Hi,
I would like to do a Group Analysis for my FA data.
For all my subjects (controls + patients) I have all the fa-masked.ANAT+CVS-to-avg35.mgz (the resampled registered image to CVS space).
I am not sure if my steps for the Group Analysis are the right one:
1) Concatenate FA from individuals into one file
mri_concat */fa-masked.ANAT+CVS-to-avg35.mgz --o GroupAnalysis-fa-masked.ANAT+CVS-to-avg35.mgz
2) Make sure the order agrees with the fsgd:
ls -l */fa-masked.ANAT+CVS-to-avg35.mgz
3) Create average of the input images for visualization: mri_average */fa-masked.ANAT+CVS-to-avg35.mgz GroupAnalysis.Average-fa-masked.ANAT+CVS-to-avg35.mgz
4) mri_glmfit --y GroupAnalysis-fa-masked.ANAT+CVS-to-avg35.mgz \ --fsgd Group_Analysis.fsgd dods --C contrast.mtx --glmdir Group_Analysis.glmdir
To view uncorrected results:
tkmedit fsaverage orig.mgz -aparc+aseg -overlay /media/1Tra/Antonella_2011/DTI_FS/dtrecon/Group_Analysis.glmdir/contrast/sig.mgh -fminmax 2 3
where -fminmax 2 3 : Voxel-wise threshold of 2 (p<.01), color saturation at 3 (p<.001)
To correct for multiple comparisons run:
5) mri_glmfit-sim \ --glmdir Group_Analysis.glmdir \ --grf 3 neg \ --cwpvalthresh .0166 \ --overwrite
Is 3 (p<.001) neg as well as .0166 (.05/3) the best voxel-wise as well as cluster wise values to be tried?
To view corrected results: tkmedit fsaverage orig.mgz \ -ov /media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.cluster.mgh \ -seg /media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.ocn.anat.mgh \ /media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.ocn.lut \ -fminmax 1.3 5
with only positive values
* p-values are p-values for cluster * Set threshold to 1.3 (p<.05)
Are this steps right?
When I run the registration over the cvs space which file is the equivalent file of the fa-tal.nii.reg from the TAL space?
MANY THANKS!
Antonella
Antonella Kis wrote:
Hi,
I would like to do a Group Analysis for my FA data.
For all my subjects (controls + patients) I have all the fa-masked.ANAT+CVS-to-avg35.mgz (the resampled registered image to CVS space).
I am not sure if my steps for the Group Analysis are the right one:
These look correct
- Concatenate FA from individuals into one file
mri_concat */fa-masked.ANAT+CVS-to-avg35.mgz --o GroupAnalysis-fa-masked.ANAT+CVS-to-avg35.mgz
- Make sure the order agrees with the fsgd:
ls -l */fa-masked.ANAT+CVS-to-avg35.mgz
- Create average of the input images for visualization: mri_average */fa-masked.ANAT+CVS-to-avg35.mgz
GroupAnalysis.Average-fa-masked.ANAT+CVS-to-avg35.mgz
- mri_glmfit --y GroupAnalysis-fa-masked.ANAT+CVS-to-avg35.mgz \ --fsgd Group_Analysis.fsgd dods --C contrast.mtx --glmdir
Group_Analysis.glmdir
To view uncorrected results:
tkmedit fsaverage orig.mgz -aparc+aseg -overlay /media/1Tra/Antonella_2011/DTI_FS/dtrecon/Group_Analysis.glmdir/contrast/sig.mgh -fminmax 2 3
where -fminmax 2 3 : Voxel-wise threshold of 2 (p<.01), color saturation at 3 (p<.001)
To correct for multiple comparisons run:
- mri_glmfit-sim \
--glmdir Group_Analysis.glmdir \ --grf 3 neg \ --cwpvalthresh .0166 \ --overwrite
Is 3 (p<.001) neg as well as .0166 (.05/3) the best voxel-wise as well as cluster wise values to be tried?
There is no answer to the voxel-wise threshold (a criticism of the method). For the CWP, you can use .05 instead of .0166 (that value is for the case where you are doing separate surface and subcortical analyses). Given that FA is very non-gaussian, a lot of people prefer to use permutation (which requires you to run a simulation -- also see my note from last week about a bug in the permutation code (not a problem if you only have one contrast))
To view corrected results: tkmedit fsaverage orig.mgz \ -ov /media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.cluster.mgh \ -seg /media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.ocn.anat.mgh \
/media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.ocn.lut \ -fminmax 1.3 5
with only positive values
* p-values are p-values for cluster
- Set threshold to 1.3 (p<.05)
Are this steps right?
When I run the registration over the cvs space which file is the equivalent file of the fa-tal.nii.reg from the TAL space?
I don't think there is one file that exists. doug
MANY THANKS!
Antonella
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Doug,
Many thanks for your help. I have one more questions please. You were telling me that:
"Given that FA is very non-gaussian, a lot of people prefer to use permutation". What this means? Do I need to run something different than:
mri_glmfit-sim \ --glmdir Group_Analysis.glmdir \ --grf 3 neg \ --cwpvalthresh .05 \ --overwrite
Thank you. Antonella
________________________________ From: Douglas N Greve greve@nmr.mgh.harvard.edu To: Antonella Kis atorok9@yahoo.com Cc: "freesurfer@nmr.mgh.harvard.edu" freesurfer@nmr.mgh.harvard.edu Sent: Monday, January 23, 2012 4:50 PM Subject: Re: [Freesurfer] mri_glm with grf over CVS space
Antonella Kis wrote:
Hi,
I would like to do a Group Analysis for my FA data.
For all my subjects (controls + patients) I have all the fa-masked.ANAT+CVS-to-avg35.mgz (the resampled registered image to CVS space).
I am not sure if my steps for the Group Analysis are the right one:
These look correct
- Concatenate FA from individuals into one file
mri_concat */fa-masked.ANAT+CVS-to-avg35.mgz --o GroupAnalysis-fa-masked.ANAT+CVS-to-avg35.mgz
- Make sure the order agrees with the fsgd:
ls -l */fa-masked.ANAT+CVS-to-avg35.mgz
- Create average of the input images for visualization:
mri_average */fa-masked.ANAT+CVS-to-avg35.mgz GroupAnalysis.Average-fa-masked.ANAT+CVS-to-avg35.mgz
- mri_glmfit --y GroupAnalysis-fa-masked.ANAT+CVS-to-avg35.mgz \
--fsgd Group_Analysis.fsgd dods --C contrast.mtx --glmdir Group_Analysis.glmdir
To view uncorrected results:
tkmedit fsaverage orig.mgz -aparc+aseg -overlay /media/1Tra/Antonella_2011/DTI_FS/dtrecon/Group_Analysis.glmdir/contrast/sig.mgh -fminmax 2 3
where -fminmax 2 3 : Voxel-wise threshold of 2 (p<.01), color saturation at 3 (p<.001)
To correct for multiple comparisons run:
- mri_glmfit-sim \
--glmdir Group_Analysis.glmdir \ --grf 3 neg \ --cwpvalthresh .0166 \ --overwrite
Is 3 (p<.001) neg as well as .0166 (.05/3) the best voxel-wise as well as cluster wise values to be tried?
There is no answer to the voxel-wise threshold (a criticism of the method). For the CWP, you can use .05 instead of .0166 (that value is for the case where you are doing separate surface and subcortical analyses). Given that FA is very non-gaussian, a lot of people prefer to use permutation (which requires you to run a simulation -- also see my note from last week about a bug in the permutation code (not a problem if you only have one contrast))
To view corrected results: tkmedit fsaverage orig.mgz \ -ov /media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.cluster.mgh \ -seg /media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.ocn.anat.mgh \ /media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.ocn.lut \ -fminmax 1.3 5
with only positive values
* p-values are p-values for cluster * Set threshold to 1.3 (p<.05)
Are this steps right?
When I run the registration over the cvs space which file is the equivalent file of the fa-tal.nii.reg from the TAL space?
I don't think there is one file that exists. doug
MANY THANKS!
Antonella
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 greve@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
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.
You can run the command below (it might take several hours). This will get a p-value based on permutation. This will probably be larger (more conservative) than the one you get from GRF. See the Nichols tutorial in HBM 2001 for more info on permutation.
doug
mri_glmfit-sim \ --glmdir Group_Analysis.glmdir \ --sim perm 10000 3 permcsd\ --cwpvalthresh .05
Antonella Kis wrote:
Hi Doug,
Many thanks for your help. I have one more questions please. You were telling me that:
"Given that FA is very non-gaussian, a lot of people prefer to use permutation". What this means? Do I need to run something different than:
mri_glmfit-sim \ --glmdir Group_Analysis.glmdir \ --grf 3 neg \ --cwpvalthresh .05 \ --overwrite
Thank you. Antonella
*From:* Douglas N Greve greve@nmr.mgh.harvard.edu *To:* Antonella Kis atorok9@yahoo.com *Cc:* "freesurfer@nmr.mgh.harvard.edu" freesurfer@nmr.mgh.harvard.edu *Sent:* Monday, January 23, 2012 4:50 PM *Subject:* Re: [Freesurfer] mri_glm with grf over CVS space
Antonella Kis wrote:
Hi,
I would like to do a Group Analysis for my FA data.
For all my subjects (controls + patients) I have all the
fa-masked.ANAT+CVS-to-avg35.mgz (the resampled registered image to CVS space).
I am not sure if my steps for the Group Analysis are the right one:
These look correct
- Concatenate FA from individuals into one file
mri_concat */fa-masked.ANAT+CVS-to-avg35.mgz --o
GroupAnalysis-fa-masked.ANAT+CVS-to-avg35.mgz
- Make sure the order agrees with the fsgd:
ls -l */fa-masked.ANAT+CVS-to-avg35.mgz
- Create average of the input images for visualization: mri_average */fa-masked.ANAT+CVS-to-avg35.mgz
GroupAnalysis.Average-fa-masked.ANAT+CVS-to-avg35.mgz
- mri_glmfit --y GroupAnalysis-fa-masked.ANAT+CVS-to-avg35.mgz \ --fsgd Group_Analysis.fsgd dods --C contrast.mtx --glmdir
Group_Analysis.glmdir
To view uncorrected results:
tkmedit fsaverage orig.mgz -aparc+aseg -overlay
/media/1Tra/Antonella_2011/DTI_FS/dtrecon/Group_Analysis.glmdir/contrast/sig.mgh -fminmax 2 3
where -fminmax 2 3 : Voxel-wise threshold of 2 (p<.01), color
saturation at 3 (p<.001)
To correct for multiple comparisons run:
- mri_glmfit-sim \
--glmdir Group_Analysis.glmdir \ --grf 3 neg \ --cwpvalthresh .0166 \ --overwrite
Is 3 (p<.001) neg as well as .0166 (.05/3) the best voxel-wise as
well as cluster wise values to be tried? There is no answer to the voxel-wise threshold (a criticism of the method). For the CWP, you can use .05 instead of .0166 (that value is for the case where you are doing separate surface and subcortical analyses). Given that FA is very non-gaussian, a lot of people prefer to use permutation (which requires you to run a simulation -- also see my note from last week about a bug in the permutation code (not a problem if you only have one contrast))
To view corrected results: tkmedit fsaverage orig.mgz \ -ov
/media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.cluster.mgh \
-seg
/media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.ocn.anat.mgh \
/media/1Tra/Antonella_2011/DTI_FS/dtrecon/group_analysis.glmdir/contrast/grf.th3.pos.sig.ocn.lut \
-fminmax 1.3 5
with only positive values
- p-values are p-values for cluster
- Set threshold to 1.3 (p<.05)
Are this steps right?
When I run the registration over the cvs space which file is the
equivalent file of the fa-tal.nii.reg from the TAL space? I don't think there is one file that exists. doug
MANY THANKS!
Antonella
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu mailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu mailto:greve@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
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.
freesurfer@nmr.mgh.harvard.edu
-
Antonella Kis -
Douglas N Greve