Malaria Detection

This project is a proof of concept, where images can be uploaded and analysed for Malaria. The model can distinguish between plasmodium falciparum and vivax. We use NIH images to train our model.

Demo

Start-up instructions

Run the docker file from a laptop/ computer (GPU-enhanced will make the model run faster). First build the dockerfile from within the malatec_app/docker folder:

docker build -t malariadetection_webapp .

Now you can run the docker file, which starts up the streamlit app:

docker run -it --rm --name malariadetection_webapp -v /home/fight/Documents/malariadetection_webapp/docker:docker  malariadetection_webapp:latest

Follow the link provided in the console and you should be able to see the streamlit app.

Pipeline

• Powered by Tensorflow and Streamlit

Publicly Available Malria Datasets

Run Cloudflare Tunnel

To access the cloudflare tunnel at demo.malariadetection.ch run. This requires some additional setup.

cloudflared tunnel run

Made with ❤️ in Switzerland ⛰️ ddd