Freesurfer

freesurfer@nmr.mgh.harvard.edu

8 participants
30078 discussions

samseg ground truth's space
by Dilek Özkan 30 Aug '22

30 Aug '22

External Email - Use Caution Hello dear all, I want to use Freesurfer-Samseg for MS lesion segmentation(cross-sectional). I got the T1 and Flair for this. I registered *flair to t1 *for every patient. My manual lesion masks are on DIR imaging. I am using FSL for registration. When I try to register my manual mask to new mask created by freesurfer samseg, fsl couldn't recognize the lesions.mgz extension. Or should I use t1 space for registration? i tried that one. and then, my registration result was really bad. As a consequence, ho should we deal with mask registration? Best, Dilek

2 4

Vertex-wise analysis using ComBat
by Rene Seiger 30 Aug '22

30 Aug '22

External Email - Use Caution Dear FreeSurfer community! We collected structural data from different scanners and we want to conduct a surface-based (vertex-wise) analysis in FreeSurfer using predefined regions of interest. The model is quite simple, as we just want to correlate a specific (e.g. behavioral) measurement from all subjects (one group) with the thickness on the surface in specific ROIs. To account for the different scanner types, we are planning to use ComBat to account for this effect. However, we are not quite sure how and at which stage we could implement ComBat using the standard FreeSurfer FSGD file approach. Any help would be highly appreciated, Thanks!

2 1

Orientation issue
by Ruifeng Dong 30 Aug '22

30 Aug '22

External Email - Use Caution Dear all, I am new to Freesurfer, using version 6.0.0. I am trying to do brain segmentation, using data from two scanners, one called "clinical" and the other called "PRISMA". The clinical scanner uses LPI orientation while the PRISMA one uses RAI. I don't know whether we need to stick to a standard orientation in Freesurfer, so I tested both of them. The result for one volunteer is attached here. In addition to the original images from two scanners, also shown are the PRISMA image flipped to LPI orientation and the clinical image flipped to RAI orientation, using the 3dresample command. [image: image.png] As seen, the volumetry looks similar among all of the four. But if Freesurfer does not require a fixed orientation, then the same image should provide exactly the same volumetry, whether we flip it or not. Am I right? (*Question 1*) If I understand correctly, there is no statistical variation in the computed volumes, i.e. different runs should give the same volume values. If Freesurfer does require a fixed orientation in the input image, then how do we know whether the left-right-flipped image is used, as they are roughly symmetric? (*Question 2)* Thank you very much! Ruifeng

2 3

Re: [Freesurfer] Thalamic nuclei exited with ERRORS
by fsbuild 30 Aug '22

30 Aug '22

External Email - Use Caution MCRv84 looks like it should run on MacOS 10.13, so in a new terminal window, I would install it directly into the freesurfer distribution with, $ export FREESURFER_HOME=/Applications/freesurfer/7.2.0$ cd $FREESURFER_HOME$ rm -rf  MCRv84 $ cd bin $ sudo FREESURFER_HOME=$FREESURFER_HOME  ./fs_install_mcr  R2014b Then source the freesurfer setup script and try running things again to see if it works differently. The crash you listed looks like Matlab trying to run Java 1.7 which generates a segmentation fault.  There is a bug posting about this on The Mathworks web site, https://secure-web.cisco.com/1YLkkrkvYufFjx5Ic_nDdTX50vjrbBOo_Gz…;  I do not have a login there to read the reply/workaround, but I do see postings that say a workaround is to run Matlab with the -nosplash flag at startup.. If you still get a crash with Java 1.7, you could try adding java 1.8 to your machine, switch to running 1.8, and see if it works differently.  There should be a Java icon in the system preferences panel on your machine which might help you manage java on your machine.  You can download Java 1.8 from the oracle web site under the MacOS entry on https://secure-web.cisco.com/1xj6rOSqU4YlWXHG3wogofsCMLA7qzZidgp9__… - R. On Aug 29, 2022, at 20:26, Yann Quidé <yannquide(a)gmail.com> wrote:        External Email - Use Caution        Hi all,Just bumping this one up.Any help would be really appreciated.Thanks in advance,YannOn 27 Aug 2022, at 10:47 am, Yann Quidé <yannquide(a)gmail.com> wrote:Hi team,I have been having troubles with processing my scans for thalamic subnuclei.In my previous post, I mentioned I was using FS7.3.2, but indeed realised that this version was not fully compatible with my OS (OSX 10.13). So I instead installed FS 7.2.0.However, I still have issues with the matlab runtime. I installed MCRv84, but received the  error <crash report deleted > It seems that FS is looking for v84 files that are missing in v93?Does anyone have an idea of what I am doing wrong?Thanks in advance.YannOn 20 Aug 2022, at 9:42 am, Yann Quidé <yannquide(a)gmail.com> wrote:Not a good news for me, but thanks!YannOn 20 Aug 2022, at 06:40, fsbuild <fsbuild(a)contbay.com> wrote:Hello Yann,It looks like the system library, /usr/lib/libSystem.B.dylib, is too old on your machine running MacOS (10.13) to work with the Matlab MCRv97 runtime environment.   MacOS 10.13 (High Sierra), 10.14 (Mojave) - are no longer supported by Apple.  Soon Apple will also drop support for MacOS10.15 (Catalina).  I think you would need to upgrade to Catalina to use the Matlab R2019b release as installed by the script ./freesurfer/bin/fs_install_mcr- R.On Aug 19, 2022, at 07:33, Yann Quidé <yannquide(a)gmail.com> wrote:        External Email - Use Caution        Hi team,I am trying to run the thalamic nuclei pipeline with FS7.3.2, on scans processed with recon-all from FS5.3 (mac OSX10.13).However, I receive the below error.Any idea how I could solve this?Thanks in advance.Yann------------------------------USER yannHOST Ians-Mac-Pro.localPROCESSID 86082 PROCESSOR x86_64OS DarwinITK_GLOBAL_DEFAULT_NUMBER_OF_THREADS 1 Darwin Ians-Mac-Pro.local 17.7.0 Darwin Kernel Version 17.7.0: Fri Oct 30 13:34:27 PDT 2020; root:xnu-4570.71.82.8~1/RELEASE_X86_64 x86_64------------------------------freesurfer-darwin-macOS-7.3.2-20220804-6354275 setenv SUBJECTS_DIR /Volumes/MacintoshHD2/mridata/IGP/ENIGMA_SZ_Thalamic_nuclei/Subjectscd /Volumes/MacintoshHD2/mridata/IGP/ENIGMA_SZ_Thalamic_nuclei/ENIGMA_BARI/Applications/freesurfer/7.3.2/bin/segmentThalamicNuclei.sh sub-BA_IGP001#--------------------------------------------#@# Thalamic Nuclei processing Fri 19 Aug 2022 20:10:24 AEST------------------------------------------Setting up environment variables---DYLD_LIBRARY_PATH is .:/Applications/freesurfer/7.3.2/MCRv97//runtime/maci64:/Applications/freesurfer/7.3.2/MCRv97//bin/maci64:/Applications/freesurfer/7.3.2/MCRv97//sys/os/maci64:/Applications/freesurfer/7.3.2/lib/gcc/lib:Invalid MEX-file '/private/tmp/MCR_3324153754/.mcrCache9.7/Segmen0/Users/developer/eugenio/GEMS2.build/bin/kvlGEMSMatlab.mexmaci64': dlopen(/private/tmp/MCR_3324153754/.mcrCache9.7/Segmen0/Users/developer/eugenio/GEMS2.build/bin/kvlGEMSMatlab.mexmaci64, 6): Symbol not found: ____chkstk_darwin  Referenced from: /private/tmp/MCR_3324153754/.mcrCache9.7/Segmen0/Users/developer/eugenio/GEMS2.build/bin/kvlGEMSMatlab.mexmaci64 (which was built for Mac OS X 10.15)  Expected in: /usr/lib/libSystem.B.dylib in /private/tmp/MCR_3324153754/.mcrCache9.7/Segmen0/Users/developer/eugenio/GEMS2.build/bin/kvlGEMSMatlab.mexmaci64Error in kvlClear (line 11)Error in SegmentThalamicNuclei (line 139)MATLAB:mex:ErrInvalidMEXFileDarwin Ians-Mac-Pro.local 17.7.0 Darwin Kernel Version 17.7.0: Fri Oct 30 13:34:27 PDT 2020; root:xnu-4570.71.82.8~1/RELEASE_X86_64 x86_64Thalamic nuclei exited with ERRORS at Fri 19 Aug 2022 20:11:24 AEST_______________________________________________Freesurfer mailing listFreesurfer@nmr.mgh.harvard..eduMailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://secure-web.cisco.com/1_dfW4osqXnmWV_vpUQGOrnVdHh-b1PBYl3XIG… mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

2 1

Re: [Freesurfer] Thalamic nuclei exited with ERRORS
by fsbuild 29 Aug '22

29 Aug '22

External Email - Use Caution Hello Yann, It looks like the system library, /usr/lib/libSystem.B.dylib, is too old on your machine running MacOS (10.13) to work with the Matlab MCRv97 runtime environment.   MacOS 10.13 (High Sierra), 10.14 (Mojave) - are no longer supported by Apple.  Soon Apple will also drop support for MacOS10.15 (Catalina).  I think you would need to upgrade to Catalina to use the Matlab R2019b release as installed by the script ./freesurfer/bin/fs_install_mcr - R. On Aug 19, 2022, at 07:33, Yann Quidé <yannquide(a)gmail.com> wrote:        External Email - Use Caution        Hi team,I am trying to run the thalamic nuclei pipeline with FS7.3.2, on scans processed with recon-all from FS5.3 (mac OSX10.13).However, I receive the below error.Any idea how I could solve this?Thanks in advance.Yann------------------------------USER yannHOST Ians-Mac-Pro.localPROCESSID 86082 PROCESSOR x86_64OS DarwinITK_GLOBAL_DEFAULT_NUMBER_OF_THREADS 1 Darwin Ians-Mac-Pro.local 17.7.0 Darwin Kernel Version 17.7.0: Fri Oct 30 13:34:27 PDT 2020; root:xnu-4570.71.82.8~1/RELEASE_X86_64 x86_64------------------------------freesurfer-darwin-macOS-7.3.2-20220804-6354275 setenv SUBJECTS_DIR /Volumes/MacintoshHD2/mridata/IGP/ENIGMA_SZ_Thalamic_nuclei/Subjectscd /Volumes/MacintoshHD2/mridata/IGP/ENIGMA_SZ_Thalamic_nuclei/ENIGMA_BARI/Applications/freesurfer/7.3.2/bin/segmentThalamicNuclei.sh sub-BA_IGP001#--------------------------------------------#@# Thalamic Nuclei processing Fri 19 Aug 2022 20:10:24 AEST------------------------------------------Setting up environment variables---DYLD_LIBRARY_PATH is .:/Applications/freesurfer/7.3.2/MCRv97//runtime/maci64:/Applications/freesurfer/7.3..2/MCRv97//bin/maci64:/Applications/freesurfer/7.3.2/MCRv97//sys/os/maci64:/Applications/freesurfer/7.3.2/lib/gcc/lib:Invalid MEX-file '/private/tmp/MCR_3324153754/.mcrCache9.7/Segmen0/Users/developer/eugenio/GEMS2.build/bin/kvlGEMSMatlab.mexmaci64': dlopen(/private/tmp/MCR_3324153754/.mcrCache9.7/Segmen0/Users/developer/eugenio/GEMS2.build/bin/kvlGEMSMatlab.mexmaci64, 6): Symbol not found: ____chkstk_darwin  Referenced from: /private/tmp/MCR_3324153754/.mcrCache9.7/Segmen0/Users/developer/eugenio/GEMS2.build/bin/kvlGEMSMatlab.mexmaci64 (which was built for Mac OS X 10.15)  Expected in: /usr/lib/libSystem.B.dylib in /private/tmp/MCR_3324153754/.mcrCache9.7/Segmen0/Users/developer/eugenio/GEMS2.build/bin/kvlGEMSMatlab.mexmaci64Error in kvlClear (line 11)Error in SegmentThalamicNuclei (line 139)MATLAB:mex:ErrInvalidMEXFileDarwin Ians-Mac-Pro.local 17.7.0 Darwin Kernel Version 17.7.0: Fri Oct 30 13:34:27 PDT 2020; root:xnu-4570.71.82.8~1/RELEASE_X86_64 x86_64Thalamic nuclei exited with ERRORS at Fri 19 Aug 2022 20:11:24 AEST_______________________________________________Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

2 3

mri_binarize error for thalamic nuclei during trac-all prep
by Makenna B McGill 29 Aug '22

29 Aug '22

External Email - Use Caution Hi Freesurfer Developers, I'm trying to troubleshoot an error message during TRACULA preprocessing that seems to be related to the mri_binarize command used to create the /dlabel/anatorig/aparc+aseg+thalnuc.nii.gz file. I'm using FS v7.3.2 on OS Monterey to run trac-all -prep and have attached both my configuration file and the log file. Here is the error output: 7.3.2 cwd /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula cmdline mri_binarize --i /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg.nii.gz --min .5 --dilate 4 --erode 2 --o /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg_mask.nii.gz sysname Darwin hostname PSYC-A30765 machine x86_64 user mbm3668 input /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg.nii.gz frame 0 nErode3d 2 nErode2d 0 output /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg_mask.nii.gz Binarizing based on threshold min 0.5 max +infinity binval 1 binvalnot 0 fstart = 0, fend = 0, nframes = 1 Found 1169534 values in range Dilating 4 voxels in 3d Eroding 2 voxels in 3d Counting number of voxels in first frame Found 1609388 voxels in final mask Writing output to /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg_mask.nii.gz Count: 1609388 1609388.000000 16777216 9.592700 mri_binarize done *printf: illegal option -- -* *usage: printf format [arguments ...]* *mri_binarize --i /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/t1prep/sub-0003107/mri/aparc+aseg.mgz --o /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg+thalnuc.nii.gz* *ERROR: must specify minimum and/or maximum threshold or match values* mri_concat --i /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg+thalnuc.nii.gz --i /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/t1prep/sub-0003107/mri/ThalamicNuclei.v13.T1.FSvoxelSpace.mgz --max --o /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg+thalnuc.nii.gz ninputs = 2 Checking inputs *error: niiRead(): error opening file /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg+thalnuc.nii.gz* *ERROR: reading /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/aparc+aseg+thalnuc.nii.gz* mri_binarize --i /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/t1prep/sub-0003107/mri/aparc+aseg.mgz --match 2 41 7 46 172 177 77 --o /Volumes/schnyer/Makenna/DODADNI/BIDS/derivatives/tracula/sub-0003107/dlabel/anatorig/White-Matter.nii.gz It seems like another user ran into the same issue ( https://secure-web.cisco.com/1udqzKCrR0Bj9BMcJxQpEnjsA37THUUcQftwPUH0tgkX_3…) I've made sure that the mri/ThalamicNuclei.v13.T1.FSvoxelSpace.mgz file is in fact there (I appreciate the update that now looks for either v12 and v13), and I've also tried running trac-all -prep without the thalamic nuclei segmentation, which was successful. I know that including this segmentation is preferred for tracts near the thalamus, so any suggestions you might have for troubleshooting this issue would be greatly appreciated. Let me know if there is any additional information I can provide. Best, Makenna *Makenna McGill *(she/her) Doctoral Student in Clinical Psychology University of Texas at Austin

1 0

Re: [Freesurfer] samseg inputs
by Cerri, Stefano 29 Aug '22

29 Aug '22

Hi Dilek, I've actually run SAMSEG-lesion on some gadolinium-enhanced scans in the paper. Look at the top row of Fig. 7 ("T1c") and at the "MSseg" dataset. I did not check the performance of the method on such data. We also don't distinguish between enhanced and non-enhanced lesions. I would probably set --lesion-mask-pattern to "0" if you want to segment most of the lesions, or --lesion-mask-pattern to "1" if you want to segment only the enhanced lesions. It is usually not advised to analyze segmentations/volumes obtained using different contrasts as you might introduce biases in your analysis. Let us know how it goes, Stefano Hello dear all, I want to use Freesurfer-Samseg for MS lesion segmentation(cross-sectional). I have read the paper, accordingly I want to try the t1-flair combination in terms of the accuracy. Do you have any experience with T1 images with contrast agent? I couldn't find anything related to those in the paper. Can I use t1 with contrast in a few patients if the patient doesn't have t1 image without contrast agent? i will declare this situation in the results. Best, Dilek

1 0

Freesurfer VM package
by Meier, Kolja 29 Aug '22

29 Aug '22

External Email - Use Caution Hello, I want to install the VM Package of Freesurfer and was asked for a passcode to upack the zip file. Could you provide me one? Best Kolja

1 0

Van Essen/Glasser Lab Connectome Post-Doc Opportunity
by Glasser, Matt 29 Aug '22

29 Aug '22

External Email - Use Caution Hi FreeSurfer Users, We have a Connectome-related Post-Doc position available immediately in the Van Essen/Glasser lab. Please see the below posting. Feel free to forward to anyone you know who might be interested. Thanks, Matt. Postdoctoral Research Associate (CIRC, Van Essen/Glasser Lab) Position Summary The Adult Aging Brain Connectome (AABC) project has a fully funded and immediately available opening for a Postdoctoral Research Associate within the Informatics, Data Analysis, and Statistics Core (IDASC). This position is based in the Van Essen/Glasser Lab<https://secure-web.cisco.com/1WZvYXfgvuq53cGKgRSB25U5IkW7IQITwIczQxqUnhtsuz…> within the Computational Imaging Research Center (CIRC)<https://secure-web.cisco.com/1uaguVfyCEHq6WsESzrPjgSjZD2SaiuQ4MEvw4TyDAbAUm…> of Mallinckrodt Institute of Radiology (MIR), at Washington University School of Medicine in St. Louis. The AABC is a longitudinal extension of the Human Connectome Project in Aging that seeks to determine the structural and functional brain connectivity trajectories associated with vulnerability and resilience to developing late life dementia. Among the IDASC’s primary aims is to continue the Human Connectome Project (HCP)’s<https://secure-web.cisco.com/1TJ34rCWxWlzejI-JhfFwndMEDUQ2t4IHs7v-4AbXCZ2U3…> innovative neuroimaging methods development, with a particular focus on HCP-Style data analysis optimized for generating multi-modal imaging data phenotypes from longitudinally acquired scans. The selected candidate will have access to the HCP’s large consortium of world experts in multiple neuroimaging domains and ample opportunities to publish research methods and scientific findings in high impact academic journals and at premier scientific conferences. The Van Essen/Glasser lab’s senior staff are highly experienced, having developed much of the HCP’s software themselves, and are excellent at training new users of HCP-Style brain imaging methods. The CIRC hosts a vibrant community of postdoctoral and graduate student trainees together with multiple additional faculty and state of the art computational resources. The position is renewable annually based on performance, and funding is available for up to 4 years. Information on being a postdoc at Washington University in St. Louis can be found at postdoc.wustl.edu/prospective-postdocs<https://secure-web.cisco.com/1Gg_mxIwFSYfuZnIdHM-VK6rM2TfCf8l5Su1ssLBefTvaR…>. Primary Duties & Responsibilities * Develop and aid in the design of longitudinally optimized HCP Pipelines for multi-modal neuroimaging data analysis, including structural, functional, diffusion, and Arterial Spin Labeling (ASL) modalities. * Develop and aid in the design of HCP Pipelines for deriving multi-modal Imaging Data Phenotypes (IDPs) at the neuranatomical levels of brain areas, functional networks, and white matter tracts. * Use the longitudinally optimized multi-modal IDPs to carry out statistical analyses relating these IDPs to behavioral and biological measures relevant for cognitive and dementia outcomes. * Serve as first author for scientific publications and presentations with primary responsibility for preparing, submitting, revising, and presenting such work. * Be a good citizen in the lab/center and follow all applicable institutional policies. Required Qualifications Ph.D. in the field of biomedical engineering, imaging science, neuroscience, or a related field. Preferred Qualifications * Record of accomplishments in at least one of the following fields: neuroimaging, medical image analysis, or a related field. * Proficiency in quantitative analytical methods and computer programming in a Unix/Linux environment (e.g., shell scripting, Matlab, Python). * Prior experience with HCP-Style methods, HCP data, and/or the HCP Pipelines. * Prior experience with the commandline implementations of FSL, FreeSurfer, and Connectome Workbench softwares. * Excellent interpersonal and communication skills, both verbal and written. * Ability to work effectively both independently and within a multidisciplinary team. Salary Range Base pay is commensurate with experience. Applicant Special Instructions To apply, please contact Dr. Matthew Glasser with the following: * Cover letter describing your research interests. * Current curriculum vitae. * Name and contact information for three references who are familiar with your work. * Two or three representative publications. Matthew F. Glasser, M.D., Ph.D. Instructor Washington University School of Medicine Mallinckrodt Institute of Radiology E-mail: glasserm(a)wustl.edu<mailto:glasserm@wustl.edu> ________________________________ The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

1 0

samseg inputs
by Dilek Özkan 29 Aug '22

29 Aug '22

External Email - Use Caution Hello dear all, I want to use Freesurfer-Samseg for MS lesion segmentation(cross-sectional). I have read the paper, accordingly I want to try the t1-flair combination in terms of the accuracy. Do you have any experience with T1 images with contrast agent? I couldn't find anything related to those in the paper. Can I use t1 with contrast in a few patients if the patient doesn't have t1 image without contrast agent? i will declare this situation in the results. Best, Dilek

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