Hi FSL
My question relates to correspondence by Niklas (my science
supervisor) below, regarding the use of t2 images in topup. I've tried
to perform topup using T2 but the unwarped images after topup come out
very distorted. And I'm wondering whether you guys can say why it
looks this way (please view the screen shots below), and if one could
try to change anything to improve upon the images?
The t2 have been FLIRTing it to the dwi b0 and merged (fslmerge) with
dwi b0 into a single dataset which is the 'imain' into topup with the
following acqparams:
1 0 0 0.05
-1 0 0 0.0001
Standard blip-up, blip-down is unfortunately not an option as the
protocol did not have reversed phase-encoding.
Thank you for any help concerning this!
Regards
Bjorn Astrom
Clinical neuroscience
Umea University, Sweden
On 2/4/16, Jesper Andersson <[log in to unmask]> wrote:
> Dear Niklas,
>
>
> I saw in a previous thread that T2-images can be used in TOPUP if you have
> not sampled the DTI 0-images with reversed phasencoding.
>
> I suppose that implies registering the T2 to the 0-image before using them
> in TOPUP. Do you have to resample the T2 to the 0-image before you do the
> registration or will the registration solve this?
>
> the crucial thing here is that the T2 and the b0 image need to be in the
> "same space", i.e. have the same matrix and voxel size. It is less critical
> that they are registered, though of course they cannot be in completely
> different positions.
>
> The easiest way to get the T2 into the same space as the b0 is by FLIRTing
> it to the b0. You can then either combine that with a bit of registration,
> i.e.
>
> flirt -ref my_b0 -in my_T2 -out my_T2_in_space_of_b0 -interp sinc
>
> or without, i.e.
>
> flirt -ref my_b0 -in my_T2 -out my_T2_in_space_of_b0 -interp sinc -applyxfm
>
> You will then need to create a 4D with my_b0 and my_T2_in_space_of_b0 and
> create an acqparams.txt file where the readout time is _much_ smaller for
> the T2 than for the b0 volumes. After that you should be able to run topup.
>
> And will the resampled T2-image be affected by the intensities in the
> T2-image while doing this?
>
> I am not sure I understand this?
>
> Jesper
>
>
>
> I do not quite follow the instructions on the FSL-page.
>
> Best
>
> Nick
>
>
>
> Niklas Lenfeldt, PhD, MSc, MAJ SWE A
> Clinical Neuroscience, UmeƄ university
> Phone: +46 70 609 66 87
> Email: [log in to unmask]<mailto:[log in to unmask]>
>
>
|