Categorical and probabilistic reasoning in medical diagnosis

https://doi.org/10.1016/0004-3702(78)90014-0Get rights and content

Abstract

Medical decision making can be viewed along a spectrum, with categorical (or deterministic) reasoning at one extreme and probabilistic (or evidential) reasoning at the other. In this paper we examine the flowchart as the prototype of categorical reasoning and decision analysis as the prototype of probabilistic reasoning. Within this context we compare PIP, INTERNIST, CASNET, and MYCIN—four of the present programs which apply the techniques of artificial intelligence to medicine. Although these systems can exhibit impressive expert-like behavior, we believe that none of them is yet capable of truly expert reasoning. We suggest that a program which can demonstrate expertise in the area of medical consultation will have to use a judicious combination of categorical and probabilistic reasoning—the former to establish a sufficiently narrow context and the latter to make comparisons among hypotheses and eventually to recommend therapy.

References (37)

  • A.R. Feinstein

    Clinical Biostatistics XXXIX. The haze of Bayes, the aerial palaces of decision analysis, and the computerized Ouija board

    Clinical Pharmacology and Therapeutics

    (April 1977)
  • B.J. Flehinger et al.

    HEME: A self-Improving Computer Program for Diagnosis-Oriented Analysis of Hematologic Diseases

    IBM J. Res. Develop.

    (November 1975)
  • G.A. Gorry

    A System for Computer-Aided Diagnosis

    Project MAC, Massachusetts Institute of Technology, Technical Report TR-44

    (September 1967)
  • P.B. Miller

    Strategy Selection in Medical Diagnosis

    Project MAC, Massachusetts Institute of Technology, Technical Report TR-153

    (September 1975)
  • A. Newell et al.

    Human Problem Solving

    (1972)
  • C. Oleson

    INTERNIST: A Computer-Based Consultation, Computer Networking in the University: Success and Potential

  • S.G. Pauker et al.

    Toward the Simulation of Clinical Cognition: Taking a Present Illnes by Computer

    The American Journal of Medicine

    (June 1976)
  • S.G. Pauker et al.

    Therapeutic decision making: a cost-benefit analysis

    New Eng. J. Med.

    (1975)
  • Cited by (246)

    • Artificial intelligence and computational modeling

      2022, 3D Lung Models for Regenerating Lung Tissue
    • Artificial intelligence in medicine: What is it doing for us today?

      2019, Health Policy and Technology
      Citation Excerpt :

      Kulikowski and Weiss discussed the CASNET and EXPERT projects in the 1980s; the former was used in the context of glaucoma care, while the latter was established to help build models for reasoning in rheumatology and endocrinology [15]. As far back as 1959, an article in Science postulated how computers might fit into the process of patient diagnosis, acknowledging that “before computers can be used effectively … we need to know more about how the physician makes a medical diagnosis,”[16] with subsequent research aiming to elucidate this process in the context of computer operation [17]. In later years, research turned towards evaluating computer-aided diagnosis in comparison with the abilities of human physicians.

    • A fuzzy inference- fuzzy analytic hierarchy process-based clinical decision support system for diagnosis of heart diseases

      2018, Expert Systems with Applications
      Citation Excerpt :

      Two CDSSs were developed in two distinguished areas including deductive and probabilistic during 1950–1960 because it was required to install CDSS on computers at that time. However, a third method, with its specific identity consisting of distinctive features of primary methods, was proposed by Ledley and Luste (1959), and Szolovits and Pauker (1978) two decades later. The HEME program, which was used for hematologic diseases diagnosis, was one of primary systems that applied the third method.

    • Checking the lists: A systematic review of electronic checklist use in health care

      2017, Journal of Biomedical Informatics
      Citation Excerpt :

      However, beyond this function it also acts as a memory aid to minimize biases and augments working and prospective memory. Flowchart checklists are beneficial where a categorical judgment of the status of a task or system can be made with reasonable certainty [13]. Clinical practice guidelines and decision support for diagnoses are often developed in the form of a flowchart, either paper-based or in electronic clinical decision support systems (CDSS).

    View all citing articles on Scopus

    This research was supported by the Department of Health, Education, and Welfare (Public Health Service) under Grant Number 1 R01 MB 00107-03.

    ∗∗

    Dr. Pauker is the recipient of a Research Career Development Award (Number 1K04GM-00349-01) from the General Medical Sciences Institute, National Institutes of Health.

    View full text