Clustering Improvement for Electrocardiographic Signals

  • Pau Micó
  • David Cuesta
  • Daniel Novák
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3617)


Holter signals are ambulatory long-term electrocardiographic (ECG) registers used to detect heart diseases which are difficult to find in normal ECG. These signals normally include several channels and its duration is up to 48 hours. The principal problem for the cardiologists consists of the manual inspection of the whole Holter ECG to find all those beats whose morphology differ from the normal cardiac rhythm. The later analysis of these abnormal beats yields a diagnostic from the pacient’s heart condition. In this paper we compare the performance among several clustering methods applied over the beats processed by Principal Component Analysis (PCA). Moreover, an outlier removing stage is added, and a cluster estimation method is included. Quality measurements, based on ECG labels from MIT-BIH database, are developed too. At the end, some results-accuracy values among several clustering algorithms is presented.


  1. 1.
    Koski, A., Juhola, M.: Segmentation of Digital Signals Based on Estimated Compression Ratio. IEEE trans. on Biomedical Engineering 43(9) (1996)Google Scholar
  2. 2.
    Micó, P., Cuesta, D., Novák, D.: Polygonal Approximation of Holter Registers: A Comparative Study for Electrocardiographic Signals Time Compression. In: Computational Intelligence in Medicine and Healthcare proc. (2005) (accepted)Google Scholar
  3. 3.
    Cuesta, D., Novák, D.: Automatic extraction of significant beats from a Holter register. In: BIOSIGNAL proceedings, pp. 3–5 (2002)Google Scholar
  4. 4.
    Cuesta, D., Novák, D., Eck, V., Pérez, C., Andreu, G.: Electrocardiogram Baseline Removal Using Wavelet Approximation. In: BIOSIGNAL proc., pp. 136–138 (2000)Google Scholar
  5. 5.
    Novák, D., Cuesta, D., Eck, V., Pérez, J.C., Andreu, G.: Denoising Electrocardiogram Signal Using Adaptive Wavelets. In: BIOSIGNAL proc., pp. 18–20 (2000)Google Scholar
  6. 6.
    Theodoridis, S., Koutroumbas, K.: Pattern Recognition. Academic Press, London (1999)Google Scholar
  7. 7.
    Rabiner, L.R., Juang, B.H.: An Introduction to Hidden Markov Models. IEEE ASSP Magazine (1986)Google Scholar
  8. 8.
    Micó, P., Cuesta, D., Novák, D.: Pre-clustering of Electrocardiographic Signals Using Ergodic Hidden Markov Models. In: Fred, A., Caelli, T.M., Duin, R.P.W., Campilho, A.C., de Ridder, D. (eds.) SSPR&SPR 2004. LNCS, vol. 3138, pp. 939–947. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  9. 9.
    Harris, R.J.: Multivariate analysis of variance. Statistics: Textbooks and monographs 137, 255–296 (1993)Google Scholar
  10. 10.
    Micó, P., Cuesta, D., Novák, D.: High-Speed Feature Extraction in Holter Electrocardiogram using Principal Component Analysis. In: BIOSIGNAL proc. (2004)Google Scholar
  11. 11.
    Mark, R., Moody, G.: MIT-BIH arrhythmia data base directory. Massachusetts Institute of Technology-Beth Israel Deaconess Medical Center (1998)Google Scholar
  12. 12.
    González, R.C., Tou, J.T.: Pattern Recognition Principles. Addison- Wesley Publishing Company, Reading (1974)zbMATHGoogle Scholar
  13. 13.
    Juan, A.: Optimización de Prestaciones en Técnicas de Aprendizaje No Supervisado y su Aplicación al Reconocimiento de Formas. PhD thesis, Universidad Politécnica de Valencia (1999)Google Scholar
  14. 14.
    Rangayyan, R.M.: Biomedical Signal Analysis. A Case-Study Approach. Wiley-IEEE Press (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Pau Micó
    • 1
  • David Cuesta
    • 1
  • Daniel Novák
    • 2
  1. 1.Department of Systems Informatics and ComputersPolytechnic School of AlcoiAlcoiSpain
  2. 2.Department of CyberneticsCzech Technical University in PragueCzech Republic

Personalised recommendations