摘要
心电图ST段是心电图诊断中一项重要指标,ST段具体形态的识别对心血管疾病诊断意义重大。针对心电图ST段形态的多样性,传统时域方法对具体形态识别显得不足,心电信号是微弱信号,易受到低频和工频信号的干扰,有效地滤除基线漂移和工频干扰为ST段准确识别提供保证。通过零相位巴特沃斯有效滤除基线漂移和工频干扰,利用神经网络与时域分析相结合的方法实现ST段多种形态的快速识别,减少神经网络输出层的形态分类,能够准确识别出ST段形态,实验结果满意,为心电图ST段诊断提供了依据。
Measurement of ECG ST-Segment plays an important role in routine ECG diagnosis. The specific form is of great significance to the diagnosis of cardiovascular disease. Based on the diversity of ST segment, the traditional time domain method is insufficient, ECG signals are weak and easy to get interference of low frequency and power frequency signal. In this paper, filtering out baseline drift and frequency interference of signals through zero phase Butterworth filter is the premise for the ST pattern recognition. Combining the neural network method with time domain analysis method was used for rapid identification of the ST segment morphology. This method can reduce output sample number. The accuracy of ST-segment shape recognition is satisfactory and necessary to diagnosis of ST segment.
出处
《计算机仿真》
CSCD
北大核心
2014年第11期349-352,共4页
Computer Simulation
关键词
神经网络
零相位巴特沃斯
心电图
Neural network
Zero phase Butterworth
Electrocardiogram