基于集成学习的室性早博识别方法

本文提出了一种集成学习方法以提升室性早搏的识别性能.MIT-BIH两个通道的数据分别经过卷积神经网络进行室性早搏心拍分类,然后按照融合规则对分类结果进行融合决策,其准确率、灵敏度和特异性分别为99.91%、98.76%、99.97%,优于已有算法的室性早搏心拍分类结果.此外,面向临床应用,本文还利用卷积神经网络和诊断规则相结合的方法实现了病人间室性早搏识别实验,在有14万多条记录的数据集上,取得的准确率、灵敏度及特异性分别为97.87%、87.94%、98.02%,验证了该算法的有效性.     

Machine learning approach to recognize ventricular arrhythmias using VMD based features

The occurrence of life-threatening ventricular arrhythmias (VAs) such as Ventricular tachycardia (VT) and Ventricular fibrillation (VF) leads to sudden cardiac death which requires detection at an early stage. The main aim of this work is to develop an automated system using machine learning tool for accurate prediction of VAs that may reduce the mortality rate. In this paper, a novel method using variational mode decomposition (VMD) based features and C4.5 classifier for detection of ventricular arrhythmias is presented. The VMD model was used to decompose the electrocardiography (ECG) signals to extract useful informative features. The method was tested for ECG signals obtained from PhysioNet database. Two standard databases i.e. CUDB (Creighton University Ventricular Tachyarrhythmia Database) and VFDB (MIT-BIH Malignant Ventricular Ectopy Database) were considered for this work. A set of time–frequency features were extracted and ranked by the gain ratio attribute evaluation method. The ranked features are subjected to support vector machine (SVM) and C4.5 classifier for classification of normal, VT and VF classes. The best detection was obtained with sensitivity of 97.97%, specificity of 99.15%, and accuracy of 99.18% for C4.5 classifier with a 5 s data analysis window. These results were better than SVM classifier result having an average accuracy of 86.87%. Hence, the proposed method demonstrates the efficiency in detecting the life-threatening VAs and can serve as an assistive tool to clinicians in the diagnosis process.

     

Detection of premature ventricular contraction arrhythmias in electrocardiogram signals with kernel methods

In this paper, we propose to investigate the capabilities of two kernel methods for the detection and classification of premature ventricular contractions (PVC) arrhythmias in Electrocardiogram (ECG signals). These kernel methods are the support vector machine and Gaussian process (GP). We propose to study these two classifiers with various feature representations of ECG signals, such as morphology, discrete wavelet transform, higher-order statistics, and S transform. The experimental results obtained on 48 records (i.e., 109,887 beats) of the MIT-BIH Arrhythmia database showed that for all feature representation adopted in this work, the GP detector trained only with 600 beats from PVC and Non-PVC classes can provide an overall accuracy and a sensitivity above 90 % on 20 records (i.e., 49,774 beats) and 28 records (i.e., 60,113 beats) seen and unseen, respectively, during the training phase.     

支持向量机方法在文本分类中的改进

提出了一种应用于文本分类的KNN和SVM相结合的算法,将SVM近似看成每类只有一个代表点的1NN分类器,对于待识别样本,如果其离支持向量机的最优分界面较远,则用SVM分类;如果其离分界面较近,采用KNN对测试样本分类,将每个支持向量作为代表点,计算待识别样本和每个支持向量的距离对其作出判断.该算法综合了KNN和SVM在分类问题中的优势,既有效地降低了分类候选的数目,又提高了文本分类的精度.最后用实验验证了该算法的有效性.     

基于多分类器投票组合的语音情感识别

为了提高语音情感的正确识别率,提出一种基于多分类器投票组合的语音情感识别新方法.在提取情感语音的韵律特征和音质特征基础上,利用投票方法将支持向量机、K近邻法和人工神经网络三种分类器构成组合分类器,实现对汉语生气、高兴、悲伤和惊奇4种主要情感类型的识别.实验结果表明,与使用单一分类器相比,组合分类器对语音情感的识别取得了87.4%的平均正确识别率,识别效果优于单一分类器.     

基于GPCA的KNNY与SVM融合的人脸识别方法

针对K近邻和支持向量机人脸识别率较低的问题,采用一种KNN和SVM融合的识别方法。提出了一种Gabor小波和主成分分析进行人脸特征提取,KNN-SVM进行分类的人脸识别方法。基于ORL和YALE人脸库中进行实验,结果表明该算法较KNN和SVM中任何一个的识别率都要高,且识别率最高可达到98.89%。     

Automatic classification of heartbeats using ECG morphology and heartbeat interval features

A method for the automatic processing of the electrocardiogram (ECG) for the classification of heartbeats is presented. The method allocates manually detected heartbeats to one of the five beat classes recommended by ANSI/AAMI EC57:1998 standard, i.e., normal beat, ventricular ectopic beat (VEB), supraventricular ectopic beat (SVEB), fusion of a normal and a VEB, or unknown beat type. Data was obtained from the 44 nonpacemaker recordings of the MIT-BIH arrhythmia database. The data was split into two datasets with each dataset containing approximately 50,000 beats from 22 recordings. The first dataset was used to select a classifier configuration from candidate configurations. Twelve configurations processing feature sets derived from two ECG leads were compared. Feature sets were based on ECG morphology, heartbeat intervals, and RR-intervals. All configurations adopted a statistical classifier model utilizing supervised learning. The second dataset was used to provide an independent performance assessment of the selected configuration. This assessment resulted in a sensitivity of 75.9%, a positive predictivity of 38.5%, and a false positive rate of 4.7% for the SVEB class. For the VEB class, the sensitivity was 77.7%, the positive predictivity was 81.9%, and the false positive rate was 1.2%. These results are an improvement on previously reported results for automated heartbeat classification systems.     

A wavelet-based ECG delineator: evaluation on standard databases

In this paper, we developed and evaluated a robust single-lead electrocardiogram (ECG) delineation system based on the wavelet transform (WT). In a first step, QRS complexes are detected. Then, each QRS is delineated by detecting and identifying the peaks of the individual waves, as well as the complex onset and end. Finally, the determination of P and T wave peaks, onsets and ends is performed. We evaluated the algorithm on several manually annotated databases, such as MIT-BIH Arrhythmia, QT, European ST-T and CSE databases, developed for validation purposes. The QRS detector obtained a sensitivity of Se = 99.66% and a positive predictivity of P+ = 99.56% over the first lead of the validation databases (more than 980,000 beats), while for the well-known MIT-BIH Arrhythmia Database, Se and P+ over 99.8% were attained. As for the delineation of the ECG waves, the mean and standard deviation of the differences between the automatic and manual annotations were computed. The mean error obtained with the WT approach was found not to exceed one sampling interval, while the standard deviations were around the accepted tolerances between expert physicians, outperforming the results of other well known algorithms, especially in determining the end of T wave.     

Effect of Multiscale PCA De-noising in ECG Beat Classification for Diagnosis of Cardiovascular Diseases

Current trends in clinical applications demand automation in electrocardiogram (ECG) signal processing and heart beat classification. This paper examines the design of an effective recognition method to diagnose heart diseases. The proposed method consists of three main modules: de-noising module, feature extraction module, and classifier module. In the de-noising module, multiscale principal component analysis (MSPCA) is used for noise reduction of the ECG signals. In the feature extraction module, autoregressive (AR) modeling is used for extracting features. In the classifier module, different classifiers are examined such as simple logistic, k-nearest neighbor, multilayer perceptron, radial basis function networks, and support vector machines. Different experiments are carried out using the MIT-BIH arrhythmia database to classify different ECG heart beats and the performance of the proposed method is evaluated in terms of several standard metrics. The experimental results show that the proposed method is able to reduce noise from the noisy ECG signals more accurately in comparison to previous methods. The numerical results indicated that the proposed algorithm achieved 99.93 % of the classification accuracy using MSPCA de-noising and AR modeling.     

基于卷积长短时记忆网络的心律失常分类方法

心律失常等慢性心血管疾病严重影响人类健康,采用心电信号(ECG)实现心律失常自动分类可有效提高该类疾病的诊断效率,降低人工成本。为此,该文基于1维心电信号,提出一种改进的长短时记忆网络(LSTM)方法实现心律失常自动分类。该方法首先设计深层卷积神经网络(CNN)对心电信号进行深度编码,提取心电信号形态特征。其次,搭建长短时记忆分类网络实现基于心电信号特征的心律失常自动分类。基于MIT-BIH心律失常数据库进行的实验结果表明,该方法显著缩短分类时间,并获得超过99.2%的分类准确率,灵敏度等评价参数均得到不同程度的提高,满足心电信号自动分类实时高效的要求。     

基于希尔伯特振动分解和卷积神经网络的融合特征心电识别算法

在信息高速发展的当代社会,5G技术的问世将极大地助力社会经济和信息化发展,而隐私安全和信息安全愈发得到重视,因此公众会对身份的识别技术提出了更高要求。然而,传统基于密码、ID卡以及新型的基于人脸和指纹的识别方法存在易丢失、遗忘和窃取或易于伪造和获取复制等问题而存在极大的安全隐患。为提高身份识别的可靠性和准确率,提出了基于希尔伯特振动分解和卷积神经网络的融合特征心电图信号识别算法。首先采用基于重叠组收缩阈值算法和平移不变的消噪算法对含噪心电信号去噪,其次利用盲源分割技术将心电信号分割成固定时长的心电片段,再次采用基于希尔伯特振动分解的时频分析方法获得心电信号的时频表示图,通过提出的心电残差卷积神经网络对时频表示图实现特征提取和降维,最后通过Softmax分类器实现分类和识别。以Physionet数据库的ECG-ID数据集验证提出算法的性能,采用10折交叉验证法得到平均识别率为99.08%。结果表明,提出的心电识别算法具有高效的识别性能和良好的应用前景。     

ECG compression method based on adaptive quantization of main wavelet packet subbands

In this study, a new compression algorithm for ECG signal is proposed based on selecting important subbands of wavelet packet transform (WPT) and applying subband-dependent quantization algorithm. To this end, first WPT was applied on ECG signal and then more important subbands are selected according to their Shannon entropy. In the next step, content-based quantization and denoising method are applied to the coefficients of the selected subbands. Finally, arithmetic coding is employed to produce compressed data. The performance of the proposed compression method is evaluated using compression rate (CR), percentage root-mean-square difference (PRD) as signal distortion, and wavelet energy-based diagnostic distortion (WEDD) as diagnostic distortion measures on MIT-BIH Arrhythmia database. The average CR of the proposed method is 29.1, its average PRD is <2.9 % and WEDD is <3.2 %. These results demonstrated that the proposed method has a good performance compared to the state-of-the-art compression algorithms.     

基于自适应快速S变换和XGBoost的心电信号精确快速分类方法

针对心电信号(ECG)传统分类方法效率较低的问题,该文提出一种基于自适应快速S变换(AFST)和XGBoost的心电信号精确快速分类方法。该方法首先通过快速定位算法确定心电信号特征频率点,再根据特征频率点自适应调节S变换窗宽因子,增强S变换的时频分辨率的同时避免迭代计算,大大减少运行时间。其次,基于自适应快速S变换的时频矩阵提取12个特征量来表征5种心电信号的特征信息,特征向量维数低,识别能力强。最后,利用XGBoost算法对特征向量进行识别。MIT-BIH心律失常数据库和患者实测数据验证表明,该方法显著地缩短了分类时间,对5种心电信号的分类准确率分别为99.59%和97.32%,适用于实际检测系统中心律失常疾病的快速诊断。     

Enhanced real-time ECG coder for packetized telecardiology applications.

A new real-time compression method for electrocardiogram (ECG) signals has been developed based on the wavelet transform approach. The method is specifically adaptable for packetized telecardiology applications. The signal is segmented into beats and a beat template is subtracted from them, producing a residual signal. Beat templates and residual signals are coded with a wavelet expansion. Compression is achieved by selecting a subset of wavelet coefficients. The number of selected coefficients depends on a threshold which has different definitions depending on the operational mode of the coder. Compression performance has been tested using a subset of ECG records from MIT-BIH Arrhythmia database. This method has been designed for real-time packetized telecardiology scenarios both in wired and wireless environments.     

SAR图像球流形局部嵌入建模及其分类方法

地物目标建模是合成孔径雷达(Synthetic Aperture Rader, SAR)图像解译和应用的关键技术之一。近年来,基于流形学习的建模方法得到发展,可望适用于依据微波散射机理成像的SAR图像建模。本文采用球流形嵌入(SLE)方法来实现SAR地物目标建模。该方法实质上是对SAR图像的高维描述或表达进行非线性降维,得到相应的低维流形结构,其分量就是SAR图像的本质特征,由于削弱了原始高维表达中的冗余信息,可用来更加精确地描述和解译地物目标,同时由于维数的降低,大大降低了计算复杂度。为验证其有效性,本文将其应用于SAR图像场景分类,采用简单的K最近邻(K nearest neighbor, KNN)分类器和支持向量机(Support Vector Machine, SVM)分类器。实验结果证明基于本文方法对SAR图像地物目标建模是有效的,有着良好的应用前景。      

Denoising and R-Peak Detection of Electrocardiogram Signal Based on EMD and Improved Approximate Envelope

The electrocardiogram (ECG ) signal is prone to various high and low frequency noises, including baseline wandering and power-line interference, which become the source of errors in QRS and in other extracted features. This paper presents a new ECG signal-processing approach based on empirical mode decomposition (EMD) and an improved approximate envelope method. To reduce the number of the initial intrinsic mode functions (IMFs), a Butterworth lowpass filter is used to eliminate high frequency noises before the EMD. To correct baseline wandering and to eliminate low frequency noises, the two last-order IMFs are abandoned. An improved approximate envelope is proposed and applied after the Hilbert transform to enhance the energy of QRS complexes and to suppress unwanted P/T waves and noises. Then, an algorithm based on the slope threshold is used for R-peak detection. The proposed denoising and R-peak detection algorithm are validated using the MIT-BIH Arrhythmia Database. The simulation results show that the proposed method can effectively eliminate the Gaussian noise, baseline wander, and power-line interference added to the ECG signal. The method can also function reliably even under poor signal quality and with long P and T peaks. The QRS detector has an average sensitivity of Se=99.94 % and a positive predictivity of +P=99.87 % over the first lead of the MIT-BIH Arrhythmia Database.     

基于心电与光电容积脉搏波特征层融合的身份识别方法

针对单模态的心电信号(ECG)或光电容积脉搏波信号(PPG)识别技术中存在的精度不高,未考虑类内相关性等问题,该文提出基于判别相关分析法(DCA)对ECG与PPG组合特征矩阵进行特征层融合以及对K-最近邻(KNN)和支持向量机(SVM)分类器在决策层融合的识别方法。实验结果表明,使用融合特征(ECG-PPG)与融合分类器(KNN-SVM)的方法对23名受试者进行分类识别的准确率可以达到98.2%,识别精度在常规环境下优于单模态识别。为多模生物特征身份识别提供了一种有效模型。     

基于小波分解和1D-GoogLeNet的心律失常检测

心电图(ECG)信号的准确分类对于心脏病的自动诊断非常重要。为了实现对心律失常的智能分类,该文提出一种基于小波分解和1D-GoogLeNet的精确分类方法。在该方法中,利用Db6小波对ECG信号进行8级分解,得到既含时域信息又有频域信息的多维数据。随后,分解的样本用作1D-GoogLeNet的输入训练该模型。在提出的1D-GoogLeNet模型中,借鉴Inception在图像特征提取中的优异性能,将2维卷积变换为1维卷积学习ECG的特征,并且简化各个Inception的结构,降低模型参数。该文提出的神经网络分类器能够有效缓解计算效率低、收敛困难和模型退化的问题。在实验中,选用MIT-BIH心律失常数据集测试所提模型的性能,对比了信号的不同分解分量组合作为输入的检测结果,当输入数据由{d2-d7}组合时,所提1D-GoogLeNet模型可以达到96.58%的平均准确率。此外,还对比了该模型与未经结构优化的简单1维GoogLeNet在数据集上的表现,前者在准确率上比后者提高了4.7%,训练效率提高了118%。     

基于支持向量机的极化SAR图像分类

与传统最大似然（ML）分类器相比，支持向量机（SVM）在小训练样本时仍具有良好的分类性能，目前已广泛应用于多个领域。该文在极化SAR特征提取的基础上，将SVM应用于极化SAR图像分类，分析了分类器参数对分类性能的影响。利用NASA／JPL实验室AIRSAR系统的L波段旧金山全极化SAR数据比较了SVM和ML的分类性能，并进一步给出了基于SVM的国内某地区双极化SAR图像分类结果。     

自适应CA-CFAR的心电信号R波检测技术

针对心电信号R波的突变特性,利用雷达信号的检测方法,本文提出一种自适应单元平均恒虚警率(cell averaging-constant false alarm rate, CA-CFAR)的R波检测方法。首先利用滤波器组对心电信号进行预处理;然后将预处理后的信号利用自适应CA-CFAR检测判决;最后由心电信号R波的间隔特性做一个不应期剔除规则的处理,得到R波的定位。对美国麻省理工学院提供的MIT-BIH数据库中心电图(Electrocardiograph, ECG)信号仿真,实验证明,自适应参考单元的CA-CFAR对MIT-BIH的ECG信号R波检测的精准率为99.842%,检测误差为0.354%。实测数据表明了算法的有效性和适用性。