Statistics review 13 Receiver operating characteristic curves 英文参考文献.docVIP

Critical Care December 2004 Vol 8 No 6 Bewick et al. Review Statistics review 13: Receiver operating characteristic curves Viv Bewick1, Liz Cheek1 and Jonathan Ball2 1Senior Lecturer, School of Computing, Mathematical and Information Sciences, University of Brighton, Brighton, UK 2Senior Registrar in ICU, Liverpool Hospital, Sydney, Australia Corresponding author: Viv Bewick, v.bewick@brighton.ac.uk Published online: 4 November 2004 Critical Care 2004, 8:508-512 (DOI 10.1186/cc3000) This article is online at /content/8/6/508 ? 2004 BioMed Central Ltd Abstract This review introduces some commonly used methods for assessing the performance of a diagnostic test. The sensitivity, specificity and likelihood ratio of a test are discussed. The uses of the receiver operating characteristic curve and the area under the curve are explained. Keywords AUROC, negative likelihood ratio, negative predictive value, positive likelihood ratio, positive predictive value, ROC curve, sensitivity, specificity Introduction who survived were correctly identified by this test. Because A simple diagnostic test for a particular disease or outcome classifies patients into two groups: those with the outcome and those without. A test is assessed by its ability to diagnose the outcome correctly, whether this is positive or negative. If the actual outcome is not evident then it may be supplied by the ‘gold standard’ test. The data given in Table 1 provide an example in which the outcome is death or survival. The patients were attending an accident and emergency unit and the venous blood analysis for the metabolic marker lactate was used in the early identification of those patients at risk for death. Patients with lactate levels greater than 1.5 mmol/l were considered to be at risk. In general, the results of a diagnostic test may be presented as shown in Table 2. both of these measures are