 |
 |
Hi Anderson, Thanks for the tip. Sorry for the beginner question. Why is there only 1 map per subject, though? Wouldn’t GM structural covariance necessitate a group-level analysis? I am referencing the following section from Coppen et al. 2016 (http://pubmedcentralcanada.ca/pmcc/articles/PMC5094265/), where the authors performed MELODIC on all subjects (patients and controls) to derive spatial component maps of GM covariations. Then, each subject gets a score that reflects network integrity (or the individual expression in each identified network). I am thinking of the workflow 1. Process the images using CAT12 in SPM to produce the modulated GM maps 2. Concatenate the GM maps from patients and controls 3. Run MELODIC on the group to extract the networks and derive the ‘network integrity’ score and perform between-group comparisons. Appreciate any feedback you might have on the method. Best Wishes, Shane,