A Novel Delay Estimation Method for Improving Corticomuscular Coherence in Continuous Synchronization Events

Jinbiao Liu, Gansheng Tan, Yixuan Sheng, Yina Wei, Honghai Liu

Abstract: Objective : While the corticomuscularcoupling between motor cortex and muscle tissue has received considerable attention, which is typically quantitative measure to evaluate neural signals synchronization in the motor control system, little work has been published regarding the effect of underlying delay of two coupled physiological signals on coherence. Methods : In this study, we developed a novel delay estimation method, named rate of voxels change (RVC), detecting time delay in two coupled physiological signals. Based on RVC framework, delay compensation was used to adjust magnitude squared coherence (MSC) image. To illustrate the effectiveness of the RVC method, we compared the estimated delays and the adjusted MSC results based on RVC method and corticomuscular coherence with time lag (CMCTL) method. Results : The simulation results suggested that RVC method was not only superior to the CMCTL method in estimating different time delays, but also has better optimization effect on MSC image. The experimental results further confirmed that delay estimated by the proposed RVC method was more in line with the underlying physiology (controls: $22.8\ {{\mathbf ms}}$ vs patients: $34.5\ {{\mathbf ms}}$ ). Meanwhile, RVC-based delay compensation could significantly optimize the MSC of specific regions. Significance : This study proved that RVC has remarkably higher reliability in detecting time delay between coupled neurophysiological signals, and the application of RVC was an improvement on the previous studies that mainly focused on biased MSC estimation.