Abstract:

In order to improve the detection reliability of effective connectivity in brain network, an fMRI （Functional Magnetic Resonance Imaging） analytical approach of effective connectivity is proposed based on the Granger causality （GC） and the principle component analysis （PCA）. In this approach, first, temporal principal components are extracted via the PCA from the fMRI signals in the region of interest, and the patterns are considered as temporal reference information. Next, the Granger causality between the reference region and each of other voxels of the brain is calculated. Then, the results are mapped into the whole brain and a Granger causality map （GCM） is thus obtained. Moreover, a theoretical derivation is performed to verify the effectiveness of the proposed approach. The proposed approach is finally used t＇o analyze the GCM of a manual movement task-induced activation in the motor area, the results verifying the correctness of theory of motor-function neural network

Key words: Granger causality, principal component analysis, effective connectivity, functional magnetic resonance imaging