Matt,
Why is it not a good idea to use FA to constrain the tracking?
I am using some fMRI activation points as seeds and targets. Basically, just
trying to see if any tracts run through there.
I guess I am used to using the classification targets since that is what I
have used in the past when using probtrack, but then again, I used to always
specify a few targets masks. From your understanding, how is it that
including multiple masks differs from using termination masks (aside from
that tracts terminate when reaching the target) and classification masks?
Maybe you can also help me with a related question. To help me better define
the seeds and targets I segmented the T1 image and used the GM as a mask on
the rendered activation map in order to cut away the WM and the artifactual
"activation". However, I am not sure if this is the best way to do this
since the activation map is a 2nd level map and is in standard space (I of
course had done the appropriate transform for the segmented mask though).
Would it be better to just segment a standard image and apply the GM from
that as a mask? I had also tried transform the standard space activation map
to the subject's space but that didn't work.
Thanks,
Carlos
|