Thank you Anderson.
A follow-on question: what is meant by density vs. volume in VBM, and are they really different? I can imagine how volume could be extrapolated at the level of single voxels, but I struggle to understand how this is used to assess the volume of larger structures.
Thank you, Andrew
Date: Fri, 3 Jul 2015 07:29:41 +0100
From: [log in to unmask]
Subject: Re: [FSL] [Repost] - VBM - determining volume change vs. tissue density changež
To: [log in to unmask]
Hi Andrew,
Please see below:
On 3 July 2015 at 05:49, Drew Faust <[log in to unmask]> wrote:
Hello FSL team,
I apologize for such a basic question, but I have searched the listserv archives and the VBM FAQs for an answer and don't find it.
I have used FSL-VBM using the standard and recommended settings (according to the VBM User Guide) to generate my template-registered grey matter images. In reading through helpful guides to VBM (e.g. Douglas Greve's http://cds.ismrm.org/protected/11MProceedings/files/ISMRM2011-8410.pdf), I find references to using VBM to measure tissue concentration or density, and then using the Jacobian determinant to measure volume of the same structural images. Using the standard VBM settings, as I understand it the output of randomise reflects the difference in tissue density or concentration between groups.
Nope. The VBM implemented in FSL uses Jacobian correction and the output refers to gray matter volume.
All the best,
Anderson
If I want to instead output the volume differences between groups, would I run
fslvbm_3_proc on the Jacobian warp field instead of on the warped FA image? If I understand the file name structure correctly, this would mean running fslvbm_3_proc on {identifier}_FA_to_target_warp.nii.gz instead of {identifier}_FA_to_target.nii.gz.
Is this correct? If not, how can I use VBM to measure volume instead of tissue density?
Thanks ever so much for the help,
Andrew Faust
