脉象特性分析和识别方法的研究

针对几种常见的脉象运用统计学习理论和方法进行了识别分类的研究。在特征提取方面本文采用了多种不同的信号分析方法，研究了脉象的特征提取。主要分析途径是，时域波形特征提取，频域与倒频域分析，小波域分析，基于AR模型的脉象信号建模以及波形的模板匹配研究。另一方面，研究了几种分类器，作为脉象识别的分类器。通过实验对不同特征的有效性和不同分类器的性能进行了对比。实验结果表明，模板匹配方法和时域特征提取方法较好。     

基于复杂度熵特征融合的高压力人群情感状态评估

长期处于压力状态容易引发各类疾病,合理有效的情感压力状态评估是进行压力情感干预的基础.脑电数据含有丰富的情感信息,针对脑电数据的压力情感特征提取问题,提出一种基于Kc复杂度、小波熵与近似熵相结合的脑电数据的压力情感特征提取方法,以Kc复杂度因子来量化脑电数据的随机程度,以小波熵和近似熵参数分别在时域和频域来量化脑电数据的复杂程度与能量分布；采用遗传算法进行全局寻优,按适者生存的原则进行支持向量机参数的选择、交叉、变异,以此优化的支持向量机融合3类不同层次的特征参数,实现压力情感状态评估.以“切水果”游戏作为压力源,采集8名被试共92组脑电信号,基于该算法来分析被试者的压力状态,最高识别率为94.12％,平均识别率82.06％.研究表明,不同脑区对压力敏感程度不同,左半球相对右半球来说,压力感受敏感.希望通过此工作,可以帮助人们有针对性地采取相应措施缓解压力,恢复身心健康.     

基于小波变换的肌肉疲劳表面肌电信号特征提取的研究

当前肌肉疲劳表面肌电信号(surface electromgography,sEMG)特征提取方法,忽略了非线性跳错信号的影响,且不能在非平稳状态下进行特征提取,存在特征提取准确度差的问题。提出基于小波变换的肌肉疲劳sEMG特征提取研究,采用小波变换对所采集的样本去噪,结合时域、频域特征分析法,融合傅里叶变换方法对肌电信号中的线性特征进行提取,根据带谱近似熵理论对非线性挑错信号进行特征回归分析,并利用拟态分解函数和希尔伯特变换法对肌电信号进行时频特征的整合提取,最终完成基于小波变换的肌肉疲劳sEMG特征提取研究。实验验证,所提方法具有可行性,且将1000个肌电信号样本分成5组,对其中的跳错信号进行特征提取,所提方法准确度较文献方法高出75%,在非平稳状态下将200个肌电信号样本分成5组进行特征提取,所提方法准确度较文献方法高出33%。由此得出,所提方法优于当前特征提取方法。     

基于熵的严重意识障碍脑电信号识别方法

本研究旨在实现对植物状态和最小意识状态脑电信号的分类识别。通过对植物状态和最小意识状态患者施加唤名刺激,采集被唤名时患者的脑电信号;然后对脑电数据进行去噪预处理、样本熵和多尺度熵的特征提取;最后将提取的数据特征送入多核学习支持向量机(SVM)中进行训练和分类。试验结果表明,严重意识障碍患者alpha波脑电特征表现显著,平均分类精度为88.24%,实现了定量化的严重意识障碍状态判定,为意识障碍程度的临床诊断提供了辅助依据。     

一种基于经验小波变换的心电信号室性早搏检测算法

针对心电信号中的室性早搏心拍检测问题,使用经验小波变换（EWT）实现心电信号的自适应分解。根据心电信号时频能量变化特征,提出了一种低复杂度的频域累积能量特征计算方法,并分析了室性早搏与正常心电信号的特征差异性。最后利用反向传播神经网络在MIT-BIH心电数据库上进行心拍样本训练与识别测试。结果表明基于EWT的特征提取避免了传统时域特征提取中的QRS波群检测过程,降低了其它干扰因素对诊断结果的影响,具有较高的分类精度与良好的鲁棒性,总体敏感度与总体阳性检测率分别达到96.55%和97.73%。     

基于光电容积脉搏波的精神疲劳评估方法

目的 针对精神疲劳难于定量评估的问题,本文探索一种非侵入式可穿戴检测方法获取人体生理参数,从而实现对人体精神疲劳的定量评估。方法 搭建光电容积脉搏波（photoplethysmography,PPG）采集平台,采集20名健康在校生的PPG信号,对PPG信号进行预处理和特征提取,获取时域、频域共143维特征。使用机器学习算法建立分类模型,对于Pearson相关系数法、F检验和relief-F得到的特征权值,选择最优的特征子集,使用降维后的特征子集训练模型,减少复杂度和过拟合概率。结果 与实际状态对比,基于该方法的单个体疲劳检测平均准确率为92.48%,多个体疲劳检测准确率最大值为92.2%,可以有效地识别精神疲劳。结论 光电容积脉搏波信号经过时域和频域分析构建的特征能够使用机器学习算法进行准确的精神疲劳状态分类评估。     

表面肌电信号特征提取方法研究发展趋势

表面肌电信号(Surface Electromyographic,s EMG)监测广泛运用于临床诊断、康复医学,它的特征信号提取是进行临床诊断的主要依据。文章在常规的时域分析、频域分析特征提取方法的基础上,对最近的研究热点内容包括时频分析方法、参数模型分析方法和非线性特征分析方法等方面进行了综述和探讨,并对表面肌电信号特征提取方法在相关领域的未来研究方向和发展趋势进行分析和展望。     

一种辅助聋儿发声训练的辅音特征提取算法的研究

为了直观地给听力语言障碍者提供发音改进信息,本文研究了利用不送气塞音时域和频域的声学特征来进行特征提取的算法,并运用到辅音训练辅助系统中.系统采用语音信号处理的方法,显示由线性预测得到的时频分布语谱图,用共振峰跟踪技术追踪不送气塞音与后接元音之间过渡音征的共振峰轨迹.由时域特征与共振峰轨迹,判别发音是否准确,并用视觉反馈的方式提供给受训者发音改进意见.系统对10人(6男,4女)的三个不送气塞音/b/、/d/、/g/加上元音/a/的发音进行了测试,获得了良好的初步结果.     

表面肌电信号的分析和特征提取

背景:表面肌电信号的检测与分析对临床诊断人体功能状况以及患者康复具有重要意义。目的:对表面肌电信号的采集、信号处理、特征分析和特征值提取方面进行分析。方法:在人体屈伸肘部的过程中,选取人体上肢4块肌肉(肱三头肌,肘肌,肱二头肌,肱桡肌)分别检测表面肌电信号,对表面肌电信号进行陷波和带通滤波等预处理(优化)。在此基础上分析表面肌电信号的特征,并应用不同的特征值提取方法对优化后的表面肌电信号进行了特征提取。结果与结论:时域方法最早应用于肌电信号分析,易提取、方法简单;频域方法提取的特征值较稳定,使得频域方法成为肌电信号处理技术的主流;以小波变换为代表的时-频分析方法因结合了时域、频域两方法的特性,在肌电信号分析方面颇有潜力。     

脑机接口中一种改进的模式识别方法

为提高脑机接口中脑电识别率，分析了特征提取方面时频特征组合法的缺点，探讨了一种改进的模式识别方法。该方法以样本类平均距离为判据，采用滑动窗优化技术，获取时域均值的最佳时间段和频域功率谱均值的最佳频率段。用经过优化的时域均值和功率谱均值组合作为特征，形成特征向量。基于该特征向量，用神经网络对脑电信号进行分类。以识别正确率为指标，将改进方法与原方法进行对比，实验结果表明改进方法能够提高脑电识别率，具有应用价值。     

Gait and surface electromyography characteristics in hurdle athletes in a total gait cycle北大核心

BACKGROUND: In recent years, the technical parameters about hurdle athletes are mainly obtained through video analysis and DLT algorithm. However, the gait and surface electromyography (sEMG) characteristics during normal walking are little reported. OBJECTIVE: To explore the changes of the gait and lower limb sEMG signals relative to gait period in hurdle athletes. METHODS: Eight male professional hurdlers were selected to perform gait and lower limb sEMG tests on the trail, and the differences in gait and sEMG signals were analyzed by mathematical statistics. RESULTS AND CONCLUSION: The gait parameters of hurdlers showed no significant differences (except step length). In the total gait cycle, along with the gait changing, the right and left side muscles of the same name moved alternately. The median frequency and average power frequency of the tapping leg in the lower limb muscles were greater than those of the swinging leg (except biceps femoris, tibialis anterior and lateral gastrocnemius), but the mean EMG and EMG integral values of the tapping leg were smaller than those of the swinging leg. At the stand phase, the median frequency and average power frequency of the tapping leg in the lower limb muscles were greater than those of the swinging leg (except tibialis anterior), but the mean EMG and EMG integral values of the tapping leg were smaller than those of the swinging leg (except soleus). At the swing phase, the median frequency and average power frequency of the tapping leg in the lower limb muscles were greater than those of the swinging leg (except tibialis anterior and lateral gastrocnemius), but the mean EMG and EMG integral values of the tapping leg were smaller than those of the swinging leg (except soleus). To conclude, there are different degrees of differences in the frequency domain and time domain of the lower limb muscles between tapping and swinging legs. Additionally, the muscle strength of the tapping leg is less than that of the swinging leg. © 2017, Journal of Clinical Rehabilitative Tissue Engineering Research. All rights reserved.     

An exploratory study to design a novel hand movement identification system

Electromyogram signal (EMG) is an electrical manifestation of contractions of muscles. Surface EMG (sEMG) signal collected from the surface of skin has been used in diverse applications. One of its usages is in pattern recognition of hand prosthesis movements. The ability of current prosthesis devices has been generally limited to simple opening and closing tasks, minimizing their efficacy compared to natural hand capabilities. In order to extend the abilities and accuracy of prosthesis arm movements and performance, a novel sEMG pattern recognizing system is proposed. To extract more pertinent information we extracted sEMGs for selected hand movements. These features constitute our main knowledge of the signal for different hand movements. In this study, we investigated time domain, time-frequency domain and combination of these as a compound representation of sEMG signal's features to access required signal information. In order to implement pattern recognition of sEMG signals for various hand movements, two intelligent classifiers, namely artificial neural network (ANN) and fuzzy inference system (FIS), were utilized. The results indicate that our approach of using compound features with principle component analysis (PCA) as dimensionality reduction technique, and FIS as the classifier, provides the best performance for sEMG pattern recognition system.     

Ant colony optimization-based feature selection method for surface electromyography signals classification

This paper presented a new ant colony optimization (ACO) feature selection method to classify hand motion surface electromyography (sEMG) signals. The multiple channels of sEMG recordings make the dimensionality of sEMG feature grow dramatically. It is known that the informative feature subset with small size is a precondition for the accurate and computationally efficient classification strategy. Therefore, this study proposed an ACO based feature selection scheme using the heuristic information measured by the minimum redundancy maximum relevance criterion (ACO-mRMR). The experiments were conducted on ten subjects with eight upper limb motions. Two feature sets, i.e., time domain features combined with autoregressive model coefficients (TDAR) and wavelet transform (WT) features, were extracted from the recorded sEMG signals. The average classification accuracies of using ACO reduced TDAR and WT features were 95.45±2.2% and 96.08±3.3%, respectively. The principal component analysis (PCA) was also conducted on the same data sets for comparison. The average classification accuracies of using PCA reduced TDAR and WT features were 91.51±4.9% and 89.87±4.4%, respectively. The results demonstrated that the proposed ACO-mRMR based feature selection method can achieve considerably high classification rates in sEMG motion classification task and be applicable to other biomedical signals pattern analysis.     

The application of Hilbert�CHuang transform in the analysis of muscle fatigue during cyclic dynamic contractions

Surface electromyography (sEMG) is a common technique used in the assessment of local muscle fatigue. As opposed to static contraction situations, sEMG recordings during dynamic contractions are particularly characterised by non-stationary (and non-linear) features. Standard signal processing methods using Fourier and wavelet based procedures demonstrate well known restrictions on time-frequency resolution and the ability to process non-stationary and/or non-linear time-series, thus aggravating the spectral parameters estimation. The Hilbert-Huang transform (HHT), comprising of the empirical mode decomposition (EMD) and Hilbert spectral analysis (HSA), provides a new approach to overcome these issues. The time-dependent median frequency estimate is used as muscle fatigue indicator, and linear regression parameters are derived as fatigue quantifiers. The HHT method is utilised for the analysis of the sEMG signals recorded over quadriceps muscles during cyclic dynamic contractions. The results are compared with those obtained by the Fourier and wavelet based methods. It is shown that HHT procedure provides the most consistent and reliable assessment of spectral and derived linear regression parameters, given the time epoch width and sampling interval in the time domain. The suggested procedure successfully deals with non-stationary and non-linear properties of biomedical signals.     

应用独立分量分析去除体表肌电中的心电干扰

体表肌电特别是从躯干获得的体表肌电往往受到被测对象自身心电信号的严重干扰。本文利用一种基于独立分量分析(ICA)的去噪方法,去除体表肌电中的心电干扰。该方法将多通道体表肌电进行独立分量分解,并用高通滤波器处理所分解出的心电独立分量以尽可能地保留其中的肌电成分,进而将去除心电干扰后的所有独立分量反向投影回原始信号空间得到去噪后的信号。仿真信号的处理结果表明,当高通滤波器的截止频率为30Hz时,该方法在有效去除心电干扰的同时使体表肌电的保真度达到最大。同时讨论了将信号的峰度(Kurtosis)值作为自动判别心电分量和肌电分量的标准的可能性。     

A discriminant bispectrum feature for surface electromyogram signal classification

This paper presents a discriminant bispectrum (DBS) feature extraction approach to surface electromyogram (sEMG) signal classification for prosthetic control. The proposed feature extraction method involves two steps: (1) the bispectrum matrix integration, and (2) the Fisher linear discriminant (FLD) projection. We compare DBS with other conventional features, such as autoregressive coefficients, root mean square, power spectral distribution and time domain statistics. First, the separability of the features is investigated by the visualization of feature distribution in the FLD subspace and quantitative measurement (Davies–Boulder clustering index). Then four linear and non-linear classifiers are used to evaluate the discriminative powers of the features in terms of classification accuracy (CA). The experimental results show that DBS has better performance than other features for identifying the motion patterns of sEMG signals, and the best CA result of DBS is 99.4%.     

握速可调式肌电假手的系统研究

针对假手无法根据使用者的意愿来控制其握合速度的问题,设计了一种握速可调的智能肌电假手。在该肌电假手系统中,电极引入的表面肌电信号,放大150倍后,在频域进行信号处理。经过阈值和等值计算,最终输出控制信号到驱动电路对假手的直流微电机进行控制。10名受试者参加了抓握试验,每人重复5次,记录输入肌电信号强度为500、1 000、1 500 mV时的输出电流、输出电压和握力等值,并拟合假手输出电压、输出电流、握力与输入肌电信号大小的曲线。拟合结果表明,假手速度与输出电压成正相关,在带有负载且不过载时,握力随着输出电流增大而增大。本假手系统具有稳定性,同时和假手握速可通过检测到的肌电信号强弱进行调节,提高了假手的灵活性。     

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。     

Linear and non-linear analysis of surface electromyograms in weightlifters

The present research was aimed at investigating the peculiarities of surface electromyogram (sEMG) signals in 12 weightlifting athletes (WLA) and 9 control subjects (control group, CG) The sEMG signals were recorded from both vastus lateralis muscles during 20 s isometric contractions made at 30% and 60% of the maximal voluntary contraction (MVC). Ground reaction force (vertical component) was recorded using a force plate. The sEMG was analysed in the frequency domain and the median frequency (MDF) was computed over successive 1 s epochs. A non-linear technique, recurrence quantification analysis was also applied to assess the presence and time course of deterministic structures in sEMG. The percentage of determinism (%DET) was used as a synthetic parameter to quantify the amount of regularly repeating sEMG waves within the signal itself (bursts). In 5 WLA the sEMG displayed a clear burst activity centred at 11 Hz. These bursts were correlated with force output oscillations and were evident both at 30% and 60%MVC. The MDF decay with time was more evident in WLA than in CG subjects. The %DET increased in WLA, this increase being more evident during 60%MVC contractions. Our results seemed to suggest a special disposition among WLA for the development of long-term changes in firing probability during sub-maximal isometric exercise. The MDF and %DET data provided indications of a greater involvement of fast twitch muscle fibres in WLA than in CG. Electronic Publication     

基于GA-RBF神经网络和sEMG的下肢动作识别方法研究

为了提高人体肌电信号对于下肢动作识别的准确率,提出一种基于遗传算法(GA)优化的径向基(RBF)神经网络分类模型.通过采集人体日常8种下肢动作的表面肌电信号并选择"sym6"系小波函数对肌电信号进行滤波预处理,使用主成分分析法(PCA)对时频域特征降维,把特征向量输入GA算法优化的RBF神经网络进行训练和识别.实验结果...