I hope it can be ignored, although dealing with the extra field of
view would add to the computatin time.
All the registration (rigid and diffeomorphic), as well as the bias
correction, is done by minimising the sum of squares. Unlike mutual
information and other related approaches, this objective function is
not influenced by all the other signal within the field of view.
There will be some effect on the estimation of the amount of noise in
the images, but I hope this is minimal.
Best regards,
-John
On 8 August 2013 09:37, Lukas Scheef <[log in to unmask]> wrote:
> Hi folks!
>
> I wonder if it would be advisable to crop data sets if a lot of neck is
> visible ? Does the neck (including its bright vessels) disturb the bias
> field correction or the halfway registration by any means? Or can it be
> ignored?
>
> Many thanks in advance for any input regarding this matter,
>
> Luke
|