mne.compute_source_morph#

mne.compute_source_morph(src, subject_from=None, subject_to='fsaverage', subjects_dir=None, zooms='auto', niter_affine=(100, 100, 10), niter_sdr=(5, 5, 3), spacing=5, smooth=None, warn=True, xhemi=False, sparse=False, src_to=None, precompute=False, verbose=None)[source]#

Create a SourceMorph from one subject to another.

Method is based on spherical morphing by FreeSurfer for surface cortical estimates [1] and Symmetric Diffeomorphic Registration for volumic data [2].

Parameters:

srcinstance of SourceSpaces | instance of SourceEstimate

The SourceSpaces of subject_from (can be a SourceEstimate if only using a surface source space).

subject_fromstr | None

Name of the original subject as named in the SUBJECTS_DIR. If None (default), then src[0]['subject_his_id]' will be used.

subject_tostr | None

Name of the subject to which to morph as named in the SUBJECTS_DIR. Default is 'fsaverage'. If None, src_to[0]['subject_his_id'] will be used.

Changed in version 0.20: Support for subject_to=None.

subjects_dirpath-like | None

The path to the directory containing the FreeSurfer subjects reconstructions. If None, defaults to the SUBJECTS_DIR environment variable.

zoomsfloat | tuple | str | None

The voxel size of volume for each spatial dimension in mm. If spacing is None, MRIs won’t be resliced, and both volumes must have the same number of spatial dimensions. Can also be 'auto' to use 5. if src_to is None and the zooms from src_to otherwise.

Changed in version 0.20: Support for ‘auto’ mode.

niter_affinetuple of int

Number of levels (len(niter_affine)) and number of iterations per level - for each successive stage of iterative refinement - to perform the affine transform. Default is niter_affine=(100, 100, 10).

niter_sdrtuple of int

Number of levels (len(niter_sdr)) and number of iterations per level - for each successive stage of iterative refinement - to perform the Symmetric Diffeomorphic Registration (sdr) transform. Default is niter_sdr=(5, 5, 3).

spacingint | list | None

The resolution of the icosahedral mesh (typically 5). If None, all vertices will be used (potentially filling the surface). If a list, then values will be morphed to the set of vertices specified in in spacing[0] and spacing[1]. This will be ignored if src_to is supplied.

Changed in version 0.21: src_to, if provided, takes precedence.

smoothint | str | None

Number of iterations for the smoothing of the surface data. If None, smooth is automatically defined to fill the surface with non-zero values. Can also be 'nearest' to use the nearest vertices on the surface.

Changed in version 0.20: Added support for ‘nearest’.

warnbool

If True, warn if not all vertices were used. The default is warn=True.

xhemibool

Morph across hemisphere. Currently only implemented for subject_to == subject_from. See notes below. The default is xhemi=False.

sparsebool

Morph as a sparse source estimate. Works only with (Vector) SourceEstimate. If True the only parameters used are subject_to and subject_from, and spacing has to be None. Default is sparse=False.

src_toinstance of SourceSpaces | None

The destination source space.

For surface-based morphing, this is the preferred over spacing for providing the vertices.
For volumetric morphing, this should be passed so that 1) the resultingmorph volume is properly constrained to the brain volume, and 2) STCs from multiple subjects morphed to the same destination subject/source space have the vertices.
For mixed (surface + volume) morphing, this is required.

New in v0.20.

precomputebool

If True (default False), compute the sparse matrix representation of the volumetric morph (if present). This takes a long time to compute, but can make morphs faster when thousands of points are used. See mne.SourceMorph.compute_vol_morph_mat() (which can be called later if desired) for more information.

New in v0.22.

verbosebool | str | int | None

Control verbosity of the logging output. If None, use the default verbosity level. See the logging documentation and mne.verbose() for details. Should only be passed as a keyword argument.

Returns:

morphinstance of SourceMorph: The mne.SourceMorph object.

Notes

This function can be used to morph surface data between hemispheres by setting xhemi=True. The full cross-hemisphere morph matrix maps left to right and right to left. A matrix for cross-mapping only one hemisphere can be constructed by specifying the appropriate vertices, for example, to map the right hemisphere to the left:

vertices_from=[[], vert_rh], vertices_to=[vert_lh, []]

Cross-hemisphere mapping requires appropriate sphere.left_right morph-maps in the subject’s directory. These morph maps are included with the fsaverage_sym FreeSurfer subject, and can be created for other subjects with the mris_left_right_register FreeSurfer command. The fsaverage_sym subject is included with FreeSurfer > 5.1 and can be obtained as described here. For statistical comparisons between hemispheres, use of the symmetric fsaverage_sym model is recommended to minimize bias [1].

New in v0.17.0.

New in v0.21.0: Support for morphing mixed source estimates.

References

Examples using `mne.compute_source_morph`#

Source reconstruction using an LCMV beamformer

Visualize source time courses (stcs)

Permutation t-test on source data with spatio-temporal clustering

2 samples permutation test on source data with spatio-temporal clustering

Repeated measures ANOVA on source data with spatio-temporal clustering

Cross-hemisphere comparison

Compute sparse inverse solution with mixed norm: MxNE and irMxNE

Morph surface source estimate

Morph volumetric source estimate

mne.compute_source_morph#

Examples using mne.compute_source_morph#

Examples using `mne.compute_source_morph`#