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

A computer program for non-parametric receiver operating characteristic analysis

S Vida1

  • 1Department of Psychiatry, McGill University, Montreal, Quebec, Canada.

Computer Methods and Programs in Biomedicine
|June 1, 1993
PubMed
Summary

This study introduces a computer program for Receiver Operating Characteristic (ROC) analysis, calculating the area under the ROC curve (AUC) without assuming normal distribution. This enhances diagnostic accuracy assessment for various medical tests.

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

Effects of added fermentable carbohydrates in the diet on intestinal proinflammatory cytokine-specific mRNA content in weaning piglets.

Journal of animal science·2006
Same author

Review of outcome measurement instruments in Alzheimer's disease drug trials: psychometric properties of behavior and mood scales.

Journal of geriatric psychiatry and neurology·2000
Same author

Vitamin B12 and folate levels and lithium administration in patients with affective disorders.

Biological psychiatry·1999
Same author

Accurate confidence intervals for measures of test performance.

Psychiatry research·1997
Same author

Prevalence of depression in Alzheimer's disease and validity of Research Diagnostic Criteria.

Journal of geriatric psychiatry and neurology·1994
Same author

An update on elder abuse and neglect.

Canadian journal of psychiatry. Revue canadienne de psychiatrie·1994

Area of Science:

  • Medical Diagnostics
  • Biostatistics

Background:

  • Sensitivity and specificity are crucial for test performance evaluation.
  • Quantifying diagnostic accuracy is challenging for tests with numerical scores due to inverse relationships between sensitivity and specificity.
  • Receiver Operating Characteristic (ROC) analysis visualizes test performance across all cutoff scores.

Purpose of the Study:

  • To present a computer program for non-parametric ROC analysis.
  • To overcome limitations of parametric methods that assume normal distribution of test scores.
  • To provide a tool for calculating and comparing diagnostic accuracy (AUC) of tests.

Main Methods:

  • Developed a computer program utilizing non-parametric formulations for ROC analysis.
  • Calculated sensitivity and specificity at all possible cutoff scores.

Related Experiment Videos

  • Plotted ROC curves and computed the area under the curve (AUC), standard error, and confidence limits.

    • Main Results:

    • The program effectively calculates AUC and its statistical measures using non-parametric methods.
    • ROC curves can be generated and AUC compared for independent and correlated samples.
    • The software does not require the assumption of normal distribution for test scores.

    Conclusions:

    • The developed program offers a flexible and robust tool for assessing diagnostic accuracy.
    • Non-parametric ROC analysis is valuable when test score distributions are unknown or non-Gaussian.
    • This approach facilitates reliable comparison of diagnostic tests.