基于差分绝对值R波检测算法与研究--聂希芸.docVIP

摘 要 心血管疾病是严重威胁人类健康的疾病之一。心电图的应用提高了心血管疾病的基础研究、临床诊断和治疗的水平，推动了现代医学的发展，为人类的健康做出了重大贡献。心电信号的准确检测与分析对心血管疾病的治疗起着关键作用，成为了当前信号处理领域的研究热点。本文以心电信号的预处理和波形检测为主要研究内容。 心电信号是一种微弱的生物电信号，且是一种复杂的非平稳随机信号，极易受各种噪声的干扰。本文重点研究了基于小波变换的去除心电信号基线漂移噪声的算法。根据心电信号和基线漂移噪声的频谱特点，本文基于小波变换的多分辨率分析，对心电信号进行多尺度分解，利用分解后得到的模拟近似信号充分逼近基线漂移噪声的特性，从而去除心电信号中的基线漂移这一低频分量。通过对MIT-BIH心律失常数据库中的心电信号进行去噪处理，验证了本文所采用算法的有效性。 心电信号的检测分析中，首要的关键问题是R波的检测，在此基础上才能进一步对心电信号进行检测和分析。传统的基于差分实现R波检测的算法，是以对心电信号的斜率进行定量分析为基础的，都需要使用阈值进行判别，而阈值的选取恰当与否将直接影响算法的检测效果。本文提出的基于差分绝对值的R波检测方法，是一种新的差分检测算法，该方法不是通过阈值来确定R波的位置，具有算法简单，计算量小的特点。通过对MIT-BIH心律失常数据库中的心电信号进行实际检测，证明了该算法具有较高的识别率，达到了较好的检测效果。本文提出的基于差分绝对值的R波检测方法，为心电信号的自动分析提供了一种新的、简单的检测手段。 关键词：心电信号；基线漂移；多分辨率分析；R波检测；差分；  Abstract Cardiovascular disease is one of diseases that seriously threaten the health of people. The application of electrocardiogram (ECG) improves the level of the basic research of cardiovascular disease, clinical diagnosis and treatment, promotes the development of modern medicine, and has made a contribution for human health. The accurate detection and analysis of electrocardiosignal plays a key role in the treatment of cardiovascular disease, has become a research hotspot in current signal processing. In this thesis, pretreatment and waveform detection of electrocardiosignal are as the main research contents. Electrocardiosignal is a weak biological signal, and is a complex non-stationary random signal which is vulnerable to the interference of all kinds of noise. This thesis was focused on the algorithm of removal of electrocardiosignal baseline drifts noise based on wavelet transform. According to the frequency spectrum characteristics of electrocardiosignal and baseline drifts noise, this thesis was based on multi-resolution analysis of wavelet transform, decomposed the electrocardiosignal to multi-scale, utilized the analog approximate signal obtained after decomposition to fully close to the characteristics of the baseline drift noise to remove the electrocardiosignal baseline dri