Hi Mark,
thank you very much for your help, everything seems to work fine now. The reason why I wanted to use a mask is that the fmri data is truncated, ie a few superior slices as well as the complete cerebellum is missing. That caused the default registration with 12 parameters (somehow needed as the distortions are quite strong) in feat to stretch the image in those regions. The idea was that when using the mask, these regions (cerebellum and top slices) are not considered during registration, so I wanted to make sure that no stretching occurs there. Anyway, the normmi seems to work fine without the mask.
Thanks,
Torsten
|