Discussing Anomalous Situations using Decision Trees: A Head Injury Case Study

A. McQuatt; D. Sleeman; P. J. D. Andrews; V. Corruble; P. A. Jones

doi:10.1055/s-0038-1634195

Methods of Information in Medicine, Table of Contents

Methods Inf Med 2001; 40(05): 373-379
DOI: 10.1055/s-0038-1634195

Original Article

Schattauer GmbH

Discussing Anomalous Situations using Decision Trees: A Head Injury Case Study

Authors

A. McQuatt

¹Department of Computing Science, University of Aberdeen, Aberdeen, Scotland
D. Sleeman

¹Department of Computing Science, University of Aberdeen, Aberdeen, Scotland
P. J. D. Andrews

²Department of Clinical Neurosciences, Western General Hospital, Edinburgh, Scotland
V. Corruble

¹Department of Computing Science, University of Aberdeen, Aberdeen, Scotland
P. A. Jones

²Department of Clinical Neurosciences, Western General Hospital, Edinburgh, Scotland

Abstract

Summary

Objectives: Predicting the outcome of seriously ill patients is a challenging problem for clinicians.

Methods: One alternative to clinical trials is to analyse existing patient data in an attempt to predict the several outcomes, and to suggest therapies. In this paper we use decision tree techniques to predict the outcome of head injury patients. The work is based on patient data from the Edinburgh Royal Infirmary which contains both background (demographic) data and temporal (physiological) data.

Results: The focus of this paper is the discussion of the anomalous cases in the decision trees with the domain experts (the clinicians).

Conclusions: These analyses led to the detection of several situations where both the data analysis and patient data collection should be enhanced, which in turn should lead to improved patient care.

Keywords

Anomalies - Decision Tree Analysis - Demographic Data - Traumatic Brain Injuries - Temporal/Physiological Data

Full Text

References

References
1 Jennett B, Bond M. Assessment of outcome after severe brain damage. A practical scale. Lancet 1975; 1: 480-4.
2 Jones PA, Andrews PJD, Midgley S, Anderson SI, Piper IR, Tocher JL, Housley AM, Corrie JA, Slattery J, Dearden NM, Miller JD. Measuring the burden of secondary Insults in Head-Injured Patients during Intensive Care. J Neurosurg Anaesthesiol 1994; 6 (Suppl. 01) 4-14.
3 Choi SC, Muizelaar JP, Barnes TY, Mamarou A, Brooks DM, Young HF. Young. Prediction tree for severely head-injured patients. J Neurosurg 1991; 75: 251-5.
4 Pilih IA, Mladenic D, Lavrac N, Prevec TS. Data analysis of patients with severed head injury. In: Intelligent Data Analysis in Medicine and Pharmacology. Boston: Kluwer Academic Publishers; 1997
5 Breiman L, Friedman JH, Olshen RA. et al. Classification and regression trees. Belmont, California: Wadsworth; 1984
6 Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical scale. Lancet 1974; 2: 81-4.
7 Gerstenbrand F, Saltuari L, Kofler M, Formisano R. Clinical evaluation of severe head injury. Neurologia 1990; 39 (Suppl. 01) 71-88.
8 Mladenic D. The learning system Magnus Assistant. BSc Thesis. Faculty of Computer and Information Sciences, University of Ljubljana; Slovenia: 1990
9 Quinlan JR. Induction of decision trees. Machine Learning 1986; 1: 81-106.
10 Quinlan JR. Probabilistic Decision Trees. In: Machine Learning: An A. I. Approach. III, Morgan Kaufmann 1990: 140-52.
11 Quinlan JR. Data Mining Tools C5.0 and See5. Rulequest Research, 1997. Web Ref: http://www.rulequest.com/see5-info.html.
12 Quinlan JR. C4.5: Programs for Machine Learning. Morgan Kaufmann 1993
13 McQuatt A. Using Machine Learning Techniques to Predict Clinical Outcome. MSc Thesis. Computing Science Department, The University of Aberdeen; 1999