A dynamic recurrent neural network for multiple muscles electromyographic mapping to elevation angles of the lower limb in human locomotion

This paper describes the use of a dynamic recurrent neural network (DRNN) for simulating lower limb coordination in human locomotion. The method is based on mapping between the electromyographic signals (EMG) from six muscles and the elevation angles of the three main lower limb segments (thigh, shank and foot). The DRNN is a fully connected network of 35 hidden units taking into account the temporal relationships history between EMG and lower limb kinematics. Each EMG signal is sent to all 35 units, which converge to three outputs. Each output neurone provides the kinematics of one lower limb segment. The training is supervised, involving learning rule adaptations of synaptic weights and time constant of each unit. Kinematics of the locomotor movements were recorded and analysed using the opto-electronic ELITE system. Comparative analysis of the learning performance with different types of output (position, velocity and acceleration) showed that for common gait mapping velocity data should be used as output, as it is the best compromise between asymptotic error curve, rapid convergence and avoidance of bifurcation. Reproducibility of the identification process and biological plausibility were high, indicating that the DRNN may be used for understanding functional relationships between multiple EMG and locomotion. The DRNN might also be of benefit for prosthetic control.

[1]  Henk A. K. Mastebroek,et al.  Plausible neural networks for biological modelling , 2001 .

[2]  Jessica Lowell.  Neural Network , 2001 .

[3]  N. A. Borghese,et al.  Kinematic determinants of human locomotion. , 1996, The Journal of physiology.

[4]  W. T. Thach,et al.  Cerebellar ataxia: torque deficiency or torque mismatch between joints? , 2000, Journal of neurophysiology.

[5]  Luís B. Almeida,et al.  Speeding up Backpropagation , 1990 .

[6]  F. Mussa-Ivaldi,et al.  The motor system does not learn the dynamics of the arm by rote memorization of past experience. , 1997, Journal of neurophysiology.

[7]  F. Lacquaniti,et al.  Motor patterns for human gait: backward versus forward locomotion. , 1998, Journal of neurophysiology.

[8]  J. F. Soechting,et al.  Evaluating an integrated musculoskeletal model of the human arm. , 1997, Journal of biomechanical engineering.

[9]  D J Ostry,et al.  Compensation for interaction torques during single- and multijoint limb movement. , 1999, Journal of neurophysiology.

[10]  Gaetan Libert,et al.  Improved Identification of the Human Shoulder Kinematics with Muscle Biological Filters , 1997, AIME.

[11]  Yasuharu Koike,et al.  Estimation of dynamic joint torques and trajectory formation from surface electromyography signals using a neural network model , 1995, Biological Cybernetics.

[12]  F. Lacquaniti,et al.  Motor Patterns in Walking. , 1999, News in physiological sciences : an international journal of physiology produced jointly by the International Union of Physiological Sciences and the American Physiological Society.

[13]  Richard R Neptune,et al.  Biomechanics and muscle coordination of human walking: part II: lessons from dynamical simulations and clinical implications. , 2003, Gait & posture.

[14]  Patrick Brézillon,et al.  Lecture Notes in Artificial Intelligence , 1999 .

[15]  F. Lacquaniti,et al.  Basal ganglia and gait control: apomorphine administration and internal pallidum stimulation in Parkinson’s disease , 1999, Experimental Brain Research.

[16]  G. Cheron,et al.  A dynamic neural network identification of electromyography and arm trajectory relationship during complex movements , 1996, IEEE Transactions on Biomedical Engineering.

[17]  C L Vaughan,et al.  A neural network representation of electromyography and joint dynamics in human gait. , 1993, Journal of biomechanics.

[18]  A. G. Feldman,et al.  Recent Tests of the Equilibrium-Point Hypothesis (λ Model) , 1998 .

[19]  F. Zajac,et al.  Contributions of the individual ankle plantar flexors to support, forward progression and swing initiation during walking. , 2001, Journal of biomechanics.

[20]  G. Ermentrout.  Dynamic patterns: The self-organization of brain and behavior , 1997 .

[21]  Arthur D Kuo,et al.  The relative roles of feedforward and feedback in the control of rhythmic movements. , 2002, Motor control.

[22]  William Craelius,et al.  The bionic man: restoring mobility. , 2002, Science.

[23]  Jean-Philippe Draye,et al.  Recurrent Neural Networks: Properties and Models , 2001 .

[24]  Francesco Lacquaniti,et al.  Early emergence of temporal co-ordination of lower limb segments elevation angles in human locomotion , 2001, Neuroscience Letters.

[25]  W. Herzog,et al.  Prediction of dynamic tendon forces from electromyographic signals: An artificial neural network approach , 1997, Journal of Neuroscience Methods.

[26]  J. McIntyre,et al.  Servo Hypotheses for the Biological Control of Movement. , 1993, Journal of motor behavior.

[27]  Jean-Philippe Draye,et al.  Self-selected modular recurrent neural networks with postural and inertial subnetworks applied to complex movements , 2002, Biological Cybernetics.

[28]  Guy Cheron,et al.  Adaptative time constant improved the prediction capacity of recurrent neural network , 1995 .

[29]  Barak A. Pearlmutter.  Learning State Space Trajectories in Recurrent Neural Networks , 1989, Neural Computation.

[30]  Francesco Lacquaniti,et al.  Development of a kinematic coordination pattern in toddler locomotion: planar covariation , 2001, Experimental Brain Research.

[31]  Giancarlo Ferrigno,et al.  Elite: A Digital Dedicated Hardware System for Movement Analysis Via Real-Time TV Signal Processing , 1985, IEEE Transactions on Biomedical Engineering.

[32]  Ferdinando A. Mussa-Ivaldi,et al.  Emergence of symmetric, modular, and reciprocal connections in recurrent networks with Hebbian learning , 1999, Biological Cybernetics.

[33]  Yasuharu Koike,et al.  Estimation of Arm Posture in 3D-Space from Surface EMG Signals Using a Neural Network Model (Special Issue on Neurocomputing) , 1994 .

[34]  D J Ostry,et al.  Are complex control signals required for human arm movement? , 1998, Journal of neurophysiology.

[35]  A. Opstal.  Dynamic Patterns: The Self-Organization of Brain and Behavior , 1995 .

[36]  A PearlmutterBarak.  Learning state space trajectories in recurrent neural networks , 1989 .

[37]  Gaetan Libert,et al.  Dynamic recurrent neural networks: a dynamical analysis , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[38]  A Berthoz,et al.  Spatial, not temporal cues drive predictive orienting movements during navigation: a virtual reality study , 2000, Neuroreport.

[39]  A. G. Feldman,et al.  Recent tests of the equilibrium-point hypothesis (lambda model). , 1998, Motor control.

[40]  R. Eckmiller.  Advanced neural computers , 1990 .

[41]  F. Lacquaniti,et al.  Individual characteristics of human walking mechanics , 1998, Pflügers Archiv.

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

[43]  Kenji Doya,et al.  Universality of Fully-Connected Recurrent Neural Networks , 1993 .

[44]  R J Full,et al.  How animals move: an integrative view. , 2000, Science.

[45]  F. Lacquaniti,et al.  Kinematic coordination in human gait: relation to mechanical energy cost. , 1998, Journal of neurophysiology.

[46]  P. Komi,et al.  Interaction between pre-landing activities and stiffness regulation of the knee joint musculoskeletal system in the drop jump: implications to performance , 2002, European Journal of Applied Physiology.

[47]  F A Mussa-Ivaldi,et al.  Central representation of time during motor learning. , 1999, Proceedings of the National Academy of Sciences of the United States of America.

[48]  J. Kelso,et al.  Intentional switching between patterns of bimanual coordination depends on the intrinsic dynamics of the patterns. , 1990, Journal of motor behavior.

[49]  Jack M. Winters,et al.  Modeling Musculoskeletal Movement Systems: Joint and Body Segmental Dynamics, Musculoskeletal Actuation, and Neuromuscular Control , 1990 .

[50]  G. Gottlieb,et al.  A Computational Model of the Simplest Motor Program. , 1993, Journal of motor behavior.