Dear SPMers,
I ran a vbm-longitudinal analysis using the new longitudinal vbm approach in spm12 & would like to validate my analysis & results.
In the study we want to examine brain-structural effects of drug-use in a longitudinal design. Therefore healthy subjects at risk were scanned twice with an interval of 12 months. At the initial assessment subjects were drug-naïve, during the follow-up some started use. Effects of the drug on brain structure was assessed comparing users who remained naïve with those who started use (we expected subjects not to differ on the initial assessment, however they might differ on the second assessment).
Analysis:
1. Pairwise registration (order: 1st: initial scan / 2nd follow-up scan)
2. Segment the avg
3. Compute c1.(from step 2) *jd (from step1)
4. Dartel the rc1 & rc2 from step 2
5. Apply MNI normalization from 4 to c1.*jd from step 3
Assessment of effects: 2-sample t-test on images from 5 (c1*jd in MNI) to assess differential changes between the groups. The analysis reveals one significant cluster – extracting ind. Values from the cluster shows positive mean values in the non-users and negative values in those who started the drug. Could this be interpreted as a relative decrease (or lack of increase) in those who started drug use?
In addition is there a way to generate the corresponding GM volume maps for both assessments? This will be useful to explore if there were differences between the groups at the initial assessment.
Thanks in advance & best regards,
Ben
|