Receiver operating characteristic analysis

A primer

Research output: Contribution to journalArticle

Abstract

In summary, the ROC curve has found many useful applications in radiology. While the statistics and mathematics behind ROC analysis can be complex, the ROC curve is fundamentally just a plot of the trade-off between sensitivity and specificity. In fact, many of the assumptions of ROC analysis (binary classification, reliance on a reference standard) are the same as those necessary to calculate sensitivity and specificity. An indication of an observer's degree of diagnostic certainty is the key additional data element that must be collected to calculate an ROC curve. Ongoing developments in ROC analysis will address more complex types of diagnostic situations and will likely expand the applicability of ROC analysis.

Original language English (US) 909-916 8 Academic Radiology 12 7 https://doi.org/10.1016/j.acra.2005.04.005 Published - Jul 2005 Yes

Fingerprint

ROC Curve
Sensitivity and Specificity
Mathematics

ASJC Scopus subject areas

• Radiology Nuclear Medicine and imaging

Cite this

In: Academic Radiology, Vol. 12, No. 7, 07.2005, p. 909-916.

Research output: Contribution to journalArticle

@article{5674964bd8c94507b0684f0d37546bdc,
title = "Receiver operating characteristic analysis: A primer",
abstract = "In summary, the ROC curve has found many useful applications in radiology. While the statistics and mathematics behind ROC analysis can be complex, the ROC curve is fundamentally just a plot of the trade-off between sensitivity and specificity. In fact, many of the assumptions of ROC analysis (binary classification, reliance on a reference standard) are the same as those necessary to calculate sensitivity and specificity. An indication of an observer's degree of diagnostic certainty is the key additional data element that must be collected to calculate an ROC curve. Ongoing developments in ROC analysis will address more complex types of diagnostic situations and will likely expand the applicability of ROC analysis.",
author = "John Eng",
year = "2005",
month = "7",
doi = "10.1016/j.acra.2005.04.005",
language = "English (US)",
volume = "12",
pages = "909--916",
issn = "1076-6332",
publisher = "Elsevier USA",
number = "7",

}

TY - JOUR

T1 - Receiver operating characteristic analysis

T2 - A primer

AU - Eng, John

PY - 2005/7

Y1 - 2005/7

N2 - In summary, the ROC curve has found many useful applications in radiology. While the statistics and mathematics behind ROC analysis can be complex, the ROC curve is fundamentally just a plot of the trade-off between sensitivity and specificity. In fact, many of the assumptions of ROC analysis (binary classification, reliance on a reference standard) are the same as those necessary to calculate sensitivity and specificity. An indication of an observer's degree of diagnostic certainty is the key additional data element that must be collected to calculate an ROC curve. Ongoing developments in ROC analysis will address more complex types of diagnostic situations and will likely expand the applicability of ROC analysis.

AB - In summary, the ROC curve has found many useful applications in radiology. While the statistics and mathematics behind ROC analysis can be complex, the ROC curve is fundamentally just a plot of the trade-off between sensitivity and specificity. In fact, many of the assumptions of ROC analysis (binary classification, reliance on a reference standard) are the same as those necessary to calculate sensitivity and specificity. An indication of an observer's degree of diagnostic certainty is the key additional data element that must be collected to calculate an ROC curve. Ongoing developments in ROC analysis will address more complex types of diagnostic situations and will likely expand the applicability of ROC analysis.

UR - http://www.scopus.com/inward/record.url?scp=22444434044&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=22444434044&partnerID=8YFLogxK

U2 - 10.1016/j.acra.2005.04.005

DO - 10.1016/j.acra.2005.04.005

M3 - Article

VL - 12

SP - 909

EP - 916