sEMG based motion pattern recognition is the focus in the rehabilitation medical engineering area. In order to get more information to characterize the sEMG signal of different upper limb motions, the signal data acquisition program is designed and the non-Gaussian characteristic of the sEMG signal is analyzed, the result shows that the sEMG signal collected is non-Gaussian signal. Bispectrum as a third-order statistics contains non-Gaussian information and the integral of bispectrum slice is extracted as feature. After that, the PCA method is adopted to reduce the bispectrum feature dimension. Then, the integral of bispectrum slice after PCA and the integral value of sEMG are combined as a multi-domain joint feature called BisIE. Finally, the experiment is executed to validate the effectiveness of the feature extraction method proposed by the SVM classifier compared with power spectrum-based multi-domain joint feature MMIE. The average classification accuracy of BisIE is about 97% and that of MMIE is about 93%. Besides, for the same subject, the classification accuracy of BisIE is higher than that of MMIE. The result shows that the proposed feature BisIE is effective in promoting sEMG-based upper limb motion recognition accuracy.
PrangeGBJanninkMJAGroothuis-OudshoornCGM. Systematic review of the effect of robot-aided therapy on recovery of the hemiparetic arm after stroke. J Rehabil Res Dev2006; 43: 161–183.
2.
ZolloLRossiniLBraviM. Quantitative evaluation of upper-limb motor control in robot-aided rehabilitation. Med Biol Eng Comput2011; 49: 1131–1144.
3.
HarwinWSMurgiaAStokesEK. Assessing the effectiveness of robot facilitated neurorehabilitation for relearning motor skills following a stroke. Med Biol Eng Comput2011; 49: 1093–1102.
4.
KrebsHIHoganNAisenML. Robot-aided neurorehabilitation. IEEE Trans Rehabil Eng1998; 6: 75–87.
5.
FengZLiPHouZ. sEMG-based continuous estimation of joint angles of human legs by using BP neural network. Neurocomputing2012; 78: 139–148.
6.
LumPSBurgarCGKenneyDE. Quantification of force abnormalities during passive and active-assisted upper-limb reaching movements in post-stroke hemiparesis. IEEE Trans Biomed Eng1999; 46: 652–662.
7.
HuangHZhouPLiG. Spatial filtering improves EMG classification accuracy following targeted muscle reinnervation. Ann Biomed Eng2009; 37: 1849–1857.
8.
ShiJCaiYZhuJ. sEMG-based hand motion recognition using cumulative residual entropy and extreme learning machine. Med Biol Eng Comput2013; 51: 417–427.
9.
LiZWangBSunF. sEMG based joint force control for an upper-limb power-assist exoskeleton robot. IEEE J Biomed Health Informat2014; 18: 1043–1050.
10.
LiZWangBYangC. Boosting-based EMG patterns classification scheme for robustness enhancement. IEEE J Biomedl Health Informat2013; 17: 545–552.
11.
TkachDHuangHKuikenTA. Study of stability of time-domain features for electromyographic pattern recognition. J Neuroeng Rehabil2010; 5: 7–21.
12.
RyaitHSAroraASAgarwalR. sEMG signal analysis at acupressure points for elbow movement. J Electromyogr Kines2011; 21: 868–876.
13.
MananasMAJaneRFizJA. Influence of estimators of spectral density on the analysis of electromyographic and vibromyographic signals. Med Biol Eng Comput2002; 40: 90–98.
14.
BilodeauMCinceraMArsenaultAB. Normality and stationarity of emg signals of elbow flexor muscles during ramp and step isometric contractions. J Electromyogr Kines1997; 7: 87–96.
15.
PengZKZhangWMYangBT. The parametric characteristic of bispectrum for nonlinear systems subjected to gaussian input. Mech Syst Signal Process2012; 36: 456–470.
16.
MohebbiMGhassemianH. Prediction of paroxysmal atrial fibrillation based on non-linear analysis and spectrum and bispectrum features of the heart rate variability signal. Comput Meth Programs Biomed2012; 105: 40–49.
17.
YuSNLeeMY. Bispectral analysis and genetic algorithm for congestive heart failure recognition based on heart rate variability. Comput Biol Med2012; 42: 816–825.
18.
XianDDuaSAcharyaRU. Classification of epilepsy using high-order spectra features and principal component analysis. J Med Syst2012; 36: 1731–1743.
19.
Sirinee T, Angkoon P, Chusak L, et al. Probability density of electromyography signal for different levels of contraction of biceps Brachii. In: Int. conf. electr. eng./electron., comput., telecommun. inf. technol., ECTI-CON 2013, Krabi, Thailand, 15–17 May 2013, p.5.
20.
AyachiFSBoudaoudSMarqueC. Evaluation of muscle force classification using shape analysis of the sEMG probability density function: A simulation study. Med Biol Eng Comput2014; 52: 673–684.
21.
ChenXZhuXZhangD. A discriminant bispectrum feature for surface electromyogram signal classification. Med Eng Phys2010; 32: 126–135.
22.
OroscoECLopezNMDiSF. Bispectrum-based features classification for myoelectric control. Biomed Signal Process Control2013; 8: 153–168.
23.
Wang N, Yang C, Lyu MR, et al. An EMG enhanced impedance and force control framework for telerobot operation in space. In: IEEE Aerosp Conf Proc 2014, Big Sky, Montana, 1–8 March 2014, pp.1–10.
24.
JiangYChenXTaoJ. Numerically simulating non-Gaussian random processes with specified PSD, skewness and kurtosis. J Syst Simul2006; 18: 1127–1130.
25.
MartinSSchneiderPSimonP. The bispectrum covariance beyond Gaussianity. Astron Astrophys2012; 540: A9–A9.
26.
Naik GR, Kumar DK and Arjunan SP. Kurtosis and negentropy investigation of myo electric signals during different MVCs. In: ISSNIP Biosignals Biorobotics Conf.: Biosignals Rob. Better Safer Living, BRC, Vitoria, Brazil, 6–8 January 2011, pp.40–43.
27.
SwamiAMendelJMMathworksI. Higher-order spectral analysis toolbox for use with MATLAB®, Natick, MA: MathWorks, 1998.
28.
SantosMGHeavensABalbiA. Multiple methods for estimating the bispectrum of the cosmic microwave background with application to the maxima data. Mon Not R Astron Soc2003; 341: 623–643.
29.
OroianT-CVizitiuC-ISerbanF. Bispectral resolution and leakage effect of the indirect bispectrum estimate for different types of 2d window functions. WSEAS Trans Signal Process2008; 4: 637–646.
30.
ZhouYChenJDongGM. Application of the horizontal slice of cyclic bispectrum in rolling element bearings diagnosis. Mech Syst Signal Process2012; 26: 229–243.
31.
SwarnkarVAbeyratneUHukinsC. Objective measure of sleepiness and sleep latency via bispectrum analysis of EEG. Med Biol Eng Comput2010; 48: 1203–1213.
32.
XieSKJinFKrishnanS. Signal feature extraction by multi-scale PCA and its application to respiratory sound classification. Med Biol Eng Comput2012; 50: 759–768.
33.
MartisRJAcharyaURMinLC. ECG beat classification using PCA, LDA, ICA and discrete wavelet transform. Biomed Signal Process Control2013; 8: 437–448.
34.
VaughanTMHeetderksWJTrejoLJ. Brain-computer interface technology: a review of the second international meeting. IEEE Trans Neural Syst Rehabil Eng2003; 11: 94–109.
35.
StulenFLucaCJD. Frequency parameters of the myoelectric signal as a measure of muscle conduction velocity. IEEE Trans Biomed Eng1981; 28: 515–523.
36.
DengZLiuBLiY. Common power spectrum estimation methods and Matlab simulation. Electron Sci Technol2014; 27: 50–52.
37.
FanTKouYLiuC. Judgement of data stability in time sequence analysis. Modern Electronics Technique2013; 36: 66–72.
38.
LauH-YTongK-YZhuH. Support vector machine for classification of walking conditions using miniature kinematic sensors. Med Biol Eng Comput2008; 46: 563–573.
39.
ShenoyPMillerKJCrawfordB. Online electromyographic control of a robotic prosthesis. IEEE Trans Biomed Eng2008; 55: 1128–1135.
40.
KhezriMJahedM. A neuro–fuzzy inference system for semg-based identification of hand motion commands. IEEE Trans Ind Electron2011; 58: 1952–1960.
