 |
 |
Hi Zhiguo, The design in the link can be used with randomise for a 2x2 repeated measurements design after a few changes regarding the exchangeability blocks. Even so, however, given that you'd like to account for nuisance variables, other less trivial changes to the design and contrasts would have to be done for the main effect of group. It is much simpler to compute the sums and subtractions -- these can be scripted (use a for-loop for each subject, calling fslmaths to do the computations, then merge in new 4D files with fslmerge). The main effect of time is computed using differences within subject. The main effect of group using the average (or the sum) within subject. The interaction with a difference for one group, and the difference with the sign swapped for the other group. All three tests will the use the same design: a 1-sample t-test adjusted for covariate(s), like this example from the manual: http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM#Single-Group_Average_with_Additional_Covariate All the best, Anderson 2015-01-21 2:49 GMT+00:00 Zhiguo Jiang <[log in to unmask]>: > After digging in the archives, i am still not clear that how to proceed > with my data. > i have a DTI data (2groups[HC,Patient] x 2 visits[time1,time2) , i would > like to know the main effect of group, the longitudinal difference within > group HC and Patient. at the same time, i would like to control for age and > education. After reading > http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM#ANOVA:_2-groups.2C_2-levels_per_subject_.282-way_Mixed_Effect_ANOVA.29, > my impression is i can't construct a 2x2 ANOVA model in randomise? i have > to do this in series of t tests? i could compute the difference images > (time2-time1) and do one sample t test seperately then i could compute the > average image (time2+time1)/2 for both groups and run two sample t test. > With the ttests i can easily add in the covariates of age and education. > is it possible that i can do all these in one model but different > contrasts? > > also i would like to ask whether i can incorporate a duration(of disease) > variate in the model to account for the variability in patient data (but no > data for control). can i just put down 0s for control and add them as one > column in the design matrix? > thanks in advance > > Tony >