Changes in Functional Connectivity of Resting-state after Motor Imagery Training Detected by Eigenvector Centrality Mapping

Motor imagery training has been indicated to be effective in motor function rehabilitation and motor skill learning. The neural mechanism underlying motor training has attracted increased neuroimaging explorations. Related neuroimaging studies demonstrated that resting-state can offer the possibility to examine the neural mechanism of motor execution training. However, motor imagery training, as another part of motor training, has been few investigated. To address this issue, eigenvector centrality mapping (ECM) method was applied to explore functional connectivity of resting-state in motor imagery training. As a data-driven analysis method, although ECM can assess the computational measurement of eigenvector centrality for capturing intrinsic neural architecture on a voxel-wise level without any prior assumptions, it is still limited in application for making pseudo enhancement in some nodes or zero centrality in all nodes. In this study, we proposed an improved ECM by adding threshold, dispersion coefficient, weighted coefficient and the initial parameters referring to Google Webpage search ranking algorithm, and applied the proposed ECM to functional connectivity measure of resting-state before and after motor imagery training. The proposed ECM showed the advantage of automatic discharge weak links and the enhancement in node ordering resolution comparing with the original ECM. The results from voxel-based comparison of the centrality between the resting-state after and before motor imagery training revealed that the significantly increased eigenvector centrality was detected in the precuneus and medial frontal gyrus for the experimental group while no significant alterations were found for the control group after training. These alterations may be related to the spatial information integration and inner state modulation of motor imagery training, and further provided new insights into the understanding of the neural mechanism underlying motor imagery training.