Dear list,
I am currently trying to do the stats after following John's recipe for
longitudinal data in spm12b. For my between-groups analysis I am doing a
Two-sample t-test and have some questions about the masking.
1. I would like to use implicit masking as I want to make sure that
nonsense voxels (e.g. outside the brain) are not considered. However, as
jd images with a range from negative to positive values are
multiplicated by the c1-averages, after this step voxels with value 0
make sense. Is it valid to add 1 to the jd images before multiplicating
with the c1-averages so that I can use an implicit mask or am I messing
up with one of the next steps by doing so? Or to ask in a different way:
Why is 1 not added to the jd images?
2. When using an explicit mask, which Toolbox is recommended for SPM12b?
3. For this longitudinal approach a threshold masking probably also does
not make sense, even if I would add 1 to the jd images (see above), correct?
4. Is there already a vbm longitudinal example available somewhere?
Thank you!
Vincent Brunsch
|