Hi,
There is an intensity normalisation step in the FSL analysis that scales the mean to a value of 10000. Hence I would expect fairly large values if you are getting a sizeable response.
All the best,
Mark
On 1 Oct 2014, at 15:01, Scott Macdonald <[log in to unmask]> wrote:
> I think I've figured it out my problem. The beta values seemed high to me when I extracted them using fslmeants/featquery only because I was expecting them to be around the same magnitude of betas collected on a different scanner on a different project... which is a false presumption, I believe. It was a coincidence that the average of betas across the entire pe.nii.gz image (as opposed to under my mask) returned a value that resemble the magnitude of a beta I would have extracted on the data from the other scanner.
|