Hi Matt and Pablo and thanks for the comments!
On Tuesday 31 March 2009 19:45:28 Matt Glasser wrote:
> Ah, I see the difficulty. Presumably one could run a 6 parameter motion
> correction step, apply all angular transforms to the gradient table, and
> then a 12 parameter eddy current correction where no transforms are applied
> (my understanding of eddy currents is that they are mostly global image
> skews and scales, or translations)?
That's also my understanding and that is why I would assume that any rotation
decomposed from the realignment matrix (12 DOFs) is due to the subjects
motion.
As far as I can imagine, the translation and scale can not be achieved by any
combination of the 3D rotation, which would interfere with the real subject
rotation. Not sure about skew.
So with the assumption that eddy currents "only" result in skew, scale, and
translation, and not the rotation, we should still be on the safe side with
rotating bvecs even for the stable subject and substantial eddy currents. In
that situation the 3D rotation should be close to zero. Would you agree?
Best,
Martin
|