Hi,
I've got 20 subjects with DTI data that I am analyzing in relation to some
clinical data. As a secondary/sub analysis, I want to look at 9 of them who
I have follow up data on. My big analysis is an ROI based analysis in which
I run tbss, create ROIs in fslview, mask the skeleton with the ROIs, and
then take the average FA values within the ROI.
I want to see if what I'm doing for the follow-up subjects is right...
basically I took the baseline and follow time points for those 9 subjects
(so, 18 FA maps total), made a new skeleton out of those maps, masked it
with the ROIs, took the average FA values, and am comparing data of average
FA within each ROI for each subject (follow value up minus baseline value).
The reason I'm wondering whether this is right or not is that the difference
scores vary wildly, so I'm not sure whether there is some big problem I'm
not seeing in how I'm doing this? Is there some other approach to a
longitudinal design with ROIs that might be better?
thanks!
Katie
_________________________________________
Katie Karlsgodt, Ph.D.
Post Doctoral Fellow, Neurogenetics Affinity Group
Dept of Psychology/Cognitive Neuroscience
University of California, Los Angeles
phone: 310-794-9673
fax: 310-794-9740
|