Dear Marco,
there is no way to do a slice-wise distortion correction with topup, nor do I think there ever will be. The principle of it would be trivial (under the assumption of no movement) but coding wise it would be a bit of work and mean changing the representation of the field. And I don’t see much application for it.
>
> I would like to perform distortion correction in a slice-wise manner (I am assuming there is no/little motion between blip up and down). Is there any trick for that?
As I say above, no. In some special cases, which I am not sure if it is applicable in your case, we have used a trick to perform 2D estimations. Examples of that has been single slice cardiac images. We have then created a fake 3D stack from multiple copies of the 2D image, run topup with movement estimation turned off (--estmov=0) and then extracted the field from the middle slice ignoring any contribution in the z-direction.
>
> I am fitting a model on a set of volumes to extract some relevant parameters. When I look at the map of the cost function at the end of fitting (sum of squared residuals), it has a weird "zebra" like appearance, meaning that in some slices the fitting is not working as well as other slices. This is not the case when I fit the data without performing distortion correction as first step.
This sounds really strange. I can’t really think what the cause could be. Are these diffusion data? Have you tried turning off movement estimation to see if it is to do with mixing of adjacent slices?
Jesper
>
> Thanks for your help.
>
> Marco
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|