help > RE: Correcting for confounders and Liebermeister statistic
Dec 16, 2019  07:12 AM | Michelle Jansen
RE: Correcting for confounders and Liebermeister statistic
Originally posted by Chris Rorden:
For nuisance regression the software will automatically select GLM rather than Liebermeister measure. Specifically, I would add a negative value to the number of permutation which will enable the Freedman-Lane method. In general, unless you have a very large sample, you may want to consider a region of interest approach, which provides more statistical power than the voxelwise approach. With regions of interest, the lesion score becomes continuous (e.g. proportion of region damaged).

Dear Chris,

Thank you very much for your quick reply and information. Does this approach then still differ from from e.g., voxel-wise non-paramatric permutation tests comparing comparing the two groups?

I figured that it would be interesting to investigate two different questions; whether there are differences in degree of damage between the two disease groups (using voxel-wise non-parametric tests) and whether damage in a certain voxel relates more to a certain group (using a VLSM approach).

Best,

Michelle 

Threaded View

TitleAuthorDate
Michelle Jansen Dec 16, 2019
Chris Rorden Dec 16, 2019
RE: Correcting for confounders and Liebermeister statistic
Michelle Jansen Dec 16, 2019