I'd suggest using the unsmoothed data to compute average values from
some brain region. Averaging the values within the ROI should deal
with the aliasing. You could use the smoothed data, but this (sort
of) has the equivalent effect of smoothing your ROI, so you obtain the
average value from some slightly blurry region.
All the best,
-John
On 15 July 2013 09:47, Antoinette Nicolle <[log in to unmask]> wrote:
> Hi SPMers,
>
> Sorry for reposting this same message but I didn't receive a response and I hope someone can advise me.
>
> I'm attempting to use Ged Ridgway's get_totals script to extract grey matter volume from some ROIs but I'm unsure whether it is appropriate to do this with smoothed images or whether I need to have unsmoothed images.
>
> I've used Dartel to get my modulated, normalised, smoothed images but, as I understand it, my images will suffer from "aliasing" artifacts if I set the smoothing to 0, 0, 0. Presumably, then, I need to use smoothed data if I want to use Dartel? But is it appropriate to extract grey matter volume from smoothed images? Perhaps if I were to smooth my images by a very small kernel (e.g. 1.5, 1.5, 1.5)?
>
> Many thanks,
> Antoinette
|