Probabilistic Diagnosis Using a Reformulation of the INTERNIST-1/QMR Knowledge Base

M. A. Shwe; B. Middleton; D. E. Heckerman; M. Henrion; E. J. Horvitz; H. P. Lehmann; G. F. Cooper

doi:10.1055/s-0038-1634846

Methods of Information in Medicine, Inhaltsverzeichnis

Methods Inf Med 1991; 30(04): 241-255
DOI: 10.1055/s-0038-1634846

Original Article

Schattauer GmbH

Probabilistic Diagnosis Using a Reformulation of the INTERNIST-1/QMR Knowledge Base

I. The Probabilistic Model and Inference Algorithms

Autoren

M. A. Shwe

¹Section on Medical Informatics, Stanford University, Stanford, CA
B. Middleton

¹Section on Medical Informatics, Stanford University, Stanford, CA
D. E. Heckerman

¹Section on Medical Informatics, Stanford University, Stanford, CA
M. Henrion

¹Section on Medical Informatics, Stanford University, Stanford, CA
E. J. Horvitz

¹Section on Medical Informatics, Stanford University, Stanford, CA
H. P. Lehmann

¹Section on Medical Informatics, Stanford University, Stanford, CA
G. F. Cooper

^*Section of Medical Informatics, University of Pittsburgh, Pittsburgh, PA, U.S.A

¹Section on Medical Informatics, Stanford University, Stanford, CA

Abstract

Abstract:

In Part I of this two-part series, we report the design of a probabilistic reformulation of the Quick Medical Reference (QMR) diagnostic decision-support tool. We describe a two-level multiply connected belief-network representation of the QMR knowledge base of internal medicine. In the belief-network representation of the QMR knowledge base, we use probabilities derived from the QMR disease profiles, from QMR imports of findings, and from National Center for Health Statistics hospital-discharge statistics.

We use a stochastic simulation algorithm for inference on the belief network. This algorithm computes estimates of the posterior marginal probabilities of diseases given a set of findings. In Part II of the series, we compare the performance of QMR to that of our probabilistic system on cases abstracted from continuing medical education materials from Scientific American Medicine. In addition, we analyze empirically several components of the probabilistic model and simulation algorithm.

Key-Words

Expert Systems - Computer-aided Diagnosis - Probabilistic Inference - Belief Networks

Volltext

Referenzen

REFERENCES
1 Miller RA, Pople HEJ, Myers JD. Internist-1: An experimental computer-based diagnostic consultant for general internal medicine. N Engl J Med 1982; 307: 468-76.
2 Miller RA. INTERNIST-1/CADUCEUS: Problems facing expert consultant programs. Meth Inform Med 1984; 23: 9-14.
3 Miller RA, McNeil MA, Challinor SM, Masarie FEJ, Myers JD. The INTERNIST-1/QUlCK MEDICAL REFERENCE project – Status report. West J Med 1986; 145: 816-22.
4 Miller R, Masarie FE, Myers JD. Quick medical reference (QMR) for diagnostic assistance. MD Computing 1986; 03: 34-48.
5 First MB, Softer LJ, Miller RA. QUICK (Quick Index to Caduceus Knowledge): Using the INTERNIST-1/CADUCEUS knowledge base as an electronic textbook of medicine. Comp Biomed Res 1985; 18: 137-65.
6 Miller RAEMF. Use of the Quick Medical Reference program as a tool for medical education. Meth Inform Med 1989; 28: 340-5.
7 Von Neumann J, Morgenstern O. Theory of Games and Economic Behavior, Princeton. New York: Princeton Univ Press; 1947
8 Spiegelhalter DJ, Franklin RCG, Bull K. Assessment, criticism and improvement of imprecise subjective probabilities for a medical expert system. In: Machine Intelligence and Pattern Recognition: Uncertainty in Artificial Intelligence 5. Henrion M, Shachter R, Kanal LN, Lemmer JF. eds. Amsterdam: North-Holland Publ Comp; 1990: 285-94.
9 Harris N, Spiegelhalter DJ, Bull K, Franklin RC. Criticizing conditional probabilities in belief networks. In: Proceedings of the Fourteenth Annual Symposium on Computer Applications in Medical Care. Miller RA. ed. Los Alamitos, CA: IEEE Comp Soc Press; 1990: 805-9.
10 Masarie Jr FE, Miller RA, Myers JD. INTERNIST-1 properties: Representing common sense and good medical practice in a computerized medical knowledge base. Comp Biomed Res 1985; 18: 458-79.
11 Heckerman DE, Horvitz EJ, Nathwani BN. Update on the Pathfinder project. In: Proceedings of the Thirteenth Annual Symposium on Computer Applications in Medical Care. Kingsland LC. ed. Los Alamitos, CA: IEEE Comp Soc Press; 1989: 203-7.
12 deDombal FT, Leaper DJ, Horrocks JC, Staniland JR, McCain AP. Human and computer-aided diagnosis of abdominal pain: Further report with emphasis on performance. Brit Med J 1974; 01: 376-80.
13 Bankowitz RA, McNeil MA, Challinor SM, Parker RC, Kapoor WN, Miller RA. A computer-assisted medical diagnostic consultation service: Implementation and prospective evaluation of a prototype. Ann Inter Med 1989; 110: 824-32.
14 Cooper GF. Current research directions in the development of expert systems based on belief networks. Appl Stoch Models Data Anal 1989; 05: 39-52.
15 Horvitz EJ, Breese JS, Henrion M. Decision theory in expert systems and artificial intelligence. J Approx Reas 1988; 02: 247-302.
16 Pearl J. Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. San Mateo, CA: Morgan Kaufman; 1988
17 Howard RA, Matheson JE. Influence diagrams. In: Readings on the Principles and Applications of Decision Analysis. Howard RA, Matheson JE. eds. Menlo Park, CA: Strategic Decisions Group; 1981: 721-62.
18 Chavez RM, Cooper GF. Hypermedia and randomized algorithms for medical expert systems. Comp Meth Progr Biomed 1990; 32: 5-16.
19 Cooper GF. NESTOR: A Computer-Based Medical Diagnostic Aid That Integrates Causal and Probabilistic Knowledge. (Ph. D. Diss). Medical Information Sciences. Stanford, CA: Stanford University; 1984
20 Andersen SK, Olesen KG, Jensen FV, Jensen F. HUGIN – A shell for building Bayesian belief universes for expert systems. In: Eleventh International Joint Conference on Artificial Intelligence. Sridharan NS. ed. San Mateo, CA: Morgan Kaufmann Publ; 1989: 1080-5.
21 Heckerman DE. A tractable inference algorithm for diagnosing multiple diseases. In: Machine Intelligence and Pattern Recognition: Uncertainty in Artificial Intelligence 5. Henrion M, Shachter R, Kanal LN, Lemmer JF. eds. Amsterdam: North-Holland Publ Comp; 1990: 163-72.
22 Henrion M. Towards efficient probabilistic diagnosis in multiply connected networks. In: Influence Diagrams, Belief Nets and Decision Analysis. Oliver RM, Smith JQ. eds. Chichester: Wiley; 1990: 385-407.
23 Peng Y, Reggia JA. A probabilistic causal model for diagnostic problem solving – part I: Integrating symbolic causal inference with numeric probabilistic inference. IEEE Trans Syst Man, and Cybern. 1987. SMC-17: 146-62.
24 Ganote DP. A Bayesian Set Covering Model Applied to Diagnosis in Blood Typing. (M. Sc. Thesis). Department of Computer Science. University of Ohio; 1989
25 Habbema JDF. Models for diagnosis and detection of combinations of diseases. In: Decision Making and Medical Care. De Dombal FT, Gremy F. eds. New York: North-Holland Publ Comp; 1976: 399-411.
26 Miller MCI, Westphal MC, Reigart JR, Barner C. Medical Diagnostic Models: A Bibliography. Ann Arbor, Mich: University Microfilms International; 1977
27 Good IJ. Good Thinking: The Foundations of Probability and Its Applications. Minneapolis, MN: Univ of Minnesota Press; 1983
28 Miller RA. Personal communication. 1987
29 Lawrence L. Detailed diagnosis and surgical procedures for patients discharged from short-stay hospitals. Vital & Helath Statistics 1986; 13.
30 Health Care Financing Administration. International Classification of Diseases (Ninth Revision) – Clinical Modification. Washington, D. C: U.S. Department of Health and Human Services, Public Health Service; 1980
31 Johnson RW. Independence and Bayesian updating methods. Artif Intell 1986; 29: 217-22.
32 Shwe MA, Middleton B, Heckerman DE, Henrion M, Horvitz EJ, Lehmann HP, Cooper GF. Probabilistic Diagnosis Using a Reformulation of the INTERNIST-1/QMR Knowledge Base – Part 1: The Probabilistic Model and Inference Algorithms. Knowledge Systems Laboratory Memo no. KSL-90-09. Stanford, CA: Stanford University; 1990
33 Dongarra JJ. Performance of Various Computers Using Standard Linear Equations Software in a FORTRAN Environment. Mathematics and Computer Science Division Technical Memorandum no. 23. Ar-gonne IL: Argonne National Laboratory; 1988
34 Cooper GF. The computational complexity of probabilistic inference using Bayesian belief networks. Artif Intell 1990; 42: 393-405.
35 Henrion M. An introduction to algorithms for inference in belief nets. In: Machine Intelligence and Pattern Recognition: Uncertainty in Artificial Intelligence 5. Henrion M, Shachter R, Kanal LN, Lemmer JF. eds. Amsterdam: North-Holland Publ Comp; 1990: 129-38.
36 Fung R, Chang KC. Weighting and integrating evidence for stochastic simulation in Bayesian networks. In: Machine Intelligence and Pattern Recognition: Uncertainty in Artificial Intelligence 5. Henrion M, Shachter R, Kanal LN, Lemmer JF. eds. Amsterdam: North-Holland Publ Comp; 1990: 209-20.
37 Shachter RD, Peot M. Simulation approaches to general probabilistic inference on belief networks. In: Machine Intelligence and Pattern Recognition: Uncertainty in Artificial Intelligence 5. Henrion M, Shachter R, Kanal LN, Lemmer JF. eds. Amsterdam: North-Holland Publ Comp; 1990: 221-31.
38 Rubinstein RY. Simulation and the Monte Carlo Method. New York: John Wiley & Sons; 1981
39 Geweke J. Bayesian inference in econometric models using Monte Carlo integration. Econometrica 1989; 57: 1317-39.
40 Heckerman DE, Miller RA. Towards a better understanding of the INTERNIST-1 knowledge base. In: MEDINFO 86. Salamon R, Blum B, Jørgensen M. eds. Amsterdam: North-Holland Publ Comp; 1986: 22-6.
41 Olesen KG, Kjaerulff U, Jensen E, Jensen FV, Falck B, Andreassen S, Andersen SK. A MUNIN networkfor the median nerve - a case study on loops. Appl Artif Intell 1989; 03: 384-404.
42 Lauritzen SL, Spiegelhalter DJ. Local computations with probabilities on graphical structures and their applications to expert systems (with discussion). J Royal Statist Soc 1988; 50: 157-224.