Skip to main content

Advertisement

Springer Nature Link
Log in

Design of sEMG-based clench force estimator in FPGA using artificial neural networks

  • S.I. : Advances in Bio-Inspired Intelligent Systems
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

    We’re sorry, something doesn't seem to be working properly.

    Please try refreshing the page. If that doesn't work, please contact support so we can address the problem.

Abstract

Hands are the main environmental manipulator for the human being. After losing a hand, the only alternative for the victim is to use a prosthesis. Despite the progress of science, the modern prosthesis has the same age-old problem of accurate force estimation. Among different kinds of force, clench force is the most important one. Because of this importance, this paper presents a hardware system that has been designed and implemented to estimate the desired clench force using surface Electromyography signals recorded from lower-arm muscles. The implementation includes a two-layer artificial neural network with a surface electromyography integrator. The neural network was trained with the Levenberg–Marquardt back propagation algorithm and was implemented in a field programmable gate array using an off-chip training method. The results from 10 datasets, recorded from five subjects, show that the hardware model is very accurate, with an average mean square error of 0.003. This suggests that the proposed design can mimic the behavior of clench force that a real limb does, and therefore this intelligent system could be a useful tool for any application related to prostheses.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Explore related subjects

Discover the latest articles and news from researchers in related subjects, suggested using machine learning.

References

  1. Ziegler-Graham K, MacKenzie EJ, Ephraim PL, Travison TG, Brookmeyer R (2008) Estimating the prevalence of limb loss in the United States: 2005 to 2050. Arch Phys Med Rehabil 89:422–429

    Article  Google Scholar 

  2. Darnall BD, Ephraim P, Wegener ST, Dillingham T, Pezzin L, Rossbach P et al (2005) Depressive symptoms and mental health service utilization among persons with limb loss: results of a national survey. Arch Phys Med Rehabil 86:650–658

    Article  Google Scholar 

  3. Parkes CM (1972) Components of the reaction to loss of a limb, spouse or home. J Psychosom Res 16:343–349

    Article  Google Scholar 

  4. Childress DS (1985) Historical aspects of powered limb prostheses. Clin Prosthet Orthot 9:2–13

    Google Scholar 

  5. Atkins DJ, Heard DC, Donovan WH (1996) Epidemiologic overview of individuals with upper-limb loss and their reported research priorities. JPO J Prosthet Orthot 8:2–11

    Article  Google Scholar 

  6. Herzog W (1996) Force-sharing among synergistic muscles: theoretical considerations and experimental approaches. Exerc Sport Sci Rev 24:173–202

    Article  Google Scholar 

  7. Herzog W, Leonard T (1991) Validation of optimization models that estimate the forces exerted by synergistic muscles. J Biomech 24:31–39

    Article  Google Scholar 

  8. Li Z-M, Latash ML, Newell KM, Zatsiorsky VM (1998) Motor redundancy during maximal voluntary contraction in four-finger tasks. Exp Brain Res 122:71–78

    Article  Google Scholar 

  9. Li Z-M, Latash M, Zatsiorsky V (1998) Force sharing among fingers as a model of the redundancy problem. Exp Brain Res 119:276–286

    Article  Google Scholar 

  10. Prilutsky B, Herzog W, Allinger T (1994) Force-sharing between cat soleus and gastrocnemius muscles during walking: explanations based on electrical activity, properties, and kinematics. J Biomech 27:1223–1235

    Article  Google Scholar 

  11. Zatsiorsky VM, Li Z-M, Latash ML (1998) Coordinated force production in multi-finger tasks: finger interaction and neural network modeling. Biol Cybern 79:139–150

    Article  Google Scholar 

  12. Latash ML (1996) The Bernstein problem: how does the central nervous system make its choices. In: Latash ML, Turvey MT (eds) Dexterity and its development. Lawrence Erlbaum Associates, Mahwah, pp 277–303

    Google Scholar 

  13. An K-N, Berglund L, Cooney WP, Chao EY, Kovacevic N (1990) Direct in vivo tendon force measurement system. J Biomech 23:1269–1271

    Article  Google Scholar 

  14. Koike Y, Kawato M (1995) Estimation of dynamic joint torques and trajectory formation from surface electromyography signals using a neural network model. Biol Cybern 73:291–300

    Article  Google Scholar 

  15. Finni T, Komi P, Lukkariniemi J (1998) Achilles tendon loading during walking: application of a novel optic fiber technique. Eur J Appl Physiol 77:289–291

    Article  Google Scholar 

  16. Walmsley B, Hodgson J, Burke R (1978) Forces produced by medial gastrocnemius and soleus muscles during locomotion in freely moving cats. J Neurophysiol 41:1203–1216

    Article  Google Scholar 

  17. Disselhorst-Klug C, Schmitz-Rode T, Rau G (2009) Surface electromyography and muscle force: limits in sEMG–force relationship and new approaches for applications. Clin Biomech 24:225–235

    Article  Google Scholar 

  18. Mostafa SS, Ahmad M, Awal MA (2012) Clench force estimation by surface electromyography for neural prosthesis hand. In: International conference on informatics, electronics and vision (ICIEV), pp 505–510

  19. Wininger M, Kim N-H, Craelius W (2008) Pressure signature of forearm as predictor of grip force. J Rehabil Res Dev 45:883–892

    Article  Google Scholar 

  20. Choi C, Kwon S, Park W, Lee H-D, Kim J (2010) Real-time pinch force estimation by surface electromyography using an artificial neural network. Med Eng Phys 32:429–436

    Article  Google Scholar 

  21. Gupta M, Jin L, Homma N (2004) Static and dynamic neural networks: from fundamentals to advanced theory. Wiley, London

    Google Scholar 

  22. Jain LC, Seera M, Lim CP, Balasubramaniam P (2014) A review of online learning in supervised neural networks. Neural Comput Appl 25:491–509

    Article  Google Scholar 

  23. Hudson DL, Cohen ME (2000) Neural networks and artificial intelligence for biomedical engineering. Wiley-IEEE Press, London

    Google Scholar 

  24. Liu MM, Herzog W, Savelberg HH (1999) Dynamic muscle force predictions from EMG: an artificial neural network approach. J Electromyogr Kinesiol 9:391–400

    Article  Google Scholar 

  25. Prentice S, Patla A, Stacey D (2001) Artificial neural network model for the generation of muscle activation patterns for human locomotion. J Electromyogr Kinesiol 11:19–30

    Article  Google Scholar 

  26. Chinnery D, Keutzer K (2002) Closing the gap between ASIC and custom: tools and techniques for high-performance ASIC design. Springer, Berlin

    Google Scholar 

  27. Dixon AM, Allstot EG, Gangopadhyay D, Allstot DJ (2012) Compressed sensing system considerations for ECG and EMG wireless biosensors. IEEE Trans Biomed Circuits Syst 6:156–166

    Article  Google Scholar 

  28. Benatti S, Casamassima F, Milosevic B, Farella E, Schonle P, Fateh S et al (2015) A versatile embedded platform for EMG acquisition and gesture recognition. IEEE Trans Biomed Circuits Syst 9:620–630

    Article  Google Scholar 

  29. (2015) BSL Hardware guide

  30. Mostafa SS, Awal MA, Ahmad M, Morgado-Dias F (2017) A method for designing EMG integrator using an FPGA. In: 5th IEEE international work conference on bioinspired intelligence (IWOBI), pp 1–6

  31. Gilles MA, Wing AM (2003) Age-related changes in grip force and dynamics of hand movement. J Mot Behav 35:79–85

    Article  Google Scholar 

  32. (2008) Electromyography II. Motor unit recruitment fatigue. Physiology lessons for use with the Biopac Student Lab Manual

  33. Chou C-J, Mohanakrishnan S, Evans JB (1993) FPGA implementation of digital filters. In: International conference on signal processing applications and technology (ICSPAT), p 1

  34. Suresh S, Sundararajan N, Savitha R (2013) Supervised learning with complex-valued neural networks. Springer, Berlin

    Book  Google Scholar 

  35. Svozil D, Kvasnicka V, Pospichal J (1997) Introduction to multi-layer feed-forward neural networks. Chemometr Intell Lab Syst 39:43–62

    Article  Google Scholar 

  36. Getting Started with ZedBoard (2014). http://zedboard.org/product/zedboard. Accessed on 29 Feb 2016

  37. Weinstein RK, Lee RH (2006) Architectures for high performance FPGA implementations of neural models. J Neural Eng 3:21–34

    Article  Google Scholar 

  38. Platform Cable USB II, Xilinx, Inc. San Jose, California, DS593 (v1.5) June 23, 2015

  39. Gadea R, Cerdá J, Ballester F, Mocholí A (2000) Artificial neural network implementation on a single FPGA of a pipelined on-line backpropagation. In: 13th international symposium on system synthesis, pp 225–230

  40. Kollmann K, Riemschneider K-R, Zeidler HC (1996) On-chip backpropagation training using parallel stochastic bit streams. In: 5th international conference on microelectronics for neural networks, pp 149–156

  41. Cox CE, Blanz WE (1992) GANGLION—a fast field-programmable gate array implementation of a connectionist classifier. IEEE J Solid-State Circuits 27:288–299

    Article  Google Scholar 

  42. Vuillemin JE, Bertin P, Roncin D, Shand M, Touati HH, Boucard P (1996) Programmable active memories: reconfigurable systems come of age. IEEE Trans Very Large Scale Integr VLSI Syst 4:56–69

    Article  Google Scholar 

  43. Aguiar L, Reis L, Morgado-Dias F (2010) Neuron implementation using system generator. In: 9th Portuguese conference on automatic control, p 1

  44. Reis L, Aguiar L, Baptista FD, Morgado-Dias F (2014) A software tool for automatic generation of neural hardware. Int Arab J Inf Technol 11:1

    Google Scholar 

  45. Cao H, Li C, Sun S, Wang W, Gao Z, Zhang X (2015) Comparison of two strategies for handgrip force prediction based on sEMG. In: International conference on mechatronics, electronic, industrial and control engineering, pp 656–659

  46. Baptista D, Morgado-Dias F (2012) A hyperbolic tangent replacement by third order polynomial approximation. In: 10th Portuguese conference on automatic control

  47. Viitasalo JH, Komi PV (1977) Signal characteristics of EMG during fatigue. Eur J Appl Physiol 37:111–121

    Article  Google Scholar 

  48. Van Boxtel A (2001) Optimal signal bandwidth for the recording of surface EMG activity of facial, jaw, oral, and neck muscles. Psychophysiology 38:22–34

    Article  Google Scholar 

Download references

Acknowledgements

All the authors acknowledge Portuguese Foundation for Science and Technology for their support through Projeto Estratégico LA 9—UID/EEA/50009/2013. S.S. Mostafa Acknowledges to ARDITI—Agência Regional para o Desenvolvimento e Tecnologia under the scope of the Project M1420-09-5369-FSE-000001—Ph.D. Studentship. Md. Esfaqur Rahman, Molla Azizur Rahman and Abdullah-Al Nahid from Khulna University are acknowledged for their valuable suggestions.

Author information

Authors and Affiliations

  1. Madeira Interactive Technologies Institute, Funchal, Portugal

    Sheikh Shanawaz Mostafa & Fernando Morgado-Dias

  2. Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Portugal

    Sheikh Shanawaz Mostafa

  3. Electronics and Communication Engineering Discipline, Khulna University, Khulna, Bangladesh

    Md. Abdul Awal

  4. Signal Processing Research and Consultancy Group, The University of Queensland, Brisbane, QLD, Australia

    Md. Abdul Awal

  5. Khulna University of Engineering and Technology, Khulna, Bangladesh

    Mohiuddin Ahmad

  6. University of Madeira, Funchal, Portugal

    Sheikh Shanawaz Mostafa & Fernando Morgado-Dias

Authors
  1. Sheikh Shanawaz Mostafa
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Md. Abdul Awal
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Mohiuddin Ahmad
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Fernando Morgado-Dias
    View author publications

    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence to Sheikh Shanawaz Mostafa.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Appendix

Appendix

Hardware/software co-design for training and calculating ANN weight (Fig. 9).

Fig. 9
figure 9

Interaction of hardware and software for force estimation and ANN weight calculation

Full size image

In this setting integrated EMG was calculated in the hardware (FPGA). However, the training of ANN was done in the software environment to calculate the weights of ANN. These weights were then used in the FPGA design. This process simplified the weight calculation.

An ANN with five hidden neurons and one output neuron was first designed on MATLAB to predict the clench force. The integrated EMG was the input of the software and it had been used to predict force (output or target) in this study. The ANN was trained using the L–M algorithm. After training, testing was performed to evaluate the system’s performance.

The network’s MSE drops rapidly as the ANN learns. The best validation performance was achieved with 0.001008 at epoch 371 (Fig. 10). Tables 4 and 5 show the force prediction results using the ANN in terms of MSE and the correlation coefficient for training, validation and testing. Negligible errors and high correlation coefficient for all subjects have been observed in both arms.

Fig. 10
figure 10

Train and test performance (MSE) of ANN weight calculation in software (MATLAB). The picture is zoomed for better view

Full size image
Table 4 MATLAB implemented force estimation result of train and validation compared with hand dynamometer data
Full size table
Table 5 MATLAB implemented force estimation results of testing and full dataset compared with hand dynamometer data
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mostafa, S.S., Awal, M.A., Ahmad, M. et al. Design of sEMG-based clench force estimator in FPGA using artificial neural networks. Neural Comput & Applic 32, 15813–15823 (2020). https://doi.org/10.1007/s00521-018-3600-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-018-3600-4

Keywords