Dear Jay Chen,
a few SPM defaults are defined in mm and have been set up for a human
brain size, for example the cut off frequency for DCT basis functions.
If you provide a mouse/rat brain, then these values will be too big for
the actual brain size of your animal.
One easy way to deal with that problem (and avoid changing SPM defaults)
is to make SPM believe the images you have are 10 times bigger (if you
assume a rat brain is about 20mm long) than in reality. This is easily
done by increasing the voxel size.
You'll just have to keep in mind that space scaling: 1mm (in SPM) =
100µm.(in reality)
Best,
Christophe
jay chen a écrit :
> Dear all:
>
> I read a paper 'Construction and evaluation of multitracer
> small animal PET probabilistic atlases for voxel based functional
> mapping of the rat brain', Journal of nuclear medicine,vol 47,Nov.
> 2006. The authors said "Data voxel size was scaled by a factor of 10
> to approximately fit the human brain size so that default parameter
> settings in SPM could be applied". Is the scaling process necessary ?
> If I don't perform this step, wil my SPM results be wrong ? Thanks
> in advance.
>
> Best Regards,
> Jay Chen
>
>
>
|