Download PDF
Methods Inf Med 1990; 29(04): 330-336
DOI: 10.1055/s-0038-1634792
ECG Interpretation Systems
Schattauer GmbH

Classification Methods for Computerized Interpretation of the Electrocardiogram

J. A. Kors
1   Department of Medical Informatics, Faculty of Medicine and Health Sciences, Erasmus University, Rotterdam, The Netherlands
,
J. H. van Bemmel
1   Department of Medical Informatics, Faculty of Medicine and Health Sciences, Erasmus University, Rotterdam, The Netherlands
› Author Affiliations
Further Information

Publication History

Publication Date:
06 February 2018 (online)

    Permissions and Reprints

Abstract

Two methods for diagnostic classification of the electrocardiogram are described: a heuristic one and a statistical one. In the heuristic approach, the cardiologist provides the knowledge to construct a classifier, usually a decision tree. In the statistical approach, probability densities of diagnostic features are estimated from a learning set of ECGs and multivariate techniques are used to attain diagnostic classification. The relative merits of both approaches with respect to criteria selection, comprehensibility, flexibility, combined diseases, and performance are described. Optimization of heuristic classifiers is discussed. It is concluded that heuristic classifiers are more comprehensible than statistical ones; encounter less difficulties in dealing with combined categories; are flexible in the sense that new categories may readily be added or that existing ones may be refined stepwise. Statistical classifiers, on the other hand, are more easily adapted to another operating environment and require less involvement of cardiologists. Further research is needed to establish differences in performance between both methods. In relation to performance testing the issue is raised whether the ECG should be classified using as much prior information as possible, or whether it should be classified on itself, explicitly discarding information other than age and sex, while only afterwards other information will be used to reach a final diagnosis. Consequences of taking one of both positions are discussed.

Key-Words

ECG Classification - Optimization - Gold Standard
 

  • REFERENCES

  • 1 Willems JL. Introduction to multivariate and conventional computer ECG analysis: Pro’s and contra’s. In: Van Bemmel JH, Willems JL. eds. Trends in Computer Processed Electrocardiograms. Amsterdam: North-Holland Publ Comp; 1977: 213-28.
    Google Scholar
  • 2 Willems JL, Abreu-Lima C, Amaud P, Van Bemmel JH. et al. Testing the performance of ECG computer programs: The CSE diagnostic pilot study. J Electro-cardiol 1987; Suppl: 73-7.
    PubMedGoogle Scholar
  • 3 Macfarlane PW, Watts MP, Podolski M, Shoat D, Lawrie TDV. The new Glasgow system. In: Willems JL, Van Bemmel JH, Zywietz C. eds. Computer ECG Analysis: Towards Standardization. Amsterdam: North-Holland Publ comp; 1986: 31-6.
    Google Scholar
  • 4 Talmon JL, Van Herpen G. Quantitative classification of T-wave abnormalities in the VCG. In: Advances in Cardiology. Vol 16. Basel: Karger, 1976: 233-6.
    Google Scholar
  • 5 Breiman L, Friedman JH, Olshen RA, Stone CI. Classification and Regression Trees. Belmont: Wadsworth, 1984.
    PubMedGoogle Scholar
  • 6 Talmon JL. A multiclass nonparametric partitioning algorithm. In: Gelsema ES, Kanal LN. eds. Pattern Recognition in Practice II. Amsterdam: North-Holland Publ Comp; 1984: 449-59.
    Google Scholar
  • 7 Talmon JL. Pattern Recognition of the ECG: A Structured Analysis (Ph.D. dissertation). Amsterdam: Free University; 1983
    Google Scholar
  • 8 Zadeh LA. Fuzzy sets as a basis for a theory of possibility. Fuzzy Sets and Systems 1978; 01: 3-28.
    CrossrefPubMedGoogle Scholar
  • 9 Degani R, Pacini G. Linguistic pattern recognition algorithms for computer analysis of ECG. Proc Biosigma 1978; 18-26.
    PubMedGoogle Scholar
  • 10 Bortolan G, Degani R. Comparative evaluation of three models of fuzzy decisionmaking in electrocardiography. In: Van Bemmel JH, Ball MJ, Wigertz O. eds. MEDINFO-83. Amsterdam: North-Holland Publ Comp; 1983: 537-40.
    Google Scholar
  • 11 Pipberger HV, Dunn RA, Cornfield J. First and second generation computer programs for diagnostic ECG and VCG classification. Proc XUth Int Coll Vectorcardio-graphicum. Brussels: Presses Acad Europ; 1972: 431-9.
    Google Scholar
  • 12 Cornfield J, Dunn RA, Batchlor CD, Pipberger HV. Multigroup diagnosis of electrocardiograms. Comp Biomed Res 1973; 06: 97-120.
    CrossrefPubMedGoogle Scholar
  • 13 Pipberger HV, McCaughan D, Littman D, Pipberger HA. et al. Clinical application of a second generation electrocardiographic computer program. Am J Cardiol 1975; 35: 597-608.
    CrossrefPubMedGoogle Scholar
  • 14 Lesaffre E. Logistic Discriminant Analysis with Applications in Electrocardiography (Ph.D. dissertation). Leuven: Catholic University; 1986
    Google Scholar
  • 15 Willems JL, Lesaffre E, Pardaens J, De Schreye D. Multigroup logistic classification of the standard 12- and 3-lead ECG. In: Willems JL, Van Bemmel JH, Zywietz C. eds. Computer ECG Analysis: Towards Standardization. Amsterdam: North-Holland Publ Comp; 1986: 203-10.
    Google Scholar
  • 16 Anderson JA. Separate sample logistic discrimination. Biometrika 1972; 59: 19-35.
    CrossrefPubMedGoogle Scholar
  • 17 Bailey JJ, Horton M. Advantages of automation of ECG analysis with conventional (heuristic) criteria. I. In: Van Bemmel JH, Willems JL. eds. Trends in Computer Processed Electrocardiograms. Amsterdam: North-Holland Publ Comp; 1977: 221-8.
    Google Scholar
  • 18 Brohet CR, Robert A, Derwael C, Fesler R. et al. Computer interpretation of pedia-reic orthogonal electrocardiograms: Statistical and deterministic classification methods. Circulation 1984; 70: 255-63.
    CrossrefPubMedGoogle Scholar
  • 19 Matthes T, Göttsch G, Zywietz C. Interactive analysis of statistical ECG diagnosis on an intelligent electrocardiograph. In: Willems JL, Van Bemmel JH, Zywietz C. eds. Computer ECG Analysis: Towards Standardization. Amsterdam: North-Holland Publ Comp; 1986: 215-20.
    Google Scholar
  • 20 Michaelis J, Wellek S, Willems JL. Reference standards for software evaluation. Meth Inform Med 1990; 29: 289-97.
    PubMedGoogle Scholar
  • 21 Willems JL, Abreu-Lima C, Arnaud P, Brohet C. et al. Evaluation of ECG interpretation results obtained by computer and cardiologists. Meth Inform Med 1990; 29: 308-16.
    PubMedGoogle Scholar
  • 22 Bourdillon PJ, Kilpatrick D. Clinicians, the Mount Sinai program and the Veterans’ Administration program evaluated against clinico-pathological data derived independently of the electrocardiogram. Eur J Cardiol 1978; 08: 395-412.
    PubMedGoogle Scholar
  • 23 Bailey JJ, Horton M, Itscoitz SB. A method for evaluating computer programs for electrocardiographic interpretation. III. Reproducibility testing and the sources of program errors. Circulation 1974; 50: 88-93.
    CrossrefPubMedGoogle Scholar
  • 24 Bailey JJ, Horton M, Itscoitz SB. The importance of reproducibility testing of computer programs for electrocardiographic interpretation: Application to the automatic vectorcardiographic analysis program. Comp Biomed Res 1976; 09: 307-16.
    CrossrefPubMedGoogle Scholar
  • 25 Dudeck J. Reproducibility of diagnostic statements in first and second generation ECG programs. In: Willems JL, Van Bemmel JH, Zywietz C. eds. Computer ECG Analysis: Towards Standardization. Amsterdam: North-Holland Publ Comp; 1986: 235-8.
    Google Scholar
  • 26 Helppi RR, Unite V, Wolf HK. Suggested minimal performance requirements and methods of performance evaluation for computer ECG analysis programs. Can Med Ass J 1973; 108: 1251-9.
    PubMedGoogle Scholar
  • 27 Toussaint GT. Bibliography on estimation of misclassification. IEEE Trans Info Theory 1974; 20: 472-9.
    CrossrefPubMedGoogle Scholar
  • 28 Jenkins G, Clifton J, Laks M. The communication problem between the computer programmer, the electrocardiographer and the ECG computer. In: Tolan GD, Pryor TA. eds. Computerized Interpretation of the ECG IV. New York: Engin Foundation; 1979: 148-61.
    Google Scholar
  • 29 Balda RA, Vallance AG, Luszcz JM, Stah-Hn FJ, Diller G. ECL: A medically oriented ECG criteria language and other research tools. In: Ostrow HG, Ripley KL. eds. Computers in Cardiology. Long Beach: IEEE Comp Soc; 1978: 481-95.
    Google Scholar
  • 30 Laks M. Experiences with electrocardiographic computer languages in the development of ECG criteria. In: Tolan GD, Pryor TA. eds. Computerized Interpretation of the ECG V. New York: Engin Foundation; 1980: 153-78.
    Google Scholar
  • 31 Rubel P, Arnaud P, Prevot D. Système d’aide à la decision. Application à I’inter-prètation automatique des vectocardio-grammes. Int J Bio-Medical Computing 1975; 06: 193-211.
    PubMedGoogle Scholar
  • 32 Kors JA, Van Herpen G, De Jong T, Van Bemmel JH. Interactive optimization of heuristic ECG classifiers (to be publ).
    PubMed
  • 33 Simonson E, Tuna N, Okamoto N, Toshi-ma H. Diagnostic accuracy of vectorcardiogram and electrocardiogram. Am J Cardiol 1966; 17: 829-78.
    CrossrefPubMedGoogle Scholar
  • 34 Linstone HA, Turoff M. The Delphi Method: Techniques and Applications. Reading: Addison-Wesley; 1975
    Google Scholar
  • 35 Kors JA, Van Bemmel JH. The Delphi method: A review of its applications in medicine. In: Barber B, Cao D, Qin D, Wagner G. eds. MEDINFO-89. Amsterdam: North-Holland Publ Comp; 1989: 156-60.
    Google Scholar
  • 36 Kors JA, Sittig AC, Van Bemmel JH. The Delphi method to validate diagnostic knowledge in computerized ECG interpretation. Meth Inform Med 1990; 29: 44-50.
    PubMedGoogle Scholar