摘要
在心电信号心律失常自动识别系统中,针对心电信号形态复杂导致特征提取困难、自动分类模型准确度低、现实应用性差的问题,设计了一种基于U-NET全卷积神经网络的心电信号语义分割的识别分类方法;该方法通过全卷积神经网络的编码运算规则,将心电信号切片数据作为输入,标签地图作为输出,可划分出信号片段中的心拍位置与类别;仿真结果表明:该方法在正常窦性搏动、左束支传导阻滞、右束支传导阻滞、房性早搏和室性早搏五分类问题中取得较高准确率,实现了对心律失常信号的有效识别。
In the arrhythmia automatic recognition system,it is difficult to extract signal features because of the complex ECGmorphology.The automatic classification model has low accuracy and adaptability.Aiming at the above problems,a recognition and classification method of ECG signal semantic segmentation based on u-net neural network is designed.With the operation rules of the full convolution neural network,the location and category of beats in the signal segments can be classified by taking ECG signal fragment data as input and label map as output.Simulation results show that the proposed method has achieved high accuracy in five classification problems of normal sinus beat,left bundle branch block,right bundle branch block,atrial premature beats and ventricular premature beats,and achieved effective recognition of arrhythmia signals.
作者
刘波
嵇启春
Liu Bo;Ji Qichun(School of Information and Control Engineering,Xi'an University of Architecture and Technology,Xi'an 710055,China)
出处
《计算机测量与控制》
2020年第6期175-179,共5页
Computer Measurement &Control
