Dear Doug,
I am trying to run "mris_preproc" as you indicated, but it fails with the following error:
MRISread($SUBJECTS_DIR/fsaverage/surf/lh.fsaverage_sym.sphere.reg): could not open file.
Why is target file searched in the "fsaverage" folder? I was expecting it is searched in the "fsaverage_sym" folder?
Thank you in advance Kind regards Paola
Paola Valsasina, MSc Research Fellow Neuroimaging Research Unit (Director Prof. M. Filippi) Institute of Experimental Neurology, Division of Neuroscience, Scientific Institute and University "Vita-Salute" San Raffaele Via Olgettina 60, 20132 Milan, Italy -------------------------------------------------------------------------- IL MIO 5XMILLE VA AL SAN RAFFAELE DI MILANO PERCHE' QUI LA RICERCA DIVENTA CURA. CF 07636600962 SE NON QUI, DOVE? Info: 5xmille@hsr.it - http://www.5xmille.org/
Disclaimer added by CodeTwo Exchange Rules 2007 http://www.codetwo.com -----Messaggio originale----- Da: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] Inviato: martedì 29 aprile 2014 18.25 A: Valsasina Paola; Freesurfer@nmr.mgh.harvard.edu Oggetto: Re: R: R: R: [Freesurfer] Filipping issue
You can't use qdec for this, you will have to use mris_preproc, mri_surf2surf, mri_glmfit, and mri_glmfit-sim. The first step is the tricky one in that you will have to run mris_preproc twice then merge the two data sets.
For the subjects that you want the left hemisphere, run mris_preproc as "normal"
mris_preproc --hemi lh --meas thickness --surfreg fsaverage_sym.sphere.reg --s subject1 --s subject2 --o subjects.lh.sm00.mgh
For the right hemi subjects run
mris_preproc --hemi lh --meas thickness --surfreg fsaverage_sym.sphere.reg --s subject3/xhemi --s subject4/xhemi --o subjects.rh.sm00.mgh
Note above that you still specify --hemi lh but that the subject name changes to subject/xhemi
Now concatenate the two files together
mri_concat subjects.lh.sm00.mgh subjects.rh.sm00.mgh --o subjects.sm00.mgh
Then proceed with smoothing, etc. When you do the glm, make sure that your FSGD file has the proper order of subjects (above it would be subject1, subject2, 3, 4)
doug
On 04/29/2014 12:07 PM, Paola Valsasina wrote:
Dear Doug,
I registered with surfreg the hemispheres I would like to analyze. How should I proceed now for the final statistical vertex-wise analysis? Should I smooth the surfaces and perform between-group comparison with
qdec?
Thank you in advance Paola
Paola Valsasina, MSc Research Fellow Neuroimaging Research Unit (Director Prof. M. Filippi) Institute of Experimental Neurology, Division of Neuroscience, Scientific Institute and University "Vita-Salute" San Raffaele Via Olgettina 60, 20132 Milan, Italy
IL MIO 5XMILLE VA AL SAN RAFFAELE DI MILANO PERCHE' QUI LA RICERCA DIVENTA CURA. CF 07636600962 SE NON QUI, DOVE? Info: 5xmille@hsr.it - http://www.5xmille.org/
Disclaimer added by CodeTwo Exchange Rules 2007 http://www.codetwo.com
-----Messaggio originale----- Da: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] Inviato: martedì 22 aprile 2014 18.20 A: Valsasina Paola; Freesurfer@nmr.mgh.harvard.edu Oggetto: Re: R: R: [Freesurfer] Filipping issue
You can register just the hemisphere you want to fsaverage_sym. But if you have a subject whose right hemi you want, you would run the surfreg with --rh, but later you would specify that you want the lh of
subject/xhemi
doug
On 04/22/2014 12:04 PM, Paola Valsasina wrote:
Sorry, I didn't note I was replying to you only.. Anyway, thank you for the reply! So I will apply the command I mentioned
to
you in my previous e-mail to create flipped RH and LH surfaces for the subjects with the right hemisphere affected. Should I register to the fsaverage_sym atlas also the subjects I don't
want
to flip (i.e., those with the left hemisphere affected), to pool them together in the statistic with the flipped subjects? Thank you again Paola
Paola Valsasina, MSc Research Fellow Neuroimaging Research Unit (Director Prof. M. Filippi) Institute of Experimental Neurology, Division of Neuroscience, Scientific Institute and University "Vita-Salute" San Raffaele Via Olgettina 60, 20132 Milan, Italy
--------------------------------------------------------------------------
IL MIO 5XMILLE VA AL SAN RAFFAELE DI MILANO PERCHE' QUI LA RICERCA DIVENTA CURA. CF 07636600962 SE NON QUI, DOVE? Info: 5xmille@hsr.it - http://www.5xmille.org/
Disclaimer added by CodeTwo Exchange Rules 2007 http://www.codetwo.com
-----Messaggio originale----- Da: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] Inviato: martedì 22 aprile 2014 17.56 A: Valsasina Paola; Freesurfer@nmr.mgh.harvard.edu Oggetto: Re: R: [Freesurfer] Filipping issue
Hi Paola, please remember to post to the list and not to us personally. thanks! I would still use the xhemi tools for this. If you have N subjects, you will get 2*N surfaces aligned to the fsaverage_sym atlas. You can then just pick the ones you want to analyze further doug
On 04/22/2014 08:34 AM, Paola Valsasina wrote:
Dear Doug,
We have followed your suggestion and we performed the recon-all analysis
for
all study subjects (till autorecon3) on non-flipped images.
I am now reading the instructions on the wiki about the xhemi tool, but
I
am
not 100% sure they apply to our case.
The wiki instructions describe a comparison between LH and RH within subjects (i.e., an analysis of cortical asymmetry), whereas we would
like
simply to L-R flip the final surfaces of patients having a RH damage, so that all patients have the damage in the LH. Then, we would like to compare them with control subjects.
For this purpose, is it sufficient to register the patients lesioned in
the
RH to the symmetric template with the --xhemi option?
(i.e. surfreg --s subject_name --t fsaverage_sym --xhemi --rh --lh)
Then, how do we perform final statistical analysis?
Thank you in advance for any suggestion! Kind regards Paola
Paola Valsasina, MSc Research Fellow Neuroimaging Research Unit (Director Prof. M. Filippi) Institute of Experimental Neurology, Division of Neuroscience, Scientific Institute and University "Vita-Salute" San Raffaele Via Olgettina 60, 20132 Milan, Italy
IL MIO 5XMILLE VA AL SAN RAFFAELE DI MILANO PERCHE' QUI LA RICERCA DIVENTA CURA. CF 07636600962 SE NON QUI, DOVE? Info: 5xmille@hsr.it - http://www.5xmille.org/
Disclaimer added by CodeTwo Exchange Rules 2007 http://www.codetwo.com
-----Messaggio originale----- Da: freesurfer-bounces@nmr.mgh.harvard.edu [mailto:freesurfer-bounces@nmr.mgh.harvard.edu] Per conto di Douglas N
Greve
Inviato: giovedì 13 febbraio 2014 21.50 A: freesurfer@nmr.mgh.harvard.edu Oggetto: Re: [Freesurfer] Filipping issue
Definitely #2. You should use the xhemi tools http://surfer.nmr.mgh.harvard.edu/fswiki/Xhemi First run them through recon-all as normal, then follow the steps on the wiki. Contact the list again to get instructions for the final analysis.
doug
On 02/13/2014 08:23 AM, Paola Valsasina wrote:
Dear List,
we are starting the analysis of cortical thickness for a group of patients who have only one side affected. The majority of patients have the left hemisphere affected, but some patients have the right hemisphere affected. In order to do a group analysis, we would like to flip the images of some patients, so that the affected side is left for all subjects.
Which is the best way to do this:
1)flip the 001.mgz volume and perform the entire recon-all pipeline with flipped images;
or
2)perform the recon-all pipeline with the non-flipped volumes and flip only the final surfaces for statistical purposes?
Thank you for any suggestion
Kind regards
Paola
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer