AIME 89 pp 125-134 | Cite as

Therapy Planning by Combining Ai and Decision Theoretic Techniques

  • Silvana Quaglini
  • Carlo Berzuini
  • Riccardo Bellazzi
  • Mario Stefanelli
  • Giovanni Barosi
Part of the Lecture Notes in Medical Informatics book series (LNMED, volume 38)

Abstract

There is an increasing interest in therapy planning systems which combine artificial intelligence (AI) and decision theoretic techniques. Medical problems often require both categorical and probabilistic reasoning, but few systems try to combine them in general and homogeneous frameworks. This work presents the therapy advisor module of an expert system designed for managing anemic patients. This module allows simplest therapeutic problems to be solved by a frame-and-rule based expert system, and more complex problems, i.e. decisions that must be taken in presence of trade-offs, to be tackled by decision-theoretic techniques. Influence diagram formalism has been chosen to model the decision problem and methods for augmenting influence diagrams in order to describe temporal processes have been investigated. Decision analysis is an integrated part of the whole system, so that AI techniques provide help to the domain expert in building and debugging its own decision model.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Lanzola, G., Stefanelli, M., Barosi, G. and Magnani, L. A knowledge systems architecture for diagnostic reasoning, in this volume, 1989Google Scholar
  2. 2.
    Fikes, R.E. and Keheler, T.R. The role of frame-basedrepresentation in reasoning. Comm. of the ACM, 28, 904–920, 1985.CrossRefGoogle Scholar
  3. 3.
    Steele, G.L. “COMMON LISP: The Language” D.E.C., 1984.Google Scholar
  4. 4.
    Pearl, J. “Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference”. Morgan Kaufmann, San Mateo, California, 1988.Google Scholar
  5. 5.
    Shachter, R.D. Evaluating Influence Diagrams, Operation Research 34,871,1986CrossRefGoogle Scholar
  6. 6.
    Lauritzen, S.L.and Spiegelhalter, D. J. Local computation with probabilities on graphical structures and their application to expert systems, Journal of the Royal Statistic Society, 50,No. 2,1988.Google Scholar
  7. 7.
    Spiegelhalter, D. Probabilistic reasoning in predictive expert systems, Uncertainty in Artificial Intelligence, 47,1986.Google Scholar
  8. 8.
    Pearl, J. Fusion, propagation and structuring in belief networks, Artificial Intelligence 29, 241, 1986.CrossRefGoogle Scholar
  9. 9.
    Cooper, F. A Method for using Belief Networks as Influence Diagrams, Proceedings AAAI 1988.Google Scholar
  10. 10.
    Chankong, V., Haimes, Y. “Multiobjective Decision Making (Theory and Methodology)”, North Holland, 1983.Google Scholar
  11. 11.
    Ripley, B.D. Stochastic Simulation. John Wiley & Sons, 1987CrossRefGoogle Scholar
  12. 12.
    Henrion, M. Propagating uncertainty by logic sampling in Bayes’ networks, Proceed. Second Workshop on Uncertainty in Artificial Intelligence, Philadelphia, 1986.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1989

Authors and Affiliations

  • Silvana Quaglini
    • 1
  • Carlo Berzuini
    • 1
  • Riccardo Bellazzi
    • 1
  • Mario Stefanelli
    • 1
  • Giovanni Barosi
    • 2
  1. 1.Dipartimento di Informatica e SistemisticaPaviaItaly
  2. 2.Dipartimento di Medicina Interna e Terapia MedicaUniversity of PaviaPaviaItaly

Personalised recommendations