Freesurfer

freesurfer@nmr.mgh.harvard.edu

16 participants
30085 discussions

gm segmentation problem
by johanna feiler 18 Jan '10

18 Jan '10

Dear freesurfer experts, After running the autorecon2 process the pial surface fails to catch up pieces of gm in the insula region. I gained little improvement with setting control points, but the problem still occurs, lasting about five slices. As an illustration, I attached a screenshot. How can I solve this problem? Thanks for your help! Johanna Feiler ______________________________________________________ GRATIS für alle WEB.DE-Nutzer: Die maxdome Movie-FLAT! Jetzt freischalten unter http://movieflat.web.de

1 0

recon-all now taking too long
by Rodrigo Zepeda 18 Jan '10

18 Jan '10

Hi, I have noticed that recently recon-all is taking a lot longer than usual to finish. It used to take around 30 to 48 hours, but the last couple that I have done are taking a whole week. It seems to get stuck in a repetitive step and it numbers each iteration like this: not handling expanded ventricles... using previously computed transform transforms/talairach.lta renormalizing sequences with structure alignment, equivalent to: -renormalize -regularize_mean 0.500 -regularize 0.500 using MR volume brainmask.mgz to mask input volume... reading 1 input volumes... logging results to talairach.log reading input volume 'norm.mgz'... reading GCA '/usr/local/freesurfer/stable4/average/RB_all_2008-03-26.gca'... freeing gibbs priors...done. average std[0] = 5.0 label assignment complete, 0 changed (0.00%) det(m_affine) = 1.30 (predicted orig area = 6.2) label assignment complete, 0 changed (0.00%) **************** pass 1 of 1 ************************ setting smoothness coefficient to 0.04 blurring input image with Gaussian with sigma=2.000... 0000: dt=0.000, rms=0.759, neg=0, invalid=766 0001: dt=252.406048, rms=0.710 (6.426%), neg=0, invalid=766 0002: dt=157.425287, rms=0.695 (2.149%), neg=0, invalid=766 0003: dt=199.529412, rms=0.687 (1.109%), neg=0, invalid=766 0004: dt=115.649123, rms=0.684 (0.433%), neg=0, invalid=766 0005: dt=517.888000, rms=0.677 (1.076%), neg=0, invalid=766 0006: dt=74.457516, rms=0.675 (0.226%), neg=0, invalid=766 0007: dt=74.457516, rms=0.675 (0.073%), neg=0, invalid=766 0008: dt=74.457516, rms=0.674 (0.143%), neg=0, invalid=766 0009: dt=74.457516, rms=0.672 (0.227%), neg=0, invalid=766 0010: dt=74.457516, rms=0.670 (0.292%), neg=0, invalid=766 0011: dt=74.457516, rms=0.668 (0.289%), neg=0, invalid=766 0012: dt=74.457516, rms=0.667 (0.258%), neg=0, invalid=766 0013: dt=74.457516, rms=0.665 (0.229%), neg=0, invalid=766 0014: dt=74.457516, rms=0.664 (0.195%), neg=0, invalid=766 0015: dt=74.457516, rms=0.663 (0.191%), neg=0, invalid=766 0016: dt=74.457516, rms=0.661 (0.213%), neg=0, invalid=766 0017: dt=74.457516, rms=0.660 (0.192%), neg=0, invalid=766 0018: dt=74.457516, rms=0.659 (0.154%), neg=0, invalid=766 0019: dt=74.457516, rms=0.658 (0.132%), neg=0, invalid=766 0020: dt=74.457516, rms=0.657 (0.119%), neg=0, invalid=766 0021: dt=74.457516, rms=0.656 (0.140%), neg=0, invalid=766 0022: dt=74.457516, rms=0.655 (0.162%), neg=0, invalid=766 0023: dt=74.457516, rms=0.654 (0.162%), neg=0, invalid=766 0024: dt=74.457516, rms=0.653 (0.138%), neg=0, invalid=766 0025: dt=74.457516, rms=0.652 (0.114%), neg=0, invalid=766 0026: dt=74.457516, rms=0.652 (0.117%), neg=0, invalid=766 0027: dt=74.457516, rms=0.651 (0.130%), neg=0, invalid=766 0028: dt=74.457516, rms=0.650 (0.138%), neg=0, invalid=766 0029: dt=74.457516, rms=0.649 (0.124%), neg=0, invalid=766 0030: dt=74.457516, rms=0.649 (0.104%), neg=0, invalid=766 0031: dt=32.368000, rms=0.649 (-0.001%), neg=0, invalid=766 blurring input image with Gaussian with sigma=0.500... 0000: dt=0.000, rms=0.649, neg=0, invalid=766 0032: dt=129.472000, rms=0.647 (0.306%), neg=0, invalid=766 0033: dt=517.888000, rms=0.645 (0.194%), neg=0, invalid=766 0034: dt=517.888000, rms=0.645 (0.016%), neg=0, invalid=766 0035: dt=517.888000, rms=0.645 (-0.929%), neg=0, invalid=766 setting smoothness coefficient to 0.15 blurring input image with Gaussian with sigma=2.000... 0000: dt=0.000, rms=0.651, neg=0, invalid=766 0036: dt=51.963303, rms=0.646 (0.707%), neg=0, invalid=766 0037: dt=216.338028, rms=0.634 (1.907%), neg=0, invalid=766 0038: dt=39.422680, rms=0.630 (0.681%), neg=0, invalid=766 0039: dt=145.152000, rms=0.627 (0.498%), neg=0, invalid=766 0040: dt=73.437288, rms=0.622 (0.677%), neg=0, invalid=766 0041: dt=75.051233, rms=0.619 (0.604%), neg=0, invalid=766 0042: dt=23.040000, rms=0.618 (0.092%), neg=0, invalid=766 0043: dt=23.040000, rms=0.618 (0.058%), neg=0, invalid=766 0044: dt=23.040000, rms=0.617 (0.112%), neg=0, invalid=766 It continues to do that for days. I have run it in 2 different computers and it makes no difference. Any ideas why this is now taking so long and how to make a bit faster? Thanks! Rodrigo

1 0

error in autorecon1
by Cindy Eckart 18 Jan '10

18 Jan '10

Dear Freesurfers, when I try to run the recon-all flow, it exits at the skull stripping stage in some subjects, giving out error messages like this: register_mri: find_optimal_transform find_optimal_transform: nsamples 3481, passno 0, spacing 8 GCAhistoScaleImageIntensities: could not find wm peak resetting wm mean[0]: 117 --> 126 resetting gm mean[0]: 74 --> 74 input volume #1 is the most T1-like using real data threshold=10.0 skull bounding box = (52, 9, 21) --> (208, 255, 220) using (104, 91, 121) as brain centroid... mean wm in atlas = 126, using box (85,61,96) --> (123, 121,145) to find MRI wm before smoothing, mri peak at 0 after smoothing, mri peak at 0, scaling input intensities by inf When I checked all prior steps, I realized that the nu.mgz was particularly dark and tried to improve it by manually setting control points in the white matter... however, without success, even under inclusion of some control points the skull strip could not be finished.... Then I found a similar problem in the mailing list archive (https://mail.nmr.mgh.harvard.edu/pipermail//freesurfer/2008-April/007603.ht…) tried the solution proposed there and altered my 001.mgz with a scaling factor of 3 (cp 001.mgz 001_orig.mgz / mri_convert --scale 3 001_orig.mgz 001.mgz / rerun recon-all) again without any success..... Finally, I have no idea how to find another solution! Has anyone an idea?? Every suggestions is highly appreciated! Thanks and greetings Cindy ___________________________________________________________ Preisknaller: WEB.DE DSL Flatrate für nur 16,99 Euro/mtl.! http://produkte.web.de/go/02/

2 1

Vertex values
by Oya, Hiroyuki 16 Jan '10

16 Jan '10

Dear Experts, Is it possible to import surface vertex index and corresponding values into Matlab? Thank you very much . Hiroyuki

2 1

SPM-VBM and freesurfer
by Lena Palaniyappan 16 Jan '10

16 Jan '10

Hi all Not sure if this question is relevant to the list - any help will be appreciated. I have a t-statistics map from comparing 2 groups in VBM approach from SPM8 I want to apply this thresholded t map as a mask ROI to see thickness differences in the same groups using freesurfer. Is this possible? Will I be better off doing FSL_VBM instead - if that is easy to run with freesurfer? cheers Lena ________________________________________________________ Lena Palaniyappan PhD student| Division of Psychiatry ( University of Nottingham) South Block 'A' floor| Queens Medical Centre | Nottingham | NG7 2UH This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system: you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

2 1

skull stripping
by laura_dg80 16 Jan '10

16 Jan '10

Hi! I have some difficulties in the Skull stripping. I'm not experienced and I'm not able to evaluate if my job is good. I visualize the result with Tkmedit and I see that some skull often remain in the eye soked region . Could you send me some example of "GOOD" and "Bad" skull stripping of this region? Thank you Laura

2 1

retinotopic mapping stimuli
by Vadim Axel 16 Jan '10

16 Jan '10

Hi, I prepare to run retinotopic mapping experiment. I need only polar angles, since I am interested in localization of V1,V2 and V3 only. So, on inflated brain I plan to locate regions' reversals (vertical and horizontal meridians). For the moment, I plan to use FreSurfer for only inflating the brain (fMRI data I analyze in SPM). I have created two different stimuli and I would like to know: 1) Since I need only vertical and horizontal meridians is it OK to have only 90 degrees sector? 2) Since I need only vertical and horizontal meridians, how effective it would be instead of Fourier phase analysis to make a GLM with a boxcar function? 3) Should my sector start straight at meridian line (as I did), or the sector should encompass it from both sides? 4) Is it optimal to make each sector block length about 10-12 secs The stimuli: http://www.youtube.com/watch?v=iW8sk44eyck http://www.youtube.com/watch?v=DcGZW1pM-28 Thanks a lot for your help, Vadim

1 0

Incomplete skull striping but correct segmentation
by Estephan Moana 15 Jan '10

15 Jan '10

I had some subjects processed through the standard recon-all pipeline that had some remaining skull just below the frontal lobe (superior part of the orbit). The segmentation as checked in tkmedit appeared to be ok, i.e. it was not considering the skull parts as white matter. The reconstructed cortical surfaces also appeared to be good in tksurfer. Playing with the watershed value did not provide much improvement, and I assume that to clean up the images adequately I would have to do so by hand. My question is: should I be concerned about improving skull stripping for these subjects, given the good segmentation and cortical reconstruction obtained? Thank you. Estephan Moana, DDS Graduate Student Oral Biology PhD Program - Neurobiology track School of Dentistry, UNC- Chapel Hill 2140 Old Dental Building, CB# 7455 Rm 2140 Chapel Hill, NC 27599 Phone: (919) 966-5680 Fax: (919) 966-5339 moana(a)email.unc.edu

1 0

Tiff images created by tksurfer
by Oya, Hiroyuki 14 Jan '10

14 Jan '10

Hello tksurfer can save multiple surface renderings of the brain as 600 x600 tiff formatted images. If anybody can tell me the relationship between center of the image and the brain, that would be appreciated. Thank you very much. Hiroyuki

1 0

Recon-ALL error
by mmoayedi＠uhnres.utoronto.ca 14 Jan '10

14 Jan '10

Hi Everyone, Here is a newish error I've encountered- does anyone know how to fix it, or what it means? status file for recon-all: Thu Jan 14 19:48:44 EST 2010 #@# MotionCor Thu Jan 14 19:48:50 EST 2010 #@# Nu Intensity Correction Thu Jan 14 19:49:05 EST 2010 #@# Talairach Thu Jan 14 19:51:03 EST 2010 #@# Talairach Failure Detection Thu Jan 14 19:52:05 EST 2010 Linux 2.6.23.1-21.fc7 #1 SMP Thu Nov 1 21:09:24 EDT 2007 i686 i686 i386 GNU/Linux recon-all exited with ERRORS at Thu Jan 14 19:52:05 EST 2010 --------------------- recon-all.error: PWD /storage/AllUsers/Nathalie/FreeSurferData/Subjects/A1/a1/mri CMD talairach_afd -T 0.005 -xfm transforms/talairach.xfm --------------------- from command line: talairach_avi done cp transforms/talairach.auto.xfm transforms/talairach.xfm #-------------------------------------------- #@# Talairach Failure Detection Thu Jan 14 19:52:05 EST 2010 /storage/AllUsers/Nathalie/FreeSurferData/Subjects/A1/a1/mri talairach_afd -T 0.005 -xfm transforms/talairach.xfm ERROR: talairach_afd: Talairach Transform: transforms/talairach.xfm ***FAILED*** (p=0.0001, pval=0.0000 < threshold=0.0050) Linux 2.6.23.1-21.fc7 #1 SMP Thu Nov 1 21:09:24 EDT 2007 i686 i686 i386 GNU/Linux recon-all exited with ERRORS at Thu Jan 14 19:52:05 EST 2010 -------------------------- As usual, any, and all help, is much appreciated!!! Massih

2 1

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