qsirecon.workflows.recon.mrtrix module

MRTrix workflows

Note that in nipype interfaces the threading-controlling attribute is nthreads, not the typical num_threads expected by nipype. To keep threading consistent between nipype and mrtrix, the nthreads attribute needs to be set in the interface and the n_procs attribute needs to be set on the Node.

qsirecon.workflows.recon.mrtrix.init_mrtrix_csd_recon_wf(inputs_dict, name='mrtrix_recon', qsirecon_suffix='', params={})[source]

Create FOD images for WM, GM and CSF.

This workflow uses mrtrix tools to run csd on multishell data. At the end, mtnormalise is run.

Parameters:

params (dict) – Parameters for the workflow. The following keys are supported: - response : dict. Parameters for estimating the response function.

The following keys are supported:

algorithm : str. The algorithm to use for response function estimation. Supported values are “dhollander” and “msmt_5tt”.

method_5tt : str. The method to use for 5tt segmentation. Supported values are “hsvs”.

fod : dict. Parameters for dwi2fod. The following keys are supported:
- algorithm : str. The algorithm to use for FOD estimation. Supported values are “msmt_csd” and “ss3t_csd”.
- max_sh : list. The maximum harmonic degree of the response function.
mtnormalize : bool. Should the FODs be mtnormalized?

Inputs:

*Default qsirecon inputs*
qsiprep_5tt_hsvs – A hybrid surface volume segmentation 5tt image aligned with the QSIRecon T1w

Outputs:

wm_txt – SH fiber response function for white matter
wm_fod – FOD SH coefficients for white matter
gm_txt – SH fiber response function for gray matter
gm_fod – FOD SH coefficients for gray matter
csf_txt – SH fiber response function for CSF
csf_fod – FOD SH coefficients for CSF
fod_sh_mif – The same file as wm_fod.