Classification and Regression of Muscle Neural Signals on Human Lower Extremities via BP_AdaBoost
Abstract
:1. Introduction
- (1)
- The correlation between rectus femoris and biceps femoris is obtained by OpenSim and SPSS, which provides a theoretical basis for the EMG signal acquisition of the lower limbs.
- (2)
- Muscle neural activation is shown to be used to identify human movement. Taking the BP neural network as the weak classifier, the BP_AdaBoost strong classifier can be used to identify different knee movements and improve the recognition rate of the BP neural network. BP_AdaBoost can also be used as a regressor, which has a good effect when the thigh EMG signal maps knee angle.
2. Materials and Methods
2.1. Muscle Selection of Lower Extremity
2.2. EMG Signal Acquisition and Noise Reduction
2.3. Feature Extraction for Weak Classification—Muscle Neural Activation
2.4. Design of BP_AdaBoost Classifier
3. Results and Discussion
3.1. Movement Classification Based on BP_Adaboost
3.2. Knee Angle Regression Based on BP_Adaboost
4. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Liu, L.; Yang, P.; Liu, J. Human Motion Intent Recognition Based on Kernel Principal Component Analysis and Relevance Vector Machine. Robot 2017, 5, 661–669. [Google Scholar]
- Sun, Z.N.; Li, Q.; Liu, Y.F. Research Progress of Computer Vision and Pattern Recognition. Front. Data Comput. 2019, 10, 16. [Google Scholar]
- Wu, X.; Fang, K.; Chen, C.; Zhang, Y. Development of a lower limb multi-joint assistance soft exosuit. Sci. China Inf. Sci. 2020, 63, 1–3. [Google Scholar] [CrossRef]
- Bai, X.J. Research on Neural Signal Acquisition and Stimulation System; Tianjin University: Tianjin, China, 2007; pp. 25–30. [Google Scholar]
- Lu, D.M. Measurement Methods of Sports Biomechanics; Beijing Sport University Press: Beijing, China, 2001; pp. 65–71. [Google Scholar]
- Han, Y.; Wang, X. Biomechanics study of human lower limb walking: Implication for design of power-assisted robot. In Proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots & Systems, Taipei, Taiwan, 18–22 October 2010. [Google Scholar]
- Xu, H.Q.; Liu, F.; Nan, J.W. Motion Research of Power-Assisted Lower Extremity Exoskeleton under Walking Process. Mach. Des. Manuf. 2019, 341, 107–110. [Google Scholar]
- Ahmed, T.; Islam, M.K. EMG Signal Classification for Detecting Neuromuscular Disorders EMG Signal Classification for Detecting Neuromuscular Disorders. In Proceedings of the First International Conference on Advances in Smart Sensor, Signal Processing and Communication Technology (ICASSCT 2021), Goa, India, 19–21 March 2021. [Google Scholar]
- Grüneberg, P. Empowering Patients in Interactive Unity with Machines: Engineering the HAL (Hybrid Assistive Limb) Robotic Rehabilitation System. In Humans and Devices in Medical Contexts; Palgrave Macmillan: Singapore, 2021. [Google Scholar]
- Kumar, H.; Goen, A. Comparative Analysis of FIR Digital Filters for Noise Reduction in EMG Signal. Int. J. Adv. Res. 2015, 3, 598–603. [Google Scholar]
- Hasan, U.C. Statistical and Practical Significance of Articles at Sports Biomechanics Conferences. Ann. Appl. Sport Sci. 2021, 9. [Google Scholar] [CrossRef]
- Chowdhury, R.H.; Reaz, M.B.I.; Ali, M.A.B.M.; Bakar, A.A.A.; Chellappan, K.; Chang, T.G. Surface Electromyography Signal Processing and Classification Techniques. Sensors 2013, 13, 12431–12466. [Google Scholar] [CrossRef]
- Kundu, B.; Naidu, D.S. Classification and Feature Extraction of Different Hand Movements from EMG Signal using Machine Leaning based Algorithms. In Proceedings of the 2021 International Conference on Electrical, Communication, and Computer Engineering (ICECCE), Kuala Lumpur, Malaysia, 12–13 June 2021. [Google Scholar]
- Ullah, K.; Kim, J.H. A mathematical model for mapping EMG signal to joint torque for the human elbow joint using nonlinear regression. In Proceedings of the International Conference on Autonomous Robots & Agents (ICARA), Wellington, New Zealand, 10–12 February 2009; IEEE: Piscataway, NJ, USA, 2009. [Google Scholar]
- Pearle, A.D.; Solomon, D.J.; Wanich, T.; Moreau-Gaudry, A.; Granchi, C.C.; Wickiewicz, T.L.; Warren, R.F. Reliability of Navigated Knee Stability Examination. Am. J. Sports Med. 2017, 35, 1315–1320. [Google Scholar] [CrossRef]
- Kim, J.J.; Cho, H.; Park, Y.; Jang, J.; Kim, J.W.; Ryu, J.S. Biomechanical influences of gait patterns on knee joint: Kinematic & EMG analysis. PLoS ONE 2020, 15, e0233593. [Google Scholar]
- Farina, D.; Merletti, R.; Nazzaro, M.; Caruso, I. Effect of joint angle on EMG variable in leg and thigh muscles. IEEE Eng. Med. Biol. Mag. 2001, 20, 62–71. [Google Scholar] [CrossRef]
- Stachaczyk, M.; Atashzar, S.F.; Farina, D. Adaptive Spatial Filtering of High-Density EMG for Reducing the Influence of Noise and Artefacts in Myoelectric Control. IEEE Trans. Neural Syst. Rehabil. Eng. 2020, 28, 1511–1517. [Google Scholar] [CrossRef] [PubMed]
- Cui, X.; Liu, H.; Wu, Q.X. Study on Spacesuit Joint Motion Assistance Based on Neuromuscular Activation Feedback. Manned Spacefl. 2019, 25, 448–453. [Google Scholar]
- Ying, C.; Qi-Guang, M.; Jia-Chen, L.; Lin, G. Advance and Prospects of AdaBoost Algorithm. Acta Autom. Sin. 2013, 39, 745–758. [Google Scholar]
- Jin, W.; Li, Z.J.; Wei, L.S.; Zhen, H. The improvements of BP neural network learning algorithm. In Proceedings of the International Conference on Signal Processing, Beijing, China, 21–25 August 2000; IEEE: Piscataway, NJ, USA, 2002. [Google Scholar]
Biceps F. | Rectus F. | Vastus I. | Vastus L. | Vastus M. | ||||||
---|---|---|---|---|---|---|---|---|---|---|
R | p | R | p | R | p | R | p | R | p | |
Biceps F. | 1 | −0.932 | 0.000 | −0.468 | 0.001 | −0.468 | 0.001 | −0.467 | 0.001 | |
Rectus F. | 1 | 0.756 | 0.000 | 0.756 | 0.000 | 0.755 | 0.000 | |||
Vastus I. | 1 | 1 | 0.000 | 1 | 0.000 | |||||
Vastus L. | 1 | 1 | 0.000 | |||||||
Vastus M. | 1 |
Regressor | Thigh-Raising | Calf-Raising | Squatting | Knee Bending | Walking | Average |
---|---|---|---|---|---|---|
Decision Tree | 35.47 | 33.35 | 23.6 | 14.15 | 22.38 | 25.79 |
SVR | 15.03 | 14.92 | 10.82 | 23.98 | 17.90 | 16.53 |
Gaussian Regression | 18.03 | 17.13 | 13.64 | 12.39 | 21.24 | 16.486 |
BP | 11.02 | 14.51 | 13.70 | 15.18 | 16.22 | 14.126 |
BP_AdaBoost | 9.02 | 4.44 | 5.7 | 8.24 | 4.52 | 6.384 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Wang, J.; Dai, Y.; Si, X. Classification and Regression of Muscle Neural Signals on Human Lower Extremities via BP_AdaBoost. Appl. Sci. 2022, 12, 5830. https://doi.org/10.3390/app12125830
Wang J, Dai Y, Si X. Classification and Regression of Muscle Neural Signals on Human Lower Extremities via BP_AdaBoost. Applied Sciences. 2022; 12(12):5830. https://doi.org/10.3390/app12125830
Chicago/Turabian StyleWang, Junyao, Yuehong Dai, and Xiaxi Si. 2022. "Classification and Regression of Muscle Neural Signals on Human Lower Extremities via BP_AdaBoost" Applied Sciences 12, no. 12: 5830. https://doi.org/10.3390/app12125830