Hi Amirhossein,
there are lots of reasons, why you see thickening, here are a few:
- scanner hardware or software update
- different hydration levels
- motion artifacts
- processing errors (e.g. skull strip failure etc)
- gradient non-linearities
- intensity inhomogeneities (most of these should
Hi Martin,
Thanks for your instructions. In my analysis when we look at Two Stage
Model results within our control group , we see increase in thickness
(thickening) more than decrease (thinning), where we expect to see the
thinning effect. Is there any explanation for this? I have 19 controls with
Hi,
yes, the first is to stack all the data into a single file (on your study
average, usually that is fsaverage). The second steps smoothes the data. You
can use different smoothing levels, depending on your data. Usually 10 or 15 is
a good number.
The data will automatically be taken from
Hi and thanks Martin for your help,
I am trying to do mixed effect analysis, should I use the registered to
template data for this ( sub1-t1.long.tempsub1 and sub1-t2.long.tempsub1)
and how should the aded.table.dat look like? Do I need to run :
mris_preproc --qdec-long qdec.table.dat --target
Hi Amirhossein,
The best way to compare would be to load all data from the longitudinal
directories (both time points), into a mixed effects model. We recommend LME
for this kind of analysis (they are usually more powerful than the simpler
2-stage approach). One problem in your design is that
Hi FreeSurfer,
What would be the best way to compare the patient group before and after
treatment with a control group? Should one use the tp1 and tp2 registered
to the template from longitudinal analysis and compare those with the
control group?
Best regards,
Amirhossein Manzouri
