Detection of Nonlinear Coupling and its Application to Cardiorespiratory Interaction

We present here a modification of the Lagrangian measures technique, which allows a reliable detection of interdependency among simultaneous measurements of different variables. This method is applied to a simulated multivariate time series and to a bivariate cardiorespiratory signal. By using this methodology, it is possible to reveal a nonlinear interaction among cardiac and respiration rhythms in pathological conditions