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

Semi-automatic learning of simple diagnostic scores utilizing complexity measures.

Martin Atzmueller1, Joachim Baumeister, Frank Puppe

  • 1Department of Computer Science, University of Würzburg, Am Hubland, 97074 Würzburg, Germany. atzmueller@informatik.uni-wuerzburg.de

Artificial Intelligence in Medicine
|October 26, 2005
PubMed
Summary

Semi-automatic learning methods, like diagnostic scores, help medical experts build understandable knowledge systems. New complexity measures allow balancing accuracy and simplicity for better diagnostic support.

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

Gastrointestinal endoscopic image style transfer using EndoStyle to improve artificial intelligence prediction models.

NPJ digital medicine·2026
Same author

Artificial intelligence assisted colorectal lesion detection in private practices a randomized controlled study.

NPJ digital medicine·2026
Same author

Correction: A complete benchmark for polyp detection, segmentation and classification in colonoscopy images.

Frontiers in oncology·2026
Same author

Enhancing transformer-based architectures with geometric deep learning for colonoscopic polyp size classification using transfer learning.

Artificial intelligence in medicine·2025
Same author

Mining exceptional social behavior on attributed interaction networks.

Machine learning·2025
Same author

Transcription of Handwritten Forms for Medical Study Documentation.

Studies in health technology and informatics·2025

Area of Science:

  • Medical Informatics
  • Artificial Intelligence in Medicine

Background:

  • Knowledge acquisition in complex medical domains is challenging and costly.
  • Domain specialists require understandable and interpretable AI models, often favoring simpler ones over complex ones.

Purpose of the Study:

  • To introduce diagnostic scores as a method for representing simple diagnostic knowledge.
  • To present an inductive learning method for diagnostic scores that can incorporate background knowledge.
  • To develop complexity measures for assessing learned diagnostic scores.

Main Methods:

  • Proposing diagnostic scores for knowledge representation.
  • Developing an inductive learning algorithm for diagnostic scores.
  • Introducing complexity measures to evaluate learned scores.

Related Experiment Videos

  • Refining the learning process with background knowledge.

    • Main Results:

    • Demonstrated the effectiveness of the diagnostic score approach using the SonoConsult system's case base.
    • Showcased how users can adjust the trade-off between accuracy and complexity.
    • Validated the utility of the proposed complexity measures.

    Conclusions:

    • Semi-automatic learning methods, particularly diagnostic scores, efficiently support domain specialists in building diagnostic knowledge systems.
    • The developed complexity measures provide an intuitive way to assess learned knowledge patterns.
    • This approach aids in creating interpretable and accurate medical knowledge systems.