T-wave end detection using neural networks and Support Vector Machines

被引:15
作者
Alexeis Suarez-Leon, Alexander [1 ,2 ]
Varon, Carolina [2 ,3 ]
Willems, Rik [2 ,4 ]
Van Huffel, Sabine [2 ,3 ]
Roman Vazquez-Seisdedos, Carlos [1 ]
机构
[1] Univ Oriente, Fac Telecommun Informat & Biomed Engn, Santiago De Cuba, Cuba
[2] Katholieke Univ Leuven, Dept Elect Engn ESAT, STADIUS Ctr Dynam Syst Signal Proc & Data Analyt, Leuven, Belgium
[3] Imec, Leuven, Belgium
[4] UZ Leuven, Leuven, Belgium
关键词
ECG; T-wave end; Neural networks; FS-LSSVM; DENSITY-ESTIMATION; CLUSTER-ANALYSIS; ECG; SIGNALS; ROBUST; CLASSIFIERS; DELINEATION; TRANSFORM; ALGORITHM; LOCATION;
D O I
10.1016/j.compbiomed.2018.02.020
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
Background and objective: In this paper we propose a new approach for detecting the end of the T-wave in the electrocardiogram (ECG) using Neural Networks and Support Vector Machines. Methods: Both, Multilayer Perceptron (MLP) neural networks and Fixed-Size Least-Squares Support Vector Machines (FS-LSSVM) were used as regression algorithms to determine the end of the T-wave. Different strategies for selecting the training set such as random selection, k-means, robust clustering and maximum quadratic (Renyi) entropy were evaluated. Individual parameters were tuned for each method during training and the results are given for the evaluation set. A comparison between MLP and FS-LSSVM approaches was performed. Finally, a fair comparison of the FS-LSSVM method with other state-of-the-art algorithms for detecting the end of the T-wave was included. Results: The experimental results show that FS-LSSVM approaches are more suitable as regression algorithms than MLP neural networks. Despite the small training sets used, the FS-LSSVM methods outperformed the state-of-the-art techniques. Conclusion: FS-LSSVM can be successfully used as a T-wave end detection algorithm in ECG even with small training set sizes.
引用
收藏
页码:116 / 127
页数:12
相关论文
共 37 条
[11]   A robust wavelet-based multi-lead electrocardiogram delineation algorithm [J].
Ghaffari, A. ;
Homaeinezhad, M. R. ;
Akraminia, M. ;
Atarod, M. ;
Daevaeiha, M. .
MEDICAL ENGINEERING & PHYSICS, 2009, 31 (10) :1219-1227
[12]   Orthogonal series density estimation and the kernel eigenvalue problem [J].
Girolami, M .
NEURAL COMPUTATION, 2002, 14 (03) :669-688
[13]   PhysioBank, PhysioToolkit, and PhysioNet - Components of a new research resource for complex physiologic signals [J].
Goldberger, AL ;
Amaral, LAN ;
Glass, L ;
Hausdorff, JM ;
Ivanov, PC ;
Mark, RG ;
Mietus, JE ;
Moody, GB ;
Peng, CK ;
Stanley, HE .
CIRCULATION, 2000, 101 (23) :E215-E220
[14]   A database for evaluation of algorithms for measurement of QT and other waveform intervals in the ECG [J].
Laguna, P ;
Mark, RG ;
Goldberg, A ;
Moody, GB .
COMPUTERS IN CARDIOLOGY 1997, VOL 24, 1997, 24 :673-676
[15]   P- and T-Wave Delineation in ECG Signals Using a Bayesian Approach and a Partially Collapsed Gibbs Sampler [J].
Lin, Chao ;
Mailhes, Corinne ;
Tourneret, Jean-Yves .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2010, 57 (12) :2840-2849
[16]   New approach for T-wave peak detection and T-wave end location in 12-lead paced ECG signals based on a mathematical model [J].
Madeiro, Joao P. V. ;
Nicolson, William B. ;
Cortez, Paulo C. ;
Marques, Joao A. L. ;
Vazquez-Seisdedos, Carlos R. ;
Elangovan, Narmadha ;
Ng, G. Andre ;
Schlindwein, Fernando S. .
MEDICAL ENGINEERING & PHYSICS, 2013, 35 (08) :1105-1115
[17]   An algorithm for QRS onset and offset detection in single lead electrocardiogram records [J].
Manriquez, Alfredo Illanes ;
Zhang, Qinghua .
2007 ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, VOLS 1-16, 2007, :541-544
[18]   Application of the phasor transform for automatic delineation of single-lead ECG fiducial points [J].
Martinez, Arturo ;
Alcaraz, Raul ;
Joaquin Rieta, Jose .
PHYSIOLOGICAL MEASUREMENT, 2010, 31 (11) :1467-1485
[19]   A wavelet-based ECG delineator:: Evaluation on standard databases [J].
Martínez, JP ;
Almeida, R ;
Olmos, S ;
Rocha, AP ;
Laguna, P .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2004, 51 (04) :570-581
[20]   Ventricular repolarization measures for arrhythmic risk stratification [J].
Monitillo, Francesco ;
Leone, Marta ;
Rizzo, Caterina ;
Passantino, Andrea ;
Iacoviello, Massimo .
WORLD JOURNAL OF CARDIOLOGY, 2016, 8 (01) :57-73