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

ROC analysis with multiple classes and multiple tests: methodology and its application in microarray studies.

Jialiang Li1, Jason P Fine

  • 1Department of Statistics and Applied Probability, National University of Singapore, Singapore.

Biostatistics (Oxford, England)
|February 29, 2008
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

MXene/RGO/Si Schottky Junction for a High-Performance Self-Powered Broadband Photodetector.

Langmuir : the ACS journal of surfaces and colloids·2026
Same author

Hippocampal GFAP in aging: Associations with AD and LATE-NC pathologies and cognitive decline in older adults.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2026
Same author

Preoperative dermatomal somatosensory evoked potentials in risk prediction of early postoperative neurological deterioration after thoracic spine surgery: a retrospective cohort study.

Journal of orthopaedic surgery and research·2026
Same author

Synergistic Design of ZnCo-MnO@NPC Cathode and ZIF-8@Zn Anode for High-Performance Aqueous Zinc-Ion Batteries.

Molecules (Basel, Switzerland)·2026
Same author

Correction: Rice growth and yield formation in heterogeneous sodic and saline-sodic soils: challenges and management strategies.

Frontiers in plant science·2026
Same author

Poly2Vec: Polymorphic Fourier-Based Encoding of Geospatial Objects for GeoAI Applications.

Proceedings of machine learning research·2026
Same journal

A Bayesian functional concurrent zero-inflated Dirichlet-multinomial regression model with application to infant microbiome.

Biostatistics (Oxford, England)·2026
Same journal

Towards optimal environmental policies: policy learning under arbitrary bipartite network interference.

Biostatistics (Oxford, England)·2026
Same journal

Multilevel functional quantile principal component analysis.

Biostatistics (Oxford, England)·2026
Same journal

Adaptive transfer learning for time-to-event modeling with applications in disease risk assessment.

Biostatistics (Oxford, England)·2026
Same journal

High-dimensional test for one-sided hypotheses.

Biostatistics (Oxford, England)·2026
Same journal

NBSR: a Negative Binomial Softmax Regression model for microRNA-seq data analysis.

Biostatistics (Oxford, England)·2026
See all related articles

This study introduces a new method for evaluating diagnostic tests with multiple classes using receiver operating characteristic (ROC) measures. It provides a practical approach for inferential procedures, improving diagnostic accuracy assessments.

Area of Science:

  • Statistics
  • Biostatistics
  • Machine Learning

Background:

  • Traditional receiver operating characteristic (ROC) curve analysis is limited to binary outcomes.
  • Extensions like volume under the surface (VUS) and hypervolume under the manifold (HUM) for multi-class diagnosis lack simple inferential procedures.
  • Calculating multi-class ROC measures can be complex due to the need for class probability assessments.

Purpose of the Study:

  • To develop a practical method for inferential procedures for multi-class ROC measures.
  • To address the limitations of existing multi-class diagnostic accuracy summarization techniques.
  • To compare ROC-based analysis with the correct classification rate, considering class prevalences.

Main Methods:

  • Estimating class probabilities using multinomial logistic regression.

Related Experiment Videos

  • Employing bootstrap inferences to account for variability in probability estimation.
  • Comparing multi-class ROC measures with the correct classification rate.

    • Main Results:

    • The proposed method provides a feasible approach for inferential procedures in multi-class diagnostic accuracy.
    • Bootstrap inferences demonstrate good performance in simulations.
    • ROC-based analysis can lead to substantially different conclusions compared to prevalence-dependent methods like correct classification rate.
    • The ROC-based analysis resulted in decreased model complexity in a tumor classification example.

    Conclusions:

    • The developed method enhances the practical utility of multi-class ROC measures for diagnostic accuracy assessment.
    • This approach offers a valuable alternative to prevalence-dependent metrics, particularly in complex classification tasks.
    • The findings suggest that ROC-based analysis can simplify models and improve interpretability in multi-class diagnostic scenarios.