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

Combining a neural network with case-based reasoning in a diagnostic system

E B Reategui1, J A Campbell, B F Leao

  • 1Department of Computer Science, University College London, UK.

Artificial Intelligence in Medicine
|January 1, 1997
PubMed
Summary
This summary is machine-generated.

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

Comment on: Bariatric surgery is expensive but improves co-morbidity: 5-year assessment of patients with obesity and type 2 diabetes.

The British journal of surgery·2021
Same author

Relationship between adverse childhood experiences and perceived discrimination in adulthood.

Journal of affective disorders·2020
Same author

Letters to the Editor.

Nutrition reviews·2019
Same author

The differential impact of adverse childhood experiences in the development of pre-diabetes in a longitudinal cohort of US adults.

Journal of diabetes and its complications·2018
Same author

Diverse approaches to the health economic evaluation of bariatric surgery: a comprehensive systematic review.

Obesity reviews : an official journal of the International Association for the Study of Obesity·2016
Same author

A five year follow-up of coloured children with kwashiorkor in Cape Town.

Journal of tropical pediatrics (London, England : 1955)·2014

This study introduces a novel hybrid diagnostic system combining neural networks (NN) and case-based reasoning (CBR). This integrated approach enhances diagnostic accuracy for complex cases like congenital heart diseases (CHD).

Area of Science:

  • Artificial Intelligence
  • Medical Diagnostics
  • Computational Medicine

Background:

  • Diagnostic systems often face challenges with complex cases.
  • Integrating symbolic reasoning with sub-symbolic methods can improve performance.
  • Neural networks (NN) and case-based reasoning (CBR) are powerful but have limitations individually.

Purpose of the Study:

  • To develop and evaluate a hybrid NN-CBR model for enhanced diagnostic systems.
  • To improve the accuracy and explainability of diagnostic reasoning.
  • To address limitations in both NN knowledge interpretation and CBR retrieval.

Main Methods:

  • A novel approach integrating NN for hypothesis generation and CBR for case retrieval.
  • NN knowledge is mapped to symbolic descriptors for credibility assessment and explanation generation.

Related Experiment Videos

  • The hybrid model was developed for congenital heart disease (CHD) diagnosis and tested on multiple datasets.

    • Main Results:

    • The hybrid NN-CBR system achieved high accuracy in diagnosing congenital heart diseases (CHD).
    • The system successfully addressed problems intractable for NN alone.
    • The approach offers solutions for NN knowledge interpretation and CBR indexing/retrieval challenges.

    Conclusions:

    • The hybrid NN-CBR model offers a significant advancement in diagnostic system capabilities.
    • This integration enhances problem-solving capacity and provides explainable reasoning.
    • The developed system demonstrates potential for improving medical diagnosis accuracy and reliability.