窦性和室性QRS波相似度计算方法的对比研究
|
摘要
目的研究几种相似性度量方法对心搏QRS波形态的区分效果,通过心搏分类为心电图,特别是24 h动态心电的诊断起到辅助作用。方法相似性度量方法包括欧氏距离、曼哈顿距离、相关系数等,对主要的几种方法从原理、算法和心搏分类的角度做了详细的分析和计算比较。结果挑选典型的窦性心搏和不同形态的室性心搏病例,用各方法对心搏进行相似性度量计算,对比结果表明,相关系数对QRS波形态差异的区分效果综合最佳。结论相似性度量方法中,相关系数最适合用于心搏形态的分类。
Objective To assess the variability of the QRS complexes based on similarity measure,and especially provide assistant diagnosis to ECG 24-hour Holter. Methods Similarity measures includes Euclidean distance, Manhattan distance,correlation technique and et al, these methods were analyzed and compared in principle, algorithm and heartbeat classification.Results Typical sinus beats and ventricular beats with different shapes were chosen to test the methods. The results indicated that the correlation technique was the best for assessing the variability of the QRS complexes. Conclusion It is demonstrated that correlation technique is the best method among all similarity measures for the classification of ECG beat.
Objective To assess the variability of the QRS complexes based on similarity measure,and especially provide assistant diagnosis to ECG 24-hour Holter. Methods Similarity measures includes Euclidean distance, Manhattan distance,correlation technique and et al, these methods were analyzed and compared in principle, algorithm and heartbeat classification.Results Typical sinus beats and ventricular beats with different shapes were chosen to test the methods. The results indicated that the correlation technique was the best for assessing the variability of the QRS complexes. Conclusion It is demonstrated that correlation technique is the best method among all similarity measures for the classification of ECG beat.
引文
[1]WANG Heng-di,ZHU Jian-min,LIANG Li,et al.A simple and accurate algorithm for identifying the R waves[J].Journal of Biomedical Engineering,2007,24(5):1019-1022.[王恒迪,朱坚民,梁丽,等.一种简单准确的R波检测算法[J].生物医学工程学杂志,2007,24(5):1019-1022.]
[2]WANG De-xu.R waves detection based on variable difference threshold[J].Scientific Research,2017,(2):00284-00285.[王德戌.基于差分阈值法的R波检测[J].科研,2017,(2):00284-00285.]
[3]WANG Qiang-wei,SUN Peng,PANG Yu,et al.Exploration of ECG signal R wave detection based on lifting wavelets[J].Life Science Instruments,2015,13(4-5):37-41.[王蔷薇,孙朋,庞宇,等.基于提升小波的心电信号R波检测算法研究[J].生命科学仪器,2015,13(4-5):37-41.]
[4]HUANG Wen-xia,LUO Hao,MA Zhan-qing.ECG beat classifier based on neural network and modified ant colony algorithm[J].Journal of Xinyang Normal University:Natural Science Edition,2008,21(4):580-583.[黄文霞,罗浩,马占卿.基于改进型群蚁神经网络的ECG心搏分类器[J].信阳师范学院学报:自然科学版,2008,21(4):580-583.]
[5]SHI Hang-rui,LIANG Ying.Research on ECG classification agorithm based on RBF neural network[J].Computer Knowledge and Technology,2017,13(19):137-139.[史航瑞,梁英.基于RBF神经网络的心电分类识别算法研究[J].电脑知识与技术,2017,13(19):137-139.]
[6]YANG Jie.The research of ECG signal detection and pattern classification methods[D].Hangzhou:Zhejiang Normal University,2014.[杨杰.心电信号的检测与模式分类方法的研究[D].杭州:浙江师范大学,2014.]
[7]Dutta S,Chatterjee A,Munshi S.Correlation technique and least square support vector machine combine for frequency domain based ECG beat classification[J].Med Eng Phys,2010,32(10):1161-1169.
[8]Güler I,譈beyh ED.A modified mixture of experts’network structure for ECG beats classification with diverse features[J].Engineering Applications of Artificial Intelligence,2005,18(7):845-856.
[9]Siddig TM,Mohmmed MA.A study of ECG signal classification using fuzzy logic control[J].International Journal of Science and Research,2014,3(2):374-380.
[10]WANG Jin-hai,SHI Meng-ying,ZHANG Xing-hua.Arrhythmia classification algorithm based on EMD and Ap En feature extraction[J].Chinese Journal of Scientific Instrument,2016,(S1):168-173.[王金海,史梦颖,张兴华.基于EMD和Ap En特征提取的心律失常分类研究[J].仪器仪表学报,2016,(s1):168-173.]
[11]LI Kun-yang,HU Guang-shu.Arrhythmia classification based on ECG analysis[J].Journal of Tsinghua University:Science and Technology,2009,49(3):416-418.[李坤阳,胡广书.基于心电图分析的心律失常分类[J].清华大学学报:自然科学版,2009,49(3):416-418.]
[12]YAN Hao-lin,AN Yong,WANG Hong-fei,et al.Feature extraction of ECG heartbeats based on convolutional neural networks[J].Computer Engineering and Design,2017,38(4):1024-1028.[颜昊霖,安勇,王宏飞,等.基于卷积神经网络的心电特征提取[J].计算机工程与设计,2017,38(4):1024-1028.]
[13]Lin KP,Chang WH.QRS feature extraction using linear prediction[J].IEEE Trans on BME,1989,36(10):1050-1057.
[14]Fook JC,Fang Q,Zhang T.A fast critical arrhythmic ECGwaveform identification method using cross-correlation and multiple template matching[J].Conf Proc IEEE Eng Med Biol,2010,2010:1922-1925.
[15]Krasteva V,Leber R,Jekova I,et al.Classification of supraventricular and ventricular beats by QRS template matching and decision tree[C].Computing in Cardiology Conference(Cin C),2014:349-352.
[16]HUANG Bi-ying,TANG Hua-ming,ZHONG Neng-zhi,et al.Automatic identification algorithm based on multi-template matching for ventricular tachycardia and supraventricular tachycardia[J].Chinese Journal of Medical Physics,2016,33(3):280-285.[黄碧莹,唐桦明,钟能枝,等.基于多模板匹配的室性心动过速与室上性心动过速自动识别算法[J].中国医学物理学杂志,2016,33(3):280-285.]
[17]CUI Yong-hua,LIANG Zheng-you.An arrhythmia detection algorithm based on feature extraction and multiple template matching[J].Journal of Guangxi Academy of Sciences,2012,28(1):7-10.[崔永华,梁正友.一种基于特征提取和多模板匹配的心律失常检测算法[J].广西科学院学报,2012,28(1):7-10.]
[18]JI Chun-lin.The research and application of feature matching algorithms for ECG data classification[D].Harbin:Harbin Institute of Technology,2014.[季春林.基于特征匹配的心电数据分类算法的研究与应用[D].哈尔滨:哈尔滨工业大学,2014.]
[19]YANG Hao,ZHANG Er-xi,JIANG Zhuo-yun.PCA face recognition study based on different distance measures[J].Journal of Shaanxi University of Technology:Natural Science Edition,2016,32(4):45-50.[杨浩,张二喜,蒋卓芸.基于距离测度的PCA人脸识别研究[J].陕西理工学院学报:自然科学版,2016,32(4):45-50.]
[20]ZHAO Yu-juan,LIU Qing-chao.Research on weighted intergration of multiple classifier based on hybrid multiple distance measurement[J].Computer Engineering,2012,38(21):171-174.[赵玉娟,刘擎超.基于混合多距离度量的多分类器加权集成研究[J].计算机工程,2012,38(21):171-174.]
[21]Suebsing A,Hiransakolwong N.Feature selection using euclidean distance and cosine similarity for intrusion detection model[C].First Asian Conference on Intelligent Information and Database Systems.IEEE Computer Society,2009:86-91.
[22]ZHU Wei-dong,DU Cheng-yong,WU Yong.A decision making method of TOPSIS based on the correlation matrix[J].Mathematics in Practice and Theory,2014,44(4):33-38.[朱卫东,杜承勇,吴勇.一种基于相关系数矩阵的TOPSIS决策方法[J].数学的实践与认识,2014,44(4):33-38.]
[2]WANG De-xu.R waves detection based on variable difference threshold[J].Scientific Research,2017,(2):00284-00285.[王德戌.基于差分阈值法的R波检测[J].科研,2017,(2):00284-00285.]
[3]WANG Qiang-wei,SUN Peng,PANG Yu,et al.Exploration of ECG signal R wave detection based on lifting wavelets[J].Life Science Instruments,2015,13(4-5):37-41.[王蔷薇,孙朋,庞宇,等.基于提升小波的心电信号R波检测算法研究[J].生命科学仪器,2015,13(4-5):37-41.]
[4]HUANG Wen-xia,LUO Hao,MA Zhan-qing.ECG beat classifier based on neural network and modified ant colony algorithm[J].Journal of Xinyang Normal University:Natural Science Edition,2008,21(4):580-583.[黄文霞,罗浩,马占卿.基于改进型群蚁神经网络的ECG心搏分类器[J].信阳师范学院学报:自然科学版,2008,21(4):580-583.]
[5]SHI Hang-rui,LIANG Ying.Research on ECG classification agorithm based on RBF neural network[J].Computer Knowledge and Technology,2017,13(19):137-139.[史航瑞,梁英.基于RBF神经网络的心电分类识别算法研究[J].电脑知识与技术,2017,13(19):137-139.]
[6]YANG Jie.The research of ECG signal detection and pattern classification methods[D].Hangzhou:Zhejiang Normal University,2014.[杨杰.心电信号的检测与模式分类方法的研究[D].杭州:浙江师范大学,2014.]
[7]Dutta S,Chatterjee A,Munshi S.Correlation technique and least square support vector machine combine for frequency domain based ECG beat classification[J].Med Eng Phys,2010,32(10):1161-1169.
[8]Güler I,譈beyh ED.A modified mixture of experts’network structure for ECG beats classification with diverse features[J].Engineering Applications of Artificial Intelligence,2005,18(7):845-856.
[9]Siddig TM,Mohmmed MA.A study of ECG signal classification using fuzzy logic control[J].International Journal of Science and Research,2014,3(2):374-380.
[10]WANG Jin-hai,SHI Meng-ying,ZHANG Xing-hua.Arrhythmia classification algorithm based on EMD and Ap En feature extraction[J].Chinese Journal of Scientific Instrument,2016,(S1):168-173.[王金海,史梦颖,张兴华.基于EMD和Ap En特征提取的心律失常分类研究[J].仪器仪表学报,2016,(s1):168-173.]
[11]LI Kun-yang,HU Guang-shu.Arrhythmia classification based on ECG analysis[J].Journal of Tsinghua University:Science and Technology,2009,49(3):416-418.[李坤阳,胡广书.基于心电图分析的心律失常分类[J].清华大学学报:自然科学版,2009,49(3):416-418.]
[12]YAN Hao-lin,AN Yong,WANG Hong-fei,et al.Feature extraction of ECG heartbeats based on convolutional neural networks[J].Computer Engineering and Design,2017,38(4):1024-1028.[颜昊霖,安勇,王宏飞,等.基于卷积神经网络的心电特征提取[J].计算机工程与设计,2017,38(4):1024-1028.]
[13]Lin KP,Chang WH.QRS feature extraction using linear prediction[J].IEEE Trans on BME,1989,36(10):1050-1057.
[14]Fook JC,Fang Q,Zhang T.A fast critical arrhythmic ECGwaveform identification method using cross-correlation and multiple template matching[J].Conf Proc IEEE Eng Med Biol,2010,2010:1922-1925.
[15]Krasteva V,Leber R,Jekova I,et al.Classification of supraventricular and ventricular beats by QRS template matching and decision tree[C].Computing in Cardiology Conference(Cin C),2014:349-352.
[16]HUANG Bi-ying,TANG Hua-ming,ZHONG Neng-zhi,et al.Automatic identification algorithm based on multi-template matching for ventricular tachycardia and supraventricular tachycardia[J].Chinese Journal of Medical Physics,2016,33(3):280-285.[黄碧莹,唐桦明,钟能枝,等.基于多模板匹配的室性心动过速与室上性心动过速自动识别算法[J].中国医学物理学杂志,2016,33(3):280-285.]
[17]CUI Yong-hua,LIANG Zheng-you.An arrhythmia detection algorithm based on feature extraction and multiple template matching[J].Journal of Guangxi Academy of Sciences,2012,28(1):7-10.[崔永华,梁正友.一种基于特征提取和多模板匹配的心律失常检测算法[J].广西科学院学报,2012,28(1):7-10.]
[18]JI Chun-lin.The research and application of feature matching algorithms for ECG data classification[D].Harbin:Harbin Institute of Technology,2014.[季春林.基于特征匹配的心电数据分类算法的研究与应用[D].哈尔滨:哈尔滨工业大学,2014.]
[19]YANG Hao,ZHANG Er-xi,JIANG Zhuo-yun.PCA face recognition study based on different distance measures[J].Journal of Shaanxi University of Technology:Natural Science Edition,2016,32(4):45-50.[杨浩,张二喜,蒋卓芸.基于距离测度的PCA人脸识别研究[J].陕西理工学院学报:自然科学版,2016,32(4):45-50.]
[20]ZHAO Yu-juan,LIU Qing-chao.Research on weighted intergration of multiple classifier based on hybrid multiple distance measurement[J].Computer Engineering,2012,38(21):171-174.[赵玉娟,刘擎超.基于混合多距离度量的多分类器加权集成研究[J].计算机工程,2012,38(21):171-174.]
[21]Suebsing A,Hiransakolwong N.Feature selection using euclidean distance and cosine similarity for intrusion detection model[C].First Asian Conference on Intelligent Information and Database Systems.IEEE Computer Society,2009:86-91.
[22]ZHU Wei-dong,DU Cheng-yong,WU Yong.A decision making method of TOPSIS based on the correlation matrix[J].Mathematics in Practice and Theory,2014,44(4):33-38.[朱卫东,杜承勇,吴勇.一种基于相关系数矩阵的TOPSIS决策方法[J].数学的实践与认识,2014,44(4):33-38.]