Freesurfer

freesurfer@nmr.mgh.harvard.edu

8 participants
30078 discussions

Subregion Segmentation of thalamus for TRACULA
by Makenna B McGill 29 Jul '22

29 Jul '22

Hi FreeSurfer Developers, I'm trying to run TRACULA in FreeSurfer v7.3 on OS Monterey. I was unable to download R2014b Runtime on my newer OS, so instead of the segmentThalamicNuclei.sh script, I tried running the newer segmentation tool segment_subregions thalamus ( https://surfer.nmr.mgh.harvard.edu/fswiki/SubregionSegmentation) This was successful, but trac-all looks for the ThalamicNuclei.v12.T1.FSvoxelSpace.mgz file, whereas the subregion segmentation produces the file ThalamicNuclei.FSvoxelSpace.mgz. I'm wondering two things: (1) is the new segmentation tool compatible with trac-all (the wiki says the results will be very close but not identical), and (2) if so, would it be appropriate to rename/link the file so that trac-all can find it, as this user did with the ThalamicNuclei.v13 file ( https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg73065.html)? Thank you in advance for any suggestions you might have, and please let me know if anything needs clarified. Best, Makenna *Makenna McGill *(she/her) Doctoral Student in Clinical Psychology University of Texas at Austin

2 2

Problem with hypothalamus subunits?
by Zeng, Victor (BIDMC - Keshavan - Psychiatry) 29 Jul '22

29 Jul '22

External Email - Use Caution Hi all, I've downloaded a new dev version of FS 7/19/2022 - and I receive an error with running subunits: mri_segment_hypothalamic_subunits --s 10893/ using 1 thread /mnt/QBERT/freesurfer/python/scripts/mri_segment_hypothalamic_subunits:1023: DeprecationWarning: get_data() is deprecated in favor of get_fdata(), which has a more predictable return type. To obtain get_data() behavior going forward, use numpy.asanyarray(img.dataobj). * deprecated from version: 3.0 * Will raise <class 'nibabel.deprecator.ExpiredDeprecationError'> as of version: 5.0 volume = np.squeeze(x.get_data()) predicting 1/1 the following problem occured when preprocessing image /..../0893/mri/nu.mgz : 'NoneType' object is not subscriptable resuming program execution segmentation saved in: /..../10893/mri/hypothalamic_subunits_seg.v1.mgz volumes saved in: /..../10893/mri/hypothalamic_subunits_volumes.v1.csv resampled images saved in: /..../10893/mri/hypothalamic_subunits_nu_resampled_1mm.v1.mgz If you use this tool in a publication, please cite: Automated segmentation of the hypothalamus and associated subunits in brain MRI B. Billot, M. Bocchetta, E. Todd, A. V. Dalca, J. D. Rohrer, J. E. Iglesias NeuroImage 2020 Is this a problem with tensorflow? It does not seem clear. No files are actually being written Victor Zeng Beth Israel Deaconess Medical Center Keshavan Lab --

2 3

Amygdala Subfields - longitudinal volumes differ a lot from the cross-sectionals
by Andrea Dell'Orco 29 Jul '22

29 Jul '22

External Email - Use Caution Dear all, I have a Cohort of subjects with multiple time points. I have segmented the T1w with the normal recon-all and sub fields and then with the longitudinal pipeline as described in the wiki. When I compare the sub fields Volumes, there is a good agreement, with the expected variability, between longitudinal and not longitudinal for the Whole Hippocampus rh and lh. The Volumes for the amygdala are on the other hand, really different, especially for the left Whole amygdala. Here some plots with the correlation longitudinal vs non-longitudinal pipeline: https://secure-web.cisco.com/14VEQLZEbs96z59Y5ty9CueQKgO9cELG9hY62vKSTRXY_0… Do someone has some idea, what can be the problem? Regards, Andrea Dell'Orco -- Andrea Dell'Orco Wissenschaftlicher Mitarbeiter Charité – Universitätsmedizin Berlin NeuroCure Clinical Research Center NCRC AG Kognitive Neurologie Charitéplatz 1, D-10117 Berlin Tel +49 30 450 560 185 Fax +49 30 450 7 560 280 Interne Besuchsadresse: CCM, Sauerbruchweg 5, E2 http://secure-web.cisco.com/1ePTkJygSqAcb6M2j8DqUJJZTkIxZCuN07o23jJpGzlCGkh…

3 3

etiv value differs if using recon all vs freesurfer dev
by Miriam Taza 28 Jul '22

28 Jul '22

External Email - Use Caution Hello, I am getting different etiv values when running recon(orange) all VS mri_sclimbic_seg (blue). My analysis uses limbic volumes extracted from the dev version but I am not sure if I should just use the etiv from recon all, given the data is tighter (see attached). Why is there this discrepancy and what do you recommend? [cid:37ebf2b4-95b9-4b28-8fe6-86715ee89d72] Thank you!

2 8

Re: [Freesurfer] Cortical Thickness measures using different ways of processing
by Ariana Fei 28 Jul '22

28 Jul '22

External Email - Use Caution I'm attaching a list of command I used to convert freesurfer cortical thickness output to native space: $Caret5_Command -file-convert -fsc2c "$FreeSurferFolder"/surf/"$hemisphere"h.thickness "$FreeSurferFolder"/surf/"$hemisphere"h.white "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii $Caret5_Command -spec-file-add "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".native.c5.spec surface_shape_file "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii $Caret5_Command -metric-set-column-name "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii 1 "$Subject"_"$Hemisphere"_Thickness $Caret5_Command -file-convert -format-convert XML_BASE64 "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii $Caret7_Command -set-structure "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii $Structure $Caret7_Command -add-to-spec-file "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject".native.wb.spec $Structure "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii $Caret7_Command -metric-palette "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii MODE_AUTO_SCALE_PERCENTAGE -pos-percent 2 98 -palette-name Gray_Interp -disp-pos true -disp-neg true -disp-zero true #cp "$AtlasSpaceFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii "$T1wFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii #commented out GW $Caret7_Command -add-to-spec-file "$T1wFolder"/"$NativeFolder"/"$Subject".native.wb.spec $Structure "$T1wFolder"/"$NativeFolder"/"$Subject"."$Hemisphere".thickness.native.shape.gii This already gives negative cortical thickness values in native space. Ariana

2 1

Unsubscribe?
by Archibald, Jessica 28 Jul '22

28 Jul '22

External Email - Use Caution Hi, I signed up last night to send a freesurfer question, but I figured it out. Now I am getting bombarded with emails from this thread and can’t see a ‘unsubscribe’ button. Can you help with this?? Thanks in advance, Jessica Jessica Archibald M.Sc. Kramer Lab- PhD candidate Experimental Medicine ICORD | Blusson Spinal Cord Center 818 W 10th Ave Vancouver, BC, Canada. [UBC E-mail Signature]

2 1

Help with improving the training of an alternative atlas for medial temporal lobe subregions
by Henzen Nicolas 28 Jul '22

28 Jul '22

External Email - Use Caution Hi all, We are aiming at training an alternative atlas of some subregions in the medial temporal lobe using FreeSurfer. As training examples, we manually segmented the subregions using tkmedit and tksurf. In this process we deleted the original segmentation labels around our new regions of interest (see Figure 1). Then, we trained the atlas and parcellated new data as follows: # Train atlases for hemi in lh rh; do mris_ca_train -n 2 \ -t ./doc/LUT.txt \ ${hemi} \ sphere.reg \ CUSTOM_ROIs.annot \ S01 S02 S03 S04 S05 S06 \ ./atlas/${hemi}.CUSTOM_ROIs.gcs done # Create annotations for hemi in lh rh; do for ID in S10 S11 S12 S13; do mris_ca_label \ -orig white \ -novar \ -t doc/ LUT.txt \ ${ID} \ ${hemi} \ sphere.reg \ ${hemi}.CUSTOM_ROIs.gcs \ $SUBJECTS_DIR/${ID}/label/${hemi}.CUSTOM_ROIs.annot done done [cid:image001.png@01D89C87.200FE420]This lead to a very inaccurate parcellation (see Figure 2), also when we filled the unlabeled region with a dummy label (see Figure 3). Figure 1. Example for manual segmentation of the 4 subregions (encircled) with deleted original segmentation labels around. [cid:image003.png@01D8A190.FBE5BED0] Figure 2. Parcellation example using the trained atlas without dummy label for unlabeled regions. [cid:image006.png@01D8A192.42B1DF00] Figure 3. Parcellation example using the trained atlas with dummy label for unlabeled regions. Any comments or suggestions on how to deal with our problem would be greatly appreciated. Many thanks for having taken the time to read and reflect Best, Nicolas

3 2

Cortical Thickness measures using different ways of processing
by Ariana Fei 28 Jul '22

28 Jul '22

External Email - Use Caution Hi, We are recently trying to get cortical thickness from freesurfer and converting them to 32k. A problem we notice is that after converting to 32k, all the cortical thickness values become negative. However, when we load freesurfer thickness files in freeview, everything appears normal. We wonder if merely inverting what we currently have in 32k would be a good idea. Alternatively, we also calculated the Euclidean distance of pial and white after they have been converted into 32k space, which should be cortical thickness. However, when comparing the inverted values and the "pial-white"values, we find that there are some differences. In this case, has anyone ever had similar issues and have suggestions on which one to trust? Thanks! Ariana

2 1

Applying bonferroni clusterwise correction for 2spaces with PALM F-contrast
by Hye Min Shin 28 Jul '22

28 Jul '22

External Email - Use Caution Hello experts, I am going through FreeSurfer's repeated measures anova tutorial using fspalm. I understand that fspalm frontend can only do t-tests at the moment and NOT F tests as noted in this thread: https://secure-web.cisco.com/1t11cJOK6FCdHxqm_O6WFHk15uAjWyctjoR7VAvgTW5hEa… So I am trying to run F-contrast on PALM but I was wondering if I also have to worry about bonferroni clusterwise correction. If I used --2spaces option when using fspalm frontend, I thought I should also try to apply the equivalent bonferroni correction when running F-test on FSL palm. For the F-test on FSL palm, it looked like I could run the following command with the addition of -f and -ftestonly flags (shown in underline): palm -i lh.thickness.mgz -i rh.thickness.mgz -d design.mat -t design.con *-f design.fts* -o myresults -n 2000 -corrcon -corrmod *-ftestonly* taken from this email thread- https://secure-web.cisco.com/1vFEFTKkUMsHsCHW3u3SsOIpxhqtkKCXdjWw6TLTufecii… Would this be equivalent to the bonferroni correction methods that fspalm uses? When I go through fspalm.log it seemed to run mri_surfcluster and mri_segstats after running fspalm with certain commands having --sig2p-max and --bonferroni-max 2 flags. Thank you in advance for your help! Best, Julia Shin

1 0

mri_jacobian questions
by Danielle Carrol 27 Jul '22

27 Jul '22

External Email - Use Caution Hello Freesurfer experts! I have been trying to send this question a few times, but it has never shown up on the archive, so I'm not sure if it has been received. I apologize if everyone has been receiving the same question multiple times! I am trying to compute the Jacobian for use in VBM. I found the following command on this email archive (Re: [Freesurfer] Jacobian mri_cvs_register (mail-archive.com) <https://secure-web.cisco.com/1PxtNE3QwqzX1SDyVeD_p5OwDGQgLiBBekceqUimQr0kRg…>: mri_jacobian <3d morph> <template volume> <output volume> I am assuming the 3d morph is the file mri/transforms/talairach.m3z, which is created during recon-all, since this file provides the nonlinear warp. However, I am confused about which file should be the “template volume”. Is this the final transformed/registered volume for each subject: ${SUBJECTS_DIR}/<subject name>/mri/brain.mgz)? Is it the fsaverage MNI template: $FREESURFER_HOME/subjects/fsaverage/mri/mni305.cor.mgz? I also thought it could be this file: freesurfer-latest/average/RB_all_2016-05-10.vc700.gca since I saw on the wiki page for recon-all that in the creation of the .m3z morph file, the nonlinear transform is computed to align with the GCA atlas, so that could be the "template" (recon-all - Free Surfer Wiki (harvard.edu) <https://secure-web.cisco.com/1dWy6gXJGfzUBSGAO9bR9NM58WNn9B3AwqHd2Y8dt6SwTh…> ). I am using version 6.0.0. Thank you for you advice and time, Danielle PhD Student University of Wisconsin-Madison

3 3

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