Print

---

last question.

what about if I implement a levene test to test the homogeneity of variance voxel-wise? 
do you suggest me to have a try or I violate any assumption of structural MR images?

thanks again for the advice,

Marco