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

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 (Rényi) 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.

[1]  Jean-Yves Tourneret,et al.  P- and T-Wave Delineation in ECG Signals Using a Bayesian Approach and a Partially Collapsed Gibbs Sampler , 2010, IEEE Transactions on Biomedical Engineering.

[2]  M. Gallegos,et al.  A robust method for cluster analysis , 2005, math/0504513.

[3]  Carlos R. Vázquez-Seisdedos,et al.  A Comparison of Different Classifiers Architectures for Electrocardiogram Artefacts Recognition , 2013, CIARP.

[4]  Yüksel Özbay,et al.  Comparison of FCM, PCA and WT techniques for classification ECG arrhythmias using artificial neural network , 2007, Expert Syst. Appl..

[5]  Johan A. K. Suykens,et al.  Optimized fixed-size kernel models for large data sets , 2010, Comput. Stat. Data Anal..

[6]  C. Matr'an,et al.  A general trimming approach to robust Cluster Analysis , 2008, 0806.2976.

[7]  J. A. Cuesta-Albertos,et al.  Trimmed $k$-means: an attempt to robustify quantizers , 1997 .

[8]  Jeffrey M Schussler,et al.  Drug-induced QT interval prolongation: mechanisms and clinical management , 2012, Therapeutic advances in drug safety.

[9]  Pablo Laguna,et al.  A database for evaluation of algorithms for measurement of QT and other waveform intervals in the ECG , 1997, Computers in Cardiology 1997.

[10]  M R Homaeinezhad,et al.  A robust wavelet-based multi-lead Electrocardiogram delineation algorithm. , 2009, Medical engineering & physics.

[11]  Mark Girolami,et al.  Orthogonal Series Density Estimation and the Kernel Eigenvalue Problem , 2002, Neural Computation.

[12]  黄亚明 PhysioBank , 2009 .

[13]  M. C. Jones,et al.  A reliable data-based bandwidth selection method for kernel density estimation , 1991 .

[14]  Senén Barro,et al.  A new approach for TU complex characterization , 2000, IEEE Transactions on Biomedical Engineering.

[15]  Aldebaro Klautau,et al.  New approach for T-wave end detection on electrocardiogram: Performance in noisy conditions , 2011, Biomedical engineering online.

[16]  Jeffrey M. Hausdorff,et al.  Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol , 2000 .

[17]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[18]  N. Ahmed,et al.  Discrete Cosine Transform , 1996 .

[19]  Matthias W. Seeger,et al.  Using the Nyström Method to Speed Up Kernel Machines , 2000, NIPS.

[20]  Johan A. K. Suykens,et al.  Bayesian Framework for Least-Squares Support Vector Machine Classifiers, Gaussian Processes, and Kernel Fisher Discriminant Analysis , 2002, Neural Computation.

[21]  María Teresa Gallegos,et al.  Maximum Likelihood Clustering with Outliers , 2002 .

[22]  Wojciech Zareba,et al.  Drug induced QT prolongation. , 2007, Cardiology journal.

[23]  A. Passantino,et al.  Ventricular repolarization measures for arrhythmic risk stratification. , 2016, World journal of cardiology.

[24]  P. Yip,et al.  Discrete Cosine Transform: Algorithms, Advantages, Applications , 1990 .

[25]  Qinghua Zhang,et al.  An Algorithm for Robust and Efficient Location of T-Wave Ends in Electrocardiograms , 2006, IEEE Transactions on Biomedical Engineering.

[26]  A. Davis,et al.  Drug-Induced QTc Prolongation. , 2017, The American journal of cardiology.

[27]  Deniz Erdogmus,et al.  Generalized information potential criterion for adaptive system training , 2002, IEEE Trans. Neural Networks.

[28]  C. D. Kemp,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[29]  W. Bystricky,et al.  Modelling T-end in Holter ECGs by 2-layer perceptrons , 2002, Computers in cardiology.

[30]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machines , 2002 .

[31]  Qinghua Zhang,et al.  An algorithm for QRS onset and offset detection in single lead electrocardiogram records , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[32]  Farzad Towhidkhah,et al.  Automated ECG Segmentation Using Piecewise Derivative Dynamic Time Warping , 2007 .

[33]  P. C. Cortez,et al.  New approach for T-wave peak detection and T-wave end location in 12-lead paced ECG signals based on a mathematical model. , 2013, Medical engineering & physics.

[34]  Jørgen K. Kanters,et al.  QT dynamics during treatment with sertindole , 2015, Therapeutic advances in psychopharmacology.

[35]  Raúl Alcaraz,et al.  Application of the phasor transform for automatic delineation of single-lead ECG fiducial points , 2010, Physiological measurement.

[36]  Pablo Laguna,et al.  A wavelet-based ECG delineator: evaluation on standard databases , 2004, IEEE Transactions on Biomedical Engineering.