Muscle activity prediction using wavelet neural network

The purpose of this study was to develop a multi dimensional wavelet neural network (WNN) approach in order to predict human lower extremity muscle activities based on ground reaction forces (GRF) and joint angles. For this purpose, four healthy subjects were taken from a previous study. The proposed approach consisted of two main parts: 1) input variable selection (IVS) and 2) network training. First, mutual information (MI) method was used to determine nine inputs including three dimensional GRFs and six joint angles as WNN inputs to predict seven number of outputs. The network was trained based on batch descent gradient algorithm using inter subject data space which provided by leave-one-out (LOO) technique. The WNN predictions for the left-out subject were compared with inverse dynamics calculations based on root mean square error (RMSE) and its percentage as well as Pearson correlation analysis (p). Results showed that multi dimensional WNN was capable to model the highly nonlinear relationship between GRF and joint angles as inputs and muscle activities as outputs.

[1]  Mustafa Yilmaz,et al.  Classification of EMG signals using wavelet neural network , 2006, Journal of Neuroscience Methods.

[2]  Abdulhamit Subasi,et al.  Wavelet neural network classification of EEG signals by using AR model with MLE preprocessing , 2005, Neural Networks.

[3]  Shahriar Gharibzadeh,et al.  Modeling force-velocity relation in skeletal muscle isotonic contraction using an artificial neural network , 2007, Biosyst..

[4]  M. Ulagammai,et al.  Application of bacterial foraging technique trained artificial and wavelet neural networks in load forecasting , 2007, Neurocomputing.

[5]  Holger R. Maier,et al.  Efficient selection of inputs for artificial neural network models , 2005 .

[6]  Holger R. Maier,et al.  Review of Input Variable Selection Methods for Artificial Neural Networks , 2011 .

[7]  J. Romkes,et al.  Changes in muscle activity in children with hemiplegic cerebral palsy while walking with and without ankle-foot orthoses. , 2006, Gait & posture.

[8]  Jorge Martins,et al.  An efficient muscle fatigue model for forward and inverse dynamic analysis of human movements , 2011 .

[9]  E. Kellis,et al.  The effects of the antagonist muscle force on intersegmental loading during isokinetic efforts of the knee extensors. , 1999, Journal of biomechanics.

[10]  Walter Herzog,et al.  Model-based estimation of muscle forces exerted during movements. , 2007, Clinical biomechanics.

[11]  Changmok Choi,et al.  Real-time pinch force estimation by surface electromyography using an artificial neural network. , 2010, Medical engineering & physics.

[12]  Scott L Delp,et al.  Generating dynamic simulations of movement using computed muscle control. , 2003, Journal of biomechanics.

[13]  W I Schöllhorn,et al.  Applications of artificial neural nets in clinical biomechanics. , 2004, Clinical biomechanics.

[14]  W Herzog,et al.  Dynamic muscle force predictions from EMG: an artificial neural network approach. , 1999, Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology.

[15]  Ping Zhang,et al.  Fuzzy Wavelet Neural Networks for City Electric Energy Consumption Forecasting , 2012 .

[16]  D B Chaffin,et al.  A back-propagation neural network model of lumbar muscle recruitment during moderate static exertions. , 1995, Journal of biomechanics.

[17]  M L Audu,et al.  A dynamic optimization technique for predicting muscle forces in the swing phase of gait. , 1987, Journal of biomechanics.

[18]  T. D.  A DYNAMIC OPTIMIZATION TECHNIQUE FOR PREDICTING MUSCLE FORCES IN THE SWING PHASE OF GAIT , 2003 .

[19]  Gaviphat Lekutai,et al.  Adaptive Self-Tuning Neuro Wavelet Network Controllers , 1997 .

[20]  T Chau,et al.  A review of analytical techniques for gait data. Part 2: neural network and wavelet methods. , 2001, Gait & posture.

[21]  Qinghua Zhang,et al.  Wavelet networks , 1992, IEEE Trans. Neural Networks.

[22]  M. Pandy,et al.  Contributions of individual muscles to hip joint contact force in normal walking. , 2010, Journal of biomechanics.

[23]  D B Chaffin,et al.  A neural network model for simulation of torso muscle coordination. , 1997, Journal of biomechanics.

[24]  Afshin Pourtaghi.  Wavelet Neural Network and Wavenet Performance Evaluation in Hydrodynamic Force Prediction due to Waves on Vertical Cylinders , 2012 .

[25]  C. Nester,et al.  Effect of rollover footwear on metabolic cost of ambulation, lower limb kinematics, kinetics, and EMG related muscle activity during walking , 2012, Journal of Foot and Ankle Research.

[26]  Kwon Son,et al.  Trunk and lower extremity muscle activity in seated weight-moving tasks: Three-dimensional analyses of intersubject and intertask differences , 1993 .

[27]  I Jonkers,et al.  A physiology-based inverse dynamic analysis of human gait using sequential convex programming: a comparative study , 2012, Computer methods in biomechanics and biomedical engineering.

[28]  S. Delp,et al.  Predicting the metabolic cost of incline walking from muscle activity and walking mechanics. , 2012, Journal of biomechanics.

[29]  Otman Basir,et al.  A neural network model for reconstructing EMG signals from eight shoulder muscles: consequences for rehabilitation robotics and biofeedback. , 2006, Journal of biomechanics.