基于小波的心电信号有效性检测方法

目的：检测采集到的信号是否为有效心电信号,提高后续心电诊断和分析的准确率。方法：将采集到的信号进行预处理,即去噪处理,主要抑制基线漂移,50Hz工频及其谐波干扰和肌电干扰;取滑动窗长度为4s,检测该段内信号是否有效。为了验证算法的准确率及对不同心电波形是否具有普遍适用性,对MIT-BIH Arrhythmia Database中48个记录,CU及MIT-BIH Noise Stress Test Database中部分记录进行了仿真、验证。结果：仿真实验证明该方法能正确区分有效和无效信号,错检率较低,实现简单,适合实时处理。结论：本方法准确率高,能减少后续心电诊断和分析的计算量并提高准确率,特别是对室颤检测,符合心电分析的要求。     

基于小波特征分析的手指动作识别研究

目的:本文利用表面肌电(sEMG)信号来研究多种手指组合动作的识别问题。方法:在对采集的四个通道sEMG信号进行降噪预处理的基础上,采用移动加窗处理方法来提取关于手指运动状态的信号活动段,再分析各个信号活动段的小波系数统计特征,进而利用多类支持向量机(SVM)分类算法来实现手指组合动作的识别。结果:动作识别率最高达到100%。结论:所采用方法能够有效地识别多种手势动作,并为后续基于肌电信号的实时人机接口系统的研究奠定了理论基础。     

基于小波特征分析的手指动作识别研究

目的：本文利用表面肌电（sEMG）信号来研究多种手指组合动作的识别问题。方法：在对采集的四个通道sEMG信号进行降噪预处理的基础上,采用移动加窗处理方法来提取关于手指运动状态的信号活动段,再分析各个信号活动段的小波系数统计特征,进而利用多类支持向量机（SVM分类算法来实现手指组合动作的识别。结果：动作识别率最高达到100％。结论：所采用方法能够有效地识别多种手势动作,并为后续基于肌电信号的实时人机接口系统的研究奠定了理论基础。     

基于支持向量机的表面肌电信号和加速度融合跌倒识别方法

目的:针对老人易跌倒和跌倒过后可能产生严重后果这一现实问题,通过将表面肌电信号和加速度融合,进一步优化采用支持向量机分类器下的包含跌倒在内的几种不同动作的分类效果。方法:提出基于表面肌电和加速度信号融合的跌倒识别算法,首先采集股直肌,股内侧肌,胫骨前肌和腓肠肌的表面肌电信号以及位于腰部的三轴加速度信号作为实验数据,然后利用滑动窗口法提取表面肌电和加速度信号的均方根值,最后针对人体日常活动和跌倒的运动特征,构建了支持向量机的分类器。结果:实验数据共计320组数据,包括3种日常活动和向前跌倒,其中160组数据作为训练集,另外160组数据作为测试集。对4种动作进行识别实验,算法的准确度为93.23%、灵敏度为92.4%、特异度为100%,达到了良好的分类效果。结论:基于支持向量机的表面肌电信号和加速度融合的跌倒识别算法分类效果良好,对于老人跌倒防护具有现实意义。     

基于多策略滑动伸缩窗特征提取方法预测蛋白质同源寡聚体

寡聚蛋白质广泛地参与多种生命活动,对其预测研究有重要的意义。文章从蛋白质序列出发,提出多策略滑动伸缩窗特征提取方法,采用“ 一对一”的多类分类策略,对蛋白质同源寡聚体进行预测研究。结果表明,在Jackknife检验下,基于支持向量机的多策略滑动伸缩窗特征和氨基酸组成成分构成的特征集在加权情况下,其总分类精度最高达到了75.37%,比单纯的氨基酸组成成分法提高10.05%,比参考文献最好特征BG_Zhang提高了3.82%。 说明多策略滑动伸缩窗特征提取方法对于蛋白质同源寡聚体分类,是一种非常有效的特征提取方法。     

心电向量均值数学模型及其Matlab实现

目的:探讨健康人群心电向量活动规律,建立带有标准差的QRS环心电向量均值数学模型,运用Matlab软件编写心电向量模型的标准源代码程序,为临床诊断领域提供科学理论依据和实现方法.方法:运用概率论与数理统计理论,抽取样本数据并进行统计分析,利用数学建模理论、借助数学软件Matlab的可视化功能实现数形转换,绘制带有标准差边界的心电向量曲线.结果:样本心电向量数据变化具有一定统计规律,QRS环具有在第一卦限(或象限)集中的趋势.心电向量活动规律与性别有关.结论:心电向量均值数学模型既能比较地准确反应出健康人群心电向量活动规律,同时模型也能较好检测出心脏病患者的异常状况.模型的使用范围可以推广到其他健康人群.该模型不仅为临床诊断提供科学的理论依据,也为心电向量四维数学模型的构建奠定坚实的理论基础。     

基于线性判别分析的基因表达数据分类方法研究

由于基因表达数据高属性维、低样本维的特点,Fisher分类器对该种数据分类性能不是很高。本文提出了Fisher的改进算法Fisher-List。该算法独特之处在于为每个类别确定一个决策阀值,每个阀值既包含总体样本信息,又含有某些对分类至关重要的个体样本信息。本文用实验证明新算法在基因表达数据分类方面比Fisher、LogitBoost、AdaBoost、k-近邻法、决策树和支持向量机具有更高的性能。     

基因表达数据小波降噪与肿瘤识别

建立了基于小波降噪和支持向量机的结肠癌基因表达数据肿瘤识别模型.对试验数据进行小波分解,并利用交叉验证的方法计算试验样本的平均分类准确率,确定小波函数与小波分解层数;引入能量阈值方法对小波分解系数进行阈值处理,达到降噪的目的;提出了基因分类贡献率与主成分分析结合的方法,提取结肠癌样本数据特征;利用支持向量机强大的非线性映射能力,实现对结肠癌样本数据的非线性分类.为了减弱样本集的划分对分类准确率的影响,本文采取Jackknife检验方法对支持向量分类器的分类器检验,其分类准确率为96.77%.试验结果证明了该方法的有效性,该方法对结肠癌的识别具有一定的参考价值.     

基于等度规映射的大鼠初级视觉皮层视像整体特征识别

整体特征是视觉信息的基本特征之一。为了认知大鼠初级视觉皮层神经元对视像整体特征的处理机制,首先给出确定有效响应时间区间的方法,确定了有效响应时间区间;然后,确定有效响应时间区间内、不同整体特征视像刺激下的神经元发放特征,并利用等度规映射进行了特征整合,进而对整合后的特征使用支持向量机进行分类;最后,将该方法应用于大鼠的初级视觉皮层视像整体特征识别,并将其分类结果分别与使用主成分分析法进行特征整合以及直接统计神经元发放特征的分类结果进行了对比。对比表明:该方法较其他两种方法,对于整体特征的识别准确率均有不同程度的明显提高。     

多生理参数监护系统设计

本研究开发一种基于Windows的多项生理参数的监护系统,并集成到下肢康复训练系统中对病人的生理信息实时监测。该系统由生理参数传感器检测人体生理信号通过单片机处理,最终通过串口通信发送到上位机进行实时显示。该设计可为动态心电、血压、血氧、体温等多参数人体信号的检测建立一个软硬件系统平台,经测试,该系统稳定且实用性强。     

SVM-based algorithm for recognition of QRS complexes in electrocardiogram

Among all electrocardiogram (ECG) components, the QRS complex is the most significant feature. This paper presents a new algorithm for recognition of QRS complexes in the electrocardiogram (ECG) based on support vector machine (SVM). Digital filtering techniques are used to remove power line interference and baseline wander in the ECG signal. SVM is used as a classifier to delineate QRS and non-QRS regions. Algorithm performance was evaluated against the standard CSE ECG database. The results indicated that the algorithm achieved 99.3% of the detection rate. The percentage of false positive and false negative was 12.4 and 0.7% respectively. It could function reliably even under the condition of poor signal quality of the ECG signal.     

SNPs Identified as Modulators of ECG Traits in the General Population Do Not Markedly Affect ECG Traits during Acute Myocardial Infarction nor Ventricular Fibrillation Risk in This Condition

Background

Ventricular fibrillation (VF) in the setting of acute ST elevation myocardial infarction (STEMI) is a leading cause of mortality. Although the risk of VF has a genetic component, the underlying genetic factors are largely unknown. Since heart rate and ECG intervals of conduction and repolarization during acute STEMI differ between patients who do and patients who do not develop VF, we investigated whether SNPs known to modulate these ECG indices in the general population also impact on the respective ECG indices during STEMI and on the risk of VF.

Methods and Results

The study population consisted of participants of the Arrhythmia Genetics in the NEtherlandS (AGNES) study, which enrols patients with a first STEMI that develop VF (cases) and patients that do not develop VF (controls). SNPs known to impact on RR interval, PR interval, QRS duration or QTc interval in the general population were tested for effects on the respective STEMI ECG indices (stage 1). Only those showing a (suggestive) significant association were tested for association with VF (stage 2). On average, VF cases had a shorter RR and a longer QTc interval compared to non-VF controls. Eight SNPs showed a trend for association with the respective STEMI ECG indices. Of these, three were also suggestively associated with VF.

Conclusions

RR interval and ECG indices of conduction and repolarization during acute STEMI differ between patients who develop VF and patients who do not. Although the effects of the SNPs on ECG indices during an acute STEMI seem to be similar in magnitude and direction as those found in the general population, the effects, at least in isolation, are too small to explain the differences in ECGs between cases and controls and to determine risk of VF.     

A wavelet optimization approach for ECG signal classification

Wavelets have proved particularly effective for extracting discriminative features in ECG signal classification. In this paper, we show that wavelet performances in terms of classification accuracy can be pushed further by customizing them for the considered classification task. A novel approach for generating the wavelet that best represents the ECG beats in terms of discrimination capability is proposed. It makes use of the polyphase representation of the wavelet filter bank and formulates the design problem within a particle swarm optimization (PSO) framework. Experimental results conducted on the benchmark MIT/BIH arrhythmia database with the state-of-the-art support vector machine (SVM) classifier confirm the superiority in terms of classification accuracy and stability of the proposed method over standard wavelets (i.e., Daubechies and Symlet wavelets).     

Identification of movement-related cortical potentials with optimized spatial filtering and principal component analysis

We present a method for detecting movement intention from multichannel electroencephalographic (EEG) recordings. Movement intention is expressed as a slow negative deflection in amplitude of the EEG signal recorded above the motor cortex. This deflection is known as a movement-related cortical potential (MRCP). Detection of movement intention implies discrimination between MRCPs and noise. The signal-to-noise ratio of MRCPs was improved by an optimized spatial filter. Features were extracted with principal component analysis or locality preserving projections from the spatially filtered signals and classification between MRCPs and noise was performed with a k-nearest neighbors algorithm, modified by adjusting the decision rule to improve specificity, and a support vector machine approach. In one representative subject the sensitivity and specificity in detection were in the range 80–90% and 98–99.5%, respectively. The method seems promising for the development of asynchronous brain–computer interfaces (BCIs) based on MRCPs.     

Classification of electrocardiogram signals with support vector machines and genetic algorithms using power spectral features

This paper proposes a new power spectral-based hybrid genetic algorithm-support vector machines (SVMGA) technique to classify five types of electrocardiogram (ECG) beats, namely normal beats and four manifestations of heart arrhythmia. This method employs three modules: a feature extraction module, a classification module and an optimization module. Feature extraction module extracts electrocardiogram's spectral and three timing interval features. Non-parametric power spectral density (PSD) estimation methods are used to extract spectral features. Support vector machine (SVM) is employed as a classifier to recognize the ECG beats. We investigate and compare two such classification approaches. First they are specified experimentally by the trial and error method. In the second technique the approach optimizes the relevant parameters through an intelligent algorithm. These parameters are: Gaussian radial basis function (GRBF) kernel parameter σ and C penalty parameter of SVM classifier. Then their performances in classification of ECG signals are evaluated for eight files obtained from the MIT–BIH arrhythmia database. Classification accuracy of the SVMGA approach proves superior to that of the SVM which has constant and manually extracted parameter.     

Development of SVM based classification techniques for the delineation of wave components in 12-lead electrocardiogram

The automatic detection of electrocardiogram (ECG) waves namely P, QRS and T-wave is important to cardiac disease diagnosis. This paper presents an application of support vector machine (SVM) as a classifier for the delineation of ECG wave components in the 12-lead ECG signal. Digital filtering techniques are used to remove power line interference and baseline wander present in the ECG signal. Gradient of the filtered ECG signal is used as a feature for the detection of QRS-complexes, P- and T-waves. The performance of the algorithm is validated using original 12-lead ECG recordings from the standard CSE ECG database. Significant detection rate is achieved. The percentage of false positive and false negative detection is low. The method successfully detects all kind of morphologies of QRS-complexes, P- and T-waves. The onsets and offsets of the detected QRS-complexes, P- and T-waves are found to be within the tolerance limits given in CSE library.     

A case of cardiac sarcoidosis mimicking Brugada syndrome

A 17-year-old boy was admitted for management of ventricular fibrillation (VF) with intermittent Brugada pattern on ECG. On evaluation, cardiac MRI revealed myocardial scar and mediastinal lymphadenopathy. 18-Fluorodeoxyglucose positron emission tomography scan showed inflammation in the heart, lungs, and lymph nodes. He was diagnosed as a case of cardiac sarcoidosis (CS) and treated with steroids. However, there was a reactivation of cardiac inflammation and the development of a second VF storm. Following catheter ablation, the patient's arrhythmia improved. This report highlights the inflammation due to CS mimicking channelopathic features.     

一种基于支持向量机的自适应肿瘤分类检测算法

根据肿瘤分类检测模型的特点,提出了一种新的算法,该算法结合使用了基因选择和数据抽取的有效方法,并在此基础上使用支持向量机对基因表达数据进行分类或者检测。其中乳腺癌的分类交叉验证结果由88．46％提高到100．0％,急性白血病的也由71．05％提高至100．0％。实验结果说明了这一方法的有效性,为在大量的基因表达数据中提高检测癌症的准确性提出了一种比较通用的方法。     

基于序列特征的人类PolⅡ启动子理论预测

基于已知的人类PolII启动子序列数据,综合选取启动子序列内容和序列信号特征,构建启动子的支持向量机分类器．分别以启动子序列的6-mer频数作为离散源参数构建序列内容特征。同时选取24个位点的3-mer频数作为序列信号特征构建PWM,将所得到的两类参数输入支持向量机对人类启动子进行预测．用10折叠交叉检验和独立数据集来衡量算法的预测能力,相关系数指标达到95％以上,结果显示结合了支持向量机的离散增量算法能够有效的提高预测成功率,是进行真核生物启动子预测的一种很有效的方法．     

Adapting subject specific motor imagery EEG patterns in space–time–frequency for a brain computer interface

In this paper we propose a new technique that adaptively extracts subject specific motor imagery related EEG patterns in the space–time–frequency plane for single trial classification. The proposed approach requires no prior knowledge of reactive frequency bands, their temporal behavior or cortical locations. For a given electrode array, it finds all these parameters by constructing electrode adaptive time–frequency segmentations that are optimized for discrimination. This is accomplished first by segmenting the EEG along the time axis with Local Cosine Packets. Next the most discriminant frequency subbands are selected in each time segment with a frequency axis clustering algorithm to achieve time and frequency band adaptation individually. Finally the subject adapted features are sorted according to their discrimination power to reduce dimensionality and the top subset is used for final classification. We provide experimental results for 5 subjects of the BCI competition 2005 dataset IVa to show the superior performance of the proposed method. In particular, we demonstrate that by using a linear support vector machine as a classifier, the classification accuracy of the proposed algorithm varied between 90.5% and 99.7% and the average classification accuracy was 96%.