Hi Livia,

the Paired Analysis wiki page that you cite, contains the steps to create the differences and then run a glm analysis on the difference files. In your case you would to the first steps exactly as on the wiki and then you can create a more complex Second FSGD File with more than one group. For that take a look at the glm tutorial where we have plenty more examples for GLM.

If the time delta between the two time points differes a lot across subjects, you may want to switch to linear mixed effects modeling. Also if you have cases that dropped out (only one time point) or cases with more than 2, mixed effects models will be the better choice. We have matlab code for that (see also the wiki).

Best, Martin

On 05/03/2017 04:53 PM, Livia Liu wrote:

Dear FS experts， I have two group and two times to do longitudinal cross analysis. I looked your answer in https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg45870.html and https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg52032.html Sorry, I can’t understand what you said” you can either subtract the two time points and then do a regular group analysis” "subtract tp1 from tp2 manually and pass the result to mri_glmfit." How to subtract the two time manually and how to do then ? Would you please tell me more details? And in my analysis, if I use the paired analysis, how to design paired.fsgd and paired-diff.fsgd? In http://surfer.nmr.mgh.harvard.edu/fswiki/PairedAnalysis I am sorry to only see how to do in one group. Looking forward to reply, and thanks very much. Kind regards, Livia

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer