In my experience it is better to do it one by one. In case of a mistake you see it immediately for every separate subject. But this does not mean, that you have to create multiple jobs using GUI every time. Use batch mode.
What I prefer is to create a separate batch file for preprocessing of every subject. You can use a simple matlab script which will set the filenames and the proper parameters of your preprocessing. At the end you should have a cell "matlabatch" with all the parameters stored in it. The structure of the cell is quite complex, but it is easy to discover if you go to View>Show m code in the batch mode when several job files are already opened there.
One more remark: in our laboratory people say that it is better to reorient the images manually so that they are more close to MNI template in advance of reprocessing. This simplifies the subsequent computations for normalization. Reorientation is easy to do in check reg. The MNI template you can find in SPM\canonical.
Good luck!
Vladimir
Volodymyr Bogdanov, PhD
Headache Research Unit
GIGA - Neurosciences
University of Liege
CHU T4 +1, Sart Tilman
Av de l'Hôpital 1
4000 Liege, Belgium
--- On Thu, 3/10/11, Valerie L.B. Ariza <[log in to unmask]> wrote:
> From: Valerie L.B. Ariza <[log in to unmask]>
> Subject: [SPM] multiple sessions
> To: [log in to unmask]
> Date: Thursday, March 10, 2011, 9:41 PM
> Hello!
> I am quite new and ignorant about spm8 secrets, so I have
> some basic questions for which I couldnt find a clear answer
> in the manual.
> I am preprocessing my subjects data (n=40), and I was
> wondering if it was possible to do all the participants and
> jobs at the same time. Someone told me this was called
> multiple session, and that I just had to add new sessions
> (one session for each subject) and then run the analysis? Is
> it the right way to do it? Can I load the data for
> each subject, and then run all the different preprocessing
> steps at the same time or is it better to do it subject by
> subject and job by job? I know we have to follow a specific
> order while preprocessing data, but is it possible to
> "pre-program" this order?
>
> Thanks for your help :)
>
> Val
>
|