Feasibility of using an artificial neural network model to estimate the elbow flexion force from mechanomyography

The goal of this study was to demonstrate the feasibility of using artificial neural network (ANN) models to estimate the elbow flexion forces from mechanomyography (MMG) under isometric muscle contraction and compare the performance of the ANN models with the performance from multiple linear regression (MLR) models. Five participants (mean±SD age=25.4±2.96 yrs) performed ten predefined and ten randomly ordered elbow flexions from 0% to 80% maximal voluntary contractions (MVCs). The MMG signals were recorded from the biceps brachii (BR) and brachioradialis (BRD), both of which contribute to elbow flexion. Inputs into the model included the root-mean-square (RMS), a temporal characterization feature, which resulted in a slightly higher signal-to-noise ratio (SNR) than when using the mean absolute value (MAV), and the zero-crossing (ZC) as spectral characterization features. Additionally, how the RMS and the ZC as model inputs affected the estimation accuracy was investigated. A cross-subject validation test was performed to determine if the established model of one subject could be applied to another subject. It was observed that the ANN model provided a more accurate estimation based on the values of the normalized root mean square error (NRMSE=0.141±0.023) and the cross-correlation coefficient (CORR=0.883±0.030) than the estimations from the MLR model (NRMSE=0.164±0.030, CORR=0.846±0.033). The estimation results from the same-subject validation test were significantly better than those of the cross-subject validation test. Thus, using an ANN model on a subject-by-subject basis to quantify and track changes in the temporal and spectral responses of MMG signals to estimate the elbow flexion force is a reliable method.

[1]  N. Hogan,et al.  Probability density of the surface electromyogram and its relation to amplitude detectors , 1999, IEEE Transactions on Biomedical Engineering.

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

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

[4]  Kenneth Levenberg.  A METHOD FOR THE SOLUTION OF CERTAIN NON – LINEAR PROBLEMS IN LEAST SQUARES , 1944 .

[5]  T. Housh,et al.  Mechanomyographic amplitude and mean power frequency versus torque relationships during isokinetic and isometric muscle actions of the biceps brachii. , 2004, Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology.

[6]  Shin-Ki Kim,et al.  A Supervised Feature-Projection-Based Real-Time EMG Pattern Recognition for Multifunction Myoelectric Hand Control , 2007, IEEE/ASME Transactions on Mechatronics.

[7]  R. Perini,et al.  Muscular sound and force relationship during isometric contraction in man , 2006, European Journal of Applied Physiology and Occupational Physiology.

[8]  C. Orizio,et al.  The surface mechanomyogram as a tool to describe the influence of fatigue on biceps brachii motor unit activation strategy. Historical basis and novel evidence , 2003, European Journal of Applied Physiology.

[9]  R. Merletti,et al.  Effect of accelerometer location on mechanomyogram variables during voluntary, constant-force contractions in three human muscles , 2006, Medical and Biological Engineering and Computing.

[10]  K. Akataki,et al.  Sex differences in mechanomyographic responses to voluntary isometric contractions. , 2006, Medicine and science in sports and exercise.

[11]  Aydin Akan,et al.  Prediction of externally applied forces to human hands using frequency content of surface EMG signals , 2010, Comput. Methods Programs Biomed..

[12]  G. Hägg,et al.  Comparison of different estimators of electromyographic spectral shifts during work when applied on short test contractions , 1991, Medical and Biological Engineering and Computing.

[13]  B. Nigg,et al.  Biomechanics of the musculo-skeletal system , 1995 .

[14]  Katsumi Mita,et al.  Mechanomyographic responses during voluntary ramp contractions of the human first dorsal interosseous muscle , 2003, European Journal of Applied Physiology.

[15]  K Mita,et al.  Electromyographic and mechanomyographic estimation of motor unit activation strategy in voluntary force production. , 2004, Electromyography and clinical neurophysiology.

[16]  Kunihiko Ito,et al.  Age‐related change in motor unit activation strategy in force production: A mechanomyographic investigation , 2002, Muscle & nerve.

[17]  Jeffrey R Stout,et al.  The consistency of ordinary least-squares and generalized least-squares polynomial regression on characterizing the mechanomyographic amplitude versus torque relationship , 2009, Physiological measurement.

[18]  C. Orizio.  Muscle sound: bases for the introduction of a mechanomyographic signal in muscle studies. , 1993, Critical reviews in biomedical engineering.

[19]  Michael Y. Hu,et al.  Forecasting with artificial neural networks: The state of the art , 1997 .

[20]  R. Major.  Biomechanics of the Musculo-Skeletal System , 1994 .

[21]  Dario Farina,et al.  Longitudinal and transverse propagation of surface mechanomyographic waves generated by single motor unit activity , 2008, Medical & Biological Engineering & Computing.

[22]  Mario Aguilar,et al.  Using EMG to anticipate head motion for virtual-environment applications , 2005, IEEE Transactions on Biomedical Engineering.

[23]  D. Barry,et al.  Muscle sounds are emitted at the resonant frequencies of skeletal muscle , 1990, IEEE Transactions on Biomedical Engineering.

[24]  Jeffrey R Stout,et al.  Reliability of absolute versus log-transformed regression models for examining the torque-related patterns of response for mechanomyographic amplitude , 2009, Journal of Neuroscience Methods.

[25]  M. Bartlett,et al.  The use of transformations. , 1947, Biometrics.

[26]  Travis W. Beck,et al.  Mechanomyographic and electromyographic time and frequency domain responses during submaximal to maximal isokinetic muscle actions of the biceps brachii , 2004, European Journal of Applied Physiology.

[27]  Kuolin Hsu,et al.  Artificial Neural Network Modeling of the Rainfall‐Runoff Process , 1995 .

[28]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[29]  Rong Song,et al.  Using recurrent artificial neural network model to estimate voluntary elbow torque in dynamic situations , 2005, Medical and Biological Engineering and Computing.

[30]  Dario Farina,et al.  Upper trapezius muscle mechanomyographic and electromyographic activity in humans during low force fatiguing and non-fatiguing contractions , 2002, European Journal of Applied Physiology.

[31]  B. Maton,et al.  Phonomyogram and electromyogram relationships with isometric force reinvestigated in man , 2004, European Journal of Applied Physiology and Occupational Physiology.

[32]  Michelle Mielke,et al.  Mechanomyographic amplitude and frequency responses during dynamic muscle actions: a comprehensive review , 2005, Biomedical engineering online.

[33]  M. Hahn.  Feasibility of estimating isokinetic knee torque using a neural network model. , 2007, Journal of biomechanics.

[34]  Dario Farina,et al.  Two-dimensional spatial distribution of surface mechanomyographical response to single motor unit activity , 2007, Journal of Neuroscience Methods.

[35]  C. Orizio,et al.  Surface mechanomyogram reflects muscle fibres twitches summation. , 1996, Journal of biomechanics.

[36]  Terry J. Housh,et al.  The effect of leg flexion angle on the mechanomyographic responses to isometric muscle actions , 1998, European Journal of Applied Physiology and Occupational Physiology.