volumes_lsq

Compute per-voxel least-squares decomposition using linear model components.

volumes_lsq [options] <file> <lm_file_1> ... <lm_file_n>

DESCRIPTION

volumes_lsq performs per-voxel least-squares decomposition of a MINC volume using a set of linear model component volumes (typically produced by volumes_lm or volumes_pca). The tool finds the least-squares coefficients that best reconstruct the input volume from the provided basis components, and outputs the coefficients.

OPTIONS

--verbose
Print progress information during processing.
--clobber
Overwrite existing output files.
--mask mask.mnc
Restrict the computation to voxels within the specified binary mask.
--fixed n
Treat the first n components as fixed with weight 1 (e.g., for the mean component).

EXAMPLES

Decompose a volume using linear model components:

volumes_lsq subject.mnc model_comp1.mnc model_comp2.mnc model_comp3.mnc

Decompose with a mask and fixed mean component:

volumes_lsq --mask brainmask.mnc --fixed 1 subject.mnc mean.mnc pc1.mnc pc2.mnc

AUTHOR

Vladimir S. Fonov - McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University.

COPYRIGHTS

Copyright &copy; 2011 by Vladimir S. Fonov

SEE ALSO

volumes_lm, volumes_pca