Dear all,
I am novice in the SPM. There are many questions regarding the interpretation of SPM analysis results modulated versus unmodulated structural images. I am studying Alzheimer subjects compared to healthy aged controls. I have used T1 WI images of same size, orientation and scanner, applied DARTEL tool as implemented in SPM8. Proportional global normalization eliminated multiple scattered areas of interest in the unmodulated date and produced better localization of the findings into MTL structures in modulated data. But still even after proportional global normalization, my globally normalized unmodulated data show larger relevant difference areas and have higher diagnostic accuracy than the globally normalized modulated. What could explain such result? Why the unmodulated demonstrated much more affected areas than the modulated? Is the following correct according to the definition of how modulation works: "perfectly registered unmodulated data should show no difference between the groups"?Does it mean that unmodulated data have some compound of registration failures? When I am adding covariates such as TIV, age and sex to both modulated and unmodulated ANCOVA with normalization, there is only slight improvement.( I think it is improvement because after that I am loosing those few voxels which were in the deep subcortical white matter.) I am looking forward to receiving your comments and thank you for your time.
Jargal Dashjamts,
PhD-student
Kyushu University, Japan/ HSUM, Mongolia
|