There should be 3 frames, so Nx1x1x3. Note that it might be reshaped, so it would be something like N/7 x 7 x 1 x nframes
David Perlman wrote:
OK, so if you load in an mgh file containing (say) thickness data, you'll get a mri struct where mri.vol is Nx1x1 where N is the number of vertices. So far so good. How is the data structured if you load in an mgh file containing surface normals?
On May 7, 2007, at 5:31 PM, Doug Greve wrote:
It's a bit of a hack to store the data in a more general format. When saving in a "volume" format, it sets the number of columns to the number of vertices and the number of rows and slices to 1. This does not work in analyze/nitfti because they cannot represent more than 32k in one dimension (sometimes we reshape to fit). So don't think of these things as volume formats. They are just ways to save arrays of data (most of which happens to be volumes).
-- -dave---------------------------------------------------------------- After all, it is not *that* inexpressible. -H.H. The Dalai Lama
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer