Dear Bruce and Co.,
I was wondering if there is a way to perform ANOVA on the cortical thickness data; for example evaluating the interaction between age, gender and three (categorical variables) levels of blood pressure? Its not obvious how this might be done using the current interface.
Mike Chee
Hi, I wanted to calculate total surface-based cortical GM, and was wondering if there was a command to do this automatically. If not, and I need to add up the volumes in ?h.aparc.stats manually, are there ones I should leave out (like unknown, corpus callosum, etc)?
Thanks,
-Aaron-
Hi Aaron,
In terms of purely anatomically, to calculate the total GM I would not include the corpus callosum (a white matter structure) or the unknown region (which is an amalgam of subcortical areas).
Best,
rahul
Hi, I wanted to calculate total surface-based cortical GM, and was
wondering if there was a command to do this automatically. If not, and I need to add up the volumes in ?h.aparc.stats manually, are there ones I should leave out (like unknown, corpus callosum, etc)?
Thanks,
-Aaron-
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Thanks,
Related to this, I noticed these values tend to be smaller than the "cortical GM" labels generated by the subcortical segmentations (by about 10% in the data I was looking at). I was just curious if there was a known reason why this would be expected.
-Aaron-
-----Original Message----- From: rahul@nmr.mgh.harvard.edu [mailto:rahul@nmr.mgh.harvard.edu] Sent: Monday, March 19, 2007 12:18 PM To: Goldman, Aaron (NIH/NIMH) [C] Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] cortical GM from parcellations
Hi Aaron,
In terms of purely anatomically, to calculate the total GM I would not include the corpus callosum (a white matter structure) or the unknown region (which is an amalgam of subcortical areas).
Best,
rahul
Hi, I wanted to calculate total surface-based cortical GM, and was
wondering if there was a command to do this automatically. If not, and
I
need to add up the volumes in ?h.aparc.stats manually, are there ones
I
should leave out (like unknown, corpus callosum, etc)?
Thanks,
-Aaron-
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Aaron,
I wouldn't trust the aseg gm labels nearly as much as the surfaces. In our experience they are quite a bit more accurate.
cheers, Bruce
On Mon, 19 Mar 2007, Goldman, Aaron (NIH/NIMH) [C] wrote:
Thanks,
Related to this, I noticed these values tend to be smaller than the "cortical GM" labels generated by the subcortical segmentations (by about 10% in the data I was looking at). I was just curious if there was a known reason why this would be expected.
-Aaron-
-----Original Message----- From: rahul@nmr.mgh.harvard.edu [mailto:rahul@nmr.mgh.harvard.edu] Sent: Monday, March 19, 2007 12:18 PM To: Goldman, Aaron (NIH/NIMH) [C] Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] cortical GM from parcellations
Hi Aaron,
In terms of purely anatomically, to calculate the total GM I would not include the corpus callosum (a white matter structure) or the unknown region (which is an amalgam of subcortical areas).
Best,
rahul
Hi, I wanted to calculate total surface-based cortical GM, and was
wondering if there was a command to do this automatically. If not, and
I
need to add up the volumes in ?h.aparc.stats manually, are there ones
I
should leave out (like unknown, corpus callosum, etc)?
Thanks,
-Aaron-
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Yea, the volumetric segmentation is only accurate to 1mm^3, whereas the surface-based has subvoxel accuracy.
doug
Bruce Fischl wrote:
Hi Aaron,
I wouldn't trust the aseg gm labels nearly as much as the surfaces. In our experience they are quite a bit more accurate.
cheers, Bruce
On Mon, 19 Mar 2007, Goldman, Aaron (NIH/NIMH) [C] wrote:
Thanks,
Related to this, I noticed these values tend to be smaller than the "cortical GM" labels generated by the subcortical segmentations (by about 10% in the data I was looking at). I was just curious if there was a known reason why this would be expected.
-Aaron-
-----Original Message----- From: rahul@nmr.mgh.harvard.edu [mailto:rahul@nmr.mgh.harvard.edu] Sent: Monday, March 19, 2007 12:18 PM To: Goldman, Aaron (NIH/NIMH) [C] Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] cortical GM from parcellations
Hi Aaron,
In terms of purely anatomically, to calculate the total GM I would not include the corpus callosum (a white matter structure) or the unknown region (which is an amalgam of subcortical areas).
Best,
rahul
Hi, I wanted to calculate total surface-based cortical GM, and was
wondering if there was a command to do this automatically. If not, and
I
need to add up the volumes in ?h.aparc.stats manually, are there ones
I
should leave out (like unknown, corpus callosum, etc)?
Thanks,
-Aaron-
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Aaron,
you can use mris_anatomical_stats -l ?h.cortex
where the cortex labl is generated automatically and should be in the subjects label dir.
cheers, Bruce
On Mon, 19 Mar 2007, Goldman, Aaron (NIH/NIMH) [C] wrote:
Hi, I wanted to calculate total surface-based cortical GM, and was wondering if there was a command to do this automatically. If not, and I need to add up the volumes in ?h.aparc.stats manually, are there ones I should leave out (like unknown, corpus callosum, etc)?
Thanks,
-Aaron-
Hi Mike,
sorry for the delay, I was unavailable for most of yesterday. There's not a way to automatically do it, but it can be done. Eg, if you have two groups (classes) and one continuous variable (eg, age), and you want to test for an interaction between group and age, then you'd set up the fsgd with two groups and one continuous variable and run with DODS. This will give you 4 regressors: (1) Intercept for Group1 (2) Intercept for Group2, (3) Age Slope for Group1, and (4) Age Slope for Group2. The interaction is just the difference between the slopes, so use a contrast of [0 0 1 -1].
doug
Mike Chee wrote:
Dear Bruce and Co.,
I was wondering if there is a way to perform ANOVA on the cortical thickness data; for example evaluating the interaction between age, gender and three (categorical variables) levels of blood pressure? Its not obvious how this might be done using the current interface.
Mike Chee
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Mike,
sorry for the delay, I was unavailable for most of yesterday. There's not a way to automatically do it, but it can be done. Eg, if you have two groups (classes) and one continuous variable (eg, age), and you want to test for an interaction between group and age, then you'd set up the fsgd with two groups and one continuous variable and run with DODS. This will give you 4 regressors: (1) Intercept for Group1 (2) Intercept for Group2, (3) Age Slope for Group1, and (4) Age Slope for Group2. The interaction is just the difference between the slopes, so use a contrast of [0 0 1 -1].
doug
Mike Chee wrote:
Dear Bruce and Co.,
I was wondering if there is a way to perform ANOVA on the cortical thickness data; for example evaluating the interaction between age, gender and three (categorical variables) levels of blood pressure? Its not obvious how this might be done using the current interface.
Mike Chee
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Doug Greve -
Goldman, Aaron (NIH/NIMH) [C] -
Mike Chee -
rahul＠nmr.mgh.harvard.edu