Nonlinear Modeling of Dynamic Cerebral Autoregulation Using Recurrent Neural Networks

The function of the Cerebral Blood Flow Autoregulation (CBFA) system is to maintain a relatively constant flow of blood to the brain, in spite of changes in arterial blood pressure. A model that characterizes this system is of great use in understanding cerebral hemodynamics and would provide a pattern for evaluating different cerebrovascular diseases and complications. This work posits a non-linear model of the CBFA system through the evaluation of various types of neural networks that have been used in the field of systems identification. Four different architectures, combined with four learning methods were evaluated. The results were compared with the linear model that has often been used as a standard reference. The results show that the best results are obtained with the FeedForward Time Delay neural network, using the Levenberg-Marquardt learning algorithm, with an improvement of 24% over the linear model (p<0.05).

[1]  D H Evans,et al.  Assessment of the thigh cuff technique for measurement of dynamic cerebral autoregulation. , 2000, Stroke.

[2]  Rune Aaslid,et al.  Comparison of Flow and Velocity During Dynamic Autoregulation Testing in Humans , 1994, Stroke.

[3]  Peter Tiňo,et al.  Learning long-term dependencies is not as difficult with NARX recurrent neural networks , 1995 .

[4]  R. Panerai,et al.  Linear and nonlinear analysis of human dynamic cerebral autoregulation. , 1999, American journal of physiology. Heart and circulatory physiology.

[5]  R. Panerai.  Assessment of cerebral pressure autoregulation in humans - a review of measurement methods , 1998, Physiological measurement.

[6]  R. Panerai,et al.  Neural network modelling of dynamic cerebral autoregulation: assessment and comparison with established methods. , 2004, Medical engineering & physics.

[7]  J F Potter,et al.  Cerebral blood flow velocity response to induced and spontaneous sudden changes in arterial blood pressure. , 2001, American journal of physiology. Heart and circulatory physiology.

[8]  José Ramón Hilera González,et al.  Redes neuronales artificiales: fundamentos, modelos y aplicaciones , 1995 .

[9]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[10]  R. Panerai,et al.  The critical closing pressure of the cerebral circulation. , 2003, Medical engineering & physics.

[11]  Xiangdong He,et al.  A New Method for Identifying Orders of Input-Output Models for Nonlinear Dynamic Systems , 1993, 1993 American Control Conference.

[12]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[13]  Jose C. Principe,et al.  Neural and adaptive systems : fundamentals through simulations , 2000 .

[14]  D. H. Evans,et al.  Frequency-domain analysis of cerebral autoregulation from spontaneous fluctuations in arterial blood pressure , 1998, Medical and Biological Engineering and Computing.

[15]  Ronney B. Panerai,et al.  A Parametric Approach to Measuring Cerebral Blood Flow Autoregulation from Spontaneous Variations in Blood Pressure , 2004, Annals of Biomedical Engineering.

[16]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..

[17]  Peter Tiño,et al.  Learning long-term dependencies in NARX recurrent neural networks , 1996, IEEE Trans. Neural Networks.

[18]  Georgios D. Mitsis,et al.  Nonlinear modeling of the dynamic effects of arterial pressure and CO2 variations on cerebral blood flow in healthy humans , 2004, IEEE Trans. Biomed. Eng..

[19]  D H Evans,et al.  Cerebral autoregulation dynamics in premature newborns. , 1995, Stroke.

[20]  Richard P. White,et al.  Grading of cerebral dynamic autoregulation from spontaneous fluctuations in arterial blood pressure. , 1998, Stroke.

[21]  V. Z. Marmarelis,et al.  Modeling of Nonlinear Physiological Systems with Fast and Slow Dynamics. II. Application to Cerebral Autoregulation , 2002, Annals of Biomedical Engineering.

[22]  Jose C. Principe,et al.  Neural and Adaptive Systems: Fundamentals through Simulations with CD-ROM , 1999 .

[23]  D. Newell,et al.  Comparison of static and dynamic cerebral autoregulation measurements. , 1995, Stroke.

[24]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.