Hi Vincenzo,
I was unable to reproduce this problem. I loaded the SPM2 fiducial
surface and mapped a SPM2 volume onto it using the enclosing voxel
method. Then I did the same using average voxel 0 neighbors. The
resulting metric columns were identical (i.e., col1-col2 = 0 for all
nodes). In any event, using average voxels doesn't make sense, unless
you want to build a cube around each voxel and average its contents. If
you just want the value of the voxel where the node is located, then use
Enclosing Voxel.
On 02/01/2007 10:18 AM, vincenzo wrote:
Dear all,
i'm trying to map my fMRI data (from SPM2) on PALS-B12 surface.
I map volume-to-surface using 'MAP to SPEC FILE WITH ATLAS',
my option are:
1)PALS_B12.RIGHT_SumsDB-HomePage_SURF-ONLY.73730 (the spec file)
2)SPM2 (the ATLAS)
3)PALSB12 MultiFid map (73730,SPM2 space).
Then I trying to use 2 different algorithms:
1)AVERAGE VOXEL (with neigh.=0)
2)ENCLOSING VOXEL
If i understood well the 2 methods have to be quite similar
But the projecton on the PALS flat maps results quite different.
There is some one can explain me why?
Am I doing something wrong?
More , when i try to project the data from a meta-analysis, the
results are again different using the 2 algorithms. The projection on
PALS FLAT MAP obtained with AVERAGE VOXEL seems to be more precise,
while the projection of original FOCI of meta-analysis are clearly
within this map.
Thank's much for the help.
--
Donna L. Dierker
(Formerly Donna Hanlon; no change in marital status -- see
http://home.att.net/~donna.hanlon for details.)
