This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Constructing Bayesian Networks for Medical Diagnosis from Incomplete and Partially Correct Statistics
July/August 2000 (vol. 12 no. 4)
pp. 509-516

Abstract—The paper discusses several knowledge engineering techniques for the construction of Bayesian networks for medical diagnostics when the available numerical probabilistic information is incomplete or partially correct. This situation occurs often when epidemiological studies publish only indirect statistics and when significant unmodeled conditional dependence exists in the problem domain. While nothing can replace precise and complete probabilistic information, still a useful diagnostic system can be built with imperfect data by introducing domain-dependent constraints. We propose a solution to the problem of determining the combined influences of several diseases on a single test result from specificity and sensitivity data for individual diseases. We also demonstrate two techniques for dealing with unmodeled conditional dependencies in a diagnostic network. These techniques are discussed in the context of an effort to design a portable device for cardiac diagnosis and monitoring from multimodal signals.

[1] G.A. Diamond and J.S. Forrester, “Analysis of Probability As An Aid in the Clinical Diagnosis of Coronary-Artery Disease,” The New England J. Medicine, vol. 300, pp. 1,350–1,359, 1979.
[2] M.J. Druzdzel and M. Henrion, “Efficient Reasoning in Qualitative Probabilistic Networks,” Proc. Eleventh Nat'l Conf. Artificial Intelligence (AAAI-93), pp. 548–553, 1993.
[3] D. Heckerman, “A Tractable Inference Algorithm for Diagnosing Multiple Diseases,” Proc. Fifth Conf. Uncertainty in Artificial Intelligence, M. Henrion, et al., eds., vol. 5, pp. 163–171, North-Holland: Elsevier Science Publishers, 1990.
[4] D. Heckerman and J. Breese, “A New Look At Causal Independence,” Proc. 10th Conf. Uncertainty in Artificial Intelligence, pp. 286–292, San Mateo, CA: Morgan Kaufmann, 1994.
[5] J.D.F Habbema, J. Hilden, and B. Bjerregaard, “The Measurement of Performance in Probabilistic Diagnosis: I. The Problem, Descriptive Tools, and Measures Based on Classification Matrices,” Methods of Information in Medicine, vol. 17, pp. 217–226, 1978.
[6] J.D.F. Habbema, J. Hilden, and B. Bjerregaard, “The Measurement of Performance in Probabilistic Diagnosis: II. Trustworthiness of the Exact Values of the Diagnostic Probabilities,” Methods of Information in Medicine, vol. 17, pp. 227–237, 1978.
[7] Netica Application User's Guide. NorsysSoftware Corp., Vancouver, BC. [http:/www.norsys.com.]
[8] Netica API Programmer's Library. NorsysSoftware Corp., Vancouver, BC. [http:/www.norsys.com.]
[9] J. Pearl, Probabilistic Reasoning in Intelligent Systems. San Mateo, Calif.: Morgan Kaufman, 1988.
[10] I. Rish and R. Dechter, “On the Impact of Causal Independence,” Proc. Stanford Spring Symp. Interactive and Mixed-Initiative Decision-Theoretic Systems, Mar. 1998.
[11] E.H. Shortliffe, “The Adolescence of AI n Medicine: Will the Field Come of Age n the '90s?” Artificial Intelligence in Medicine, vol. 5, pp. 93–106, NorthHolland: Elsevier Science Publishers, 1993.
[12] D.J. Spiegelhalter, “Probabilistic Expert Systems in Medicine,” Statistical Science, vol. 2, no.1, pp. 3–44, 1987.
[13] S. Srinivas, “A Generalization of the Noisy-OR Model,” Proc. Ninth Conf. Uncertainty in Artificial Intelligence, pp. 208–218, San Mateo, CA: Morgan Kaufmann, 1993.
[14] M.A. Shwe, B. Middleton, D.E. Heckerman, M. Henrion, E.J. Horvitz, H.P. Lehmann, and G.F. Cooper, “Probabilistic Diagnosis Using A Reformulation of the INTERNIST-1/QMR Knowledge Base,” Methods of Information in Medicine, vol. 30, pp. 241–255, 1991.
[15] H.R. Warner, D.K. Sorenson, and O. Bouhaddou, Knowledge Eng., in Health Informatics. Berlin, New York: Springer Verlag, 1997.
[16] M.P. Wellman, “Fundamental Concepts of Qualitative Probabilistic Networks,” Artificial Intelligence, vol. 44, pp. 257-304, 1990.
[17] H. Yu, P.J. Haug, M.J. Lincoln, C.W. Turner, and H.R. Warner, “Clustered Knowledge Representation: Increasing the Reliability of Computerized Expert Systems,” Proc. 12th Symp. Computer Applications in Medical Care, 1988.

Index Terms:
Bayesian networks, medical diagnosis, combining risk factors, leaky noisy-OR nodes.
Citation:
Daniel Nikovski, "Constructing Bayesian Networks for Medical Diagnosis from Incomplete and Partially Correct Statistics," IEEE Transactions on Knowledge and Data Engineering, vol. 12, no. 4, pp. 509-516, July-Aug. 2000, doi:10.1109/69.868904
Usage of this product signifies your acceptance of the Terms of Use.