Patients Classification by Risk Using Cluster Analysis and Genetic Algorithms

  • Max Chacón
  • Oreste Luci
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2905)


Knowing a patient’s risk at the moment of admission to a medical unit is important for both clinical and administrative decision making: it is fundamental to carry out a health technology assessment. In this paper, we propose a non-supervised learning method based on cluster analysis and genetic algorithms to classify patients according to their admission risk. This proposal includes an innovative way to incorporate the information contained in the diagnostic hypotheses into the classification system. To assess this method, we used retrospective data of 294 patients (50 dead) admitted to two Adult Intensive Care Units (ICU) in the city of Santiago, Chile. An area calculation under the ROC curve was used to verify the accuracy of this classification. The results show that, with the proposed methodology, it is possible to obtain an ROC curve with a 0.946 area, whereas with the APACHE II system it is possible to obtain only a 0.786 area.


Genetic Algorithm Health Technology Assessment Adult Intensive Care Unit Diagnostic Hypothesis Genetic Algorithm Grouping 
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.


  1. 1.
    Iezzoni, L.I.: An introduction to risk adjustment. Am. J. Med. Qual. 11, 8–11 (1996)Google Scholar
  2. 2.
    Marvin, N., Bower, M., Rowe, J.E. and AI Group.: De Montfort University, Milton Keynes, UK,:An evolutionary approach to constructing prognostic models. Artif. Intell. Med. 15, 155–165 (1999)CrossRefGoogle Scholar
  3. 3.
    Evans, R.W.: Health care technology and the inevitability of resource allocation and rationing decisions. JAMA 249, 2047–2053 (1983)CrossRefGoogle Scholar
  4. 4.
    Le Gall, J.R., Lemeshow, S., Saulnier, F.: new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 270, 2957–2963 (1993)CrossRefGoogle Scholar
  5. 5.
    Knaus, W.A., Zimmerman, J.E., Wagner, D.P., Draper, E.A., Lawrence, D.E.: APACHE-acute physiology and chronic health evaluation: a physiologically based classification system. Crit. Care Med. 9, 591–597 (1981)CrossRefGoogle Scholar
  6. 6.
    Shwartz, M., Iezzoni, L.I., Moskowitz, M.A., Ash, A.S., Sawitz, E.: The importance of comorbidities in explaining differences in patient costs. Med. Care 34, 767–782 (1996)CrossRefGoogle Scholar
  7. 7.
    Dybowski, R., Weller, P., Chang, R., Gant, V.: Prediction of outcome in critically ill patients using artificial neural network synthesised by genetic algorithm. Lancet 347, 1146–1150 (1996)CrossRefGoogle Scholar
  8. 8.
    Peña-Reyes, C.A., Sipper, M.: Evolutionary computation in medicine: an overview. Artif. Intell. Med. 19, 1–23 (2000)CrossRefGoogle Scholar
  9. 9.
    Sierra, B., Larranaga, P.: Predicting survival in malignant skin melanoma using Bayesian networks automatically induced by genetic algorithms. An empirical comparison between different approaches. Artif. Intell. Med. 14, 215–230 (1998)Google Scholar
  10. 10.
    Horbar, J.D., Onstad, L., Wright, E.: Predicting mortality risk for infants weighting 501- 1500 grams at birth: A National Institute of Health Neonatal Research Network report. Crit. Care Med. 21, 12–18 (1993)CrossRefGoogle Scholar
  11. 11.
    Knaus, W.A., Draper, E.A., Wagner, D.P., Zimmerman, J.E.: APACHE II: a severity of disease classification system. Crit. Care Med. 13, 818–829 (1985)CrossRefGoogle Scholar
  12. 12.
    Weinstein, M.C., Fineberg, H.V.: Clinical Decision Analysis. W.B. Saunders Company, Philadelphia (1980)Google Scholar
  13. 13.
    Aldenderfer, M., Blashfield, R.: Cluster Analysis. Sage University Paper, California (1984)Google Scholar
  14. 14.
    Goldberg, D.E.: Genetic Algorithms in Search, Optimization Machine Learning. Addison-Wesley, Reading (1989)zbMATHGoogle Scholar
  15. 15.
    Calinski, T., Harabasz, J.: A Dendrite Method For Cluster Analysis. Communications in Statistics 3, 1–27 (1974)MathSciNetCrossRefGoogle Scholar
  16. 16.
    Ding, H., El-Keib, A.A., Smith, R.E.: Optimal Clustering of Power Networks Using Genetic Algorithms, TVGA Report No. 92001, University of Alabama, Tuscaloosa, AL (1992)Google Scholar
  17. 17.
    Falkenauer, E.: Genetic Algorithms and Grouping Problems. John Wiley & Sons, Chichester (1999)Google Scholar
  18. 18.
    Becker, R.B., Zimmerman, J.E.: ICU scoring systems allow prediction of patient outcomes and comparison of ICU performance. Crit. Care Clin. 12, 503–514 (1996)CrossRefGoogle Scholar
  19. 19.
    Thibault, G.: Prognosis and clinical predictive models for critically ill patients, Brockton-West Roxbury Veterans Affairs Medical Center, West Roxbury, Massachusetts (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Max Chacón
    • 1
  • Oreste Luci
    • 1
  1. 1.Informatic Engineering DepartmentUniversity of Santiago de ChileSantiagoChile

Personalised recommendations