This is a containarized version of the tool developed by Michael Bernier. Credits go to him for the great work! See also his paper:
Bernier, M., Cunnane, S. C., & Whittingstall, K. (2018). The morphology of the human cerebrovascular system. Human Brain Mapping. https://doi.org/10.1002/hbm.24337
After cloning this repository, cd into its folder and build the Docker image using:
sudo docker build --no-cache -t <<YOUR PREFERED IMAGE TAG>> .
Or pull it from my Dockerhub repository as a Singularity container using:
singularity pull --name <<YOUR PREFERED IMAGE TAG>>.simg Docker://rhaast/braincharter
It's mostly the same procedure to run the pipeline as the original. Inside your ToF or SWI data folder, run the following command:
docker exec <<MOUNTING OPTIONS>> <<your image tag>> $PATH_TO/extract_vessels.sh <<PARAM1>> <<PARAM2>> <<PARAM3>>
singularity exec <<MOUNTING OPTIONS>> <<your image tag>>.simg $PATH_TO/extract_vessels.sh <<PARAM1>> <<PARAM2>> <<PARAM3>>
Where:
- Param1: name of the file (ex: ToF.nii.gz -> ToF)
- Param2: suffix of the data (ex: ToF.nii.gz -> .nii.gz)
- Param3: type of the data (ex: TOF or SWI or OTHER) -> SWI means dark vessels, ToF means bright vessels, OTHER won't skullstrip and skip some process and assume bright blood.
- (Optional) Mounting options: e.g. -B <> ...
So, for example:
docker exec rhaast/braincharter:latest bash extract_vessels.sh ToF 'nii.gz' TOF
or
singularity exec braincharter-vasculature.simg bash extract_vessels.sh ToF 'nii.gz' TOF
- Make it compatible with BIDS data format and convert to BIDS-app