Analysis and Design of On-sensor ECG Processors for Realtime Detection of Cardiac Anomalies Including VF,VT, and PVC

Cardiovascular disease remains the main cause of death, and great efforts are spent on the design of ECG (electrocardiogram) body sensors these years. Essential components such as analog frontend and wireless transceivers have been integrated on a compact IC with micro-Watt power consumption. To provide timely warning against the fatal vascular signs, based on the Chaotic Phase Space Differential (CPSD) algorithm, heterogeneous VLSI processors are implemented and integrated to extract the abnormal ECG characteristics for VF (Ventricular Fibrillation), VT (Ventricular Tachycardia) and PVC (Premature Ventricular Contraction). The on-sensor processing reduces 98.0% power of wireless data transmission for raw ECG signals. The application specific processor is designed to accelerate CPSD algorithm with 1.7μW power while the OpenRISC is integrated to provide the system flexibility. The architecture is realized on the FPGA platform to demonstrate the detection of the abnormal ECG signals in realtime.     

基于回归分析理论的辐射源个体识别技术

针对目前辐射源个体识别未能将信号特征与硬件组成相联系的问题，该文使用高阶谱分析和变分模态分解(VMD)两种特征提取手段，进行研究分析，采用围线双谱积分以及改进变分模态分解技术对半实物平台仿真信号以及软件仿真(ADS)输出信号进行特征提取并分析。通过软件仿真定量分析辐射源相位噪声以及功率放大电路非线性失真对信号无意调制特征的影响，对变量进行相关性分析，并对其中显著相关的变量进行回归拟合，得到其相关回归函数。然后利用硬件与特征的相关性，改进传统支持向量机(SVM)分类器，构建相关性权重支持向量机分类器。最后分别以软件仿真输出信号以及半实物仿真平台实测信号为样本进行验证，结果表明，同信噪比下权重支持向量机与传统支持向量机相比分类准确率提升在10%以上。     

Detecting ventricular tachycardia and fibrillation by complexity measure

Sinus rhythm (SR), ventricular tachycardia (VT) and ventricular fibrillation (VF) belong to different nonlinear physiological processes with different complexity. In this study, we present a novel, and computationally fast method to detect VT and VF, which utilizes a complexity measure suggested by Lempel and Ziv [1]. For a specific window length (i.e., the length of data segment to be analyzed), the method first generates a 0-1 string by comparing the raw electrocardiogram (ECG) data to a selected suitable threshold. The complexity measure can be obtained from the 0-1 string only using two simple operations, comparison and accumulation. When the window length is 7 s, the detection accuracy for each of SR, VT, and VF is 100% for a test set of 204 body surface records (34 SR, 85 monomorphic VT, and 85 VF). Compared with other conventional time- and frequency-domain methods, such as rate and irregularity, VF-filter leakage, and sequential hypothesis testing, the new algorithm is simple, computationally efficient, and well suited for real-time implementation in automatic external defibrillators (AED's).     

A simple time domain algorithm for the detection of ventricular fibrillation in electrocardiogram

Ventricular fibrillation (VF) is the most serious variety of arrhythmia which requires quick and accurate detection to save lives. In this paper, we propose a new time domain algorithm, called threshold crossing sample count (TCSC), which is an improved version of the threshold crossing interval (TCI) algorithm for VF detection. The algorithm is based on an important feature of the VF signal which relies on the random behavior of the electrical heart vector. By two simple operations: comparison and count, the technique calculates an effective measure which is used to separate life-threatening VF from other heart rhythms. For assessment of the performance of the algorithm, the method is applied on the complete MIT-BIH arrhythmia and CU databases, and a promising good performance is observed. Seven other classical and new VF detection algorithms, including TCI, have been simulated and comparative performance results in terms of different quality parameters are presented. The TCSC algorithm yields the highest value of the area under the receiver operating characteristic curve (AUC). The new algorithm shows strong potential to be applied in clinical applications for faster and accurate detection of VF.     

Intelligent hybrid approaches for human ECG signals identification

This paper presents hybrid approaches for human identification based on electrocardiogram (ECG). The proposed approaches consist of four phases, namely data acquisition, preprocessing, feature extraction and classification. In the first phase, data acquisition phase, data sets are collected from two different databases, ECG-ID and MIT-BIH Arrhythmia database. In the second phase, noise reduction of ECG signals is performed by using wavelet transform and a series of filters used for de-noising. In the third phase, features are obtained by using three different intelligent approaches: a non-fiducial, fiducial and a fusion approach between them. In the last phase, the classification approach, three classifiers are developed to classify subjects. The first classifier is based on artificial neural network (ANN). The second classifier is based on K-nearest neighbor (KNN), relying on Euclidean distance. The last classifier is support vector machine (SVM) classification accuracy of 95% is obtained for ANN, 98 % for KNN and 99% for SVM on the ECG-ID database, while 100% is obtained for ANN, KNN, and SVM on MIT-BIH Arrhythmia database. The results show that the proposed approaches are robust and effective compared with other recent works.     

An algorithm used for ventricular fibrillation detection without interrupting chest compression

Ventricular fibrillation (VF) is the primary arrhythmic event in the majority of patients suffering from sudden cardiac arrest. Attention has been focused on this particular rhythm since it is recognized that prompt therapy, especially electrical defibrillation, may lead to a successful outcome. However, current versions of automated external defibrillators (AEDs) mandate repetitive interruptions of chest compression for rhythm analyses since artifacts produced by chest compression during cardiopulmonary resuscitation (CPR) preclude reliable electrocardiographic (ECG) rhythm analysis. Yet, repetitive interruptions in chest compression are detrimental to the success of defibrillation. The capability for rhythm analysis without requiring "hands-off" intervals will allow for more effective resuscitation. In this paper, a novel continuous-wavelet-transformation-based morphology consistency evaluation algorithm was developed for the detection of disorganized VF from organized sinus rhythm (SR) without interrupting the ongoing chest compression. The performance of this method was evaluated on both uncorrupted and corrupted ECG signals recorded from AEDs obtained from out-of-hospital victims of cardiac arrest. A total of 232 patients and 31,092 episodes of either VF or SR were accessed, in which 8195 episodes were corrupted by artifacts produced by chest compressions. We also compared the performance of this method with three other established algorithms, including VF filter, spectrum analysis, and complexity measurement. Even though there was a modest decrease in specificity and accuracy when chest compression artifact was present, the performance of this method was still superior to other reported methods for VF detection during uninterrupted CPR.     

Ventricular tachycardia and fibrillation detection by a sequentialhypothesis testing algorithm

An algorithm for detecting ventricular fibrillation (VF) and ventricular tachycardia (VT) by the method of sequential hypothesis testing is presented. The algorithm first generates a binary sequence by comparing the signal to a threshold. The probability distribution of the time intervals of the binary sequence is obtained, and Wald's sequential hypothesis testing procedure is next employed to discriminate the arrhythmias. Sequential hypothesis testing of 85 cases resulted in identification of 1) 97.64% VF and 97.65% VT episodes after 5 s, and 2) 100% identification of both VF and VT after 7 s. The desired false positive and false negative error probabilities can be preprogrammed into the algorithm. An important feature of the sequential method is that extra time for detection can be traded off for improved accuracy, and vice versa.     

A two-stage discrimination of cardiac arrhythmias using a total least squares-based prony modeling algorithm

In this paper, we describe a new approach for the discrimination among ventricular fibrillation (VF), ventricular tachycardia (VT) and superventricular tachycardia (SVT) developed using a total least squares (TLS)-based Prony modeling algorithm. Two features, dubbed energy fractional factor (EFF) and predominant frequency (PF), are both derived from the TLS-based Prony model. In general, EFF is adopted for discriminating SVT from ventricular tachyarrhythmias (i.e., VF and VT) first, and PF is then used for further separation of VF and VT. Overall classification is achieved by performing a two-stage process to the indicators defined by EFF and PF values, respectively. Tests conducted using 91 episodes drawn from the MIT-BIH database produced optimal predictive accuracy of (SVT, VF, VT) = (95.24%, 96.00%, 97.78%). A data decimation process is also introduced in the novel method to enhance the computational efficiency, resulting in a significant reduction in the time required for generating the feature values.     

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。     

多特征分类的PolSAR图像飞机目标检测方法

针对目前有关极化合成孔径雷达(Polarimetric Synthetic Aperture Radar, PolSAR)的飞机目标检测算法虚警较多、自适应性较差的问题,给出一种复杂大场景中PolSAR图像多特征分类的飞机目标检测方法。该方法分为线下分类器训练和飞机目标检测两部分。使用Filter特征选择结合穷举法筛选出分类性能高的飞机极化特征训练SVM (Support Vector Machine, SVM)分类器;利用异化散射功率提取疑似飞机目标,进一步提取多个极化特征送入SVM分类获得检测结果。利用UAVSAR系统采集的多幅实测数据进行实验,并与现有的PolSAR图像飞机目标检测算法进行对比,结果表明该方法能够有效检测出飞机目标,并且虚警和漏警较少,方法自适应性有所提高。      

A Feature Reduced Intrusion Detection System with Optimized SVM Using Big Bang Big Crunch Optimization

The swift proliferation in traffic across computer networks has led to certain types of attacks and intrusions, raising a serious global concern of information security. Attack detection is possible by monitoring and observing occurrences in intrusion detection systems, however these systems tend to suffer from problem of curse of dimensionality, high false alarm rate, high time complexity and low detections. In order to overcome these limitations, we propose a feature reduced intrusion detection system employing optimized SVM as a classifier. Feature Reduction has been performed by fusing ranked features from information gain and chi square in a way that it has helped in retaining only important features and discarding the rest. The study further proposes an optimized version of SVM classifier using Big Bang Big Crunch (BBBC) optimization that simulates the big bang and big crunch theory of evolution of universe. BBBC has helped in finding an optimal set of SVM parameters quickly that are further used for classification. We also experimented with a number of fitness functions for gauging the performance of IDS and propose a new fitness function based on the weighted F1 score of various traffic classes. KDD-99 dataset has been used for experimentation and analysis. The paper further experiments the effects of under-sampling and oversampling of various traffic classes on the proposed IDS performance and recommends that maintaining a desired ratio for a mix of under-sampling and over-sampling of desired classes produces the best results.

     

基于机器视觉的印刷套准识别方法研究*

针对印刷套准检测存在的精度低、速度慢的问题,提取了印刷标志图像的Tamura纹理特征：粗糙度、对比度和方向度,以描述其印刷标志套准或套不准特征;设计了支持向量机的分类器对印刷标志图像进行套准识别,并采用高斯径向基核函数用于非线性数据的分类。实验结果证明,采用建议的印刷标志图像特征提取和分类方法,识别准确率达到90%,识别时间为0.032751秒。本文建议的方法在识别准确率和识别速度上都优于人工检测和文献8的方法。     

基于感兴趣区域和HOG-MBLBP特征的交通标识检测

交通标识检测中样本类别间的不平衡常常导致分类器的检测性能弱化,为了克服这一问题,该文提出一种基于感兴趣区域和HOG-MBLBP融合特征的交通标识检测方法。首先采用颜色增强技术分割提取出自然背景中交通标识所在的感兴趣区域;然后对标识样本库提取HOG-MBLBP融合特征,并用遗传算法对SVM交叉验证进行参数的优化选取,以此来训练和提升SVM分类器性能;最后将提取的感兴趣区域图像的HOG-MBLBP特征送入训练好的SVM多分类器,进行进一步的精确检测和定位,剔除误检区域。在自建的中国交通标识样本库上进行了实验,结果表明所提方法能达到99.2%的分类准确度,混淆矩阵结果也表明了该方法的优越性。     

基于One-ClassSVM的机载塔康测距信息异常检测方法研究

针对多源导航信息融合系统中导航传感器数据保障的问题，本文提出了一种基于 One-Class SVM 的机载塔康测距信息异常检测方法。首先，提取机载塔康测距信息的时域参数构成特征样本空间；然后，采用 One-Class SVM 训练出机载塔康测距信息正常状态时的模型，通过发现非正常状态的样本进行异常检测。利用模拟的机载塔康测距数据进行方法验证，实验结果表明：该异常检测方法对机载塔康测距信息中的噪声有一定的鲁棒性，可以满足实际应用的需要。     

Novel ECG Signal Classification Based on KICA Nonlinear Feature Extraction

Electrocardiogram (ECG) signal feature extraction is important in diagnosing cardiovascular diseases. This paper presents a new method for nonlinear feature extraction of ECG signals by combining principal component analysis (PCA) and kernel independent component analysis (KICA). The proposed method first uses PCA to decrease the dimensions of the ECG signal training set and then employs KICA to calculate the feature space for extracting the nonlinear features. Support vector machine (SVM) is utilized to determine the nonlinear features of the ECG signal classification. Genetic algorithm is also used to optimize the SVM parameters. The proposed method is advantageous because it does not require a huge amount of sampling data, and this technique is better than traditional strategies to select optimal features in the multi-domain feature space. Computer simulations reveal that the proposed method yields more satisfactory classification results on the MIT–BIH arrhythmia database, reaching an overall accuracy of 97.78 %.     

Real-time sleep apnea detection by classifier combination

To find an efficient and valid alternative of polysomnography (PSG), this paper investigates real-time sleep apnea and hypopnea syndrome (SAHS) detection based on electrocardiograph (ECG) and saturation of peripheral oxygen (SpO(2)) signals, individually and in combination. We include ten machine-learning algorithms in our classification experiment. It is shown that our proposed SpO (2) features outperform the ECG features in terms of diagnostic ability. More importantly, we propose classifier combination to further enhance the classification performance by harnessing the complementary information provided by individual classifiers. With our selected SpO(2) and ECG features, the classifier combination using AdaBoost with Decision Stump, Bagging with REPTree, and either kNN or Decision Table achieves sensitivity, specificity, and accuracy all around 82% for a minute-based real-time SAHS detection over 25 sleep-disordered-breathing suspects' full overnight recordings.     

基于改进熵特征的心律失常分类研究

心律失常类型的判断对心血管疾病的防治十分重要,针对波动散布熵(multiscale f luctuation dispersion entropy,FDE)在进行心律失常分类识别时尺度单一、不能全面反映心律失常 信息等不足,通 过改进FDE特征,提出一种基于自适应多尺度波动散布熵(adaptive multiscale fluctuati on dispersion entropy,AMFDE)的心律失常分类方法。首先在计算FDE特征前利用优化K值的变分模态分 解(variational mode decomposition,VMD)对信号进行分解,以获取预设数量的固有模态 分量(IMF),然后 提取各尺度子序列的FDE作为分类特征,并采用遗传算法(genetic algorithm, GA)对支持向量机(SVM)的惩罚 因子c和 核函数参数g进行寻优,最后通过GA-SVM模型进行模式识别。计算结 果表明, 所提方法对4类心律识 别的平均准确率达到95.3%,灵敏度达到95.4% ,特 异性达到 98.4%,相比自适应多尺度散布熵(adaptive multiscale dispersion entropy, AMDE)等其他方法优势明显,可以实现对心 电(electrocardiogram, ECG)信号的准确分类。     

一种基于SVM主动学习的卡通视频检测方法

通过分析卡通与非卡通视频在视觉上的差异,对视频片断提取了MPEG-7描述子等8组视觉特征来构造卡通视频的特征空间;并将主动相关反馈技术引入到支撑向量机(SVM)算法中,设计了一种基于主动学习的卡通视频检测分类方法。利用大量实际视频片断所做的测试实验结果表明,该文选取的特征对卡通和非卡通视频有较好的区分能力;且与单纯的SVM算法以及传统相关反馈和SVM算法结合的方法相比,该文算法在检测性能上有较大的优势。     

Electrocardiogram based neonatal seizure detection

A method for the detection of seizures in the newborn using the electrocardiogram (ECG) signal is presented. Using a database of eight recordings, a method was developed for automatically annotating each 1-min epoch as "nonseizure" or "seizure". The system uses a linear discriminant classifier to process 41 heartbeat timing interval features. Performance assessment of the method showed that on a patient-specific basis an average accuracy of 70.5% was achieved in detecting seizures with associated sensitivity of 62.2% and specificity of 71.8%. On a patient-independent basis the average accuracy was 68.3% with sensitivity of 54.6% and specificity of 77.3%. Shifting the decision threshold for the patient-independent classifier allowed an increase in sensitivity to 78.4% at the expense of decreased specificity (51.6%), leading to increased false detections. The results of our ECG-based method are comparable with those reported for EEG-based neonatal seizure detection systems and offer the benefit of an easier acquisition methodology for seizure detection.     

Automated processing of the single-lead electrocardiogram for the detection of obstructive sleep apnoea

A method for the automatic processing of the electrocardiogram (ECG) for the detection of obstructive apnoea is presented. The method screens nighttime single-lead ECG recordings for the presence of major sleep apnoea and provides a minute-by-minute analysis of disordered breathing. A large independently validated database of 70 ECG recordings acquired from normal subjects and subjects with obstructive and mixed sleep apnoea, each of approximately eight hours in duration, was used throughout the study. Thirty-five of these recordings were used for training and 35 retained for independent testing. A wide variety of features based on heartbeat intervals and an ECG-derived respiratory signal were considered. Classifiers based on linear and quadratic discriminants were compared. Feature selection and regularization of classifier parameters were used to optimize classifier performance. Results show that the normal recordings could be separated from the apnoea recordings with a 100% success rate and a minute-by-minute classification accuracy of over 90% is achievable.