Dear FSL list,
 I have completed my voxel-wise analysis between three groups (two patient and one control) and I have very promising group data.  I therefore was curious about the individual subject data and created Z-scores for all my patients using the mean and standard deviation from my control group.  I was  surprised to find that none of my subjects seemed to deviate from my mean (by a large amount) and at first I believed this was due to subtle differences that were only being detected at the group level.  However, I forgot that in my original design matrix for the group comparison I used age and gender as covariates which brings me to my question.  Is there a way to regress out the effects of age and gender on the individual FA skeletons so that I can generate Z-scores based on the adjusted data?

Thank you,
Jeremy