I want to ask a related question: We study group differences between patients and controls, and correlations of symptoms and thickness in patients only. We would like to display (and extract) the common regions between two analyses in ONE figure. Which template can we use? We created two sets of ROIs in Qdec: one set is defined using an averge from all subjects; the other set is defined using an average of patients only. The ROI.labels from one set can not be loaded correctly on the average tamplate of the other set. The problem remains even if we convert both sets of ROI.labels to individual spaces. can we use fsaverage for both analyses? thanks, Xin Wang
________________________________ From: Doug Greve [mailto:gr...@nmr.mgh.harvard.edu] Sent: Mon 1/5/2009 2:23 PM To: Siddharth Srivastava Cc: Freesurfer Mailing List Subject: Re: [Freesurfer] questions about region-wise analysis data Siddharth Srivastava wrote: Hi all, I am trying to understand the --surf flag in the mri_glmfit command, and the 2 parameters that follow it. The tutorial mentions --surf average lh . 1) what should "average" contain? should it point to the average of the population, of fsaverage, or is it just a flag? This is the name of the average subject as created by make_average_subject. If you did not make an average subject, then use fsaverage. 2) lh: is this just the specification of the hemisphere to be processed or should it point to a valid set of files. Just the hemisphere If i am doing a group analysis of a measure (say, thickness), and i have smoothed thickness maps registered to a population specific template in a subdir, how should i specify --surf and subsequebt parameters? The input is specified with with --y. The --surf just tells glmfit that it is a surface so that it can compute the FWHM appropriately. Further, i also get an error saying --C command not found (for the contrast vector). why is this happenning? Can't say without the cmd line, but I'm guessing you have a rouge space in your script best regards, sid. On Sun, Dec 28, 2008 at 12:24 PM, Nick Schmansky <ni...@nmr.mgh.harvard.edu> wrote: Sid, A group analysis can be performed from the command-line using mri_glmfit directly. See this tutorial: http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis You can use a different average subject, other than fsaverage, although it doesnt improve an analysis to use a subject from your group. Also, you can create a symlink from $FREESURFER_HOME/subjects/fsaverage to your SUBJECTS_DIRS, to spare having to copy the data. Nick On Sun, 2008-12-28 at 12:01 -0800, Siddharth Srivastava wrote: > Hi Nick, > Thanks, It was my mistake. It works exactly as you say. > I still have to look at qdec documentation, but could i ask if it is > possible to perform the group analysis in a batch mode, without > the GUI, i.e ? > > regarding the movement of data, the first time i did it, i got a > screen full > of memory addresses and corresponding routines, followed by a crash ( > exit from wish, back to shell). I am not able to replicate it, seems > to work > now. > > Also, is it possible to register everything to a different surface, > other > that fsaverage.. for example to one of the subjects? Can this be > located > outside the relative paths reachable via $SUBJECTS_DIR ? In fact, can > fsaverage be located outside? For the processing i did, i had to copy > the fsaverage as one of the subjects, so that it can be located > within > the current context of $SUBJECTS_DIR . > > Thanks again, for the timely help. > sid. > > > On Sun, Dec 28, 2008 at 9:17 AM, Nick Schmansky > <ni...@nmr.mgh.harvard.edu> wrote: > Sid, > > The .mgh files, in the case of the commands you ran, store > surface data > corresponding to the 'fsaverage' subject. So you would first > load the > inflated surface for fsaverage (tksurfer fsaverage lh > inflated), then > you would use the Load Overlay menu option to load the .mgh > file(s) you > created. You probably will have to adjust the color > thresholds (View- > >Configure->Overlay). For the lgi data, you can use: > > recon-all -s subj -qcache -measure pial_lgi > > which will sample the subjects lgi data to the fsaverage > surface, at > which point you could average that (although typically a > statistical > analysis is performed on the set of subjects sampled to > fsaverage, see > out group analysis slides and tutorial pages). > > You should be able to move subject data around to other > directories > without any problems. Does tksurfer really halt after that > 'not in > scripts dir' message? That is a common message and not an > error. Can > you open tksurfer for the sample bert subject? > > Nick > > > > > On Sat, 2008-12-27 at 19:14 -0800, Siddharth Srivastava wrote: > > Hi Nick, > > I finally got a chance to use the sequence of > command > > that you had provided below. In continuation of this thread: > > > > a) The output of using the command mentioned in reply 2) > below > > is an MGH file (and not faces/vertices combination)? How is > it > > possible to get a smoothed surface > > that i can load as an overlay in tksurfer? I would > ultimately > > like to average all the smoothed lgi mappings as one single > > surface per hemisphere. Is this possible in freesurfer? In > any case, > > although > > my pial_lgi surfaces look ok, i do not see anything in > > the .fwhmxx.fsaverage.mgh files, when > > i load them using tkmedit. what can be going wrong? I used > fwhm = > > 15. > > > > b) Are there known issues with migrating the data set > processed > > partiallty > > at one mount point, to some different location? Due to lack > of space, > > i had to do some recon-all processing in one directory, and > then > > copied the tree to a different > > location for subsequent steps. I get tksurfer errors > "surfer: not in > > "scripts" dir ==> using cwd for session root", > > followed by, what i think, a read attempt of the > corresponding T1.mgz > > file. No window is drawn. > > I googled with the error string, and it seems there was a > similar > > error under discussion, which > > had to do with moving files and absolute paths. I was not > able to > > understand the fix, though, > > and hence i wanted to confirm with the list what exactly has > to be > > done in this situation. > > > > Thanks, > > sid. > > > > On Fri, Dec 5, 2008 at 10:42 AM, Nick Schmansky > > <ni...@nmr.mgh.harvard.edu> wrote: > > Sid, > > > > 1) when recon-all is finished on a subject, there > are a number > > of stats > > found in the <subj>/stats directory. for the > additional ones > > you want, > > you can use mris_anatomical_stats. see > 'mris_anatomical_stats > > --help'. > > ex. mris_anatomical_stats -a aparc -t lgi subjid lh > > see also 'aparcstats2table' to group a collection of > subjects. > > > > 2) you can use the -qcache option of recon-all > (normally used > > to > > generate files used by the qdec app) to handle the > resampling > > of data to > > a common surface (fsaverage). example: > > > > recon-all -s subj -qcache -measure lgi -fwhm 0 -hemi > lh > > > > will create subj/surf/lh.lgi.fwhm0.fsaverage.mgh > > > > where fwhm is the amount of smoothing (0mm in this > case means > > no > > smoothing). > > then, once you have this file for every subject, to > get a > > mean, you can > > use: > > mri_concat --i subj1/surf/lh.lgi.fwhm0.fsaverage.mgh > \ > > subj2/surf/lh.lgi.fwhm0.fsaverage.mgh \ > > subj3/surf/lh.lgi.fwhm0.fsaverage.mgh ... \ > > --mean \ > > --o mygroup.lh.lgi.fwhm0.mean.fsaverage.mgh > > > > 3) there is no way to change the colors in the > > tkmedit/tksurfer colormap > > (other than the color options available). people > change the > > thresholds > > to get the nice pictures. > > > > Nick > > > > > > > > > > > > On Thu, 2008-12-04 at 19:50 -0800, Siddharth > Srivastava wrote: > > > Hi everyone, > > > while the machine is busy > crunching > > through our > > > data set, i would > > > like to prepare myself from the next step of the > analysis, > > which is a > > > region-wise > > > analysis of the computed metrics by the freesurfer > pipeline. > > Being a > > > complete > > > novice in the use of this software, i would like > to know how > > do i go > > > about > > > doing the following: > > > > > > 1) With the parcellations already available in > freesurfer, i > > would > > > like to > > > extract metrics (curvature, lgi, thickness) > etc. for > > each label, > > > and > > > process them outside freesurfer (matlab, > since i am > > more used to > > > it). Is it possible to do this within > freesurfer or > > without, i.e > > > read these > > > values on the pial/white matter/inflated > surfaces, ans > > also > > > extract ones > > > corresponding to a particular label? > > > 2) How do i go about generating a group average of > the > > various maps > > > that freesurfer generates > (curvature/thickness/lgi) ? > > As i > > > understand, > > > the analysis is done entirely in the subject > space. Can > > i apply a > > > transformation > > > to the generated surfaces for averaging? Or > does it > > require a re- > > > processing > > > after registering/reslicing all the data to a > common > > space? > > > 3) I am still struggling with a proper colormap, > which can > > bring out > > > the structure in out data. I see very nice > renderings > > in > > > various > > > publications, of data generated with > freesurfer. Is it > > possible > > > to tweak > > > the colormap to get such a rendering? > > > > > > Thanking you all in anticipation, > > > sid. > > > > > _______________________________________________ > > > Freesurfer mailing list > > > Freesurfer@nmr.mgh.harvard.edu > > > > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > > > > _______________________________________________ > > Freesurfer mailing list > > Freesurfer@nmr.mgh.harvard.edu > > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ________________________________ _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer -- Douglas N. Greve, Ph.D. MGH-NMR Center gr...@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422 In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
