SPM "new segment" tissues definition

[bclenet]

When using the new segmentation from SPM in Nipype, we currently write the following lines:

# NEW SEGMENT - Unified segmentation using tissue probability maps in SPM12.
spm_tissues_file = join(SPMInfo.getinfo()['path'], 'tpm', 'TPM.nii')
segmentation = Node(NewSegment(), name = 'segmentation')
segmentation.inputs.write_deformation_fields = [False, True]
segmentation.inputs.tissues = [
    [(spm_tissues_file, 1), 2, (True, False), (True, False)], # Grey matter
    [(spm_tissues_file, 2), 2, (True, False), (True, False)], # White matter
    [(spm_tissues_file, 3), 2, (True, False), (True, False)], # CSF
    [(spm_tissues_file, 4), 3, (True, False), (True, False)], # Bone
    [(spm_tissues_file, 5), 4, (True, False), (True, False)], # Soft tissue
    [(spm_tissues_file, 6), 2, (True, False), (True, False)] # Air / background
]

where:

• spm_tissues_file is the tissue probability map file TPM.nii of SPM
• each element of the segmentation.inputs.tissues represents a probability map ; the second element of each being the Number of Gaussians (cf. SPM manual page 49) for each tissue.

Currently, I'm not able to find if there are default values for the number of gaussians. And if not, which ones we should set.

The manual (page 50) mentions:

Typical numbers of Gaussians could be two for grey matter, two for white matter, two for CSF, three for bone, four for other soft tissues and two for air (background).

[bclenet]

@cmaumet, could you please let me know if you have an idea about that ?
Thanks !

[bclenet]

For now, this is needed by:

• 98BT
• 0I4U
• V55J