Observer Performance Methods for Diagnostic Imaging(Imaging in Medical Diagnosis and Therapy) H 590 p. 17
Chakraborty, Dev P. 著
目次
Preliminaries Introduction Clinical tasks Imaging device development and its clinical deployment Image quality vs. task performance Why physical measures of image quality are not enough Model observers Measuring observer performance: four paradigms Hierarchy of assessment methods Overview of the book and how to use it PART I The binary paradigm Introduction Decision vs. truth: the fundamental 2x2 table of ROC analysis Sensitivity and specificity Reasons for the names sensitivity and specificity Estimating sensitivity and specificity Disease prevalence Accuracy Positive and negative predictive values Example: calculation of PPV and NPV PPV and NPV are irrelevant to laboratory tasks Modeling the binary task Introduction Decision variable and decision threshold Changing the decision threshold: Example I Changing the decision threshold: Example II The equal-variance binormal model The normal distribution Demonstration of the concepts of sensitivity and specificity Inverse variation of sensitivity and specificity The ROC curve Assigning confidence intervals to an operating point Variability in sensitivity and specificity: the Beam et al study The ratings paradigm Introduction The ROC counts table Operating points from counts table Relation between ratings paradigm and the binary task Ratings are not numerical values A single "clinical" operating point from ratings data The forced choice paradigm Observer performance studies as laboratory simulations of clinical tasks Discrete vs. continuous ratings: the Miller study The BIRADS ratings scale and ROC studies The controversy Empirical AUC Introduction The empirical ROC plot Empirical operating points from ratings data AUC under the empirical ROC plot The Wilcoxon statistic Bamber’s theorem The importance of Bamber’s theorem Appendix 5.A: Details of Wilcoxon theorem Binormal model Introduction The binormal model Least-squares estimation Maximum likelihood estimation (MLE) Expression for area under ROC curve Appendix 6.A: Expressions for partial and full area under the binormal ROC Sources of variability affecting AUC Introduction Three sources of variability Dependence of AUC on the case sample Estimating case-sampling variability using the DeLong method Estimating case-sampling variability of AUC using the bootstrap method Estimating case-sampling variability of AUC using the jackknife method Estimating case-sampling variability of AUC using a calibrated simulator Dependence of AUC on the reader’s expertise Dependence of AUC on the modality Effect on empirical AUC of variations in thresholds and numbers of bins Empirical vs. fitted AUCs PART II Hypothesis Testing Introduction Hypothesis testing for a single-modality single-reader ROC study Type-I errors One-sided vs. two sided tests Statistical power Some comments on the code Why is alpha chosen to be 5% Dorfman-Berbaum-Metz-Hillis (DBMH) analysis Co-author: Xuetong Zhai, MS Introduction Random and fixed factors Reader and case populations and data correlations Three types of analyses General approach The Dorfman-Berbaum-Metz (DBM) method Random-reader random-case (RRRC) analysis Fixed-reader random-case (FRRC) analysis Random-reader fixed-case (RRFC) analysis DBMH Analysis: Example 1 DBMH Analysis: Example 2 Validation of DBMH analysis Meaning of pseudovalues Obuchowski-Rockette-Hillis (ORH) analysis Co-author: Xuetong Zhai, MS Introduction The single reader multiple treatment model The multiple reader multiple treatment model ORH model Special cases: fixed-reader and fixed-case analyses Example of ORH analysis Comparison of ORH and DBMH methods Sample size estimation for ROC studies Introduction Statistical power Sample size estimation Dependence of statistical power on estimates of model parameters Formulae for random-reader random-case (RRRC) sample size estimation Formulae for fixed-reader random-case (FRRC) sample size estimation Formulae for random-reader fixed-case (RRFC) sample size estimation Example 1 Example 2 Details of the sample size estimation process Cautionary notes: the Kraemer et al paper Prediction accuracy of sample size estimation method On the unit of effect-size: a proposal PART III The FROC paradigm Introduction Location specific paradigms The FROC paradigm as a search task A pioneering FROC study in medical imaging Population and binned FROC plots The "solar" analogy: search vs. classification performance Empirical operating characteristics derivable from FROC data Introduction Latent vs. actual marks Formalism: the FROC plot Formalism: the alternative FROC (AFROC) plot The EFROC plot Formalism: the inferred ROC plot Formalism: the weighted-AFROC (wAFROC) plot Formalism: the AFROC1 plot Formalism: the weighted-AFROC1 (wAFROC1) plot Example: "raw" FROC plots Confusion about location-level "true-negatives" Example: binned FROC plots Example: "raw" FROC / AFROC plots Example: binned AFROC plots Example: binned FROC/AFROC/ROC plots Recommendations Computation and meanings of empirical FROC FOM-statistics and AUC measures Introduction Empirical AFROC FOM-statistic Empirical weighted-AFROC FOM-statistic Two Theorems Understanding the AFROC and wAFROC empirical plots Physical interpretation of AFROC-based FOMs Visual Search Paradigms Introduction Grouping and labeling ROIs in an image Recognition/Finding vs. detection Two visual search paradigms Determining where the radiologist is looking The Nodine - Kundel search model Analyzing simultaneously acquired eye-tracking & FROC data The radiological search model (RSM) Introduction The radiological search model (RSM) Physical interpretation of RSM parameters Model re-parameterization Equation Chapter 16 Section 1 Predictions of the RSM Introduction Inferred integer ROC ratings Constrained end-point property of the RSM-predicted ROC curve The RSM-predicted ROC curve Example: RSM-predicted ROC/pdf curves The RSM-predicted FROC curve The RSM-predicted AFROC curve Quantifying search performance Quantifying lesion-classification performance The FROC curve is a poor descriptor of search performance Evidence for the RSM Fitting RSM to FROC/ROC data and key findings Introduction FROC Likelihood function IDCA Likelihood function ROC Likelihood function RSM vs. PROPROC and CBM, and a serendipitous finding Reason for serendipitous finding Analyzing FROC data and sample size estimation Introduction Example analysis of a FROC dataset Plotting wAFROC curves Single fixed-factor Analysis Crossed treatment analysis Sample size estimation: wAFROC FOM PART IV Proper ROC models Introduction Theorem: slope of ROC equals likelihood ratio Theorem: likelihood ratio observer maximizes AUC Proper vs. improper ROC curves Degenerate datasets The likelihood ratio observer PROPROC formalism The contaminated binormal model The bigamma model The bivariate binormal model and its software implementation (CORROC2) Introduction The bivariate binormal model The multivariate probability density function Visualizing the bivariate probability density function Estimating bivariate binormal model parameters CORROC2 software Application to a real dataset Comparing performance of standalone CAD to a group of radiologists interpreting the same cases Introduction The Hupse-Karssemeijer et al. study Extending the analysis to random cases Ambiguity in interpreting a point-based FOM Results using full-area measures Design and calibration of a single-modality multiple-reader decision variable simulator and using it to validate the proposed CAD analysis methodCo-author: Xuetong Zhai, MSIntroductionBivariate contaminated binormal model (BCBM)Single-modality multiple-reader decision variable simulatorCalibration, validation of simulator and testing its NH behavior
カート
カートに商品は入っていません。