Is any information about the manually-parcellated (CMA) training set used to create the spherical template used in fsv4.*? The information reported in Cerebral Cortex (2004) 14(1):11-22 reports 36 MRI volumes from an ongoing study.
Does this sound about right? Or have there been any updates to the spherical template since then?
Many Thanks for any info, Jenifer
Hi Jennifer,
the spherical stuff has nothing to do with the CMA. And in fact the spherical registration itself is not based on any manual labelings, just estimates of cortical geometry from a set of training subjects. The cortical parcellations do depend on manual labelings, but this is also not the CMA. Rather there are two sets - one done by Rahul Desikan (39 subjects I think) and the other by Christophe Destrieux (12 or 14 subjects). The current spherical template we use is from the same 39 that Rahul labeled.
cheers, Bruce
On Wed, 22 Apr 2009, Juranek, Jenifer wrote:
Is any information about the manually-parcellated (CMA) training set used to create the spherical template used in fsv4.*? The information reported in Cerebral Cortex (2004) 14(1):11-22 reports 36 MRI volumes from an ongoing study.
Does this sound about right? Or have there been any updates to the spherical template since then?
Many Thanks for any info, Jenifer
Ah...Excellent! Thanks so much for clearing things up for me.
Cheers, Jenifer
-----Original Message----- From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] Sent: Wednesday, April 22, 2009 11:55 AM To: Juranek, Jenifer Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] spherical template: CMA training set
Hi Jennifer,
the spherical stuff has nothing to do with the CMA. And in fact the spherical registration itself is not based on any manual labelings, just estimates of cortical geometry from a set of training subjects. The cortical parcellations do depend on manual labelings, but this is also not the CMA. Rather there are two sets - one done by Rahul Desikan (39 subjects I think) and the other by Christophe Destrieux (12 or 14 subjects). The current spherical template we use is from the same 39 that Rahul labeled.
cheers, Bruce
On Wed, 22 Apr 2009, Juranek, Jenifer wrote:
Is any information about the manually-parcellated (CMA) training set used to create the spherical template used in fsv4.*? The information reported in Cerebral Cortex (2004) 14(1):11-22 reports 36 MRI volumes from an ongoing study.
Does this sound about right? Or have there been any updates to the spherical template since then?
Many Thanks for any info, Jenifer
And just to bring this discussion full circle, how does the training set for the subcortical segmentation relate?? Is the training set (and associated manual labeling) that is used for the subcortical segmentation composed of the same 39 subjects used for both the current spherical template and the Desikan cortical atlas?
thanks, Mike H.
On Wed, 2009-04-22 at 12:55 -0400, Bruce Fischl wrote:
Hi Jennifer,
the spherical stuff has nothing to do with the CMA. And in fact the spherical registration itself is not based on any manual labelings, just estimates of cortical geometry from a set of training subjects. The cortical parcellations do depend on manual labelings, but this is also not the CMA. Rather there are two sets - one done by Rahul Desikan (39 subjects I think) and the other by Christophe Destrieux (12 or 14 subjects). The current spherical template we use is from the same 39 that Rahul labeled.
cheers, Bruce
On Wed, 22 Apr 2009, Juranek, Jenifer wrote:
Is any information about the manually-parcellated (CMA) training set used to create the spherical template used in fsv4.*? The information reported in Cerebral Cortex (2004) 14(1):11-22 reports 36 MRI volumes from an ongoing study.
Does this sound about right? Or have there been any updates to the spherical template since then?
Many Thanks for any info, Jenifer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
there is a lot of overlap but they are not quite the same. I think 30-35 of them are the same, but a few are different.
On Wed, 22 Apr 2009, Michael Harms wrote:
And just to bring this discussion full circle, how does the training set for the subcortical segmentation relate?? Is the training set (and associated manual labeling) that is used for the subcortical segmentation composed of the same 39 subjects used for both the current spherical template and the Desikan cortical atlas?
thanks, Mike H.
On Wed, 2009-04-22 at 12:55 -0400, Bruce Fischl wrote:
Hi Jennifer,
the spherical stuff has nothing to do with the CMA. And in fact the spherical registration itself is not based on any manual labelings, just estimates of cortical geometry from a set of training subjects. The cortical parcellations do depend on manual labelings, but this is also not the CMA. Rather there are two sets - one done by Rahul Desikan (39 subjects I think) and the other by Christophe Destrieux (12 or 14 subjects). The current spherical template we use is from the same 39 that Rahul labeled.
cheers, Bruce
On Wed, 22 Apr 2009, Juranek, Jenifer wrote:
Is any information about the manually-parcellated (CMA) training set used to create the spherical template used in fsv4.*? The information reported in Cerebral Cortex (2004) 14(1):11-22 reports 36 MRI volumes from an ongoing study.
Does this sound about right? Or have there been any updates to the spherical template since then?
Many Thanks for any info, Jenifer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Juranek, Jenifer -
Michael Harms