Predicting Patient’s Long-Term Clinical Status after Hip Arthroplasty Using Hierarchical Decision Modelling and Data Mining

B. Zupan; J. Demšar; D. Smrke; K. Boižkov; V. Stankovski; I. Bratko; J. R. Beck

doi:10.1055/s-0038-1634460

Methods of Information in Medicine, Table of Contents

Methods Inf Med 2001; 40(01): 25-31
DOI: 10.1055/s-0038-1634460

Original Article

Schattauer GmbH

Predicting Patient’s Long-Term Clinical Status after Hip Arthroplasty Using Hierarchical Decision Modelling and Data Mining

B. Zupan

¹Faculty of Computer and Information Sciences, University of Ljubljana and Slovenia, USA

²J. Stefan Institute, Ljubljana, Slovenia, USA

¹Faculty of Computer and Information Sciences, University of Ljubljana and Slovenia, USA

,

J. Demšar

¹Faculty of Computer and Information Sciences, University of Ljubljana and Slovenia, USA

,

D. Smrke

⁴Department of Traumatology, University Clinical Center, Ljubljana, Slovenia

,

K. Boižkov

⁴Department of Traumatology, University Clinical Center, Ljubljana, Slovenia

,

V. Stankovski

⁴Department of Traumatology, University Clinical Center, Ljubljana, Slovenia

,

I. Bratko

¹Faculty of Computer and Information Sciences, University of Ljubljana and Slovenia, USA

²J. Stefan Institute, Ljubljana, Slovenia, USA

,

J. R. Beck

³Office of Information Technology, Baylor College of Medicine, Houston, TX, USA

› Author Affiliations

Abstract

Construction of a prognostic model is presented for the long-term outcome after femoral neck fracture treatment with implantation of hip endoprosthesis. While the model is induced from the follow-up data, we show that the use of additional expert knowledge is absolutely crucial to obtain good predictive accuracy. A schema is proposed where domain knowledge is encoded as a hierarchical decision model of which only a part is induced from the data while the rest is specified by the expert. Although applied to hip endoprosthesis domain, the proposed schema is general and can be used for the construction of other prognostic models where both follow-up data and human expertise is available.

Keywords

Harris Hip Score - Hip Arthroplasty - Prognostic Models - Data Mining - Hierarchical Decision Models

Full Text

References

References
1 Bellazzi R, Zupan B. Intelligent data analysis in medicine and pharmacology: A position statement. In: IDAMAP-98. Brighton: UK; 1988: 1-4.
2 Bohanec M, Rajkovič V. DEX: An expert system shell for decision support. Sistemica 1990; 1 (Suppl. 01) 145-57.
3 Bohanec M, Zupan B. Rajkovič Hierarchical multi-attribute decision models and their application in health care. In: Medical Informatics Europe ’99. IOS Press; 1999: 670-5.
4 Cestnik B, Bratko I. On estimating probabilities in tree pruning. In: Proc European Working Session on Learning EWSL-91. Berlin: Springer-Verlag; 1991. 1: 138-50.
5 Dietterich TG. Approximate statistical tests for comparing supervised classification learning algorithms. Neural Computa 1989; 10: 1895-924.
6 Domingos P, Pazzani M. Beyond independence: conditions for the optimality of the simple Bayesian classifier. In: Proceedings of the Thirteenth International Conference on Machine Learning. Bari: Morgan Kaufmann; 1996: 105-12.
7 Everitt BS. The Analysis of contingency tables. London: Chapman and Hall; 1977
8 Harris WH. Traumatic arthritis of the hip after dislocation and acetabular fractures: Treatment by mold arthroplasty: end result study using a new method of result evaluation. J Bone Joint Surg 1969; 51-A: 737-55.
9 James SE, Gallannaug SC. Bi-articular hemiarthroplasty of the hip: a 7-year follow-up. Injury 1991; 22 (Suppl. 05) 391-3.
10 Johnston RC, Fitzgerald RH, Poss R, Muller ME, Sledge CB. Clinical and radiographic evaluation of total hip replacement. A standard system of terminology for reporting results. J Bone Joint Surg 1990; 72 (Suppl. 02) 161-8.
11 Kononenko I, Bratko I, Kukar M. Application of machine learning to medical diagnosis. In: Machine Learning and Data Mining: Methods and Applications. Chichester: John Wiley & Sons; 1998: 389-408.
12 Koval KJ, Skovron ML, Aharonoff GB, Zukerman JD. Predictions of functional recovery after hip fracture in the elderly. Clin Orthop 1998; 348: 22-8.
13 Koval KJ, Zukerman JD. Hip fractures are an increasingly important public health problem (editorial). Clin Orthop 1998; 348: 2.
14 Lavrač N, Džeroski S, Pirnat V, Križman V. The utility of background knowledge in learning medical diagnostic rules. Appli Artifi Intelli 1993; 7: 273-93.
15 Lubsen J, Pool J, van der Does E. A practical device for the application of a diagnostic or prognostic function. Method Inform Med 1978; 17: 127-9.
16 Lyons AR. Clinical outcomes and treatment of hip fractures. Am J Med 1997; 103 2A 51S-63.
17 Michie D, Spiegelhalter DJ, Taylor CC. ed. Machine learning, neural and statistical classification. London: Ellis Horwood; 1994
18 Niblett T, Bratko I. Learning decision rules in noisy domains. In: Expert Systems 86. Cambridge University Press; 1986: 15-8. (Proc. EWSL 1986, Brighton).
19 Quinlan R. Induction of decision trees. Machine Learning 1986; 1 (Suppl. 01) 81-106.
20 Saaty TL. Multicriteria decision making: The analytic hierarchy process. Pittsburgh: RWS Publications; 1992
21 Shapiro AD. Structured induction in expert systems. Wokingham, UK: Turing Institute Press in association with Addison-Wesley Publishing Company; 1987
22 Zupan B, Bohanec M, Demšar J, Bratko I. Learning by discovering concept hierarchies. Artifi Intelli 1999; 109 1-2 211-42.
23 Zupan B, Demšar J, Kattan MW, Beck JR, Bratko I. Machine learning for survival analysis: A case study on recurrence of prostate cancer. In: AIMDM-99. Horn W. et al., ed. Heidelberg: Springer; 1999: 346-55.
24 Zupan B, Džeroski S. Acquiring and validating background knowledge for machine learning using function decomposition. Artifici Intelli Med 1998; 14 1-2 101-17.