Freesurfer December 2011

freesurfer@nmr.mgh.harvard.edu

98 participants
157 discussions

freeview label
by Klein, Holger 25 Dec '11

25 Dec '11

Dear Freesurfers, Merry Christmas to all of you!!! I have a little problem with freeview: As attachment there are three screenshots, which show two labels of the 2009 atlas (Primary auditory cortex und primary visual cortex). I used brain.mgz as "input" volume and chose the labels in the ROI tab in freeview - but the screenshots look like as if there is a vast part of the labeled cortex not colored? What could be the reason for this? Thx a lot & best wishes for 2012, Holger

2 1

Talairach transform issue; V 5.0
by Andy Mayer 24 Dec '11

24 Dec '11

Hello Freesurfer Experts- We recently reran several subjects from scratch through the stable Freesurfer 5.0 pipeline who had previously been processed with the stable 4.5 pipeline. We used tkregister2 for QA of the 5.0 data and noticed that the talairach transform was off (misaligned) for the first 5 subjects we examined. All subjects passed the auto Talairach check, and their 4.5 data was not misaligned. Applying the talairach.xfm matrix from the 4.5 pipeline corrected the problem. Looking at the recon-all.cmd files from the different versions, it looks like the only difference for the tailairach_avi step was that the 4.5 version called for nu.mgz whereas the 5.0 version called for orig.mgz. I reran a couple of 5.0 subjects using talairach_avi --i nu.mgz –xfm transforms/talairach.auto.xfm Doing this fixed the transform and subsequent alignment problems (copied above to talairach.xfm and re-ran tkregister2). I have read on the list that the talairach.xfm is not used in the majority of data streams (although table indicates otherwise). Our plans were to use the data for both cortical thickness analyses as well as MEG analyses (source mapping). I would appreciate it if someone could tell me: 1) The best way to rerun the current data that has been run through 5.0. We have already run about 50 of 100 subjects through 5.0. 2) The best way to fix our recon-all program for 5.0 so that the correct transform is generated for new subjects. 3) In addition, we also observed the following error in the log file and could not find any information on what it may mean (all checked subjects had error) *********************************************** Computing MAP estimate using 3481 samples... *********************************************** IFLAG= -1 LINE SEARCH FAILED. SEE DOCUMENTATION OF ROUTINE MCSRCH ERROR RETURN OF LINE SEARCH: INFO= 3 POSSIBLE CAUSES: FUNCTION OR GRADIENT ARE INCORRECT OR INCORRECT TOLERANCESoutof QuasiNewtonEMA: ************************************************** EM alignment process ... Computing final MAP estimate using 382743 samples. ************************************************** IFLAG= -1 LINE SEARCH FAILED. SEE DOCUMENTATION OF ROUTINE MCSRCH ERROR RETURN OF LINE SEARCH: INFO= 3 POSSIBLE CAUSES: FUNCTION OR GRADIENT ARE INCORRECT OR INCORRECT TOLERANCESpass 2 through quasi-newton minimization... Thanks for your time! Andrew

2 2

Noise filtering before recon-all
by Courtney Haswell 23 Dec '11

23 Dec '11

We have some T1 images that are kind of noisy that were acquired several years ago. We applied a noise-reduction filter (FSL's SUSAN) to the T1 images before beginning the recon-all segmentation process. Is there any reason not to do this type of smoothing before recon-all, or is there any option in the recon-all pipeline to filter noise? Is it recommended to apply a filter before recon-all in this case? Thanks, Courtney

2 1

GLM question
by Govindarajan, Koushik Athreya 23 Dec '11

23 Dec '11

Hello group, I am a new user of Freesurfer and just learning to process data through the pipeline. Sorry if this is an amateur question. I was just trying to look at cortical thickness differences between a patient group and a control group. So, my fsgd file would just list the input files and the corresponding diagnosis. Now, if I were to design a contrast matrix to look at cortical thickness differences between the 2 groups, I understand that I would have only 2 regressors . So, my contrast matrix would just be [-1 1]? I tried doing that and I got a matrix read error for the contrast file. Thanks in advance for your help. Regards Koushik

2 2

about labels
by LAOUCHEDI MAKHLOUF 23 Dec '11

23 Dec '11

hi eveyone i am quite new in freesurfer, is it possible to generate aparc.a2009s+aseg.mgz from aparc.a2005s+aseg.mgz without re running recon-all thanks

2 1

Surface-based cortical volume calculation
by _andreia_＠sapo.pt 23 Dec '11

23 Dec '11

Hello, Recalling these emails: "The methods are somewhat different. For the value in the aseg.stats table, the method is to compute the total volume inside the pial surface and subtract the total volume inside the white surface. For mris_anatomical_stats, the method is to compute the thickness times area of each vertex. This method will probably underestimate the total volume because it uses the area of the white surface when it should use the area of the surface in the middle between the white and pial surfaces. I've added this to the list of known issues on our release page. doug Alexopoulos, Dimitrios wrote: > I have generated surfaces using the the centos4 build (version 5.0) > and want to confirm that my surface-based GM and WM volumes are correct. > For the surface-based GM calculation I originally used > 'mris_anatomical_stats -l lh.cortex.label subjectID hemi' > (run from within the 'label' subdirectory) and for WM i used > 'mris_wm_volume subjectID hemi' (run from within > the 'surf' subdirectory). > When I add the calculated left/right cortical volumes, I get a total > that is different from what is output in > the 'aseg.stats' file, which in version 5.0 is noted to contain > total surface-based GM volume > (Cortex, CortexVol: Total cortical gray matter volume (based on > surface-stream). > What are the correct GM and WM surface-based volumes? > Thanks. Jim" I question came up to me: The surfaces in the hippocampus/amygdala are inaccurate and should be ignored. However, in version 5.0 the cortical volume is surface-based, thus it takes into account the surfaces in the hippocampus/amygdala, is this correct? If so, it is expected that an error is introduced in the surface-based calculation of cortical volume. Has anyone checked the influence of this error? Or FS compensates for the inaccuracy of the surface estimation of these regions somehow? Thanks! Andreia

3 6

FsFast Functional Connectivity with subject-specific surface seeds
by konglq＠nmr.mgh.harvard.edu 22 Dec '11

22 Dec '11

Hi Freesurfers, We are trying to do a resting-state functional connectivity analysis using freesurfer. Instead of using the aparc+aseg segmentations as seeds, we would like to use a subject-specific surface labels that we have defined apriori. In the fcseed-config step, we then don't have a "segid" as in the anatomical seeds case. What should we specify as the -segid and -seg (segmentation)? We also looked into the ROI-based options using "-roi" but this takes seeds from output of "funcroi-config" and requires that we specify a contrast and an analysis. How should we just specify a surface label without generating a new analysis and contrast? Any help or suggestions are appreciated. Lingqiang

2 3

mri_matrix_multiply to move functional images to common space
by Andrew Dumas 22 Dec '11

22 Dec '11

Hi, I'm trying to move a bunch of partial FOV functional images (occipital cortex only with oblique coronal slices) into common space. For each subject, I have a registration file (register.dat) to register functional images to anatomicals, and I'd like to incorporate the talairach.xfm registration so that I have a matrix that registers the functional images to the mni305 volume. I've been using the following, but registrations aren't coming out properly: mri_matrix_multiply -im register.dat -iim <subject>/mri/transforms/talairach.xfm -om reg_to_common.dat I'm checking them using tkregister2, but the results don't look great. Is there a better/easier way to do this? Thanks! Andrew

2 3

Re: [Freesurfer] mri_matrix_multiply to move functional images to common space
by Douglas N Greve 22 Dec '11

22 Dec '11

Yea, you need to do it in the two steps you mention. You can do that from one command line with something like bbregister --s subject --mov epi.nii --init-{fsl,header} --bold --int aascout.nii --o register.dat The --int is an "intermediate" volume. doug Andrew Dumas wrote: > That did the trick, thank you! As for using bbregister, I haven't had > any luck using bbregister for the partial volume EPIs... they're always > off. We have a strange protocol where only the functionals get the > 32-channel coil, so right before the EPIs get done I have an AAscout > image. I've been using bbregister to align the FS MPRAGE scans with the > AAscout (first frame), then using that registration on the EPIs. It > seems to come out a LOT better than using bbregister on EPIs directly. > > Andrew > > > On Tue, 2011-12-20 at 14:58 -0500, Douglas N Greve wrote: > >> Hi Andrew, use mri_vol2vol with the --tal --talres 2 or 2mm option (or >> --talres 1 for 1mm isotropic) option. This will put it into an MNI305 >> space. So something like >> >> mri_vol2vol --reg register.dat --mov yourvolume.nii --tal --talres 2 --o >> yourvolume.mni305.nii >> >> If you did not create the register.dat with bbregister, I would strongly >> suggest it. It works really well on the partial volume files. If you do, >> you should specify your current register.dat as the init registration. >> >> doug >> >> Andrew Dumas wrote: >> >>> Hi, >>> >>> I'm trying to move a bunch of partial FOV functional images (occipital >>> cortex only with oblique coronal slices) into common space. For each >>> subject, I have a registration file (register.dat) to register >>> functional images to anatomicals, and I'd like to incorporate the >>> talairach.xfm registration so that I have a matrix that registers the >>> functional images to the mni305 volume. I've been using the following, >>> but registrations aren't coming out properly: >>> >>> mri_matrix_multiply -im register.dat -iim >>> <subject>/mri/transforms/talairach.xfm -om reg_to_common.dat >>> >>> I'm checking them using tkregister2, but the results don't look great. >>> Is there a better/easier way to do this? Thanks! >>> >>> Andrew >>> >>> _______________________________________________ >>> 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: www.nmr.mgh.harvard.edu/facility/filedrop/index.html

1 0

labels broblem
by LAOUCHEDI MAKHLOUF 22 Dec '11

22 Dec '11

hi everyone i run recon-all -all on some subjects using the 5.1 version of freesurfer and instead of getting aparc.a2009s+aseg.mgz i got aparc.a2005s+aseg.mgz. the components of the cingulate cortex are quite different in both files a2005s and a2009s : in a2009s i used in my work: 11108 ctx_lh_G_and_S_cingul-Mid-Post 11109 ctx_lh_G_cingul-Post-dorsal 11110 ctx_lh_G_cingul-Post-ventral and in a2005s the whole labals of the cingulate cortex are : 1103 ctx-lh-G_cingulate-Isthmus 1104 ctx-lh-G_cingulate-Main_part 1200 ctx-lh-G_cingulate-caudal_ACC 1201 ctx-lh-G_cingulate-rostral_ACC 1205 ctx-lh-S_cingulate-caudal_ACC 1206 ctx-lh-S_cingulate-rostral_ACC 1207 ctx-lh-S_cingulate-posterior 1002 ctx-lh-caudalanteriorcingulate 1147 ctx-lh-S_cingulate-Main_part_and_Intracingulate 1148 ctx-lh-S_cingulate-Marginalis_part is there any way to get the a2009s cingulate components labels using the a2005s labels in the aparc.a2005s+aseg.mgz. thanks

2 1

← Newer
1
2
3
4
5
6
7
...
16
Older →

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 December 2011