RSS-Feed abonnieren
DOI: 10.1055/s-0038-1634813
Bayesian Diagnostic Probabilities without Assuming Independence of Symptoms
We wish to thank Mr. S. J. Nixon MB, ChB, BSc, FRCS of the General Surgical Unit, Western General Hospital, Edinburgh, and Dr. A. A. Gunn, formerly of the Bangour Hospital, Roxburgh, for their expert advice and for making available data originally collected at Bangour Hospital. We also wish to thank Yiqun Gu r’cor writing computer programs and Jlbyce Smith for typesetting this paper.Publikationsverlauf
Publikationsdatum:
08. Februar 2018 (online)
The paper describes an application of Bayes’ Theorem to the problem of estimating from past data the probabilities that patients have certain diseases, given their symptoms. The data consist of hospital records of patients who suffered acute abdominal pain. For each patient the records showed a large number of symptoms and the final diagnosis, to one of nine diseases or diagnostic groups. Most current methods of computer diagnosis use the “Simple Bayes” model in which the symptoms are assumed to be independent, but the present paper does not make this assumption. Those symptoms (or lack of symptoms) which are most relevant to the diagnosis of each disease are identified by a sequence of chi-squared tests. The computer diagnoses obtained as a result of the implementation of this approach are compared with those given by the “Simple Bayes” method, by the method of classification trees (CART), and also with the preliminary and final diagnoses made by physicians.
-
REFERENCES
- 1 Thatcher AR. Contribution to the discussion on a paper by Lauritzen and Spiegelhalter. J Royal Statist Soc, Series B 1988; 50: 196.
- 2 Thatcher AR. Computer Models of Probabilistic Reasoning: Bayes’ Theorem Without Assuming Independence. Technical Report No. 88/1 Heriot-Watt University, Edinburgh. 1988
- 3 Cumberbatch J, Heaps HS. A disease-conscious method for sequential diagnosis by use of disease probabilities without assumption of symptom independence. Int J Biomed Comp 1976; 07: 66-78.
- 4 Breiman L, Friedman JH, Olshen RA, Stone CJ. Classification and Regression Trees. Wadsworth Cal. 1984
- 5 De Dombal FT. Computer-aided diagnosis of acute abdominal pain. Brit Med J 1972; 02: 9-13.
- 6 Adams ID. Computer-aided diagnoses of acute abdominal pain: a multicentre study. Brit Med J 1986; 293: 800-4.
- 7 Spiegelhalter DJ, Knill-Jones RP. Statistical and knowledge-based approaches to clinical decision-support systems with an application in gastro-enterology (with discussion). Royal Statist Soc, Series A 1984; 147: 35-77.
- 8 Lauritzen SL, Spiegelhalter DJ. Local computations with probabilities on graphical structures and their application to expert systems. J Royal Statist Soc, Series B 1988; 50: 157-224.
- 9 Duda RO, Gasching J, Hart PE. Model Design in the Prospector Consultant System for Mineral Exploration Expert Systems in Micro Electronic Age. Edinburgh University Press. 1979
- 10 Cox PR, Broughton RK. Micro Expert Users Manual (Version 2.1). ISIS Systems Ltd. 1981
- 11 Gammerman A, Lanin M. Multivariate information analysis of qualitative factors. Bull Math Appl Genetics 1978; 61: 99-104.
- 12 Stuart A, Ord JK. Kendall’s Advanced Theory of Statistics. Vol.1. London: Griffin; 1987
- 13 Thatcher AR. Relationships between Bayesian and confidence limits for predictions. J Royal Statist Soc, Series B 1964; 26: 176-92.
- 14 Pearson ES, Hartley HO. Biometrika Tables for Statisticians. Vol. 1. Cambridge University Press; 1954
- 15 Gammerman A, Thatcher AR. Bayesian Inference in an Expert System without Assuming Independence. Technical Report 88/11, Heriot-Watt University, Edinburgh. 1988