Nobody has a clue ??
Hello,
I encountered a problem when using a batch script when analyzing SPECT
images.
I have 3 groups with 2 images per people. I want to do something like
the paired-test but with an indication for the belonging to a peculiar
group. So when comparing group 1 to group2, I used 2 parameters for the
group flags, 2 parameters for the kind of image and n parameters for the
number of couples of images I'm comparing.
When I set the contrast in the xCon part of the SPM structure, Matlab
(7) indicates that it is not a good contrast and that it will be
projected, and I get a vector with something like (0.8, -0.8, 0, 0, 0.2,
-0.2, ...) instead of(1, -1, 0, ...)
Did someone encounter this behaviour or can explain what I'm doing wrong ?
Matthieu Brucher
|