心音检测系统中综合评估方法的设计及实现

发布时间：2019-06-05 03:32

【摘要】：心音信号是人体重要的生理信号,它蕴含着心脏各个部分状态的大量生理病理信息,所以对心音信号的采集和研究具有重要的意义。本文的心音检测系统通过采集心音信号,自动分析其心率和心音综合指数,并且采用语音播报分析结果,为用户自助检测心音提供一种新的平台。本文首先从心音的频域能量谱入手,分析了心音的频域能量谱分布,给出了心音能量分布指标计算方法,然后再结合心音的心率、心率齐整性提出了心音综合指数(CITF),CITF是一种全面反应心音在时域和频域综合特性的量化指标。同时设计出了CITF的计算方法,为心音综合评估提供了一种新方法。为了克服心音实际采集后会出现噪声变为伪峰值或第一心音峰值不明显等异常情况,本文通过改进选择排序算法、设计自适应阈值及预设峰值跨度,获得一种出错率较低的心率计算方法——常阈值法(CDT)。本文设计、实现了一种心音检测系统。该心音检测系统包括采集模块、语音播报模块、主控模块,并且将CITF方法、心率CDT方法应用于该系统中。完成了系统的心音采集及预处理、心音综合指数计算、心率计算、语音播报等功能,同时还对该系统进行了小型化封装,便于携带使用。最后本文通过样本跟踪实验和多种情况下的重复实验,测试了心音检测系统的准确性和稳定性。通过实验结果分析提出了系统升级方案,在原方案的基础上,(1)设计动态阈值获得一种改进的心率计算方法;(2)计算出心音的时间序列符号熵,将符号熵应用于心音综合指数中。通过对比实验表明,系统改进方案对提升分析心音的准确性和稳定性具有明显的效果。本文提出的心音综合评估方法以及心音检测系统方案具有可移植性,它们可应用于不同性能的硬件平台,具有一定的推广和实用价值。
[Abstract]:Heart sound signal is an important physiological signal of human body, which contains a large number of physiological and pathological information about the state of each part of the heart, so it is of great significance to collect and study the heart sound signal. The heart sound detection system in this paper automatically analyzes the heart rate and heart sound comprehensive index by collecting the heart sound signal, and provides a new platform for the user to detect the heart sound by using the voice broadcast analysis results. In this paper, starting with the frequency domain energy spectrum of heart sound, the frequency domain energy spectrum distribution of heart sound is analyzed, and the calculation method of heart sound energy distribution index is given, and then the heart sound synthesis index (CITF), is put forward according to the heart rate and heart rate uniformity of heart sound. CITF is a quantitative index of the comprehensive characteristics of heart sound in time domain and frequency domain. At the same time, the calculation method of CITF is designed, which provides a new method for the comprehensive evaluation of heart sound. In order to overcome the abnormal situation that the noise becomes pseudo peak value or the first heart sound peak value is not obvious after the heart sound is actually collected, this paper designs the adaptive threshold and the preset peak span by improving the selection sorting algorithm. A method for calculating heart rate with low error rate, constant threshold method (CDT).), is obtained. In this paper, a heart sound detection system is designed and implemented. The heart sound detection system includes acquisition module, voice broadcast module and main control module, and the CITF method and heart rate CDT method are applied to the system. The functions of heart sound acquisition and preprocessing, heart sound comprehensive index calculation, heart rate calculation, voice broadcast and so on are completed. At the same time, the system is miniaturized and packaged, which is easy to carry and use. Finally, the accuracy and stability of the heart sound detection system are tested by sample tracking experiment and repeated experiment in many cases. Through the analysis of the experimental results, the system upgrade scheme is put forward. On the basis of the original scheme, (1) an improved heart rate calculation method is obtained by designing the dynamic threshold; (2) the symbolic entropy of heart sound is calculated, and the symbolic entropy is applied to the comprehensive index of heart sound. The comparative experiments show that the improved scheme has obvious effect on improving the accuracy and stability of heart sound analysis. The comprehensive evaluation method of heart sound and the scheme of heart sound detection system proposed in this paper have portability, and they can be applied to hardware platforms with different performance, and have certain popularization and practical value.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：R540.4;TN912.3

【相似文献】