Valentina,
The randomization in the algorithms accounts for about a 0.02% change in thickness results. You can disable randomness (or rather, force seeding the rng with same number) by including the flag -norandomness with recon-all. This flag is available in v3.0.4.
In older versions, do this to disable randomness: create a file called xopts containing this text:
mris_smooth -seed 1234 mris_sphere -seed 1234 mris_fix_topology -seed 1234 mris_ca_label -seed 1234
and then include this flag with recon-all:
-expert xopts
This causes the flag '-seed 1234' to be included with each of the listed binaries, which is what the -norandomness flag would do. If the -expert xopts option is included with each run of the same subject, then the results should be exactly the same (this is how our nightly testing is conducted).
fyi, you can confirm that the flags appear in your runtime commands by including the flag -dontrun with recon-all, which causes recon-all to just printout the commands that it will run.
Nick
On Thu, 2006-12-21 at 10:09 -0500, Durastanti, Valentina (NIH/NINDS) [F] wrote:
Could you please quantify "almost"? could you also please explain me how can I eventually disable the randomization?
I have another question that I've already put on the list but the answer never came, maybe is not clear enough: I have two classes (patients and healthy volunteers) and one variable (age), if I want correct for age the difference in cortical thickness between these two groups how can I set the contrast vector?
Thanks Valentina
-----Original Message----- From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] Sent: Wednesday, December 20, 2006 4:33 PM To: Durastanti, Valentina (NIH/NINDS) [F] Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] variability
not quite, but almost. There is some randomization that you can disable if you wanted to On Wed, 20 Dec 2006, Durastanti, Valentina (NIH/NINDS) [F] wrote:
Hi, I have a question: If I run the same subject twice, without editing, should I aspect to obtain exactly the same results in term of regional and whole cortical thickness or something could change?
Thanks Valentina
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer