dipy_classify_tissue

Synopsis

Extract tissue from a volume.

Usage

dipy_classify_tissue [OPTIONS] input_files

Input Parameters

• input_files

  Path to the input volumes. This path may contain wildcards to process multiple inputs at once.

General Options

• --bvals_file

  Path to the b-values file. Required for ‘dam’ method. (default: None)

• --method

  Method to use for tissue extraction. Options are:

  • ‘hmrf’: Markov Random Fields modeling approach.

  • ‘dam’: Directional Average Maps, proposed by [1].

  ‘hmrf’ method is recommended for T1w images, while ‘dam’ method is recommended for DWI Multishell images (single shell are not recommended). (default: None)

• --wm_threshold

  The threshold below which a voxel is considered white matter. For data like HCP, threshold of 0.5 proves to be a good choice. For data like cfin, higher threshold values like 0.7 or 0.8 are more suitable. Used for ‘dam’ method. (default: 0.5)

• --b0_threshold

  The intensity threshold for a b=0 image. used only for ‘dam’ method. (default: 50)

• --low_signal_threshold

  The threshold below which a voxel is considered to have low signal. Used only for ‘dam’ method. (default: 50)

• --nclass

  Number of desired classes. Used only for ‘hmrf’ method. (default: None)

• --beta

  Smoothing parameter, the higher this number the smoother the output will be. Used only for ‘hmrf’ method. (default: 0.1)

• --tolerance

  Value that defines the percentage of change tolerated to prevent the ICM loop to stop. Default is 1e-05. If you want tolerance check to be disabled put ‘tolerance = 0’. Used only for ‘hmrf’ method. (default: 1e-05)

• --max_iter

  Fixed number of desired iterations. Default is 100. This parameter defines the maximum number of iterations the algorithm will perform. The loop may terminate early if the change in energy sum between iterations falls below the threshold defined by tolerance. However, if tolerance is explicitly set to 0, this early stopping mechanism is disabled, and the algorithm will run for the specified number of iterations unless another stopping criterion is met. Used only for ‘hmrf’ method. (default: 100)

Output Options

• --out_dir

  Output directory. (default: current directory)

• --out_tissue

  Name of the tissue volume to be saved. (default: tissue_classified.nii.gz)

• --out_pve

  Name of the pve volume to be saved. (default: tissue_classified_pve.nii.gz)

References

[1]

Hu Cheng, Sharlene Newman, Maryam Afzali, Shreyas Fadnavis, and Eleftherios Garyfallidis. Segmentation of the brain using direction-averaged signal of DWI images. Magnetic Resonance Imaging, 69:1–7, June 2020. URL: https://doi.org/10.1016/j.mri.2020.02.010, doi:10.1016/j.mri.2020.02.010.

Garyfallidis, E., M. Brett, B. Amirbekian, A. Rokem, S. Van Der Walt, M. Descoteaux, and I. Nimmo-Smith. Dipy, a library for the analysis of diffusion MRI data. Frontiers in Neuroinformatics, 1-18, 2014.