AI Techniques to Estimate Muscle Force and Joint Torque for Upper Extremity

Abstract:

This paper is motivated by works done in the area of robot-assisted stroke rehabilitation. To map the EMG to joint torque, evolutionary techniques with suitable mathematical models are proposed. These models have unknown adjustable parameters, and the values of these parameters are obtained using nonlinear regression methods such as GA and SA. Five subjects took part in the experiments and they were asked to perform non-fatiguing and variable force maximal voluntary contractions and sub-maximal voluntary contractions. The recorded EMG signals data of various joints are entered to the model, to estimate the best fit values for the unknown parameters. Once these values of the parameters are obtained they are applied into the model and thus the joint torque is estimated. Predictions made by the proposed techniques are well correlated with experimental data.

• Periodical: Key Engineering Materials (Volumes 467-469)
• Pages: 788-793
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.467-469.788
• Online since: February 2011
• Authors: S. Parasuraman, Arif Wicaksono Oyong, Veronica Lestari Jauw
• Keywords: Electromyography (EMG), Genetic Algorithm (GA), Robotics, Simulated Annealing (SA), Stroke Rehabilitation

References

[1] N. Hogan, R Mann, Myoelectric signal processing: optimal estimation applied to electromyography-Part1: derivation of the optimal myoprocessors, IEEE Transactions on Biomedical Engineering, vol. BME-27, No. 7, July (1980).

DOI: 10.1109/tbme.1980.326652

[2] Siegler S, Hillstrom HJ, Freedman W, Moskowitz G, Effect of myoelectric signal processing on the relationship between muscle force and processed EMG, Am J Phys Med. 64(3): 130-149, Jun (1985).

[3] P. M. H. Rack and D. R. Westbury, The effects of length and stimulus rate on tension in the isometric cat soleus muscle, J. Physiol. 204; 443460, (1969).

DOI: 10.1113/jphysiol.1969.sp008923

[4] F.E. Zajac, Muscle coordination of movement: a perspective, J Biomech, 26 SuppII: 109- 24. (1993).

[5] A.L. Hof and IW. van der Berg, EMG to force processing III: estimation of model parameters for the human triceps surae muscle and assessment of the accuracy by means of a torque plate, J Biomech, pp.771-785, 14, (1981).

DOI: 10.1016/0021-9290(81)90033-6

[6] Laurene V. Fausett, Fundamentals of Neural Networks: Architectures, Algorithms and Applications, Dec. 19, (1993).

[7] Y. Koike and M. Kawato, Estimation of dynamic joint torques and trajectory formation from surface electromyography signals using a neural network model, BioI Cybernet 73 pp.291-300, (1995).

DOI: 10.1007/bf00199465

[8] Laurene V. Fausett, Nonlinear Regression Analysis and Its Applications, Wiley Series in Probability and Statistics, (1998).

[9] Gallant, A. R., Nonlinear Regression, The American Statistician 29: 7381. (1975).

[10] Goffe WL, Ferrier GD, Rogers J. Global optimization of statistical functions with simulated annealing. J Econometrics; 60: 65–99, (1994).

DOI: 10.1016/0304-4076(94)90038-8

[11] H. Motulsky, A. Christopoulos, Fitting Models to Biological Data using Linear and Nonlinear Regression, Oxford University Press, New York, USA, (2004).

DOI: 10.1002/pst.167