Advertisement

Accurate Classification of ECG Patterns with Subject-Dependent Feature Vector

  • Piotr AugustyniakEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 403)

Abstract

Correct and accurate classification of ECG patterns in a long-term record requires optimal selection of feature vector. We propose a machine learning algorithm that learns from short randomly selected signal strips and, having an approval from a human operator, classifies all remaining patterns. We applied a genetic algorithm with aggressive mutation to select few most distinctive features of ECG signal. When applied to the MIT-BIH Arrhythmia Database records, the algorithm reduced the initial feature space of 57 elements to 3–5 features optimized for a particular subject. We also observe a significant reduction of misclassified beats percentage (from 2.7 % to 0.7 % in average for SVM classifier and three features) with regard to automatic correlation-based selection.

Keywords

Feature Vector Mother Population Syntactic Model Signal Strip Heartbeat Classification 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Notes

Acknowledgments

This scientific work is supported by the AGH University of Science and Technology in year 2015 as a research project No. 11.11.120.612.

References

  1. 1.
    Augustyniak, P.: The use of shape factors for heart beats classification in holterrecordings. Proc. Comput. Med. Zakop. 2–6(05), 47–52 (1997)Google Scholar
  2. 2.
    Augustyniak, P.: Adaptive discrete ECG representation—comparing variable depth decimation and continuous non-uniform sampling. Comput. Cardiol. 29, 165–168 (2005)Google Scholar
  3. 3.
    Augustyniak, P.: Wearable wireless heart rate monitor for continuous long-term variability studies. J. Electrocardiol. 44(2), 195–200 (2011)CrossRefGoogle Scholar
  4. 4.
    Chang, K.C., Lee, R.G., Wen, C., Yeh, M.F.: Comparison of similarity measures for clustering electrocardiogram complexes. Comput. Cardiol. 32, 759–762 (2005)Google Scholar
  5. 5.
    de Chazal, P., O’Dwyer, M., Reilly, R.B.: Automatic classification of heartbeats using ecg morphology and heartbeat interval features. IEEE Trans. Biomed. Eng. 51(7), 1196–1206 (2004)CrossRefGoogle Scholar
  6. 6.
    Holland, J.H., Reitman, J.S.: Cognitive systems based on adaptive algorithms. ACM SIGART Bull. 63, 43–49 (1977)Google Scholar
  7. 7.
    Jaworek, J., Augustyniak, P.: A cardiac telerehabilitation application for mobile devices. Comput. Cardiol. 38, 241–244 (2011)Google Scholar
  8. 8.
    Jokić, S., Krčo, S., Delić, V., Sakač, D., Lukić, Z., Loncar-Turukalo, T.: An efficient approach for heartbeat classification. Comput. Cardiol. 2010(37), 991–994 (2010)Google Scholar
  9. 9.
    Kittler, J.: Feature set search algorithms. In: Pattern Recognition and Signal Processing, pp. 41–60. Sijthoff and Noordhoff, Alphen aan den Rijn (1978)Google Scholar
  10. 10.
    Kononenko, I.: Estimating attributes: analysis and extensions of relief. In: De Raedt, L., Bergadano, F. (eds.) Machine Learning: ECML-94, pp. 171–182. Springer, Berlin (1994)Google Scholar
  11. 11.
    Lemay, M., Jacquemet, V., Forclaz, A., Vesin, J.M., Kappenberger, L.: Spatiotemporal QRST cancellation method using separate QRS and T-Waves templates. Comput. Cardiol. 32, 611–614 (2005)Google Scholar
  12. 12.
    Llamedo-Soria, M., Martinez, J.P.: An ECG classification model based on multilead wavelet transform features. Comput. Cardiol. 34, 105–108 (2007)Google Scholar
  13. 13.
    Llamedo-Soria, M., Martinez, J.P.: Analysis of multidoma in features for ECG classification. Comput. Cardiol. 36, 561–564 (2009)Google Scholar
  14. 14.
    Llamedo, M., Khwaja, A., Martinez, J.P.: Analysis of 12-lead classification models for ECG classification. Comput. Cardiol. 37, 673–676 (2010)Google Scholar
  15. 15.
    Llamedo, M., Martinez, J.: Heartbeat classification using feature selection driven by database generalization criteria. IEEE Trans. Biomed. Eng. 58(3), 616–625 (2011)Google Scholar
  16. 16.
    MacQueen, J.: Some methods for classification and analysis of multivariate observations. In: Le Cam, L.M., Neyman, J. (eds.) Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, vol. 1, pp. 281–297. University of California Press, Berkeley (1967)Google Scholar
  17. 17.
    Mallat, S., Zhong, S.: Characterization of signals from multiscale edges. IEEE Trans. Pattern Anal. Mach. Intell. 14(7), 710–732 (1992)CrossRefGoogle Scholar
  18. 18.
    Mensing, S., Bystricky, W., Safer, A.: Identifying and measuring representative QT intervals in predominantly non-normal ECGs. Comput. Cardiol. 33, 361–364 (2006)Google Scholar
  19. 19.
    Moody, G.B.: The MIT-BIH Arrhythmia Database CD-ROM, 3rd Edn. Harvard-MIT Division of Health Sciences and Technology, Cambridge (1997)Google Scholar
  20. 20.
    O’Dwyer, M., de Chazal, P., Reilly, R.I.: Beat classification for use in arrhythmia analysis. Comput. Cardiol. 27, 395–398 (2000)Google Scholar
  21. 21.
    Rejer, I.: Genetic algorithms in EEG feature selection for the classification of movements of the left and right hand. In: Proceedings CORES2013, pp. 579–589 (2013). doi: 10.1007/978-3-319-00969-8-57
  22. 22.
    Rejer, I.: Genetic algorithm with aggressive mutation for feature selection in BCI feature space pattern. Anal. Appl. (2014). doi: 10.1007/s10044-014-0425-3
  23. 23.
    Rodriguez-Sotelo, J.L., Cuesta-Frau, D., Castellanos-Dominguez, G.: An improved method for unsupervised analysis of ECG beats based on WT features and J-Means clustering. Comput. Cardiol. 34, 581–584 (2007)Google Scholar
  24. 24.
    Tibshirani, R.: Regression shrinkage and selection via thelasso. J. Stat. Soc. Ser. B 58(1), 267–288 (1996)MathSciNetzbMATHGoogle Scholar
  25. 25.
    Vansteenkiste, E., Houben, R., Pizurica, A., Philips, W.: Classifying electrocardiogram peaks using new wavelet domain features. Comput. Cardiol. 35, 853–856 (2008)Google Scholar
  26. 26.
    Vapnik, V.N.: The Nature Of Statistical Learning Theory. Springer, New York (1995)CrossRefzbMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.AGH University of Science and TechnologyKrakowPoland

Personalised recommendations