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

[Random sample size in ROC analyses].

W Steinbach, K Richter, H Cobet

    Rofo : Fortschritte Auf Dem Gebiete Der Rontgenstrahlen Und Der Nuklearmedizin
    |August 1, 1985
    PubMed
    Summary

    Receiver Operating Characteristic (ROC) analysis effectiveness depends on sample size. For reliable diagnostic efficiency assessment using ROC curves, employ random sample groups of at least 200 images each.

    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

    The effect of storage on allergen and microbial agent levels in frozen house dust.

    Allergy·2003
    Same author

    Dose reduction of inhaled corticosteroids under concomitant medication with montelukast in patients with asthma.

    The European respiratory journal·2002
    Same author

    Montelukast versus fluticasone: effects on lung function, airway responsiveness and inflammation in moderate asthma.

    The European respiratory journal·2002
    Same author

    Cerebral toxoplasmosis after combined liver-pancreas-kidney and liver-pancreas transplantation.

    Transplantation proceedings·2002
    Same author

    Predictors of cat allergen (Fel d 1) in house dust of German homes with/without cats.

    Journal of investigational allergology & clinical immunology·2002
    Same author

    [European Community Respiratory Health Survey in Adults (ECRHS)].

    Pneumologie (Stuttgart, Germany)·2002

    Area of Science:

    • Medical Imaging Analysis
    • Diagnostic Accuracy Studies

    Context:

    • Receiver Operating Characteristic (ROC) analysis is a valuable tool for evaluating diagnostic test performance.
    • Assessing diagnostic efficiency in medical imaging requires careful consideration of sample size and data variability.

    Purpose:

    • To investigate the impact of random sample size on the diagnostic efficiency assessment using ROC analysis for thoracic images.
    • To compare the diagnostic performance of experienced radiologists with a semi-automatic classification system.

    Summary:

    • Satisfactory convergence between radiologist ratings and ROC curves was observed when random sample groups comprised approximately 200 images or more.
    • Smaller sample sizes (<200 images) resulted in increased scatter and reduced reliability of rating performance.
    • A semi-automatic classifier achieved a performance index comparable to that of a low-performing human evaluator.

    Impact:

    • Highlights the critical role of adequate sample size in ensuring the reliability of ROC analysis for diagnostic efficiency.
    • Provides benchmarks for sample size selection in future diagnostic imaging studies.
    • Suggests potential for semi-automatic systems to augment or assist in diagnostic image interpretation.

    Related Experiment Videos