fedi_dmri_outliers

Synopsis

Volume, slice and voxel weighting and outlier detection using multiple methods: SOLID, Gaussian Mixture Model (GMM), SHORE-based, angular, and correlation with neighbors.

Usage

fedi_dmri_outliers [-h] -d <file> -b <file> -a <file> -e <file> -o <file>
                   [-s <file>] [-f <file>] [-m <file>] [-k <file>]
                   [-t <list>] [-c <str>] [-l <str>] [-z <file>]
                   [-n <file>] [-y <file>] [-g <file>] [-r <file>]

Options

Help

  • -h, –help Show this help message and exit

Mandatory

  • -d, –dmri <file> Path to dMRI file

  • -b, –dmrigmm <file> Path to dMRI file for GMM

  • -a, –bval <file> Path to bval file

  • -e, –bvec <file> Path to bvec file

  • -o, –outpath <file> Output directory path

Optional

  • -s, –spred <file> Path to spred file

  • -f, –spredgmm <file> Path to spred file for GMM

  • -m, –mask <file> Path to mask file, required for GMM weighting

  • -k, –maskgmm <file> Path to mask file, required for GMM weighting

  • -t, –thresholds <list> Lower and upper modified Z-score thresholds as ‘lower,upper’

  • -c, –zscoremetric <str> Modified Z-score metric: var, mean, or iod

  • -l, –scalingmethod <str> Scaling method for slice weights: linear or sigmoid

  • -z, –fsliceweights_mzscore <file> Output filename for slice weights using modified Z-score

  • -n, –fsliceweights_angle_neighbors <file> Output filename for slice weights using angular comparison with neighbors

  • -y, –fsliceweights_corre_neighbors <file> Output filename for slice weights using correlation with neighbors

  • -g, –fsliceweights_gmmodel <file> Output filename for slice weights using Gaussian Mixture Model (GMM)

  • -r, –fvoxelweights_shorebased <file> Output 4D .nii.gz file of voxel weights using SHORE-based residuals

References

Snoussi, Haykel, Davood Karimi, Onur Afacan, Mustafa Utkur, and Ali Gholipour. HAITCH: A framework for distortion and motion correction in fetal multi-shell diffusion-weighted MRI. Imaging Neuroscience 2025.