Jung,
> I just start use the snpm and have the stupid problem in the setup
> statistic model. I have 12 subjects' PET, all in the rest condition. One
> subject is patient and others are controls. I try to find the different
> region between this subject and controls. So I think it maybe fit the
> snpm statistic model ?
Yes, you can analyze this type of data with SnPM.
> I choice multiple subjects, one conditions, one scan per subjects and
> go on the following setting.( is it right ? or I should choice multiple
> groups, one scan per subject ?)
Use two groups, one scan per subject. One of your groups will
consist of a single subject.
> Than after compute and I try to see the result. I choice +ve and
> -ve effect and try to see the MIP figure. It shows the permutation
> distribution figure. But when I click the "return" I can not see the
> MIP projection. However, in the previously write down image portion,
> I can use the display in spm99 and see the figure ? why ?
You need to *enter* return, that is, just press return in the
Matlab console window. Sorry for the awkward interface; this
is required so that you have a chance to look at the permutation
distribution.
Hope this helps.
-Tom
-- Thomas Nichols -------------------- Department of Biostatistics
http://www.sph.umich.edu/~nichols University of Michigan
[log in to unmask] 1420 Washington Heights
-------------------------------------- Ann Arbor, MI 48109-2029
|