Recurrent neural networks employing Lyapunov exponents for analysis of ECG signals

An approach based on the consideration that electrocardiogram (ECG) signals are chaotic signals was presented for automated diagnosis of electrocardiographic changes. This consideration was tested successfully using the nonlinear dynamics tools, like the computation of Lyapunov exponents. Recurrent neural network (RNN) was implemented and used as basis for detection of variabilities of ECG signals. Four types of ECG beats (normal beat, congestive heart failure beat, ventricular tachyarrhythmia beat, atrial fibrillation beat) obtained from the PhysioBank database were classified. Decision making was performed in two stages: computing features which were then input into the RNN and classification using the RNN trained with the Levenberg-Marquardt algorithm. The research demonstrated that the Lyapunov exponents are the features which are well representing the ECG signals and the RNN trained on these features achieved high classification accuracies.

[1]  Gustavo Deco,et al.  Dynamics extraction in multivariate biomedical time series , 1998, Biological Cybernetics.

[2]  Yasser M. Kadah,et al.  Study of features based on nonlinear dynamical modeling in ECG arrhythmia detection and classification , 2002, IEEE Transactions on Biomedical Engineering.

[3]  H. Abarbanel,et al.  LYAPUNOV EXPONENTS IN CHAOTIC SYSTEMS: THEIR IMPORTANCE AND THEIR EVALUATION USING OBSERVED DATA , 1991 .

[4]  Bidyut Baran Chaudhuri,et al.  Efficient training and improved performance of multilayer perceptron in pattern classification , 2000, Neurocomputing.

[5]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[6]  Jiann-Shing Shieh,et al.  Intracranial pressure model in intensive care unit using a simple recurrent neural network through time , 2004, Neurocomputing.

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

[8]  Roberto Battiti,et al.  First- and Second-Order Methods for Learning: Between Steepest Descent and Newton's Method , 1992, Neural Computation.

[9]  Mahantapas Kundu,et al.  Knowledge-based ECG interpretation: a critical review , 2000, Pattern Recognit..

[10]  B. A. Harvey,et al.  Neural network-based EKG pattern recognition , 2002 .

[11]  Lalit Gupta,et al.  Classification of temporal sequences via prediction using the simple recurrent neural network , 2000, Pattern Recognit..

[12]  Karsten Sternickel,et al.  Automatic pattern recognition in ECG time series , 2002, Comput. Methods Programs Biomed..

[13]  Eckmann,et al.  Liapunov exponents from time series. , 1986, Physical review. A, General physics.

[14]  A. Casaleggio,et al.  Estimation of Lyapunov exponents of ECG time series—The influence of parameters , 1997 .

[15]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[16]  Donald C. Wunsch,et al.  Recurrent neural network based prediction of epileptic seizures in intra- and extracranial EEG , 2000, Neurocomputing.

[17]  Martin Casdagli,et al.  Nonlinear prediction of chaotic time series , 1989 .

[18]  Elif Derya Übeyli,et al.  Recurrent neural networks employing Lyapunov exponents for EEG signals classification , 2005, Expert Syst. Appl..

[19]  Elif Derya Übeyli.  Usage of eigenvector methods in implementation of automated diagnostic systems for ECG beats , 2008, Digit. Signal Process..

[20]  S. T. Hamde,et al.  Feature extraction from ECG signals using wavelet transforms for disease diagnostics , 2002, Int. J. Syst. Sci..

[21]  Pineda,et al.  Generalization of back-propagation to recurrent neural networks. , 1987, Physical review letters.

[22]  L. Tsimring,et al.  The analysis of observed chaotic data in physical systems , 1993 .

[23]  A. Wolf,et al.  Determining Lyapunov exponents from a time series , 1985 .

[24]  Elif Derya Übeyli,et al.  ECG beat classifier designed by combined neural network model , 2005, Pattern Recognit..

[25]  Nurettin Acir.  Classification of ECG beats by using a fast least square support vector machines with a dynamic programming feature selection algorithm , 2005, Neural Computing & Applications.

[26]  Elif Derya Übeyli.  Detection of electrocardiogram beats using a fuzzy similarity index , 2007, Expert Syst. J. Knowl. Eng..

[27]  R. Schiffer,et al.  Recurrent neural network-based approach for early recognition of Alzheimer's disease in EEG , 2001, Clinical Neurophysiology.

[28]  Simon Haykin,et al.  Detection of signals in chaos , 1995, Proc. IEEE.

[29]  Lalit Gupta,et al.  Investigating the prediction capabilities of the simple recurrent neural network on real temporal sequences , 2000, Pattern Recognit..

[30]  Jürgen Fell,et al.  Nonlinear analysis of continuous ECG during sleep II. Dynamical measures , 2000, Biological Cybernetics.

[31]  Elif Derya Übeyli,et al.  An expert system for detection of electrocardiographic changes in patients with partial epilepsy using wavelet‐based neural networks , 2005, Expert Syst. J. Knowl. Eng..

[32]  Elif Derya Übeyli.  Implementing wavelet transform/mixture of experts network for analysis of electrocardiogram beats , 2008, Expert Syst. J. Knowl. Eng..

[33]  Jeffrey M. Hausdorff,et al.  Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol , 2000 .

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

[35]  Sawada,et al.  Measurement of the Lyapunov spectrum from a chaotic time series. , 1985, Physical review letters.

[36]  K. Narayanan,et al.  On the evidence of deterministic chaos in ECG: Surrogate and predictability analysis. , 1998, Chaos.

[37]  Michael G. Strintzis,et al.  ECG pattern recognition and classification using non-linear transformations and neural networks: A review , 1998, Int. J. Medical Informatics.