Detection of abnormality in the electrocardiogram without prior knowledge by using the quantisation error of a self-organising map, tested on the European ischaemia database

  • E. A. Fernández
  • P. Willshaw
  • C. A. Perazzo
  • R. J. Presedo
  • S. Barro


Most systems for the automatic detection of abnormalities in the ECG require prior knowledge of normal and abnormal ECG morphology from pre-existing databases. An automated system for abnormality detection has been developed based on learning normal ECG morphology directly from the patient. The quantisation error from a self-organising map ‘learns’ the form of the patient's ECG and detects any change in its morphology. The system does not require prior knowledge of normal and abnormal morphologies. It was tested on 76 records from the European Society of Cardiology database and detected 90.5% of those first abnormalities declared by the database to be ischaemic. The system also responded to abnormalities arising from ECG axis changes and slow baseline drifts and revealed that ischaemic episodes are often followed by long-term changes in ECG morphology.


ECG analysis Morphology change detection Ischaemia detection Self-organizing map Neural networks 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Andrew, J., Hopkirk, J., Leader, S., Uhl, G. S., Hickman, J. R., andFischer, J. (1984): ‘Limitation of exercise-induced R wave amplitude changes in detecting coronary artery disease in asymptomatic men’,J. Amer. College Cardiol.,3, pp. 821–826Google Scholar
  2. Bezerianos, A., Vladutu, L., andPapadimitriou, S. (2000): ‘Hierarchical state space partitioning with a network self-organizing map for recognition of ST-T segment changes’,Med. Biol. Eng. Comput.,38, pp. 406–415Google Scholar
  3. Bonoris, P. E., Greenberg, P. S., Christison, G. W., Castellanet, M. J., andEllestad, M. H. (1978): ‘Evaluation of R wave amplitude changes versus ST-segment depression in stress testing’,Circulation,57, pp. 904–910Google Scholar
  4. David, D., Naito, M., Michelson, E., Watanbe, Y., Chen, C. C., Morganroth, J., andShaffenburg, M. (1982): ‘Intramiocardial conduction: a major determinant of R-wave amplitude during acute myocardial ischaemia’,Circulation,65, pp. 161–167Google Scholar
  5. Fernández, E. A., Presedo, J., andBarro, S. (1999): ‘An ECG ischemic detection system based on self-organizing maps and sigmoid function pre-processing stage’,Artificial Intelligence in Medicine. Lecture Notes in Artificial Intelligence,1620, pp. 207–216 (Springer-Verlag, Berlin, Heidelberg, New York, 1999)Google Scholar
  6. Ham, F. M., andSoowham, H. (1996): ‘Classification of cardiac arrhythmias using fuzzy ARTMAP’,IEEE Trans. Biomed. Eng.,43, pp. 425–430CrossRefGoogle Scholar
  7. Jager, F., Moody, G. B., andMark, R. G. (1998): ‘Detection of transient ST segment episodes during ambulatory ECG monitoring’,Comput. Biomed. Res.,31, pp. 305–322CrossRefGoogle Scholar
  8. Kohonen, T. (1997): ‘Self-organizing maps’ (Springer-Verlag, Berlin, Heidelberg, New York, 1997), 2nd ednGoogle Scholar
  9. Laguna, P., Moody, G. B., García, J., Goldberger, A. L., andMark, R. G. (1999): ‘Analysis of the ST-T complex of the electrocardiogram using the Karhunen-Loeve transform: adaptive monitoring and alternans detection’,Med. Biol. Eng. Comput.,37, pp. 175–189Google Scholar
  10. Laister, M. J., andRiggs, R. J. (1984): ‘A comparison of template matching and feature extraction of ECG analysis’ in ‘Computers in cardiology 1984’ (IEEE Computer Society Press), pp. 101–104Google Scholar
  11. Maglaveras, N., Stamkopoulos, T., Diamantaras, K., Pappas, C., andStrintzis, M. G. (1998a): ‘ECG pattern recognition and classification using non-linear transformations and neural networks: a review’,Int. J. Med. Informatics,52, pp. 191–208CrossRefGoogle Scholar
  12. Maglaveras, N., Stamkopoulos, T., Pappas, C., andStrintzis, M. G. (1998b): ‘An adaptive backpropagation neural network for real-time ischemic episodes detection: development and performance analysis using the European ST-T database’,IEEE Trans. Biomed. Eng.,45, pp. 805–813CrossRefGoogle Scholar
  13. Moody, G. B., andMark, R. G. (1982): ‘Development and evaluation of a 2-lead ECG analysis program’ in ‘Computers in cardiology 1982’, pp. 39–44Google Scholar
  14. Myers, J., Ahnve, S., Froelicher, V., andSullivan, M. (1985): ‘Spatial R wave amplitudes changes during exercise: relation with left ventricular ischaemia and function’,J. Amer. College Cardiol,6, pp. 603–608Google Scholar
  15. Ng, F., Wong, S., Almeida, D., Mora, F., Hernandez, A., andPassariello, G. (1997): ‘Diagnostic value of R wave amplitude changes in coronary artery disease’ in ‘Computers in cardiology 1997’ (IEEE Computer Society Press), pp. 457–460Google Scholar
  16. Polak, M. J., Zhou, S. H., Rautaharju, P. M., Armstrong, W. W., andChaitman, B. R. (1995): ‘Adaptive logic network compared with backpropagation network in automated detection of ischaemia from resting ECG’ in ‘Computers in cardiology 1995’ (IEEE Computer Society Press), pp. 217–220Google Scholar
  17. Presedo, J., Fernández, E. A., Vila, J., andBarro, S. (1996a) ‘Cycles of ECG parameter evolution during ischemic episodes’ in ‘Computers in cardiology 1996’, pp. 489–492Google Scholar
  18. Presedo, J., Vila, J., Barro, S., Palacios, F., Ruiz, R., Taddei, A., andEmdim, M. (1996b): ‘Fuzzy modelling of the expert's knowledge in ECG-based ischaemia detection’,Fuzzy Sets Syst.,77, pp. 63–75CrossRefGoogle Scholar
  19. Rappaport, S. H., Gillick, L., Moody, G. B., andMark, R. G., (1982): ‘QRS morphology classification: quantitative evaluation of different strategies’ in ‘Computers in cardiology 1982’, pp. 33–38Google Scholar
  20. Silipo, R., andMarchesi, C. (1996): ‘Neural networks architecture for extensive long term ECG analysis’ in ‘Computers in cardiology 1996’, pp. 13–16Google Scholar
  21. Silipo, R., Laguna, P., Marchesi, C., andMark, R. G. (1995): ‘ST-T segment change recognition using artificial neural networks and principal component analysis’ in ‘Computers in cardiology 1995’, pp. 213–216Google Scholar
  22. Taddei, A., Biagini, G., Emdim, M., Mazzei, M. G., Pisani, P., Roggero, N., Varanini, M., Mark, R. G., Moody, G. B., Braaksma, L., Zeelenberg, C., andMarchesi, C. (1991): ‘The European ST-T Database: development, distribution and use’ in ‘Computers in cardiology 1991’ (IEEE Computer Society Press), pp. 177–180Google Scholar
  23. Taddei, A., Constantin, G., Silipo, R, Emdim, M., andMarchesi, C. (1995): ‘A system for the detection of ischemic episodes in ambulatory ECG’ in ‘Computers in cardiology 1995’ (IEEE Computer Society Press), pp. 705–708Google Scholar
  24. Taddei, A., Varanini, M., Macerata, A., Emdim, M., andMarchesi, C. (1992): ‘Detection of ischemic episodes through principal component analysis’,Biotelemetry,XII, pp. 77–84Google Scholar
  25. Thakor, N., Gramatikov, B., andMita, M. (1993): ‘Multiresolution wavelet analysis of ECG during ischaemia and reperfusion’ in ‘Computers in cardiology 1993’ (IEEE Computer Society Press), pp. 895–898Google Scholar
  26. Vila, J., Presedo, J., Delgado, M., Barro, S., Ruiz, R., andPalacios F. (1997): ‘SUTIL: intelligent ischaemia monitoring system’,Int. J. Med. Informatics,47, pp. 193–214CrossRefGoogle Scholar
  27. Wang, J. Y. J. (1983): ‘Application of pattern recognition techniques to QRS complex classification—a review’ in ‘Computers in cardiology 1983’ (IEEE Computer Society Press), pp. 77–82Google Scholar

Copyright information

© IFMBE 2001

Authors and Affiliations

  • E. A. Fernández
    • 1
  • P. Willshaw
    • 1
  • C. A. Perazzo
    • 1
  • R. J. Presedo
    • 2
  • S. Barro
    • 2
  1. 1.Favaloro UniversityBuenos AiresArgentina
  2. 2.University of Santiago de CompostelaCompostelaSpain

Personalised recommendations