> Is there any way to segment the hippocampus from a MR conventional IR
> sequence image? What kind of template (T1, T2, ...) should I try? Is there
> any method to improve the image quality?
There's a whole universe of people who spend their lives trying to develop
automated hippocampal segmentation approaches. SPM can do a vaguely good job
of finding grey matter, but then you'd need to seperate the hippocampal grey
matter from the other stuff. Finding out what works best for a particular
type of data needs to be done by empirical exploration, but I would
personally try doing it using SPM5.
Some of the new stuff I'm working on may eventually be applicable to automated
hippocampal segmentation, but it's quite a long way from being ready for
general consumption (also needs writing up first).
Best regards,
-John
|