Re: [Freesurfer] Saving results in volume space

[Martin Juneja]

Thank you so much Douglas. That's really helpful.

On Mon, Sep 18, 2017 at 2:41 PM, Douglas N Greve 
wrote:

>
>
> On 09/18/2017 05:35 PM, Martin Juneja wrote:
> > Dear Douglas,
> >
> > Thanks a lot for your reply.
> > Could you please clarify why I have to add 0.3 because I am using a
> > positive sign? How did
> > Do you mean if I want to use threshold of 3, then adding 0.3 to this
> > so threshold of 3.3 should be used ?
> The sig map is unsigned. If it were signed, then the p-values would be
> half, and the sig values would be 0.3 greater (sig=-log10(p)). When you
> run glmfit-sim with a signed value (pos, neg), then effectively adjusts
> the p-values to be signed. However, when you view the sig maps, no such
> adjustment takes place.
> >
> > If I use --sim-sign abs, do you think then there is no need to add a
> > number to threshold? Whats the best approach to run this command?
> Correct, if you use abs, you do not need to adjust because abs is
> two-tailed. The answer to your second question is totally up to you.
> Usually, people use a signed  test if they had an a priori hypothesis
> about the direction of the effect.
> >
> > I am really confused here and would really appreciate any
> > clarification on this.
> >
> > Thanks a lot.
> >
> > On Mon, Sep 18, 2017 at 2:20 PM, Douglas N Greve
> > > wrote:
> >
> >
> >
> > On 09/15/2017 02:13 PM, Martin Juneja wrote:
> > > Hi experts,
> > >
> > > I have three questions. I would really appreciate any help.
> > >
> > > (1). I ran following commands to estimate relationships between LGI
> > > and Behav data:
> > >
> > >  mris_preproc --fsgd Behav.fsgd --target fsaverage --hemi lh --meas
> > > pial_lgi --out lh.Behav_LGI.mgh
> > >
> > > mri_surf2surf --hemi lh --s fsaverage --sval lh.Behav_LGI.mgh
> --fwhm
> > > 10 --cortex --tval lh.Behav_LGI.10.mgh
> > >
> > > mri_glmfit --y lh.Behav_LGI.10.mgh --fsgd Behav.fsgd dods --C
> > > Corr-Behav-cor1.mtx --surf fsaverage lh --cortex --glmdir
> > > lh.Behav_LGI_P.glmdir --eres-save
> > >
> > > mri_glmfit-sim_nonortho --glmdir lh.Behav_LGI_P.glmdir --sim
> > perm 2000
> > > 1.3 permcsd --sim-sign pos --cwpvalthresh .05 --perm-resid
> > --overwrite
> > >
> > > Finally, cluster summary gives me one clusters with peak at the
> > > fusiform gyrus as following:
> > > ClusterNo  Max   VtxMax   Size(mm^2)  MNIX   MNIY  MNIZCWP
> > >  CWPLowCWPHi   NVtxsWghtVtx   Annot
> > >1   -4.045   10153  32101.81-36.5 -39.5  -15.1  0.00050
> > >  0.0  0.00100  64731  -106698.82  fusiform
> > >
> > > As you can see number of voxels is 64731, so I when I viewed the
> > > results in FreeView, its showing me a big cluster, almost all
> > over the
> > > brain (please see attached screen shot).
> > >
> > > Could you please tell me why I am getting these sort of results,
> and
> > > how can I fix this? If I change cluster forming threshold from
> > 1.3 to
> > > 3, would that work? I wanted to make sure before I re-run the
> > analysis
> > > because mri_glmfit-sim_nonortho command takes about a day to
> finish.
> > I don't know whether it would work, but it would make it better.
> > You can
> > also just load the sig.mgh file and change the voxel-wise thresholds
> > until you get reasonably sized clusters, then run using that
> > threshold.
> > If you do this, set the glmfti-sim threshold to the value your
> > selected
> > during viewing but add 0.3 (needed because you are using a
> > positive sign).
> > >
> > > (2). How can I save output cluster e.g. fusiform gyrus in above
> > > example in NIFTI MNI volume space so that I can use this cluster
> for
> > > functional connectivity analysis?
> > You're probably better of doing your fmri analysis in surface space
> > because mapping back and forth is a little tricky. But if you want
> > to do
> > it, you can do
> > something like
> > mri_label2vol --annot /path/to/annot/in/glmfit/dir --temp
> > $FSLDIR/data/standard/MNI152_T1_2mm.nii.gz --reg
> > $FREESURFER_HOME/average/mni152.register.dat --fill-ribbon --subject
> > fsaverage --hemi lh --o output.segmentation.in
> > .mni152space.nii.gz
> >
> > The output segmenation will have all the clusters from the annot
> file.
> > Remember, this mapping is an approximation
> >
> > >
> > > (3). If I want to use fusiform gyrus from standard FreeSurfer
> > > parcellation rather than from my results, how can I save that in
> > > standard MNI volume space?
> > same as above
> > >
> > > Thank you so much !
> > >
> > >
> > > ___
> > > Freesurfer mailing list
> > > 

Re: [Freesurfer] Saving results in volume space

[Douglas N Greve]

On 09/18/2017 05:35 PM, Martin Juneja wrote:
> Dear Douglas,
>
> Thanks a lot for your reply.
> Could you please clarify why I have to add 0.3 because I am using a 
> positive sign? How did
> Do you mean if I want to use threshold of 3, then adding 0.3 to this 
> so threshold of 3.3 should be used ?
The sig map is unsigned. If it were signed, then the p-values would be 
half, and the sig values would be 0.3 greater (sig=-log10(p)). When you 
run glmfit-sim with a signed value (pos, neg), then effectively adjusts 
the p-values to be signed. However, when you view the sig maps, no such 
adjustment takes place.
>
> If I use --sim-sign abs, do you think then there is no need to add a 
> number to threshold? Whats the best approach to run this command?
Correct, if you use abs, you do not need to adjust because abs is 
two-tailed. The answer to your second question is totally up to you. 
Usually, people use a signed  test if they had an a priori hypothesis 
about the direction of the effect.
>
> I am really confused here and would really appreciate any 
> clarification on this.
>
> Thanks a lot.
>
> On Mon, Sep 18, 2017 at 2:20 PM, Douglas N Greve 
> > wrote:
>
>
>
> On 09/15/2017 02:13 PM, Martin Juneja wrote:
> > Hi experts,
> >
> > I have three questions. I would really appreciate any help.
> >
> > (1). I ran following commands to estimate relationships between LGI
> > and Behav data:
> >
> >  mris_preproc --fsgd Behav.fsgd --target fsaverage --hemi lh --meas
> > pial_lgi --out lh.Behav_LGI.mgh
> >
> > mri_surf2surf --hemi lh --s fsaverage --sval lh.Behav_LGI.mgh --fwhm
> > 10 --cortex --tval lh.Behav_LGI.10.mgh
> >
> > mri_glmfit --y lh.Behav_LGI.10.mgh --fsgd Behav.fsgd dods --C
> > Corr-Behav-cor1.mtx --surf fsaverage lh --cortex --glmdir
> > lh.Behav_LGI_P.glmdir --eres-save
> >
> > mri_glmfit-sim_nonortho --glmdir lh.Behav_LGI_P.glmdir --sim
> perm 2000
> > 1.3 permcsd --sim-sign pos --cwpvalthresh .05 --perm-resid
> --overwrite
> >
> > Finally, cluster summary gives me one clusters with peak at the
> > fusiform gyrus as following:
> > ClusterNo  Max   VtxMax   Size(mm^2)  MNIX   MNIY  MNIZCWP
> >  CWPLowCWPHi   NVtxsWghtVtx   Annot
> >1   -4.045   10153  32101.81-36.5 -39.5  -15.1  0.00050
> >  0.0  0.00100  64731  -106698.82  fusiform
> >
> > As you can see number of voxels is 64731, so I when I viewed the
> > results in FreeView, its showing me a big cluster, almost all
> over the
> > brain (please see attached screen shot).
> >
> > Could you please tell me why I am getting these sort of results, and
> > how can I fix this? If I change cluster forming threshold from
> 1.3 to
> > 3, would that work? I wanted to make sure before I re-run the
> analysis
> > because mri_glmfit-sim_nonortho command takes about a day to finish.
> I don't know whether it would work, but it would make it better.
> You can
> also just load the sig.mgh file and change the voxel-wise thresholds
> until you get reasonably sized clusters, then run using that
> threshold.
> If you do this, set the glmfti-sim threshold to the value your
> selected
> during viewing but add 0.3 (needed because you are using a
> positive sign).
> >
> > (2). How can I save output cluster e.g. fusiform gyrus in above
> > example in NIFTI MNI volume space so that I can use this cluster for
> > functional connectivity analysis?
> You're probably better of doing your fmri analysis in surface space
> because mapping back and forth is a little tricky. But if you want
> to do
> it, you can do
> something like
> mri_label2vol --annot /path/to/annot/in/glmfit/dir --temp
> $FSLDIR/data/standard/MNI152_T1_2mm.nii.gz --reg
> $FREESURFER_HOME/average/mni152.register.dat --fill-ribbon --subject
> fsaverage --hemi lh --o output.segmentation.in
> .mni152space.nii.gz
>
> The output segmenation will have all the clusters from the annot file.
> Remember, this mapping is an approximation
>
> >
> > (3). If I want to use fusiform gyrus from standard FreeSurfer
> > parcellation rather than from my results, how can I save that in
> > standard MNI volume space?
> same as above
> >
> > Thank you so much !
> >
> >
> > ___
> > 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: 

Re: [Freesurfer] Saving results in volume space

[Martin Juneja]

Dear Douglas,

Thanks a lot for your reply.
Could you please clarify why I have to add 0.3 because I am using a
positive sign? How did
Do you mean if I want to use threshold of 3, then adding 0.3 to this so
threshold of 3.3 should be used ?

If I use --sim-sign abs, do you think then there is no need to add a number
to threshold? Whats the best approach to run this command?

I am really confused here and would really appreciate any clarification on
this.

Thanks a lot.

On Mon, Sep 18, 2017 at 2:20 PM, Douglas N Greve 
wrote:

>
>
> On 09/15/2017 02:13 PM, Martin Juneja wrote:
> > Hi experts,
> >
> > I have three questions. I would really appreciate any help.
> >
> > (1). I ran following commands to estimate relationships between LGI
> > and Behav data:
> >
> >  mris_preproc --fsgd Behav.fsgd --target fsaverage --hemi lh --meas
> > pial_lgi --out lh.Behav_LGI.mgh
> >
> > mri_surf2surf --hemi lh --s fsaverage --sval lh.Behav_LGI.mgh --fwhm
> > 10 --cortex --tval lh.Behav_LGI.10.mgh
> >
> > mri_glmfit --y lh.Behav_LGI.10.mgh --fsgd Behav.fsgd dods --C
> > Corr-Behav-cor1.mtx --surf fsaverage lh --cortex --glmdir
> > lh.Behav_LGI_P.glmdir --eres-save
> >
> > mri_glmfit-sim_nonortho --glmdir lh.Behav_LGI_P.glmdir --sim perm 2000
> > 1.3 permcsd --sim-sign pos --cwpvalthresh .05 --perm-resid --overwrite
> >
> > Finally, cluster summary gives me one clusters with peak at the
> > fusiform gyrus as following:
> > ClusterNo  Max   VtxMax   Size(mm^2)  MNIX   MNIY   MNIZCWP
> >  CWPLowCWPHi   NVtxsWghtVtx   Annot
> >1   -4.045   10153  32101.81-36.5  -39.5  -15.1  0.00050
> >  0.0  0.00100  64731  -106698.82  fusiform
> >
> > As you can see number of voxels is 64731, so I when I viewed the
> > results in FreeView, its showing me a big cluster, almost all over the
> > brain (please see attached screen shot).
> >
> > Could you please tell me why I am getting these sort of results, and
> > how can I fix this? If I change cluster forming threshold from 1.3 to
> > 3, would that work? I wanted to make sure before I re-run the analysis
> > because mri_glmfit-sim_nonortho command takes about a day to finish.
> I don't know whether it would work, but it would make it better. You can
> also just load the sig.mgh file and change the voxel-wise thresholds
> until you get reasonably sized clusters, then run using that threshold.
> If you do this, set the glmfti-sim threshold to the value your selected
> during viewing but add 0.3 (needed because you are using a positive sign).
> >
> > (2). How can I save output cluster e.g. fusiform gyrus in above
> > example in NIFTI MNI volume space so that I can use this cluster for
> > functional connectivity analysis?
> You're probably better of doing your fmri analysis in surface space
> because mapping back and forth is a little tricky. But if you want to do
> it, you can do
> something like
> mri_label2vol --annot /path/to/annot/in/glmfit/dir --temp
> $FSLDIR/data/standard/MNI152_T1_2mm.nii.gz --reg
> $FREESURFER_HOME/average/mni152.register.dat --fill-ribbon --subject
> fsaverage --hemi lh --o output.segmentation.in.mni152space.nii.gz
>
> The output segmenation will have all the clusters from the annot file.
> Remember, this mapping is an approximation
>
> >
> > (3). If I want to use fusiform gyrus from standard FreeSurfer
> > parcellation rather than from my results, how can I save that in
> > standard MNI volume space?
> same as above
> >
> > Thank you so much !
> >
> >
> > ___
> > 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
>
> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
> FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2
> www.nmr.mgh.harvard.edu/facility/filedrop/index.html
> Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
>
> ___
> Freesurfer mailing list
> Freesurfer@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.
>
>
___
Freesurfer mailing list
Freesurfer@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 

Re: [Freesurfer] Saving results in volume space

[Douglas N Greve]

On 09/15/2017 02:13 PM, Martin Juneja wrote:
> Hi experts,
>
> I have three questions. I would really appreciate any help.
>
> (1). I ran following commands to estimate relationships between LGI 
> and Behav data:
>
>  mris_preproc --fsgd Behav.fsgd --target fsaverage --hemi lh --meas 
> pial_lgi --out lh.Behav_LGI.mgh
>
> mri_surf2surf --hemi lh --s fsaverage --sval lh.Behav_LGI.mgh --fwhm 
> 10 --cortex --tval lh.Behav_LGI.10.mgh
>
> mri_glmfit --y lh.Behav_LGI.10.mgh --fsgd Behav.fsgd dods --C 
> Corr-Behav-cor1.mtx --surf fsaverage lh --cortex --glmdir 
> lh.Behav_LGI_P.glmdir --eres-save
>
> mri_glmfit-sim_nonortho --glmdir lh.Behav_LGI_P.glmdir --sim perm 2000 
> 1.3 permcsd --sim-sign pos --cwpvalthresh .05 --perm-resid --overwrite
>
> Finally, cluster summary gives me one clusters with peak at the 
> fusiform gyrus as following:
> ClusterNo  Max   VtxMax   Size(mm^2)  MNIX   MNIY   MNIZCWP   
>  CWPLowCWPHi   NVtxsWghtVtx   Annot
>1   -4.045   10153  32101.81-36.5  -39.5  -15.1  0.00050 
>  0.0  0.00100  64731  -106698.82  fusiform
>
> As you can see number of voxels is 64731, so I when I viewed the 
> results in FreeView, its showing me a big cluster, almost all over the 
> brain (please see attached screen shot).
>
> Could you please tell me why I am getting these sort of results, and 
> how can I fix this? If I change cluster forming threshold from 1.3 to 
> 3, would that work? I wanted to make sure before I re-run the analysis 
> because mri_glmfit-sim_nonortho command takes about a day to finish.
I don't know whether it would work, but it would make it better. You can 
also just load the sig.mgh file and change the voxel-wise thresholds 
until you get reasonably sized clusters, then run using that threshold. 
If you do this, set the glmfti-sim threshold to the value your selected 
during viewing but add 0.3 (needed because you are using a positive sign).
>
> (2). How can I save output cluster e.g. fusiform gyrus in above 
> example in NIFTI MNI volume space so that I can use this cluster for 
> functional connectivity analysis?
You're probably better of doing your fmri analysis in surface space 
because mapping back and forth is a little tricky. But if you want to do 
it, you can do
something like
mri_label2vol --annot /path/to/annot/in/glmfit/dir --temp 
$FSLDIR/data/standard/MNI152_T1_2mm.nii.gz --reg 
$FREESURFER_HOME/average/mni152.register.dat --fill-ribbon --subject 
fsaverage --hemi lh --o output.segmentation.in.mni152space.nii.gz

The output segmenation will have all the clusters from the annot file. 
Remember, this mapping is an approximation

>
> (3). If I want to use fusiform gyrus from standard FreeSurfer 
> parcellation rather than from my results, how can I save that in 
> standard MNI volume space?
same as above
>
> Thank you so much !
>
>
> ___
> 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

Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/

___
Freesurfer mailing list
Freesurfer@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.