Freesurfer

freesurfer@nmr.mgh.harvard.edu

16 participants
30084 discussions

Version differences
by Martin Ystad 03 Mar '09

03 Mar '09

Hello, I'm re-running some of my volumes previously processed through an older version of FreeSurfer (in 2004/2005 - version 3.01 I think). At the time they were processed through autorecon2 and 3, and problems that appeared were addressed by manual editing. However, when I re-process the volumes on version 4.2.0 in freesurfer, new segmentation errors appear, although I have kept my old manual edits from last time (see pictures). At the same time, we experience more than usual segmentation errors on new volumes that we process. Actually, many of these errors appeared after we updated from version 4.1 to 4.2. Is there any settings that I can tweak to make the segmentation algorithms behave more like they did before, or do I have to roll back to a previous version? Thanks Martin Ystad

2 1

Problem while fixing the Surface Topology
by Pradeep Reddy Ramana 03 Mar '09

03 Mar '09

Hello FS Team, I had a question recently regarding an error encountered in the freesurfer recon-all. I believe error is the result of: mris_fix_topology -mgz -sphere qsphere.nofix -ga FirstOrderPoly lh CORRECTING DEFECT 3 (vertices=10731, convex hull=2557) normal vector of length zero at vertex 90051 with 0 faces vertex 90051 has 0 face No such file or directory recon-all exited with ERRORS I will appreciate very much if you can suggest a possible fix or a rerun with some set of flags etc. The partial log and a previous post by somebody else is here: http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg09492.html Thanks, Pradeep -- Pradeep Kumar Reddy. Raamana PhD Student - Biomedical Engg. Medical Image Analysis Lab, Simon Fraser University, Burnaby BC - V5A1S6 - Canada. Work : +1 778 782 5509 "What is written without effort is, in general, read without pleasure." - Samuel Johnson

2 2

Configure Overlay Display for >2 Colors
by Jeff Sadino 02 Mar '09

02 Mar '09

Hello Surfers, I have an overlay file with values ranging from -4.63 (n=138477) to 1.34 (n=25365). I would like to display this data with more than the two colors of the Green/Red, Heat, Blue/Red, or complex Color Scales in the Configure Overlay dialog. So far, no luck, even with toying with the min, max, mid, and piecewise options every way I can think of. Is there some way to get a fuller color spectrum on my overlay? Thank you, Jeff Sadino _________________________________________________________________ Hotmail® is up to 70% faster. Now good news travels really fast. http://windowslive.com/online/hotmail?ocid=TXT_TAGLM_WL_HM_70faster_032009

2 2

recon-all with MT images?
by Lucia Garrido 02 Mar '09

02 Mar '09

Dear Freesurfers, I was wondering whether it would be possible to run recon-all with MT images? My final aim is to compare cortical thickness between two groups. I've done this with T1 weighted images, but was wondering whether i could also do with MT images. Segmentation in SPM worked really well on these images. Thank you very much. Best wishes, Lucia

2 2

clusterwise correction for multiple comparisons
by Devdutta W 02 Mar '09

02 Mar '09

Hi, In the mri_glmfit-sim command, when it does a "clusterwise correction" is it correcting across the whole brain? In other words, does the term "clusterwise" mean that it is only correcting within that cluster or is it correcting across the whole brain, cluster-by-cluster (instead of, say, voxel-by-voxel)? While doing this analysis, (after displaying the .annot file in tksurfer as stated in the documentation), I noticed that our display did not match what was shown in the docs. In the documentation, the picture we are supposed to see only displays the clusters (all clusters, regardless of the significance). Ours displayed cortical parcellations in addition to the clusters. This may just be a display issue and the analysis might still be correct, but I just want to be sure. Has anyone else encountered this? Thanks for any light anyone can shed on these questions. Devdutta

2 2

aparc-stats wth curv-file as input?
by Florian Gerstl 02 Mar '09

02 Mar '09

Dear FS-Team, I have mapped some PET-data on the corresponding subject's cortical surface. Is it possible to get some statistics from the resulting curv-file using the FS-parcellation as ROIs? Thanks in advance, Florian

4 5

¿write_annotation?
by Rafa x 02 Mar '09

02 Mar '09

Dear FS team, I used read_annotation(lh.aparca.annot) for obtaining [vertexs, label, colortable]. After that, I have modified these variables and would like to write a new annotation file. Is there the function "write_annotation" (or something similar)? Thanks in advance. Best, Rafa

1 0

Normal vector of length zero
by Pradeep Reddy Ramana 01 Mar '09

01 Mar '09

Hello Freesurfer Experts, I ran into the following error when running Freesurfer for one of our subjects, during: mris_fix_topology -mgz -sphere qsphere.nofix -ga FirstOrderPoly lh I could only find a partial answer to a previous post on a similar problem - pasted at the end. Following that, I looked at lh.orig.nofix surface - screenshots attached. I am not sure whether its alright. No rh surfaces were created. Please let me know how to correct this problem and proceed further. Quick response will be greatly appreciated, Pradeep ======= Tail end of scripts/recon-all.log ========== #-------------------------------------------- #@# Fix Topology lh Fri Feb 27 19:29:56 PST 2009 cp ../surf/lh.orig.nofix ../surf/lh.orig cp ../surf/lh.inflated.nofix ../surf/lh.inflated /ensc/CORTEX/PROJECTS/UNSW/WEIWEN_PARF/PROCESSED_DATA/scanner_differences/FirstOrderPoly/scripts mris_fix_topology -mgz -sphere qsphere.nofix -ga FirstOrderPoly lh reading spherical homeomorphism from 'qsphere.nofix' using genetic algorithm with optimized parameters ************************************************************* Topology Correction Parameters retessellation mode: genetic search number of patches/generation : 10 number of generations : 10 surface mri loglikelihood coefficient : 1.0 volume mri loglikelihood coefficient : 10.0 normal dot loglikelihood coefficient : 1.0 quadratic curvature loglikelihood coefficient : 1.0 volume resolution : 2 eliminate vertices during search : 1 initial patch selection : 1 select all defect vertices : 0 ordering dependant retessellation: 0 use precomputed edge table : 0 smooth retessellated patch : 2 match retessellated patch : 1 verbose mode : 0 ************************************************************* INFO: assuming .mgz format $Id: mris_fix_topology.c,v 1.43 2007/01/05 16:57:16 nicks Exp $ $Id: mrisurf.c,v 1.557.2.10 2008/03/10 13:59:45 nicks Exp $ before topology correction, eno=-264 (nv=105340, nf=211208, ne=316812, g=133) using quasi-homeomorphic spherical map to tessellate cortical surface... Correction of the Topology Finding true center and radius of Spherical Surface...done Surface centered at (0,0,0) with radius 100.0 in 8 iterations marking ambiguous vertices... 45195 ambiguous faces found in tessellation segmenting defects... 81 defects found, arbitrating ambiguous regions... analyzing neighboring defects... -merging segment 6 into 0 -merging segment 7 into 0 -merging segment 0 into 18 -merging segment 39 into 28 -merging segment 37 into 30 -merging segment 54 into 46 -merging segment 53 into 59 -merging segment 72 into 64 -merging segment 71 into 70 72 defects to be corrected 0 vertices coincident reading input surface /ensc/CORTEX/PROJECTS/UNSW/WEIWEN_PARF/PROCESSED_DATA/scanner_differences/FirstOrderPoly/surf/lh.qsphere.nofix... reading brain volume from brain... reading wm segmentation from wm... Computing Initial Surface Statistics -face loglikelihood: -9.2008 (-4.6004) -vertex loglikelihood: -6.9827 (-3.4913) -normal dot loglikelihood: -3.5884 (-3.5884) -quad curv loglikelihood: -6.0560 (-3.0280) Total Loglikelihood : -25.8279 CORRECTING DEFECT 0 (vertices=73, convex hull=66) After retessellation of defect 0, euler #=-61 (80515,236855,156279) : difference with theory (-69) = -8 CORRECTING DEFECT 1 (vertices=212, convex hull=29) After retessellation of defect 1, euler #=-60 (80531,236912,156321) : difference with theory (-68) = -8 CORRECTING DEFECT 2 (vertices=14, convex hull=24) After retessellation of defect 2, euler #=-59 (80534,236928,156335) : difference with theory (-67) = -8 CORRECTING DEFECT 3 (vertices=10731, convex hull=2557) normal vector of length zero at vertex 90051 with 0 faces vertex 90051 has 0 face No such file or directory Linux ilpc9.cs.sfu.ca 2.6.18-92.1.22.el5 #1 SMP Tue Dec 16 11:57:43 EST 2008 x86_64 x86_64 x86_64 GNU/Linux recon-all exited with ERRORS at Fri Feb 27 21:07:28 PST 2009 ======================================== ------------------ Previous post with a similar problem ------------------- [Freesurfer] Problem with latest dev version (crashes correcting errors) Tim Jarvis jarv0075 at umn.edu Mon Feb 6 12:41:21 EST 2006 * Previous message: [Freesurfer] Problem with latest dev version (crashes correcting errors) * Next message: [Freesurfer] Problem with latest dev version (crashes correctingerrors) * Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] For the lh, there is no cerebellum and no skull, looks to be correct. I don't see any rh* files though? brewster 14%> ls /data/brewster2/tjarvis/thesis_data/software/freesurfer/subjects/icbm/surf/ lh.defect_borders lh.defect_chull lh.defect_labels lh.inflated lh.inflated.nofix lh.orig lh.orig.nofix lh.qsphere.nofix lh.smoothwm.nofix On 6 Feb 2006, Bruce Fischl wrote: > yes, mri_ca_register takes a long time. The good news is it works really > well :>. As for the mris_topology_fixing problem, what does the > ?h.orig.nofix surface look like? Does it include cerebellum or skull? Does > it look approximately correct? > > On Mon, 6 Feb 2006, Tim Jarvis wrote: > > > Gave the new dev version a spin with autorecon-all, couple of things... > > > > For the mri_ca_register step, it took a total of 791 minutes! Is this > > normal or could this be a hint as to things going wrong early in the > > pipeline? > > > > Also, the program eventually crashed during the error correction, below are > > the lines that seem like they may have had some effect: > > > > Anyone else experience this problem? > > > > > > > > Thanks > > -- Pradeep Kumar Reddy. Raamana PhD Student - Biomedical Engg. Medical Image Analysis Lab, Simon Fraser University, Burnaby BC - V5A1S6 - Canada. Work : +1 778 782 5509 Freedom from the desire for an answer is essential to the understanding of a problem - Jiddu Krishnamurti

1 1

effect sizes and SE
by Lars M. Rimol 01 Mar '09

01 Mar '09

Hi, Is there a way to get the standard error of the adjusted means for each group in a glm with a categorical predictor? -- yours, LMR

1 0

scale axis range
by Juranek, Jenifer 27 Feb '09

27 Feb '09

Hi, Running fsv4.2.0 on RHEL4. Within qdec, the "graph average ROI" provides valuable information in scatterplot format. Is it possible for users to define the axis range (especially the Y-axis)? For some unknown reason, we have encountered several instances where data points are not displayed since the scale range of the automatic plot does not capture the entire range of individual subject data. Restarting qdec, re-running the design, and reloading the label does not alter the outcome. Any thoughts or quick fixes?

1 0

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