Hello everyone,
Recently, I was doing some resting state functional connectivity analysis in FSL. But now I have met some problems with my GLM steup. First, I will explain what I want to do, I have four groups of subjects, including schizophrenia group, depression group , bipolar group and one healthy control group (suppose each group has two patients, however, actually, in my data set, it is more than 2, and the subject numbers are different among each group). There were no repeated measurements. I want to know the common affected brain regions among each patient group compared with healthy control group, I also want to know the differently affected brain regions in the patient groups compared with HCs, and finally I want to know the relationship between some cognitive function deficit with the affected brain areas. The following is my design mat,
A B C D E
1 0 0 0 -6
1 0 0 0 -3
0 1 0 0 7
0 1 0 0 8
0 0 1 0 -4
0 0 1 0 -2
0 0 0 1 -3
0 0 0 1 -5
(A=schizophrenia, B=depression, C=bipolar, D=healthy controls, E=cognitive function)
Now I am confused by the contrasts, according to fslwiki and other's suggestion, I created the contrasts like this
(A-D) 1 0 0 -1 (F-test)
(B-D) 0 1 0 -1 (F-test)
(C-D) 0 0 1 -1 (F-test)
(meanHC) -1 -1 -1 1(t-test)
(meanA) 1 0 0 1(t-test)
(meanB) 0 1 0 1(t-test)
(meanC) 0 0 1 1(t-test)
[HC-(A+B+C)] -2 -2 -2 -2(t-test)
To tell the truth, I have no idea what I am doing and how to explain my contrasts. I also don't know how to add the covariate of the cognitive function to the contrast mat. Do you think my contrasts are right? Can you kindly tell me how to add the covariate analysis to the contrast mat?
Sorry for asking such stupid questions to bother you, looking forward for your reply.
Best,
Hua
|