Hi, Josh, I think you can run the autorecon1, subcortseg and no segstats if you don't want the statistics file (aseg.stats). Also, use -openmp 8 can speed up quite a lot by multi-threading. For a regular T1 brain image, I can the running time to get aseg.mgz in about 4-5 hours.

recon-all -i subject autorecon1 -subcortseg -nosegstats -openmp 8

cheers, Natalie

On Thu, Mar 28, 2013 at 12:33 PM, Bruce Fischl fischl@nmr.mgh.harvard.eduwrote:

Hi Josh,

yes, I think that's correct.

Bruce On Thu, 28 Mar 2013, Joshua Lee wrote:

...

In fact, couldn't I stop after calabel if all I want is the aseg.mgz?

Josh

On Thu, Mar 28, 2013 at 11:18 AM, Joshua Lee jkilee@ucdavis.edu wrote: Hi All,

  Am I right in reading the dev table that if I just want the
  aseg.mgz, (not the stats output), I only need to run auto-recon1
  and auto-recon2?

  I need to run 250+ brains just to get the aseg volume, so if I
  can shave off 4-10 hours off each that would be huge.

---

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.

nope, you wouldn't benefit any further I don't think On Thu, 28 Mar 2013, Joshua Lee wrote:

Hey thanks,

I haven't worried about multi threading typically since I just run more brains simultaneously. On my computer, I have sufficient 64 gigs RAM (64 gigs) and 24 CPU/cores (not sure which)  to run 20 recon-all scripts simultaneously at full speed. Would I still benefit from the multi threading? I am a little confused about cores vs cpus

I created a copy of the recon-all script that stops at calabel step, and am running that now.

Josh

On Thu, Mar 28, 2013 at 12:58 PM, Natalie Han zhan1@stanford.edu wrote: Hi, Josh, I think you can run the autorecon1, subcortseg and no segstats if you don't want the statistics file (aseg.stats). Also, use -openmp 8 can speed up quite a lot by multi-threading. For a regular T1 brain image, I can the running time to get aseg.mgz in about 4-5 hours.  recon-all -i subject autorecon1 -subcortseg -nosegstats -openmp 8

cheers, Natalie

On Thu, Mar 28, 2013 at 12:33 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote: Hi Josh,

  yes, I think that's correct.

  Bruce
  On Thu, 28 Mar 2013, Joshua Lee wrote:

  > In fact, couldn't I stop after calabel if all I want is
  the aseg.mgz?
  > -
  > Josh
  >
  >
  > On Thu, Mar 28, 2013 at 11:18 AM, Joshua Lee
  <jkilee@ucdavis.edu> wrote:
  >       Hi All,
  >
  >       Am I right in reading the dev table that if I just
  want the
  >       aseg.mgz, (not the stats output), I only need to
  run auto-recon1
  >       and auto-recon2?
  >
  >       I need to run 250+ brains just to get the aseg
  volume, so if I
  >       can shave off 4-10 hours off each that would be
  huge.
  >
  >
  >
  >

---

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.