I'm curious. Would you recommend the same procedures if the scan time points were greatly distanced apart (e.g., baseline, 6 wk, 1 yr)? I suspect not, but just wanted to double-check. I've taken the tact of realigning and coregistering within each scan visit, but then moving all to common coregistration based upon the mid-point visit (6 wk scan).
Regards,
Jeff Browndyke
-----------------------------------------------------------------------
Jeff Browndyke, Ph.D.
Clinical & Research Neuropsychologist
Durham VA & Duke University Medical Centers
[log in to unmask] / [log in to unmask]
office: (919) 286-0411 ext. 4656
cell: (336) 264-4222
On Feb 16, 2013, at 5:14 AM, "H. Nebl" <[log in to unmask]> wrote:
> Dear Drew,
>
>
> there was an almost identical question recently, see https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1302&L=spm&F=&S=&P=15936
>
> In your case, I would suggest to realign the 12 functional sessions onto each other (one batch, 12 sessions), and also realign the 12 structural volumes to get a mean structural volume. Then coregister the structural mean onto the functional mean, normalize (or maybe better segment) the structural mean and reapply these normalization parameters onto the functional volumes.
>
> This should lead to better overlap of the different sessions, because you use within-modality registrations (= realignment) based on the whole data set, instead of preprocessing separately and then somehow coregister the means onto each. And you should get nice structural means with high SNR this way, which might increase quality of the segmentation.
>
>
> In case the raw data look very different between sessions or days (due to very different placement inside the scanner and thus different susceptibility artefacts, deformations, ...), you could try to preprocess the different days separately. It would be a good idea to create a structural mean out of the two volumes of a particular day still.
>
>
> Oblique slice orientation shouldn't matter for SPM in general, but make sure that the orientation of the volumes are somewhat in agreement with MNI space (otherwise you might have to manually reorient the volumes onto AC - PC).
>
>
> Hope this helps,
>
> Helmut
|