

• File System is read only, so no images can temporarely be saved
• Tensorflow 2.1.0 can't be installt, the build of the container will fail. Use tensorflow 2.0.0

The following files should be converted in a zip archive for the next step:

• main.py
• requirements.txt
• modelV3.h5

For the Code a Cloud Storage Bucket is needed. See /docs/CreateStorageBucket.png

Setup the Cloud function in the following way: