The use of receiver operating characteristic curves in biomedical informatics

@article{citeulike:465479, abstract = {Receiver operating characteristic (ROC) curves are frequently used in biomedical informatics research to evaluate classification and prediction models for decision support, diagnosis, and prognosis. ROC analysis investigates the accuracy of a model's ability to separate positive from negative cases (such as predicting the presence or absence of disease), and the results are independent of the prevalence of positive cases in the study population. It is especially useful in evaluating predictive models or other tests that produce output values over a continuous range, since it captures the trade-off between sensitivity and specificity over that range. There are many ways to conduct an ROC analysis. The best approach depends on the experiment; an inappropriate approach can easily lead to incorrect conclusions. In this article, we review the basic concepts of ROC analysis, illustrate their use with sample calculations, make recommendations drawn from the literature, and list readily available software.}, author = {Lasko, Thomas A. and Bhagwat, Jui G. and Zou, Kelly H. and Ohno-Machado, Lucila }, citeulike-article-id = {465479}, doi = {10.1016/j.jbi.2005.02.008}, journal = {Journal of Biomedical Informatics}, keywords = {biology, complex, information, kdd, math, stochastic}, month = {October}, number = {5}, pages = {404--415}, posted-at = {2008-05-09 09:22:01}, priority = {2}, title = {The use of receiver operating characteristic curves in biomedical informatics}, url = {http://dx.doi.org/10.1016/j.jbi.2005.02.008}, volume = {38}, year = {2005} }

TY - JOUR ID - citeulike:465479 TI - The use of receiver operating characteristic curves in biomedical informatics JF - Journal of Biomedical Informatics VL - 38 IS - 5 SP - 404 EP - 415 N2 - Receiver operating characteristic (ROC) curves are frequently used in biomedical informatics research to evaluate classification and prediction models for decision support, diagnosis, and prognosis. ROC analysis investigates the accuracy of a model's ability to separate positive from negative cases (such as predicting the presence or absence of disease), and the results are independent of the prevalence of positive cases in the study population. It is especially useful in evaluating predictive models or other tests that produce output values over a continuous range, since it captures the trade-off between sensitivity and specificity over that range. There are many ways to conduct an ROC analysis. The best approach depends on the experiment; an inappropriate approach can easily lead to incorrect conclusions. In this article, we review the basic concepts of ROC analysis, illustrate their use with sample calculations, make recommendations drawn from the literature, and list readily available software. KW - biology KW - complex KW - information KW - kdd KW - math KW - stochastic AU - Lasko, Thomas AU - Bhagwat, Jui AU - Zou, Kelly AU - Ohno-Machado, Lucila PY - 2005/10// UR - http://dx.doi.org/10.1016/j.jbi.2005.02.008 ER -