Determining Effective Connectivity from FMRI Data Using a Gaussian Dynamic Bayesian Network

Two techniques that are based on the Bayesian network, Gaussian Bayesian network (BN) and discrete dynamic Bayesian network (DBN), have recently been used to determine the effective connectivity from functional magnetic resonance imaging (fMRI) data in an exploratory manner and provide a new method for the interactions among brain regions. However, Gaussian BN ignores the temporal relationships of interactions among brain regions, while discrete DBN loses a great of information by discretizing data. In this study, we proposed Gaussian DBN, which is based on Gaussian assumptions, to capture the temporal characteristics of connectivity with less associated loss of information. Synthetic data were generated to validate the effectiveness of this method, and the results were compared with discrete DBN. The result demonstrated that our method is both more robust than discrete DBN and an improvement over BN.