I have a question about extracting functional data from structurally defined cortical labels. Here's what I'm doing:
1) Run a gammafit analysis;
2) Define cortical ROIs by using mri_annotation2label to make labels from aparc.annot;
3) Define sub-cortical ROIs by using mri_cor2label to make labels from aseg.mgz;
4) Extract functional data from labels using func2roi-sess;
5) Summarize the data with roisummary-sess.
I usually view the cortical ROIs on the inflated (or pial) surface and the subcortical ROIs in the volume, and they usually look great. But when I view the cortical labels in the volume, they seem to consist of a 1-voxel thick strip hugging the white matter; they don't seem to cover the gray matter at all. Does this mean that I am extracting functional data from a 1-voxel thick strip (!), or am I just mis-understanding something about how a cortical label appears in the volume? Also, I know that "white" is the default surface in mri_annotation2label, but if I switch to "pial", make a label, and view it in the volume, it still looks like a 1-voxel thick strip, just around the outside edge of the gray matter (instead of around the inside edge if I use "white" as the surface).
My simple mind wonders why these cortical labels-which look great on the inflated brain-don't cover the region between the pial and white surfaces (i.e., the gray matter) in the volume. Am I making a basic error here? If so, what do I need to do to extract functional data from cortically-defined ROIs? I am running version 4.0.2.
Thanks!
Dan D.
Yes, it is that 1mm3 voxel on the white (or pial) surface. For sampling into the functional volume, this is usually not an issue as the functional voxels are generally much bigger.
You can change the cortical sampling depth by running mri_label2label with --projfrac (eg, --projfrac 0.5 for midway between the white and pial).
doug
Dan Dillon wrote:
I have a question about extracting functional data from structurally defined cortical labels. Here's what I'm doing:
Run a gammafit analysis;Define cortical ROIs by using mri_annotation2label to makelabels from aparc.annot;
Define sub-cortical ROIs by using mri_cor2label to makelabels from aseg.mgz;
Extract functional data from labels using func2roi-sess;Summarize the data with roisummary-sess.I usually view the cortical ROIs on the inflated (or pial) surface and the subcortical ROIs in the volume, and they usually look great. But when I view the cortical labels in the volume, they seem to consist of a 1-voxel thick strip hugging the white matter; they don't seem to cover the gray matter at all. Does this mean that I am extracting functional data from a 1-voxel thick strip (!), or am I just mis-understanding something about how a cortical label appears in the volume? Also, I know that "white" is the default surface in mri_annotation2label, but if I switch to "pial", make a label, and view it in the volume, it still looks like a 1-voxel thick strip, just around the outside edge of the gray matter (instead of around the inside edge if I use "white" as the surface).
My simple mind wonders why these cortical labels--which look great on the inflated brain--don't cover the region between the pial and white surfaces (i.e., the gray matter) in the volume. Am I making a basic error here? If so, what do I need to do to extract functional data from cortically-defined ROIs? I am running version 4.0.2.
Thanks!
Dan D.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Doug. So if I'm understanding you, the 1mm3 voxel overlaps the larger (3 x 3 x 3 mm) functional voxel and so when I pull out the functional data it comes from the entire functional voxel? Is the combination of mri_annotation2label and func2roi-sess a common way to extract data from cortical ROIs? Is there a set of flags that generally works well that I should use to make sure I'm getting data from all the gray matter?
Thanks!
Dan
-----Original Message----- From: Doug Greve [mailto:greve@nmr.mgh.harvard.edu] Sent: Monday, March 17, 2008 4:56 PM To: Dan Dillon Cc: 'Freesurfer Mailing List' Subject: Re: [Freesurfer] Question re: extracting functional data from cortical ROIs
Yes, it is that 1mm3 voxel on the white (or pial) surface. For sampling into the functional volume, this is usually not an issue as the functional voxels are generally much bigger.
You can change the cortical sampling depth by running mri_label2label with --projfrac (eg, --projfrac 0.5 for midway between the white and pial).
doug
Dan Dillon wrote:
I have a question about extracting functional data from structurally defined cortical labels. Here's what I'm doing:
1) Run a gammafit analysis;
2) Define cortical ROIs by using mri_annotation2label to make labels from aparc.annot;
3) Define sub-cortical ROIs by using mri_cor2label to make labels from aseg.mgz;
4) Extract functional data from labels using func2roi-sess;
5) Summarize the data with roisummary-sess.
I usually view the cortical ROIs on the inflated (or pial) surface and the subcortical ROIs in the volume, and they usually look great. But when I view the cortical labels in the volume, they seem to consist of a 1-voxel thick strip hugging the white matter; they don't seem to cover the gray matter at all. Does this mean that I am extracting functional data from a 1-voxel thick strip (!), or am I just mis-understanding something about how a cortical label appears in the volume? Also, I know that "white" is the default surface in mri_annotation2label, but if I switch to "pial", make a label, and view it in the volume, it still looks like a 1-voxel thick strip, just around the outside edge of the gray matter (instead of around the inside edge if I use "white" as the surface).
My simple mind wonders why these cortical labels-which look great on the inflated brain-don't cover the region between the pial and white surfaces (i.e., the gray matter) in the volume. Am I making a basic error here? If so, what do I need to do to extract functional data from cortically-defined ROIs? I am running version 4.0.2.
Thanks!
Dan D.
_____
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Yes, it comes from all of a functional voxel that a label point falls into. You can require that a given functional voxel be filled to a certain fraction with -labelfillthresh to func2roi-sess. I don't that will fill the ribbon with a label, but what you can do is to use mri_cor2label on aparc+aseg to get a ribbon-filled label for the cortical rois.
doug
Dan Dillon wrote:
Hi Doug. So if I'm understanding you, the 1mm3 voxel overlaps the larger (3 x 3 x 3 mm) functional voxel and so when I pull out the functional data it comes from the entire functional voxel? Is the combination of mri_annotation2label and func2roi-sess a common way to extract data from cortical ROIs? Is there a set of flags that generally works well that I should use to make sure I'm getting data from all the gray matter?
Thanks!
Dan
-----Original Message----- From: Doug Greve [mailto:greve@nmr.mgh.harvard.edu] Sent: Monday, March 17, 2008 4:56 PM To: Dan Dillon Cc: 'Freesurfer Mailing List' Subject: Re: [Freesurfer] Question re: extracting functional data from cortical ROIs
Yes, it is that 1mm3 voxel on the white (or pial) surface. For sampling into the functional volume, this is usually not an issue as the functional voxels are generally much bigger.
You can change the cortical sampling depth by running mri_label2label with --projfrac (eg, --projfrac 0.5 for midway between the white and pial).
doug
Dan Dillon wrote:
I have a question about extracting functional data from structurally defined cortical labels. Here's what I'm doing:
Run a gammafit analysis;Define cortical ROIs by using mri_annotation2label to makelabels from aparc.annot;
Define sub-cortical ROIs by using mri_cor2label to makelabels from aseg.mgz;
Extract functional data from labels using func2roi-sess;Summarize the data with roisummary-sess.I usually view the cortical ROIs on the inflated (or pial) surface and the subcortical ROIs in the volume, and they usually look great. But when I view the cortical labels in the volume, they seem to consist of a 1-voxel thick strip hugging the white matter; they don't seem to cover the gray matter at all. Does this mean that I am extracting functional data from a 1-voxel thick strip (!), or am I just mis-understanding something about how a cortical label appears in the volume? Also, I know that "white" is the default surface in mri_annotation2label, but if I switch to "pial", make a label, and view it in the volume, it still looks like a 1-voxel thick strip, just around the outside edge of the gray matter (instead of around the inside edge if I use "white" as the surface).
My simple mind wonders why these cortical labels--which look great on the inflated brain--don't cover the region between the pial and white surfaces (i.e., the gray matter) in the volume. Am I making a basic error here? If so, what do I need to do to extract functional data from cortically-defined ROIs? I am running version 4.0.2.
Thanks!
Dan D.
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
In order to help us help you, please follow the steps in:
surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
freesurfer@nmr.mgh.harvard.edu
-
Dan Dillon -
Doug Greve