Quickstart

1. Install Kubeflow on your favorite cloud provider (or locally).
2. Create a Jupyter Notebook Server.
3. Upload pipeline/DICOM Images to Basis Vectors Pipe.iynb
4. Run all cells- links should appear in the last cell which you can click to see your run output.

Docker Files

rawkintrevo/covid-prep-docim

• Expects to find DOCIMs in /data directory. Reads and converts to 3d matrices, and then vectors. Vectors written to /mnt/data/s.csv.

rawkintrevo/covid-basis-vectors

• Conatains "fat" jar that is built with calculate-basis-vectors/ source. Reads /data/s.csv and outputs basis vectors and linear combinations at /data/drmVt.txt and /data/drmU.txt.

Data

Non COVID-19 CT Scans: https://www.via.cornell.edu/databases/simbadb.html

COVID-19 Positive CT Scans: https://coronacases.org (accessed via other/collect-data-from-coronacases.org)

We're all in this together.

If you would like to collaborate on upcoming works regarding the COVID pandemic, please reach out to papers at aboriginal-armadillo d0t com.