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

Accuracy curves: an alternative graphical representation of probability data.

R Detrano1

  • 1V. A. Medical Center, Long Beach, CA 90822.

Journal of Clinical Epidemiology
|January 1, 1989
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

Abdominal aortic diameter and vascular atherosclerosis: the Multi-Ethnic Study of Atherosclerosis.

European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery·2011
Same author

Validation of thoracic quantitative computed tomography as a method to measure bone mineral density.

Calcified tissue international·2004
Same author

Medicine based on scientific evidence versus medicine based on evangelical marketing.

The American journal of cardiology·2001
Same author

Active serum vitamin D levels are inversely correlated with coronary calcification.

Circulation·1997
Same author

Electron beam computed tomographic coronary calcium as a predictor of coronary events: comparison of two protocols.

Circulation·1997
Same author

Predictive value of electron beam computed tomography.

Circulation·1997

Receiver operating characteristic (ROC) curves assess model discrimination but not accuracy. A new accuracy curve complements ROC curves, revealing model reliability and probability calibration in medical applications.

Area of Science:

  • Medical statistics
  • Clinical model evaluation

Background:

  • Receiver operating characteristic (ROC) curves are standard for evaluating probability model discrimination in medicine.
  • However, ROC curves do not fully represent a model's predictive accuracy or reliability.

Purpose of the Study:

  • To introduce a supplementary graphical method to assess the accuracy of probability models.
  • To differentiate between model discrimination and calibration.

Main Methods:

  • Derivation of a novel accuracy curve.
  • Comparison of the accuracy curve with the traditional ROC curve.
  • Application to a clinical example.

Main Results:

  • The accuracy curve is coincident with the ROC curve for reliable models.

Related Experiment Videos

  • The accuracy curve lies above the ROC curve if probabilities are overestimated.
  • The accuracy curve lies below the ROC curve if probabilities are underestimated.

    • Conclusions:

    • The proposed accuracy curve provides a more comprehensive assessment of probability model performance.
    • This graphical tool enhances the evaluation of model reliability beyond discrimination alone.