Hi,
I'm afraid that the noresample option doesn't do what you want it to.
It is all about whether resampling occurs internally when looking for
the right transformation - the final output dimensions are *not* affected
by this option.
But you can easily do what you want by firstly running Flirt to get a good
transformation and save the matrix - which the GUI does automatically
(don't worry about the output voxel size for now). Then, use the
ApplyXFM gui to apply this saved transform to your input image, and
specify the size of the output image either by copying the size from
another image, or by setting the number and size of the voxels yourself.
I'd recommend the latter with half the voxel size of the avg152 and twice
the number of voxels (same FOV).
That should do the trick.
All the best,
Mark
On Wednesday, November 6, 2002, at 07:58 AM, Edward Vessel wrote:
> Hi there folks -
>
> I just started using FLIRT to perform spatial normalization, and have a
> question about the "noresample" option.
>
> I have been using the avg152T1_brain file as a standard (which has 2mm
> voxels), and the images I am normalizing are hires anatomical images with
> 1mm
> voxels. I would prefer to keep their high resolution when normalized,
> since
> this will help me better select anatomical regions of interest.
>
> However, when I use the "noresample" option, it doesn't seem to change the
> output resolution, but instead still creates a 2mm voxel dataset. Am I
> doing
> something wrong?
>
> Here is the command line I used:
>
> flirt -noresample -in s02_hiresanat_4FSL_brain.hdr -ref avg152T1_brain.hdr
> -out s02_hiresanat_norm.hdr
>
> thanks!
> Ed
>
> --
> Ed Vessel
> U. of Southern California [log in to unmask]
> Dept. of Neuroscience
> HNB, 3641 Watt Way http://geon.usc.edu/~vessel
> Los Angeles, CA 90089-2520
> (213) 740-6102
