EMGdi signal enhancement based on ICA decomposition and wavelet transform

Diaphragmatic electromyogram (EMGdi) signal plays an important role in the diagnosis and analysis of respiratory diseases. However, EMGdi recordings are often contaminated by electrocardiographic (ECG) interference, which posing serious obstacle to traditional denoising approaches due to overlapped spectra of these signals. In this paper, a novel method based on wavelet transform and independent component analysis (ICA) is proposed to remove the ECG interference from noisy EMGdi signals. With the proposed method, the original independent components of contaminated EMGdi signal were first obtained with ICA. Then the ECG components contained were removed by a specially designed wavelet domain filter. After that, the purified independent components were reconstructed back to the original signal space by ICA to obtain clean EMGdi signals. Experimental results achieved on practical clinical data show that the proposed approach is better than several traditional methods include wavelet transform (WT), ICA, digital filter and adaptive filter in ECG interference removing.     

A novel compression algorithm for electrocardiogram signals based on the linear prediction of the wavelet coefficients

This paper describes a new algorithm for electrocardiogram (ECG) compression. The main goal of the algorithm is to reduce the bit rate while keeping the reconstructed signal distortion at a clinically acceptable level. It is based on the compression of the linearly predicted residuals of the wavelet coefficients of the signal. In this algorithm, the input signal is divided into blocks and each block goes through a discrete wavelet transform; then the resulting wavelet coefficients are linearly predicted. In this way, a set of uncorrelated transform domain signals is obtained. These signals are compressed using various coding methods, including modified run-length and Huffman coding techniques. The error corresponding to the difference between the wavelet coefficients and the predicted coefficients is minimized in order to get the best predictor. The method is assessed through the use of percent root-mean square difference (PRD) and visual inspection measures. By this compression method, small PRD and high compression ratio with low implementation complexity are achieved. Finally, we have compared the performance of the ECG compression algorithm on data from the MIT-BIH database.     

Dynamic network signal processing using latent threshold models

We discuss multivariate time series signal processing that exploits a recently introduced approach to dynamic sparsity modelling based on latent thresholding. This methodology induces time-varying patterns of zeros in state parameters that define both directed and undirected associations between individual time series, so generating statistical representations of the dynamic network relationships among the series. Following an overview of model contexts and Bayesian analysis for dynamic latent thresholding, we exemplify the approach in two studies: one of foreign currency exchange rate (FX) signal processing, and one in evaluating dynamics in multiple electroencephalography (EEG) signals. These studies exemplify the utility of dynamic latent threshold modelling in revealing interpretable, data-driven dynamics in patterns of network relationships in multivariate time series.     

基于小波包收缩的心电信号除噪方法研究

文章提出了一种新的基于小波包分析的心电信号的除噪方法。讨论了小波包收缩消噪的原理、阈值的选取以及阈值的量化规则。比较了选择不同的阈值以及不同的阈值量化规则对信号消噪的效果。结果表明基于小波包分析的小波包收缩除噪技术在保持信号奇异性的同时能有效的去除心电信号的噪声。     

基于多类SVM和小波变换的ECG检测、分类算法

提出了一种基于多类SVM和小波变换的ECG检测、分类算法,利用小波变换的时频特性,最大限度地避免了原始ECG信号的各种噪声干扰,有效提取ECG的特征量作为多类SVM的输入训练样本。实验结果表明,将小波变换和SVM相结合,可以有效地将不同病患的ECG信号识别出来。     

Recognizing diseases with multivariate physiological signals by a DeepCNN-LSTM network

The usage of multivariate time series to identify diseases plays an important role in the medical field, as it can help medical staff to improve diagnose accuracy and reduce medical costs. Current research shows that deep Convolutional Neural Networks (CNN) can automatically capture features from raw data and Long Short-Term Memory (LSTM) networks can manage and learn temporal dependence between time series data such as physiological signals. In this work, we propose a deep learning framework called DeepCNN-LSTM by combining the CNN and LSTM to leverage their respective advantages for disease recognition, allowing itself to characterize complex temporal varieties with multiple autoencoded features. In particular, we use stationary wavelet transform together with median filter to preprocess low-frequency signal data, and introduce sliding window to segment physiological time series before model training for performance improvement on the training speed as well as the accuracy for recognizing diseases. In addition, we validate our model on a hybrid benchmark dataset collecting from MIMIC and Fantasia databases and set up four kinds of comparative experiments. Empirical evaluations on the benchmark dataset demonstrate that the proposed model outperforms other competitive approaches.

     

Comparing and classifying one-dimensional spatial patterns: an application to laser altimeter profiles

Numerical analyses of remotely sensed data may valuably contribute to an understanding of the vegetation/land surface interface by pointing out at which scales a given variable displays a high level of spatial variability. Thus, there is a need of methods aimed at classifying many one-dimensional signals, such as airborne laser profiles, on the basis of their spatial structure. The present paper proposes a theoretical framework ensuring a consistent combination of a multi-scale pattern characterization, based on the Haar wavelet variance (also called in ecology Two Terms Local Variance, TTLV), with two multivariate techniques such as principal components analysis (PCA) and hierarchical cluster analysis. We illustrate our approach by comparing and classifying 257 laser profiles, with a length of 64 measurements (448 m), that were collected by the BRGM in French Guiana over three main landscape units with distinct geomorphological and ecological characteristics. We calculate for each profile a scalogram that summarized the multi-scale pattern and analyze the structural variability of profiles via a typology and a classification of one-dimensional patterns. More than 80% of the variability between spatial patterns of laser profiles has been summarized by two PCA axes, while four classes of spatial patterns were identified by cluster analysis. Each landscape unit was associated with one or two dominant classes of spatial patterns. These results confirmed the ability of the method to extract landscape scaling properties from complex and large sets of remotely sensed data.     

一种深度小波过程神经网络及在时变信号分类中的应用

针对多通道非线性时变信号分类问题,提出一种基于稀疏自编码器的深度小波过程神经网络(SAE-DWPNN)。通过构建一种多输入/多输出的小波过程神经网络(WPNN),实现对时变信号的多尺度分解和对过程分布特征的初步提取;通过在WPNN隐层之后叠加一个SAE深度网络,对所提取的信号特征进行高层次的综合和表示,并基于softmax分类器实现对时变信号的分类。SAE-DWPNN将现有过程神经网络扩展为深度结构,同时将深度SAE网络在信息处理机制上扩展到时间域,扩展了两类模型的信息处理能力。该网络可提取多通道时序信号的分布特征及其结构特征,并保持样本特征的多样性,提高了对信号时频特性和结构特征的分析能力。文中分析了SAE-DWPNN的性质,给出了综合训练算法。以基于12导联ECG信号的7种心血管疾病分类诊断为例,实验结果验证了模型和算法的有效性。     

心电信号的小波变换识别方法

本文把小波变换应用于心电信号的识别。探讨了伸缩尺度和伪频率(译自pseudo-frequency)之间的关系；利用二进双正交样条小波对室扑信号按Mallat算法进行小波分解；提出了心室扑动和心室颤动信号的小波变换识别方法。     

基于FPGA的多通道心电信号实时滤波与压缩

为了改善多通道心电(ECG)信号滤波的质量和保证数据传输速率,实现实时采集与处理,提出了一种基于现场可编程门阵列(FPGA)的ECG信号滤波和压缩处理方法.对AD采集的ECG信号进行FIR低通和四层Coif1小波滤波处理;由多级树集合分裂(SPIHT)模块进行压缩;将压缩数据经通用串行总线(USB)传入上位机.根据Modelsim仿真和Altera Arria V FPGA的实验结果表明:经过数字滤波与压缩,ECG信号的信噪比(SNR)可提升7.4 dB,数据压缩率可达13.4.方法可实现对64通道的ECG信号实时滤波和压缩处理.     

卡尔曼滤波在心电信号QRS波检测中的尝试

心电信号QRS波的检测方法很多，但在准确性与实时性方面都不太好，该文中将心电信号按照QRS波周期进行分割，利用RLS算法的自适应AR建模，为心电信号建立模型，再利用kalman滤波算法对心电信号进行滤波和预测，在保证R波探测率的同时提高了探测的速度。针对心率不齐或者QRS波周期产生波动的情况，程序中利用各QRS波周期的相似性，求其互相关，以确定周期T，同时对T进行自适应建模，以便对下一周期预测。经过试验，取得了比较好的效果。     

Optimal signal multi-resolution by genetic algorithms to support artificial neural networks for exchange-rate forecasting

Detecting the features of significant patterns from historical data is crucial for good performance in time-series forecasting. Wavelet analysis, which processes information effectively at different scales, can be very useful for feature detection from complex and chaotic time series. In particular, the specific local properties of wavelets can be useful in describing the signals with discontinuous or fractal structure in financial markets. It also allows the removal of noise-dependent high frequencies, while conserving the signal bearing high frequency terms of the signal. However, one of the most critical issues to be solved in the application of the wavelet analysis is to choose the correct wavelet thresholding parameters. If the threshold is small or too large, the wavelet thresholding parameters will tend to overfit or underfit the data. The threshold has so far been selected arbitrarily or by a few statistical criteria.

This study proposes an integrated thresholding design of the optimal or near-optimal wavelet transformation by genetic algorithms (GAs) to represent a significant signal most suitable in artificial neural network models. This approach is applied to Korean won/US dollar exchange-rate forecasting. The experimental results show that this integrated approach using GAs has better performance than the other three wavelet thresholding algorithms (cross-validation, best basis selection and best level tree).     

Optimal selection of wavelet basis function applied to ECG signal denoising

Over the years ElectroCardioGram (ECG) signal has been used to assess the cardiovascular condition of humans. In practice, real time acquisition and transmission of the ECG may contain noise signals superimposed on it. In general, the signal processing algorithms employed for denoising provide optimal performance and eliminate the high frequency noise between any two beats contained in a continuous ECG signal. Despite their optimal performance, the signal processing algorithms significantly attenuate the peaks of characteristics wave of the ECG signal. This paper presents a selection procedure of mother wavelet basis functions applied for denoising of the ECG signal in wavelet domain while retaining the signal peaks close to their full amplitude. The obtained wavelet based denoised ECG signals retain the necessary diagnostics information contained in the original ECG signal.     

A new music-empirical wavelet transform methodology for time–frequency analysis of noisy nonlinear and non-stationary signals

The goal of signal processing is to estimate the contained frequencies and extract subtle changes in the signals. In this paper, a new adaptive multiple signal classification-empirical wavelet transform (MUSIC-EWT) methodology is presented for accurate time–frequency representation of noisy non-stationary and nonlinear signals. It uses the MUSIC algorithm to estimate the contained frequencies in the signal and build the appropriate boundaries to create the wavelet filter bank. Then, the EWT decomposes the time-series signal into a set of frequency bands according to the estimated boundaries. Finally, the Hilbert transform is applied to observe the evolution of calculated frequency bands over time. The usefulness and effectiveness of the proposed methodology are validated using two simulated signals and an ECG signal obtained experimentally. The results demonstrate clearly that the proposed methodology is immune to noise and capable of estimating the optimal boundaries to isolate the frequencies from noise and estimate the main frequencies with high accuracy, especially the closely-spaced frequencies.     

小波变换在心电信号滤波器中的应用研究

讨论了基于小波变换的增强ECG信号的滤波算法,通过处理小波变换的细小尺度和粗大尺度减小了50Hz的工频干扰、基线漂移和随机噪声。实验结果表明了此方法的可行性。     

心电模板构造方法及其在心电去噪中的应用

针对强噪声心电去噪,提出了基于心电模板的去噪方法。首先,将小波变换模极大值方法和信号相干平均技术相结合,构造出心电模板信号。然后给出了两种基于心电模板的强噪声心电去噪方法：直接重建法和间接相关法。最后将该方法与基于小波软、硬阈值去噪方法进行了对比,结果显示所得信号波形平滑度更好、信噪比更高。基于心电模板的去噪方法能够有效去除心电强噪声,为心电弱特征信息成分的准确提取奠定了基础。同时研究提供的心电模板构造方法也可用于其他准周期性生理信号,为强噪声生理信号去噪提供了一种有益思路。     

基于自适应多元多尺度熵的心电信号分类研究

针对目前单通道心电信号识别精度不高,现存多元分解方法效果不佳、多元非线性心电信号分析复杂等问题,提出了一种基于自适应多元多尺度色散熵的心电信号分类方法。首先利用频谱分析,创新性地引入了正弦辅助多元经验模态分解方法,对心电信号进行分解得到多元模态分量;然后结合多模态分解和色散熵的优越性,通过累加多元本征模态分量代替粗粒化采样,提出了自适应多元多尺度色散熵的方法获取特征熵值。最后将特征输入到多个分类器上进行分类,通过实验对比分析,在模拟信号和MIT-BIH数据上验证该方案的有效性。     

基于小波变换的心电信号降噪处理

实测的心电信号不可避免地存在一些强干扰和噪声,为了实现准确地提取反映心电信号的特征信息,该文应用一维离散小波变换实现了对心电信号的降噪处理。实验研究结果表明,该方法能够有效地去除心电信号中的噪声,从而为心电信号特征信息的提取奠定了理论基础。     

小波包滤波在齿轮故障诊断中的应用

为快速准确地判断齿轮故障的类型,提出了小波包滤波和神经网络相结合进行齿轮故障分类的方法。介绍了小波包去噪的原理和神经网络的设计方法,对阈值算法和神经网络优化算法作了改进,得到了不含噪声的信号和准确的故障分类方法。仿真结果表明,基于小波包滤波的神经网络方法具有更高的准确性和稳定性,可以满足工业故障诊断的要求。     

A DSP based frequency domain approach for classification of transmission line faults

This paper employs a digital signal processing (DSP) based frequency domain approach using wavelet multi-resolution analysis (MRA) to overcome difficulties such as fault inception angle, fault impedance and fault distance associated with conventional time domain approach employing voltage and current based measurements for fault classification in case of digital relaying of transmission line. The frequency domain approach for fault classification algorithm uses wavelet MRA technique to extract the features of the current signals based on harmonics generated at the instant of occurrence of fault due to abrupt change of currents in a three phase transmission line. Since choice of particular wavelet plays a vital role for extracting features of generated harmonics, therefore an attempt has been made in this proposed research to extensively investigate using 16 wavelets to establish the superiority of Db4 wavelet over other standard wavelets for accurate fault classification.