[Mrtrix-discussion] realignment question

[Donald Tournier]

Hi Samuel,

The problem is with the way you're using mrtransform: by default all
it will do is modify the transformation matrix in the header - it
doesn't actually reslice the data. When you combine the supposedly
realigned data with mrcat, the output data set will have the transform
of the first input data set, and all the other transforms are ignored.
Hence your problem: the voxel intensities themselves haven't actually
changed at all...

To actually get mrtransform to do the reslicing, add the "-template
image" option, with your reference data set specified as the image.
The reference is the one you've realigned everything to. Then you
should be able to use mrcat to combine all the data properly.

Hope that helps,

Donald.


2009/6/5 "Samuel Gröschel" <S.Groeschel at gmx.org>:
> Hi Donald and mrtrix-users,
>
> A simple question regarding realignment of diffusion images.
>
> Is there an easy way to realign DTI images and then further analyse them in mrtrix?
>
> I have 3 DW-aquisitions from a subject and combine them using 'mrcat'. But there is (a little) motion between the aquisitions.
>
> The approach I chose is to realign(estimate) the images in SPM and then use the transformation matrix (as 4x4 matrix using the function spm_matrix) to apply in mrtrix using 'mrtransform'. However, this effectly does not change anything, there is still the motion in the resulting combined image.
>
> Is it the wrong approach or just some mistake in using mrtransform correctly?
>
> Thanks for your comments,
>
> Samuel
> _______________________________________________
> Mrtrix-discussion mailing list
> Mrtrix-discussion at www.nitrc.org
> http://www.nitrc.org/mailman/listinfo/mrtrix-discussion
>



-- 
Jacques-Donald Tournier (PhD)
Brain Research Institute, Melbourne, Australia
Tel: +61 (0)3 9496 4078