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

An exponential model used for optimal threshold selection on ROC curves.

W L England1

  • 1Center for Health Systems Research and Analysis, University of Wisconsin-Madison.

Medical Decision Making : an International Journal of the Society for Medical Decision Making
|April 1, 1988
PubMed
Summary

A novel two-parameter exponential equation accurately models receiver operating characteristic (ROC) curves without distributional assumptions. This method simplifies area under the curve calculations and aids in determining optimal diagnostic thresholds.

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

Cost and volume trends in health care facility construction.

Health care financing review·1991
Same author

Strategies for screening for colorectal carcinoma.

Medical decision making : an international journal of the Society for Medical Decision Making·1989
Same author

Cost effectiveness in the detection of renal artery stenosis.

Journal of general internal medicine·1988
Same author

Surgery or angioplasty for cost-effective renal revascularization?

Medical decision making : an international journal of the Society for Medical Decision Making·1987
Same author

Setting health maintenance organization capitation rates for Medicaid in Wisconsin.

Health care financing review·1987
Same author

Immunization to prevent insulin-dependent diabetes mellitus? The economics of genetic screening and vaccination for diabetes.

Annals of internal medicine·1981

Area of Science:

  • Biostatistics
  • Medical Diagnostics
  • Machine Learning

Background:

  • Receiver operating characteristic (ROC) curve analysis is crucial for evaluating diagnostic test performance.
  • Existing ROC modeling methods often rely on specific distributional assumptions or complex calculations.
  • There is a need for flexible and computationally efficient ROC modeling techniques.

Purpose of the Study:

  • To introduce a new two-parameter exponential equation for ROC curve modeling.
  • To demonstrate the model's ability to calculate the area under the curve (AUC) as a function of its parameters.
  • To show the utility of the model's derivative in optimizing decision thresholds for diagnostic tests.

Main Methods:

  • Developed a two-parameter exponential equation to model ROC curves.

Related Experiment Videos

  • Performed computer simulations with 75 ROC curves to compare model fit with maximum likelihood estimation (MLE).
  • Applied the model to existing ROC curve data from scientific literature.

    • Main Results:

    • The proposed model demonstrated a fit comparable to the MLE method in simulations.
    • The model successfully fitted ROC curve data from the literature.
    • The model's equation directly relates the true-positive ratio to the false-positive ratio.
    • The model's first derivative was identified as useful for threshold optimization.

    Conclusions:

    • The two-parameter exponential equation offers a flexible and effective method for ROC curve modeling.
    • The model simplifies AUC calculation and provides a tool for optimizing diagnostic decision thresholds, particularly for sequential testing.
    • This approach avoids restrictive distributional assumptions, enhancing its applicability across various diagnostic scenarios.