Backgound
on VBM
Voxel-based morphometry (VBM) is a technique that
uses a voxel-by-voxel comparison of tissue concentration between
different groups of subjects. This involves spatial normalization
of all images to the same stereotactic space, extracting gray or
white matter segments, smoothing the images, and finally carrying
out statistical analyses to make inferences about the data. Corrections
for multiple comparisons are made using the theory of Gaussian Random
Fields.
Here is a presentation
on VBM Theory and Methods .
Readings
on VBM Methods
- A. Mechelli, C.J. Price, K.J. Friston, and
J. Ashburner. Voxel-Based Morphometry of the Human Brain: Methods
and Applications. Current Medical Imaging Reviews, pp 105-113,
2005. PDF
- J. Ashburner and K.J. Friston. Why voxel-based
morphometry should be used. NeuroImage, 14(6):1238-1243, 2001.
PDF
- J. Ashburner and K.J. Friston. Voxel-Based
Morphometry -- The Methods. NeuroImage, 11:805-821, 2000. PDF
The
Bookstein Controversy
This refers to the debated issue of imperfectly
registered images, and the impact this has on SPM results. The document
by Thacker also addresses some other pitfalls of VBM that one should
be aware of, before conducting any of these analyses.
- A. Mechelli, C.J. Price, K.J. Friston, and
J. Ashburner. Voxel-Based Morphometry of the Human Brain: Methods
and Applications. Current Medical Imaging Reviews, pp 105-113,
2005. PDF
- Bookstein F. "Voxel-based
Morphometry" should not be used with imperfectly registered
images. PDF
- Thacker, N.A. A Critical Analysis of
Voxel-Based Morphometry (2005). PDF
|
