Sign inSign up

kojix2/covid-net

By kojix2

•Updated about 6 years ago

Image
0

349

kojix2/covid-net repository overview

⁠covid-net-docker

Docker Cloud Build Status Docker Image Size Docker Image Version (latest by date) Source Github

:couple: COVID-Net⁠ - An open source project to find COVID-19 from chest x-ray images using deep learning.

:orange: This docker image contains a pre-trained model. There is no need for additional downloads.

⁠Docker pull or build

⁠pull
docker pull kojix2/covid-net
⁠Or build it yourself...
git clone https://github.com/kojix2/covid-net-docker
cd covid-net-docker
sudo docker build -t kojix2/covid-net .

⁠Docker run

Mount the current directory as a volume in Docker so that you can access the chest x-ray images placed in the current directory.

sudo docker run -it -v $(pwd):/tmp/share kojix2/covid-net bash

Then, you will see the message below.

________                               _______________                
___  __/__________________________________  ____/__  /________      __
__  /  _  _ \_  __ \_  ___/  __ \_  ___/_  /_   __  /_  __ \_ | /| / /
_  /   /  __/  / / /(__  )/ /_/ /  /   _  __/   _  / / /_/ /_ |/ |/ / 
/_/    \___//_/ /_//____/ \____//_/    /_/      /_/  \____/____/|__/


WARNING: You are running this container as root, which can cause new files in
mounted volumes to be created as the root user on your host machine.

To avoid this, run the container by specifying your user's userid:

$ docker run -u $(id -u):$(id -g) args...

⁠Inference

python inference.py \
    --weightspath models/COVIDNet-CXR4-A \
    --metaname model.meta \
    --ckptname model-18540 \
    --imagepath assets/ex-covid.jpeg

Or

python inference.py \
    --weightspath models/COVIDNet-CXR4-A \
    --metaname model.meta \
    --ckptname model-18540 \
    --imagepath /tmp/share/your-chest-image.jpg

Result:

assets/ex-covid.jpeg

ex-covid

Prediction: COVID-19
Confidence
Normal: 0.031, Pneumonia: 0.189, COVID-19: 0.780

⁠Inference Severity

python inference_severity.py \
    --weightspath_geo models/COVIDNet-SEV-GEO \
    --weightspath_opc models/COVIDNet-SEV-OPC \
    --metaname model.meta \
    --ckptname model \
    --imagepath assets/ex-covid.jpeg

Or

python inference_severity.py \
    --weightspath_geo models/COVIDNet-SEV-GEO \
    --weightspath_opc models/COVIDNet-SEV-OPC \
    --metaname model.meta \
    --ckptname model \
    --imagepath /tmp/share/your-chest-image.jpg

Result:

assets/ex-covid.jpeg

ex-covid

Geographic severity: 0.519
Geographic extent score for right + left lung (0 - 8): 4.155
For each lung: 0 = no involvement; 1 = <25%; 2 = 25-50%; 3 = 50-75%; 4 = >75% involvement.
Opacity severity: 0.388
Opacity extent score for right + left lung (0 - 6): 2.329
For each lung: 0 = no opacity; 1 = ground glass opacity; 2 =consolidation; 3 = white-out.

⁠Download pre trained models

https://github.com/lindawangg/COVID-Net/blob/master/docs/models.md⁠

⁠Want to make covid-net-docker better?

  • Report bugs
  • Fix bugs and submit pull requests

⁠Contributors

@sue445, @hareudon, @inductor and ruby-jp slack members.

⁠Disclaimer

COVID-NET is not a production-ready solution.

Tag summary

Content type

Image

Digest

Size

1.5 GB

Last updated

about 6 years ago

docker pull kojix2/covid-net