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 new parametric method based on S-distributions for computing receiver operating characteristic curves for

Albert Sorribas1, Jaume March, Javier Trujillano

  • 1Departament de Ciències Mèdiques Bàsiques, Universitat de Lleida, Av. Rovira Roure 44, 25198-Lleida, Spain. albert.sorribas@cmb.udl.es

Statistics in Medicine
|July 12, 2002
PubMed
Summary

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

Hybrid model based on fuzzy logic and classification trees for the prediction of mortality of critical trauma patients: Pre-hospital variables from the RETRAUCI registry.

BMC medical research methodology·2026
Same author

The genome sequence of the red-legged partridge, <i>Alectoris rufa</i> Linnaeus 1758.

F1000Research·2025
Same author

Development and validation of a diagnostic prediction model for heart failure-related pleural effusions: the BANCA score.

ERJ open research·2025
Same author

Methodological Review of Classification Trees for Risk Stratification: An Application Example in the Obesity Paradox.

Nutrients·2025
Same author

Factors related to in-hospital mortality in patients with refractory Status Epilepticus.

Journal of clinical neuroscience : official journal of the Neurosurgical Society of Australasia·2025
Same author

Multiscale Mathematical Modeling in Systems Biology: A Framework to Boost Plant Synthetic Biology.

Plants (Basel, Switzerland)·2025

This study introduces the S-distribution, a flexible statistical model, to accurately estimate Receiver Operating Characteristic (ROC) curves for diagnostic tests. The S-distribution method provides a parametric approach for constructing ROC curves from sample data, improving diagnostic test evaluation.

Area of Science:

  • Biostatistics
  • Medical Diagnostics
  • Statistical Modeling

Background:

  • Receiver Operating Characteristic (ROC) curves are essential for evaluating diagnostic test performance.
  • Estimating continuous ROC curves from sample data often results in step-line approximations due to unknown underlying distributions.
  • Existing methods for ROC curve estimation face challenges with unknown distributions and potential data transformation distortions.

Purpose of the Study:

  • To introduce the S-distribution as a flexible parametric model for constructing ROC curves.
  • To demonstrate the utility of the S-distribution in accurately modeling unknown distributions for diagnostic variables.
  • To enable straightforward computation of ROC curves and confidence bands using the S-distribution.

Main Methods:

Related Experiment Videos

  • The S-distribution, defined by a differential equation, was employed to model the cumulative distribution of diagnostic variables.
  • S-distributions were fitted to sample data from diseased and healthy populations.
  • ROC curves were computed as solutions to differential equations derived from the fitted S-distributions.

Main Results:

  • The S-distribution provides a flexible family of distributions capable of modeling unknown and classical statistical distributions.
  • Fitting S-distributions to sample data allowed for straightforward parametric ROC curve computation.
  • The S-distribution based method enables the calculation of pointwise confidence bands for ROC curves and their areas.

Conclusions:

  • The S-distribution offers a robust parametric approach for estimating ROC curves, particularly when underlying distributions are unknown.
  • This method overcomes limitations of empirical and binormal approaches by providing a flexible and accurate modeling framework.
  • The S-distribution based method is a valuable tool for enhancing the evaluation of diagnostic test performance.