Anbin Xiong

A novel HCI based on EMG and IMU

The technology of human-computer interaction (HCI) is developing rapidly in tandem with the advancement of information and biological technologies. Many new types input device are introduced into this field; some of them are aimed to benefit special groups of people like old or disabled persons. In the meantime, Electromyography (EMG) and Inertia Measure Unit (IMU) have been readily available and extensively applied in control systems in many fields. In this paper, we propose a novel EMG-IMU based mouse controller that controls cursor movements based on IMU signals. The displacement of the cursor is determined by integrating the acceleration signal from the IMU, which moves with the operators arm. The mouse operations such as left click, right click and wheel scroll, are commanded through EMG signals. The pattern recognition algorithm, Linear Discriminant Analysis (LDA), is adopted to classify the EMG data into several clusters, which correspond to the pre-defined mouse operations. Experimental results have indicated that the proposed mouse controller can achieve an accuracy of 88%.

A State-Space EMG Model for the Estimation of Continuous Joint Movements

A state-space electromyography (EMG) model is developed for continuous motion estimation of human limb in this paper. While the general Hill-based muscle model (HMM) estimates only joint torque from EMG signals in an “open-loop” form, we integrate the forward dynamics of human joint movement into the HMM, and such an extended HMM can be used to estimate the joint motion states directly. EMG features are developed to construct measurement equations for the extended HMM to form a state-space model. With the state-space HMM, a normal closed-loop prediction-correction approach such as the Kalman-type algorithm can be used to estimate the continuous joint movement from EMG signals, where the measurement equation is used to reject model uncertainties and external disturbances. Moreover, we propose a new normalization approach for EMG signals for the purpose of rejecting the dependence of the motion estimation on varying external loads. Comprehensive experiments are conducted on the human elbow joint, and the improvements of the proposed methods are verified by the comparison of the EMG-based estimation and the inertial measurement unit measurements.

Application of the Chaos Theory in the Analysis of EMG on Patients with Facial Paralysis

Surface electromyography (sEMG) has been widely applied to disease diagnosis, pathologic analysis and rehabilitation evaluation. It is the nonlinear summation of the electrical activity of the motor units in a muscle and can reflect the state of neuromuscular function. Traditional linear and statistical analysis methods have some significant limitations due to the short-term stationary and lower signal-noise ratio of sEMG. In this paper, we introduce chaotic analysis into the field of sEMG process to investigate the hidden nonlinear characteristics of sEMG of patients with facial paralysis. sEMG on the bilateral masseter, levator labii superioris and frontalis of the 21 patients is recorded. Chaotic analysis is employed to extract new features, including correlation dimension, Lyapunov exponent, approximate entropy and so on. We discover the maximum Lyapunov exponents are all greater than 0, indicating that sEMG is a chaotic signal; correlation dimensions of sEMG on healthy sides are all smaller than that of diseased sides; and inversely, the approximate entropies of healthy sides are all greater than that of diseased sides. Consequently, chaotic analysis can provide a new insight into the complexity of the EMG and may be a vital indicator of diagnosis and recovery assessment of facial paralysis.

Feasibility of EMG-based ANN controller for a real-time virtual reality simulation

Estimation of the joint angle from the surface electromyography (sEMG) is a quite complex task due to the complicated relationship between the kinematical variables and the raw sEMG. In this paper, we build a sEMG-to-motion model with the artificial neural network (ANN). EMG features, including Integral of absolute value (IAV), Zero crossing (ZC), Auto-regression coefficients (ARC), Median frequency (MDF), are extracted as the input of the ANN, and the output of the ANN is the operators elbow joint angle and the wrist motion. In addition, a 3D upper extremity model, which is built in SolidWorks and then transformed into MATLAB, will imitate the operators motion simultaneously with the estimations of the ANN. Thus, we accomplish a virtual reality system to realize the real-time simulation and validate the effectiveness of the sEMG-to-motion model. Experiment results show that the system achieves well in model accuracy, hardware compatibility and real time performance with a small mean square error of 1.921 degrees.

sEMG based quantitative assessment of acupuncture on Bell's palsy: an experimental study

Acupuncture is a traditional Chinese therapeutic method, recognized by western medicine as an important complementary therapy. However, the efficacy assessment of this traditional intervention has been mainly empirical rather than scientific. In this paper, we propose a surface electromyography (sEMG) based approach for quantitative assessment of acupuncture on Bell’s palsy. The assessment is made by comparing the muscle activities of healthy with diseased ones. A feature-vector of four individual sEMG features is introduced for more comprehensive representation of muscle activity. A clustering technique is then proposed to calculate quantitative differences between the muscle activities of the healthy and diseased sides. A regressive model is also proposed to predict the recovery trend of a patient based on the quantitative assessments during his/her clinical acupuncture history. As reported in the paper, a total of 20 selected Bell’s palsy patients have participated in the extensive experiments that were conducted to verify the performance of the proposed method.摘要针刺是一种传统的中医治疗手段, 已经在临床上得到广泛的应用. 但是, 针刺疗效目前缺乏客观精确的量化评估方法. 本文提出一种基于表面肌电信号的针刺治疗面瘫疗效量化评估方法, 主要包括以下几个部分: 首先, 采集患者健康侧与患侧的表面肌电信号, 并提取特征, 组建特征向量; 然后, 采用主元分析和 k均值方法, 对特征向量进行降维聚类, 以类别中心距离作为评估针刺疗效的指标; 最后, 利用类别中心距离构造自回归模型, 预测患者的康复趋势. 20 名面瘫患者参与到本实验中, 验证上述方法. 结果表明, 本文方法预测精度达到 92.8%以上.

Assessment of the Effectiveness of Acupuncture on Facial Paralysis Based on sEMG Decomposition

Although acupuncture has been extensively and routinely applied in clinic around the world, its use is conventionally based on empirical knowledge rather than scientific evidence. In this paper, we investigate the surface electromyographic (sEMG) activity on the face of patients with facial paralysis to validate the effectiveness of acupuncture. sEMG decomposition method is employed to quantify of the differences between the healthy and diseased sides, which includes 2 parts: to decompose sEMG into motor units action potential trains (MUAPTs) with Gaussian Mixture Model (GMM); to assess the recovery of patients with facial paralysis according to the differences between the MUAPTs of healthy and diseased sides. Finally, an auto-regression model is used to predict the recovery trends. Results indicate that the proposed method can assess the effectiveness of acupuncture on facial paralysis and achieve a high accuracy of 90.94% to predict the recovery trends. c Springer International Publishing Switzerland 2015.

Classification of Gesture Based on Semg Decomposition: A Preliminary Study

Multi-channel surface electromyography (sEMG) recognition has been investigated extensively by researchers over the past several decades. However, due to the nature of sEMG sensors, the more sensors are used, the greater chance for the sEMG to be influenced by environment noise. Furthermore, it is not feasible to use multi-sensors in some cases because of the bulky size of the sensors and the limited area of muscles. This paper proposes a novel sEMG recognition method based on the decomposition of single-channel sEMG. At first, sEMG is acquired while the participant does 5 predetermined hand gestures. Then, this signal is decomposed into its component motor unit potential trains (MUAPTs), which includes 4 steps: 2-order differential filtering, spikes detection, dimension reduction and clustering with Gaussian Mixture Model (GMM). Finally, 5 MUAPTs are obtained and used for hand gestures classification: four features, integral of absolute value (IAV), maximum value (MAX), median value of non-zero value (NonZeroMed) and index of NonZeroMed (Ind) are extracted to form feature matrix, which is then classified with the algorithm of Linear Discriminate Analysis (LDA). The classification results indicate this method can achieve an accuracy of 74.7% while the accuracy of traditional classification method for single-channel sEMG is about 52.6%.