MVPA using ANTsR?

[ptsii]

Apologies if this isn't the right place to ask this question: It seems that ANTsR (and leveraging R generally) has the tools to allow one to do an MVPA (multivariate pattern analysis) with fMRI data. Has anyone done this, and if so, can you point me to any code that has worked out how this has been done?

For context: I have fMRI data where subjects were given 56 stimuli that fall into 2 categories (28 in each). I want to do a MVPA with leave-one-out cross-validation, using a searchlight method (with, say, a 4 voxel diameter sphere).

Any thoughts would be appreciated.

-Tom