脉搏波信号采集质量实时分析研究及设备开发

发布时间：2018-04-26 04:17

本文选题：脉搏波 + 波形特征　；参考：《北京工业大学》2014年硕士论文

【摘要】：在脉搏波检测设备的研究与应用过程中，干扰和运动伪影导致的脉搏波信号质量降低一直是研究者面临的一个难题。由于脉搏波信号易受到外界或本身的干扰，导致信号质量下降，不能提供准确的心血管功能信息，，从而不能有效地得到有临床价值的信息。同时，检测人员也必须经过培训以获得一定的脉搏波检测知识和经验才能准确采集脉搏波信号。这些都成为脉搏波血流动力学检测技术在家庭保健应用和远程医疗应用的困难因素。为了在采集过程中评估脉搏波信号的采集质量，获得准确的心血管功能信息，提高采集自动化程度，本课题从脉搏波波形特征出发，提出脉搏波波形标准化、归一化方法。使用脉搏波波形特征参数实现对脉搏波波形特征的量化描述。通过讨论脉搏波波形特征参数具有的生理意义，并结合大量临床脉搏波实例，对脉搏波波形特征进行分类。以此为基础，研究脉搏波采集质量实时评价模型及算法。针对脉搏波监测过程中，实时分析监控脉搏波采集质量的实际工程需求，研究设计实时脉搏波单波判别技术，包括脉搏波数据处理队列中的单波判别算法、经验心率初判算法、基于经验心率的实时单波判别算法。提出脉搏波质量评价参数及算法，包括脉搏主频能量比、干扰频率能量比、基线漂移累计量、周期变异量、单波数据点特征变异量、单波主波峰宽度变异量。通过多种分析手段与方法建立脉搏波采集质量实时评价模型，并对模型做出了逻辑性验证。在设备开发方面，研究设计基于ARM CORTEX A8的脉搏波信号采集质量实时分析装置硬件系统。该系统使用以三星S5PV210处理器为CPU的核心板电路及外围拓展电路组合的设计方案。核心板部分采用三星S5PV210作为主处理器，采用了2.0mm间距的双排针引出CPU大部分常用功能引脚。核心板的硬件资源还包括512M DDR2内存和1GB闪存(SLC)。设计的外围电路部分主要包括AD采集电路、声卡电路、LCD电路、USB电路、网络接口电路、电源电路等。开发基于Android平台的脉搏波质量评价实时分析软件，其总体结构分为六层设计，分别是传感层、表示层、业务逻辑层、数据访问层、服务层、网络通信层。其主要功能涵盖AD采集、声卡采集、实时质量评价、SQLite数据库功能、网络通信功能等。最后设计了双盲诱发干扰检测实验，用于评价模型及设备的有效性。通过对实验数据重复性及变异性指标的分析，讨论在诱发干扰条件下脉搏波采集质量实时分析模型的有效性。
[Abstract]:In the research and application of pulse wave detection equipment, it is a difficult problem for researchers to reduce the quality of pulse wave signal caused by interference and motion artifact. Because the pulse wave signal is easily interfered by the outside world or itself, the signal quality is decreased, and the accurate cardiovascular function information can not be provided, so the clinical value information can not be obtained effectively. At the same time, the examiner must be trained to acquire certain knowledge and experience of pulse wave detection in order to collect pulse wave signal accurately. These are the difficult factors for the application of pulse wave hemodynamics in home health care and telemedicine applications. In order to evaluate the quality of pulse wave signal acquisition, obtain accurate cardiovascular function information and improve the automatic degree of acquisition, this paper presents a method of pulse wave standardization and normalization based on the characteristics of pulse wave waveform. The characteristic parameters of pulse wave are used to describe the characteristic of pulse wave. By discussing the physiological significance of pulse wave characteristic parameters and combining with a large number of clinical pulse wave examples, the pulse wave characteristics are classified. Based on this, the real-time evaluation model and algorithm of pulse wave acquisition quality are studied. In the process of pulse wave monitoring, the real engineering requirement of monitoring pulse wave acquisition quality is analyzed in real time. The single wave discrimination technology of real time pulse wave is studied and designed, including the single wave discrimination algorithm in pulse wave data processing queue and the experience heart rate initial judgement algorithm. Real-time single-wave discrimination algorithm based on experiential heart rate. The parameters and algorithm of pulse wave quality evaluation are presented, including pulse main frequency energy ratio, interference frequency energy ratio, baseline drift accumulation, periodic variation, characteristic variation of single wave data point and variation of peak width of single wave. A real-time evaluation model of pulse wave acquisition quality is established by various analytical methods and the logical verification of the model is made. In the aspect of equipment development, the hardware system of pulse signal acquisition and quality real-time analysis device based on ARM CORTEX A8 is studied and designed. This system uses the design scheme of the core board circuit with Samsung S5PV210 processor as CPU and peripheral expansion circuit. The core board uses Samsung S5PV210 as the main processor and the double row needle with 2.0mm spacing to lead to most of the commonly used functional pins of CPU. The hardware resources of the core board also include 512m DDR2 memory and 1GB flash memory. The peripheral circuit includes AD acquisition circuit, sound card circuit and LCD circuit, USB circuit, network interface circuit, power supply circuit and so on. A real-time analysis software for pulse wave quality evaluation based on Android platform is developed. Its overall structure is divided into six layers: sensing layer, presentation layer, business logic layer, data access layer, service layer and network communication layer. Its main functions include AD acquisition, sound card acquisition, real-time quality evaluation and SQLite database function, network communication function and so on. Finally, a double-blind interference detection experiment is designed to evaluate the effectiveness of the model and equipment. By analyzing the repeatability and variability of experimental data, the validity of the real-time analysis model of pulse wave acquisition quality under induced interference is discussed.
【学位授予单位】：北京工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.23;R318.6

【参考文献】