Motion Classification of EMG Signals Based on Wavelet Packet Transform and LS-SVMs Ensemble

This paper presents an effective method for motion classification using the surface electromyographic (sEMG) signal collected from the forearm. Given the nonlinear and time-varying nature of EMG signal, the wavelet packet transform (WPT) is introduced to extract time-frequency joint information. Then the multi-class classifier based on the least squares support vector machine (LS-SVM) is constructed and verified in the various motion classification tasks. The results of contrastive experiments show that different motions can be identified with high accuracy by the presented method. Furthermore, compared with other classifiers with different features, the performance indicates the potential of the SVM techniques combined with WPT in motion classification.

Multifractal analysis of surface EMG signals for assessing muscle fatigue during static contractions

This study is aimed at assessing muscle fatigue during a static contraction using multifractal analysis and found that the surface electromyographic （SEMG） signals characterized multiffactality during a static contraction. By applying the method of direct determination ofthef（a） singularity spectrum, the area of the multifractal spectrum of the SEMG signals was computed. The results showed that the spectrum area significantly increased during muscle fatigue. Therefore the area could be used as an assessor of muscle fatigue. Compared with the median frequency （MDF）--the most popular indicator of muscle fatigue, the spectrum area presented here showed higher sensitivity during a static contraction. So the singularity spectrum area is considered to be a more effective indicator than the MDF for estimating muscle fatigue.

A Hybrid Model Based on ResNet and GCN for sEMG-Based Gesture Recognition

The surface electromyography(sEMG)is one of the basic processing techniques to the gesture recognition because of its inherent advantages of easy collection and non-invasion.However,limited by feature extraction and classifier selection,the adaptability and accuracy of the conventional machine learning still need to promote with the increase of the input dimension and the number of output classifications.Moreover,due to the different characteristics of sEMG data and image data,the conventional convolutional neural network(CNN)have yet to fit sEMG signals.In this paper,a novel hybrid model combining CNN with the graph convolutional network(GCN)was constructed to improve the performance of the gesture recognition.Based on the characteristics of sEMG signal,GCN was introduced into the model through a joint voting network to extract the muscle synergy feature of the sEMG signal.Such strategy optimizes the structure and convolution kernel parameters of the residual network(ResNet)with the classification accuracy on the NinaPro DBl up to 90.07%.The experimental results and comparisons confirm the superiority of the proposed hybrid model for gesture recognition from the sEMG signals.

Microelectronic neural bridging of toad nerves to restore leg function

The present study used a microelectronic neural bridge comprised of electrode arrays for neural signal detection, functional electrical stimulation, and a microelectronic circuit including signal amplifying, processing, and functional electrical stimulation to bridge two separate nerves, and to restore the lost function of one nerve. The left leg of one spinal toad was subjected to external mechanical stimulation and functional electrical stimulation driving. The function of the left leg of one spinal toad was regenerated to the corresponding leg of another spinal toad using a microelectronic neural bridge. Oscilloscope tracings showed that the electromyographic signals from controlled spinal toads were generated by neural signals that controlled the spinal toad, and there was a delay between signals. This study demonstrates that microelectronic neural bridging can be used to restore neural function between different injured nerves.

Terrain Identification for Prosthetic Knees Based on Electromyographic Signal Features

The features of electromyographic （EMG） signals were investigated while people walking on different terrains, including up and down slopes, up and down stairs, and during level walking at different speeds, The features were used to develop a terrain identification method. The technology can be used to develop an intelligent transfemoral prosthetic limb with terrain identification capability, The EMG signals from 8 hip muscles of 13 healthy persons were recorded as they walked on the different terrains. The signals from the sound side of a transfemoral amputee were also recorded. The features of these signals were obtained using data processing techniques with an identification process developed for the identification of the terrain type. The procedure was simplified by using only the signals from three muscles. The identification process worked well in an intelligent prosthetic knee in a laboratory setting.