Features Classification of EMG Signal Envelope under Colored Noise

Resumen: An analysis of the electromyography (EMG) signal features is provided to realize whether a subject has made some motion or not. In this paper we determine such features and specify the motion characteristics required in the prothesis robotics. An accurate features analysis requires the EMG signal envelope, which is highly affected by diverse artifacts and unknown non Gaussian noise. It is shown that noise and artifacts can be efficiently suppressed if to use filtering algorithms developed for colored measurement noise (CMN). An efficient filtering algorithm is presented to remove the EMG envelope artifacts. It is also demonstrated that the EMG signal features can be extracted with more accuracy under the CMN. Extensive experimental investigations are provided using diverse EMG signal data.Feature analysis in biomedical signals often requires the calculation of the envelope. However, envelope acquisition methods extract undesirable artifacts; Therefore, many researchers develop extraction techniques. We will present filtration processes to remove EMG envelope artifacts, which estimates the linear envelope of the signal at the output. Finally, we will know if the EMG envelope gives the optimal features for an accurate prediction.

¿Cómo citar?

Sandra Marquez, Yuriy S. Shmaliy & Oscar G Ibarra-Manzano. Features Classification of EMG Signal Envelope under Colored Noise. Memorias del Congreso Nacional de Control Automático, pp. 1-6, 2020.

Palabras clave

EMG signals, envelope, filtering, features, classification

• I. Rubin, J. R. Daube, Rapid MUAP quantization, in: AANEM Workshop, pp. 1–7, 2008.
• Reaz, M. Hussain, and F. Mohd-Yasin, “Techniques of emg signalanalysis: Detection, processing, classification, and applications,” Biological Procedures Online, vol. 8, pp. 11–35, 2006.
• Merlo, D. Farina, and R. Merletti, “A fast and reliable technique formuscle activity detection from surface emg signals,” IEEE Trans. Biomed. Eng, vol. 50, no. 3, pp. 316–323, 2003.
• Tallison, P. Godfrey, G. Robinson, “EMG signal amplitude assessment during abdominal bracing and hollowing,” J. Elec- tromyography Kinesiology, vol. 8, no. 1, pp. 51–57, 1996.
• D’Alessio and S. Conforto, “Extraction of the envelope from surface EMG signals,” IEEE Eng. Med. Biol. Mag., vol. 20, no. 6, pp. 55–61, 2001.
• Zhang, R. Shiavi, M. A. Hunt and J. Chen, “Clustering analysis and pattern discrimination of EMG linear envelopes,” IEEE Trans. Biomed. Eng, vol. 38, no. 8, pp. 777-784, Aug. 1991.
• Thongpanja, A. Phinyomark, F. Quaine, Y. Laurillau, C. Limsakul, and P. Phukpattaranont, “Probability density functions of stationary surface EMG signals in noisy environments,” IEEE Trans. Inf. Theory, vol. 63, no. 6, pp. 1547–1557, 2016.
• Jang, J. Kim, S. Lee, and Y. Choi, “EMG-based continuous control scheme with simple classifier for electric-powered wheelchair,” IEEE Trans. Instrum. Elec., vol. 65, no. 7, pp. 3695–3705, 2016
• Ramírez, M. Ruano, C. Younes, “Análisis de bioseñales: Enfoque técnico en el análisis clínico de señales fonocardiografíacas, 1st Edition,” Univ. Nacional de Colombia, vol. 1, pp. 1–1, 2016
• de A. Rocha, J. C. do Carmo and F. Assis de O. Nascimento, “Weighted-Cumulated S-EMG Muscle Fatigue Estimator,” IEEE J. Biomedical and Health Informatics, vol. 22, no. 6, pp. 1854-1862, Nov. 2018.
• Xie, Z. Wang, “Mean frequency derived via Hilbert-Huang transform with application to fatigue EMG signal analysis,” Computer Methods and Programs in Biomedicine, vol. 82, no.2, pp. 312–320, 2006.
• Kutilek, J. H´ybl, J. Mare˘s, V. Socha, P.Smrcka, Pavel, “A myoelectric prosthetic arm controlled by a sensor-actuator loop,” Acta Polytechnica, vol. 54, pp.197-204, 2014.
• Chen and H. Yaru, “Feature extraction and classification of ehgbetween pregnancy and labour group using hilbert-huang transform andextreme learning machine,” Computational and Mathematical Methodsin Medicine, vol. 2017, pp. 1–9, 02 2017
• Kleissen and G. Zilvold, “Estimation uncertainty in ensemble averagesurface emg profiles during gait,” J. Electromyography Kinesiology, vol. 4, pp. 83–94, 1994
• Stirn, T. Jarm, Tomaz, V. Peter Kapus, V. Strojnik, “Evaluation of mean power spectral frequency of EMG signal during 100 metre crawl,” Europ. J. Sport Science, vol.13, pp. 1-10, 2011.
• Chien Hung, H. Wen Vincent Young, C.Yen Wang, Y. HungWang, P. Lei Lee, J. Horng Kang, and M. Tzung Lo, “Quantifying spasticity with limited swinging cycles using pendulum test based onphase amplitude coupling,” IEEE Trans. Nucl. Sci, vol. 24, pp. 1–1, 2016.
• M. Studer, R. J. P. Figueiredo, G. S. Moschytz, “An algorithm for sequential signal estimation and system identification for EMG signals”, IEEE Trans. Biomed. Eng, vol. 31, no.3, pp.285-295, 1984.
• Zhan, S. Guo, K. M. Kendrick, J. Feng, “Filtering noise for synchronised activity in multi-trial electrophysiology data using Wiener and Kalman filters,” BioSystems, vol.96, no.1, pp.1–13, 2009.
• M. Lopez, F. di Sciascio, C. M. Soria, M. E. ´ Valentinuzzi, “Robust EMG sensing system based on data fusion for myoelectric control of a robotic arm,” BioMedical Eng OnLine, vol.8, no.5, pp. 1-13, 2009.
• S. L. Tsui, J. Q. Gan, S. J. Roberts, “A selfpaced braincomputer interface for controlling a robot simulator: an online event labelling paradigm and an extended Kalman filter based algorithm for online training,” Med Biol Eng Comput, vol. 47, no.3, pp.257-265, 2009
• L. Menegaldo, “Real-time muscle state estimation from EMG signals during isometric contractions using Kalman filters,” Biol Cybern, vol. 111, no. 5-7, pp. 335–346, 2017
• Triwiyantoa, O. Wahyunggoroa, H. A. Nugrohoa, H. Herianto, “Muscle fatigue compensation of the electromyography signal for elbow joint angle estimation using adaptive feature,” Computers Electr. Eng., vol. 71, pp. 284–293, 2018.
• Uribe-Murcia, Y. S. Shmaliy, C. K. Ahn and S. Zhao, ”Unbiased FIR Filtering for Time-Stamped Discretely Delayed and Missing Data,” in IEEE Transactions on Automatic Control.2019.
• Aviles, J. Rodriguez, M. Herrera, G. Martinez, UCI Machine Learning Repository. Bogota: Iniversidad Militar Nueva Granada, Tecno Parque SENA nodo Manizales, 2012
• Shmaliy, “An iterative kalman-like algorithm ignoring noise and initialconditions,” IEEE Trans. Signal Process., vol. 59, pp. 2465–2473, 2011
• Watson JC, Dyck PJ. Peripheral Neuropathy: A Practical Approach to Diagnosis and Symptom Management. Mayo Clin Proc 2015;90:940-51
• Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG, Mietus JE, Moody GB, Peng C-K, Stanley HE. PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals (2003). Circulation. 101(23):e215-e220.
• Atzori M., Gijsberts A., Heynen S., Mittaz Hager A.-G., Deriaz O., Van der Smagt P., Castellini C., Caputo B., and M¨uller H. 2012 IEEE International Conference on Biomedical Robotics and Biomechatronics (BioRob 2012).
• S. Shmaliy, F. Lehmann, S. Zhao, and C. K. Ahn, “Comparing robustness of the kalman, H∞, and UFIR filters,” IEEE Trans. Signal Process, vol. 66, no. 13, pp. 3447–3458, 2018.
• S. Shmaliy, “Unbiased FIR filtering of discrete time polynomial state space models,” IEEE Trans. on Signal Process., vol. 57, no. 4, pp. 1241-1249, 2009.
• Shmaliy, A. Marienko, and A. Savchuk, “Gpsbased optimal kalmanestimation of time error, frequency offset, and aging,” 31st Precise Time and Time Interval (PTTI) Systems and Application Mtg, pp. 431–440, 1999
• Marquez-Figueroa, Y. S. Shmaliy, and O. IbarraManzano, “Optimal extraction of EMG signal envelope and artifacts removal assuming colored measurement noise,” Biomed. Signal Process. Control, vol. 57 (to be published).
• M. Atzori, A. Gijsberts, C. Castellini, B. Caputo, H. A. Mittaz, S. Elsig, G. Giatsidis, F. Bassetto, H. Müller, Electromyography data for non-invasive naturally controlled robotic hand protheses. Scientific Data 1:140053 (2014). URL https://doi.org/10.1038/sdata.2014.53
• Lyu, M., Chen, W.-H., Ding, X., Wang, J., Pei, Z., Zhang, B. (2019). Development of an EMGControlled Knee Exoskeleton to Assist Home Rehabilitation in a Game Context. Frontiers in Neurorobotics, 13.