Available:*
Library | Item Barcode | Call Number | Material Type | Item Category 1 | Status |
---|---|---|---|---|---|
Searching... | 30000010184987 | R853.S7 B76 2007 | Open Access Book | Book | Searching... |
On Order
Summary
Summary
There are numerous advantages to using Bayesian methods in diagnostic medicine, which is why they are employed more and more today in clinical studies. Exploring Bayesian statistics at an introductory level, Bayesian Biostatistics and Diagnostic Medicine illustrates how to apply these methods to solve important problems in medicine and biology.
After focusing on the wide range of areas where diagnostic medicine is used, the book introduces Bayesian statistics and the estimation of accuracy by sensitivity, specificity, and positive and negative predictive values for ordinal and continuous diagnostic measurements. The author then discusses patient covariate information and the statistical methods for estimating the agreement among observers. The book also explains the protocol review process for cancer clinical trials, how tumor responses are categorized, how to use WHO and RECIST criteria, and how Bayesian sequential methods are employed to monitor trials and estimate sample sizes.
With many tables and figures, this book enables readers to conduct a Bayesian analysis for a large variety of interesting and practical biomedical problems.
Table of Contents
Preface | p. ix |
Acknowledgments | p. xi |
Author | p. xiii |
1 Introduction | p. 1 |
1.1 Introduction | p. 1 |
1.2 Statistical Methods in Diagnostic Medicine | p. 1 |
1.3 Preview of Book | p. 2 |
1.4 Datasets for the Book | p. 4 |
1.5 Software | p. 4 |
References | p. 5 |
2 Diagnostic Medicine | p. 7 |
2.1 Introduction | p. 7 |
2.2 Imaging Modalities | p. 7 |
2.3 Activities in Diagnostic Imaging | p. 11 |
2.4 Accuracy and Agreement | p. 12 |
2.5 Developmental Trials for Imaging | p. 14 |
2.6 Protocol Review and Clinical Trials | p. 15 |
2.6.1 The Protocol | p. 16 |
2.6.2 Phase I, II, and III Clinical Designs | p. 16 |
2.7 The Literature | p. 18 |
References | p. 18 |
3 Other Diagnostic Procedures | p. 21 |
3.1 Introduction | p. 21 |
3.2 Sentinel Lymph Node Biopsy for Melanoma | p. 21 |
3.3 Tumor Depth to Diagnose Metastatic Melanoma | p. 22 |
3.4 Biopsy for Nonsmall Cell Lung Cancer | p. 23 |
3.5 Coronary Artery Disease | p. 24 |
References | p. 24 |
4 Bayesian Statistics | p. 27 |
4.1 Introduction | p. 27 |
4.2 Bayes Theorem | p. 28 |
4.3 Prior Information | p. 29 |
4.4 Posterior Information | p. 33 |
4.5 Inference | p. 36 |
4.5.1 Introduction | p. 36 |
4.5.2 Estimation | p. 36 |
4.5.3 Testing Hypotheses | p. 38 |
4.5.3.1 Introduction | p. 38 |
4.5.3.2 Binomial Example of Testing | p. 39 |
4.5.3.3 Comparing Two Binomial Populations | p. 40 |
4.5.3.4 Sharp Null Hypothesis for the Normal Mean | p. 41 |
4.6 Sample Size | p. 41 |
4.6.1 Introduction | p. 41 |
4.6.2 A One-Sample Binomial for Response | p. 42 |
4.6.3 A One-Sample Binomial with Prior Information | p. 44 |
4.6.4 Comparing Two Binomial Populations | p. 45 |
4.7 Computing | p. 45 |
4.7.1 Introduction | p. 45 |
4.7.2 Direct Methods of Computation | p. 46 |
4.7.3 Gibbs Sampling | p. 49 |
4.7.3.1 Introduction | p. 49 |
4.7.3.2 Common Mean of Normal Populations | p. 50 |
4.7.3.3 MCMC Sampling with WinBUGS[Registered] | p. 53 |
4.8 Exercises | p. 56 |
References | p. 57 |
5 Bayesian Methods for Diagnostic Accuracy | p. 59 |
5.1 Introduction | p. 59 |
5.2 Study Design | p. 60 |
5.2.1 The Protocol | p. 60 |
5.2.2 Objectives | p. 60 |
5.2.3 Background | p. 61 |
5.2.4 Patient and Reader Selection | p. 61 |
5.2.5 Study Plan | p. 62 |
5.2.6 Number of Patients | p. 63 |
5.2.7 Statistical Design and Analysis | p. 63 |
5.2.8 References | p. 64 |
5.3 Bayesian Methods for Test Accuracy: Binary and Ordinal Data | p. 65 |
5.3.1 Introduction | p. 65 |
5.3.2 Classification Probabilities | p. 65 |
5.3.3 Predictive Values | p. 68 |
5.3.4 Diagnostic Likelihood Ratios | p. 69 |
5.3.5 ROC Curve | p. 69 |
5.4 Bayesian Methods for Test Accuracy: Quantitative Variables | p. 73 |
5.4.1 Introduction | p. 73 |
5.4.2 The Spokane Heart Study | p. 73 |
5.4.3 ROC Area | p. 74 |
5.4.4 Definition of the ROC Curve | p. 77 |
5.4.5 Choice of Optimal Threshold Value | p. 78 |
5.5 Clustered Data: Detection and Localization | p. 79 |
5.5.1 Introduction | p. 79 |
5.5.2 Bayesian ROC Curve for Clustered Information | p. 80 |
5.5.3 Clustered Data in Mammography | p. 82 |
5.6 Comparing Accuracy between Modalities | p. 84 |
5.7 Sample Size Determination | p. 87 |
5.7.1 Introduction | p. 87 |
5.7.2 Discrete Diagnostic Scores | p. 88 |
5.7.2.1 Binary Tests | p. 88 |
5.7.2.2 Multinomial Outcomes | p. 91 |
5.7.3 Sample Sizes: Continuous Diagnostic Scores | p. 92 |
5.7.3.1 One ROC Curve | p. 92 |
5.7.3.2 Two ROC Curves | p. 95 |
5.8 Exercises | p. 97 |
References | p. 99 |
6 Regression and Test Accuracy | p. 101 |
6.1 Introduction | p. 101 |
6.2 Audiology Study | p. 102 |
6.2.1 Introduction | p. 102 |
6.2.2 Log Link Function | p. 102 |
6.2.3 Logistic Link | p. 103 |
6.2.4 Diagnostic Likelihood Ratio | p. 105 |
6.3 ROC Area and Patient Covariates | p. 107 |
6.3.1 Introduction | p. 107 |
6.3.2 ROC Curve as Response to Therapy | p. 108 |
6.3.3 Diagnosing Prostate Cancer | p. 110 |
6.4 Exercises | p. 111 |
References | p. 111 |
7 Agreement | p. 113 |
7.1 Introduction | p. 113 |
7.2 Agreement for Discrete Ratings | p. 114 |
7.2.1 Binary Scores | p. 114 |
7.2.2 Other Indices of Agreement | p. 116 |
7.2.3 A Bayesian Version of McNemar | p. 117 |
7.2.4 Comparing Two Kappa Parameters | p. 117 |
7.2.5 Kappa and Stratification | p. 119 |
7.2.6 Multiple Categories and Two Readers | p. 120 |
7.2.7 Multiple Categories | p. 122 |
7.2.8 Agreement and Covariate Information | p. 123 |
7.3 Agreement for a Continuous Response | p. 126 |
7.3.1 Introduction | p. 126 |
7.3.2 Intra-Class Correlation Coefficient | p. 127 |
7.3.2.1 One-Way Random Model | p. 127 |
7.3.2.2 Two-Way Random Model | p. 131 |
7.3.3 Regression and Agreement | p. 132 |
7.4 Combining Reader Information | p. 134 |
7.5 Exercises | p. 136 |
References | p. 139 |
8 Diagnostic Imaging and Clinical Trials | p. 141 |
8.1 Introduction | p. 141 |
8.2 Clinical Trials | p. 142 |
8.2.1 Introduction | p. 142 |
8.2.2 Phase I Designs | p. 142 |
8.2.3 Phase II Trials | p. 143 |
8.2.4 Phase III Trials | p. 145 |
8.3 Protocol | p. 145 |
8.4 Guidelines for Tumor Response | p. 146 |
8.5 Bayesian Sequential Stopping Rules | p. 148 |
8.6 Software for Clinical Trials | p. 152 |
8.6.1 CRM Simulator for Phase I Trials | p. 153 |
8.6.2 Multc Lean for Phase II Trials | p. 153 |
8.7 Examples | p. 154 |
8.7.1 Phase I Trial for Renal Cell Carcinoma | p. 154 |
8.7.2 An Ideal Phase II Trial | p. 156 |
8.7.3 Phase II Trial for Advanced Melanoma | p. 158 |
8.8 Exercises | p. 162 |
References | p. 163 |
9 Other Topics | p. 165 |
9.1 Introduction | p. 165 |
9.2 Imperfect Diagnostic Test Procedures | p. 166 |
9.2.1 Extreme Verification Bias | p. 166 |
9.2.2 Verification Bias | p. 170 |
9.2.3 Estimating Test Accuracy with No Gold Standard | p. 173 |
9.3 Test Accuracy and Survival Analysis | p. 178 |
9.4 ROC Curves with a Non-Binary Gold Standard | p. 180 |
9.5 Periodic Screening in Cancer | p. 182 |
9.5.1 Inference for Sensitivity and Transition Probability | p. 182 |
9.5.2 Bayesian Inference for Lead-Time | p. 186 |
9.6 Decision Theory and Diagnostic Accuracy | p. 189 |
9.7 Exercises | p. 192 |
References | p. 193 |
Index | p. 195 |