Abstract
We explore the effect of using bagged decision tree (BDT) as an ensemble learning method with proposed time-domain feature extraction methods on electrocardiogram (ECG) arrhythmia beat classification comparing with single decision tree (DT) classifier. RR interval is the main property which defines irregular heart rhythm, and its ratio to the previous value and difference from mean value are used as morphological feature extraction methods. Form factor, its ratio to the previous value and difference from mean value are used to express ECG waveform complexity. In addition, skewness and second-order linear predictive coding coefficients are added to the feature vector of 56,569 ECG heart beats obtained from MIT–BIH arrhythmia database as time-domain feature extraction methods. The quarter of ECG heart beat samples are used as test data for DT and BDT. The performance measures of these classifiers are evaluated using the metrics such as accuracy, sensitivity, specificity and Kappa coefficient for both classifiers, and the performance of BDT classifier is examined for number of base learners up to 75. The BDT results in more predictive performance than DT according to the performance measures. BDT with 69 base learners has 99.51 % of accuracy, 97.50 % of sensitivity, 99.80 % of specificity and 0.989 of Kappa coefficient while DT gives 98.78, 96.05, 99.57 and 0.975 %, respectively. These metrics show that the suggested BDT increases the numbers of successfully identified arrhythmia beats. Moreover, BDT with at least three base learners has higher distinguishing capability than DT.
Similar content being viewed by others
References
Sache FB (2004) Computational cardiology: modeling of anatomy, electrophysiology, and mechanics. Springer, Germany
Homaeinezhad MR, Atyabi SA, Tavakkoli E, Toosi HN, Ghaffari A, Ebrahimpour R (2012) ECG arrhythmia recognition via a neuro-SVM-KNN hybrid classifier with virtual QRS image-based geometrical features. Expert Syst Appl 39:2047–2058
Zhang H, Zhang LQ (2005) ECG analysis based on PCA and support vector machines. ICNN&B 2:743–747
Sandoe E, Sigurd B (1991) Arrhythmia–a guide to clinical electrocardiology. Publishing Partners, Bingen
Kim J, Shin HS, Shin K, Lee M (2009) Robust algorithm for arrhythmia classification in ECG using extreme learning machine. Biomed Eng Online 8:31. doi:10.1186/1475-925X-8-31
Özçift A (2011) Random forests ensemble classifier trained with data resampling strategy to improve cardiac arrhythmia diagnosis. Comput Biol Med 41:265–271
Brause RW (2001) Medical analysis and diagnosis by neural networks, In Computer Science Department. Frankfurt a.m, Germany
Rangayyan RM (2001) Biomedical signal analysis: a case-study approach. Wiley-IEEE Press, USA
Yu SN, Chen YH (2009) Noise-tolerant electrocardiogram beat classification based on higher order statistics of sub-band components. Artif Intell Med 46:165–178
Tsipouras MG, Fotiadis DI, Sideris D (2005) An arrhythmia classification system based on the RR-interval signal. Artif Intell Med 33:237–250
Jekova I, Bortolan G, Christov I (2008) Assessment and comparison of different methods for heartbeat classification. Med Eng Phys 30:248–257
Asl BM, Setarehdan SK, Mohebbi M (2008) Support vector machine-based arrhythmia classification using reduced features of heart rate variability signal. Artif Intell Med 44:51–64
Chen SW (2007) Complexity-measure—based sequential hypothesis testing for real–time detection of lethal cardiac arrhythmias. Eurasip J Adv Sig Pr 1–8. doi:10.1155/2007/20957
Chiu CC, Lin TH, Liau BY (2005) Using correlation coefficient in ECG waveform for arrhythmia detection. Biomed Eng App Bas C 17:37–42
He T, Clifford G, Tarassenko L (2006) Application of independent component analysis in removing artefacts from the electrocardiogram. Neural Comput Appl 15:105–116
Engin M (2004) ECG beat classification using neuro-fuzzy network. Pattern Recogn Lett 25:1715–1722
Chawla MPS (2009) A comparative analysis of principal component and independent component techniques for electrocardiograms. Neural Comput Appl 18:539–556
Karimifard S, Ahmadian A (2011) A robust method for diagnosis of morphological arrhythmias based on Hermitian model of higher order statistics. Biomed Eng Online 10:1–18
Park KS, Cho BH, Lee DH, Song SH, Lee JS, Chee YJ, Kim IY, Kim SI (2008) Hierarchical support vector machine based heartbeat classification using higher order statistics and hermite basis function. Comput Cardiol 35:229–232
Al-Fahoum AS, Howitt I (1999) Combined wavelet transformation and radial basis neural networks for classifying life-threatening cardiac arrhythmias. Med Biol Eng Comput 37:566–573
Yu SN, Chen YH (2007) Electrocardiogram beat classification based on wavelet transformation and probabilistic neural network. Pattern Recognit Lett 28:1142–1150
Yu SN, Chen YH (2008) Selection of higher order sub-band features for ECG beat classification. In: 16th EUSIPCO
Mert A, Kilic N, Akan A (2011) Support vector machines with reduced dimensionality using independent component analysis for breast cancer classification. In: ELMAR 2011 proceedings, pp 37–40
Thanapatay D, Suwansaroj C, Thanawattano C (2010) ECG beat classification method for ECG printout with principle components analysis and support vector machines. ICEIE 1:72–75
Mousa R, Munib Q, Moussa A (2005) Breast cancer diagnosis system based on wavelet analysis and fuzz-neural. Expert Syst Appl 28:713–723
Xiao Q, Jian CW, Fei GD (2011) ECG signal classification based on BPNN. ICEICE 2:1362–1364
Cavalin PR, Sabourin R, Suen CY (2011) Dynamic selection approaches for multiple classifier systems. Neural Comput Appl. doi:10.1007/s00521-011-0737-9
Breiman L (1996) Bagging predictors. Mach Learn 24:123–140
Zhu X, Yang Y (2008) A lazy bagging approach classification. Pattern Recogn 41:2980–2992
Freund Y, Shapire RE (1997) A decision—theoretic generalization of on line learning and an application to boosting. J Comput Syst Sci 55:119–139
Das R, Sengur A (2010) Evaluation of ensemble methods for diagnosis of valvular heart disease. Expert Syst Appl 37:5110–5115
Hothorn T, Lausen B (2003) Bagging tree classifiers for laser scanning images: a data- and simulation- based strategy. Artif Intell Med 27:65–79
Moon H, Ahn H, Kodell RL, Baek S, Lin CJ, Chen JJ (2007) Ensemble methods for classification of patients for personalized medicine with high-dimensional data. Artif Intell Med 41:197–207
Ho TK (1998) The random subspace method for constructing decision forests. IEEE T Pattern Anal 20:832–844
MIT–BIH arrhythmia database. http://www.physionet.org/physiobank/database/html/mitdbdir/mitdbdir.htm. Accessed 26 May 2012
Osowski S, Linh TH (2001) ECG beat recognition using fuzzy hybrid neural network. IEEE T Bio Med Eng 48:1265–1271
PhysioBank ATM. http://physionet.org/cgi-bin/atm/ATM?database=mitdb&tool=plot_waveforms. Accessed 26 May 2012
Jerez-Aragones JM, Gomez-Ruiz JA, Ramaos-Jimenez G, Munoz-Perez J (2003) A combined neural network and decision tree model for prognosis of breast cancer relapse. Artif Intell Med 27:45–63
Zhang Y, Zhong S (2012) A privacy-preserving algorithm for distributed training of neural network ensembles. Neural Comput Appl. doi:10.1007/s00521-012-1000-8
Webb AR, Coppesy KD (2011) Statistical pattern recognition. Wiley, Malvern
Tumer K, Ghosh J (1996) Error correlation and error reduction in ensemble classifiers. Connect SCI 8:385–404
Freidman JH (1997) On bias, variance, 0/1-loss, and the curse-of-dimensionality. Data Min Knowl Disc 1:55–57
Cohen J (1968) Weighted Kappa: nominal scale agreement with provision for scaled disagreement or partial credit. Psychol Bull 70:213–220
Fleiss JL (1981) Statistical methods for rates and proportions. Wiley, New York
Berdinas BG, Betanzos AA (2002) Empirical evaluation of a hybrid intelligent monitoring system using different measures of effectiveness. Artif Intell Med 24:71–96
Karpagachelvi S, Arthanari M, Sivakumar M (2011) Classification of electrocardiogram signals with support vector machines and extreme learning machine. Neural Comput Appl. doi:10.1007/s00521-011-0572-z
Lagerholm M, Peterson C, Braccini G, Ebendrandt L, Sornmo L (2000) Clustering ECG complexes using hermite functions and self-organizing maps. IEEE T Bio Med Eng 47:838–848
Dokur Z, Olmez T (2001) ECG beat classification by a hybrid neural network. Comput Method Progr Bio 66:167–181
Acknowledgments
This work was partially supported by The Research Fund of The University of Istanbul. Project numbers: IRP-11824 and UDP-25231.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Mert, A., Kılıç, N. & Akan, A. Evaluation of bagging ensemble method with time-domain feature extraction for diagnosing of arrhythmia beats. Neural Comput & Applic 24, 317–326 (2014). https://doi.org/10.1007/s00521-012-1232-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-012-1232-7