It is a rare case. Only this I found
Path analysis of multivariate time series fMRI data with subject-level covariates
by Zhang, Yue, Ph.D., STATE UNIVERSITY OF NEW YORK AT STONY BROOK, 2007, 124 pages; 3335134
Also look at this link
Unified Structural Equation Modeling Approach for the Analysis of Multisubject, Multivariate Functional MRI Data (Jieun Kim, Wei Zhu, Linda Chang, Peter M. Bentler, and Thomas Ernst)