 |
 |
Hi all, I am doing a longitudinal analysis of structural (T1) scans taken before and after an intervention in older adults. I was looking for advice on how to conduct the group level analysis & I came across a presentation (see first and last slides attached), which suggests that I can model normalised & smoothed 'jd_'.nii files (produced from pairwise longitudinal registration) at the group level for TBM analysis, or "combine with seg of avg" to do VBM. For the former I did a multiple regression using the Jacobian rate (jd_ images) and examined the contrast '1' (t-contrast on the mean effect within the regression design matrix as I am interested in the group as a whole) and I am not sure how to interpret the results; do significant clusters represent an increase in volume between times 1 & 2? Secondly, I was wondering how you combine the segmented average image to do VBM - for example, should you average the 'smwc1avg_' with 'jd_' in ImCalc? I would prefer to do a VBM analysis than TBM so if this is the incorrect approach I would really appreciate advice on how I should be doing it instead. Thank you in advance, Georgia O'Callaghan Postdoctoral Research Fellow, School of Psychology and Institute of Neuroscience, Trinity College Dublin Email: [log in to unmask]