摘要
快速准确是心电(ECG)智能识别的发展趋势,然而现有的很多方法都很难在识别速度和精度上同时达到满意的效果。为此本文提出一种基于移动窗口运算的QRS复合波快速识别方法。该方法选取80ms宽的移动窗口,在窗口内只做简单的差运算和积运算,计算简单、运算量小,能很好地抑制P波、T波及其它噪声的影响。该方法在不进行数字滤波去噪的情况下对MIT-BIH心电数据库进行试验,识别QRS复合波的准确率高达99.6%,实现了快速、准确识别QRS复合波的目标,能满足ECG实时监测设备智能识别的要求。
A fast and accurate intelligent identification is the developing trend of electrocardiogram (ECG) research. However, there are few methods by which satisfactory results could be obtained both in speed and in accuracy. A fast identification method of QRS wave was proposed based on moving window operation in this study. An 80 ms wide moving window was employed, in which simple difference and product operations were carried out with simple and less computation, and a very good inhibition of P and T waves and other noises was realized. Then the method was investigated with data from the MIT-BIH arrhythmia database with absence of digital filtering de noising, and the identification accuracy of QRS complex reached 99.6%. The results showed that a rapid and accurate identifica- tion of QRS complex could be realized, which would meet the requirements for studying the real-time ECG monito- ring equipment.
出处
《生物医学工程学杂志》
EI
CAS
CSCD
北大核心
2013年第5期988-992,共5页
Journal of Biomedical Engineering
关键词
移动窗口
QRS波
实时智能识别
Moving window
QRS wave
Real-time intelligent identification