A Novel Approach for Modeling Neural Responses to Joint Perturbations Using the NARMAX Method and a Hierarchical Neural Network

The human nervous system is an ensemble of connected neuronal networks. Modeling and system identification of the human nervous system helps us understand how the brain processes sensory input and controls responses at the systems level. This study aims to propose an advanced approach based on a hierarchical neural network and non-linear system identification method to model neural activity in the nervous system in response to an external somatosensory input. The proposed approach incorporates basic concepts of Non-linear AutoRegressive Moving Average Model with eXogenous input (NARMAX) and neural network to acknowledge non-linear closed-loop neural interactions. Different from the commonly used polynomial NARMAX method, the proposed approach replaced the polynomial non-linear terms with a hierarchical neural network. The hierarchical neural network is built based on known neuroanatomical connections and corresponding transmission delays in neural pathways. The proposed method is applied to an experimental dataset, where cortical activities from ten young able-bodied individuals are extracted from electroencephalographic signals while applying mechanical perturbations to their wrist joint. The results yielded by the proposed method were compared with those obtained by the polynomial NARMAX and Volterra methods, evaluated by the variance accounted for (VAF). Both the proposed and polynomial NARMAX methods yielded much better modeling results than the Volterra model. Furthermore, the proposed method modeled cortical responded with a mean VAF of 69.35% for a three-step ahead prediction, which is significantly better than the VAF from a polynomial NARMAX model (mean VAF 47.09%). This study provides a novel approach for precise modeling of cortical responses to sensory input. The results indicate that the incorporation of knowledge of neuroanatomical connections in building a realistic model greatly improves the performance of system identification of the human nervous system.

[1]  a.R.V.,et al.  Human Neuroanatomy , 1954, Neurology.

[2]  J. Schoukens,et al.  First estimates of Wiener and Hammerstein systems using multisine excitation , 2001, IMTC 2001. Proceedings of the 18th IEEE Instrumentation and Measurement Technology Conference. Rediscovering Measurement in the Age of Informatics (Cat. No.01CH 37188).

[3]  C G Billings,et al.  The prediction of in-flight hypoxaemia using non-linear equations. , 2013, Respiratory medicine.

[4]  M Scherg,et al.  Somatotopy of human hand somatosensory cortex revealed by dipole source analysis of early somatosensory evoked potentials and 3D-NMR tomography. , 1995, Electroencephalography and clinical neurophysiology.

[5]  Sheng Chen,et al.  Orthogonal least squares methods and their application to non-linear system identification , 1989 .

[6]  Andreas Daffertshofer,et al.  A General Approach for Quantifying Nonlinear Connectivity in the Nervous System Based on Phase Coupling , 2016, Int. J. Neural Syst..

[7]  Tzyy-Ping Jung,et al.  Independent Component Analysis of Electroencephalographic Data , 1995, NIPS.

[8]  Rik Pintelon,et al.  Identification of a Wiener–Hammerstein system using the polynomial nonlinear state space approach , 2009 .

[9]  Stephen A. Billings,et al.  Identification of nonlinear time-varying systems using an online sliding-window and common model structure selection (CMSS) approach with applications to EEG , 2016, Int. J. Syst. Sci..

[10]  O. Nelles.  Nonlinear System Identification: From Classical Approaches to Neural Networks and Fuzzy Models , 2000 .

[11]  Karl J. Friston,et al.  a K.E. Stephan, a R.B. Reilly, , 2007 .

[12]  Mariano Sigman,et al.  CUDAICA: GPU Optimization of Infomax-ICA EEG Analysis , 2012, Comput. Intell. Neurosci..

[13]  Frans C. T. van der Helm,et al.  Frequency Domain Characterization of the Somatosensory Steady State Response in Electroencephalography , 2015 .

[14]  Sheng Chen,et al.  Representations of non-linear systems: the NARMAX model , 1989 .

[15]  Roger W. Brockett.  Volterra series and geometric control theory , 1976, Autom..

[16]  Alfred C. Schouten,et al.  Stretch Evoked Potentials in Healthy Subjects and After Stroke: A Potential Measure for Proprioceptive Sensorimotor Function , 2015, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[17]  M. Breakspear,et al.  Multi-frequency phase locking in human somatosensory cortex. , 2011, Progress in biophysics and molecular biology.

[18]  G Cheron,et al.  Somatosensory evoked potentials to finger stimulation in healthy octogenarians and in young adults: wave forms, scalp topography and transit times of parietal and frontal components. , 1980, Electroencephalography and clinical neurophysiology.

[19]  David T. Westwick,et al.  Identification of Hammerstein models with cubic spline nonlinearities , 2004, IEEE Transactions on Biomedical Engineering.

[20]  Karl J. Friston.  The labile brain. I. Neuronal transients and nonlinear coupling. , 2000, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[21]  M. Breakspear.  Dynamic models of large-scale brain activity , 2017, Nature Neuroscience.

[22]  Karl J. Friston,et al.  Population dynamics: Variance and the sigmoid activation function , 2008, NeuroImage.

[23]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[24]  Robert Oostenveld,et al.  FieldTrip: Open Source Software for Advanced Analysis of MEG, EEG, and Invasive Electrophysiological Data , 2010, Comput. Intell. Neurosci..

[25]  Johan Schoukens,et al.  Modeling the Nonlinear Cortical Response in EEG Evoked by Wrist Joint Manipulation , 2018, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[26]  Fritz Buchthal,et al.  Evoked action potentials and conduction velocity in human sensory nerves , 1966 .

[27]  Rohitash Chandra,et al.  Multi-step-ahead chaotic time series prediction using coevolutionary recurrent neural networks , 2016, 2016 IEEE Congress on Evolutionary Computation (CEC).

[28]  Alfred C. Schouten,et al.  A Generalized Coherence Framework for Detecting and Characterizing Nonlinear Interactions in the Nervous System , 2016, IEEE Transactions on Biomedical Engineering.

[29]  F. V. D. van der Helm,et al.  Quantifying Nonlinear Contributions to Cortical Responses Evoked by Continuous Wrist Manipulation , 2017, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[30]  D. Westwick,et al.  Separable Least Squares Identification of Nonlinear Hammerstein Models: Application to Stretch Reflex Dynamics , 2001, Annals of Biomedical Engineering.