Dear Andrew,
Glad you are liking the new version.
It is true that BBR will do a registration to the structural, and therefore must have a structural image to work with.
However, it does not produce a result in structural space. The algorithm can do this, but we have chosen to keep things in the functional space (but undistorted) in order to be more consistent with the way that FEAT has always worked before, and to keep the file sizes down.
So the filtered_func image will _not_ be in structural space and will require transformation to that space if that's what you want.
If you are following the standard FEAT procedure, you do not need to do anything yourself, as FEAT will do all the necessary transformations/resamplings at the start of the second-level analysis.
All the best,
Mark
On 7 Sep 2012, at 14:51, andrew Persichetti <[log in to unmask]> wrote:
> Hello all,
>
> The new version looks great!
>
> I just wanted to clarify something: Now, when I b0 unwarp my functional data feat uses bbr. The feat user guide says:
>
> 'B0 unwarping is now carried out using BBR, which also performs simultaneous registration to the structural image'
>
> Does that mean that, unlike the previous version of feat, v5.0 will automatically apply the xfm to the main structural image by virtue of BBR's algorithm, at this stage of the processing?
>
> In the registration section of my prestats only processing, I have only specified a main structural image. Will my filtered func images already be in this space? Or will I need to apply the xfm later - before running a second level analysis?
>
> Thanks in advance for your help!
>
> Best,
> Andrew
>
|