Hi Carlos,
> I am just wondering what the DTI processing piepline is when using B0
> correction? I have already unwarped the DTI data and am wondering if I
> should have instead eddy corrected first, then BETed, then applied
> the B0
> correction? Would it be OK to eddy correct unwapred, BETed data?
I think in general you would be better off doing eddy correction
first, and then do the B0 (fieldmap) correction.
The reason for this is that, as a first order approximation, the
field "follows the subject" in the scanner. Hence if the subject
moves during the experiment one would ideally need to register
(align) the fieldmap individually with each and everyone of the
images prior to applying the fieldmap.
If you run eddy_correct first it will correct for the movements (as
well as eddy current-induced distortions) and all the images will be
in the same space. That means that you will only have to register the
fieldmap to one of the images (e.g. the average of your b=0 images),
and then it can be applied to all of them.
If there is no subject movement in your data the order should not
matter.
Good luck Jesper
|