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

Computation of multifactorial receiver operator and predictive accuracy characteristics

K Hnatkova1, J D Poloniecki, A J Camm

  • 1Department of Cardiological Sciences, St. George's Hospital Medical School, London, UK.

Computer Methods and Programs in Biomedicine
|March 1, 1994
PubMed
Summary

This study introduces an efficient algorithm for computing multivariate receiver operator characteristics, simplifying complex data analysis. The method enhances risk stratification for conditions like acute myocardial infarction, improving patient outcome predictions.

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

Adverse effects of amiodarone at low dose: plus ça change.

Journal of the American College of Cardiology·1997
Same author

Atrial premature beats preceding episodes of paroxysmal atrial fibrillation: factorial analysis of a prediction system.

Pacing and clinical electrophysiology : PACE·1997
Same author

Exercise-induced T-wave alternans as a marker of high risk in patients with hypertrophic cardiomyopathy.

Japanese circulation journal·1997
Same author

Elevated Chlamydia pneumoniae antibodies, cardiovascular events, and azithromycin in male survivors of myocardial infarction.

Circulation·1997
Same author

Chlamydia pneumoniae and coronary heart disease.

BMJ (Clinical research ed.)·1997
Same author

Amiodarone after myocardial infarction: EMIAT and CAMIAT trials.

Lancet (London, England)·1997

Area of Science:

  • Biomedical Engineering
  • Computational Statistics
  • Medical Informatics

Background:

  • Receiver operator characteristics (ROC) are crucial for diagnostic test evaluation.
  • Univariate ROC analysis is straightforward, but multivariate ROC computation is computationally intensive, increasing exponentially with data dimensionality.
  • Existing methods struggle with high-dimensional data, limiting their clinical application.

Purpose of the Study:

  • To develop and present an efficient algorithm for computing multivariate receiver operator characteristics (MROC) and related predictive characteristics.
  • To demonstrate the algorithm's utility in a clinical risk stratification study.
  • To address the computational challenges associated with high-dimensional data in diagnostic accuracy assessment.

Main Methods:

Related Experiment Videos

  • The algorithm employs pre-sorting of data within each dimension and stratification into groups with 100% positive and negative case separation.
  • It leverages computational efficiency concepts to handle multivariate data complexity.
  • A risk stratification study involving 539 acute myocardial infarction survivors was conducted using signal-averaged electrocardiogram indices.

Main Results:

  • The algorithm significantly reduces computational complexity for multivariate ROC analysis.
  • The risk stratification study successfully identified patients at high risk of early death post-myocardial infarction.
  • Specific computing times for the algorithm in the study are detailed, demonstrating its practical efficiency.

Conclusions:

  • The developed algorithm provides an efficient solution for multivariate ROC analysis, overcoming previous computational barriers.
  • This method facilitates improved risk stratification in clinical settings, particularly for complex conditions.
  • The algorithm's efficiency and applicability are validated through a real-world clinical study.