[Detection study of walking segments of children with cerebral-palsy based on surface electromyographic signals].

In this study, surface electromyography (sEMG) of the lower limbs of cerebral-palsy (CP) subjects in gait cycle was recorded and its parameters of gait cycle characters were analyzed to assess their clinical severity. Three algorithms, including integrated profile (IP), sample-entropy (SampEN) and smooth nonlinear energy operator (SNEO) algorithm, were applied to calculate the duration of walking sEMG segments in simulated SEMG signals. After that, the efficiency and accuracy were compared among these three algorithms.

[Surface electromyogram denoising using adaptive wavelet thresholding].

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.

[Research on surface electromyographic signal decomposition based on the level of contraction force].

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.

[Research on the surface electromyography signal decomposition based on multi-channel signal fusion analysis].

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.

Interpreting sign components from accelerometer and sEMG data for automatic sign language recognition.

The identification of constituent components of each sign gesture is a practical way of establishing large-vocabulary sign language recognition (SLR) system. Aiming at developing such a system using portable accelerometer (ACC) and surface electromyographic (sEMG) sensors, this work proposes a method for automatic SLR at the component level. The preliminary experimental results demonstrate the effectiveness of the proposed method and the feasibility of interpreting sign components from ACC and sEMG data.

[Study on the classification of motor unit action potentials from single-channel surface EMG signal based on the wavelet analysis].

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.

Exploration on the feasibility of building muscle-computer interfaces using neck and shoulder motions.

This paper investigates the feasibility of building muscle-computer interfaces starting from surface Electromyography (SEMG) -based neck and shoulder motion recognition. In order to reach the research goal, a real-time SEMG sensing, processing and classification system was developed firstly. Then two types of SEMG recognition experiments, namely user-specific and user-independent classification, were designed and conducted on seven kinds of neck and shoulder motions to explore the feasibility of using these motions as input commands of muscle-computer interfaces.

Dynamic gesture recognition based on multiple sensors fusion technology.

This paper investigates the roles of a three-axis accelerometer, surface electromyography sensors and a webcam for dynamic gesture recognition. A decision-level multiple sensor fusion method based on action elements is proposed to distinguish a set of 20 kinds of dynamic hand gestures. Experiments are designed and conducted to collect three kinds of sensor data stream simultaneously during gesture implementation and compare the performance of different subsets in gesture recognition.

Feasibility of building robust surface electromyography-based hand gesture interfaces.

This study explored the feasibility of building robust surface electromyography (EMG)-based gesture interfaces starting from the definition of input command gestures. As a first step, an offline experimental scheme was carried out for extracting user-independent input command sets with high class separability, reliability and low individual variations from 23 classes of hand gestures. Then three types (same-user, multi-user and cross-user test) of online experiments were conducted to demonstrate the feasibility of building robust surface EMG-based interfaces with the hand gesture sets recommended by the offline experiments.

The decomposition of surface EMG signals based on blind source separation of convolved mixtures.

The decomposition of surface EMG signals can provide valuable information about the recruitment and firing of motor units from surface EMG recordings. According to the physiologic characteristic of the surface EMG signals generation, a method of the decomposition of SEMG signals based on the technique of convolved mixing blind source separation was proposed. Using simulated SEMG signals, the performance of the decomposition algorithm was analyzed and compared with that of the decomposition technique adopting Independent Component Analysis (ICA).

[Application of high frequency component in classification of different mental tasks].

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.

[Decomposition of EMG signals based on combination of information diffusion theory and fuzzy neural network].

Objective: To solve the problem of large samples and contradictory samples in EMG during high level muscle contraction.

Method: By means of recording EMG during muscle contraction with linearly increasing force instead of constant force, basic MUAP templates were obtained with the combination of information diffusion theory and fuzzy neural network. Samples were compressed and contradictory samples were eliminated.

[Study of the average FR of the MUAP's in a couple of agonist-antagonist muscles].

"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.