Dear SPM community,
We have just compared the mask-images of our subjects (1st-level)
normalized with and without prior segmentation and are very surprised to
find large differences in the covered brain volume between the masks. When
segmentation has been conducted, our masks show a large drop out in the
frontal (especially orbitofrontal), temporal, and parietal lobes. So we are
worrying that some of our key regions of interest will be masked out due to
segmentation.
Are there any advices how to deal with this problem?
In particular, we would like to know:
1.) What's the added value of segmentation, if it is accompanied by greater
drop out in certain brain regions at the same time?
2.) We consider changing the spm_default file with regard to the default
value of spm_mask from .8 to .5. Are there any disadvantages we have to
take into account by changing this value? Is there a cookbook approach for
determining the value of spm_mask?
Any advice would be most welcome!
Kristin
|