mrtpipelines

MRTrix3 processing diffusion and generating tractography of subject data from data collected.

Contents

• Introduction
• Disclaimer
• Installation
  • Containerized package
• Usage Required arguments Optional arguments
• Support
• References

Introduction

Details regarding usage and workflows coming soon.

• genDhollanderTractography
  • Performs preprocessing to geneate whole-brain tractography following the Dhollander response algorithm.

More information regarding algorithms used can be found from the ¹MRTrix3 website.

Disclaimer

This branch of mrtpipelines is still undergoing development. While the pipeline can be used in its current state, it is possible for the project to undergo major changes.

For HCP datasets, please see the HCP branch.

Installation

Development of this project was written in Python3 and makes use of Nipype.

To install the package on your system, the following commands should be run:

git clone https://github.com/kaitj/mrtpipelines
pip install -r requirements.txt
python setup.py install

Containerized package

This pipeline is also available within a container via both Docker and Singularity.

To use the Docker container, run the following command:

docker pull kaitj/mrtpipelines

To use the Singularity container, users will have to build the container from the recipe found within the container directory. To do so, run the following command:

singularity build mrtpipelines_0.0.3.img Singularity.0.0.2

Note: sudo may be required to pull or build container

It is highly advised to run this pipeline through the available container. Some functionality may be lost if run locally due to custom additions to dependencies, which may yet to be implemented in original software.

Usage

Shown here is an example of the command line interface to run the pipeline:

Usage: genDhollanderTractography <bids dir> <template_fod> <subject list/subject id>

Or if running through singularity:

Usage: singularity exec <singularity_img> genDhollanderTractography <bids dir> <template_fod> /
<subject list/subject id>

Required arguments

bids_dir                Directory with input dataset, formatted according to BIDS

template_fod            A path to the template FOD file for registration of subjects

participant_label       A file containing label(s) of participant(s) to perform pipeline execution on

Note there may be pipeline specific arguments if using a different tracking algorithm (eg. 5-tissue segmentation for ACT pipeline)

Optional arguments

-s      Number of streamlines to generate for each subject(s)

-l      Maxinum harmonic degree(s) for response function estimation (eg. -l 0 8 8)

-w      Work directory.
        Defaults to <bids_dir>/derivatives/MRTrix/work

-o      Output directory.
        Defaults to <bids_dir>/derivatives/MRTrix/out

-n      Number of threads to use for pipeline execution where applicable

-h      Display help documentation

Support and communication

All bugs, concerns, and requests for features can be requested via the github repository found here.

References

[1] J.-D. Tournier, F. Calamante, A. Connelly. MRtrix: Diffusion tractography in crossing fiber regions. Int J Imag Syst Tech 22(2012):53-66.