基于小波包分解的不同状态下脑电信号分析

本文针对脑电信号的非平稳性,引入小波包分解理论处理临床脑电.根据脑电信号的不同节律特性,提出应用小波包分解构造不同频率特性的时变滤波器,提取脑电信号不同节律的动态特性,并由此构造各种节律的动态脑电地形图.为了研究不同脑功能状态下脑电信号各种节律的动态特性,文中对两组不同的临床脑电数据进行分析,比较两种状态下各种节律的动态特性.实验结果表明,利用小波包分解对脑电信号进行滤波,能够有效提取临床脑电不同节律的动态特性,为分析脑电信号提供一条新的途径.     

Directional dual-tree complex wavelet packet transforms for processing quadrature signals

Quadrature signals containing in-phase and quadrature-phase components are used in many signal processing applications in every field of science and engineering. Specifically, Doppler ultrasound systems used to evaluate cardiovascular disorders noninvasively also result in quadrature format signals. In order to obtain directional blood flow information, the quadrature outputs have to be preprocessed using methods such as asymmetrical and symmetrical phasing filter techniques. These resultant directional signals can be employed in order to detect asymptomatic embolic signals caused by small emboli, which are indicators of a possible future stroke, in the cerebral circulation. Various transform-based methods such as Fourier and wavelet were frequently used in processing embolic signals. However, most of the times, the Fourier and discrete wavelet transforms are not appropriate for the analysis of embolic signals due to their non-stationary time–frequency behavior. Alternatively, discrete wavelet packet transform can perform an adaptive decomposition of the time–frequency axis. In this study, directional discrete wavelet packet transforms, which have the ability to map directional information while processing quadrature signals and have less computational complexity than the existing wavelet packet-based methods, are introduced. The performances of proposed methods are examined in detail by using single-frequency, synthetic narrow-band, and embolic quadrature signals.     

小波分析理论在脑电分析中的应用

小波变换是一种把时间、频率（或尺度）两域结合起来的分析方法。它具有：（１）多分辨率；（２）相对带宽恒定；（３）适当地选择基本小波，可使小波在时、频两域都具有表征信号局部特征的能力的特点，被誉为“分析信号的显微镜”。本系统以Ｗｉｎｄｏｗｓ为操作系统平台，将小波变换用于脑电信号分析，实现病历管理，１００Ｈｚ脑电信号采样，１０分钟脑电数据存储等功能，是一个在Ｗｉｎｄｏｗｓ３．１下开发的脑电分析系统。从脑电信号小波变换后的波形可以看出，各尺度信号不仅反映信号的频率信息，同时也反映信号的时间信息，意即反映此时ＥＥＧ的状态。而传统的傅里叶分析只能获得信号的整体频谱，不能反映时域信息     

基于熵的动态收缩sEMG信号疲劳特征分析

频谱分析方法常被用来检测肌肉疲劳过程。本文将频率分析和非线性动力学方法结合起来，基于表面肌电（sEMG）信号在不同频率分布不均匀的特点将信号能量分解到不同频带。以此计算功率谱／小波包和熵相结合的功率谱熵／小波包熵来衡虽系统的复杂度，进而衡量肌肉的疲劳程度，为用EMG信号研究动态收缩过程中的肌肉疲劳程度提供了新的分析手段和方法。文中方法也适用于萁它生物医学信号的分析。     

用于脑瘫儿下肢康复装置的表面肌电信号特征提取方法研究

采用时域、频域、时频域和熵的特征提取方法,找到适合脑瘫儿表面肌电信号的特征提取方法.通过在训练过程加一个阻力得到四个不同肌肉活性的训练阶段数据,对数据进行预处理和特征提取,然后用因子分析法对所提取的特征进行分析,实验结果显示本研究所提出的时域、时频域和熵特征部分适用于脑瘫儿,频域特征不适用于脑瘫儿.本研究结果对脑瘫儿的康复训练有很大的帮助.     

Event-related oscillations are 'real brain responses'--wavelet analysis and new strategies.

The EEG consists of the activity of an ensemble of generators producing rhythmic activity in several frequency ranges. These oscillators are active usually in a random way. However, by application of sensory stimulation these generators are coupled and act together in a coherent way. This synchronization and enhancement of EEG activity gives rise to 'evoked' or 'event-related oscillations'. The compound evoked potential manifests as superimposition of evoked rhythms in the EEG frequencies ranging from delta to gamma ('natural frequencies of the brain'). The superimposition principle is described with efficient strategies and by utilization of an efficient algorithm. The wavelet analysis confirms the results of the combined analysis procedure obtained by using the amplitude frequency characteristics (AFCs) and digital filtering. The AFC and adapted digital filtering methods are based on the first approach to analyze average evoked potentials. In contrast, the wavelet analysis is based on signal retrieval and selection among a large number of sweeps recorded in a given physiological or psychological experiment. By combining all these results and concepts, it can be stated that the wavelet analysis underlines and extends the expression that alpha-, theta-, delta-, and gamma-responses described in this report are the most important brain responses related to psychophysiological functions. The wavelet analysis confirms once more the expression 'real signals' which we attribute to EEG frequency responses of the brain. It will be demonstrated that the delta, theta, and alpha responses (i.e. the rhythms 'predicted' by digital filtering) are real brain oscillations. The frequency components of the event-related potential vary independently of each other with respect to: (a) their relation to the event; (b) their topographic distribution; and (c) with the mode of the physiological measurements.     

震颤信号分析的研究现状及展望

震颤是人身体某一个或多个功能区肌肉的节律性、不自主振动,是运动神经元异常同步化的结果。用信号处理的方法检测分析震颤患者加速度(accelerometer,ACC)、表面肌电(electromyography,EMG)、脑电(electroencephalography,EEG)信号对震颤临床诊断、等级评定、疾病早期发现等方面具有重要意义。介绍了时域分析、频域分析、人工神经网络、高阶谱、近似熵、模糊、浑沌、判别分析等方法在震颤信号研究中的应用情况,最后展望了震颤信号分析的应用前景。     

基于独立成分和小波分析的胎儿心电提取

背景：在胎儿心电信号的采集过程中,会受到母体和其他噪声的强干扰,如何快捷与有效地提取出胎儿心电将成为重要的研究课题。 目的：采用结合独立成分分析和小波分析的方法对来自于同一母体的观测信号进行独立分量分离,得到有效的胎儿心电。 方法：结合独立成分分析和小波分析的算法进行胎儿心电的特征提取,首先对含噪信号进行小波变换,去除奇异信号和非平稳随机信号,然后对小波重构后的信号运用快速独立成分分析算法进行成分分析。 结果与结论：在胎儿心电信号的采集过程中,会受到母体和其他噪声的强干扰,但这些信号都是随机的,不相关的,可以认为它们间是相互独立的。采用结合独立成分和小波分析的方法对来自于同一母体的观测信号进行独立分量分离,得到有效的胎儿心电。实验证明该方法是一种有效的方法。     

EEG signals classification of epileptic patients via feature selection and voting criteria in intelligent method

Epileptic disease can be diagnosed by using intelligent methods on the Electroencephalograph (EEG) signals. In this paper, wavelet packet transform (WPT) was used in each of the frequency bands and wavelet coefficients were obtained, then the energy and entropy function was done on the wavelet coefficients and used as initial feature vectors. In the next step, eight and 15 features from 30 initial energy and entropy features were selected as the final features because their receiver operating characteristic (ROC) curve areas were higher than others. There were seven classifier inputs. These seven classifiers consisted of four artificial neural networks (ANN) with different structures, support vector machines (SVM), K-nearest neighbours (KNN) and a hybrid network. Each classifier was trained by 0.5, 0.8 and 0.9 EEG signals. After the training process, a fusion network based on a voting criteria was used to make the algorithm robust against the possible changes in each classifier and increase the classification accuracy. Finally, the algorithm was tested by other EEG signals. As a result, normal and epileptic classes were detected with total classification accuracy of 99–100%.     

Quantification of 1H–MRS signals based on sparse metabolite profiles in the time‐frequency domain

MRS is an analytical approach used for both quantitative and qualitative analysis of human body metabolites. The accurate and robust quantification capability of proton MRS (¹H–MRS) enables the accurate estimation of living tissue metabolite concentrations. However, such methods can be efficiently employed for quantification of metabolite concentrations only if the overlapping nature of metabolites, existing static field inhomogeneity and low signal‐to‐noise ratio (SNR) are taken into consideration. Representation of ¹H–MRS signals in the time‐frequency domain enables us to handle the baseline and noise better. This is possible because the MRS signal of each metabolite is sparsely represented, with only a few peaks, in the frequency domain, but still along with specific time‐domain features such as distinct decay constant associated with T ₂ relaxation rate. The baseline, however, has a smooth behavior in the frequency domain. In this study, we proposed a quantification method using continuous wavelet transformation of ¹H–MRS signals in combination with sparse representation of features in the time‐frequency domain. Estimation of the sparse representations of MR spectra is performed according to the dictionaries constructed from metabolite profiles. Results on simulated and phantom data show that the proposed method is able to quantify the concentration of metabolites in ¹H–MRS signals with high accuracy and robustness. This is achieved for both low SNR (5 dB) and low signal‐to‐baseline ratio (?5 dB) regimes.     

基于CNN和频率切片小波变换的T波形态分类

心电实时监控是心血管疾病防治的重要手段.心电图中T波的变化是心肌缺血和心脏猝死等疾病的重要表征,T波形态自动识别是心电远程监控中一个重要问题.由于实时监护用心电的强噪声背景影响,传统的T波特征提取与分类算法遭遇瓶颈.提出一种结合切片频率小波变换和卷积神经网络的T波形态识别算法,包括:自动定位R波波峰位置与T波终点位置,...     

Analysis of rhythmic patterns produced by spinal neural networks

A network of spinal neurons known as central pattern generator (CPG) produces the rhythmic motor patterns required for coordinated swimming, walking, and running in mammals. Because the output of this network varies with time, its analysis cannot be performed by statistical methods that assume data stationarity. The present work uses short-time Fourier (STFT) and wavelet-transform (WT) algorithms to analyze the nonstationary rhythmic signals produced in isolated spinal cords of neonatal rats during activation of the CPGs. The STFT algorithm divides the time series into consecutive overlapping or nonoverlapping windows and repeatedly applies the Fourier transform across the signal. The WT algorithm decomposes the signal using a family of wavelets varying in scale, resulting in a set of wavelet coefficients presented onto a continuous frequency range over time. Our studies revealed that a Morlet WT algorithm was the tool of choice for analyzing the CPG output. Cross-WT and wavelet coherence were used to determine interrelations between pairs of time series in time and frequency domain, while determining the critical values for statistical significance of the coherence spectra using Monte Carlo simulations of white-noise series. The ability of the cross-Morlet WT and cross-WT coherence algorithms to efficiently extract the rhythmic parameters of complex nonstationary output of spinal pattern generators over a wide range of frequencies with time is demonstrated in this work under different experimental conditions. This ability can be exploited to create a quantitative dynamic portrait of experimental and clinical data under various physiological and pathological conditions.     

The effectiveness of the wavelet transforms method in the heart sounds analysis

Heart sounds can be used more efficiently by medical doctors when they are displayed visually, rather through a conventional stethoscope. Heart sounds provide clinicians with valuable diagnostic and prognostic information. Although heart sound analysis by auscultation is convenient as a clinical tool, heart sound signals are so complex and non-stationary that they are very difficult to analyse in time or frequency domains. We have studied the extraction of features from heart sounds in the time-frequency domain for recognition of heart sounds through time-frequency analysis. The application of wavelet transform for the heart sounds is thus described. The performance of continuous wavelet transform, discrete wavelet transform and packet wavelet transform is discussed in this paper. After these transformations, we can compare normal and abnormal heart sounds to verify clinical usefulness of our extraction methods for recognition of heart sounds.     

基于小波包分析的意识任务特征提取与分类

将基于小波包变换的多尺度分析方法应用于自发脑电 (EEG)的特征提取。在对 3种意识任务的脑电信号进行多级小波包分解的基础上 ,将不同尺度空间的能量信号作为特征值 ,组成不同意识任务的特征向量 ,并利用径向基函数神经网络进行分类测试。结果表明 ,小波包变换方法的分类正确率高于自回归模型方法。小波包分析方法可以作为不同意识任务脑电信号特征提取的一种新方法 ,具有较强的稳定性     

A comparison of different feature extraction methods for diagnosis of valvular heart diseases using PCG signals

This article presents a novel method for diagnosis of valvular heart disease (VHD) based on phonocardiography (PCG) signals. Application of the pattern classification and feature selection and reduction methods in analysing normal and pathological heart sound was investigated. After signal preprocessing using independent component analysis (ICA), 32 features are extracted. Those include carefully selected linear and nonlinear time domain, wavelet and entropy features. By examining different feature selection and feature reduction methods such as principal component analysis (PCA), genetic algorithms (GA), genetic programming (GP) and generalized discriminant analysis (GDA), the four most informative features are extracted. Furthermore, support vector machines (SVM) and neural network classifiers are compared for diagnosis of pathological heart sounds. Three valvular heart diseases are considered: aortic stenosis (AS), mitral stenosis (MS) and mitral regurgitation (MR). An overall accuracy of 99.47% was achieved by proposed algorithm.     

Use of Neural Networks for Identification of Artifacts in Electroencephalographic Signals Decomposed Using Wavelet Packet Transform

An algorithm for automatic interference (artifact) detection is suggested. This algorithm detects interference (artifact) as a component of EEG signals. The algorithm is based on approximation of electrophysiological signal using neural network model decomposition using the wavelet packet transform.     

An intelligent system for diagnosis of the heart valve diseases with wavelet packet neural networks

In this paper, an intelligent system is presented for interpretation of the Doppler signals of the heart valve diseases based on the pattern recognition. This paper especially deals with combination of the feature extraction and classification from measured Doppler signal waveforms at the heart valve using the Doppler Ultrasound. Because of this, a wavelet packet neural network model developed by us is used. The model consists of two layers: wavelet and multi-layer perceptron. The wavelet layer is used for adaptive feature extraction in the time-frequency domain and is composed of wavelet packet decomposition and wavelet packet entropy. The multi-layer perceptron used for classification is a feed-forward neural network. The performance of the developed system has been evaluated in 215 samples. The test results showed that this system was effective in detecting Doppler heart sounds. The correct classification rate was about 94% for abnormal and normal subjects.     

脑电信号数据压缩及棘波识别的小波神经网络方法

对小波神经网络及其算法研究的基础上,提出了一种对脑电信号压缩表达和痫样脑电棘波识别的新方法。实验结果显示,小波网络在大量压缩数据的同时,能够较好的恢复原有信号。另外,在脑电信号的时频谱等高线图上,得到了易于自动识别的棘波和棘慢复合波特征说明此方法在电生理信号处理和时频分析方面有着光明的应用前景。     

Electroencephalogram signals processing for topographic brain mapping and epilepsies classification

In this study, topographic brain mapping and wavelet transform-neural network method are used for the classification of grand mal (clonic stage) and petit mal (absence) epilepsies into healthy, ictal and interictal (EEGs). Preprocessing is included to remove artifacts occurred by blinking, wandering baseline (electrodes movement) and eyeball movement using the Discrete Wavelet Transformation (DWT). De-noising EEG signals from the AC power supply frequency with a suitable notch filter is another job of preprocessing. In experimental data, the preprocessing enhanced speed and accuracy of the processing stage (wavelet transform and neural network). The EEGs signals are categorized to normal and petit mal and clonic epilepsy by an expert neurologist. The categorization is confirmed by Fast Fourier Transform (FFT) analysis and brain mapping. The dataset includes waves such as sharp, spike and spike-slow wave. Through the Counties Wavelet Transform (CWT) of EEG records, transient features are accurately captured and separated and used as classifier input. We introduce a two-stage classifier based on the Learning Vector Quantization (LVQ) neural network location in both time and frequency contexts. The brain mapping used for finding the epilepsy locates in the brain. The simulation results are very promising and the accuracy of the proposed classifier in experimental clinical data is ∼80%.     

基于小波包变换的癫痫脑电棘波检测

为了能够较好地实现癫痫患者脑电的棘波检测,提出一种将棘波物理特征（幅度、频率）和小波包变换结合的算法,用于癫痫患者脑电信号的棘波检测。首先利用小波包变换对癫痫脑电信号进行小波包分解,将脑电波频率（0~30 Hz）划分为3层;其次根据脑电波的频率范围重构第三层节点频率S（3, 0）（0~10.85 Hz）、S（3, 1）（10.85~21.7 Hz）、S（3, 2）（21.7~32.55 Hz）的脑电信号;最后取棘波的幅度作为检测阈值分别提取癫痫患者健康期、癫痫发作间期及癫痫发作期的棘波。实验结果证明,当数据的采样频率为173.61 Hz、信号长度为23.6 s时,该算法能够提取不同癫痫患者在不同时期的棘波信号,该算法棘波的误检率为12.02%、漏检率为11.70%。因此,本文所采用的算法在癫痫棘波检测中具有良好的效果。