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

Case-based explanation of non-case-based learning methods.

R Caruana1, H Kangarloo, J D Dionisio

  • 1Department of Radiological Sciences, University of California, Los Angeles, USA.

Proceedings. AMIA Symposium
|November 24, 1999
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

Measurement of serum acetylsalicylic acid in a porcine model of aspirin overdose.

Veterinary and human toxicology·1996
Same author

Augmentation of hypoxic pulmonary vasoconstriction by N(G)-L-methyl-arginine in a rabbit model of thermal injury.

The American journal of emergency medicine·1996
Same author

Counseling to reduce road injuries and deaths.

Canadian family physician Medecin de famille canadien·1996
Same author

Pacemaker patient-triggered event recording: accuracy, utility, and cost for the pacemaker follow-up clinic.

Pacing and clinical electrophysiology : PACE·1996
Same author

Acute lung injury fibroblast migration and invasion of a fibrin matrix is mediated by CD44.

The Journal of clinical investigation·1996
Same author

Unusual features of osteoarticular tuberculosis in children.

The Journal of pediatrics·1996
Same journal

Progressive display of very high resolution images using wavelets.

Proceedings. AMIA Symposium·2002
Same journal

The Chronus II temporal database mediator.

Proceedings. AMIA Symposium·2002
Same journal

Gene expression levels in different stages of progression in oral squamous cell carcinoma.

Proceedings. AMIA Symposium·2002
Same journal

An assessment of the visibility of MeSH-indexed medical web catalogs through search engines.

Proceedings. AMIA Symposium·2002
Same journal

Filtering for medical news items using a machine learning approach.

Proceedings. AMIA Symposium·2002
Same journal

Enriching the structure of the UMLS semantic network.

Proceedings. AMIA Symposium·2002
See all related articles

We developed a method to explain complex artificial intelligence (AI) models using similar cases from training data. This approach aids understanding AI predictions in fields like medicine.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Explainable AI

Background:

  • Non-case-based learning methods, such as artificial neural networks and decision trees, are powerful but often lack transparency.
  • Understanding the reasoning behind AI predictions is crucial, especially in high-stakes domains like healthcare.

Purpose of the Study:

  • To introduce a novel method for generating case-based explanations for non-case-based machine learning models.
  • To leverage trained models as distance metrics for identifying similar training cases.

Main Methods:

  • Utilizing the trained machine learning model itself as a distance metric.
  • Identifying training set cases most similar to the case requiring explanation based on the model's learned representation.
  • Applying this to artificial neural nets and decision trees.

Related Experiment Videos

Main Results:

  • Demonstrated the feasibility of generating case-based explanations for complex models.
  • The method effectively uses the model's internal structure to find relevant similar cases.
  • Successfully applied to artificial neural nets and decision trees.

Conclusions:

  • The proposed method provides interpretable, case-based explanations for non-case-based AI models.
  • This approach is particularly valuable in medical domains, enhancing trust and understanding of AI-driven predictions.
  • Facilitates the integration of complex AI into clinical practice by providing understandable justifications.