Freesurfer

freesurfer@nmr.mgh.harvard.edu

17 participants
30086 discussions

CUDA PC build
by Chiu, Bryan (PHTH) 22 Apr '14

22 Apr '14

Hi, Trying to build a CUDA enabled PC for the lab to do processing for a variety of neuroimaging programs including Freesurfer and FSL. Want to know input from others about the 'recommended' PC and what the 'standard' is based on the current PC prices and technology. Sample build is from NCIX Canada, if curious. SKU Description Price Amount 84500 ASUS Z87-A ATX LGA1150 Z87 DDR3 3PCI-E16 2PCI-E1 2PCI CrossFireX/SLI SATA3 USB3.0 HDMI Motherboard [Reg. $159.99] $154.99 $154.99 93608 ADATA Xpg V1.0 16GB DDR3-1600 CL9-9-9-24 2X8GB Dual Channel Memory Kit Black [Reg.$189.99] $149.99 $149.99 82953 ASUS DRW-24F1ST 24X SATA DVD Writer Black $24.98 $24.98 65701 Seagate Barracuda 1TB 7200RPM 64MB SATA 6Gbps 3.5in Internal Hard Drive - OEM [Reg. $73.44] $64.99 $64.99 76367 ASUS GeForce GTX 660 OC DirectCU 1020MHZ 2GB 6.0GHZ GDDR5 2xDVI HDMI DisplayPort PCI-E Video Card $229.99 $459.98 81327 Intel Core i5 4670K Unlocked Quad Core 3.4GHZ Processor LGA1150 Haswell 6MB Cache Retail [Reg. $269.98] $249.99 $249.99 38650 Corsair Professional HX850W 850W ATX 12V 70A 24PIN ATX Modular Power Supply Active PFC 140mm Fan $169.98 $169.98 67326 Antec Three Hundred Two Mid Tower Gaming Case 302 ATX 3X5.25 6X3.5INT No PS Front USB3.0 & Audio [Reg. $77.98] $69.99 $69.99 In short -- runs with 16 Gb RAM, 3.4 GHz Quad core, with 2 GTX 660s in SLI. Final quote comes out to $1506 CAD. This is my first time attempting to build a PC of this type, so I still have tons of questions regarding minor details. I think spec wise the computer is good, unless others have recommendations from previous experience. Most likely will be running a flavour of Linux on this rig. Any suggestions for what the 'standard' for this build would be? I'm thinking CentOS based on older webmail threads, but I would prefer Ubuntu if possible. I would also like it to be user friendly so the technologically challenged researchers can get the software up and running. Is there anything else I should think about before I order? I want to make sure I have all of my bases covered before ordering and getting this rig going. Regards, - Bryan Chiu

1 0

recon all problem
by Danielle Miller 22 Apr '14

22 Apr '14

Hi Freesurfer Experts, I seem to be having some difficulty with reconing a healthy young adult. I have tried running this on multiple different computers and multiple different versions (5.3 and 5.1) and all return the same weird recon (see attached image). I have used the same sequence and commands on another subject and the recon returned fine. This participant did not have excess motion and the orig.mgz file looks fine. Please let me know if there is any more information I could provide that would be of use. Commands used: 5.3: recon-all -all -FLAIR (FLAIR image) -subjid (SUBJID) -FLAIRpial -hippo-subfields recon-all -all -subjid (SUBJID) 5.1: recon-all -all -subjid (SUBJID) -hippo-subfields recon-all -all -subjid (SUBJID) Thanks, Danielle Miller -- Ph.D. Program in Behavioral Neuroscience Boston University School of Medicine L-815 72 E. Concord St Boston, MA 02118 VA Boston Healthcare System Jamaica Plain Memory Disorders Research Center 150 South Huntington Ave D11-103 Boston, MA 02130 OFFICE:(857) 364-2130

2 1

Re: [Freesurfer] viewing certain regions
by Douglas N Greve 22 Apr '14

22 Apr '14

HI Courtney, please post to the list. Thanks! Get the codes from $FREESURFER_HOME/FreeSurferColorLUT.txt. Also, you can use multiple numbers to match all the ROIs you want at once (if that is what you want). Eg, --match 2 26 17 doug On 04/22/2014 03:16 PM, Courtney Haswell wrote: > Thanks for the reply. I was able to make the hippocampus and amygdala masks easily, but was having some trouble for the other ones. > > Looking in the "aparc.annot.ctab" file, I see that caudalanteriorcingulate is labeled 2 and rostralanteriorcingulate is 26. So then is this the correct way to get the ROI extracted: > mri_binarize --i $SUBJDIR/aparc+aseg.mgz --o $ROIDIR/3_rostralACC.mgz --match 26 > mri_binarize --i $SUBJDIR/aparc+aseg.mgz --o $ROIDIR/3_caudalACC.mgz --match 2 > > > > I also did the caudal ACC (that is in aparc.annot) the same way but that ROI covers all of the white matter in the right hemisphere so I think I did something wrong. > mri_binarize --i $SUBJDIR/aparc+aseg.mgz --o $ROIDIR/3_caudalACC.mgz --match 2 > > mri_mask $SUBJDIR/aparc+aseg.mgz $ROIDIR/3_caudalACC.mgz $ROIDIR/acc_mask5.mgz > > > > Thank you, Courtney > > > > > Use mri_binarize with the --match option to create a mask of the > regions you want. Then use mri_mask applied to aparc+aseg to create a > new volume with only the regions you want > doug > On 04/22/2014 09:06 AM, Courtney Haswell wrote: > > Hi! > > > > I want to view only certain segmented regions on one of our subject's > > brains. It seems like it should be simple but I am struggling to find > > the steps. I want to display only hippocampus, amygdala, medial > > orbitofrontal, lateral orbitofrontal, sulcus/gyri anterior cingulate, > > subcallosal gyrus, and rACC and cACC. > > > > How do I extract these labels so that I can only show just these > > regions on the brain? I extracted stats for these areas, but want to > > visually show the locations. > > > > Thank you, > > Courtney -- Douglas N. Greve, Ph.D. MGH-NMR Center greve(a)nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422 Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/

1 0

2 available spots in Boston FreeSurfer course
by Allison Stevens 22 Apr '14

22 Apr '14

We have had some last minute drop outs and there are now 2 spots available in the upcoming Boston FreeSurfer course (May 5-7th). If you would like to sign up for these spots, please register here: https://surfer.nmr.mgh.harvard.edu/cgi-bin/fsurfer/register.8263.cgi Note that credit card cannot be used to pay for the course. Also, this course will not have the diffusion analysis component (it will focus on fMRI analysis). Allison

2 3

viewing certain regions
by Courtney Haswell 22 Apr '14

22 Apr '14

Hi! I want to view only certain segmented regions on one of our subject's brains. It seems like it should be simple but I am struggling to find the steps. I want to display only hippocampus, amygdala, medial orbitofrontal, lateral orbitofrontal, sulcus/gyri anterior cingulate, subcallosal gyrus, and rACC and cACC. How do I extract these labels so that I can only show just these regions on the brain? I extracted stats for these areas, but want to visually show the locations. Thank you, Courtney

2 1

Re: [Freesurfer] RAS to MNI coordinate conversion function?
by Douglas N Greve 22 Apr '14

22 Apr '14

Hi John, please remember to post to the list. thanks! Yes, that should work. Use read read_surf.m to get the proper coordinates. doug On 04/20/2014 08:08 PM, John Pyles wrote: > Hi Doug, > > Thanks so much for the reply. > > I had looked through the wiki on coordinates but because I'm not a big > Freesurfer user was unsure which was the correct transformation. > > It seems like for what I need to do the example below is appropriate? > So in theory I could write a Matlab function using the calculation > below correct? > > Thanks again. > > Best, > John > > > ------------ > 2. I have an RAS point on the surface (tkrR tkrA tkrS) ("Vertex RAS" > from tksurfer) and want to compute the MNI305 RAS that corresponds to > this point: > > MNI305RAS = TalXFM*Norig*inv(Torig)*[tkrR tkrA tkrS 1]' > > TalXFM: subject/orig/transforms/talairach.xfm Norig: mri_info > --vox2ras orig.mgz Torig: mri_info --vox2ras-tkr orig.mgz > ------------ > > > ______________________________________ > > John Pyles, Ph.D. > Research Scientist > Center for the Neural Basis of Cognition > Department of Psychology > Carnegie Mellon University > email: jpyles(a)cmu.edu > phone: 206.552.0107 > ______________________________________ > > > On Fri, 18 Apr 2014, Douglas Greve wrote: > >> >> There is not a simple function, but you can look at our wiki page on >> coordinate systems. There are some examples there on >> tranforming from one coordinate to another >> doug >> >> >> On 4/18/14 2:40 PM, John Pyles wrote: >> Hi, >> >> I am not a heavy Freesurfer user and I haven't had good luck figuring >> this out through the FreeSurfer documentation and I >> thought an expert might have a fast answer. If anyone has an easy >> solution it would be very much appreciated. >> >> While I use FreeSurfer for automatic segmentation, I don't normally >> use if for analysis. However for an ECoG project I'm >> using parts of FreeSurfer to aid in electrode coregistration. >> >> Basically the issue is that I have electrode coordinates in what I >> believe are FreeSurfer RAS space and I need to convert >> them to MNI space. I have many of these electrode coordinates for >> each subject so I need a FreeSurfer or Matlab function to >> automate this process. Is there a simple way to do this? >> >> The *very* kludgy way I figured out how to do this "manually" is as >> follows (I'm sure this is going to seem very dumb to a >> FreeSurfer user, but I needed a quick solution that worked for a >> figure). >> 1) Enter the coordinate output from my electrode coregistration >> routines manually into the RAS field in Freeview. >> 2) Take the "brain volume" coordinates (all positive) in Freeview and >> enter them into the "Volume Coordinate" field in >> TkMedit. >> 3) Copy the TAL coordinates output in MkEdit when the brain volume >> coordinate is entered. >> 4) Convert the TAL coordinates to MNI in Matlab with tal2mni.m >> >> It seems like there should be a simple function to do this coordinate >> conversion using the transformation info from >> recon-all that I'm assuming both Freeview and TkMedit use, but I >> can't seem to find it? >> >> Any help would be appreciated. Thanks again! >> >> Best, >> John >> >> ______________________________________ >> >> John Pyles, Ph.D. >> Research Scientist >> Center for the Neural Basis of Cognition >> Department of Psychology >> Carnegie Mellon University >> ______________________________________ >> >> >> >> >> >> >> >> _______________________________________________ >> Freesurfer mailing list >> Freesurfer(a)nmr.mgh.harvard.edu >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> >> >> >> -- Douglas N. Greve, Ph.D. MGH-NMR Center greve(a)nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422 Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/

1 0

ROI analysis in Matlab
by Reza Rajimehr 22 Apr '14

22 Apr '14

Hi, For my dataset, it appears to be easier to do an ROI analysis in Matlab (instead of using funcroi commands). I just want to make sure the procedure described below is right: 1) I have a label/ROI on the right hemisphere surface of a subject (e.g. MT from the motion localizer). I convert it to a mask file (MT.mask.nii.gz). I load it to Matlab using MRIread. In an fMRI experiment, I have 3 conditions. For the right hemisphere vertices of the subject, I compute cespct.nii.gz for the contrasts of each condition vs. baseline (1_vs_0 , 2_vs_0 , 3_vs_0). Then I load the three cespct files to Matlab using MRIread. 3) For the vertices that are 1 in the MT.mask.nii.gz, I calculate the averaged cespct across those vertices, in each contrast. So I will have % signal change values for the three conditions. 4) I repeat this for all subjects and do statistics across subjects. Any comments? Reza

2 1

rmanova contrast file error
by Harry Hallock [hhal5293＠uni.sydney.edu.au] 22 Apr '14

22 Apr '14

Dear Experts, I am running a rmanova using FSGD file and mri_glmfit. I have done the mris_preproc step, and am currently trying to do the mri_glmfit step. I keep getting an error that appears to be related to the contrast files and matricies (highlighted in yellow below) I have also attched the FSGD file and a contrast file. Everything is in ASCII. The output from the command terminal is below: mri_glmfit --glmdir lh.rmanova --y lh.thickness.mgh --fsgd rmanova.fsgd doss --C tp1_vs_tp2.mtx --C tp1_vs_tp3.mtx --C tp2_vs_tp3.mtx --C tp_effect.mtx --C mean.mtx --surface fsaverage lh gdfReadHeader: reading rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 tp1_vs_tp2 0 0.840168 1 tp2_vs_tp3 0 0.766965 Class Means of each Continuous Variable 1 ECT_001 0.0000 0.0000 2 ECT_002 0.0000 0.0000 3 ECT_003 0.0000 0.0000 4 ECT_004 0.0000 0.0000 5 ECT_005 0.0000 0.0000 6 ECT_006 0.0000 0.0000 7 ECT_007 0.0000 0.0000 8 ECT_008 0.0000 0.0000 9 ECT_009 0.0000 0.0000 10 ECT_010 0.0000 0.0000 11 ECT_011 0.0000 0.0000 12 ECT_012 0.0000 0.0000 13 ECT_013 0.0000 0.0000 14 ECT_014 0.0000 0.0000 15 ECT_015 0.0000 0.0000 16 ECT_016 0.0000 0.0000 17 ECT_017 0.0000 0.0000 18 ECT_018 0.0000 0.0000 19 ECT_019 0.0000 0.0000 20 ECT_020 0.0000 0.0000 21 ECT_021 0.0000 0.0000 22 ECT_022 0.0000 0.0000 23 ECT_023 0.0000 0.0000 24 ECT_024 0.0000 0.0000 INFO: gd2mtx_method is doss Reading source surface /media/My_Book/MyDATA/ECT_Structure_For_Harry/FreeSurfer_Analysis/priorEDIT/T1/fsaverage/surf/lh.white Number of vertices 163842 Number of faces 327680 Total area 65416.648438 AvgVtxArea 0.399267 AvgVtxDist 0.721953 StdVtxDist 0.195470 $Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /media/My_Book/MyDATA/ECT_Structure_For_Harry/FreeSurfer_Analysis/priorEDIT/T1 cmdline mri_glmfit --glmdir lh.rmanova --y lh.thickness.mgh --fsgd rmanova.fsgd doss --C tp1_vs_tp2.mtx --C tp1_vs_tp3.mtx --C tp2_vs_tp3.mtx --C tp_effect.mtx --C mean.mtx --surface fsaverage lh sysname Linux hostname fs-OptiPlex-755 machine i686 user fs FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /media/My_Book/MyDATA/ECT_Structure_For_Harry/FreeSurfer_Analysis/priorEDIT/T1/lh.thickness.mgh logyflag 0 usedti 0 FSGD rmanova.fsgd labelmask /media/My_Book/MyDATA/ECT_Structure_For_Harry/FreeSurfer_Analysis/priorEDIT/T1/fsaverage/label/lh.cortex.label maskinv 0 glmdir lh.rmanova IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory lh.rmanova Loading y from /media/My_Book/MyDATA/ECT_Structure_For_Harry/FreeSurfer_Analysis/priorEDIT/T1/lh.thickness.mgh INFO: gd2mtx_method is doss Saving design matrix to lh.rmanova/Xg.dat Normalized matrix condition is 2.67942 Matrix condition is 32.198 Found 149955 points in label. Pruning voxels by thr: 0.000000 Found 148274 voxels in mask Saving mask to lh.rmanova/mask.mgh Reshaping mriglm->mask... search space = 73670.427324 MatrixReadTxT: could not scan value [2][1] ERROR: loading C tp1_vs_tp2.mtx Any help would be greatly appreciated. Thanks, Harry Harry Hallock | MPhil Candidate Regenerative Neuroscience Group | Brain & Mind Research Institute THE UNIVERSITY OF SYDNEY Room 408, Level 4, M02K | 94 Mallett St Camperdown | NSW | 2050 | Australia T +61 2 9114 4138 F +61 9351 0930 E hhal5293(a)uni.sydney.edu.au | W www.rng.org.au

2 1

resetting wm-mean [CA Normalize]
by Mark Plantz 21 Apr '14

21 Apr '14

Hi freesurfers, I am running into an error where the white matter and gray matter means of my atlas are automatically setting to 0 during the #@# CA Normalize pathway. I copied a code segment below to help explain better. Has anyone else run into this problem? Is there a way to manually change these averages? *#@# CA Normalize Mon Apr 14 17:03:07 CDT 2014* */Applications/freesurfer/subjects/CMH-11780/mri* *\n mri_ca_normalize -c ctrl_pts.mgz -mask brainmask.mgz nu.mgz /Applications/freesurfer/average/atlas2.gca transforms/talairach.lta norm.mgz \n* *writing control point volume to ctrl_pts.mgz* *using MR volume brainmask.mgz to mask input volume...* *reading 1 input volume* *reading atlas from '/Applications/freesurfer/average/atlas2.gca'...* *reading transform from 'transforms/talairach.lta'...* *reading input volume from nu.mgz...* *resetting wm mean[0]: 0 --> 0* *resetting gm mean[0]: 0 --> 0* *input volume #1 is the most T1-like* *using real data threshold=27.0* *skull bounding box = (64, 79, 51) --> (194, 166, 204)* *using (107, 108, 128) as brain centroid...* *mean wm in atlas = 0, using box (91,97,109) --> (122, 118,146) to find MRI wm* *before smoothing, mri peak at 110* *after smoothing, mri peak at 111, scaling input intensities by 0.000* *scaling channel 0 by 0* *using 136509 sample points...* *INFO: compute sample coordinates transform* * 1.000 0.000 0.000 0.000;* * 0.000 1.000 0.000 0.000;* * 0.000 0.000 1.000 0.000;* * 0.000 0.000 0.000 1.000;* *INFO: transform used* Thank you for the help. - MP

2 1

An overwrite issue with selxavg3-sess
by Reza Rajimehr 21 Apr '14

21 Apr '14

I have done a functional analysis for some contrasts using selxavg3-sess. In the contrast folders, I have added some files. Now I want to do the same analysis for some new contrasts. My guess is that selxavg3-sess computes all the contrasts again, and I am worried that the new folders overwrite the previous ones and I lose the files that I have manually added. If this is the case, is there a way to run selxavg3-sess for some specified contrasts? Thanks, Reza

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