Estimation of Coupling and Directionality between Signals Applied to Physiological Uterine EMG Model and Real EHG Signals

Several measures have been proposed to detect the strength and direction of relationships between biosignals. In this paper we study two nonlinear methods that are widely used in functional and effective connectivity analysis: nonlinear correlation coefficient (h 2) and general synchronization (H). The performance of both methods is tested on two dimensional coupled synthetic nonlinear Rossler model. The best method for these synthetic Rossler signals is then applied for the first time to signals generated by a physiological uterine EMG model. It was then applied to real uterine EMG recorded during pregnancy and labor in order to compare their synchronization and propagation direction. The best performing method was tested on the physiological model data, and the results obtained on real signals are encouraging. It may possibly be of use for solving the open questions of the relationships between uterine EMG at different places on the uterus, and may provide a way to localize their sources.