Freesurfer

freesurfer@nmr.mgh.harvard.edu

30132 discussions

Postdoctoral Fellowship: Multimodal Neuroimaging -MGH/Harvard Medical School
by sheraz＠nmr.mgh.harvard.edu 26 Jul '11

26 Jul '11

Postdoctoral Fellowship: Multimodal Neuroimaging -MGH/Harvard Medical School Job Description A postdoctoral position is available with the TRANSCEND Research Program (www.transcendresearch.org) at the Martinos Center for Biomedical Imaging in Charlestown, MA (www.martinos.org) which is affiliated with the Massachusetts General Hospital, Harvard Medical School and MIT. We are seeking a candidate with a strong basis in magnetic resonance imaging. Emphasis will be on MRI (DTI, spectroscopy, morphometry, ASL as well as resting state fMRI), and on co-registering MEG with MRI. This position will involve analysis of existing multimodal imaging data and collection of new data. The emphasis of the postdoctoral fellowship will be analysis of existing datasets with secondary activity in piloting data for new studies. It will involve working closely with a multidisciplinary team and with children, and will also involve some research oriented analysis of data collected for clinical purposes. After initial phase-in, ample opportunity will also be provided to the candidate to self-explore and lead research. Datasets to be analyzed include: MRI (including DTI and 1H-spectroscopy) and MEG data on 6-12 and teenage matched autism spectrum and control subjects with phenotyping data MRI data ( (morphometry, DTI, spectroscopy) plus laboratory and phenotyping data) on 70 children with autism plus epilepsy and/or mitochondrial dysfunction, along with one or more overnight EEGs on each patient data from children ages 2-10 with and without autism. Overall objectives: To perform multimodal analyses of research and clinical research data, to develop new approaches for performing these analyses, and to design pipelines for data analysis. To write papers and grants which will be high priorities all along the way and will be actively supported by senior faculty. To take advantage of the world class faculty and facilities of the Martinos Center for Biomedical Imaging to perform the above activities to their maximal potential. The programs emphasis is on pathophysiologically grounded brain research and application of advanced imaging acquisition and analysis techniques to neurological and sensory aspects of autism spectrum disorders. Requirements: Candidates must have PhD in neuroscience, physics, biomedical engineering, electrical engineering, computer science or other related fields. Prior experience in MRI analysis is required. Experience with EEG will be an added advantage. Salary will be consistent with Massachusetts General Hospital, Harvard Medical School policies for Postdoctoral trainees and will range between $45,000 to $55,000 depending upon qualifications and experience. Compensation also includes full staff benefits, including health insurance, and vacation time. Contact: Interested applicants may send a CV and statement of interest addressing background and specific pertinence of the candidates interest to Dr. Martha R. Herbert at mherbert1(a)partners.org and cc transcend(a)partners.org.

1 0

ROI analysis with label in both hemispheres -still looking for an answer
by Katie Bettencourt 26 Jul '11

26 Jul '11

I'm still looking for an answer on this. If the answer is you can't do it, I'm ok with that, I just would like to know! Thanks! Hi all, > > I was wondering if anyone had any idea if it was possible to combine > together ROIs from the separate hemispheres into one ROI (for example LH V1 > and RH V1) and then perform a functional ROI analysis on it. I know how to > combine ROIs (through labels_union) and how to do a functional ROI analysis > (using func2roi-sess and roisummary-sess), but I don't know if it is > possible to do it with a label that runs across both hemispheres. Thoughts? > > Thanks, > Katie >

3 5

mri_surfcluster coordinates
by Chris Hyatt 26 Jul '11

26 Jul '11

Dear Freesurfers, I know that Qdec outputs coordinates in a "Freesurfer specific space", rather than in Talairach space, per this old thread https://mail.nmr.mgh.harvard.edu/pipermail/freesurfer/2008-November/008878.… However, I found that mri_surfcluster (Freesurfer v5.0) in its summary output also outputs coordinates in this same "Freesurfer specific space", even though the columns indicate TalX, TalY and TalZ. I was going to use this summary info to list the location of the statistical maxima of each cluster in a cortical thickness p-map. Is this a bug in mri_surfcluster? Any suggestions on how to get the coordinate data? Here is the summary info for the left hemisphere (note the positive TalX values, when all X values should be negative): # Cluster Growing Summary (mri_surfcluster) # $Id: mri_surfcluster.c,v 1.51.2.1 2011/03/28 15:32:36 greve Exp $ # $Id: mrisurf.c,v 1.693.2.2 2011/04/27 19:21:05 nicks Exp $ # CreationTime 2011/07/23-17:08:39-GMT # cmdline mri_surfcluster --in ./lh.CvsA.glmdir/group.diff/sig.mgh --subject fsaverage --hemi lh --surf inflated --thmin 1.301 --sum lh.CvsA.thickness.summary --oannot ./lh.CvsA.thickness.annot # cwd /prod4/other/freesurfer/glm_cth/age_of_onset # sysname Linux # hostname everest # machine x86_64 # FixVertexAreaFlag 1 # FixSurfClusterArea 1 # # Input ./lh.CvsA.glmdir/group.diff/sig.mgh # Frame Number 0 # srcsubj fsaverage # hemi lh # surface inflated # SUBJECTS_DIR /prod4/other/freesurfer/subjects # SearchSpace_mm2 62822.2 # SearchSpace_vtx 163842 # Bonferroni 0 # Minimum Threshold 1.301 # Maximum Threshold infinity # Threshold Sign abs # AdjustThreshWhenOneTail 1 # Area Threshold 0 mm^2 # Overall max 2.53286 at vertex 81529 # Overall min -2.67513 at vertex 24566 # NClusters 36 # Total Cortical Surface Area 62822.2 (mm^2) # FixMNI = 1 # # ClusterNo Max VtxMax Size(mm^2) TalX TalY TalZ NVtxs 1 -2.675 24566 287.98 7.6 95.9 0.5 533 2 2.533 81529 83.49 -7.4 -28.4 50.2 199 3 2.530 6795 386.88 -15.4 -86.6 -9.4 677 4 -2.496 129236 79.10 -27.0 27.1 22.6 166 5 -2.483 138360 227.18 -5.7 84.6 2.6 343 6 2.481 18891 211.45 3.4 -96.6 -26.2 373 7 2.460 33958 277.19 37.6 5.5 15.4 354

2 1

Qdec does not change the color bar though the color scare is reversed on the brain.
by Knut J Bjuland 26 Jul '11

26 Jul '11

Hi I have a problem when using qdec. When I check the option reverse values there color on the brain switch color, but the color bar stays the same. Is there anything that might be wrongs in my files? Is there anything that I might edit in Qdec configuration file? I do not know what kind of log that are required and I would be happy to send them. Cheers Knut J Bjuland figure attachment http://vedlegg.uib.no/?id=73435c1da0c1f0c1ba792601c4fa1199 http://vedlegg.uib.no/?id=20e06d09dcc450ff4abcd85e1e5ab1c5

2 3

Converting from volume label RAS to Talairach
by Jed Singer 26 Jul '11

26 Jul '11

I'm at my wits' end with what seems like it should be an easy question. I have a lot of volume labels that have been saved for a lot of brains. I want the Talairach coordinates corresponding to those points. It seems like mri_label2label should be able to do this, by putting in 'talairach' as the target subject, but this returns an error that the subject 'talairach' isn't found: mri_label2label --srcsubject m00045 --srclabel label/all.label --trgsubject talairach --trglabel all_talairach.label --regmethod volume It seems like I should be able to do it manually by loading the talairach.xfm generated by recon-all and multiplying that matrix by the RAS points in the labels; this yields results that are pretty close to those given in tkmedit, but not that close. Usually within a few, sometimes off by more. Probably my inability to do this reflects some misunderstanding or lack of information, but I've been poring over documentation, forum, and wiki for hours now and feel like I've hit a wall. Any suggestions? Thanks so much, Jed

3 3

Re: [Freesurfer] Cortical segmentation error
by Carolina Valencia 26 Jul '11

26 Jul '11

Hi Bruce, thanks for your explanation. I was thinking the the segmentation region is used to calculate cortical thickness, by the way, is it used for calculate something? such as the GM volumes? I want to know this to be sure if in some point I have to fix this. About the heterotopias, is a good output to have someones classified as WM, other as GM and sometimes are outlined by the yellow line and sometimes are not? Best regards, Carolina 2011/7/19 Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> > Hi Carolina, > > it looks mostly to me like the surfaces are in the right place (the yellow > and red lines). Don't look at the colored overlay - that is the aseg voxel > segmentation and not what we typically use for cortical morphometry. Only > look at the yellow and red lines. > > > cheers > Bruce > On Tue, 19 Jul 2011, Carolina Valencia wrote: > > tiff I guess, I took the pictures from FS >> >> 2011/7/19 Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> >> what format are the files in the archive in? They don't have an >> extension >> >> On Tue, 19 Jul 2011, Carolina Valencia wrote: >> >> >> I hope this attachement works.I also noticed that some heterotopias >> are classified as WM, other as GM and sometimes are outlined by the yellow >> line and >> sometimes are not. >> >> Best regards, >> >> Carolina >> >> 2011/7/19 Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> >> Hi Carolina, >> >> can you gzip it or use compression? Usually people can send a >> few images to the list. >> >> cheers >> Bruce >> On Tue, 19 Jul 2011, Carolina Valencia wrote: >> >> Dear FS experts, >> I'm trying to analyse a case of cortical dysplasia, but after >> recon-all I see some errors in tkmedit that I don't know what to edit. I >> found some >> errors >> in cortical and >> subcortical segmentation, I don't find in the failure mode >> presentation a way to solve this. >> I saw the pial surface (yellow line) with acceptable accuracy >> but the GM (the red one) in some regions are misplaced (WM classified as GM >> and some >> GM not >> classified (empty >> color)) >> I tried to attached an image to illustrate the problem but I >> don't succeed, because the mailing list reject my email. How I can send you >> an image of >> ~700kb? >> >> Thanks a lot, >> >> Carolina >> >> >> >> >> 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<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. >> >> >> >> >> > >

2 3

Re: [Freesurfer] Longitudinal Analysis Error
by Martin Reuter 26 Jul '11

26 Jul '11

Hi Silvia, it is never a good idea to mix versions and we did not test any such combinations. Also because 5.1 is much better than 4.5 I would definitely recommend to rerun. However, I still believe that it should be possible to use the 5.1 post-processing scripts to e.g. create rate maps for each subject and qdec to analyze them even with 4.5 processed data. Best, Martin On Tue, 2011-07-26 at 10:45 +0200, Silvia Juanes wrote: > Hi Martin, > > We used the preprocessing pipeline from 4.5 version and then we used > the 5.1 version to perform the longitudinal analysis. Do you think > that we should repeat the preprocessing using 5.1 version? or do you > know any other solution to do the analysis with the 4.5 preprocessing > data? > > Best, > > Silvia > > 2011/7/25 Martin Reuter <mreuter(a)nmr.mgh.harvard.edu> > Hi Silvia, > > qdec does not support any longitudinal analysis (only cross > sectional). > So I assume you either compare groups cross sectionally (e.g. > at > baseline) or you constructed your own longitudinal measure > (e.g. percent > change, within each subject) and then do a cross sectional > analysis of > that statistic? > Both these things should work. I think your qdec table is > setup > incorrectly? > > Best, Martin > > > On Tue, 2011-07-19 at 12:45 +0200, Silvia Juanes wrote: > > Dear all, > > We are preprocessed our data to perform a longitudinal study > using > > longitudinal pipeline of Freesurfer 4.5. However, when we > try to do > > the group-analysis using qdec (Freesurfer 5.1), we obtain > the > > following error: > > > > > > ERROR: QdecGlmDesign::GenerateContrasts: factor 1 must have > 2 levels > > ninputs = 64 > > Checking inputs > > nframestot = 64 > > Allocing output > > Done allocing > > nframes = 64 > > Writing to /home/barbara/subjects/qdec/Untitled/y.mgh > > gdfReadHeader: > reading /home/barbara/subjects/qdec/Untitled/qdec.fsgd > > INFO: gd2mtx_method is dods > > Reading source > surface /home/barbara/subjects/fsaverage/surf/lh.white > > ERROR: no contrasts specified. > > Error in Analyze: command failed: mri_glmfit > > --y /home/barbara/subjects/qdec/Untitled/y.mgh > > --fsgd /home/barbara/subjects/qdec/Untitled/qdec.fsgd dods > > --glmdir /home/barbara/subjects/qdec/Untitled --surf > fsaverage lh > > > > > > > > Has anyone experienced a similar problem and is there any > way to correct it? > > > > > > Best, > > > > > > > > > _______________________________________________ > > Freesurfer mailing list > > Freesurfer(a)nmr.mgh.harvard.edu > > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > > 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] Error in qdec longitudinal 5.1
by Martin Reuter 26 Jul '11

26 Jul '11

Hi Natalia, /TemplateSub1/surf/lh.thckness.fwhm10.fsaverage.mgh there is an 'i' missing in thckness (but that might be just a typo). also this is a regular thickness file. You cannot analyze the thickness in the base. That is meaningless, you need to select one of the other files in qdec, e.g. the spc. (there is a selection field). Best, Martin On Tue, 2011-07-26 at 14:01 +0200, Natalia h wrote: > Hi Martin, > > > I rechecked the .Qdecrc file, I put it as suggested on the tutorials. > inside my qdec folder, I put a .Qdecrc file and contains the 4 lines. > But still qdec cannot recognize the new types of files, am not sure > why. > > > Thanks again > Natalia > > > On Tue, Jul 26, 2011 at 11:25 AM, Natalia h > <natalia.himalaya(a)gmail.com> wrote: > Hi Martin, > > Thanks for the reply. my .qdecrc file is correct. > What I tried to say is that, > the error appear when I click analyze button. (qdec 1.4) > Error in Analyize: Couln´t open > file /TemplateSub1/surf/lh.thckness.fwhm10.fsaverage.mgh > > > > p.s TemplateSub1 is the -base name > > > > While I looked back into this folder, Indeed there is no such > file > What i have are > lh.long.thickness-spc-fwhm10.fsaverage.mgh > lh.long.thickness-rate-fwhm10.fsaverage.mgh > lh.long.thickness-pc1-fwhm10.fsaverage.mgh > ....etc > > > > > > > The qdec 1.4 seems to be not able to see these files... > > > Regards > Natalia > > > On Mon, Jul 25, 2011 at 5:08 PM, Martin Reuter > <mreuter(a)nmr.mgh.harvard.edu> wrote: > > Hi Natalia, > > > > check your qdec/.QDECRC file, it needs to contain: > > > > MEASURE1 = long.thickness-avg > > MEASURE2 = long.thickness-rate > > MEASURE3 = long.thickness-pc1 > > MEASURE4 = long.thickness-spc > > > > probably you put a .spc instead of -spc > > > > Best, Martin > > > > On Fri, 2011-07-22 at 18:51 +0200, Natalia h wrote: > >> Dear list, > >> > >> I got an error after following the newer version of the > longitudinal > >> stream 5.1. > >> > >> long_mris_slopes worked without problems for my data which > was > >> longtudinally processed in 4.5. > >> > >> I also split the qdec.table to the cross for being able to > be loaded > >> in qdec GUI. > >> > >> Somehow I got the error that the > >> Templatesubxx/surf/xh.thickness.fwhm.xx.fsaverage.mgh. > >> > >> I went back to check inside the Templatesubxx/surf has only > the > >> xh.long.thickness-xxx.fwhmx.fsaverage.mgh > >> > >> Is there something I am missing? > >> > >> Thanks a lot for the help > >> _______________________________________________ > >> Freesurfer mailing list > >> Freesurfer(a)nmr.mgh.harvard.edu > >> > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > >> > >> > > > > > > > > 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

lesion question
by Rebecca McCourt 25 Jul '11

25 Jul '11

Hello everyone, I have several patients with large cortical lesions (haemorrhage). I want to have subcortical structure volumes reflect the lesion, i.e. show up as hypointensities in aseg.mgz (which they're not right now). However I also want to calculate cortical thickness, and the WM & GM surfaces around the lesion are way off. The lesion is much too big to fill manually in wm.mgz. Any suggestions? - Rebecca

1 0

Re: [Freesurfer] Dimension error - mri_segstats
by Jaichander S 25 Jul '11

25 Jul '11

Hi, For others who are looking for the solution to the problem I mentioned earlier - Running mri_vol2surf in step3 of the instructions given here ( http://surfer.nmr.mgh.harvard.edu/fswiki/VolumeRoiCorticalThickness) replacing '--reshape' with '--rf 6' did the trick for me. I tried this after going throught the help for mri_vol2surf command. Cheers, Jaichander From: Jaichander S <jaichander.s(a)gmail.com> > Subject: [Freesurfer] Dimension error - mri_segstats > To: freesurfer(a)nmr.mgh.harvard.edu > Message-ID: > <CABc7hshQM0Gszc0_uC8GEY85wVdVKq6rgXM7YyEdFwu8YkFgKA(a)mail.gmail.com > > > Content-Type: text/plain; charset="iso-8859-1" > > Hi freesurfers, > > I performed ROI cortical thickness measurements earlier on an older version > of FS as described on > http://surfer.nmr.mgh.harvard.edu/fswiki/VolumeRoiCorticalThickness and I > was able to get the required values. > > Now I have the latest version of FS. I draw an ROI mask using FSL software > and then use it to follow the instructions on the tutorial. The ROI and the > initial T1 image are in the same coordinate space (with different > dimensions) as is required by the procedure. After following all the steps > I > get > > ERROR: dimension mismatch between input volume and seg > input 27307 1 6 > seg 23406 1 7 > > when running mri_segstats. > > This has been reported by some other people already on the mailing list but > the suggestions do not work for me. Any light that you can shed to sidestep > this dimensions problem will be greatly appreciated. > > Is there any other workflow to define a volume ROI and then get the > cortical > thickness in the area. > > Thanks! > Jaichander >

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