节点文献
基于心音信号遥测的心肌变力性与变时性的研究
Study of Myocardial Inotropism and Chronotropism Based on Telemetering Heart Sound Signal
【作者】 陈历明;
【作者基本信息】 重庆大学 , 生物医学工程, 2004, 博士
【摘要】 心音是人体最重要的生理信号之一,心音听诊的历史已有180多年。但是由于缺乏能够客观量化的指标,使得心音信号检测在诊断和评估心功能方面仍然只是一种辅助手段。国际上关于心音与心肌收缩能力关系的动物实验以及有创和无创临床对照研究表明,第一心音的幅值是心肌收缩能力的标准量度,这提示了可以利用心音信号来研究心肌的变力性;而心音信号与心电信号之间存在严格同步关系的生理机制又提示可以利用心音信号来研究心肌的变时性。本论文以心音信号作为研究对象,开发了便携式心音信号遥测系统,并利用该系统分别研究了受试者在完成规定负荷的心音图运动试验后以及在静息状态下5分钟内的心肌变力性与变时性,以便为运动医学、临床医学等领域更为全面、客观地诊断和评估心功能提供可以量化的指标,从而进一步拓展心音信号检测技术的应用范围以及应用价值。本论文主要包括以下有特色的研究工作和结论:(1)为了适应在运动场地、医院病房等移动场所通过测试心音信号来研究心肌变力性与变时性的需求,研制了便携式心音信号遥测系统。该系统由遥测发射机、遥测接收机、笔记本计算机等三部分组成,工作在433MHz免申请的开放频段,能同时测试10名受试者的心音信号;遥测发射机由受试者随身携带,每次发射一个心动周期的心音信号,并以心电信号的R波同步脉冲作为数据包的帧头标记,这样作为系统主机的笔记本计算机只需采用简单算法就能快速而准确地从接收到的遥测心音信号中提取研究心肌变力性和变时性所需的心力数据和心率数据,而无需遥测心电信号;遥测接收机通过专用的USB接口转换模块,实现了与笔记本计算机的USB口之间的高速数据传输,同时又无需额外的工作电源。性能测试表明,该便携式心音信号遥测系统达到了设计要求,并已被应用到在移动场地开展心肌变力性与变时性的研究中。(2)设计了一个由一维形态滤波器和基于Stein似然无偏估计的小波滤波器构建的组合式数字滤波器,来分别滤除小幅度的射频干扰脉冲和遥测系统的随机噪声以及心杂音。测试结果表明,该组合式数字滤波器能显著滤除遥测心音信号中的主要噪声,同时又能保证心音信号的主要成分基本不受损失。(3)设计了基于心电信号R波辅助定位的心音信号特征提取算法和基于包络提取与识别的独立心音信号的特征提取算法。对于用R波定位的心音信号特征提取算法而言,提出了能够比经典的心电信号定位法更为快速、准确地识别和确定S1和S2幅值的算法;对于独立心音信号的特征提取算法而言,利用心音信号与语音信号在产生机制和形态上的相似性,借鉴了语音信号研究领域在音节分割与识<WP=6>别上比较成熟的算法,提出了基于规格化平均Shannon能量算法提取的能量包络来识别和确定心音信号特征参数的算法,并获得了较高的识别率。(4)利用自行开发的便携式心音信号遥测系统,采集了50名体育系学生和30名普通系学生在完成了规定负荷的心音图运动试验后5分钟内的心力恢复趋势和心率恢复趋势。通过对数据进行三次样条插值、均匀重采样、小波滤波、差分运算等处理后提取了能够反映大负荷运动后心肌变力性与变时性动态特性的特征指数。测试结果表明,一方面体育系学生高水平的心脏功能不仅体现在具有更大的心力储备上(p<0.001), 而且还体现在具有更快的心力恢复速度上(p<0.001),而另一方面其心率储备与普通系学生并无显著差异,这与大负荷运动下心脏储备的生理机制是基本吻合的。测试结果也说明了基于心音信号遥测的心音图运动试验与其它常用的运动心功能测试方法相比,具有简单、无创、快捷、能够提供可以客观量化的心力指标和心率指标的优势,适用于在运动医学、临床医学等领域对心功能进行日常测试和评估。(5)根据心肌收缩能力与心率等心血管生理参数都受到自主神经系统调控的生理机制,开展了心力变异性的研究工作。针对目前国内外在心力变异性研究方面缺乏统一的分析方法和评估指标的现状,提出以心率变异性的研究方法作为参照,从20名健康大学生的心音信号中提取了5分钟的心率变异性信号和心力变异性信号,来同时开展静息状态下心肌变时性与变力性的静态特性研究。从时域分析和基于Marple算法的AR模型功率密度谱估计的结果看,心力变异性在形态上与心率变异性有着较大的相似性和较为一致的变化规律。因此可以得出初步结论,静息状态下的心肌变力性也能反映自主神经系统对心脏活动的调节功能,这就为下一步开展基于心音信号测试的心力变异性在临床诊断与自主神经系统受损相关的心血管疾病上的应用研究奠定了基础。
【Abstract】 Heart sound is one of the most important physiological signal in human body, the history of cardiac auscultation is more than 180 years. However, the heart sound detection is still only an accessorial tool to diagnose or evaluate cardiac function, because it lacks objective and quantitative indices. Internationally, animal experiments as well as invasive and noninvasive clinical trials for studying the relationship between heart sound and cardiac contractility showed that the amplitude of first heart sound is the standard measurement of cardiac contractility, which implys that heart sound signal can be used to study myocardial inotropism; The strict synchrony between ECG signal and heart sound signal implys that heart sound signal can be used to study myocardial chronotropism. This dissertation takes heart sound as study subject. A portable heart sound signal telemetry system is developed to study the subjects’ five-minute myocardial inotropism and chronotropism following phonocardiogram exercise test and under quiescent status respectively, in order to provide quantitative indices for diagnosing or evaluating cardiac function in the fields such as clinical medicine and sports medicine, aiming to further expand the application fields and values of the heart sound signal detection technique. It mainly includes the following distinctive research work and conclusions:A portable heart sound signal telemetry system is developed to satisfy the demand of studying myocardial inotropism and chronotropism in mobile places such as sports site and hospital wards. It consists of telemetry transmitter, telemetry receiver, notebook computer. This telemetry system operates at the licence-free open frequency band of 433MHz and can telemeter 10 subjects’ heart sound signals simultaneously ; The telemetry transmitter is carried by subject, and transmits a cardiac cycle-long heart sound signal each time with ECG’s R wave synchronous pulse as the head mark of the telemetry data packet, So the notebook computer only needs a simple algorithm to extract cardiac contractility signal and heart rate signal quickly and exactly from the received telemetry data without telemetering ECG signal; By using specific-application USB interface module, the telemetry receiver realizes high speed data transmission with the notebook computer’s USB port without extra power supply. Performance test shows that this telemetry system has reached the design requirement, and can be applied to study myocardial inotropism and chronotropism in mobile sites. <WP=8>A combined digital filter is constructed by an one-dimension morphological filter and a wavelet filter based on Stein Unbiased Risk Estimation rule to filter out small amplitude radio frequency interference pulses and system’s random noise as well as cardiac murmur respectively. Test shows that it can filter out the main noise in the telemetry heart sound signal significantly, while the main components of the heart sound signal can be kept without obvious loss.Two heart sound signal’s feature extraction algorithms are designed, one uses ECG’s R wave to help localization, another is based on independent heart sound signal. Compared with the classical R wave localizaton algorithm, this dissertation’s algorithm has the advantage of recognizing and determining the amplitude of S1 and S2 more quickly and exactly. For the feature extraction algorithm of independent heart sound signal, according to the similarity of waveform and generation mechanism between heart sound and speech, Some mature algorithms of segmenting and extracting speech signal’s feature have been used as reference, and an algorithm based on using average Shannon energy to exact heart sound signal’s energy envelope is put forward to recognize and determine the feature parameters of heart sound signal, which has better recognition rate.(4) By applying the developed portable heart sound signal telemetry system, five- minute continuous cardiac contractility and heart rate signals were sampled respectively from 50 students in
【Key words】 Heart Sound Signal; Myocardial Inotropism; Myocardial Chronotropism; Telemetry; Feature Extraction;