Kelly H Zou, Aiyi Liu, Andriy I Bandos

Statistical Evaluation of Diagnostic Performance

Topics in ROC Analysis

• Gebundenes Buch

Produktbeschreibung

An important method for statistical validation is the receiver operating characteristic (ROC) analysis. This visual tool is used in a variety of clinical areas, including laboratory testing, epidemiology, radiology, and bioinformatics, for evaluating diagnostic tests. This book gives a historical overview of the empirical and nonparametric ROC method for continuous diagnostic and classification data. It introduces methods for estimating and comparing ROC curves based on diagnostic test results and covers both semiparametric and parametric models. The authors develop likelihood-based algorithms for estimating an ROC curve and its characteristics under these models. They also present methods for sample size calculations and Monte Carlo simulations. The text includes many real clinical examples, with R code provided for all of them.

Produktdetails

• Verlag: CRC Press
• Seitenzahl: 246
• Erscheinungstermin: 1. Juni 2010
• Englisch
• Abmessung: 243mm x 164mm x 20mm
• Gewicht: 514g
• ISBN-13: 9781439812228
• ISBN-10: 1439812225
• Artikelnr.: 29952422

Autorenporträt

Kelly H. Zou, Aiyi Liu, Andriy I. Bandos, Lucila Ohno-Machado, Howard E. Rockette

Inhaltsangabe

Introduction: Background and Introduction. Methods for Univariate and
Multivariate Data: Diagnostic Rating Scales. Monotone Transformation
Models. Combination and Pooling of BiomarkersBayesian ROC Methods. Advanced
Approaches and Applications: Sequential Designs of ROC Experiments.
Multireader ROC Analysis. Free-Response ROC Analysis. Machine Learning and
Predictive Modeling. Discussions and Extensions: Summary and Challenges.
Section Appendices Symbols, Notations and Assumptions. Appendix A:
Conventions. Appendix B: Notations. Appendix C: Abbreviations. Appendix D:
Definitions and Terminologies. Appendix E: Remarks. Index.