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 Apr 29:2022:6414664.
doi: 10.1155/2022/6414664. eCollection 2022.

Myoelectric Pattern Recognition Performance Enhancement Using Nonlinear Features

Md Johirul Islam  1   2 , Shamim Ahmad  3 , Fahmida Haque  4 , Mamun Bin Ibne Reaz  4 , Mohammad A S Bhuiyan  5 , Khairun Nisa' Minhad  5 , Md Rezaul Islam  1
Affiliations

Myoelectric Pattern Recognition Performance Enhancement Using Nonlinear Features

Md Johirul Islam et al. Comput Intell Neurosci. .

Abstract

The multichannel electrode array used for electromyogram (EMG) pattern recognition provides good performance, but it has a high cost, is computationally expensive, and is inconvenient to wear. Therefore, researchers try to use as few channels as possible while maintaining improved pattern recognition performance. However, minimizing the number of channels affects the performance due to the least separable margin among the movements possessing weak signal strengths. To meet these challenges, two time-domain features based on nonlinear scaling, the log of the mean absolute value (LMAV) and the nonlinear scaled value (NSV), are proposed. In this study, we validate the proposed features on two datasets, the existing four feature extraction methods, variable window size, and various signal-to-noise ratios (SNR). In addition, we also propose a feature extraction method where the LMAV and NSV are grouped with the existing 11 time-domain features. The proposed feature extraction method enhances accuracy, sensitivity, specificity, precision, and F1 score by 1.00%, 5.01%, 0.55%, 4.71%, and 5.06% for dataset 1, and 1.18%, 5.90%, 0.66%, 5.63%, and 6.04% for dataset 2, respectively. Therefore, the experimental results strongly suggest the proposed feature extraction method, for taking a step forward with regard to improved myoelectric pattern recognition performance.

PubMed Disclaimer

Conflict of interest statement

All authors declare that there are no conflicts of interest regarding the publication of this paper.

Figures

Figure 1
Figure 1
Applications of myoelectric pattern recognition. (a) Prosthetic hand source: https://mcopro.com/blog/resources/arm-hand-prosthetics/ (accessed on 05 Apr. 2022). (b) Game controller (source: https://www.miro.ing.unitn.it/emg-remote-control-of-a/ (accessed on 05 Apr. 2022).
Figure 2
Figure 2
The EMG data acquisition system. (a, b) The schematic circuit diagram and an EMG signal acquisition system.
Figure 3
Figure 3
The frequency spectrum of EMG signal acquisition system: (a) noise and (b) EMG signal.
Figure 4
Figure 4
The EMG signal of dataset 2 in the time domain for ten-finger movements collected from two channels.
Figure 5
Figure 5
The EMG signal acquisition settings. (a, b) The electrode placement and the finger movements.
Figure 6
Figure 6
The frequency spectrum of an EMG signal of dataset 2.
Figure 7
Figure 7
The feature extraction procedure of the proposed time-domain features.
Figure 8
Figure 8
The forward feature selection algorithm.
Figure 9
Figure 9
Block diagram of the myoelectric pattern recognition system.
Figure 10
Figure 10
The scatter plot and RES index of different feature extraction methods for subject 1 of dataset 1: (a) FS1, (b) FS2, (c) FS3, (d) FS4, and (e) the proposed method.
Figure 11
Figure 11
The F1 score of different feature extraction methods: (a) dataset 1 and (b) dataset 2.
Figure 12
Figure 12
The F1 score enhancement of existing feature extraction methods with the LMAV and NSV: (a) dataset 1 and (b) dataset 2.
Figure 13
Figure 13
Movement-wise performance enhancement by using the LMAV and NSV using LDA classifier: (a) dataset 1 and (b) dataset 2.
Figure 14
Figure 14
The impact of the LMAV and NSV on F1 score with variable window size using LDA classifier: (a) dataset 1 and (b) dataset 2.
Figure 15
Figure 15
The impact of the LMAV and NSV on F1 with variable SNR score using LDA classifier: (a) dataset 1 and (b) dataset 2.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Shen C., Pei Z., Chen W., Wang J., Zhang J., Chen Z. Toward generalization of sEMG-based pattern recognition: a novel feature extraction for gesture recognition. IEEE Transactions on Instrumentation and Measurement . 2022;71:1. doi: 10.1109/TIM.2022.3141163. - DOI
    1. Castellini C., Van Der Smagt P. Surface EMG in advanced hand prosthetics. Biological Cybernetics . Jan. 2009;100(1):35–47. doi: 10.1007/s00422-008-0278-1. - DOI - PubMed
    1. Cheng W., Sun Y., Li G., Jiang G., Liu H. Jointly network: a network based on CNN and RBM for gesture recognition. Neural Computing & Applications . 2019;31(S1):309–323. doi: 10.1007/s00521-018-3775-8. - DOI
    1. Arieta A. H., Yokoi H., Arai T., Yu W. Study on the effects of electrical stimulation on the pattern recognition for an EMG prosthetic application. Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings . 2005;7:6919–6922. doi: 10.1109/iembs.2005.1616097. - DOI - PubMed
    1. Wei W., Hong H., Wu X. A hierarchical view pooling network for multichannel surface electromyography-based gesture recognition. Computational Intelligence and Neuroscience . 2021;2021 doi: 10.1155/2021/6591035.6591035 - DOI - PMC - PubMed

LinkOut - more resources