Dynamic modeling of spinal electromyographic activity during various conditions

The surface electromyographic (sEMG) signals recorded along the spine during a rocking motion created by feedback coupling between the motion of the spine and the efferent nerve fibers at the dural attachment points are giving various linear dynamical models of the ARIMA type. The most significant dynamical phenomenon is the nonlinear switching among the various linear models. The switchings represent transitions among qualitatively different modes of the motion of the spine, referred to as Levels 1, 2, 3. Statistical analysis reveals a definite relationship between the qualitatively assessed levels and the various quantitatively relevant models. Finally, it is shown that the higher the level of care, the more reliable the model, that is, the better the model is able to predict the motion as specified by the sEMG signal.