Dear FreeSurfer experts,
We have a longitudinal study design with up to 5 time
points. However, we both want to do a baseline
cross-sectional analysis and later a longitudinal
analysis. All subjects have been run through the
longitudinal pipeline (only the base was edited). We
want to start analysing baseline cross-sectionally using
QDEC. I have read through the wiki guides but would like
to make sure that we are doing it correctly. We have
done/plan to do the following:
1. The [CROSS], edited [BASE] and [LONG] is
processed.
2. We will create a qdec table called qdec.table.dat.
However, because we want to use the improved
longitudinal runs [LONG] (i.e., because the [BASE] has
been edited] the first column in the QDEC table (i.e.,
fsid) has been changed to include the longitudinal names
(please see below). In addition, we have included the
line SUBJECTS_DIR /my/path/to/subject/data to
let FreeSurfer know that it is the longitudinal
directories that should be used (is this the correct why
to let FreeSurfer know that it is the "longitudinal
runs" which should be used?).
SUBJECTS_DIR /my/path/to/subject/data
fsid Gender Group Age
d115a.long.d115_base 1 1 19
d120a.long.d120_base 0 1 22
...
2. We transfer the default fsaverage to the subject
directory.
3. To presmooth the data onto the target we run the
following command: recon-all -long <tpid>
<baseid> -qcache
When the above steps are done we plan to load it into
QDEC.
Is the above steps correct for doing a
cross-sectional group analysis in QDEC?
Best regards,
Kasper Jessen
MD, PhD student