That's a good question, but I'm not sure there is a good answer to it.
SPM doesn't interpolate images of p-values.
One thing you could try would be to "unsquash" the values to make them
behave more like a Gaussian field, interpolate that and then squash
the result (see eg http://en.wikipedia.org/wiki/Logit ). Otherwise,
you may need to look at the mathematics that generates the statistic
and try to reverse engineer an interpolation scheme. Interpolating
contrast images is easy, but interpolating things like sums of squares
images needs a bit more information than just the sums of squares
image itself.
Best regards,
-John
On 22 October 2010 10:37, Anders Eklund <[log in to unmask]> wrote:
> Dear SPMers,
>
> I've calculated a volume with corrected p-values in Matlab (not SPM) for my
> fMRI dataset (3.75 mm voxels). Now I want to interpolate these p-values to
> show them on a T1 volume (1 mm voxels), how should I do the interpolation? I
> can't use ordinary interpolation since an area of a slice can look something
> like this
>
> 1.00 0.80 1.00 1.00 1.00
> 1.00 0.70 0.80 1.00 1.00
> 1.00 0.01 0.02 0.03 1.00
> 1.00 0.08 0.05 1.00 1.00
>
> and the interpolated p-values with ordinary interpolation then become very
> wrong. How is this handled in SPM?
>
> I know that p-values normally not are visualized, but I want to compare two
> different statistical approaches and their test values are quite diffucult
> to compare.
>
> /Anders
>
> --
> --
> -----------------------------------------------------------------------
>
> Anders Eklund
> Phd student
>
> Medical Informatics, Department of Biomedical Engineering
> CMIV, Center for Medical Image Science and Visualization
>
> Tel: +46 73 6003790 mail: [log in to unmask]
> Fax: +46 13 101902 web: http://www.wanderineconsulting.com/
> -----------------------------------------------------------------------
>
|