Hi Mahinda
1. The subvoxel accuracy has been verified experimentally in Dianas Rosas'
paper:
http://www.ncbi.nlm.nih.gov/pubmed/11889230
where she measured the thickness under a microscope (ex vivo) and compared
it to our results and found they agreed within a couple of hundred microns.
It's easy to see why this would work. Imagine you have a voxel that is all
gray matter and the next voxel is all csf, so the pial boundary is between
them. Now imagine that there is some atrophy and the boundary moves into
the voxel. The voxel intensity will decline (on a T1) giving us the
information we need to detect this type of effect.
2. No we can't really distinguish loss of tissue from changes in tissue
properties. This is true of most of MRI.
3. You mean manually measuring the thickness? This is very very difficult
to do, an IMO would result in a significantly less accurate result. The
thickness has to be measured in 3D and not necessarily along any of the
cardinal viewing planes
cheers
Bruce
On Tue, 29 Sep 2015,
Mahinda Yogarajah wrote:
Dear Freesurfer team,
I have a manuscript under review which has used Freesurfer to compare
cortical thickness differences between groups, controls and patients with a
specific genetic mutation. There are robust, statistically significant group
differences in biologically plausible regions. However one of the reviewers
has taken offence at aspects of the methods, and before I put together a
response I was hoping for help from the Freesurfer team.
(1) In my methods I have stated that “the maps produced are not restricted
to the voxel resolution of the original data and thus are capable of
detecting sub-millimetre differences between groups,”
The reviewer takes umbrage with the fact that partial volume effects prevent
an MRI acquired with 1.1x1.1x1.1 mm3 voxel size from providing a metric
measuring sub-millimetric changes.
I was wondering how the Freesurfer team might compose a response to this ?
(2) The reviewer also asks the question as to how one can differentiate loss
of cortical volume (thickness in this case) from alterations in the contents
of the volume: changes in cell type, cell shape or cell volume. He/she asks
how one defines the tissue classes across which the spatial intensity
gradients are applied.
The first part of this question seems slightly circular to me (!). I would
answer the second part by pointing out that the initial surface is refined
to follow the intensity gradients between the white and gray matter (this is
referred to as the white surface). The white surface is then nudged to
follow the intensity gradients between the gray matter and CSF (this is the
pial surface). The cortical thickness is the distance in between them.
(3) Finally he/she suggests that one might confirm cortical thickness
differences by measuring sulcal thickness or subarachnoid space volumes over
those areas where I see significant results. This seems like over kill to me
- any suggestions as how to make a firm rebuttal ?
Thanks,
Mo
_______________________________________________
Freesurfer mailing list
[email protected]
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.
