Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Certainty factor theory and its implementation in a medical expert system shell.

Q Dan1, J Dudeck

  • 1Institute for Medical Informatics, University Giessen, Germany.

Medical Informatics = Medecine Et Informatique
|April 1, 1992
PubMed
Summary

This study critiques the probabilistic interpretations of the MYCIN model's certainty factors, highlighting issues with belief and disbelief measures. It also examines evidence combination schemes and proposes modifications for expert systems.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

[Changes of myocardial calcium currents in rats with myocardial injury induced by running exercise during acute hypoxia].

Nan fang yi ke da xue xue bao = Journal of Southern Medical University·2022
Same author

[Value of fragmented QRS wave in evaluating the prognosis of patients with coronary artery disease].

Zhonghua yi xue za zhi·2017
Same author

Dynasore enhances the formation of mitochondrial antiviral signalling aggregates and endocytosis-independent NF-κB activation.

British journal of pharmacology·2015
Same author

The Giessen Tumor Documentation System (GTDS)--review and perspectives.

Methods of information in medicine·2006
Same author

Representation of practice guidelines with XML--modeling with XML schema.

Methods of information in medicine·2002
Same author

Informed consent for cancer registration.

The Lancet. Oncology·2002

Area of Science:

  • Artificial Intelligence
  • Medical Informatics
  • Expert Systems

Background:

  • The MYCIN model is a foundational expert system for handling uncertainty in medical diagnosis.
  • Probabilistic interpretations of certainty factors have been proposed but face challenges.
  • Existing evidence combination schemes in expert systems have limitations.

Purpose of the Study:

  • To critically evaluate the probabilistic interpretations of the MYCIN model's certainty factors.
  • To analyze the validity of arguments regarding evidence updating and combination in expert systems.
  • To propose improvements for uncertainty handling in HIS-oriented expert systems.

Main Methods:

  • Review and critique of existing literature on MYCIN model interpretations (Adams, Shortliffe, Heckerman, Horvitz).

Related Experiment Videos

  • Analysis of probabilistic interpretations and their implications for evidence updating.
  • Examination of the isomorphic mapping between likelihood ratio and EMYCIN's evidence combination.
  • Description of a modified certainty factor mechanism for an expert system shell.

    • Main Results:

    • Identified inappropriate probabilistic interpretations of belief and disbelief measures in the MYCIN model.
    • Demonstrated that Heckerman's argument about non-commutative evidence-updating is incorrect.
    • Highlighted significant restrictions in the applicability of likelihood ratio evidence combination schemes.
    • Presented a modified certainty factor mechanism for practical implementation.

    Conclusions:

    • The probabilistic interpretations of MYCIN's certainty factors, as proposed by Shortliffe and Heckerman, are problematic.
    • Certain evidence combination schemes used in expert systems are highly restrictive.
    • Modifications to the certainty factor mechanism can enhance the performance of HIS-oriented expert systems.