Dear Christian Gaser,
I am trying to analyse longitudinal VBM data to assess 6-month treatment effects on GM volumes compared to a waiting list.
I started of with the pairwise longitudinal registration in SPM but then discovered CAT12 which is recommended for short term training effects. I have a few questions:
- What is the major difference between the SPM-registration and CAT12? Should I indeed use CAT12 for 6-month treatment effects?
- I noticed that the SPM-registration corrects for the time difference of the scans. Time differences between scans in my sample varies between 3 and 10 months (although the majority is about 6.5 months). Would you recommend to correct for this or not?
- I would like to create individual GM-difference images for further analysis (i.e. to assess whether treatment success is associated with regional GM-changes in a single or multiple regression analysis). For the SPM-registration it is suggested to multiply the jd-difference images with the GM-segments and then normalize and smooth these images. But this image is not created with CAT12. What would be the procedure for this in CAT12?
- I would like to save the jacobian determinant and deformation fields, but I can't find this option in the longitudinal batch-job (see below). How do you save/find these?
matlabbatch{1}.spm.tools.cat.tools.long.subj.mov = {T1_image1; T1_image2};
matlabbatch{1}.spm.tools.cat.tools.long.opts.tpm = {'.../spm12/tpm/TPM.nii'};
matlabbatch{1}.spm.tools.cat.tools.long.opts.affreg = 'mni';
matlabbatch{1}.spm.tools.cat.tools.long.extopts.APP = 1;
matlabbatch{1}.spm.tools.cat.tools.long.extopts.LASstr = 0.5;
matlabbatch{1}.spm.tools.cat.tools.long.extopts.gcutstr = 0.5;
matlabbatch{1}.spm.tools.cat.tools.long.extopts.cleanupstr = 0.5;
matlabbatch{1}.spm.tools.cat.tools.long.extopts.darteltpm = {'.../spm12/toolbox/cat12/templates_1.50mm/Template_1_IXI555_MNI152.nii'};
matlabbatch{1}.spm.tools.cat.tools.long.extopts.vox = 1.5;
matlabbatch{1}.spm.tools.cat.tools.long.output.surface = 0;
matlabbatch{1}.spm.tools.cat.tools.long.modulate = 1;
matlabbatch{1}.spm.tools.cat.tools.long.dartel = 0;
Thanks a lot!
Marieke
|