Dear Torsten,
I'm very surprised that you get no masking at all.
Have you turned the BET off and set the brain/background
threshold to 0? If you had either of these set differently
then I would expect you to still get masking of your data.
Can you please confirm what settings you have for
brain/background threshold (in the Misc tab) and BET
(in the Pre-stats tab)? And can you check to see if there
is *no* masking or if the mask is just a bit too generous?
All the best,
Mark
On 21 Jul 2011, at 12:18, Torsten Ruest wrote:
> Dear Mark,
>
> thanks very much for this. I have to confess that's not quite what I wanted as it seems that the filtered fmri image now is not masked at all, ie there's still some remains from inproper betting. What I'd like is that the fieldmap mask is not being eroded - it's OK to mask the fmri with the registered magnitude image, ie threshold all non-zero voxels within the magnitude image and use this to mask the fmri. How could I do this?
> I suppose it must be between
> # refine mask (remove edge voxels where signal is poor)
> and
> # now demean
>
> Thanks very much in advance for your advice,
>
> Cheers,
>
>
> Torsten
>
|