Using data mining to Improve Clinical Decision Support

  • K. E. Burn-Thornton
  • S. I. Thorpe
  • J. Attenborough
Conference paper


In this paper we report that it is important to train clinical decision support systems on as large a data set as possible. We describe how the diagnosis of heart disease, from 12-lead ECGs, may be improved by up to 32%, with respect to diagnosis using a small data set, by the use of data mining algorithms trained on a sufficiently large data set. We also describe how these suitably trained, and chosen, data mining algorithms appear to be capable of provision of clinical decision support by virtue of the short time, 1-3 seconds, required to perform patient state classification.


Data Mining Acute Myocardial Infarction Classification Accuracy Knowledge Discovery Clinical Decision Support 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Sittig DF, Stead WW. Computer-Based Physician Order Entry–The State of the Art. J. Amer. Med. Inform. Assoc. 1994; 1: 108–123.CrossRefGoogle Scholar
  2. 2.
    Clayton PD, Hripcsack G. Decision Support in Health-Care. Int. J. Bio-Med Comp 1995; 39: 59–66.CrossRefGoogle Scholar
  3. 3.
    Burn-Thornton KE, Edenbrandt L. Pinpointing the Key Indicators of 12-lead ECGs Using data mining Computers & Medicine 1998; 31; 293–303.Google Scholar
  4. 4.
    Burn-Thornton KE, Garibaldi J, Hamer P: data mining Algorithms: Applicability to DataSet Types. In: INC 98, Plymouth, U.K, 1998.Google Scholar
  5. 5.
    Mulholland G, Brewer B: Improving Your Skills in 12 Lead ECG Interpretation. Ems, New York, 1996.Google Scholar
  6. 6.
    Montgomery AA, Fahey TA. Systematic Review of the use of Computers in the Management of Hypertension. J. Epidemiology and Community Health 1998; 52: 520–525.CrossRefGoogle Scholar
  7. 7.
    Similar Y, Musen MA. Knowledge-based Temporal Abstraction in Clinical Domains. Artif. Intell. In Med. 1996; 8: 267–298.Google Scholar
  8. 8.
    Knirsch CA, Jain NL, PablosMendez A, Friedman C, Hripcsak A. Respiratory Isolation of Tuberculosis Patients using Clinical guidelines and an Automated Clinical Decision Support System. Infection Control and Hospital Epidemiology 1998; 19: 94–100.CrossRefGoogle Scholar
  9. 9.
    Pappas C, Mavromatis A, Maglaveras N, Tsikotis A, Pangalos G, Ambrosiadou V. Cardiological database management system as a mediator to clinical decision support. Meth. Inf. in Med. 1996; 35: 52–58.Google Scholar
  10. 10.
    Willems J, Abreu-Lima C, Arnaud P. The Diagnostic Performance of Computer Programs for Interpretation of Electrocardiograms. N. Engl. J. Med. 1991; 325: 1767–1773.CrossRefGoogle Scholar
  11. 11.
    Heden B, Ohlin H, Rittner R, Edenbrandt L: Acute Myocardial Infarction Detected in the 12-lead ECG by Artificial Neural Networks. Circulation 1997; 96: 1798–1802.CrossRefGoogle Scholar
  12. 12.
    Gordon C, Johnson P, Waite C, Veloso M: Algorithm and Care Pathway: Clinical Guidelines and Healthcare Processes. Lect Notes in A. I. 1997; 121: 66–69.Google Scholar
  13. 13.
    Petersen J: Business Applications of Statistics for data mining. In: Proc. UNICOM Seminar on data mining, London, July 1995, pp 157–165.Google Scholar
  14. 14.
    Leon M, Valamudi P. Data Warehouse vendors do data mining Info. World 1996; 18: pp. 38–42.Google Scholar
  15. 15.
    Bounds D, Lloyd P, Mathew B. A Comparison of Neural Networks and Other Pattern Recognition Approaches to the Diagnosis of Low Back Disorders. Neural Networks 1990; 3: 583–591.CrossRefGoogle Scholar
  16. 16.
    Waddell G: A New Clinical Model for the Treatment of Lower Back-pain. Spine 1998; 12: 632–644.CrossRefGoogle Scholar
  17. 17.
    McClean S, Scotney B, Shapcott M: Using background Knowledge with Attribute-Orientated data mining In: IEE Colloquium on Knowledge Discovery and data mining,, London, May 1998, pp 1:1–4.CrossRefGoogle Scholar
  18. 18.
    McSherry D. Attribute-value Distribution as a Strategy for Increasing the Efficiency of data mining, In: IEE Colloquium on Knowledge Discovery and data mining, London, May 1998, pp 2:1–3.Google Scholar
  19. 19.
    Liu X. Strategies for Outlier Analysis. In: IEE Colloquium on Knowledge Discovery and data mining, London, May 1998, pp 3:1–3.Google Scholar
  20. 20.
    Chen X, Petrounias I: An Architecture for Temporal data mining. In: IEE Colloquium on Knowledge Discovery and data mining, In: WE Colloquium on Knowledge Discovery and data mining, London, May 1998, pp 4:1–4.Google Scholar
  21. 21.
    Liu X, Cheng G, Wu J: Noise and Uncertainty Management in Intelligent Data Modelling. In: Proc. 12th Nat. Conf. on Artificial Intelligence [AAAI-94], 1994, pp 263–268.Google Scholar
  22. 22.
    Barnet V. The Ordering of Multivariate Data. J. Roy. Statis. Soc. A. 1976; 39: 318–54.CrossRefGoogle Scholar
  23. 23.
    Huber P: Robust Statistics. 1st edition, John Wiley, London, 1981.MATHCrossRefGoogle Scholar
  24. 24.
    Barnet V, Lewis T: Outliers in Statistical Data 1st edition, John Wiley, London, 1994.Google Scholar
  25. 25.
    Hawkins D: Identification of Outliers, Chapman & Hall, 1st edition, London, 1980.MATHGoogle Scholar
  26. 26.
    Chen A, Tseng F: Evaluating Aggregate Operation over Imprecise Data: In: IEE Trans. On Knowledge Discovery and Data Engineering, 1996; 8: 273–284.Google Scholar
  27. 27.
    McSherry D. Knowledge Discovery by Inspection. Decision Support Systems 1997; 21: 43–47.CrossRefGoogle Scholar
  28. 28.
    Quinlan J: Bagging and Boosting and C4.5. In: Proc. Of 13th Nat. Conf. On AI, AAI Press, 1996, pp 725–730.Google Scholar
  29. 29.
    Kohavi R, John G: Wrappers for Feature Subset Selection. All special issue on relevance, 1997.Google Scholar
  30. 30.
    Imielinski T, Manila H: A Database Perspective on Knowledge Discovery. CACM, 1996, 39: 58–64.Google Scholar
  31. 31.
    Saraee M, Theodoulidis B: Knowledge Discovery in Temporal Databases. In: Proc. In: IEE Colloqium on Knowledge Discovery in Databases, London, May, 1995, pp 1:1–4.CrossRefGoogle Scholar
  32. 32.
    Chen X, Petrounias L Heathfield H: Discovering Temporal Association Rules in Temporal Databases. In: Proc. LADT, Berlin, Germany, 1998.Google Scholar
  33. 33.
    Enas G, Choi S. Choice of the Smoothing Parameter and Efficiency of the k-mi Neighbour Classification. Comput. Math. Applic. 1986; 12A: 235–244.CrossRefGoogle Scholar
  34. 34.
    Friedman J: Regularised Discriminant Analysis, J. Amer. Statist. Assoc. 1989; 84: pp. 165–175.MathSciNetCrossRefGoogle Scholar
  35. 35.
    Quinlan J. Simplifying decision trees, Int. J. Man-Machine Studies 1987; 27: 221–234.CrossRefGoogle Scholar
  36. 36.
    Michie D, Spiegelhalter DJ, Taylor CC [Eds]: Machine Learning & Statistical Classification: Ellis Horwood, London, 1994.MATHGoogle Scholar
  37. 37.
    Burn-Thornton KE, Thorpe SI, Edenbrandt L: Improving Clinical Decision Support Using data mining In Proc. SPIE ‘99, Orlando, April 1999, vol. 3495, pp 207–15.Google Scholar
  38. 38.
    Burn-Thornton KE, Thorpe SI, Singh H, Macleod N: A Methodology for Determining the Minimum Size of Data Set Required for Accurate Domain Analysis, in press.Google Scholar

Copyright information

© Springer-Verlag London 2000

Authors and Affiliations

  • K. E. Burn-Thornton
    • 1
  • S. I. Thorpe
    • 1
  • J. Attenborough
    • 1
  1. 1.Data Mining Group, Department of Computer ScienceUniversity of Durham, Science LaboratoriesDurhamUK

Personalised recommendations