Methods Inf Med 1992; 31(02): 90-105
DOI: 10.1055/s-0038-1634867
Original Article
Schattauer GmbH

Toward Normative Expert Systems: Part I The Pathfinder Project

D. E. Heckerman
1   Department of Computer Science, University of California, Los Angeles CA
,
E. J. Horvitz
2   Palo Alto Laboratory, Rockwell International Science Center, Palo Alto CA
,
B. N. Nathwani
3   Department of Pathology, University of Southern California, Los Angeles CA, USA
› Institutsangaben
Weitere Informationen

Publikationsverlauf

Publikationsdatum:
08. Februar 2018 (online)

Abstract:

Pathfinder is an expert system that assists surgical pathologists with the diagnosis of lymph-node diseases. The program is one of a growing number of normative expert systems that use probability and decision theory to acquire, represent, manipulate, and explain uncertain medical knowledge. In this article, we describe Pathfinder and our research in uncertain-reasoning paradigms that was stimulated by the development of the program. We discuss limitations with early decision-theoretic methods for reasoning under uncertainty and our initial attempts to use non-decision-theoretic methods. Then, we describe experimental and theoretical results that directed us to return to reasoning methods based in probability and decision theory.

 
  • REFERENCES

  • 1 Heckerman DE, Horvitz EJ, Nathwani BN. Update on the Pathfinder project. In: Proceedings of the Thirteenth Symposium on Computer Applications in Medical Care. Silver Spring: IEEE Computer Society Press; 1989: 203-7.
  • 2 Heckerman DE, Horvitz EJ, Nathwani BN. The Pathfinder Project. Technical Report KSL-90-08. Stanford: Computer Science Group, Section on Medical Informatics, Stanford University; 1990
  • 3 Heckerman DE, Horvitz EJ, Nathwani BN. Pathfinder Research Directions. Technical Report KSL-89-64. Stanford: Computer Science Group, Section on Medicial Informatics, Stanford University; 1985
  • 4 Nathwani BN, Heckerman DE, Horvitz EJ, Lincoln TL. Integrated expert systems and videodisc in surgical pathology: An overview. Human Pathology 1990; 21: 11-27.
  • 5 Rosai RJ, Ackerman LV. The pathology of tumors, part II: Diagnostic techniques. Cancer 1979; 29: 22-39.
  • 6 Levin HS. Operating room consultation by the pathologist. Urology Clinics of North America 1985; 12: 549-56.
  • 7 Byrne GE. Rappaport classification of non-Hodgkin’s lymphoma: Histologic features and clinical significance. Cancer Treatment Reports 1977; 61: 935-44.
  • 8 Coltman CA, Gams RA, Glick JH, Jenkin RD. Lymphoma. In: Hoogstraten B. ed. Cancer Research Impact of the Cooperative Groups. Masson Publ; 1980: 39-84.
  • 9 Jones SE, Butler JJ, Byrne GE, Coltman CA, Moon TE. Histopathologic review of lymphoma cases from the Southwest Oncology Group. Cancer 1977; 39: 1071-6.
  • 10 Kim H, Zelman RJ, Fox MA, Bennett JM, Berard CW, Butler JJ, Byrne GE, Dorfman RF, Hartsock RJ, Lukes RJ, Mann RB, Neiman RS, Rebuck JW, Sheehan WW, Variakojis D, Wilson JF, Rappaport H. Pathology panel for lymphoma clinical studies: A comprehensive analysis of cases accumulated since its inception. J Nat Cane Instit 1982; 68: 43-67.
  • 11 Bartlett FD. Thinking. New York: Basic Books; 1958
  • 12 Elstein AS, Loupe MJ, Erdman JG. An experimental study of medical diagnostic thinking. J Structural Learning 1971; 02: 45-53.
  • 13 Elstein AS. Clinical judgment: Psychological research and medical practice. Science 1976; 194: 696-700.
  • 14 Elstein AS, Shulman LS, Sprafka SA. Medical Problem Solving: An Analysis of Clinical Reasoning. Cambridge, MA: Harvard University Press; 1978
  • 15 Hacking I. The Emergence of Probability. New York: Cambridge University Press; 1975
  • 16 Bayes T. An essay towards solving a problem in the doctrine of chances. Biometrika 1958; 46: 293-8. (Reprint of original work of 1763).
  • 17 Howard RA. Decision analysis: Perspectives on inference, decision, and experimentation. Proceedings of the IEEE 1970; 58: 632-43.
  • 18 Pearl J. How to do with probabilities what people say you can’t. In: Proceedings of the Second Conference on Artificial Intelligence Applications. Silver Spring: IEEE Computer Press; 1985: 6-12.
  • 19 Spiegelhalter DJ. Probabilistic reasoning in predictive expert systems. In: Kanal LN, Lemmer JF. eds. Uncertainty in Artificial Intelligence. New York: North-Holland Publ Comp; 1986: 47-67.
  • 20 Von Neumann J, Morgenstern O. Theory of Games and Economic Behavior. Princeton: Princeton University Press; 1947
  • 21 Keeney RL, Raiffa H. Decisions with Multiple Objectives: Preferences and Value Tradeoffs. New York: Wiley and Sons; 1976
  • 22 McNeil BJ, Pauker SG, Sox HC, Tversky A. On the elicitation of preferences for alternative therapies. N Engl J Med 1982; 306: 1259-62.
  • 23 Howard RA. On making life and death decisions. In: Schwing RC, Albers Jr WA. eds. Societal Risk Assessment. New York: Plenum Publ; 1980: 89-113.
  • 24 Howard RA. Microrisks for medical decision analysis. Internat J Technol Assessment in Health Care 1989; 05: 357-70.
  • 25 Tribus M. Rational Descriptions, Decisions, and Designs. New York: Pergamon Press; 1969
  • 26 Ledley RS, Lusted LB. Reasoning foundations of medical diagnosis. Science 1959; 130: 9-21.
  • 27 Gorry GA, Barnett GO. Experience with a model of sequential diagnosis. Comput Biomed Res 1968; 01: 490-507.
  • 28 Howard RA. Value of information lotteries. IEEE Transactions of Systems Science and Cybernetics 1967; SSC-3 (01) 54-60.
  • 29 Howard RA. Decision Analysis, EES231 Course Notes. Stanford: Department of Eingineering-Economic Systems, Stanford University; 1985
  • 30 Warner HR, Toronto AF, Veasy LG, Stephenson R. A mathematical approach to medical diagnosis: Application to congenital heart disease. J Am Med Assoc 1961; 177: 177-83.
  • 31 De Dombal FT, Leaper DJ, Staniland JR, McCann AP, Horrocks JC. Computer-aided diagnosis of acute abdominal pain. Br Med J 1972; 02: 9-13.
  • 32 Gorry GA. Computer-assited clinical decision making. Meth Inform Med 1973; 12: 45-51.
  • 33 Shortliffe EH. Computer-based Medical Consultations: MYCIN. New York: North-Holland Publ Comp; 1976
  • 34 Fryback DG. Bayes’ theorem and conditional nonindependence of data in medical diagnosis. Comput Biomed Res 1978; 11: 423-34.
  • 35 Norusis MJ, Jacquez JA. Diagnosis 1: Symptom nonindependence in mathematical models for diagnosis. Comput Biomed Res 1975; 08: 156-72.
  • 36 Szolovits P. Artificial intelligence in medicine. In: Szolovits P. ed. Artificial Intelligence In Medicine. Boulder: Westview Press; 1982: 1-19.
  • 37 Davis R. Consultation, knowledge acquisition, and instruction. In: Szolovits P. ed. Artificial Intelligence In Medicine. Boulder: Westview Press; 1982: 57-78.
  • 38 Shortliffe EH, Buchanan BG. A model of inexact reasoning in medicine. Mathematical Biosciences 1975; 23: 351-79.
  • 39 Miller RA, Pople EP, Myers JD. INTER-NIST-1: An experimental computer-based diagnostic consultant for general internal medicine. N Engl J Med 1982; 307: 476-86.
  • 40 Miller RA, McNeil MA, Challinor SM, Masarie FE, Myers JD. The INTERNIST-1/Quick Medicial Reference project: Status report. West J Med 1986; 145: 816-22.
  • 41 Dempster AP. Upper and lower probabilities induced by a multivalued mapping. Ann Math Statist 1967; 38: 325-39.
  • 42 Shafer G. A Mathematical Theory of Evidence. Princeton: Princeton University Press; 1976
  • 43 Shafer G. Savage revisited. Statistical Science 1986; 01: 463-501.
  • 44 Barnett JA. Computational methods for a mathematical theory of evidence. In: Proceedings of the Seventh International Joint Conference on Artificial Intelligence; Vancouver: 1981: 868-75.
  • 45 Gordon J, Shortliffe EH. A method for managing evidential reasoning in a hierarchical hypothesis space. Artif Intell 1985; 26: 323-57.
  • 46 Shafer G. Probability judgment in artificial intelligence. In: Proceedings of the Workshop on Uncertainty and Probability in Artificial Intelligence. Mountain View: Association for Uncertainty in Artificial Intelligence; 1985: 91-8. Also in: Kanal L, Lemmer J, eds. Uncertainty in Artificial Intelligence. New York: North-Holland Publ Comp, 1986: 127-36.
  • 47 Hummel R, Landy M. Evidence as opinions of experts. In: Proceedings of the Second Workshop on Uncertainty in Artificial Intelligence. Mountain View: Association for Uncertainty in Artificial Intelligence; 1986: 135-43. Also in: Kanal L, lemmer J, eds. Uncertainty in Artificial Intelligence 1. New York: North-Holland Publ Comp, 1988: 43-53.
  • 48 Smets P. Constructing the pignistic probability function in a context of uncertainty. In: Proceedings of the Fifth Workshop on Uncertainty in Artificial Intelligence. Mountain View: Association for Uncertainty in Artificial Intelligence; 1989: 319-26. Also in: Henrion M, Shachter R, Kanal L, Lemmer J, eds. Uncertainty in Artificial Intelligence 5. New York: North-Holland Publ Comp, 1990: 29-39.
  • 49 Zadeh L. Is probability theory sufficient for dealing with uncertainty in AI: A negative view. In: Kanal LN, Lemmer JF. eds. Uncertainty in Artificial Intelligence. New York: North-Holland Publ Comp; 1986: 103-16.
  • 50 Heckerman DE. An empirical comparison of three inference methods. In: Proceedings of the Fourth Workshop on Uncertainty in Artificial Intelligence. Mountain View: Association for Uncertainty in Artificial Intelligence; 1988: 158-69. Also in: Shachter R, Levitt T, Kanal L, Lemmer J, eds. Uncertainty in Artificial Intelligence 4. New York: North-Holland Publ Comp, 1990: 283-302.
  • 51 Heckerman DE. Probabilistic interpretations for MYCIN’S certainty factors. In: Proceedings of the Workshop on Uncertainty and Probability in Artificial Intelligence. Mountain View: Association for Uncertainty in Artificial Intelligence; 1985: 9-20. Also in: Kanal L, Lemmer J, eds. Uncertainty in Artificial Intelligence. New York: North-Holland Publ Comp, 1986: 167-96.
  • 52 Grosof B. Evidential confirmation as transformed probability. In: Proceedings of the Workshop on Uncertainty and Probability in Artificial Intelligence. Mountain View: Association for Uncertainty in Artificial Intelligence; 1985: 185-92. Also in: Kanal L, Lemmer J, eds. Uncertainty in Artificial Intelligence. New York: North-Holland Publ Comp, 1986: 153-66.
  • 53 Heckerman DE, Miller RA. Towards a better understanding of the INTERNIST-1 knowledge base. In: Proceedings of Med-info 86; New York: North-Holland Publ Comp; 1986: 27-31.
  • 54 Henrion M. Towards efficient probabilistic inference in multiply connected belief networks. In: Oliver RM, Smith JQ. eds. Influence Diagrams, Belief Nets, and Decision Analysis. New York: Wiley and Sons; 1990. chapter 17.
  • 55 Heckerman DE. A tractable algorithm for diagnosing multiple diseases. In: Proceedings of the Fifth Workshop on Uncertainty in Artificial Intelligence. Mountain View: Association for Uncertainty in Artificial Intelligence; 1989: 174-81. Also in: Henrion M, Shachter R, Kanal L, Lemmer J, eds. Uncertainty in Artificial Intelligence 5. New York: North-Holland Publ Comp, 1990: 163-71.
  • 56 Shwe M, Middleton B, Heckerman D, Henrion M, Horitz E, Lehmann H, Cooper G. Probabilistic diagnosis using a reformulation of the INTERNIST-1/QMR knowledge base, part I: The probabilistic model and inference algorithms. Meth Inform Med 1991; 30: 241-50.
  • 57 Middleton B, Shwe M, Heckerman D, Henrion M, Horvitz E, Lehmann H, Cooper G. Probabilistic diagnosis using a reformulation of the INTERNIST-1/QRM knowledge base, part II: Evaluation of diagnostic performance. Meth Inform Med 1991; 30: 256-67.
  • 58 Johnson R. Independence and Bayesian updating methods. In: Proceedings of the Workshop on Uncertainty and Probability in Artificial Intelligence. Mountain View: Association for Uncertainty in Artificial Intelligence; 1985: 28-30. Also in: Kanal L, Lemmer J, eds. Uncertainty in Artificial Intelligence. New York: North-Holland Publ Comp, 1986: 197-201.
  • 59 Shachter RD, Heckerman DE. Thinking backward for knowledge acquisition. AI Magazine 1987; 08: 55-63.
  • 60 Tversky A, Kahneman D. Causal schemata in judgments under uncertainty. In: Kahneman D, Slovic P, Tversky A. eds. Judgement Under Uncertainty: Heuristics and Biases. New York: Cambridge University Press; 1982
  • 61 Horvitz EJ, Breese JS, Henrion M. Decision theory in expert systems and artificial intelligence. Internat J Approximate Reasoning 1988; 02: 247-302.
  • 62 Ramsey FP. Truth and probability. In: Braithwaite RB. ed. The Foundations of Mathematics and other Logical Essays. London: Humanities Press; 1931. Reprinted in Kyburg and Smokier 1964
  • 63 De Finetti B. La prevision: See lois lo-giques, ses sources subjectives. In: Annales de l’Institut Henri Poincaré 1937; 07: 1568 Translated in Kyburg and Smokier, 1964.
  • 64 Kyburg HE, Smokier HE. Studies in Subjective Probability. New York: Wiley and Sons; 1964
  • 65 Cox R. Probability, frequency and reasonable expectation. Am J Physics 1946; 14: 1-13.
  • 66 Savage LI. The Foundations of Statistics. New York: Dover; 1954
  • 67 Howard RA. Risk preference. In: Howard RA, Matheson JE. eds. Readings on the Principles and Applications of Decision Analysis, volume II. Menlo Park: Strategic Decisions Group; 1970: 629-63.
  • 68 Tversky A, Kahneman D. Judgment under uncertainty: Heuristics and biases. Science 1974; 185: 1124-31.
  • 69 Kahneman D, Slovic P, Tversky A. eds. Judgment Under Uncertainty: Heuristics and Biases. New York: Cambridge University Press; 1982
  • 70 Gorry GA, Kassirer JP, Essig A, Schwartz WB. Decision analysis as the basis for computer-aided management of acute renal failure. Am J Med 1973; 55: 473-84.
  • 71 Buchanan BG, Shortliffe EH. eds. Rule-Based Expert Systems: The MYCIN Experiments of the Stanford Heuristic Programming Project. Reading: Addison-Wesley; 1984
  • 72 Zadeh L. Personal communication. 1986.
  • 73 Heckerman D, Nathwani B. An evaluation of the diagnostic accuracy of Pathfinder. Comput Biomed Res 1992; 15: 56-79.
  • 74 Cooper GF. NESTOR: A Computer-based Medical Diagnostic Aid that Integrates Causal and Probabilistic Knowledge. PhD thesis, Stanford: Medical Computer Science Group, Stanford University; 1984. Report HPP-84-48.
  • 75 Spiegelhalter DJ, Knill-Jones RP. Statistical and knowledge-based approaches to clinical decision support systems, with an application in gastroenterology. J Royal Statist Soc 1984; 147: 35-77.
  • 76 Reggia JA, Perricone BT. Answer justification in medical decision support systems based on Bayesian classification. Comput Biology Med 1985; 15: 161-7.
  • 77 Ben-Bassat M, Carlson VK, Puri VK, Davenport MD, Schriver JA, Latif MM, Smith R, Lipnick EH, Weil MH. Pattern-based interactive diagnosis of multiple disorders: The MEDAS system. IEEE Transactions on Pattern Analysis and Machine Intelligence 1980; 02: 148-60.
  • 78 Andreassen S, Woldbye M, Falck B, Andersen SK. MUNIN: A causal probabilistic network for interpretation of electromyographic findings. In: Proceedings of the Tenth International Joint Conference on Artificial Intelligence. San Mateo: Morgan Kaufmann; 1987: 366-72.
  • 79 Nino-Murcia G, Shwe M. An expert system for diagnosis of sleep disorders. In: Chase M, Lydic R, O’Connor C. eds. Sleep Research, volume 20. Los Angeles: Brain Information Service; 1991: 433.
  • 80 Kanal L, Lemmer J. eds. Uncertainty in Artificial Intelligence. New York: North-Holland Publ Comp; 1986
  • 81 Lemmer J, Kanal L. eds. Uncertainty in Artificial Intelligence 2. New York: North-Holland Publ Comp; 1988
  • 82 Kanal L, Levitt T, Lemmer J. eds. Uncertainty in Artificial Intelligence 3. New York: North-Holland Publ Comp; 1989
  • 83 Shachter R, Levitt T, Kanal L, Lemmer J. eds. Uncertainty in Artificial Intelligence 4. New York: North-Holland Publ Comp; 1990
  • 84 Henrion M, Shachter R, Kanal L, Lemmer J. eds. Uncertainty in Artificial Intelligence 5. New York: North-Holland Publ Comp; 1990