Open Access Open Access  Restricted Access Subscription Access

ECG Beat Classification using the Integration of S-transform, PCA and Artificial Neural Network


Affiliations
1 Mallabhum Institute of Technology, Bishnupur - 722122, West Bengal, India
2 RCC Institute of Information Technology, Kolkata - 700015, West Bengal, India
 

Electrocardiography is an important tool in diagnosing the condition of the heart. In this paper, we propose a scheme to integrate the Stockwell Transform (ST), Principal Component Analysis (PCA) and Neural Networks (NN) for ECG beat classification. The ST is employed to extract the morphological features. In addition, PCA is among considerable techniques for data reduction. A Back Propagation Neural Network (BPNN) is employed as classifier. ECG samples attributing to six different beat types are sampled from the MIT-BIH arrhythmias database for experiments. In this paper comparative study of performance of six structures such as FCM-NN, PCA-NN, FCM-ICA-NN, FCM-PCA-NN, ST-NN and ST-PCA-NN are investigated. The test results suggest that ST-PCA-NN structure can perform better and faster than other techniques.

Keywords

Artificial Neural Network, ECG, Principal Component Analysis, S-Transform.
User
Notifications
Font Size

  • Acharya R, Bhat PS, Iyengar SS, Roo A, Dua S. Classification of heart rate data using artificial neural network and fuzzy equivalence relation. The Journal of the Pattern Recognition Society. 2002; 36(1):61–8.
  • De Chazal P, Reilly RB. Automatic classification of ECG beats using waveform shape and heart beat interval features. In IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP’03); Hong Kong, China. 2003. p. 269–72.
  • Osowski S, Linh TH. ECG beat recognition using fuzzy hybrid neural network. IEEE Transaction on Biomedical Engineering. 2001; 48(11):1265–71.
  • Guler I, Ubeyli ED. ECG beat classifier designed by combined neural network model. Pattern Recognition. 2005; 38:199–208.
  • Engin M. ECG beat classification using neuro-fuzzy network. Pattern Recognition Letters. 2004; 25:1715–22.
  • Owis MI, Youssef A-BM, Kadah YM. Characterization of ECG signals based on blind source separation. Medical and Biological Engineering and Computing. 2002; 40:557–64.
  • Yu SN, Chou K-T. Integration of independent component analysis and neural networks for ECG beat classification. Expert Systems with Applications. 2008; 34:2841–6.
  • Ceylan R, Ozbay Y. Comparision of FCM, PCA and WT techniques for classification ECG arrhythmias using artificial neural network. Expert Systems with Applications. 2007; 33:286–95.
  • Patra D, Das MK, Pradhan S. Integration of FCM, ICA and neural networks for ECG signal classification. IAENG International Journal of Computer Science. 2010; 36:3.
  • Stockwell R, Mansinha L, Lowe R. Localization of the complex spectrum: The s transform. IEEE Transactions on Signal Processing. 1996 Apr; 44(4): 998–1001.
  • Stockwell R. why use the s-transform in pseudo-differentials operatores: PDEs and time frequency analysis. Ser Fields Institute Communications. Wong, editors. AMS. 2007; 52:279–309.
  • Stockwell R. A basic efficient representation of the s-transform. Digital Signal processing. 2007; 17:371–93.
  • Schimmel M, Gallart J. The inverse S-transform in filters with time-frequency localization. IEEE Trans. Signal Processing. 2005; 53(11):4417–22.
  • Haykin S. Neural networks. New Delhi: Pearson Education Asia; 2002.
  • The MIT-BIH arrhythmias database. Available from: http://physionet.org/ physiobank/database/mitdb/
  • Pan J, Tompkins WJ. A real time QRS detection algorithm. IEEE Transaction on Biomedical Engineering. 1985; 32(3):230–6.

Abstract Views: 399

PDF Views: 209




  • ECG Beat Classification using the Integration of S-transform, PCA and Artificial Neural Network

Abstract Views: 399  |  PDF Views: 209

Authors

Manab Kumar Das
Mallabhum Institute of Technology, Bishnupur - 722122, West Bengal, India
Anup Kumar Kolya
RCC Institute of Information Technology, Kolkata - 700015, West Bengal, India

Abstract


Electrocardiography is an important tool in diagnosing the condition of the heart. In this paper, we propose a scheme to integrate the Stockwell Transform (ST), Principal Component Analysis (PCA) and Neural Networks (NN) for ECG beat classification. The ST is employed to extract the morphological features. In addition, PCA is among considerable techniques for data reduction. A Back Propagation Neural Network (BPNN) is employed as classifier. ECG samples attributing to six different beat types are sampled from the MIT-BIH arrhythmias database for experiments. In this paper comparative study of performance of six structures such as FCM-NN, PCA-NN, FCM-ICA-NN, FCM-PCA-NN, ST-NN and ST-PCA-NN are investigated. The test results suggest that ST-PCA-NN structure can perform better and faster than other techniques.

Keywords


Artificial Neural Network, ECG, Principal Component Analysis, S-Transform.

References