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

Sampling strategies in a statistical approach to clinical classification

Y Yang1, C G Chute

  • 1Section of Medical Information Resources, Mayo Clinic/Foundation, Rochester, Minnesota 55905, USA.

Proceedings. Symposium on Computer Applications in Medical Care
|January 1, 1995
PubMed
Summary

This study optimizes sampling strategies for Expert Network (EexNet) in patient record classification. Efficient methods achieve high accuracy with reduced training data and computational cost.

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

The WHO ICD-11 Classification of Dermatological Diseases: a new comprehensive online skin disease taxonomy designed by and for dermatologists.

The British journal of dermatology·2021
Same author

Health Concept and Knowledge Management: Twenty-five Years of Evolution.

Yearbook of medical informatics·2016
Same author

Genetic variation among 82 pharmacogenes: The PGRNseq data from the eMERGE network.

Clinical pharmacology and therapeutics·2016
Same author

A genome-wide association study identifies variants in KCNIP4 associated with ACE inhibitor-induced cough.

The pharmacogenomics journal·2015
Same author

Harmonization of detailed clinical models with clinical study data standards.

Methods of information in medicine·2014
Same author

Design and anticipated outcomes of the eMERGE-PGx project: a multicenter pilot for preemptive pharmacogenomics in electronic health record systems.

Clinical pharmacology and therapeutics·2014

Area of Science:

  • Medical Informatics
  • Statistical Learning
  • Machine Learning

Background:

  • Patient record classification is crucial for healthcare.
  • Expert Network (EexNet) is a statistical learning system used for this task.
  • Large-scale applications require efficient computational costs and high accuracy.

Purpose of the Study:

  • To investigate sampling strategies for EexNet.
  • To optimize classification accuracy at an affordable computational cost.
  • To improve EexNet's performance in large-scale patient record classification.

Main Methods:

  • Analysis of EexNet learning curves based on training resources, set size, vocabulary, and category coverage.
  • Evaluation of a novel method combining different sampling strategies.

Related Experiment Videos

  • Testing with a large training corpus.

    • Main Results:

    • Nearly-optimal classification accuracy (average precision) achieved with a small training set.
    • Fast real-time response demonstrated.
    • Effective human-machine interaction capabilities.

    Conclusions:

    • Optimized sampling strategies significantly enhance EexNet performance.
    • High accuracy and computational efficiency are achievable for large-scale patient record classification.
    • The proposed method supports practical implementation in clinical settings.