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

Continuous versus categorical data for ROC analysis: some quantitative considerations.

R F Wagner1, S V Beiden, C E Metz

  • 1Office of Science and Technology, Center for Devices and Radiological Health, Food and Drug Administration, Rockville, MD 20857, USA.

Academic Radiology
|April 11, 2001
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

Dermatological depictions in animated movies.

The British journal of dermatology·2018
Same author

Twenty-first-century skin findings response.

Journal of the European Academy of Dermatology and Venereology : JEADV·2018
Same author

Correspondence.

Journal of the National Medical Association·2010
Same author

Radiographic screen-film noise power spectrum: calibration and intercomparison.

Applied optics·2010
Same author

Radiographic screen-film noise power spectrum: variation with microdensitometer slit length.

Applied optics·2010
Same author

Psychotic symptoms and emotional distress in patients with Guillain-Barré syndrome.

European neurology·2002
Same journal

Deep Learning for Opportunistic Vertebral Fracture Detection on Routine Thoraco-abdominal Computed Tomography: A Systematic Review and Hierarchical Summary Receiver Operating Characteristic Meta-analysis of Patient-level Diagnostic Test Accuracy.

Academic radiology·2026
Same journal

"Where are They Now?": A Single Institution's 10-Years Experience with an Integrated Nuclear Radiology Fellowship.

Academic radiology·2026
Same journal

Dual-layer Spectral Detector CT Quantitative Parameters for Predicting Tumor Budding Grade and Prognosis in Stage Ⅱ Colorectal Cancer.

Academic radiology·2026
Same journal

Promotion from Associate Professor to Full Professor Should Not Be Monolithic: A National Bibliometric Study by Radiology Subspecialty.

Academic radiology·2026
Same journal

Technological Lag of Digitization for Patient Image Transfer.

Academic radiology·2026
Same journal

Prognostic Value of Coronary Sinus Flow and Aortic Pressure Gradient Quantified by 4D Flow CMR in AMI.

Academic radiology·2026
See all related articles

Using quasi-continuous rating scales in receiver operating characteristic (ROC) analysis improves diagnostic performance measurement precision. This method enhances data accuracy compared to discrete scales, benefiting diagnostic studies.

Area of Science:

  • Measurement Science
  • Medical Imaging Analysis
  • Diagnostic Accuracy Studies

Background:

  • Traditional diagnostic studies often use discrete rating scales (5-7 categories).
  • There is ongoing debate regarding the optimal rating scale for receiver operating characteristic (ROC) curve analysis.
  • Quasi-continuous rating scales are increasingly advocated for improved data collection in ROC analysis.

Purpose of the Study:

  • To quantitatively analyze the impact of data discretization on ROC curve analysis from a measurement science perspective.
  • To investigate the effect of rating scale quantization on the precision of diagnostic performance measurements.
  • To compare the performance of quasi-continuous versus discrete rating scales in ROC studies.

Main Methods:

  • Developed a model to assess the effect of data discretization on the variance of noisy data.

Related Experiment Videos

  • Conducted Monte Carlo simulations of multiple-reader, multiple-case ROC experiments.
  • Analyzed simulation data using components-of-variance models to evaluate quantization effects.

    • Main Results:

    • Discretization into five categories significantly reduces measurement precision in single-reader ROC studies.
    • The negative impact of discretization on precision may be less pronounced in multi-reader studies.
    • Quantization of rating scales introduces measurement error affecting ROC analysis outcomes.

    Conclusions:

    • Employing quasi-continuous rating scales enhances the precision of diagnostic performance measurements in ROC studies.
    • More precise measurements facilitate a more efficient allocation of healthcare resources.
    • Adoption of superior measurement methods, like quasi-continuous scales, is crucial for advancing diagnostic accuracy.