Freesurfer September 2017

freesurfer@nmr.mgh.harvard.edu

109 participants
199 discussions

TMI(total motion index) of diffusion MRI
by 이 정환 06 Oct '17

06 Oct '17

Hello I’m attempting to analysis diffusion MRI using TRACULA. As mentioned on article "Spurious group differences due to head motion in a diffusion MRI study", I tried to calculate TMI(total motion index) first. The problem is that, upper and lower quartile of Percent Bad Slices are both 0. Upper and lower quartile of avg dropout score are both 1. Calculating TMI on some subjects using those values figured out at questionable results because difference of upper and lower quartile values is used as denominator in formula. I those cases, should I exclude Percent Bad slices and avg dropout score calculating TMI? Does anyone have any thoughts on it? Thanks in advance.

2 1

TRACULA trac-all prep error
by Fred Sampedro 06 Oct '17

06 Oct '17

Dear TRACULA users, I'm trying to preprocess a DTI subject in TRACULA which has an already reconed-all error-free T1. I followed the tutorial, but trac-all -prep exited with errors very fast. Please find attached the log and error files, and the dmrirc. Also bvals and bvecs (were obtained from dcm2nii). The raw dti.nii.gz has 16 volumes, the first being b0 and then 15 directions. Any ideas on what could be happening would be much appreciated, Thanks a lot in advance, Fred-

2 1

Test-retest reliability in longitudinal data
by Worker, Amanda 04 Oct '17

04 Oct '17

Hi all, I have a longitudinal dataset that I'd like to calculate test-retest reliability for. However, I am not sure whether to calculate this for the cross-sectionally processed data or longitudinal data? It would seem to make sense to use the cross-sectional data, as the time points are independent, but then it means that the test-retest results would not be applicable in a dataset processed longitudinally. On the other hand, calculating reliability metrics for data processed in the exact same way as will be used in further studies seems to make sense also, but would calculating test-retest on the longitudinally processed data bias the results, as the data points are not fully independent? Does anyone have any idea of the best way forward? Thanks, A

4 6

generating automatic functional label
by Yagmur Ozdemir 19 02 Oct '17

02 Oct '17

Hello FreeSurfer developers, This is a piece of code I came across for producing a functional label with using time course estimates from MEG/EEG data, for freesurfer use. I am wondering if something similar is available to estimate a functional label/region from fMRI data with similar tresholding. https://martinos.org/mne/stable/auto_examples/inverse/plot_label_from_stc.h… Best, Idil

2 3

fwhm in area analysis
by Daniel Yang 02 Oct '17

02 Oct '17

Hi FreeSurfer experts, When I specify fwhm=15 to analyze area, mri_surfcluster said that it cannot find fwhm31 in fsaverage/${hemi}/cortex/fwhm31/... This did not happen when I used FreeSurfer 5.3.0 I am wondering if FreeSurfer 6.0.0 pre-smooths the surface area? Thanks, Daniel

3 3

qdec error
by Seung-Gul Kang (강승걸) 02 Oct '17

02 Oct '17

Dear Freesurfer, I have a question during qdec. I tried to compare the surface area between two groups using qdec program. Strange to say, if I control some combination of confounding factors, then the error develops and qdec is closed. When I control age, sex, HAMD (hamilton depression) score, and mean surface area, there was no error. When I control age and sex, there was no error. When I control only surface area, there was no error. However, when I control age, sex, and mean surface area, then error happens. Could you let me how to manage this problem? I am looking forward to receive your answer. Thank you. Best, Seung-Gul ---------------------------------------------------------------------------------below is the messages containing error SUBJECTS_DIR is '/mnt/hgfs/share/suicide' lh-Avg-Intercept-area ----------------------- Does the average area differ from zero? Nuisance factors: age area 1.000 1.000 1.000 1.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000; lh-Diff-NS-S-Intercept-area ----------------------- Does the average area, accounting for sex, differ between NS and S? Nuisance factors: age area 1.000 -1.000 1.000 -1.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000; lh-Diff-M-F-Intercept-area ----------------------- Does the average area, accounting for diagnosis, differ between M and F? Nuisance factors: age area 1.000 1.000 -1.000 -1.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000; lh-X-diagnosis-sex-Intercept-area ----------------------- Is there a diagnosis--sex interaction in the mean area? Nuisance factors: age area 1.000 -1.000 -1.000 1.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000; ninputs = 38 Checking inputs nframestot = 38 Allocing output Done allocing nframes = 38 Writing to /mnt/hgfs/share/suicide/qdec/area_LH_AA/y.mgh gdfReadHeader: reading /mnt/hgfs/share/suicide/qdec/area_LH_AA/qdec.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 age 41.5263 12.8201 1 area 163656 15524.4 Class Means of each Continuous Variable 1 diagnosisNS-sexM 36.2000 176040.8391 2 diagnosisS-sexM 40.6250 170897.0234 3 diagnosisNS-sexF 46.5556 151977.3351 4 diagnosisS-sexF 42.9091 156686.6932 INFO: gd2mtx_method is dods Reading source surface /mnt/hgfs/share/suicide/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 /home/sgk cmdline mri_glmfit --y /mnt/hgfs/share/suicide/qdec/area_LH_AA/y.mgh --fsgd /mnt/hgfs/share/suicide/qdec/area_LH_AA/qdec.fsgd dods --glmdir /mnt/hgfs/share/suicide/qdec/area_LH_AA --surf fsaverage lh --label /mnt/hgfs/share/suicide/fsaverage/label/lh.aparc.label --C /mnt/hgfs/share/suicide/qdec/area_LH_AA/contrasts/lh-Avg-Intercept-area.mat --C /mnt/hgfs/share/suicide/qdec/area_LH_AA/contrasts/lh-Diff-NS-S-Intercept-area.mat --C /mnt/hgfs/share/suicide/qdec/area_LH_AA/contrasts/lh-Diff-M-F-Intercept-area.mat --C /mnt/hgfs/share/suicide/qdec/area_LH_AA/contrasts/lh-X-diagnosis-sex-Intercept-area.mat sysname Linux hostname localhost.localdomain machine x86_64 user sgk FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /mnt/hgfs/share/suicide/qdec/area_LH_AA/y.mgh logyflag 0 usedti 0 FSGD /mnt/hgfs/share/suicide/qdec/area_LH_AA/qdec.fsgd labelmask /mnt/hgfs/share/suicide/fsaverage/label/lh.aparc.label maskinv 0 glmdir /mnt/hgfs/share/suicide/qdec/area_LH_AA IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /mnt/hgfs/share/suicide/qdec/area_LH_AA Loading y from /mnt/hgfs/share/suicide/qdec/area_LH_AA/y.mgh INFO: gd2mtx_method is dods Saving design matrix to /mnt/hgfs/share/suicide/qdec/area_LH_AA/Xg.dat Normalized matrix condition is 2083.95 Matrix condition is 1e+08 Found 148151 points in label. Pruning voxels by thr: 0.000000 Found 148025 voxels in mask Saving mask to /mnt/hgfs/share/suicide/qdec/area_LH_AA/mask.mgh Reshaping mriglm->mask... search space = 73649.347769 DOF = 26 Starting fit and test Fit completed in 0.0923167 minutes Computing spatial AR1 on surface Residual: ar1mn=0.997068, ar1std=0.001719, gstd=7.468720, fwhm=17.587492 Writing results lh-Avg-Intercept-area maxvox sig=3.56625 F=17.7049 at index 25775 0 0 seed=1506511386 lh-Diff-NS-S-Intercept-area maxvox sig=3.42944 F=16.7076 at index 10791 0 0 seed=1506511386 lh-Diff-M-F-Intercept-area maxvox sig=-3.12066 F=14.5487 at index 103184 0 0 seed=1506511386 lh-X-diagnosis-sex-Intercept-area maxvox sig=3.53629 F=17.4843 at index 158713 0 0 seed=1506511386 mri_glmfit done ninputs = 4 Checking inputs nframestot = 4 Allocing output Done allocing nframes = 4 Writing to /mnt/hgfs/share/suicide/qdec/area_LH_AA/contrasts.sig.mgh reading colortable from annotation file... colortable with 36 entries read (originally /autofs/space/terrier_001/users/nicks/freesurfer/average/colortable_desikan_killiany.txt) reading colortable from annotation file... colortable with 36 entries read (originally /autofs/space/terrier_001/users/nicks/freesurfer/average/colortable_desikan_killiany.txt) gdfReadHeader: reading /mnt/hgfs/share/suicide/qdec/area_LH_AA/y.fsgd INFO: ignoring tag Creator INFO: ignoring tag SUBJECTS_DIR INFO: ignoring tag SynthSeed INFO: NOT demeaning continuous variables Continuous Variable Means (all subjects) 0 age 41.5263 12.8201 1 area 163656 15524.3 Class Means of each Continuous Variable 1 diagnosisNS-sexM 36.2000 176040.6000 2 diagnosisS-sexM 40.6250 170897.1250 3 diagnosisNS-sexF 46.5556 151977.4444 4 diagnosisS-sexF 42.9091 156686.6364 MatrixMultiply: m1 is null! No such file or directory [sgk@localhost ~]$

2 2

Can someone tell me why my subcorticals look like this?
by Jeff Crawford 02 Oct '17

02 Oct '17

Hi list! I was wondering what could cause this type of subcortical effect during FS preprocessing? I am referring to the white blank space where half the thalamus, part of the putamen, and all of the globus pallidus are not visable and Freesurfer just sort of "assumes" that they exist with its segmentation. My theory is that the dimensions of the Nii file I am using are being stretched and it is causing this "assumption" of white space. My dimensions on this Nii files are X = 160 , Y = 240, and Z = 256. This would lead me to believe that the FSaverage brain that my nii is being warped to match has a greater x axis than my nii. Is this correct? If so, is there any fsaverage options that has smaller dimensions? I assume this is an issue for analysis of subcortical measures, but should this effect cortical regions at all? Hope you all are well and my message finds someone promptly, Jeffrey Crawford

2 3

bad midline cut
by Gabor Perlaki 01 Oct '17

01 Oct '17

Dear all, Here is my question again. Somehow, it didn't appear on the Freesurfer forum. I'm resending it without the attachments. > I've a subject, where the midline cut is bad (see the attached pictures), > therefore the segmented left lateral ventricle also includes considerable > number of voxels from the right one. > Is it possible to correct it somehow? >

2 15

cortical marker
by N Saf 30 Sep '17

30 Sep '17

Dear FreeSurfer experts, I have 2 questions: 1- I try to measure average convexity for temporal lobe by FreeSurfer, I can see the values for each vertex when I overlay the e.g. lh.sulc in lh.inflated, but I wonder if I can find these values somewhere else like in stats files? 2- Is FreeSurfer compute metric distortion for a specific ROI (e.g. ROIs segmented in parcellation ) according following equation? and if that's the case where can I find these values of metric distortion? Metric distortion = k ∗ area of a triangle on registered sphere/ area of triangle on original gray/white interface surface Where k = total surface area of original gray/white interface surface/ total surface area of individual sphere could you help me to solve the problem, I am looking forward to receive your answer. Thank you for your help. Best Regards, Nazanin

2 3

sim-sign abs/ cluster summary results confirmation
by miracle ozzoude 29 Sep '17

29 Sep '17

Hello, I ran cortical thickness analyses using the following contrast and fsgd. After that, I corrected for multiple comparison using monte carlo 5000 with cluster forming threshold of 2 and unsigned/2-tailed t-test (abs). I wanted to confirm my interpretation of the results. 1) Since my contrast is (-1 1 0 0 representing control>stroke), it should be red/yellow and if blue, it should be control<stroke. However, I see clusters with colors different from the aformentioned (see attached screenshot). How should I interpret them? 2) I want to view the uncorrected p-values, which of the columns in the summary text file should I concentrate, and should the p < or > 0.05? 3) Looking at the CWP column in the summary text file, can I conclude that the obtained regions/clusters are significant? This is because non of the cluster-wise p-values are below 0.05 (--cwp 0.05). 4) Which of the columns represent the sign of my contrast i.e. where Control>Stroke and Control<Stroke because i used abs? Max? FSGD file GroupDescriptorFile Title StrokevsControl Class1 Stroke Class2 Control Variable bmi Input 6001 Stroke 20 Input 8001 Control 30 Contrast: -1 1 0 0 (Control>Stroke removing the effects of bmi) Thank you, Paul

2 5

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 September 2017