Q Score Analysis

Docker Build + Run

cd QScore
docker build . -t q_score_container
docker run -t -v path/to/incoming:/incoming --name q_score -e PYTHONUNBUFFERED=1 q_score_container

The Dockerfile expects a mount to /incoming, as two of the ENV variables point to it (NIFTI_DIR and DATA_DIR)

• Note: if the NIFTI output is not going to be to /incoming and instead a different mount, please change the Dockerfile's NIFTI_DIR ENV variable and also make sure the location is passed in as a volume during the docker run

PYTHONUNBUFFERED is set to 1 to allow output to the console for debug purposes.

Usage

Once the container is running, you can start feeding it data. To kick off an analysis, you first need to place a NIFTI file into the local file location that /incoming is mounted to. This will trigger the NIFTI listener to start an analysis. The NIFTI file must be a gunzipped NIFTI, and must be named with the series description. The permutations and feat analyses will occur on the NIFTI file.

The number of permutations is set in the q_score/permutations/permutations.py file.

Eventually, the program outputs a q score which is then posted to "http://localhost:5000/data/q_score" and a complience score which is posted to "http://localhost:5000/data/compliance_score"