Identification of nonstationary dynamics in physiological recordings

Abstract. We present a novel framework for the analysis of time series from dynamical systems that 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 transition 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 previously unclassified details of the transitions. In particular, a parameter is found that is potentially helpful for vigilance monitoring. We expect that the method will generally be useful for the analysis of nonstationary dynamical systems, which are abundant in medicine, chemistry, biology and engineering.

[1]  Klaus-Robert Müller,et al.  Analysis of switching dynamics with competing neural networks , 1995 .

[2]  A. Rechtschaffen,et al.  A manual of standardized terminology, technique and scoring system for sleep stages of human subjects , 1968 .

[3]  James D. Hamilton Time Series Analysis , 1994 .

[4]  Klaus Pawelzik,et al.  Hidden Markov mixtures of experts with an application to EEG recordings from sleep , 1999 .

[5]  Athanasios Kehagias,et al.  Time-Series Segmentation Using Predictive Modular Neural Networks , 1997, Neural Computation.

[6]  James P. Crutchfield,et al.  Geometry from a Time Series , 1980 .

[7]  M Lambertz,et al.  Reticular formation of the lower brainstem. A common system for cardiorespiratory and somatomotor functions: discharge patterns of neighboring neurons influenced by somatosensory afferents. , 1983, Journal of the autonomic nervous system.

[8]  Klaus-Robert Müller,et al.  Annealed Competition of Experts for a Segmentation and Classification of Switching Dynamics , 1996, Neural Computation.

[9]  Klaus Pawelzik,et al.  Segmentation and Identification of Switching Dynamics with Competing Neural Networks , 1994 .

[10]  H M Praetorius,et al.  Adaptive segmentation of EEG records: a new approach to automatic EEG analysis. , 1977, Electroencephalography and clinical neurophysiology.

[11]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[12]  E. Wolpert A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. , 1969 .

[13]  T. Bohlin Analysis of EEG signals with changing spectra using a short-word Kalman estimator , 1977 .

[14]  K. Pawelzikzy,et al.  Improving Short-term Prediction with Competing Experts , 1995 .

[15]  Andreas S. Weigend,et al.  Taking time seriously: hidden Markov experts applied to financial engineering , 1997, Proceedings of the IEEE/IAFE 1997 Computational Intelligence for Financial Engineering (CIFEr).

[16]  F. Takens Detecting strange attractors in turbulence , 1981 .

[17]  J. S. Barlow,et al.  Computerized EEG pattern classification by adaptive segmentation and probability density function classification. Clinical evaluation. , 1985, Electroencephalography and clinical neurophysiology.

[18]  Steven J. Nowlan,et al.  Mixtures of Controllers for Jump Linear and Non-Linear Plants , 1993, NIPS.

[19]  Yoshua Bengio,et al.  An Input Output HMM Architecture , 1994, NIPS.

[20]  K. Pawelzikzy,et al.  Analysing Physiological Data from the Wake-sleep State Transition with Competing Predictors , 1995 .

[21]  G. Micheletti,et al.  Brainstem Control of Wakefulness and Sleep, M Steriade, RW McCarley. Plenum, San Diego (1990), 1 vol, 499 p. , 1994 .

[22]  J Rittweger,et al.  Common slow modulation of respiration, arterial blood pressure and cortical activity during sleep onset while napping. , 1999, Clinical physiology.

[23]  Kaneko,et al.  Chaotic but regular posi-nega switch among coded attractors by cluster-size variation. , 1989, Physical review letters.

[24]  Geoffrey E. Hinton,et al.  Adaptive Mixtures of Local Experts , 1991, Neural Computation.

[25]  Ashok N. Srivastava,et al.  Nonlinear gated experts for time series: discovering regimes and avoiding overfitting , 1995, Int. J. Neural Syst..

[26]  M. Steriade,et al.  Brainstem Control of Wakefulness and Sleep , 1990, Springer US.

[27]  Gottfried Mayer-Kress,et al.  Dimensions and Entropies in Chaotic Systems , 1986 .

[28]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[29]  P.A. Karjalainen,et al.  Estimation of event-related synchronization changes by a new TVAR method , 1997, IEEE Transactions on Biomedical Engineering.

[30]  Rose,et al.  Statistical mechanics and phase transitions in clustering. , 1990, Physical review letters.

[31]  J. Trinder,et al.  Respiratory instability during sleep onset. , 1992, Journal of applied physiology.

[32]  L. Glass,et al.  Oscillation and chaos in physiological control systems. , 1977, Science.

[33]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[34]  M Lambertz,et al.  Cardiac rhythmic patterns in neuronal activity are related to the firing rate of the neurons: I. Brainstem reticular neurons of dogs. , 1995, Journal of the autonomic nervous system.

[35]  H. Kantz,et al.  Nonlinear time series analysis , 1997 .

[36]  Jens Kohlmorgen Analyse schaltender und driftender Dynamik mit neuronalen Netzen , 1998 .

[37]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[38]  W. Herrmann,et al.  Kritische Bemerkungen zu den Regeln von Rechtschaffen und Kales über die visuelle Auswertung von EEG-Schlafableitungen , 1982 .