Abstract
We propose a novel interpretation and usage of Neural Network (NN) in modeling physiological signals, which are allowed to be nonlinear and/or nonstationary. The method consists of training a NN for the k-step prediction of a physiological signal, and then examining the connection-weight-space (CWS) of the NN to extract information about the signal generator mechanism. We define a novel feature, Normalized Vector Separation (γ ij ), to measure the separation of two arbitrary states “i” and “j” in the CWS and use it to track the state changes of the generating system. The performance of the method is examined via synthetic signals and clinical EEG. Synthetic data indicates that γ ij can track the system down to a SNR of 3.5 dB. Clinical data obtained from three patients undergoing carotid endarterectomy of the brain showed that EEG could be modeled (within a root-means-squared-error of 0.01) by the proposed method, and the blood perfusion state of the brain could be monitored via γ ij , with small NNs having no more than 21 connection weight altogether.
Similar content being viewed by others
References
Cisse Y, KinouchiY, Nagashino H, Akutagawa M (2002) BP neural networks approach for identifying biological signal source in circadian data fluctuations. IEICE Trans Inf Syst E85-D(3):568–576
Elbert T, Ray WJ, Kowalik ZJ, Skinner JE, Graf KE, Birbaumer N (1994) Chaos and physiology: deterministic chaos in excitable cell assemblies. Physiol Rev 74:1–47
Funahashi K (1989) On the approximate realization of continuous mappings by neural networks. Neural Netw 2:183–192
Gorman RP, Sejnowski TJ (1988) Analysis of hidden units in a layered network trained to classify sonar targets. Neural Netw 1:75–89
Hinton GE (1992) How neural networks learn from experience. Sci Am 267:105–109
Hornik K, Stinchombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Netw 2:359–366
Japer H (1958) Ten-twenty electrode system of the international federation. Electroencephalogr Clin Neurophysiol 10:371–375
Jenkins GM, Watts DG (1969) Spectral analysis and its applications, Holden-Day series in time series analysis. Holden-Day, London
Kay SM (1988) Modern spectral estimation: theory and application. Prentice-Hall, NJ
Kirby SD, Eng P, Danter W, George CF, Francovic T, Ruby RR, Ferguson KA (1999) Neural network prediction of obstructive sleep apnea from clinical criteria. Chest 116:409–415
Li YC, Liu L, Chiu WT, Jian WS (2000) Neural network modeling for surgical decisions on traumatic brain injury patients. Med Infor 57:1–9
Mackey C, Glass L (1977) Oscillation and chaos in physiological control systems. Science 197:287–289
Manuca R, Casdagli MC, Savit RS (1998) Nonstationarity in epileptic EEG and implications for neural dynamics. Math Biosci 147(1):1–22
Murray JD (1993) Mathematical biology, 2nd edn. Springer, Berlin
Oppenheim AV, Schafer RW (1989) Discrete-time signal processing. Prentice-Hall, Englewood Cliffs, NJ
Oud M (2002) Internal-state analysis in a layered artificial neural network trained to categorize lung sounds, man and cybernetics, Part A. IEEE Trans Syst 32:757–760
Packard NJ, Crutchfield JP, Farmer JD, Shaw RS (1980) Geometry from a time series. Phys Rev Lett 45:712–716
Rieke C, Mormann F, Andrzejak RG, Kreuz T, David P, Elger CE, Lehnertz K (2003) Discerning nonstationarity from nonlinearity in seizure-free and preseizure EEG recordings from epilepsy patients. IEEE Trans Biomed Eng 50:634–639
Rumelhart DE, Hinton GE, Williams RJ (1986) Learning internal representation by error propagation. In: Rumelhart DE, McClelland JL, the PDP Research Group (eds) Parallel distributed processing. MIT press, Cambridge, pp 318–362
Takens F (1981) Detecting strange attractors in turbulence. In: Rand DA, Young LS (eds) Lecture notes in math, vol 898. Springer-Verlag, Berlin
Varady P, Micsik T, Benedek S, Benyo Z (2002) A novel method for the detection of apnea and hypopnea events in respiration signals. IEEE Trans Biomed Eng 49:936–942
Watanabe E, Nakasako N, Mitani Y (1997) A prediction method of non-stationary time series data by using a modular structured neural network. IEICE Trans Fundamentals E80-A(6):971–976
Yu H, Bang S (1997) An improved time series prediction by applying the layer-by-layer learning method to FIR neural networks. Trans Soc Comput Simul Int 14:1717–1729
Acknowledgments
This work was partly supported by Grants-in-Aid for Scientific Research #16560353 and #16700440 from Japan Society of Promotion of Science. The authors would like to thank The University of Queensland, Australia for access to facilities and hosting the first author as a Research Scholar during the work. Authors also thank Dr Fumio Shichijo (Department of Neurology, Suzue hospital, Tokushima, Japan) for providing them with clinical data for the work of this project.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Emoto, T., Akutagawa, M., Abeyratne, U.R. et al. Tracking the states of a nonlinear and nonstationary system in the weight-space of artificial neural networks. Med Bio Eng Comput 44, 146–159 (2006). https://doi.org/10.1007/s11517-005-0019-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11517-005-0019-8