Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Oct 20:16:1020086.
doi: 10.3389/fnins.2022.1020086. eCollection 2022.

Natural grasping movement recognition and force estimation using electromyography

Baoguo Xu  1 , Kun Zhang  1 , Xinhao Yang  1 , Deping Liu  1 , Cong Hu  2 , Huijun Li  1 , Aiguo Song  1
Affiliations

Natural grasping movement recognition and force estimation using electromyography

Baoguo Xu et al. Front Neurosci. .

Abstract

Electromyography (EMG) generated by human hand movements is usually used to decode different action types with high accuracy. However, the classifications of the gestures rarely consider the impact of force, and the estimation of the grasp force when performing natural grasping movements is so far overlooked. Decoding natural grasping movements and estimating the force generated by the associated movements can help patients to improve the accuracy of prosthesis control. This study mainly focused on two aspects: the classification of four natural grasping movements and the force estimation of these actions. For this purpose, we designed an experimental platform where subjects could perform four common natural grasping movements in daily life, including pinch, palmar, twist, and plug grasp, to complete target profiles. On the one hand, the results showed that, for natural grasping movements with different levels of force (three levels at 20, 50, and 80%), the average accuracy could reach from 91.43 to 97.33% under five classification schemes. On the other hand, the feasibility of force estimation for natural grasping movements was demonstrated. Furthermore, in the process of force estimation, we confirmed that the regression performance about plug grasp was the best, and the average R 2 could reach 0.9082. Besides, we found that the regression results were affected by the speed of force application. These findings contribute to the natural control of myoelectric prosthesis and the EMG-based rehabilitation training system, improving the user's experience and acceptance.

Keywords: action decoding; electromyography (EMG); force estimation; grasping force; natural grasping movements.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

FIGURE 1
FIGURE 1
Experimental setup for four natural grasping movements. (A) The experimental table with a screen in front. (B) EMG electrode setup on right arm (left: front, right: back). (C) Force sensor measurement direction.
FIGURE 2
FIGURE 2
The experimental paradigm based on the audio and visual cues and data synchronization scheme.
FIGURE 3
FIGURE 3
Three profiles to apply force in the first experiment. Blue lines represent the target force profiles.
FIGURE 4
FIGURE 4
Four profiles to apply force in the second experiment. Blue lines represent the target profile. In the random profile, the red dotted line represents that the subject should not exert force beyond this range.
FIGURE 5
FIGURE 5
Average recognition rate about the five classification schemes under different total numbers of channels and the box diagram of subject one. The horizontal axis represents the total number of channels. SCH1: four-class classification for four actions without using force information. SCH2: eight-class classification for four actions with level 1 and level 2 of force. SCH3: eight-class classification for four actions with level 2 and level 3 of force. SCH4: eight-class classification for four actions with level1 and level3 of force. SCH5: 12-class classification for four actions with level1, level2, and level3 of force. The horizontal axis represents the total number of channels.
FIGURE 6
FIGURE 6
The accuracy of each subject under five classification schemes when the number of channels was four.
FIGURE 7
FIGURE 7
Regression results of the four grasping forces with different step lengths (A) and window lengths (B).
FIGURE 8
FIGURE 8
Force estimation results of different natural grasping movements (Take the step-climbing profile as an example). All the combinations were tested and the range of R2 is indicated. The bars were plotted with the mean values.
FIGURE 9
FIGURE 9
The estimation result of palmar force under four profiles (Take subject five as an example). (A) A rectangular profile (R2 = 0.9375). (B) A triangle profile (R2 = 0.9751). (C) A step-climbing profile (R2 = 0.9832). (D) A random profile (R2 = 0.9485).
FIGURE 10
FIGURE 10
The estimation results of four grasping forces under the step-climbing profile (Take subject three as an example). (A) Pinch force (R2 = 0.9343). (B) Palmar force (R2 = 0.9423). (C) Twist force (R2 = 0.9694). (D) Plug force (R2 = 0.9755).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Blana D., van den Bogert A. J., Murray W. M., Ganguly A., Krasoulis A., Nazarpour K., et al. (2020). Model-based control of individual finger movements for prosthetic hand function. IEEE Trans. Neural Syst. Rehabil Eng. 28 612–620. 10.1109/TNSRE.2020.2967901 - DOI - PMC - PubMed
    1. Bog M. F., Erkocevic E., Niemeier M. J., Mathiesen J. R., Smidstrup A., Kamavuako E. N. (2011). “Investigation of the linear relationship between grasping force and features of intramuscular EMG,” in Proceedings of the 15th Nordic-Baltic conference on biomedical engineering and medical physics (NBC 2011), Aalborg. 10.1016/j.jneumeth.2009.09.006 - DOI
    1. Boudaoud S., Allouch S., Al Harrach M., Marin F. (2015). On the benefits of using HD-sEMG technique for estimating muscle force. Comput. Methods Biomech. Biomed. Eng. 18 1890–1891. 10.1080/10255842.2015.1070578 - DOI - PubMed
    1. Castellini C., Gruppioni E., Davalli A., Sandini G. (2009). Fine detection of grasp force and posture by amputees via surface electromyography. J. Physiol. Paris 103 255–262. 10.1016/j.jphysparis.2009.08.008 - DOI - PubMed
    1. Castellini C., Smagt P. V. D., Sandini G., Hirzinger G. (2008). “Surface EMG for force control of mechanical hands,” in Proceedings of the 2008 IEEE international conference on robotics and automation, Pasadena, CA, 725–730.

LinkOut - more resources