We have a couple of questions about implementing the optimized VBM protocol
in SPM2b.
1. In the defaults for both normalization and segmentation, we have to set a
cutoff value for the DCT basis functions. The default values are 25 for
normalization and 30 for segmentation. What does this mean and how can we
determine what is appropriate for VBM analysis? or should we simply use the
default values? Does it matter that the defaults are different between the
two processes? Any references regarding this would be greatly appreciated.
2. In SPM99, the modulation of the segmented images was performed after
creation of the unmodulated segmented images in order to address questions
regarding volume differences. In SPM2b, preserving concentration or volume
is an option in the initial normalization procedures. Therefore, we have the
option to perform the VBM procedures twice, once preserving concentration to
address density differences and once preserving volume to address volume
differences. Alternatively, we can preserve concentration and perform
modulation after creation of the unmodulated segmented images using the
matlab code as before). What is the appropriate way to proceed for VBM in
SPM2b?
Jacqueline Wood and MVittoria Spampinato
**************************************************
Jacqueline N. Wood, Ph.D.; MVittoria Spampinato, M.D.
Cognitive Neuroscience Section
National Institute of Neurological Disorders and Stroke
National Institutes of Health
Building 10; Room 5C205
10 Center Drive; MSC 1440
Bethesda
MD 20892-1440
U.S.A.
Tel: 301-402-0060
Fax: 301-480-2909
email: [log in to unmask]
**************************************************
|