Just a word of caution - VBM results may or may not replicate with other volumetric methods. Coregistration artifacts? It can happen.

BRAINS is my suggestion. For disclosure - I was involved in developing it, I customize it, and I use it in my business every day to provide measures for researchers. So I may a little biased :) It is hosted at NITRC, developed at Iowa. A new version, Beta 4, is now available that I feel is good. It normally takes both T1 and T2 images as a minimum, but there is a T1-only version of the workup, too, that I will use in the command line below. If you have a T2 on these subjects, use that in addition to the T1.

http://www.nitrc.org/projects/brains

Un-tar it into /opt. You will need to have a few prerequisites installed, such as tcl 8.4. I have run it on CentOS 6 and several newer versions of Ubuntu. The easiest way to run it is by creating a batch file for each scan with the following command. You can put more than one scan's command in the batch file, too.

Brains::AutoWorkup::WorkupT1Only PathForOutputDirectory PathtoT1Image {110} {0} {10} ""

Then run this by first setting up the environment

source /opt/BRAINS2/bin/brains3_setup.sh (for bash, .csh for cshell)
brains3 -b batchFileName.b3

It will take about 45 minutes or so per scan, depending on your computer. The T1 image will be ACPC aligned, inhomogeneity corrected, intensity normalized, and tissue classified. Output masks of the brain and a number of subcortical structures will be created, which can be reviewed in Slicer, freeview, or BRAINSTracerQT that comes with BRAINS (you need qt4 installed). I prefer the simplicity of freeview (which comes with FreeSurfer 5.1), but I know others like Slicer. Measurements will be in csv files in the output directory, 10_AUTO.110 under the output directory you specified in the command line. I suggest with ANY method that you do a visual inspection of every scan before you use the data. One would think that goes without saying, but it happens. Results for any method can be variable depending on scan parameters, quality issues, unforeseen issues, etc...

The other option is FreeSurfer, but I am less inclined to suggest that. Let us know how things work for you and if you have any questions along the way.

Ron Pierson

Brain Image Analysis, LLC