Name

brain_rsa - Characterise brain activation’s representation dissimilarity matrix (RDM).

Synopsis

brain_rsa [-h] -nif NeuroImageFile -bmask BrainMask
          [-roi ROI [ROI ...]] [-cate Category] [-metric Metric]
          [-zscore] -out Output

Arguments

Required Arguments

Argument	Discription
nif	brain activation image
bmask	brain mask image Each non-zero value will be regarded as a ROI label. The activation pattern in each ROI will be used to calculate RDM.
out	output filename with suffix as .rdm.h5

Optional Arguments

Argument	Discription
roi	Specify which ROI labels in bmask will be used; Default using all labels.
cate	a .stim.csv file which contains category information (i.e. ‘label’ item) If used, do rsa category-wisely that average activation pattern before calculating the distance. And the row/column order of RDM is organized from small to big according to the ‘label’. Note: the ‘label’ here is an item in the .stim.csv file rather than the label in ‘-roi’ option!
metric	Specify metric used to calculate distance. Default: euclidean
zscore	Standardize feature values for each sample by using zscore.

Outputs

The output is a .rdm.h5 file, which contains each ROI’s RDM.

Examples

Each volume in nif.nii.gz is an activation map of each stimulus.
Calculate correlation distance for each pair of stimuli using the activation pattern of ROI1 (voxels with label 1 in bmask.nii.gz) and ROI3 (voxels with label 3 in bmask.nii.gz) successively.
Save results to out.rdm.h5

brain_rsa -nif nif.nii.gz -bmask bmask.nii.gz -roi 1 3 -metric correlation -out out.rdm.h5