[pymvpa] Multi Modal Analysis

[Jeffrey Stout]

Hi MVPA users,

I am trying to perform multivariate analysis on multi-modal data (T1,
FLAIR, and PET).  All have been warped to MNI space and resliced to 2mm
isotropic voxels.  I may be wrong but I haven't seen much documented on the
website integrating multiple modalities into an mvpa analysis. (I have seen
separate analysis of different types of data (fmri or meg) but not
integrated into a single dataset/analysis).

Currently I have stacked the images in the Z direction into one nifti file
using fslmerge.  I also had to stack 3 copies of a brainmask in the Z
direction to properly mask out the brain regions.  Originally, I had
thought of just concatenating the data into time dimension so that I could
use a single mask - but I am not sure if this is appropriate or not.  The
other reason for not using the time dimension is that I would like to add
TBSS data from fsl into the analysis which would necessitate a different
mask (something that I could do if I stacked the data in the Z direction).

The data is reading in correctly and preliminary analyses seem to be
working.  I know that this is not an optimal solution.  Is there a better
method already packaged into mvpa?


Thank you,
Jeff Stout
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/attachments/20140409/0b3fc4e1/attachment.html>