BCIL

Brain Connectomics Imaging Libraries

Installation

1. Requires FSL (>=6.0), WORKBENCH (>=2.0.0), FreeSurfer (>=6.0.0) , matlab (>=2022b), R (>=4.3.0)

• matlab toolbox: DSE decomposition (https://github.com/asoroosh/DVARS)
• R toolbox: ggplot2
• jq
• ImageMagick

2. Donwload zip file of bcil and unzip the file
3. Configure the setting file ([unzipped directory]/bcilconf/settings.sh) for your environments

Default settings are:

export CARET7DIR=/usr/local/workbench/
export HCPPIPEDIR=/mnt/pub/devel/HCPPipeline
export FREESURFER_HOME=/usr/local/freesurfer-v0.0.0
export FSLDIR=/usr/local/fsl
export DVARSDIR=/mnt/pub/devel/git/DVARS-master

Please edit the paths using nano (or whatever editor you like) for your environments. CARET7DIR is the path to Workbench, HCPPIPEDIR to HCP pipeline, FREESURFER_HOME to FreeSurfer, and DVARSDIR to DSE decomposition.

4. Set path to [unzipped directory]/bin

export PATH=[unzipped directory]/bin:$PATH

5. Run commands in the directory, [unzipped directory]/bin

Followings are useful for QCing an individual subject:

$ hcppipe_qc

, which generates images and brain MRI quality metrics (BQM), and for group-wise QC:

$ hcppipe_gqc

, which generates control charts for many BQM across subjects and sites (or protocols).

Note that confidence levels in each chart are currently created by a conventional Shewhart's method based on assumptions of normality (ordinary metrics) or Poisson distribution (count data). Fully non-parametric method is under development for future release!