Identification of Non-stationary Dynamics in Physiological Recordings

We present a novel framework for the analysis of time series from dynamical systems which alternate between different operating modes. The method simultaneously segments and identifies the dynamical modes by using predictive models. In extension to previous approaches, it allows an identification of smooth transitions between successive modes. The method can be used for analysis, diagnosis, prediction, and control. In an application to EEG and respiratory data recorded from humans during afternoon naps, the obtained segmentations of the data agree with the sleep stage segmentation of a medical expert to a large extent. However, in contrast to the manual segmentation, our method does not require a-priori knowledge about physiology. Moreover, it has a high temporal resolution and reveals many previously unclassified details of the transitions. In particular, a parameter is found that is potentially useful for vigilance monitoring. We expect that the method will generally be useful for th..