摘要
T波终点识别是心电信号特征提取中的重要组成部分,但是由于T波部分容易受到噪声的干扰,再加上T波形态的变化使得T波终点识别相对比较困难。为了能准确快速地识别T波终点,该文以最小二乘估计为基础,首先快速定位T波顶点,再对T波后面的等电位线进行修正,得到符合实际情况的等电位线,最后根据等电位线与T波最大陡线的交点求得T波终点的参考点,并对参考点位置进行调整得到T波终点。由于最小二乘估计运算量小,算法的运算速度得到提高且具有一定的抗噪性,利用算法对QT数据库进行试验,T波检出率与T波终点识别精度都取得了很好的效果。
T-wave endpoint identification is an important part of ECG signal feature extraction.However,because the T-wave part is easily disturbed by noise,coupled with changes in the shape of the T-wave,it is relatively difficult to identify the T-wave endpoint.In order to accurately and quickly identify T-wave endpoint,this paper is based on least-squares estimation.First,the apex of the T wave is quickly located,and then the equipotential line behind the T wave is corrected to obtain the equipotential line that conforms to the actual situation.Finally,the reference point for T-wave endpoint is calculated according to the intersection of the equipotential line and the maximum steepness line of the T wave,and the position of the reference point is adjusted to obtain T-wave endpoint.Due to the small calculation amount of the least squares estimation,the calculation speed of the algorithm is improved and has a certain degree of noise immunity.The algorithm is used to test the QT database,and good results are achieved in both the T-wave detection rate and the T-wave endpoint identification accuracy.
出处
《科技创新与应用》
2025年第36期22-25,共4页
Technology Innovation and Application
关键词
最小二乘估计
等电位线
最大陡线
心电信号
抗嗓性
least squares estimation
equipotential line
maximum steepness line
electrocardiogram signal
noise resistance
