Mike,
I have run such tests in the past (which showed about 2-3% variability in hippo volume due to randomness) and Doug has done similar tests, but admittedly running a large scale analysis and showing the results on a wiki page would be useful. Something like running our Buckner40 and/or ADNI60 data set, making 20 copies of each subject and including the -randomness flag, and plotting the mean/std of the volume and surface rois.
Nick
Nick might have, not sure. Are you volunteering Mike :)
On Wed, 20 Jun 2012, Michael Harms wrote:
I'm curious: For comparison to the results in that paper, has anyone quantified the variability that results when one runs the same FS version repeatedly on the same subject, but with a different random seed each time? That is, how much of the difference is related to math libraries vs. intrinsic variability that arises from the components of FS that use a random seed?
cheers, -MH
On Wed, 2012-06-20 at 10:43 -0400, Bruce Fischl wrote:
Hi Peter
thanks for the info. Feel free to post this on one of the *many* blogs howling for our blood :) I think that the effects in the paper reflect default floating point settings in gcc on 32-bit vs. 64-bit, although we haven't really investigated as it doesn't seem like a wise use of limited resources (since no one would ever do a study that way in any case).
Bruce
On Wed, 20 Jun 2012, Peter J. Molfese wrote:
We run things on an Xgrid cluster that is now a mixture of Macs running Snow Leopard (10.6.8) and Lion (10.7.x). We found the results given from asegstats2table and aparcstats2table are identical after running 100 subjects on both Mac OS X 10.7.4 and 10.6.8 with Freesurfer 5.1. I also ran a few subjects on the cluster and on individual computers (not on the cluster so to speak) and the results are identical.
Best, Peter
Hi,
The paper entitled
?The Effects of FreeSurfer Version, Workstation Type, and Macintosh Operating System Version on Anatomical Volume and Cortical Thickness Measurements?, PLoSONE, Vol 7(6), e38234 (2012)
may be of interest to all of you. It can be found at:
http://dx.plos.org/10.1371/journal.pone.0038234
Cheers, Ed
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer