Receiver-Operating Characteristic Analysis for Evaluating Diagnostic Tests and Predictive Models

By Kelly H. Zou, A. O’Malley, L. Mauri. Circulation, 2007

1,338 citations recorded by the source corpus.

Subfields: Cardiology and Cardiovascular Medicine, Electrical and Electronic Engineering, Surgery

View Paper: https://www.semanticscholar.org/paper/7b417d2d58930c763381b6673009fc20ab21d13a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cardiology and Cardiovascular Medicine

1. J. Stoltzfus
2. Nancy R. Cook
3. Jay P. Singh
4. P. Ridker

In Electrical and Electronic Engineering

1. J. Stoltzfus
2. Tat-seng Chua
3. Hanwang Zhang
4. Xiangnan He

In Surgery

1. J. Stoltzfus
2. Nancy R. Cook

Citations in the corpus

• Statistical evaluation of prognostic versus diagnostic models: beyond the ROC curve. (2008, background): “The most popular measure of discrimination using such a range is the receiver operating characteristic (ROC) curve, a plot of sensitivityvs1specificity( 8 ).AtypicalROCcurveis shown in Fig. 1. Area under the curve (AUC) is also known as the c-statistic or c index, and can range from 0.5(nopredictive”
• Prediction of In-hospital Mortality in Emergency Department Patients With Sepsis: A Local Big Data-Driven, Machine Learning Approach. (2016, methodology): “40 Model comparison was performed in STATA using the command roc-comp to evaluate significance via chi-square statistics.”
• Logistic regression: a brief primer. (2011, methodology): “The AUROC is often used when one wants to consider different cut points for classification to maximize both sensitivity and specificity.(3,18)”
• Attributed Social Network Embedding (2017, methodology): “To judge the ranking quality, we employ the area under the ROC curve (AUROC) [48], which is widely used in IR community to evaluate a ranking list.”
• Exercise standards for testing and training: a scientific statement from the American Heart Association. (2013, positive): “0, the more accurate the test.(163) From a meta-analysis of multiple studies, the standard exercise test cut point of 0.”
• MR-IMPACT: comparison of perfusion-cardiac magnetic resonance with single-photon emission computed tomography for the detection of coronary artery disease in a multicentre, multivendor, randomized trial. (2008, neutral): “Any drugs—n (%) 44 (98) 49 (98) 47 (98) 42 (93) 45 (100) Beta-blockers—n (%) 30 (67) 37 (74) 38 (79) 36 (80) 37 (82) Statins—n (%) 36 (80) 34 (68) 34 (71) 27 (60) 30 (67) ACEI—n (%) 21 (47) 29 (58) 21 (44) 23 (51) 20 (44) Diuretics—n (%) 9 (20) 12 (24) 13 (27) 10 (22) 5 (11) Ca-CB—n (%) 13 (29) 11 (”