Freesurfer

freesurfer@nmr.mgh.harvard.edu

10 participants
30079 discussions

Re: [Freesurfer] reprocessing on FS4.0 takes much longer for some scans
by Pedro Paulo Oliveira Jr 12 Nov '07

12 Nov '07

I'm sure Bruce has much more experience on the subject than me. Also I'm not part of Martinos development team. I'll just add my personal experience: 1) Your running times seens very high. I never saw 72 hours in my studies. Notice that Itanium (and Itanium 2) cpu are far distinct than x86 and x86-64, and even running correctly x86 code its performance in fpu falls in my experience to 50%. Since FS code is not compiled to Itanium arch I think the stated above can explain the problem. 2) Regarding memory my experience is that 1GB usually does fine if you have swap space. 3) about 3.0.5 and 4.0.1 differences in running time my data is: 22 hrs vs 31 hrs average. It depends on the "extended" topological fix 4.0.1 does. It would be nice to know your cluster configuration and a sample of your data to give you a better advice. best regards, Pedro Paulo Oliveira Jr. -- mens. original -- Assunto: Re: [Freesurfer] reprocessing on FS4.0 takes much longer for some scans De: Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> Data: 11/11/2007 21:42 Hi Nicolas, 1. No, sorry, the amount of RAM is sometimes dependent on the individual anatomy, and in any case can't be predefine. 2. Not sure about the itaniums. We have no real experience. 72 hours does sound pretty long. That's I think about what it used to take on our old athlons. Can you extend the time limit? As for the random stopping of recon-all, we have seen that sometimes as well, and are trying to track it down. It seems pretty mysterious, as a binary will exit with a nonzero exit code according to the shell even though the last printf in the code has been executed and the next statement is an exit(0). Bruce On Mon, 12 Nov 2007, Nicolas Cherbuin wrote: > Hi, > > At the beginning of the year I processed 400 scans on a linux cluster. It had a reported ram limit of 1 Gig but it could cope with the slight excess of FS3 and a processing time limit of 48 hours (which was fine for 99% of the scans). Most scans went through autorecon-all without problem. > > I am now trying to reprocess the scans with the new version but I am running into a number of problems. On the same linux cluser processing the same scans, some scans (~30%) run without problem. The rest either fail because they exceed the memory limit or because they take much more than 48 hours (the jobs are being killed and the logs report only the left and sometimes part of the right hemisphere being processed). > > Since the documentation makes clear that FS works best with 2 Gig of ram, I have switched to an Itinium cluster with 2 Gig ram limit and 48 hours processing time limit. When I compare the logs of the same scans processed on both systems the Itinium cluster seems to take longer and although I am still running tests it appears that for at least some scans autoreconall might take 70+ hours. > > Here are my questions: > > 1. On the linux cluster can I tell freesurfer not to exceed a certain ram allocation? and if yes how? > > 2. Do the problems I have on the Itinium cluster suggest that FS is badly configured on this system? and if yes where should we look? (I don't have access to this system's configuration and I have to feedback to the system managers to fix eventual problems.) > > Thank you very much for your help and for sharing these great tools with us. > > Nic > > _______________________________________________ > Freesurfer mailing list > Freesurfer(a)nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > _______________________________________________ Freesurfer mailing list Freesurfer(a)nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

1 1

fMRI raw data time course
by raij＠nmr.mgh.harvard.edu 09 Nov '07

09 Nov '07

Hi Doug, I am troubleshooting an fMRI measurement and would need to see the raw data time course. I have previously used sliceview-sess for this purpose, but for some reason I can no longer make it work. Well, sliceview-sess itself opens fine if I define the analysis and contrast names, but pushing "r" has no effect (should evoke raw data display), and the command line raw data request gives a vague error. I already analysed the data in both bhdr and nii formats, without improvement (a time point exclusion file was used during analysis). -v1 or -v2 makes no difference. Any suggestions would be much appreciated! Bests, Tommi ******************** on machine cuzco nmr-std-env (same problem with nmr-dev-env) cd /space/cuzco/1/users/raij/avml_fmri_mctogether/avml14_session_BRISI12/ setenv SUBJECTS_DIR /space/cognito/5/users/raij/subjects_mri/ setenv SUBJECT avml14 [cuzco:avml14_session_BRISI12] (nmr-std-env) sliceview-sess -sf sessid -df sesspar -raw 011 -motioncor -------------------------------------------------------------- sliceview-sess logfile is /autofs/space/cuzco_001/users/raij/avml_fmri_mctogether/avml14_session_BRISI12/log/sliceview-sess.log -------------------------------------------------------------- INFO: no slice specified, assuming mosaic grep: /autofs/space/cuzco_001/users/raij/avml_fmri_mctogether/avml14_session_BRISI12/session.info: No such file or directory avml14_session_BRISI12 ---------------------------------------------- /autofs/space/cuzco_001/users/raij/avml_fmri_mctogether/avml14_session_BRISI12 yakview -i /autofs/space/cuzco_001/users/raij/avml_fmri_mctogether/avml14_session_BRISI12/bold/011/fmc -r /autofs/space/cuzco_001/users/raij/avml_fmri_mctogether/avml14_session_BRISI12/bold/011/fmc -sn mos -thmin 2 -thmax 7 -title avml14_session_BRISI12 -v1 -mgh ---------------------------------------------- yakview: \$Id: yakview,v 1.11.2.1 2007/10/05 20:42:55 greve Exp $ ERROR: cannot find [cuzco:avml14_session_BRISI12] (nmr-std-env) --- Tommi Raij, M.D., Ph.D. MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging Bldg 149, 13th St Charlestown, MA 02129 U.S.A.

1 0

hardware for computing Q for Paul Raines
by Kirk, Gregory 09 Nov '07

09 Nov '07

Hi Paul, We have been plowing a lot of data with freesurfer, FSL, etc. and are thinking of moving up to a hunkier computer system, anything from 4 processor workstation to a minimal cluster. Wonder if you know a couple URL's of stuff you equipment you think is decent for a starter system G.

2 1

equipment
by Feng-Xian Yan 09 Nov '07

09 Nov '07

Hi, I want to use freesurfer to quantity cortical thickness, and I want to use the software under Linux fedora core 4. What equipment for computer do you recommend? Think you.

2 1

optseq2 questions
by Susan Mosher 08 Nov '07

08 Nov '07

Greetings Freesurfers, I am using optseq2 for the first time to generate a stimulus order for my trials and figure out the ideal ITI between trials. I have a few questions about how this works. 1. I tried specifying the PSD max as 18 seconds and minimum of 2 seconds. My ntp is 216 and my TR is set at 2 seconds. I have 9 events specified, each 14 seconds long with two repetitions of each. Here is the command I ran: optseq2 --ntp 216 --tr 2 --psdwin 2 18 --ev ev1 14 2 --ev ev2 14 2 --ev ev3 14 2 --ev ev4 14 2 --ev ev5 14 2 --ev ev6 14 2 --ev ev7 14 2 --ev ev8 14 2 --ev ev9 14 2 --o test --nsearch 1000 --nkeep 100 In the par files that are generated, many of the NULL time durations are much longer than the PSD max I specified (i.e., I specified 18 seconds and some are as long as 58 seconds). Is this simply because the duration of my trials is so long that 18 seconds is not long enough of a max PSD? Does optseq2 just ignore my max PSD request if it is unrealistic? 2. Overall, I will have nine runs. While I realize that the mean ITI will remain the same in each par file, I want to minimize the duration of any single given ITI. As such, if I pseudorandomly select only 9 par files with the lowest maximum ITIs, is this acceptable? Is there some reason I would not want to do this? 3. If I wanted to change my TR from 2 seconds to 1.8 seconds, would it be acceptable to use the same trial order/ITI randomizations as I selected that were generated with a 2 second TR, or would I have to generate new ones? This is more a matter of whether I have to make new stimulus lists than running optseq2 again. Many thanks, ~Sue Mosher * * ** *** ***** ******** ************* Susan M. Mosher Program in Neuroscience Boston University 2 Cummington Street, Room 209F Boston, MA 02215-2425 smosher(a)bu.edu * * ** *** ***** ******** *************

2 1

dicom question
by Earl Zastrow 07 Nov '07

07 Nov '07

Hi, I'm a new user and I'm having problem with some of the dicom data sets I have. Namely, I got the following error message when I try to convert some of the data sets to .mgz: file not found or unknown file type for file /home/uwcem/cpratoom/BrainMRI/PA2/SE1/EXP0001.dcm I looked at the dicom info of EXP0001.dcm and discovered that the Manufacturer field is blank, compare to the one that did convert to .mgz without errors. Does that have anything to do with the error that I got? Thank you for your time. Regards, Earl Zastrow -- Earl Zastrow Graduate Research Assistant University of Wisconsin - Madison Computation Electromagnetics Laboratory (UWCEM) cpratoom(a)cae.wisc.edu (608) 265-0651

3 4

skullstripping woes
by Matt Colvin 07 Nov '07

07 Nov '07

Dear FreeSurfer wizards, I have a brain that keeps ending in error during the skullstripping portion of recon1, in version 4.0.1 of freesurfer for linux CentOS. Usually the error is directly preceded by a message about a "Segmentation fault". I have tried tweaking it manually in a couple different ways, including the -multistrip option but so far to no avail. The original message looked like this: ****************TEMPLATE DEFORMATION**************** second estimation of the COG coord: x=127,y=154, z=126, r=12861 iterations Segmentation fault Linux tammi 2.6.9-11.ELsmp #1 SMP Wed Jun 8 17:54:20 CDT 2005 i686 i686 i386 GNU/Linux recon-all exited with ERRORS at Mon Oct 29 21:36:33 CST 2007 I have attached a copy of the log if anyone is interested and wants to take a look. As always I would be very grateful is anybody out there has an idea about what is going wrong. --Matt Colvin

2 1

Manual edits to aseg.mgz/brainmask.mgz
by Chris Watson 07 Nov '07

07 Nov '07

Hello all, I was wondering if there is a quick way to make edits to multiple slices of the aseg.mgz (or the brainmask.mgz) volume. Specifically, I am working on my first subject, and it seems that on almost every slice, pieces of the dura/skull/arachnoid/what-have-you get segmented as cortex (less often, pieces of dura are included within the boundary of the pial surface). I've worked on some slices by manually removing/de-segmenting the pixels that need to go, but this is proving to be extraordinarily time-consuming (some of the slices I've worked on can take 10 minutes easily; perhaps I am obsessing too much, but am trying to get it as close to perfect as it should look). For over 100 slices, even 5 minutes/slice of editing will take several hours. I tried changing the mri_watershed parameter on another subject, but that didn't seem to fix anything. Any tips to save me hours will be greatly appreciated. -Chris

3 5

OT - animating ?h.pial or aparc+aseg
by Martin Kavec 07 Nov '07

07 Nov '07

Hi, sorry to bring an off topic, but I could not find a solution myself. I was asked to prepare a short fly-by animation of a ?h.pial for a TV who interviewed my boss last week. Would you have any recommendation on how to do that? The ?h.pial is a mesh, which is ideal, because the surface of the pial would be nice and smooth. I tried to export aparc+aseg to Analyze and read it into OsiriX, where I can easily produce the animation, but the surface was not smooth, as one would expect. So that is not the solution. Has anyone experience with this kind of task? Thanks, Martin

2 3

OpenGL
by Daniel Wakeman 06 Nov '07

06 Nov '07

Hello, I am configuring a Freesurfer installation for two different types of computers (headless servers and workstations). For both sets of computers we want any non-OpenGL commands that are run to first go through python scripts which load balance (these are working fine); however we have directly included the standard tkmedit* tksurfer* tkregister2* scuba* scuba2* and qdec* in the bin directory of freesurfer for the workstations. We still do not seem able to get these applications to work. Are there some dependencies, which I am missing for each of these besides tkmedit.bin etc.? Thank You, Dan

3 2

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer