Dear Elisabeth,
well, there's no simple answer because there are various options for longitudinal analysis. However, in case the data is quantitatively interpretable it is possible to simply subtract the images from different time points / testing for differences in intensity (like comparing two grey matter volumes, beta images obtained from single-subject fMRI models).
Thus you could go with a registration step to register individual T1-weighted "uniform" volumes, e.g. with the longitudinal toolbox (probably turning off the bias regularisation when relying on uniform volumes from MP2RAGE), then you would have to transform *some* volumes into a standard space. An additional smoothing step might be useful before forwarding the files into analysis.
The exact settings / preprocessed files are going to depend on your preferences, see https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=spm;e608f95c.1311 for several options.
Hope this helps,
Helmut
|