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

Reader studies for validation of CAD systems.

Brandon D Gallas1, David G Brown

  • 1NIBIB/CDRH Laboratory for the Assessment of Medical Imaging Systems, FDA, Silver Spring, MD 20993-0002, United States. brandon.gallas@fda.hhs.gov

Neural Networks : the Official Journal of the International Neural Network Society
|January 25, 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

Methodological considerations for evaluating deep learning segmentation models in digital pathology whole-slide images.

Journal of medical imaging (Bellingham, Wash.)·2026
Same author

Medical Imaging Data Strategies for Catalyzing AI Medical Device Innovation.

Journal of imaging informatics in medicine·2025
Same author

Prioritizing cases from a multi-institutional cohort for a dataset of pathologist annotations.

Journal of pathology informatics·2024
Same author

The Role of Water Networks in Phosphodiesterase Inhibitor Dissociation and Kinetic Selectivity.

ChemMedChem·2024
Same author

Training pathologists to assess stromal tumour-infiltrating lymphocytes in breast cancer synergises efforts in clinical care and scientific research.

Histopathology·2024
Same author

FDA Fosters Innovative Approaches in Research, Resources, and Collaboration.

Nature machine intelligence·2024
Same journal

Exploiting audio-visual modalities in videos: Object detection via multi-stage bilateral coupling network.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Reliability-aware modality completion with cross-modal distillation for federated learning with missing modalities.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

IGFD-Net: Illumination-guided frequency decoupling for polarization image fusion.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Multiple-Strategies dung beetle optimizer and its applications in engineering optimization and bankruptcy prediction.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Aggregating global-scale pixel-wise forgery cues within a graph.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Finite-Time intermittent control for secure synchronization of Neutral-Type stochastic delayed neural networks under aperiodic DoS attacks.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study introduces a new statistical method to compare how well readers detect abnormalities in medical images with and without computer-assisted diagnosis (CAD) systems, accounting for reader variability.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Statistical Analysis

Background:

  • Evaluating computational intelligence (CI) systems for human operators requires accounting for human variability.
  • Medical imaging computer-assisted diagnosis (CAD) systems aim to improve diagnostic performance.
  • Reader variability is a key factor in assessing CAD system effectiveness.

Purpose of the Study:

  • To outline a method for comparing reader detection performance with and without CAD systems.
  • To extend existing multi-reader multi-case (MRMC) variance analysis methods.
  • To develop a method capable of analyzing arbitrary study designs, not just fully-crossed designs.

Main Methods:

  • Utilizing receiver operating characteristic (ROC) experiments to compare reader performance.

Related Experiment Videos

  • Summarizing performance using the reader-averaged area under the ROC curve (AUC).
  • Applying advanced multi-reader multi-case (MRMC) variance analysis to account for random readers, cases, and correlations.
  • Presenting a novel method for MRMC variance analysis that accommodates arbitrary study designs.

    • Main Results:

    • The proposed method allows for statistically powerful comparisons of reader performance in CAD studies.
    • The method extends MRMC analysis beyond fully-crossed designs to arbitrary study configurations.
    • Computer simulations demonstrated the method's utility and assessed the statistical power of various designs.

    Conclusions:

    • The developed method provides a flexible and statistically robust approach for evaluating CAD systems in medical imaging.
    • This work enhances the ability to accurately assess the impact of CAD on diagnostic performance while considering human variability.
    • The findings support more sophisticated study designs for future CAD system evaluations.