Model Selection for Diagnosis and Treatment Using Temporal Influence Diagrams

  • Gregory M. Provan
Conference paper
Part of the Lecture Notes in Statistics book series (LNS, volume 89)

Abstract

This paper describes the model-selection process for a temporal influence diagram formalization of the diagnosis and treatment of acute abdominal pain. The temporal influence diagram explicitly models the temporal aspects of the diagnosis/treatment process as the findings (signs and symptoms) change over time. Given the complexity of this temporal modeling process, there is uncertainty in selecting (1) the model for the initial time interval (from which the process evolves over time); and (2) the stochastic process describing the temporal evolution of the system. Uncertainty in selecting the initial Bayesian network model is addressed by sensitivity analysis. The choice of most appropriate stochastic process to model the temporal evolution of the system is also discussed briefly.

Keywords

Expense Perforation Appendicitis Gastroenteritis Salpingitis 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    J. Beck and S. Pauker. The Markov Process in Medical Prognosis. Medical Decision Making, 3: 419–458, 1983.CrossRefGoogle Scholar
  2. [2]
    C. Berzuini, R. Bellazi, and S. Quaglini. Temporal Reasoning with Probabilities. In Proc. Conf. Uncertainty in Artificial Intelligence, pages 14–21, 1989.Google Scholar
  3. [3]
    J.R. Clarke. A Concise Model for the Management of Possible Appendicitis. Medical Decision Making, 4: 331–338, 1984.CrossRefGoogle Scholar
  4. [4]
    P. Dagum, R. Shachter, and L. Fagan. Modeling Time in Belief Networks. Technical Report STAN- KSL-91-49, Stanford University, Knowledge Systems Laboratory, November 1991.Google Scholar
  5. [5]
    F.T. de Dombal, D. J. Leaper, J.R. Staniland, A.P. McCann, and J.C. Horrocks. Computer-aided Diagnosis of Acute Abdominal Pain. British Medical Journal, 2: 9–13, 1972.CrossRefGoogle Scholar
  6. [6]
    T. Dean and K. Kanazawa. A Model for Reasoning about Persistence and Causation. Computational Intelligence, 5 (3): 142–150, 1989.CrossRefGoogle Scholar
  7. [7]
    R.A. Howard and J.E. Matheson, Influence diagrams. In R. Howard and J. Matheson, editors, The Principles and Applications of Decision Analysis, pages 720–762. Strategic Decisions Group, CA, 1981.Google Scholar
  8. [8]
    H.I. Pass and J.D. Hardy. The appendix. In Hardy’s Textbook of Surgery, pages 574–581. J.B. LippincottCo., 2nd edition, Philadelphia, 1988.Google Scholar
  9. [9]
    J. Pearl. Probabilistic Reasoning in Intelligent Systems. Morgan Kaufmann, 1988.Google Scholar
  10. [10]
    G. M. Provan and D. Poole. The Utility of Consistency-Based Diagnostic Techniques. In Proc. Conf on Principles of Know. Representation and Reasoning, pages 461–472, 1991.Google Scholar
  11. [11]
    G.M. Provan. Tradeoffs in Knowledge-Based Construction of Probabilistic Models. 1993, submitted.Google Scholar
  12. [12]
    G.M. Provan and J.R. Clarke. Dynamic Network Construction and Updating Techniques for the Diagnosis of Acute Abdominal Pain. IEEE Transactions on Pattern Analysis and Machine Intelligence, 15: 299–306, March 1993.CrossRefGoogle Scholar
  13. [13]
    S. Schwartz, J. Baron, and J. Clarke. A Causal Bayesian Model for the Diagnosis of Appendicitis. In L. Kanal and Lemmer J., editors, Proc. Conf Uncertainty in Artificial Intelligence, pages 229–236, 1986.Google Scholar
  14. [14]
    S. Srinivas and J. Breese. IDEAL: A Software Package for Analysis of Influence Diagrams. In Proc. Conf. Uncertainty in Artificial Intelligence, pages 212–219, 1990.Google Scholar

Copyright information

© Springer-Verlag New York, Inc. 1994

Authors and Affiliations

  • Gregory M. Provan
    • 1
  1. 1.Computer and Information Science DepartmentUniversity of PennsylvaniaPhiladelphiaUSA

Personalised recommendations