> we have fmri data with possible strong mvt during acquisition!
> What is the best way to obtain the best preprocessing for those data?
This is a tricky one to give a simple answer for, as it depends on your
experimental design, as well as the artifacts you may or may not have in the
data. You may want to try modelling out the estimated movement in the design
matrix. If you don't want to try this, then you may be able to achieve
cleaner movement corrected data using the realign and unwarp option.
If the movement just occurs during some of the scans, then the offending scans
can be effectively removed from the design matrix by including additional
columns of zeros in the matrix, with a value of one corresponding to the bad
scans.
Best regards,
-John
|