Build context for a Flywheel Gear which runs a specialized version of the ANTs Cortical Thickness pipeline for older patients. The container/gear employs a template for older individuals provided by Nick Tustison and a script for masking out or cropping neck voxels using c3d provided by Paul Yushkevich and Sandhitsu Das.

This gear is based on another Docker image developed by Phil Cook which contains the primary run script. It can be run outside Flywheel. For more information, see the Github repository.

• bids-filter-file: A JSON file describing custom BIDS input filters using PyBIDS. Must contain 't1w' key (note lowercase 't'). See example_filter.json. More info here.
• t1w_anatomy: T1-weighted anatomical NIfTI file. When provided this file will be used in place of any T1w images found in the current session's BIDS dataset.
• mni-labels-*: One or more generic label images in the MNI152NLin2009cAsym space, to be warped to the subject space.
• mni-cortical-labels-*: One or more cortical label images in the MNI152NLin2009cAsym space, to be propagated to the subject's cortical mask. Use this option if the label set contains only cortical labels.

• BIDS-*: (subject, session, acq, run): Specify the minimum set of BIDS fields for uniquely distinguishing exactly one T1w image in the BIDS dataset. Will be overridden by t1w_anatomy input or bids-filter-file input.
• denoise: Run denoising withing the ANTS Cortical Thickness pipeline. (Recommended).
• num-threads: Maximum number of CPU threads to use. Set to 0 to use as many threads as there are cores.*
• run-quick: whether or not to use a quick registration step. (Not recommended).
• trim-neck-mode: three options related to neck trimming:
  • mask: sets neck region to 0
  • crop: crops neck region from image
  • none: uses the unaltered input image

antsCorticalThickness.sh guide and example.

Reference: Tustison et al, 2014.