动态ECG分析中QRS波检测算法的研究
摘要
随着计算机技术、数字信号处理技术、人工智能等先进理论的发展，心电计算机自动分析技术的研究也不断向纵深发展。将心电自动分析技术用于动态心电图中，使用计算机对所记录的长达24小时的心电数据进行自动分析处理，可以大大减轻海量数据给医生带来的负担。
QRS波检测是动态心电分析中的关键。然而，动态心电检查中伴随的干扰、不同病人心电波形之间的差异，以及同一病人心电随时间的变异等因素，给QRS波检测带来了很大的困难。另一方面，要在较短的时间内对24小时的心电数据进行处理，速度和存储量的要求限制了算法的复杂性，因而，一个既能方便、快速地实现，又能满足准确度要求的QRS波检测算法，仍然是众多研究者不断追求的目标。
本文的研究工作，就是围绕适用于动态心电分析的QRS波检测方案而展开，主要包括以下几方面：
首先，对动态心电自动分析进行综述，并对动态心电分析中QRS波检测面临的问题进行了分析。
从生理学角度出发，探讨心电产生的原理和正常心电图各波的形态和意义，对临床常见的正常和异常心电QRS波形态进行研究，并对QRS波的特点进行了归纳和小结。
根据动态心电分析的特点，分析了适用于动态心电处理的QRS波检测算法应当满足的要求。
对历年来QRS波检测算法的研究成果进行分类回顾和小结，尤其是对基于滤波和阈值检测的方法进行了深入研究，通过对各类方法的特点和对于动态心电分析的适用性的讨论，从而得出结论：只有基于滤波和阐值检测的方法才能满足实时检测的要求。
提出了一种基于包络波形的QRS波检测方案，该方案也是基于滤波和阈值检测思想的，因而具有较快的处理速度；同时，经过MIT-BIH和AHA心电数据库部分数据的测试，该方案的准确度分别达到了99.11％和97.09％以上。实验证明，该方案在速度和准确度等两方面都达到了满意的效果。
With the development of computer technology, digital signal processing technology, as well as theory of artificial intelligence, the technology of computerized ECG analysis has been developed quite a lot since recent years. This technology has been adopted in ambulatory ECG, which did automatic processing and analysis to long ECG recordings by computer, and has relieved technicians from the heavy burden caused by vast volume of ECG data.
QRS detection is the key of computerized ambulatory ECG analysis. But, much factors, such as the noises accompanied with ECG recordings, the difference between patients, even the variations of the ECG signal in one patient, has brought much difficulty in QRS detection. On the other side, in order to process the large volume of ambulatory ECG data in a short duration, the speed and the capacity of memory has limited the complexity of the algorithm. Thus, an even more accurate and high-speed algorithm of QRS detection which can be realized easily is still what most researches seeks for.
The main research work is involved in the scheme of QRS detection for ambulatory ECG, which includes the following aspects:
Based on summarization on the technology of ambulatory ECG analysis, the problems in QRS detection for ambulatory ECG are analyzed.
The origin of electrocardiogram, as well as the shape and signification of normal ECG is discussed. Further more, the common shape of QRS wave in clinical is studied, and conclusion are drawn for its characteristic.
According to the specialty of ambulatory ECG processing, its requirement for QRS detection algorithm is analyzed in detail.
Various kinds of algorithm for QRS detection developed in decades has been reviewed, especially the kind based on the concept of filter-threshold detection. After discussion and investigation in trait and applicability of each method, conclusion can be drawn that only the methods based on the concept of filter-threshold detection can meet the needs for real time application.
A new scheme of QRS detection based on the idea of envelope is brought forward in this paper. The scheme is based upon the concept of filter-threshold detection, thus obtains high speed. The MIT-BIH arrhythmia database and AHA ventricular
arrhythmia database are used to evaluate the method. Result shows that the accuracy of the scheme reaches 99.11% and 97.09% respectively. According to the evaluation in both aspects of speed and precision, it reveals that the scheme can meet the requirement of ambulatory ECG analysis with satisfaction.
QRS波检测是动态心电分析中的关键。然而，动态心电检查中伴随的干扰、不同病人心电波形之间的差异，以及同一病人心电随时间的变异等因素，给QRS波检测带来了很大的困难。另一方面，要在较短的时间内对24小时的心电数据进行处理，速度和存储量的要求限制了算法的复杂性，因而，一个既能方便、快速地实现，又能满足准确度要求的QRS波检测算法，仍然是众多研究者不断追求的目标。
本文的研究工作，就是围绕适用于动态心电分析的QRS波检测方案而展开，主要包括以下几方面：
首先，对动态心电自动分析进行综述，并对动态心电分析中QRS波检测面临的问题进行了分析。
从生理学角度出发，探讨心电产生的原理和正常心电图各波的形态和意义，对临床常见的正常和异常心电QRS波形态进行研究，并对QRS波的特点进行了归纳和小结。
根据动态心电分析的特点，分析了适用于动态心电处理的QRS波检测算法应当满足的要求。
对历年来QRS波检测算法的研究成果进行分类回顾和小结，尤其是对基于滤波和阈值检测的方法进行了深入研究，通过对各类方法的特点和对于动态心电分析的适用性的讨论，从而得出结论：只有基于滤波和阐值检测的方法才能满足实时检测的要求。
提出了一种基于包络波形的QRS波检测方案，该方案也是基于滤波和阈值检测思想的，因而具有较快的处理速度；同时，经过MIT-BIH和AHA心电数据库部分数据的测试，该方案的准确度分别达到了99.11％和97.09％以上。实验证明，该方案在速度和准确度等两方面都达到了满意的效果。
With the development of computer technology, digital signal processing technology, as well as theory of artificial intelligence, the technology of computerized ECG analysis has been developed quite a lot since recent years. This technology has been adopted in ambulatory ECG, which did automatic processing and analysis to long ECG recordings by computer, and has relieved technicians from the heavy burden caused by vast volume of ECG data.
QRS detection is the key of computerized ambulatory ECG analysis. But, much factors, such as the noises accompanied with ECG recordings, the difference between patients, even the variations of the ECG signal in one patient, has brought much difficulty in QRS detection. On the other side, in order to process the large volume of ambulatory ECG data in a short duration, the speed and the capacity of memory has limited the complexity of the algorithm. Thus, an even more accurate and high-speed algorithm of QRS detection which can be realized easily is still what most researches seeks for.
The main research work is involved in the scheme of QRS detection for ambulatory ECG, which includes the following aspects:
Based on summarization on the technology of ambulatory ECG analysis, the problems in QRS detection for ambulatory ECG are analyzed.
The origin of electrocardiogram, as well as the shape and signification of normal ECG is discussed. Further more, the common shape of QRS wave in clinical is studied, and conclusion are drawn for its characteristic.
According to the specialty of ambulatory ECG processing, its requirement for QRS detection algorithm is analyzed in detail.
Various kinds of algorithm for QRS detection developed in decades has been reviewed, especially the kind based on the concept of filter-threshold detection. After discussion and investigation in trait and applicability of each method, conclusion can be drawn that only the methods based on the concept of filter-threshold detection can meet the needs for real time application.
A new scheme of QRS detection based on the idea of envelope is brought forward in this paper. The scheme is based upon the concept of filter-threshold detection, thus obtains high speed. The MIT-BIH arrhythmia database and AHA ventricular
arrhythmia database are used to evaluate the method. Result shows that the accuracy of the scheme reaches 99.11% and 97.09% respectively. According to the evaluation in both aspects of speed and precision, it reveals that the scheme can meet the requirement of ambulatory ECG analysis with satisfaction.
引文
[1]张开滋，郭继鸿，刘海祥，顾菊康编著，《临床心电信息学》，湖南科学技术出版社，2002。
[2]陈清启，杨庭树，卢喜烈，吴祥编著，《心电图学》，山东科学技术出版社，2002。
[3]李翠微，郑崇勋，“ECG自动分析技术的发展”，国外医学生物医学分册，1994：17(1)：15～22。
[4]肖咏梅，陈杭，葛霁光，“心电自动分析技术”，生物医学工程学杂志，2000：17(3)：339～342。
[5]吴杰，“心电图的计算机分析及其评价方法”，世界医疗器械杂志，1998：4(11)：40～52。
[6]徐有福，“心电图自动诊断”，国外医学生物医学分册，1985：8(2)：69～82。
[7]陈新主编，《临床心律失常学电生理和治疗》，人民卫生出版社，2000。
[8]屈建石，“24小时连续记录Holter的现状和一些技术问题”，北京生物医学工程，1993：12(4)：197～209。
[9]张镜如，乔健天编，《生理学》，人民卫生出版社，2000。
[10]Joseph T Catalano , 《Guide To ECG Analysis》, by J.B.Lippincott Company, 1993.
[11]O. Pahlm, L. Sornmo, "Software QRS Detection In Ambulatory Monitoring-A Review", Med. & Biol. Eng.& Comput, 1984: 22: 289～297.
[12]S. Jalaleddine, C. Hutchens, "Ambulatory ECG Wave Detection for Automated Analysis: A Review ", ISA Transactions, 1987: 26(4): 33～43。
[13]Bert-Uwe Kohler, Carsten Henning, Reinhold Orglmeister, "The Principles Of Software QRS Detection", IEEE Engineering In Medicine and Biology, 2002: Jan/Feb, 42～57.
[14]邱庆军，蒋景英，虞启琏，“QRS波检测方法的回顾与新进展”，中国医疗器械杂志，1999：23(6)：344～347。
[15]温凌峰，孟兆辉，张永红，白净，“QRS波群检测方法的新进展”，国外医学生物医学分册，2001：24(5)：193～197。
[16]N. V. Thakor, J. G. Webster, W. J. Tompkins, "Optimal QRS Detector", Med. & Biol. Eng.& Comput, 1983: 21: 343～350.
[17]W. A. H. Englese, C. Zeelenberg, "A Single Scan Algorithm for QRS-Detection and Feature Extraction", Computers In Cardiology, 1979: 37～42.
[18]F. E. M. Brekelmans, C. D. R. de Vaal, "A QRS Detection Scheme for Multichannel ECG Devices", Computers In Cardiology, 1981: 437～441.
[19]M. E. Nygards, L. Sornmo, "Delineation Of the QRS Complex Using the Envelope of
the e.c.g.", Med. &Biol. Eng.& Comput, 1983: 21: 538～547.
[20].A. Ligtenberg, M. Kunt, "A Robust-Digital QRS-Detection Algorithm for Arrhythmia Monitoring", Comput. Biomed. Res., 1983: 16: 273～286.
[21]M. L. Ahlstrom, W. J. Tompkins, "Automated High-Speed Analysis of Holter Tapes with Microcomputers", IEEE Trans. Biomed. Eng., 1983: 30(10): 651～657.
[22]J. Pan, W. J. Tompkins, "A Real-Time QRS Detection Algorithm", IEEE Trans. Biomed. Eng., 1985: 32(3): 230～236。
[23]F. Gritzali, "Towards A Generalized Scheme for QRS Detection in ECG Waveforms", Single Processing, 1988: 15(2): 183～192.
[24]S. Suppappola, Y. Sun, "A Comparison of Three QRS Detection Algorithm Using the AHA ECG Database", Annual International Conference of the IEEE EMBS, 1991: 13(2): 586～587.
[25]S. Suppappola, Y. Sun, "Nonlinear Transformation of ECG signal for Digital QRS Detection: A Quantitative Analysis", IEEE Trans. Biomed. Eng., 1994:41(4): 397～400.
[26]P. O. Boljesson, et.al, "Adaptive QRS Detection Based on Maximum: A Posteriori Estimation", IEEE Trans. Biomed. Eng., 1982: 29(5): 341～350.
[27]L. Sommo, O. Pahlnn, M. Nygards, "Adaptive QRS Detection: A Study of Performance", IEEE Trans. Biomed. Eng., 1985: 32(6): 392～401.
[28]P. E. Trahanias, "An approach to QRS complex detection using mathematical morphology," IEEE Trans. Biomed. Eng., 1993: 40(2): 201～205。
[29]S. A. Coast, R. M. Stem, et.al., "An Approach to Cardiac Arrhythmia Analysis Using Hidden Markov Models", IEEE Trans. Biomed. Eng., 1990: 37(9): 826～835.
[30]S. M. Collins, R. C. Arzbaecher, "An Efficient Algorithm for Waveform Analysis Using the Correlation Coefficient", Comput. Biomed. Res., 1981: 14: 381～389.
[31]A. Ruha, S. Sallinen, S. Nissila, "A Real-Time Microprocessor QRS Detector System with A 1-ms Timing Accuracy for the Measurement of Ambulatory HRV", IEEE Trans. Biomed. Eng., 1997: 44(3): 159～167.
[32]J. K. Udupa, I. S. Murthy, "Syntactic Approach to ECG Rhythm Analysis", IEEE Trans. Biomed. Eng., 1980: 27: 370～375.
[33]E. Skordalakis, "Syntactic ECG Processing: A Review", Pattern Recognition, 1986: 19(4): 305～313.
[34]G. Papakonstantinous, E. Skordalakis, F. Gritzali, "An Attribute Grammar for QRS Detection", Pattern Recognition, 1986: 19(4): 297～303.
[35]C. Li, C. Zheng, C. Tai, "Detection of ECG Characteristic Points Using Wavelet Transforms", IEEE Trans. Biomed. Eng., 1995: 42(1): 21～28.
[36]S. Kadambe, R. Murray," Wavelet Transform-Based QRS Complex Detector", IEEE Trans. Biomed. Eng., 1999: 46(7): 838～848.
[37]J. S. Sahambi, S. N. Tandom, R. K. P. Bhatt, "Using Wavelet Transform for ECG Characterization." IEEE Eng.Med. Biol.Mag., 1997:16: 77～83.
[38]V. X. Afonso, W. J. Tompkins, "ECG Beat Detection Using Filter Banks", IEEE Trans. Biomed. Eng., 1999: 46(2): 192～202.
[39]张永红译，“神经网络在医学信号处理方面的应用”，国外医学生物医学分册，1993：16(4)：202～216。
[40]Q. Xue, Y. H. Hu, W. J. Tompkins, "Neural-Network-Based Adaptive Matched Filtering for QRS Detection", IEEE Trans. Biomed. Eng., 1992: 39(4): 317～329.
[41]G. Vijaya, V. Kumar, H. K. Verma, "ANN-Based QRS-Complex Analysis of ECG", J.Med. Eng. Technol., 1998: 22(4): 160～167.
[42]洪玮，“ECG波形分类算法研究”，浙江大学硕士学位论文，2001。
[43]段新昱，林家瑞，司海芹，“ECG监视过程中的消噪问题”，国外医学生物医学工程分册，1992：15(3)：138。
[44]沈凤麟，陈和晏主编，《生物医学随机信号处理》，中国科学技术大学出版社，1999。
[45]石瑛，“实时心电信号检测和聚类分析”，浙江大学硕士学位论文，1998。
[46]N. V. Thakor, J. G. Webster, W. J. Tompkins, "Estimation of QRS Complex Power Spectra for Design of A QRS Filter", IEEE Trans. Biomed. Eng., 1984:31(11): 702～706.
[47]吴杰，“心电图数据库的开发和价值”，中华心律失常学杂志，2000：4(1)：76～77。
[48]朱泽煌，胡广书，郭恒，崔子经，“MIT-BIH心电数据库的开发及用作检测标准”，中国生物医学工程学报，1993：12(4)：244～249。
[49]J. L. Willems, P. Arnaud, J. H. van Bemmel, et.al., "Common Standards for Quantitative Electrocardiography:Goals and Main Results", Methods of Information in Medicine, 1990: 29: 263～271.
[50]P. S. Hamilton, W. J. Tompkins, "Quantitive Investigation of QRS Detection Rules Using the MIT/BIH Arrhythmia Database", IEEE Trans. Biomed. Eng., 1986: 33(12): 1157～1165.
[51]C. L. Chung, K. P. Lin, T. H. Tao,et.al., "Validation of Automated Arrhythmia Detection for Holter ECG", Proceedings - 20th annual international conference, IEEE/EMBS, 1998:101～103.
[2]陈清启，杨庭树，卢喜烈，吴祥编著，《心电图学》，山东科学技术出版社，2002。
[3]李翠微，郑崇勋，“ECG自动分析技术的发展”，国外医学生物医学分册，1994：17(1)：15～22。
[4]肖咏梅，陈杭，葛霁光，“心电自动分析技术”，生物医学工程学杂志，2000：17(3)：339～342。
[5]吴杰，“心电图的计算机分析及其评价方法”，世界医疗器械杂志，1998：4(11)：40～52。
[6]徐有福，“心电图自动诊断”，国外医学生物医学分册，1985：8(2)：69～82。
[7]陈新主编，《临床心律失常学电生理和治疗》，人民卫生出版社，2000。
[8]屈建石，“24小时连续记录Holter的现状和一些技术问题”，北京生物医学工程，1993：12(4)：197～209。
[9]张镜如，乔健天编，《生理学》，人民卫生出版社，2000。
[10]Joseph T Catalano , 《Guide To ECG Analysis》, by J.B.Lippincott Company, 1993.
[11]O. Pahlm, L. Sornmo, "Software QRS Detection In Ambulatory Monitoring-A Review", Med. & Biol. Eng.& Comput, 1984: 22: 289～297.
[12]S. Jalaleddine, C. Hutchens, "Ambulatory ECG Wave Detection for Automated Analysis: A Review ", ISA Transactions, 1987: 26(4): 33～43。
[13]Bert-Uwe Kohler, Carsten Henning, Reinhold Orglmeister, "The Principles Of Software QRS Detection", IEEE Engineering In Medicine and Biology, 2002: Jan/Feb, 42～57.
[14]邱庆军，蒋景英，虞启琏，“QRS波检测方法的回顾与新进展”，中国医疗器械杂志，1999：23(6)：344～347。
[15]温凌峰，孟兆辉，张永红，白净，“QRS波群检测方法的新进展”，国外医学生物医学分册，2001：24(5)：193～197。
[16]N. V. Thakor, J. G. Webster, W. J. Tompkins, "Optimal QRS Detector", Med. & Biol. Eng.& Comput, 1983: 21: 343～350.
[17]W. A. H. Englese, C. Zeelenberg, "A Single Scan Algorithm for QRS-Detection and Feature Extraction", Computers In Cardiology, 1979: 37～42.
[18]F. E. M. Brekelmans, C. D. R. de Vaal, "A QRS Detection Scheme for Multichannel ECG Devices", Computers In Cardiology, 1981: 437～441.
[19]M. E. Nygards, L. Sornmo, "Delineation Of the QRS Complex Using the Envelope of
the e.c.g.", Med. &Biol. Eng.& Comput, 1983: 21: 538～547.
[20].A. Ligtenberg, M. Kunt, "A Robust-Digital QRS-Detection Algorithm for Arrhythmia Monitoring", Comput. Biomed. Res., 1983: 16: 273～286.
[21]M. L. Ahlstrom, W. J. Tompkins, "Automated High-Speed Analysis of Holter Tapes with Microcomputers", IEEE Trans. Biomed. Eng., 1983: 30(10): 651～657.
[22]J. Pan, W. J. Tompkins, "A Real-Time QRS Detection Algorithm", IEEE Trans. Biomed. Eng., 1985: 32(3): 230～236。
[23]F. Gritzali, "Towards A Generalized Scheme for QRS Detection in ECG Waveforms", Single Processing, 1988: 15(2): 183～192.
[24]S. Suppappola, Y. Sun, "A Comparison of Three QRS Detection Algorithm Using the AHA ECG Database", Annual International Conference of the IEEE EMBS, 1991: 13(2): 586～587.
[25]S. Suppappola, Y. Sun, "Nonlinear Transformation of ECG signal for Digital QRS Detection: A Quantitative Analysis", IEEE Trans. Biomed. Eng., 1994:41(4): 397～400.
[26]P. O. Boljesson, et.al, "Adaptive QRS Detection Based on Maximum: A Posteriori Estimation", IEEE Trans. Biomed. Eng., 1982: 29(5): 341～350.
[27]L. Sommo, O. Pahlnn, M. Nygards, "Adaptive QRS Detection: A Study of Performance", IEEE Trans. Biomed. Eng., 1985: 32(6): 392～401.
[28]P. E. Trahanias, "An approach to QRS complex detection using mathematical morphology," IEEE Trans. Biomed. Eng., 1993: 40(2): 201～205。
[29]S. A. Coast, R. M. Stem, et.al., "An Approach to Cardiac Arrhythmia Analysis Using Hidden Markov Models", IEEE Trans. Biomed. Eng., 1990: 37(9): 826～835.
[30]S. M. Collins, R. C. Arzbaecher, "An Efficient Algorithm for Waveform Analysis Using the Correlation Coefficient", Comput. Biomed. Res., 1981: 14: 381～389.
[31]A. Ruha, S. Sallinen, S. Nissila, "A Real-Time Microprocessor QRS Detector System with A 1-ms Timing Accuracy for the Measurement of Ambulatory HRV", IEEE Trans. Biomed. Eng., 1997: 44(3): 159～167.
[32]J. K. Udupa, I. S. Murthy, "Syntactic Approach to ECG Rhythm Analysis", IEEE Trans. Biomed. Eng., 1980: 27: 370～375.
[33]E. Skordalakis, "Syntactic ECG Processing: A Review", Pattern Recognition, 1986: 19(4): 305～313.
[34]G. Papakonstantinous, E. Skordalakis, F. Gritzali, "An Attribute Grammar for QRS Detection", Pattern Recognition, 1986: 19(4): 297～303.
[35]C. Li, C. Zheng, C. Tai, "Detection of ECG Characteristic Points Using Wavelet Transforms", IEEE Trans. Biomed. Eng., 1995: 42(1): 21～28.
[36]S. Kadambe, R. Murray," Wavelet Transform-Based QRS Complex Detector", IEEE Trans. Biomed. Eng., 1999: 46(7): 838～848.
[37]J. S. Sahambi, S. N. Tandom, R. K. P. Bhatt, "Using Wavelet Transform for ECG Characterization." IEEE Eng.Med. Biol.Mag., 1997:16: 77～83.
[38]V. X. Afonso, W. J. Tompkins, "ECG Beat Detection Using Filter Banks", IEEE Trans. Biomed. Eng., 1999: 46(2): 192～202.
[39]张永红译，“神经网络在医学信号处理方面的应用”，国外医学生物医学分册，1993：16(4)：202～216。
[40]Q. Xue, Y. H. Hu, W. J. Tompkins, "Neural-Network-Based Adaptive Matched Filtering for QRS Detection", IEEE Trans. Biomed. Eng., 1992: 39(4): 317～329.
[41]G. Vijaya, V. Kumar, H. K. Verma, "ANN-Based QRS-Complex Analysis of ECG", J.Med. Eng. Technol., 1998: 22(4): 160～167.
[42]洪玮，“ECG波形分类算法研究”，浙江大学硕士学位论文，2001。
[43]段新昱，林家瑞，司海芹，“ECG监视过程中的消噪问题”，国外医学生物医学工程分册，1992：15(3)：138。
[44]沈凤麟，陈和晏主编，《生物医学随机信号处理》，中国科学技术大学出版社，1999。
[45]石瑛，“实时心电信号检测和聚类分析”，浙江大学硕士学位论文，1998。
[46]N. V. Thakor, J. G. Webster, W. J. Tompkins, "Estimation of QRS Complex Power Spectra for Design of A QRS Filter", IEEE Trans. Biomed. Eng., 1984:31(11): 702～706.
[47]吴杰，“心电图数据库的开发和价值”，中华心律失常学杂志，2000：4(1)：76～77。
[48]朱泽煌，胡广书，郭恒，崔子经，“MIT-BIH心电数据库的开发及用作检测标准”，中国生物医学工程学报，1993：12(4)：244～249。
[49]J. L. Willems, P. Arnaud, J. H. van Bemmel, et.al., "Common Standards for Quantitative Electrocardiography:Goals and Main Results", Methods of Information in Medicine, 1990: 29: 263～271.
[50]P. S. Hamilton, W. J. Tompkins, "Quantitive Investigation of QRS Detection Rules Using the MIT/BIH Arrhythmia Database", IEEE Trans. Biomed. Eng., 1986: 33(12): 1157～1165.
[51]C. L. Chung, K. P. Lin, T. H. Tao,et.al., "Validation of Automated Arrhythmia Detection for Holter ECG", Proceedings - 20th annual international conference, IEEE/EMBS, 1998:101～103.