External Email - Use Caution
To the kind attention of the FreeSurfer team,
I have some questions concerning a GLM analysis I’m running and any help would be greatly appreciated! I’m interested in identifying cortical regions where older age is associated with greater atrophy. However, unlike the example reported in the tutorial page, I do not have two groups (male and females) but just one group (patients, regardless of their gender).
- Firstly, I’ve generated an “age.fsgd" file with the following structure:
Class Main Variables Age Input subj 1 Main 61 Input subj 2 Main 73 ….
And so on.
- Secondly, I’ve assembled the data using the mris_preproc command (including resampling, concatenating and smoothing).
- Thirdly, I’ve proceeded with the glm analysis using the mri_glm command. However, here I’ve changed the fsgd option, and in particular:
—fsgd age.fsgd doss (I thought the DOSS design was more appropriate in my case). FIRST QUESTION: is this correct?
Then, I’ve set the contrast in the "lh-Avg-thickness-age-Cor.mtx” file as: 0 1. SECOND QUESTION: is this correct? In this case I’ve set the intercept value as 0 and the slope value as 1.
Many thanks for your help,
Pilar