If it would seem reasonable to assume that there was no change of
brain shape between scanning sessions, then I'd suggest simply adding
data from the different sessions as different sessions in the
realignment. This does within-session alignment, and then aligns all
the sessions together.
Best regards,
-John
On 27 January 2012 03:28, Istvan Akos Morocz <[log in to unmask]> wrote:
> dear SPM list,
>
> what may be the optimal procedure to realign/normalize numerous sets
> of fMRI data for the same subject acquired on the same scanner but at
> different dates ? we like to prepare these data for a longitudinal
> brain states analysis - not necessarily in a normalized brain space
> like MNI152 but if possible in a subject-specific brain space.
>
> in this respect it is not clear to me if bread+butter realignment
> would do the job for us, or if a normalization to a subject-specific
> beforehand created template brain is the way to go. we may prefer to
> do the analysis on non-smoothed but realigned and perhaps normalized
> data as outlined above. moreover, our spatio-temporal analysis
> approach is likely fairly susceptible to spatially distributed
> artefacts that either originate in the unavoidable varying field
> inhomogeneities introduced with each new scan date and position - or
> in those introduced with the spatial preprocessing procedures
> addressed here. is there a general consensus in the community about
> such ?
>
> thanks much for feedback, bye, pisti
>
>
> .,-*~'`^`'~*-,._.,-*~'`^`'~*-,.
>
> [log in to unmask]
> http://cafe.spl.harvard.edu
> http://www.pistikem.org
>
> `'~*-,._.,-*~'`^`'~*-,._.,-*~'`
|