I am looking for other reports, theory or implementation, of tests for
distributed effects in statistical maps.
I am particularly interested in modeling the weighted volume of all voxels in
an SPM above a threshold value, which has been termed the
"excedence mass" -- by whom I am not sure.
So far, I am aware of:
1. the "set-level" test implemented in SPM, which addresses the NUMBER of
clusters above that expected from the Euler characteristic
[Friston, K. J., A. Holmes, et al. (1996). “Detecting activations
in PET
and fMRI: levels of inference and power.” Neuroimage 4(3 Pt 1): 223-35.]
2. Worsley's test, which relates the MSOS of all voxels in an SPM to the
resels
in the analysis
[Worsley, K. J., J. B. Poline, et al. (1995). “Tests for distributed,
nonfocal brain activations.” Neuroimage 2(3): 183-94.]
3. Logigian et al, who implemented a weighted volume-type test on SPECT data
[Logigian, E. L., K. A. Johnson, et al. (1997). “Reversible cerebral
hypoperfusion in Lyme encephalopathy.” Neurology 49(6): 1661-70.]
I believe there is a report of a similar test using MR anatomic data, which I
have not found.
many thanks for any leads anyone can provide--
|