Freesurfer

freesurfer@nmr.mgh.harvard.edu

3 participants
30135 discussions

Cortical thickness average
by Sahil Bajaj 25 Mar '16

25 Mar '16

Hello FreeSurfer community, I calculated cortical thickness for 100 controls following the instructions described here: https://surfer.nmr.mgh.harvard.edu/fswiki/VolumeRoiCorticalThickness So I have 100 lh.thickness.fsaverage.mgh, one for each subject. I was wondering if there is any way to average the thickness over these 100 lh.thickness.fsaverage.mgh files to get one final lh.thickness.fsaverage.mgh I am interested in calculating cortical thickness for 15 previously defined ROIs, but I do not want to calculate 15x100 text files using mri_segstats i.e. 15 text files for each subject and then average over those but I would rather prefer to average over 100 lh.thickness.fsaverage.mgh files first and then calculate thickness for each ROI to get 11 text files. Any help would be really appreciated. Thanks, Sahil -- ------------------------------------------------- Sahil Bajaj Post-doctoral Fellow Nantz National Alzheimer's Center, Department of Neurology The Houston Methodist Research Institute (THMRI) Houston, TX, USA. E-mail:sahil.brain@gmail.com <E-mail%3Asahil.phy(a)gmail.com>

2 1

Re: [Freesurfer] FA and MD *positively* correlated within same regions ?
by Elijah Mak 25 Mar '16

25 Mar '16

Hi Kirstie, Yes indeed there were 3 subjects at the bottom-left of the plot. To account for PV, I eroded the boundaries using *-seg-erode 1* during the *mri_segstats* stage, and this resulted in a few subjects with no data for the WM entorhinal region (although these were saved as "0" in the .stats file). Do people generally do this for DTI studies? All good now. Many thanks again, Kirstie. Best Wishes, Elijah

2 1

FA and MD *positively* correlated within same regions ?
by Elijah Mak 24 Mar '16

24 Mar '16

Hi Freesurfer Community, I recently used mri_segstats to extract FA and MD as suggested here: http://freesurfer.net/fswiki/FsTutorial/Diffusion Using the aparc+aseg and wmparc (resampled to diffusion space), I looked at some of the correlations between FA and MD in the same regions, and came across a few interesting ones. wm_lh_entorhinal FA & wm_lh_entorhinal MD: r = 0.6, p <0.001 ctx_lh_entorhinal FA & ctx_lh_entorhinal MD: r = -0.3, p < 0.020 wm_lh_precuneus FA & wm_lh_precuneus MD: r = -0.6, p <0.001 As I understand, a clear-cut interpretation especially with regards to FA changes is not so easy! But could someone discuss / speculate as to why FA and MD could be positively correlated within the entorhinal WM region? That was found separately within a group of a elderly healthy controls and a group of dementia subjects. Thank you for your time! Best Wishes, Elijah

3 2

Re: [Freesurfer] Trac-all -path Segmentation error
by Anastasia Yendiki 24 Mar '16

24 Mar '16

Hi Katherine - The dlabel/ directory is created during the -prep step, when the structural segmentation gets mapped from T1 to diffusion space. However bedpost may still run fine without the dlabel/ directory. It seems that the problem is fairly early in the -prep step. Any errors there? Best, a.y On Wed, 23 Mar 2016, Katherine Damme wrote: > Hey Anastasia, > > > Although the bedpost appears to complete properly the dlabel/diff is not > being created during the bedp step. I am not sure how or why this error is > occurring. That would, it appears, definitely be the cause of the failure. > However, I am not getting an error during the prep or the bedpost stage, and > so I am still completely lost any advice would be deeply appreciated! > > Thank you, > > KD > > On Wed, Jan 13, 2016 at 6:07 PM, Anastasia Yendiki > <ayendiki(a)nmr.mgh.harvard.edu> wrote: > > Hi Katherine - Do the freesurfer segmentations look ok? What > about the registration from structural to diffusion? You can > inspect those quickly with: > > freeview dmri/dtifit_FA.nii.gz > dlabel/diff/aparc+aseg.bbr.nii.gz:colormap=lut:opacity=.3 > > Hope this helps, > a.y > > On Sun, 3 Jan 2016, Katherine Damme wrote: > > First I should note that using the same scripts I > completed 52 participants successfully and am > currently trying to > batch process the newest 42 participants that are > only failing at the path level. Upon Visual > Inspection they all > appear normal (with the exception of the one > attached). After carefully comparing a random sample > of 5 from each > group images from the files that failed versus the > files that were successful, I found: > > 1. The range of values in the mean_ph1samples in the > successful ( i.e. -3.00 to 87.30 ) differed from the > failures > (i.e. -2 to 17) (the successful samples varied on > their high end from 50-114 but the failures were on > the 5-24 > range) > 2. The range of values in the mean_th1samples in the > successful ( -0.02 to 19.12) differed from the > failures > (-0.00267 to 7.176) (the successful samples varied > on their high end from 19-17 but the failures were > on the 3-10 > range) > 3. Some, but not all (2/5), mean_S0samples, had a > abnormal range in successful (0 to 19998.4) compared > to the > unsuccessful (0 to 709.96) > > However, I understand that the diffusion units are > likely arbitrary and not sure these patterns will > hold in a > larger sample. > > I know this is a lot of confusing information and so > in response to your question. After an initial > inspection > there are two types of output: > 1. data that appears visually normal with a slightly > altered range. > Should I be concerned about the range if all other > data looks fine? > > 2. data that appears visually normal with a normal > range and bizarre merged files. > What do you think could be a cause of these odd > merged files? > > > > > On Tue, Dec 22, 2015 at 1:54 PM, Anastasia Yendiki > <ayendiki(a)nmr.mgh.harvard.edu> wrote: > > Hi Katherine - Have you looked at any of the > intermediate images? Do they look normal? > > a.y > > On Mon, 21 Dec 2015, Katherine Damme wrote: > > Hello Everyone! > > It looks as though my prep and bedp > stages completed properly, but the prep stage ends > in a > segmentation > errror. > > See attached and let me know your > thoughts. I would prefer not to have to redo all > steps if > possible! > Thank you! > > > > > The information in this e-mail is intended only for > the person to whom it is > addressed. If you believe this e-mail was sent to > you in error and the e-mail > contains patient information, please contact the > Partners Compliance HelpLine at > http://www.partners.org/complianceline . If the > e-mail was sent to you in error > but does not contain patient information, please > contact the sender and properly > dispose of the e-mail. > > > > > >

1 0

Re: [Freesurfer] Cortical thickness average
by Elijah Mak 24 Mar '16

24 Mar '16

Hi Sahil, I did something similar recently. I believe you can use mri_concat with the -mean option to create an average map of the group. Another option might be make_average_subject? https://surfer.nmr.mgh.harvard.edu/fswiki/make_average_subject Best Wishes, Elijah

2 1

Re: [Freesurfer] FA and MD *positively* correlated within same regions ?
by Elijah Mak 24 Mar '16

24 Mar '16

Hi Alshikho <http://www.mail-archive.com/search?l=freesurfer@nmr.mgh.harvard.edu&q=from:…> , Yes, I am sure :) Same sequence for everyone. Just to add. FA and MD were significantly negatively correlated for the other regions including the fusiform, precuneus, hippocampus, thalamus, etc. So I am very curious about what could be going on with the ERC here? Best Wishes, Elijah

2 1

ROI within subject
by Ji Won Bang 24 Mar '16

24 Mar '16

Dear. Freesurfer experts. Hi. How are you? I have a question about making ROI. This might be a naive question, but any input would be greatly appreciated. I preprocessed scans per run, thus I have register.dof6.dat file inside all run epi directories. Then, I made ROIs per single subject based on the single subject's functional contrast map using the command: tksurfer replay06 lh inflated -curv -gray I overlayed the single subject's contrast result sig.nii.gz (obtained from 3 runs) and the registration file in 1 run(register.dof6.dat in 1 among 3 runs), then I cut the plane and then flattenr the cortex, made ROI. While trying to check the ROIs, I realized that this ROI seems to be slightly different depending on which register.dof6.dat I use (I tried all 3 runs' register.dof6.dat when overlaying). tkmedit -f $SUBJECTS_DIR/$SUBJECT/mri/norm.mgz -overlay $DATA_DIR/$SUBJECT/roimask/test.nii -overlay-reg $DATA_DIR/$SUBJECT/bold_retino/025/register.dof6.dat -fthresh 0.01 I thought ideally, this functional ROI should be the same across all runs if thes register files are perfectly registering each functional sans to anatomical scan. Is this a usual way when people make ROI based on functional contrast map per each subject? The way I did in the past did not have this kind of problem because I did motion correction as targetting the first EPI (aligning all images to the first EPI) and then used only 1 (same) register.dat file.. Please share your idea with me. Thank you very much. Best, Ji Won

1 0

errors message in QAtools
by Lars M. Rimol 24 Mar '16

24 Mar '16

Hi, I'm running QAtools on a set of subjects processed with FS 5.3.0, and consistently get this message when recon_checker is running: Processing each subcortical label ... (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error (standard_in) 1: syntax error etc. *And in the recon_check file I notice several of my subjects have this error message:* ERROR: Recon-all steps missing in subject <SUBJ_ID> status file Expected to see the following step(s): #@# Remove Neck ti. 22. mars 15:02:33 +0100 2016 in the recon-all-status.log file before (line # 12): --> CA Reg Inv *I ran recon-all -all, and the recon-all log file seems to indicate neck removal was peformed*; from recon-all.log: mri_remove_neck -radius 25 nu.mgz transforms/talairach.m3z /cluster/software/VERSIONS/freesurfer-5.3.0/average/RB_all_2008-03-26.gca nu_noneck.mgz erasing everything more than 25 mm from possible brain reading atlas '/cluster/software/VERSIONS/freesurfer-5.3.0/average/RB_all_2008-03-26.gca'... reading input volume 'nu.mgz'... reading transform 'transforms/talairach.m3z'... -- writing output to nu_noneck.mgz... nonbrain removal took 0 minutes and 51 seconds. #-------------------------------------- #@# SkullLTA ti. 22. mars 14:45:17 +0100 2016 mri_em_register -skull -t transforms/talairach.lta nu_noneck.mgz /cluster/software/VERSIONS/freesurfer-5.3.0/average/RB_all_withskull_2008-03-26.gca transforms/talairach_with_skull_2.lta ======= NUMBER OF OPENMP THREADS = 1 ======= aligning to atlas containing skull, setting unknown_nbr_spacing = 5 -- reading 'nu_noneck.mgz'... freeing gibbs priors...done. bounding unknown intensity as < 20.2 or > 943.7 total sample mean = 92.0 (1443 zeros) ************************************************ spacing=8, using 3481 sample points, tol=1.00e-05... Apart from this, the snapshots look ok (except for one subject I asked about in a previous e-mail) and there are no outliers in any of the subjects (according to QAtools). Any idea why I get this error message? Thank you! sincerely yours, Lars M. Rimol, PhD Senior researcher, Norwegian Advisory Unit for functional MRI Department of Radiology, St. Olav's University hospital 7006 Trondheim, Norway

1 0

OpenMP and GPU implementation in freesurfer 6 beta
by Francis Tyson Thomas 24 Mar '16

24 Mar '16

Hi, I have been working on speeding up the the hippocampal segmentation in freesurfer 6 beta and I have been playing around with both -openmp and -use-gpu flags. I noticed that freesurfer does seem to make use of all the threads in freesurfer 6 irrespective of passing the -openmp flag or not. Is that the right behaviour. I found this by analysing the cpu usage in both the cases. Also, it looks like development for GPU usage has been halted for now and so I was trying to use the CUDA 5 for getting it working under ubuntu 14.04. I have been not successful that as the cuda device isn't getting selected. Do you have any recommendations on that? And lastly, there is the itkthreads option available to -hippocampal-subfields-T1, so can I combine all three or two of these options to minimize the execution time - like GPU and itkthreads or GPU, OPENMP and itkthreads? Also the v6 beta is no longer available on your website. Is there any particular reason as to why it was removed and can I continue using the copy that I have for hippocampal segmentation ? Thanks, Tyson

2 7

Surface based analysis using FA maps
by Alshikho, Mohamad J. 24 Mar '16

24 Mar '16

Dear FS experts, I would like to run surface based analysis using FA maps. I am not exactly sure that what I am doing is totally right specially step 2 . I used the following commands: 1. I registered the FA maps to T1 using the command: bbregister --s bert --mov dtifit_FA.nii --bold --init-fsl --reg reg.dat 2. I concatenated the subject's FA maps using the command : mris_preproc -target fsaverage -hemi lh -iv .bert1/dtifit_FA.nii reg.dat -iv bert2/dtifit_FA.nii reg.dat .... --projfrac 0.5 --out lh.FA.mgz 3. Then smoothing mri_surf2surf --hemi lh --s fsaverage --fwhm 5 --cortex --sval lh.FA.mgz --tval lh.FA_5mm.mgz 4. Then GLM analysis using mri_glmfit 5. Correction for multiple comparison mri_glmfit-sim Kindly, are these steps correct? I used the previous steps to do surface based analysis using PET images and it worked very well. While for the FA maps it seems that there is something wrong and I can't figure out what it is! I highly appreciate any input on this! Best, Mohamad

3 8

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