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

A model for single and multiple knowledge based networks.

G Bologna1

  • 1Swiss Institute of Bioinformatics, Rue Michel Servet 1, 1211 Geneva, Switzerland. guido.bologna@isb-sib.ch

Artificial Intelligence in Medicine
|August 2, 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

Corrigendum to "Circulating Cancer Stem Cell-Derived Extracellular Vesicles as a Novel Biomarker for Clinical Outcome Evaluation".

Journal of oncology·2020
Same author

Circulating Cancer Stem Cell-Derived Extracellular Vesicles as a Novel Biomarker for Clinical Outcome Evaluation.

Journal of oncology·2019
Same author

Cryopreservation effects on Wharton's Jelly Stem Cells proteome.

Stem cell reviews and reports·2014
Same author

Intracranial haemorrhage in patients on antithrombotics: clinical presentation and determinants of outcome in a prospective multicentric study in Italian emergency departments.

Cerebrovascular diseases (Basel, Switzerland)·2006
Same author

A study on rule extraction from several combined neural networks.

International journal of neural systems·2001
Same author

Shortage of circulating naive CD8(+) T cells provides new insights on immunodeficiency in aging.

Blood·2000
Same journal

Real-time EEG-based epileptic seizure prediction using artificial intelligence: A systematic review.

Artificial intelligence in medicine·2026
Same journal

R-peak detection and ECG data compression scheme based on empirical mode decomposition and wavelet transform.

Artificial intelligence in medicine·2026
Same journal

CastNet: A three-channel EEG-based deep learning model for cross-subject depression detection.

Artificial intelligence in medicine·2026
Same journal

State-of-the-art TinyML approaches for colorectal cancer detection: Current advances, challenges, and future directions.

Artificial intelligence in medicine·2026
Same journal

JRadiEvo: A Japanese radiology report generation model enhanced by evolutionary optimization of model merging.

Artificial intelligence in medicine·2026
Same journal

Causally-informed deep learning towards explainable and generalizable outcome prediction in critical care.

Artificial intelligence in medicine·2026
See all related articles

This study introduces a method for extracting understandable rules from neural networks, improving their interpretability. The Discretized Interpretable Multi-Layer Perceptron (DIMLP) model shows comparable accuracy to standard neural networks and outperforms other methods in medical diagnosis tasks.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Biology

Background:

  • Neural networks often function as black boxes, hindering the explanation of their decisions.
  • Existing research on rule extraction primarily focuses on single neural networks, with limited investigation into combined networks.

Purpose of the Study:

  • To develop a method for translating symbolic rules into a Discretized Interpretable Multi-Layer Perceptron (DIMLP) model.
  • To enable rule extraction from single and multiple combined neural networks.
  • To enhance the interpretability of neural network responses.

Main Methods:

  • Characterizing discriminant hyperplane frontiers for rule extraction.
  • Translating symbolic rules into the DIMLP model.

Related Experiment Videos

  • Applying single DIMLP networks and ensembles to medical diagnosis and prognosis datasets.

    • Main Results:

    • Unordered rules extracted in polynomial time with 100% matching accuracy on training data.
    • DIMLP models demonstrated comparable accuracy to standard Multi-Layer Perceptrons (MLP) across 17 medical datasets.
    • DIMLP networks significantly outperformed the CN2 algorithm on eight problems, particularly in non-Hodgkin lymphoma diagnosis (96.1% accuracy).

    Conclusions:

    • The DIMLP approach effectively extracts symbolic rules from neural networks, enhancing model interpretability.
    • DIMLP networks offer a powerful and accurate alternative for medical diagnosis and prognosis tasks.
    • Ensembles of DIMLP networks show superior performance in complex classification problems like lymphoma diagnosis.