Dear SPMers,
I am having trouble normalizing my data and need some advice. I have coregistered my T1 and
functional images together.
1- If I estimate the normalization parameters with the T1 image (either using the normalize or
segment functions) and use those parameters to write the functionals and T1, I get large
susceptibility artifacts in the functional images and lose most of my signal from the orbital and
medial temporal areas.
2- If I use the EPI template and the mean functional image to estimate my normalization
parameters, I can preserve most of the functional signal and avoid the susceptibility artifacts.
However, if I apply those same parameters to the T1 image, it looks very distorted and does not
line up well with the functional images.
3- I have also tried using the segment feature on the mean functional image to estimate my
parameters and apply those to the functionals and the T1 I get a similar effect as #2 (preserved
functional signal, distorted T1).
What is the ideal way to both preserve the functional signal while still maintaining proper
localization with the T1 images? Should I simply use separate parameters to normalize the T1 and
functional images?
Any advice is appreciated,
Brion Woroch
University of Illinois
|