Re: [Freesurfer] Labels and parcellations

[Douglas N Greve]

On 05/08/2013 05:24 PM, Tudor Popescu wrote:
> Sorry, one more iteration:
>
> 1) Command "mri_annotation2label --subject 01 --hemi lh --aparc.a2009s 
> --outdir $SUBJECTS_DIR/01/myLabelsDestrieux" doesn't recognise the 
> aparc.2009s flag (I tried a few other variations with no luck)!
The syntax is --annot aparc.2009s
If you run mri_annotation2label with --help it will tell you the syntax.
>
> 1') Assuming the above command works and I get .label files 
> corresponding to all Destrieux structures, what is the exact call to 
> mri_glmfit that will do the two contrasts I mentioned (group 
> comparison of CT, and comparison of CT-behaviour correlation)?
>
> 1'') How should the design matrix & contrast files (arguments for 
> calling mri_glmfit) be created, for these analyses?
Have you looked at our group analysis tutorials?
>
> 2) ..so when would I use a mask, in a ROI analysis, as opposed to a label?
Functionally, there is no difference. If you had a process that 
generated a mask, then you would just use that. You can convert any mask 
to a label with mri_cor2label. mri_glmfit will give you the same result.
>
> 3) I'm still not sure as to the answer to my question: regardless of 
> what space I do it in, does it pay off to create the label/mask that 
> I"ll use in the ROI analysis /manually /for each individual subject 
> (i.e. taking into account individual anatomy), as opposed to just 
> taking it from an atlas and using /the same /label/mask with /all 
> /subjects?
If you are using it as a mask for voxel-wise analysis, then you need a 
mask in the common space. You could get this mask by drawing the mask in 
the common space. Or you could draw the mask on each individual, then 
map the masks to the common space. But you would need to then merge all 
those masks together into a single binary mask. I don't know whether one 
would be better than the other, but the individual masks will surely be 
a lot more work.
>
> 4) Bingo - this creates an image as I wanted, but now the mouse cursor 
> no longer tells me the structure above which I hover the mouse (so a 
> legend to each color, much like in Fig 1 of the Desikan paper)
I don't understand what you mean here. The name of the structure should 
appear in the control window. It does not appear in the image window 
like in Fig 1.
doug
>
> I really appreciate your patience Doug, thanks :)
> Tudor
>
> On 8 May 2013 18:56, Douglas N Greve  > wrote:
>
>
> On 05/08/2013 01:28 PM, Tudor Popescu wrote:
>
>
> 2) ..and when would you use a surface mask versus a
> surface label?
>
> It is just a matter of convenience.
>
> Sorry, I meant I don't really understand what the difference
> is between a label and a mask, and in which circumstances each
> should be used. I tried finding an answer on the wiki but no
> luck..
>
> A label is a text file with a list of points in the label. A mask
> is binary file (eg, mgh or nii) with 1s where the mask is and 0s
> otherwise.
>
>
> 3) (it seems you skipped over this one:) )
>
> Sorry:) what are you going to do with the ROI?
>
> Use it mri_glmfit, in order to compare cortical thickness in
> that ROI between groups, and also (in a different design) to
> compare the correlations between thickness and a behavioural
> predictor
>
> If you're doing the final analysis in the group space, then just
> use the ROI in the group space.
>
>
> 4) This opens up a coronal slice rather than a 3D
> image as in
> the picture I linked to. Do I need to do anything else in
> tkmedit after it opens? (I tried pressing some
> buttons, and
> some menu options, with no luck)
>
> that was generated with tksurfer, something like tksurfer
> subject
> lh pial -aparc
>
> Using that command, I get a brain
>  that
> has only little of its surface coloured, unlike the image
> 
> 
> that I linked to, in which it seems each major structure is
> identified in its own colour, and these colours fill up the
> entire brain (there's no grey surface left)
>
> Change the annotation mode to be filled instead of outline mode.
> In the tksurfer control window, it is the 4th button from the
> right on the top row. It will say "Draw Filled Labels" when you
> put the mouse over it
>
>
>
>
> On 8 May 2013 16:54, Douglas N Greve
>  
>  >
>  

Re: [Freesurfer] Labels and parcellations

[Tudor Popescu]

Sorry, one more iteration:

1) Command "mri_annotation2label --subject 01 --hemi lh --aparc.a2009s
--outdir $SUBJECTS_DIR/01/myLabelsDestrieux" doesn't recognise the
aparc.2009s flag (I tried a few other variations with no luck)!

1') Assuming the above command works and I get .label files corresponding
to all Destrieux structures, what is the exact call to mri_glmfit that will
do the two contrasts I mentioned (group comparison of CT, and comparison of
CT-behaviour correlation)?

1'') How should the design matrix & contrast files (arguments for calling
mri_glmfit) be created, for these analyses?

2) ..so when would I use a mask, in a ROI analysis, as opposed to a label?

3) I'm still not sure as to the answer to my question: regardless of what
space I do it in, does it pay off to create the label/mask that I"ll use in
the ROI analysis *manually *for each individual subject (i.e. taking into
account individual anatomy), as opposed to just taking it from an atlas and
using *the same *label/mask with *all *subjects?

4) Bingo - this creates an image as I wanted, but now the mouse cursor no
longer tells me the structure above which I hover the mouse (so a legend to
each color, much like in Fig 1 of the Desikan paper)

I really appreciate your patience Doug, thanks :)
Tudor

On 8 May 2013 18:56, Douglas N Greve  wrote:

>
> On 05/08/2013 01:28 PM, Tudor Popescu wrote:
>
>>
>> 2) ..and when would you use a surface mask versus a surface label?
>>
>> It is just a matter of convenience.
>>
>> Sorry, I meant I don't really understand what the difference is between a
>> label and a mask, and in which circumstances each should be used. I tried
>> finding an answer on the wiki but no luck..
>>
> A label is a text file with a list of points in the label. A mask is
> binary file (eg, mgh or nii) with 1s where the mask is and 0s otherwise.
>
>
>> 3) (it seems you skipped over this one:) )
>>
>> Sorry:) what are you going to do with the ROI?
>>
>> Use it mri_glmfit, in order to compare cortical thickness in that ROI
>> between groups, and also (in a different design) to compare the
>> correlations between thickness and a behavioural predictor
>>
> If you're doing the final analysis in the group space, then just use the
> ROI in the group space.
>
>>
>> 4) This opens up a coronal slice rather than a 3D image as in
>> the picture I linked to. Do I need to do anything else in
>> tkmedit after it opens? (I tried pressing some buttons, and
>> some menu options, with no luck)
>>
>> that was generated with tksurfer, something like tksurfer subject
>> lh pial -aparc
>>
>> Using that command, I get a brain > sRcnX93M/8-05-2013_18-25-04.**html>
>> that has only little of its surface coloured, unlike the image <
>> http://brainybehavior.com/**neuroimaging/wp-content/**
>> uploads/2010/05/jt_rh_pial_**freesurfer_tksurfer.png>
>> that I linked to, in which it seems each major structure is identified in
>> its own colour, and these colours fill up the entire brain (there's no grey
>> surface left)
>>
> Change the annotation mode to be filled instead of outline mode. In the
> tksurfer control window, it is the 4th button from the right on the top
> row. It will say "Draw Filled Labels" when you put the mouse over it
>
>>
>>
>>
>> On 8 May 2013 16:54, Douglas N Greve
>> > > >
>> >
>> >>
>> wrote:
>>
>>
>> On 05/08/2013 11:43 AM, Tudor Popescu wrote:
>> > Hello all,
>> >
>> > I have some more burning questions, for which I thank you in
>> advance!!
>> >
>> > 1) I expected that "mri_annotation2label --subject 01
>> --hemi lh
>> > --outdir $SUBJECTS_DIR/myLabels" would give me labels
>> corresponding to
>> > structures of the standard FreeSurfer atlas (Destrieux),
>> instead it
>> > produced labels with completely different names, e.g.
>> > "lh.postcentral.label" instead of the Destrieux name
>> > "lh_S_postcentral". Exactly how different are they, and
>> which ones
>> > should be used in mri_glmfit for ROI analyses?
>> That is the Desikan/Killiany atlas it uses by default.
>> Tell it to use
>> aparc.a2009s to get Destrieux
>> >
>> > 2) Does it make sense to use mri_glmfit with volumetric
>> masks, or
>> > should it only be used with labels?
>> If you are analyzing surface data you have to give it a
>> surface
>> mask or
>> a surface la

Re: [Freesurfer] Labels and parcellations

[Douglas N Greve]

On 05/08/2013 01:28 PM, Tudor Popescu wrote:
>
> 2) ..and when would you use a surface mask versus a surface label?
>
> It is just a matter of convenience.
>
> Sorry, I meant I don't really understand what the difference is 
> between a label and a mask, and in which circumstances each should be 
> used. I tried finding an answer on the wiki but no luck..
A label is a text file with a list of points in the label. A mask is 
binary file (eg, mgh or nii) with 1s where the mask is and 0s otherwise.
>
> 3) (it seems you skipped over this one:) )
>
> Sorry:) what are you going to do with the ROI?
>
> Use it mri_glmfit, in order to compare cortical thickness in that ROI 
> between groups, and also (in a different design) to compare the 
> correlations between thickness and a behavioural predictor
If you're doing the final analysis in the group space, then just use the 
ROI in the group space.
>
> 4) This opens up a coronal slice rather than a 3D image as in
> the picture I linked to. Do I need to do anything else in
> tkmedit after it opens? (I tried pressing some buttons, and
> some menu options, with no luck)
>
> that was generated with tksurfer, something like tksurfer subject
> lh pial -aparc
>
> Using that command, I get a brain 
>  that 
> has only little of its surface coloured, unlike the image 
> 
>  
> that I linked to, in which it seems each major structure is identified 
> in its own colour, and these colours fill up the entire brain (there's 
> no grey surface left)
Change the annotation mode to be filled instead of outline mode. In the 
tksurfer control window, it is the 4th button from the right on the top 
row. It will say "Draw Filled Labels" when you put the mouse over it
>
>
>
> On 8 May 2013 16:54, Douglas N Greve
> mailto:gr...@nmr.mgh.harvard.edu>
>  >> wrote:
>
>
> On 05/08/2013 11:43 AM, Tudor Popescu wrote:
> > Hello all,
> >
> > I have some more burning questions, for which I thank you in
> advance!!
> >
> > 1) I expected that "mri_annotation2label --subject 01
> --hemi lh
> > --outdir $SUBJECTS_DIR/myLabels" would give me labels
> corresponding to
> > structures of the standard FreeSurfer atlas (Destrieux),
> instead it
> > produced labels with completely different names, e.g.
> > "lh.postcentral.label" instead of the Destrieux name
> > "lh_S_postcentral". Exactly how different are they, and
> which ones
> > should be used in mri_glmfit for ROI analyses?
> That is the Desikan/Killiany atlas it uses by default.
> Tell it to use
> aparc.a2009s to get Destrieux
> >
> > 2) Does it make sense to use mri_glmfit with volumetric
> masks, or
> > should it only be used with labels?
> If you are analyzing surface data you have to give it a
> surface
> mask or
> a surface label.
> >
> > 3) Say I want to do a ROI analysis for a particular
> brain structure;
> > how much is there to gain by manually (visually) creating a
> label/mask
> > for that structure in each subject (i.e. taking into account
> > individual anatomy), as opposed to just using an atlas
> to create a
> > single label/mask for use with all subjects?
> >
> > 4) I used the command "tkmedit 01 aparc+aseg.mgz" (I
> also tried
> adding
> > " -aux brainmask.mgz -surfs –aseg"), thinking that it would
> display an
> > image like this
> >
>
> 
> ,
> > i.e. with all the main cortical structures displayed in
> different
> > colours. Does aparc+aseg.mgz not contain all the necessary
> information
> > to produce such an image?
> try tkmedit subject orig.mgz -seg aparc+aseg.mgz
> >
> > 5) Which mask file corresponds to the brain extraction
> procedure of
> > recon-all? I see that a subject's /mri/brainmask.mgz is
> not a binary
> > map, as expectd, but instead a BETted (skull-stripped)
> version
> of that
> > subject's structural.
> We don't create a binary mask. You can with mri_binarize --i
> brainmask.mgz --min 0.5 --o 

Re: [Freesurfer] Labels and parcellations

[Tudor Popescu]

> 2) ..and when would you use a surface mask versus a surface label?
>>
> It is just a matter of convenience.

Sorry, I meant I don't really understand what the difference is between a
label and a mask, and in which circumstances each should be used. I tried
finding an answer on the wiki but no luck..

 3) (it seems you skipped over this one:) )
>>
> Sorry:) what are you going to do with the ROI?

Use it mri_glmfit, in order to compare cortical thickness in that ROI
between groups, and also (in a different design) to compare the
correlations between thickness and a behavioural predictor


>  4) This opens up a coronal slice rather than a 3D image as in the picture
>> I linked to. Do I need to do anything else in tkmedit after it opens? (I
>> tried pressing some buttons, and some menu options, with no luck)
>>
> that was generated with tksurfer, something like tksurfer subject lh pial
> -aparc
>
Using that command, I get a
brainthat
has only little of its surface coloured, unlike the
imagethat
I linked to, in which it seems each major structure is identified in
its own colour, and these colours fill up the entire brain (there's no grey
surface left)




>
>> On 8 May 2013 16:54, Douglas N Greve > gr...@nmr.mgh.harvard.**edu >> wrote:
>>
>>
>> On 05/08/2013 11:43 AM, Tudor Popescu wrote:
>> > Hello all,
>> >
>> > I have some more burning questions, for which I thank you in
>> advance!!
>> >
>> > 1) I expected that "mri_annotation2label --subject 01 --hemi lh
>> > --outdir $SUBJECTS_DIR/myLabels" would give me labels
>> corresponding to
>> > structures of the standard FreeSurfer atlas (Destrieux), instead it
>> > produced labels with completely different names, e.g.
>> > "lh.postcentral.label" instead of the Destrieux name
>> > "lh_S_postcentral". Exactly how different are they, and which ones
>> > should be used in mri_glmfit for ROI analyses?
>> That is the Desikan/Killiany atlas it uses by default. Tell it to use
>> aparc.a2009s to get Destrieux
>> >
>> > 2) Does it make sense to use mri_glmfit with volumetric masks, or
>> > should it only be used with labels?
>> If you are analyzing surface data you have to give it a surface
>> mask or
>> a surface label.
>> >
>> > 3) Say I want to do a ROI analysis for a particular brain structure;
>> > how much is there to gain by manually (visually) creating a
>> label/mask
>> > for that structure in each subject (i.e. taking into account
>> > individual anatomy), as opposed to just using an atlas to create a
>> > single label/mask for use with all subjects?
>> >
>> > 4) I used the command "tkmedit 01 aparc+aseg.mgz" (I also tried
>> adding
>> > " -aux brainmask.mgz -surfs –aseg"), thinking that it would
>> display an
>> > image like this
>> >
>> > uploads/2010/05/jt_rh_pial_**freesurfer_tksurfer.png
>> >,
>> > i.e. with all the main cortical structures displayed in different
>> > colours. Does aparc+aseg.mgz not contain all the necessary
>> information
>> > to produce such an image?
>> try tkmedit subject orig.mgz -seg aparc+aseg.mgz
>> >
>> > 5) Which mask file corresponds to the brain extraction procedure of
>> > recon-all? I see that a subject's /mri/brainmask.mgz is not a binary
>> > map, as expectd, but instead a BETted (skull-stripped) version
>> of that
>> > subject's structural.
>> We don't create a binary mask. You can with mri_binarize --i
>> brainmask.mgz --min 0.5 --o binarymask.mgz
>>
>> doug
>> >
>> > Thank you so much for your help!
>> > Tudor
>> >
>> >
>> > __**_
>> > 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

Re: [Freesurfer] Labels and parcellations

[Douglas N Greve]

On 05/08/2013 12:02 PM, Tudor Popescu wrote:
> Thanks Doug. Follow-ups below
>
> 2) ..and when would you use a surface mask versus a surface label?
It is just a matter of convenience.
>
> 3) (it seems you skipped over this one:) )
Sorry:) what are you going to do with the ROI?
>
> 4) This opens up a coronal slice rather than a 3D image as in the 
> picture I linked to. Do I need to do anything else in tkmedit after it 
> opens? (I tried pressing some buttons, and some menu options, with no 
> luck)
that was generated with tksurfer, something like tksurfer subject lh 
pial -aparc
>
> On 8 May 2013 16:54, Douglas N Greve  > wrote:
>
>
> On 05/08/2013 11:43 AM, Tudor Popescu wrote:
> > Hello all,
> >
> > I have some more burning questions, for which I thank you in
> advance!!
> >
> > 1) I expected that "mri_annotation2label --subject 01 --hemi lh
> > --outdir $SUBJECTS_DIR/myLabels" would give me labels
> corresponding to
> > structures of the standard FreeSurfer atlas (Destrieux), instead it
> > produced labels with completely different names, e.g.
> > "lh.postcentral.label" instead of the Destrieux name
> > "lh_S_postcentral". Exactly how different are they, and which ones
> > should be used in mri_glmfit for ROI analyses?
> That is the Desikan/Killiany atlas it uses by default. Tell it to use
> aparc.a2009s to get Destrieux
> >
> > 2) Does it make sense to use mri_glmfit with volumetric masks, or
> > should it only be used with labels?
> If you are analyzing surface data you have to give it a surface
> mask or
> a surface label.
> >
> > 3) Say I want to do a ROI analysis for a particular brain structure;
> > how much is there to gain by manually (visually) creating a
> label/mask
> > for that structure in each subject (i.e. taking into account
> > individual anatomy), as opposed to just using an atlas to create a
> > single label/mask for use with all subjects?
> >
> > 4) I used the command "tkmedit 01 aparc+aseg.mgz" (I also tried
> adding
> > " -aux brainmask.mgz -surfs –aseg"), thinking that it would
> display an
> > image like this
> >
> 
> ,
> > i.e. with all the main cortical structures displayed in different
> > colours. Does aparc+aseg.mgz not contain all the necessary
> information
> > to produce such an image?
> try tkmedit subject orig.mgz -seg aparc+aseg.mgz
> >
> > 5) Which mask file corresponds to the brain extraction procedure of
> > recon-all? I see that a subject's /mri/brainmask.mgz is not a binary
> > map, as expectd, but instead a BETted (skull-stripped) version
> of that
> > subject's structural.
> We don't create a binary mask. You can with mri_binarize --i
> brainmask.mgz --min 0.5 --o binarymask.mgz
>
> doug
> >
> > Thank you so much for your help!
> > Tudor
> >
> >
> > ___
> > 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.
>
>

-- 
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/

Re: [Freesurfer] Labels and parcellations

[Tudor Popescu]

Thanks Doug. Follow-ups below

2) ..and when would you use a surface mask versus a surface label?

3) (it seems you skipped over this one:) )

4) This opens up a coronal slice rather than a 3D image as in the picture I
linked to. Do I need to do anything else in tkmedit after it opens? (I
tried pressing some buttons, and some menu options, with no luck)

On 8 May 2013 16:54, Douglas N Greve  wrote:

>
> On 05/08/2013 11:43 AM, Tudor Popescu wrote:
> > Hello all,
> >
> > I have some more burning questions, for which I thank you in advance!!
> >
> > 1) I expected that "mri_annotation2label --subject 01 --hemi lh
> > --outdir $SUBJECTS_DIR/myLabels" would give me labels corresponding to
> > structures of the standard FreeSurfer atlas (Destrieux), instead it
> > produced labels with completely different names, e.g.
> > "lh.postcentral.label" instead of the Destrieux name
> > "lh_S_postcentral". Exactly how different are they, and which ones
> > should be used in mri_glmfit for ROI analyses?
> That is the Desikan/Killiany atlas it uses by default. Tell it to use
> aparc.a2009s to get Destrieux
> >
> > 2) Does it make sense to use mri_glmfit with volumetric masks, or
> > should it only be used with labels?
> If you are analyzing surface data you have to give it a surface mask or
> a surface label.
> >
> > 3) Say I want to do a ROI analysis for a particular brain structure;
> > how much is there to gain by manually (visually) creating a label/mask
> > for that structure in each subject (i.e. taking into account
> > individual anatomy), as opposed to just using an atlas to create a
> > single label/mask for use with all subjects?
> >
> > 4) I used the command "tkmedit 01 aparc+aseg.mgz" (I also tried adding
> > " -aux brainmask.mgz -surfs –aseg"), thinking that it would display an
> > image like this
> > <
> http://brainybehavior.com/neuroimaging/wp-content/uploads/2010/05/jt_rh_pial_freesurfer_tksurfer.png
> >,
> > i.e. with all the main cortical structures displayed in different
> > colours. Does aparc+aseg.mgz not contain all the necessary information
> > to produce such an image?
> try tkmedit subject orig.mgz -seg aparc+aseg.mgz
> >
> > 5) Which mask file corresponds to the brain extraction procedure of
> > recon-all? I see that a subject's /mri/brainmask.mgz is not a binary
> > map, as expectd, but instead a BETted (skull-stripped) version of that
> > subject's structural.
> We don't create a binary mask. You can with mri_binarize --i
> brainmask.mgz --min 0.5 --o binarymask.mgz
>
> doug
> >
> > Thank you so much for your help!
> > Tudor
> >
> >
> > ___
> > 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 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.

Re: [Freesurfer] Labels and parcellations

[Douglas N Greve]

On 05/08/2013 11:43 AM, Tudor Popescu wrote:
> Hello all,
>
> I have some more burning questions, for which I thank you in advance!!
>
> 1) I expected that "mri_annotation2label --subject 01 --hemi lh 
> --outdir $SUBJECTS_DIR/myLabels" would give me labels corresponding to 
> structures of the standard FreeSurfer atlas (Destrieux), instead it 
> produced labels with completely different names, e.g. 
> "lh.postcentral.label" instead of the Destrieux name 
> "lh_S_postcentral". Exactly how different are they, and which ones 
> should be used in mri_glmfit for ROI analyses?
That is the Desikan/Killiany atlas it uses by default. Tell it to use 
aparc.a2009s to get Destrieux
>
> 2) Does it make sense to use mri_glmfit with volumetric masks, or 
> should it only be used with labels?
If you are analyzing surface data you have to give it a surface mask or 
a surface label.
>
> 3) Say I want to do a ROI analysis for a particular brain structure; 
> how much is there to gain by manually (visually) creating a label/mask 
> for that structure in each subject (i.e. taking into account 
> individual anatomy), as opposed to just using an atlas to create a 
> single label/mask for use with all subjects?
>
> 4) I used the command "tkmedit 01 aparc+aseg.mgz" (I also tried adding 
> " -aux brainmask.mgz -surfs –aseg"), thinking that it would display an 
> image like this 
> ,
>  
> i.e. with all the main cortical structures displayed in different 
> colours. Does aparc+aseg.mgz not contain all the necessary information 
> to produce such an image?
try tkmedit subject orig.mgz -seg aparc+aseg.mgz
>
> 5) Which mask file corresponds to the brain extraction procedure of 
> recon-all? I see that a subject's /mri/brainmask.mgz is not a binary 
> map, as expectd, but instead a BETted (skull-stripped) version of that 
> subject's structural.
We don't create a binary mask. You can with mri_binarize --i 
brainmask.mgz --min 0.5 --o binarymask.mgz

doug
>
> Thank you so much for your help!
> Tudor
>
>
> ___
> 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] Labels and parcellations

[Tudor Popescu]

Hello all,

I have some more burning questions, for which I thank you in advance!!

1) I expected that "mri_annotation2label --subject 01 --hemi lh --outdir
$SUBJECTS_DIR/myLabels" would give me labels corresponding to structures of
the standard FreeSurfer atlas (Destrieux), instead it produced labels with
completely different names, e.g. "lh.postcentral.label" instead of the
Destrieux name "lh_S_postcentral". Exactly how different are they, and
which ones should be used in mri_glmfit for ROI analyses?

2) Does it make sense to use mri_glmfit with volumetric masks, or should it
only be used with labels?

3) Say I want to do a ROI analysis for a particular brain structure; how
much is there to gain by manually (visually) creating a label/mask for that
structure in each subject (i.e. taking into account individual anatomy), as
opposed to just using an atlas to create a single label/mask for use with
all subjects?

4) I used the command "tkmedit 01 aparc+aseg.mgz" (I also tried adding "
-aux brainmask.mgz -surfs –aseg"), thinking that it would display an image
like 
this,
i.e. with all the main cortical structures displayed in different colours.
Does aparc+aseg.mgz not contain all the necessary information to produce
such an image?

5) Which mask file corresponds to the brain extraction procedure of
recon-all? I see that a subject's /mri/brainmask.mgz is not a binary map,
as expectd, but instead a BETted (skull-stripped) version of that subject's
structural.

Thank you so much for your help!
Tudor
___
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.