7 includes a global scale factor. Usually your anatomical and functional
images are acquried in such a way that the brain appears the same size in each
and 6 degrees of freedom (3 translations, 3 rotations) is enough, but if for
some reason they weren't (e.g. scanner wrote a bad qform/sform) the 7th degree
of freedom (global scale) will fix that for you. On the other hand, if there is
signal dropout around the border of your functional image, the 7th dof may end
up inflating your functional image to match the anatomical, making it much
larger than it should be, so use it with caution!
On Wednesday, May 16, 2012 23:04:46 you wrote:
> Hello,
>
> I am using FLIRT to align two anatomical images acquired from the same
> subject: T2-weighted anatomical inplanes (grid = 256 x 256 x 44, voxel
> size = 0.750 x 0.750 x 3.000 mm) and a T1-weighted MPRAGE volume (192 x
> 256 x 160 voxels, 0.938 x 0.938 x 1.000 mm). My first inclination was to
> use 6 dof; however, I have also seen people use 7 dof for this purpose.
> Is there any rule that typically guides decisions of whether to use 6/7
> dof for this kind of intra-subject alignment? I tried both options and
> any differences in the resulting .mat files appear to be very small.
>
> Thanks for your time,
>
> Karin
|