Hi Vincent,

yes, it is tp2-tp1 / time_delta

so thinning is negative, thickening positive. I'll correct the help text, the tutorial is right.

It knows about the time point ordering because of the time column.

Best, Martin

On 08/02/2013 09:04 PM, vbrunsch@nmr.mgh.harvard.edu wrote:

Dear Freesurfer experts,

I have followed the LongitudinalTutorial with two time points + two groups and now, after looking at the output of the qdec analysis, I just want to make sure I haven't made a sign error as for the control group the difference is as expected around 0 but for the experimental group I see negative values (= thinning?) all over the place which seems weird, just want to make sure I got this right:

I set the time variable to 0 for the baseline scan and to 1 for the after-intervention-scan. Is it correct that for do-rate the calculation is then only <thickness of tp2> - <thickness of tp1> and not the other way around? I ask because in the long_mris_slopes -help it says that do-rate would yield the "thinning in mm/time" and if this means that a positive value stands for thinning then this would contradict what I understood from the Tutorial where it says for the rate "thick2-thick1" is calculated, so a positive value should mean "thickening".

Also, how does long_mris_slopes know which one is tp1 and which one is tp2? I assume it takes the first entry in the fsid column of long.qdec.table.dat as tp1 and the second entry below as tp2?

Hm, I see no other possibilities of making a sign error during the whole process, did anyone report a similar issue?

Thanks for your help! Best, Vincent

---

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