I'll leave 1-3 for Doug and Nick. As for 4, this is a QA step to make sure nothing failed in the spherical registration, or the segmentation/surface reconstruction in that region.
cheers Bruce
On Thu, 11 Jun 2009 nmaleki@mclean.harvard.edu wrote:
Dear all, 1-Is there a way to use mris_anatomical_stats to calculate mean ROI values on the re-sampled and smoothed individual subject data that is produced in the pre-processing step, rather than mapping the ROI back to each subjects’ space? If so what is the command? 2-If you define a ROI where there is statistically significant difference (between 2 groups) at the group level on your average subject and then map the ROI back to each subject’s space and measure the mean value of say thickness in that ROI, should you expect to find the same significant difference between the groups that you are comparing? What does it mean if you don’t for some ROIs? 3- Is there a way to smooth the vertices in a ROI similar to what is done at the preprocessing level?
4- There is a comment on wiki (qdec, ROI analysis) that indicates the usefulness of mapping the ROI back to each ubject's space to 'check the integirty of your reults'. Could you please elaborate more on this? Thank you, Nasim