基于最佳小波包的表面肌电信号分类方法

针对表面肌电信号的分类问题，采用最佳小波包分解构造最能体现分类能力的小波包基。用Fisher线性判别分析对肌电信号各个子空间的相对能量特征进行降维处理，然后利用BP神经网络进行分类识别。实验表明该方法能够有效地从伸肌和屈肌采集的两道肌电信号中识别前臂内旋、前臂外旋、握拳和展拳四种运动模式，是一种稳定、有效的特征提取方法，为非平稳生理信号的分析提供了新的手段。     

早搏信号的小波模极大多重分形分析

本文运用基于小波模极大值的多重分形分析方法,研究心脏房性早搏(APB)信号、室性早搏(PVC)信号及正常心电(ECG)信号的多重分形特征。通过分析多重分形谱得出:三种信号都具有不同程度的多重分形特性;正常ECG信号的分形程度最强,PVC信号次之,APB信号最弱。t检验结果表明,此方法得出的三种信号分形谱宽度差异具有显著性,对临床医学诊断区分APB、PVC信号有很好的借鉴意义。     

基于多尺度最大李雅普诺夫指数的表面肌电信号模式识别

为了提高动作表面肌电信号的识别率,提出一种将最大李雅普诺夫指数和多尺度分析结合的方法。从非线性和非平稳的角度出发,引入多尺度最大李雅普诺夫指数特征,并应用到人体前臂6类动作表面肌电信号的模式识别中。首先利用希尔伯特-黄变换,对原始信号进行经验模态分解,即多尺度分解;然后利用非线性时间序列分析方法,计算多尺度最大李雅普诺夫指数;最后将多尺度最大李雅普诺夫指数作为特征向量,输入支持向量机进行识别。平均识别率达到97.5%,比利用原始信号的最大李雅普诺夫指数进行识别时提高了3.9%。结果表明,利用多尺度最大李雅普诺夫指数对动作表面肌电信号进行模式识别效果良好。     

基于BP神经网络的手势动作表面肌电信号的模式识别

手势语言在日常生活中有着广泛的应用,本研究利用手势动作时从前臂4块肌肉上获取的4路表面肌电（SEMG）信号,经特征提取并采用BP神经网络,对8种手势动作模式进行了识别。鉴于BP网络具有较强的模式分类能力,而特征提取（幅度绝对值均值、AR模型系数、过零率）又利用了多路肌电信号的信息,实验结果取得了较高的识别正确率,表明所采用的方法是有效的。     

心脑疾病对同步十二导联心电图多重分形谱的影响

对心脑疾病人群的同步十二导联ECG(心电图)进行多重分形特性分析,发现不同导联的多重分形曲线互不重叠。计算其十二导联平均的多重分形奇异强度分布范围以及分布范围的十二个导联间的离散特性,发现不同人群中存在互为交叉而有明显不同的结果。用十二导联多重分形Δα的平均值Δα及其离散度δα(取Δα的标准差)两个参量来描述其多重分形谱特征。发现健康人与心脏病人Δα接近,但δα相差较大;健康人与脑损伤患者δα接近,但Δα相差较大。预示着多重分形特性受到神经自律和心脏组织结构的自谐特性的双重控制,特征参数Δα与神经控制相对应,δΔ与心脏组织结构自谐特性的各向异性相对应。     

EMG classification using wavelet functions to determine muscle contraction

Surface electromyogram (SEMG) is a complex signal and is influenced by several external factors/artifacts. The electromyogram signal from the stump of the subject is picked up through surface electrodes. It is amplified and artifacts are removed before digitising it in a controlled manner so that minimum signal loss occurs due to processing. As removing these artifacts is not easy, feature extraction to obtain useful information hidden inside the signal becomes a different process. This paper presents methods of analysing SEMG signals using discrete wavelet Transform (DWT) for extracting accurate patterns of the signals and the performance of the used algorithms is being analysed rigorously. The obtained results suggest a root mean square difference (RMSD) value for the denoising and quality of reconstruction of the SEMG signal. The result shows that the best mother wavelets for tolerance of noise are second order of symmlets and bior6.8. Results inferred that bior6.8 suitable for the classification and analysis of SEMG signals of different arm motions results in a classification accuracy of 88.90%.     

Multifractal analysis of ventricular fibrillation and ventricular tachycardia

A study of ventricular fibrillation and ventricular tachycardia was undertaken using multifractal analysis. By applying the method of direct determination of the f(alpha) singularity spectrum, the value of the area of the VF and VT singularity spectrum was calculated. The comparison between the results showed that the value of the area of the VF singularity spectrum tended to be larger than that of the value of the area of the VT singularity spectrum. This makes the multifractal singularity spectrum a powerful criterion for discriminating between VF and VT.     

基于多重分形去趋势波动分析的脑电信号特征提取及分类方法

目的:针对脑电信号普遍存在的数据维度高、难以预测的问题,提出一种多重分形去趋势波动分析特征提取方法与长短时记忆网络（LSTM）相结合的脑电信号分类方法。方法:首先对信号样本进行多重分形去趋势波动分析计算得到脑电信号样本的多重分形谱,计算广义Hurst指数hq和广义维数Dq之间的函数关系;然后对多重分形谱进行分析,找出最具代表性的坐标值作为信号的特征向量;最后将其用于LSTM进行训练和分类测试。实验采用波恩大学采集的经过处理的癫痫脑电数据集。结果:当训练样本占总体样本比例超过10%之后,LSTM分类器的测试准确率均稳定在98%以上;当占比超过80%时LSTM分类器的测试准确率达到了100%;即使训练样本较少时也有95%之上的准确率。结论:该算法有良好的准确率和稳定性。     

A novel feature extraction for robust EMG pattern recognition

This paper presents the detailed evaluation and classification of Surface Electromyogram (SEMG) signals at different upper arm muscles for different operations. After acquiring the data from selected locations, interpretation of signals was done for the estimation of parameters using simulated algorithm. First, different types of arm operations were analysed; then statistical techniques were implemented for investigating muscle force relationships in terms of amplitude estimation. The classification (Artificial Neural Network) based results have been presented for detecting different pre-defined arm motions in order to discriminate SEMG signals. The outcome of research indicates that a neural network classifier performs best with an average classification rate of 92.50%. Finally, the result also inferred the operations which were observed to be easy for arm recognition and the study is a step forward to develop powerful, flexible and efficient prosthetic designs.     

基于表面肌电信号的手势识别与分析

目的 研究利用前臂及手部表面肌电( surface electromyography,sEMG)信号进行手势识别的方法,以及不同 手势下拇指、食指的关节角度,探讨 sEMG 信号控制外骨骼手的可行性。 方法 采集 20 名健康右利手受试者右侧 前臂及手部 6 块肌肉 sEMG 信号。 提取 sEMG 信号的时域特征值,对比人工神经网络( artificial neural network, ANN)、K-近邻(K-nearest neighbor, KNN)、决策树(decision tree, DT)、随机森林( random forest, RF)和支持向量机(support vector machine, SVM)等多种分类器对 6 种日常手势进行识别。 同时,采用 Vicon 摄像机跟踪系统捕捉右手拇指、食指运动轨迹,计算拇指、食指关节角度。 结果 利用前臂及手部 sEMG 信号可以实现 6 种手势的模式识别,其中 ANN 分类器的分类预测效果最好,测试集预测精度可达 97. 9% ,Kappa 系数可达 0. 975。 同时,计算得到不同手势下拇指、食指的关节角度,并进行不同手势下关节角度相关性分析。 结论 利用前臂及手部 sEMG 信号进 行手势识别,能够实现具有几乎完全一致的分类预测结果。 研究结果证明了 sEMG 信号手势识别应用于外骨骼手 控制的可行性。     

基于卷积混合盲源分离技术的表面肌电信号分解研究

根据表面肌电信号（SEMG）形成的生理学特性，采用一种基于卷积混合过程的盲源分离技术来分析隐含在SEMG信号中的运动单位动作电位信息，利用仿真的SEMG信号对这种算法的分解性能进行实验研究，并与采用瞬时混合过程的独立分量分析（ICA）算法的分解性能进行比较，同时将该算法应用于真实SEMG信号的分解实验。研究结果表明，无论是对模拟SEMG信号还是真实SEMG信号，采用卷积混合盲源分离技术的分解方法均能得到较明显的分解效果，且该方法较符合表面肌电信号的形成过程，因而具有重要的研究价值。     

心律失常的多重分形去趋势波动分析

目的心律失常的及早诊断对及时救助病人具有重要意义.方法基于多重分形去趋势波动分析方法,本文对正常心电信号及窦、房性心律失常信号进行分析.结果发现三种信号都具有不同程度的长程相关性和多重分形特性,且在波动函数的阶数为正值时,三种信号的长程相关特性最为明显.通过比较三种信号的多重分形谱,发现正常心电信号的多重分形谱宽度最小,窦性心动过缓信号次之,心房颤动宽度最大.结论此研究结果对临床医学诊断区分心律失常与正常心电信号有很好的借鉴意义.     

Signal feature extraction by multi-scale PCA and its application to respiratory sound classification

Respiratory sound (RS) signals carry significant information about the underlying functioning of the pulmonary system by the presence of adventitious sounds. Although many studies have addressed the problem of pathological RS classification, only a limited number of scientific works have focused in multi-scale analysis. This paper proposes a new signal classification scheme for various types of RS based on multi-scale principal component analysis as a signal enhancement and feature extraction method to capture major variability of Fourier power spectra of the signal. Since we classify RS signals in a high dimensional feature subspace, a new classification method, called empirical classification, is developed for further signal dimension reduction in the classification step and has been shown to be more robust and outperform other simple classifiers. An overall accuracy of 98.34% for the classification of 689 real RS recording segments shows the promising performance of the presented method.     

基于多重分形去趋势涨落的心室纤颤和心动过速分析

本文运用多重分形去趋势涨落的分析方法,研究心动过速、心室纤颤和正常心电信号的多重分形特征,用以有效区分上述三种信号。通过分析心动过速、心室纤颤和正常心电信号的赫斯特指数、Renyi指数和多重分形谱,得出三种信号都具有不同程度的长程相关性和多重分形特性,在波动函数的阶数大于0时,三种信号的长程相关特性区别明显。通过分析多重分形谱,发现心室纤颤的多重分形谱比心动过速的多重分形谱宽,正常心电信号的多重分形谱最小。以上研究结果将对临床医学诊断识别心动过速和心室纤颤号信号有很好的借鉴意义。     

Study on Singularity of Chaotic Signal Based on Wavelet Transform

INTRODUCTION The significant information of a signal is often carried by singular characteristics or irregular struc-tures of the signal, for example, the most important information of ECG(electrocardiogram) or EEG isoften presented at the transient points of a signal, such as those points near peaks. The singular charac-teristics of these transient points are more obvious than the smooth parts of signals. Therefore,to studythe singularity of a signal is a meaningful work. Those analysi…     

A Novel Approach Based on Data Redundancy for Feature Extraction of EEG Signals

Feature extraction and classification for electroencephalogram (EEG) in medical applications is a challenging task. The EEG signals produce a huge amount of redundant data or repeating information. This redundancy causes potential hurdles in EEG analysis. Hence, we propose to use this redundant information of EEG as a feature to discriminate and classify different EEG datasets. In this study, we have proposed a JPEG2000 based approach for computing data redundancy from multi-channels EEG signals and have used the redundancy as a feature for classification of EEG signals by applying support vector machine, multi-layer perceptron and k-nearest neighbors classifiers. The approach is validated on three EEG datasets and achieved high accuracy rate (95–99 %) in the classification. Dataset-1 includes the EEG signals recorded during fluid intelligence test, dataset-2 consists of EEG signals recorded during memory recall test, and dataset-3 has epileptic seizure and non-seizure EEG. The findings demonstrate that the approach has the ability to extract robust feature and classify the EEG signals in various applications including clinical as well as normal EEG patterns.     

Objective motor response onset detection in surface myoelectric signals

Precise detection of discrete motor events like the onsets of voluntary muscle contractions is a prerequisite for various psychophysiological approaches in sensorimotor system analysis. In biomedical research and clinical diagnosis, motor events frequently are determined from surface electromyographic (SEMG) signals by some computerized detection algorithm. However, little is known about the reliability and accuracy of these methods, which frequently rely on intuitive and heuristic criteria. Therefore, the systematic approach to computerized detection of discrete motor events from SEMG signals presented by this paper fills a basic gap in EMG signal processing. Based upon a dynamic process model for the SEMG signal, a formal detection scheme is established which incorporates the essential processing modules common to the majority of algorithms. In addition, using concepts of statistically optimal change detection in random processes, a new model-based algorithm is presented which serves as a reference for optimum performance. The validity of this concept is demonstrated for the specific example of accurate detection of muscle activation onsets in rapid voluntary contractions; the estimation error (i.e., the deviation between estimated and "true" onset time) was evaluated by statistical simulations for three representative methods. Results show a substantial decrease of performance of traditional methods in the case of highly variable dynamic muscle activation profiles and/or superimposed activation patterns (e.g., due to a secondary motor task simultaneously executed by the same muscle). The model-based approach provided significantly more accurate results, even when the exact model parameters were unknown but estimated from the SEMG signal actually measured. It is concluded that the detection algorithm has to be critically taken into consideration during interpretation of motor events resolved from SEMG signals. The process model together with the corresponding statistically optimal detector represents an efficient tool for selecting appropriate detection algorithms for a particular experimental condition, and it allows a quantitative assessment of their performance.     

Surface electromyogram spectral characterization and motor unit activity during voluntary ramp contraction in men

Summary The relationships were investigated between the surface electromyographic (SEMG) power spectrum analysed by the 20 order autoregressive model (AR spectrum) and underlying motor unit (MU) activity during isometric contractions increasing linearly from 0% to 80% maximal voluntary contraction. Intramuscular spikes and SEMG signals were recorded simultaneously from biceps brachii muscle; the former were analysed by a computer-aided intramuscular MU spike amplitude-frequency (ISAF) histogram and the latter subjected to AR spectral analysis. Results indicated that there was a positive correlation between the force output and the mean amplitude of the ISAF histogram but not with the mean frequency. These changes were accompanied by changes in relative power of the high frequency (100–200 Hz) peak (HL) in the AR spectrum. It was also found that there was a positive correlation between the mean amplitude of the ISAF histogram and the HL value. These data suggested that the power of the high frequency peak in the AR spectrum of the SEMG signal preferentially reflected the progressive recruitment of underlying MU according to their size. Differences between the AR spectrum and the spectrum estimated by fast Fourier transform algorithm have also been discussed.     

Feature dimensionality reduction for myoelectric pattern recognition: A comparison study of feature selection and feature projection methods

This study investigates the effect of the feature dimensionality reduction strategies on the classification of surface electromyography (EMG) signals toward developing a practical myoelectric control system. Two dimensionality reduction strategies, feature selection and feature projection, were tested on both EMG feature sets, respectively. A feature selection based myoelectric pattern recognition system was introduced to select the features by eliminating the redundant features of EMG recordings instead of directly choosing a subset of EMG channels. The Markov random field (MRF) method and a forward orthogonal search algorithm were employed to evaluate the contribution of each individual feature to the classification, respectively. Our results from 15 healthy subjects indicate that, with a feature selection analysis, independent of the type of feature set, across all subjects high overall accuracies can be achieved in classification of seven different forearm motions with a small number of top ranked original EMG features obtained from the forearm muscles (average overall classification accuracy >95% with 12 selected EMG features). Compared to various feature dimensionality reduction techniques in myoelectric pattern recognition, the proposed filter-based feature selection approach is independent of the type of classification algorithms and features, which can effectively reduce the redundant information not only across different channels, but also cross different features in the same channel. This may enable robust EMG feature dimensionality reduction without needing to change ongoing, practical use of classification algorithms, an important step toward clinical utility.     

EMG feature assessment for myoelectric pattern recognition and channel selection: A study with incomplete spinal cord injury

Myoelectric pattern recognition with a large number of electromyogram (EMG) channels provides an approach to assessing motor control information available from the recorded muscles. In order to develop a practical myoelectric control system, a feature dependent channel reduction method was developed in this study to determine a small number of EMG channels for myoelectric pattern recognition analysis. The method selects appropriate raw EMG features for classification of different movements, using the minimum Redundancy Maximum Relevance (mRMR) and the Markov random field (MRF) methods to rank a large number of EMG features, respectively. A k-nearest neighbor (KNN) classifier was used to evaluate the performance of the selected features in terms of classification accuracy. The method was tested using 57 channels’ surface EMG signals recorded from forearm and hand muscles of individuals with incomplete spinal cord injury (SCI). Our results demonstrate that appropriate selection of a small number of raw EMG features from different recording channels resulted in similar high classification accuracies as achieved by using all the EMG channels or features. Compared with the conventional sequential forward selection (SFS) method, the feature dependent method does not require repeated classifier implementation. It can effectively reduce redundant information not only cross different channels, but also cross different features in the same channel. Such hybrid feature-channel selection from a large number of EMG recording channels can reduce computational cost for implementation of a myoelectric pattern recognition based control system.