基于ELM的室性早搏检测算法
|
摘要
计算机辅助室性早搏检测对室早的早诊断、早治疗十分关键,而基于SVM的室早检测方法存在训练速度慢、分类效果不稳定等问题.提出了一种基于极限学习机的计算机辅助室早检测算法,该算法首先对心电图像进行预处理,去除噪声后进行QRS波检测,然后建立室早特征模型并提取特征,最后基于极限学习机(extreme learning machine,ELM)进行室早检测.利用MIT_BIH的Arrhythmia心电数据库的心电信号对该算法进行了测试,结果表明与SVM相比ELM在分类速度及分类准确度上都有明显的优势.
Extreme learning machine(ELM)is an easy to use and effective learning algorithm of single-hidden layer feed forward neural networks(SLFNs).The classical learning algorithm in neural networks,e.g.back propagation,requires setting several user-defined parameters and may produce the local minimum.However,the extreme learning machine only requires setting the number of hidden neurons and the activation function.It does not need to adjust the input weights and hidden layer biases during the implementation of the algorithm,and it produces only one optimal solution. Therefore,ELM has the advantages of fast learning speed and good generalization performance.In this paper,ELM is introduced in predicting PVC,Nine features for the classification contains heart rate,morphology and wavelet energy of the ECG.And by comparing with SVM,its feasibility and advantages in PVC prediction are analysed.The experimental result s shows that ELM is much more accurate than SVM,and it has obvious advantages in parameter selection and learning speed.
Extreme learning machine(ELM)is an easy to use and effective learning algorithm of single-hidden layer feed forward neural networks(SLFNs).The classical learning algorithm in neural networks,e.g.back propagation,requires setting several user-defined parameters and may produce the local minimum.However,the extreme learning machine only requires setting the number of hidden neurons and the activation function.It does not need to adjust the input weights and hidden layer biases during the implementation of the algorithm,and it produces only one optimal solution. Therefore,ELM has the advantages of fast learning speed and good generalization performance.In this paper,ELM is introduced in predicting PVC,Nine features for the classification contains heart rate,morphology and wavelet energy of the ECG.And by comparing with SVM,its feasibility and advantages in PVC prediction are analysed.The experimental result s shows that ELM is much more accurate than SVM,and it has obvious advantages in parameter selection and learning speed.
引文
[1]Huang G,Zhou H,Ding X,et al.Extreme learning machine for regression and multiclass classification.IEEE Trans on Systems,Man,and Cybernetics,Part B:Cybernetics,2012,42(2):513-529
[2]Huang G,Qin Y,Siew C.Extreme learning machine:Theory and applications.Neurocomputing,2006,70(1):489-501
[3]王之琼,康雁,于戈,等.基于特征选择ELM的乳腺肿块检测算法.东北大学学报:自然科学版,2013,34(6):792-796
[4]沈利,杨杰,周越.基于支持向量机的室性早搏检测.生物医学工程杂志,2005,22(1):78-81
[5]宋莉,孟庆建,张光玉,等.基于波形特征和SVM的心电信号自动分类方法研究.中国医学物理学杂志,2010,27(4):2043-2046
[6]菜磊,程国建,潘华贤.极限学习机在岩性识别中的应用.计算机工程与设计,2010,31(9):2010-2012
[7]任杰,杨丽晓.基于小波变换系数的心电信号基线漂移噪声去除方法.医疗卫生装备,2010,31(11):24-26
[8]李延军,严洪,王增丽.心电基线漂移去除方法的比较研究.航天医学与医学工程,2009,22(5):381-386
[9]高玉宝,查代奉,江金龙.一种新的最优小波基选择准则及其应用.通信技术,2008,41(11):185-187
[10]魏宝琴,李白萍.最优小波基的选取原则.甘肃科研,2008,41(11):42-43
[11]韦高.基于小波变换的心电信号去噪法.中国医学装备,2010,8(7):9-11
[12]Donoho D L.De-noising by soft-threshold.IEEE Trans on Information Theory,1995,41(3):613-626
[13]潘磊,张军,邹采荣.基于小波变换的一种心电信号去噪算法.科技咨询导报,2007,5(12):582-586
[14]David L,Jain M.Idea apatial adaptation by wavelet shrinkage.Biometrika,1994,81(3):425-455
[15]Mitchell T M.机器学习.曾华军,张银奎,译.北京:机械工业出版社,2003
[16]徐晨,赵瑞珍,甘小冰.小波分析应用算法.北京:科学出版社,2004
[17]Gunawan D.Denoising images using wavelet transform//Proc of IEEE Pacific Rim Conf on Communications Computers and Signal Processing.Piscataway,NJ:IEEE,1999:83-85
[18]Chui C K,Chne G.Kalman Filtering with Real-Time Applications.Berlin:Springer,1991
[19]赵治栋,潘敏,郭希山,等.基于小波包收缩的心电信号除噪方法研究.计算机工程与应用,2002,19(22):19-24
[20]张小丽.基于离散小波变换门限消噪法阈值确定的研究.科技信息,2009:24(1):433
[21]陈光亮.心电图的计算机智能分析系统.北京:北方工业大学,2006
[22]王立传,陈裕泉,潘敏.QRS波检测技术的进展.航天医学与医学工程,2006,19(3):231-234
[23]王文,孙世双,周勇.基于小波变换的心电图QRS波群检测方法研究.北京生物医学工程,2002,4(21):241-247
[24]万相奎,秦树人,梁小容,等.小波变换在心电信号特征提取中的应用.北京生物医学工程,2005,24(6):410-413
[25]林薇,吴效明,朱维宗,等.应用小波变换模极大值检测ECG特征点.医疗卫生装备,2004,10(10):149-151
[26]李立策,郭兴明,王景灿,等.基于DSP的心电信号R波检测新算法.微计算机信息,2008,24(11):3831-3834
[2]Huang G,Qin Y,Siew C.Extreme learning machine:Theory and applications.Neurocomputing,2006,70(1):489-501
[3]王之琼,康雁,于戈,等.基于特征选择ELM的乳腺肿块检测算法.东北大学学报:自然科学版,2013,34(6):792-796
[4]沈利,杨杰,周越.基于支持向量机的室性早搏检测.生物医学工程杂志,2005,22(1):78-81
[5]宋莉,孟庆建,张光玉,等.基于波形特征和SVM的心电信号自动分类方法研究.中国医学物理学杂志,2010,27(4):2043-2046
[6]菜磊,程国建,潘华贤.极限学习机在岩性识别中的应用.计算机工程与设计,2010,31(9):2010-2012
[7]任杰,杨丽晓.基于小波变换系数的心电信号基线漂移噪声去除方法.医疗卫生装备,2010,31(11):24-26
[8]李延军,严洪,王增丽.心电基线漂移去除方法的比较研究.航天医学与医学工程,2009,22(5):381-386
[9]高玉宝,查代奉,江金龙.一种新的最优小波基选择准则及其应用.通信技术,2008,41(11):185-187
[10]魏宝琴,李白萍.最优小波基的选取原则.甘肃科研,2008,41(11):42-43
[11]韦高.基于小波变换的心电信号去噪法.中国医学装备,2010,8(7):9-11
[12]Donoho D L.De-noising by soft-threshold.IEEE Trans on Information Theory,1995,41(3):613-626
[13]潘磊,张军,邹采荣.基于小波变换的一种心电信号去噪算法.科技咨询导报,2007,5(12):582-586
[14]David L,Jain M.Idea apatial adaptation by wavelet shrinkage.Biometrika,1994,81(3):425-455
[15]Mitchell T M.机器学习.曾华军,张银奎,译.北京:机械工业出版社,2003
[16]徐晨,赵瑞珍,甘小冰.小波分析应用算法.北京:科学出版社,2004
[17]Gunawan D.Denoising images using wavelet transform//Proc of IEEE Pacific Rim Conf on Communications Computers and Signal Processing.Piscataway,NJ:IEEE,1999:83-85
[18]Chui C K,Chne G.Kalman Filtering with Real-Time Applications.Berlin:Springer,1991
[19]赵治栋,潘敏,郭希山,等.基于小波包收缩的心电信号除噪方法研究.计算机工程与应用,2002,19(22):19-24
[20]张小丽.基于离散小波变换门限消噪法阈值确定的研究.科技信息,2009:24(1):433
[21]陈光亮.心电图的计算机智能分析系统.北京:北方工业大学,2006
[22]王立传,陈裕泉,潘敏.QRS波检测技术的进展.航天医学与医学工程,2006,19(3):231-234
[23]王文,孙世双,周勇.基于小波变换的心电图QRS波群检测方法研究.北京生物医学工程,2002,4(21):241-247
[24]万相奎,秦树人,梁小容,等.小波变换在心电信号特征提取中的应用.北京生物医学工程,2005,24(6):410-413
[25]林薇,吴效明,朱维宗,等.应用小波变换模极大值检测ECG特征点.医疗卫生装备,2004,10(10):149-151
[26]李立策,郭兴明,王景灿,等.基于DSP的心电信号R波检测新算法.微计算机信息,2008,24(11):3831-3834