Surface electromyogram denoising using adaptive wavelet thresholding / 生物医学工程学杂志

Surface electromyogram (sEMG) may have low signal to noise ratios. An adaptive wavelet thresholding technique was developed in this study to remove noise contamination from sEMG signals. Compared with convention- al wavelet thresholding methods, the adaptive approach can adjust thresholds based on different signal to noise ratios of the processed signal, thus effectively removing noise contamination and reducing distortion of the EMG signal. The advantage of the developed adaptive thresholding method was demonstrated using simulated and experimental sEMG recordings.

Research on the surface electromyography signal decomposition based on multi-channel signal fusion analysis / 生物医学工程学杂志

The decomposition method of surface electromyography (sEMG) signals was explored by using the multi-channel information extraction and fusion analysis to acquire the motor unit action potential (MUAP) patterns. The action potential waveforms were detected with the combined method of continuous wavelet transform and hypothesis testing, and the effective detection analysis was judged with the multi-channel firing processes of motor units. The cluster number of MUAPs was confirmed by the hierarchical clustering technique, and then the decomposition was implemented by the fuzzy k-means clustering algorithms. The unclassified waveforms were processed by the template matching and peel-off methods. The experimental results showed that several kinds of MUAPs were precisely extracted from the multi-channel sEMG signals. The space potential distribution information of motor units could be satisfyingly represented by the proposed decomposition method.

Research on surface electromyographic signal decomposition based on the level of contraction force / 生物医学工程学杂志

Aiming at the difficulty of surface electromyography (SEMG) signal decomposition, we in this paper proposed a method of gradual processing based on contraction force level of muscle. At first, SEMG signals were recorded at different levels of muscle contraction force. Then, the SEMG data recorded at minimum level of contraction force were decomposed adopting the conventional methods. Further, the data at higher level of contraction force was decomposed using the templates and inter-pulse interval (IPI) information resulted from the previous composition performed at lower level of contraction force. Such procedure was iteratively performed level by level until the SEMG data at the maximal level of contraction force were successfully decomposed. The experimental results showed that the proposed method was effective in decomposing SEMG data, offering a valuable solution to the difficulty in obtaining templates at relatively high level of muscle contraction force. The complexity of SEMG decomposition in the case of high level of contraction force could also be reduced to a certain extent by using the proposed method.

Study on the classification of motor unit action potentials from single-channel surface EMG signal based on the wavelet analysis / 生物医学工程学杂志

A method of motor unit action potentials (MUAP) detection and classification was introduced to explore the firing information of recruited motor units in the neural muscular system. Based on the continuous wavelet transform, the first order Hermite-Rodriguez (HR) function was used as the mother wavelet, and the binary hypothesis testing algorithm was combined to detect and localize the MUAP waveforms in the surface electromyography (sEMG) signal. Then, the fuzzy k-means clustering and minimum distance classifying algorithms were applied to the primary clustering of the detected MUAPs. Finally, the template matching method was used to solve the problem of the unclassified waveforms. The experimental results showed that the kinds of MUAP information from the recorded sEMG signal could be acquired by waveform detection and pattern recognition. The proposed method does not require multi-channel sEMG signals; it just utilizes the single channel signal to analyze the MUAPs, and it can improve the decomposition efficiency.

Identification of Model Parameters Basing on Matched Processing between Simulated and Recorded sEMG Signals / 航天医学与医学工程

Objective To identify the model parameters of surface Electromyography (sEMG) by comparison between simulated and recorded signals. Methods A physiological model of sEMG signal was established basing on several logical hypothetical conditions, such as motor unit action potentials (MUAP), motor unit recruitment and firing behavior caused by excitation, architecture of volume conductor and other simulated factors. According to the matched shapes between the simulated and recorded sEMG signals, a group of model parameters was obtained; according to the similar power spectrum variations of real sEMG signals, decreased muscle fiber conduction velocity (MFCV) was applied to simulate the sEMG signals of the fatigued muscle. Results The experimental results showed that the simulated superimposed MUAP shapes could be matched with the recorded MUAPs satisfactorily by adjusting some proper physiological parameters of the model. When the MFCV of each fiber was assumed to decrease, the mean and median frequency (MNF, MDF) of the simulated sEMG signals declined, and this phenomenon was very similar to that of the recorded sEMG signals and could be used to interpret the muscle fatigue process. Conclusion This model provides an effective approach to simulate real sEMG signals, and the simulated signals can also be used to help the analysis of recorded sEMG signals.

Quantitative relationship between needle electrode electromyography signal of AR model parameter and local muscular fatigue / 中国组织工程研究

BACKGROUND: Research on muscular fatigue has extensive value of application in fundamental research about neuromuscular system, handicapped rehabilitation engineering, objective evaluation on physiotherapeutic effect, scientific training of athletes and ergonomics etc.OBJECTIVE: To study partial muscular fatigue by using AR model parameter of needle electrode electromyography (NEMG) signal and try to reveal the quantitative relationship between local muscular fatigue process and AR model coefficient of NEMG signal.DESIGN: Human NEMG signal was taken as the subject, the changing rule of myoelectric characteristics parameter in local muscular fatigue process was studied.SETTING: NML Laboratory of China University of cience and Technology.PARTICIPANTS: 4 eases of NEMG signals were obtained from NEMG signal collection system manufactured by NML Laboratory of China University of Science and Technology. Four volunteers that selected were healthy males and their tibialis anterior muscles were tested.METHODS: Based on the theory of random signal parameter model,NEMG signals were modeled and parameter was selected, the trend of parameter of NEMG signal changing with the time increasing was studied in the muscular fatigue process. Relative programs from MATLAB language toolbox were adopted for programming.MAIN OUTCOME MEASURES: The trend of α1 parameter of NEMG signal changing with the time (fatigue process) increasing. RESULTS: There was a correlation between local museular fatigue of human body and the changing trend of α1 parameter of NEMG signal tested from the muscle, that is, α1 parameter of NEMG signal increased with the time (fatigue process) increasing.CONCLUSION: By means of the increasing trend of α1 parameter of NEMG signal with the time (fatigue process) increasing, muscular fatigue state can be better evaluated.

A Study on the Decomposition of Surface EMG Signals Based on Second Order Non-stationary Source Separation / 航天医学与医学工程

Objective To investigate the decomposition method of surface EMG(sEMG)signals based on Blind Source Separation and to detect the the motor unit action potential(MUAP)information.Methods Utilizing the sEMG signals recorded at low muscle contraction force(10% MVC),the methods of second order non-stationary source separation(SEONS)and FastICA were explored to analyze the sEMG signals decomposition.Results The experiment results showed that the MUAP information could be acquired by spike detection and pattern recognition after the decomposition of recorded sEMG signals using the proposed algorithm and FastICA method,but a little difference occurred due to the complexity of sEMG signals.Conclusion The non-stationary characteristic of sEMG signals is considered by the SEONS algorithm,and the proposed method can be applied in the sEMG signals decomposition.

Application of high frequency component in classification of different mental tasks / 生物医学工程学杂志

Electroencephalogram (EEG) signals of different mental tasks were preprocessed using Independent Component Analysis (ICA). Auto-Regressive (AR) model was used to extract the feature, and Back-Propagation (BP) network as the classifier. When features were extracted from 20-100 Hz high frequency range, the classification accuracy was the same as that taken from the whole frequency range and was more higher than the result of 2-35 Hz normal EEG rhythm. The explanation of this phenomenon is: brain displays different rhythm assimilation during different mental task under the effect of 60 Hz power frequency, so the high frequency components of EEG include more mental modulated information which is useful for improving the classification accuracy. The result presents a new evidence for the brain rhythm assimilation phenomenon and gives a novel feature extraction method for realizing high accuracy real-time BCI based on mental task.

Study of the average FR of the MUAP's in a couple of agonist-antagonist muscles / 生物医学工程学杂志

"Common Drive" is presented recently as a new concept used to explore the control mechanism of neuromuscular system. In this paper, the average firing rate (FR) of the motor unit action potential (MUAP) is estimated by means of decomposition technique for needle electromyographic (NEMG) signals obtained from elbow joint agonist-antagonist muscle pair with constant contraction force. The change tendency and correlation of the average FR with time are studied. The results of the experiment show that, no matter flexion or extension of the elbow joint, the average FR of both motor units(MUs) in the couple of agonist and antagonist descends with time, and the variations of their amplitude and fluctuation are highly correlated. This indicates that when two antagonist muscles are activated simultaneously to stiffen a joint, the nervous system views them as one unit and controls them in similar fashion. It also confirms the existence of "Common Drive" phenomenon at joint level.