Hi Jaiashre,
when switching scanners within a longitudinal study you will see effects that are caused by the scanner (or a scanner / age, scanner /disease, scanner / motion interaction). If the majority of your subjects is scanned on the same scanner for most time points, then you may be able to correct for scanner differences by adding a scanner co-var in your LME statistic. This would adjust for a global scanner effect. You may also need to include interaction terms, if there is evidence that the scanner affects different groups/ages/disease severity differently.
Best, Martin
On Oct 31, 2016, at 6:15 PM, Jaiashre Sridhar jaiashre.sridhar@northwestern.edu wrote:
Hi FreeSurfer Team,
We are doing a longitudinal analysis with three time points which are six months apart. The subjects at any time point were scanned on either one of the two scanners (both Siemens 3T TIM TRIO) and we used FreeSurfer v5.1 for our analysis. Most of the subjects show decrease in cortical volume over time except a few of them, who show a huge increase in volume at their 6th-month visit (or at their 12th-month in a couple of cases). When we tried to look at the reconstructed surfaces of the long brains in Freeview, the gray matter and white matter boundaries of the 6th-month were bigger overall than the boundaries of the baseline and 12th-month. Could you share with us your thoughts on what factors (with MPRAGE/Scanner/FreeSurfer) might have caused this problem?
Thank you, Jaiashre.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer