Dear Li Ming,
On Tue, 27 Sep 2022 12:05:03 +0100, Li Ming <[log in to unmask]> wrote:
>Dear CAT12 experts, I am working on a longitudinal VBM study, using the longitudinal segment module of CAT12 for preprocessing, but I have a few questions and would appreciate your help
>
>1. Is it must to perform AC-PC reorient before segment?
No, this is done automatically inside CAT12.
>2. Our data are from Asian people, should affine regularization use European template or East Asian template
The affine registration is not that much different that the defaults should be changed. Furthermore, in CAT12 we use several affine registrations to make this more stable.
>3. Should Crop neck nii data be used, or the original nii data be used,should I use dcm2nii or dcm2niix to convert DICOM?
No corpping please. Leave the images untocuhed. DCM conversion should not affect the data if you always use the same tool.
>4. If there are three time points and two groups, and I want to analyze time and the interaction effect of time and group, how should the contrast matrix of flexible design be designed?
This is explained in the CAT12 manual for two time points and can easily extended (k is number of time points):
T-test
1 0 -1 -1 0 1
or
-1 0 1 1 0 -1
F-Test
[eye(k)-1/k 1/k-eye(k)]
>5. The results of different CAT12 versions are different. Is the result of CAT12.8 more accurate than that of CAT12.7 and CAT12.6?
Yes, there were some major changes in the long. pipeline between 12.6 and 12.8. Please use the most recent version.
>6. Does the longitudinal data have to be preprocessed by longitudinal segment module? Is it possible to use ordinary segment module for preprocessing?
Yes, definitely the long. pipeline which is optimized for that kind of data and necessary.
Best,
Christian
>
>Looking forward to your help, thank you very much
|