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

ARTMAP-IC and medical diagnosis: Instance counting and inconsistent cases.

G A Carpenter1, N Markuzon

  • 1Center for Adaptive Systems and Department of Cognitive and Neural Systems, Boston University, 677 Beacon Street, Boston, MA 02215, USA.

Neural Networks : the Official Journal of the International Neural Network Society
|March 29, 2003
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

Automated analysis of crackles in patients with interstitial pulmonary fibrosis.

Pulmonary medicine·2011
Same author

Neural networks: introduction to the 1 December 1987 issue of Applied Optics.

Applied optics·2010
Same author

ART 2: self-organization of stable category recognition codes for analog input patterns.

Applied optics·2010
Same author

S-TREE: self-organizing trees for data clustering and online vector quantization.

Neural networks : the official journal of the International Neural Network Society·2001
Same author

Networks and groups within the genus Neisseria: analysis of argF, recA, rho, and 16S rRNA sequences from human Neisseria species.

Molecular biology and evolution·1999
Same author

Ventricular outflow tract reconstructions with cryopreserved cardiac valve homografts. A single surgeon's 10-year experience.

Annals of surgery·1996

The ARTMAP-IC neural network improves medical diagnosis by enhancing fuzzy ARTMAP with distributed prediction and instance counting. Its new match tracking algorithm excels with sparse data, outperforming other methods in accuracy and efficiency.

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Medical Informatics

Background:

  • Complex database prediction, particularly in medical diagnosis, requires advanced computational models.
  • Existing fuzzy ARTMAP systems lack robust prediction capabilities for sparse or inconsistent data.
  • Accurate and efficient predictive models are crucial for clinical decision support.

Purpose of the Study:

  • To introduce the ARTMAP-IC neural network, enhancing fuzzy ARTMAP with distributed prediction and category instance counting.
  • To develop a novel ARTMAP match tracking algorithm (MT-) for improved prediction accuracy with challenging datasets.
  • To evaluate the performance of ARTMAP-IC against established machine learning algorithms on medical databases.

Main Methods:

  • Implementation of ARTMAP-IC, integrating distributed prediction and instance counting into the fuzzy ARTMAP architecture.

Related Experiment Videos

  • Development and application of a new match tracking algorithm (MT-) to handle sparse and inconsistent data.
  • Comparative simulation analysis using four medical databases: Pima Indian diabetes, breast cancer, heart disease, and gall bladder removal.

    • Main Results:

    • ARTMAP-IC demonstrated predictive accuracy equal to or superior to logistic regression, K nearest neighbour (KNN), and other benchmark algorithms.
    • The new MT- algorithm approximated real-time network differential equations more effectively than MT+, compressing memory without performance degradation.
    • A voting strategy, combined with instance counting and distributed representations, enhanced prediction confidence estimates.

    Conclusions:

    • ARTMAP-IC offers a powerful and efficient solution for complex medical diagnosis and database prediction tasks.
    • The enhanced match tracking algorithm significantly improves ARTMAP's ability to handle real-world, imperfect data.
    • ARTMAP-IC's fast, stable, and scalable dynamics, coupled with confidence estimation, make it a promising tool for clinical applications.