摘要
心电图数据是分析人体心脏病理的基础,针对心电数据的QRS波形特征提取问题,提出一种基于改进小波变换的特征提取方法。首先,选用小波函数Coiflet对采集信号2层分解,然后进行去噪处理。最大限度地保留心电信号有用特征成分,采用极大值与斜率双重检测法进行QRS波中R峰的定位,进而准确提取QRS特征。最后利用MIT-BIH数据库验证算法的准确性和有效性。实验结果表明,该算法针对QRS波形特征识别精度达到了99.661%,具有更高的有效性。
ECG data is the basis of human heart pathological analysis.In allusion to the QRS waveform feature extraction of ECG data,a feature extraction method based on improved wavelet transform is proposed.The wavelet function Coiflet is selected to perform 2⁃layer decomposition of the ECG signal,and then signal denoising is performed.The useful characteristic components of ECG signal is retained to the greatest extent.The maximum value and slope double detection method is used to locate the R peak in the QRS wave,so as to accurately extract the QRS feature.The MIT⁃BIH database is used to verify the accuracy and effectiveness of the algorithm.The experiment results show that the proposed algorithm has a high accuracy for the QRS waveform feature recognition,and its accuracy reaches to 99.661%.Therefore,it is of higher effectiveness.
作者
侯晓晴
仝泽友
刘晓文
HOU Xiaoqing;TONG Zeyou;LIU Xiaowen(School of Information and Control Engineering,China University of Mining and Technology,Xuzhou 221000,China;Internet of Things(Perception Mine)Research Center,China University of Mining and Technology,Xuzhou 221000,China)
出处
《现代电子技术》
北大核心
2020年第13期57-61,共5页
Modern Electronics Technique
基金
2017年国家重点研发计划项目:矿山安全生产物联网关键技术与装备研发(2017YFC0804401)。
关键词
QRS波识别
特征提取
心电信号
小波变换
信号去噪
R峰定位
QRS wave recognition
feature extraction
ECG signal
wavelet transform
signal denoising
R peak positioning
