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

PGMC: a framework for probabilistic graphic model combination.

Chang An Jiang1, Tze-Yun Leong, Kim-Leng Poh

  • 1Medical Computing Laboratory, National University of Singapore.

AMIA ... Annual Symposium Proceedings. AMIA Symposium
|June 17, 2006
PubMed
Summary

This study introduces a novel framework for integrating multiple probabilistic graphical models, enhancing decision-making in biomedicine. The approach preserves key relationships, simplifying model extension and application in areas like heart disease diagnosis.

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

Medical Artificial Intelligence and Human Values.

The New England journal of medicine·2024
Same author

Artificial Intelligence in Medicine.

The New England journal of medicine·2023
Same author

Subdomain Adaptation With Manifolds Discrepancy Alignment.

IEEE transactions on cybernetics·2021
Same author

Modeling Multi-View Dependence in Bayesian Networks for Alzheimer's Disease Detection.

Studies in health technology and informatics·2019
Same author

Knowledge-Driven Interpretation of Multi-View Data in Medicine.

Studies in health technology and informatics·2018
Same author

A Deep Learning Approach to Neuroanatomical Characterisation of Alzheimer's Disease.

Studies in health technology and informatics·2018

Area of Science:

  • Biomedical informatics
  • Artificial intelligence in medicine
  • Probabilistic graphical models

Background:

  • Biomedical decision-making frequently requires updating existing models with new evidence.
  • Integrating multiple probabilistic graphical models (PGMs) is complex and time-consuming.
  • Existing methods may not effectively preserve conditional independence relationships crucial for model accuracy.

Purpose of the Study:

  • To propose a new framework for the effective and incremental integration of multiple PGMs.
  • To minimize the time and effort needed for customizing and extending existing models.
  • To preserve conditional independence relationships inherent in Bayesian networks and influence diagrams.

Main Methods:

  • A four-step algorithm systematically combines qualitative and quantitative aspects of different PGMs.

Related Experiment Videos

  • Three heuristic methods for target variable generation are described to reduce model complexity.
  • The framework focuses on preserving conditional independence relationships.

    • Main Results:

    • Preliminary results demonstrate the feasibility of the proposed integration framework.
    • The framework successfully preserves conditional independence relationships during model combination.
    • Case study in heart disease diagnosis shows potential for real-world application.

    Conclusions:

    • The proposed framework offers an effective approach for integrating multiple PGMs in biomedicine.
    • This method simplifies model customization and extension, saving time and effort.
    • The framework shows promise for improving diagnostic accuracy and decision support systems.