Klienditugi: 7440010 (E-R 10-18)

E-raamat: Statistical Evaluation of Medical Tests for Classification and Prediction

4.00/5 (14 hinnangut Goodreads-ist)

Margaret Sullivan Pepe (Professor of Biostatistics, University of Washington; Fred Hutchinson Cancer Research Center, Washington, USA)

Formaat: PDF+DRM
Sari: Oxford Statistical Science Series
Ilmumisaeg: 13-Mar-2003
Kirjastus: Oxford University Press
Keel: eng
ISBN-13: 9780191588617

Teised raamatud teemal:

Formaat - PDF+DRM
Hind: 85,20 €*
* hind on lõplik, st. muud allahindlused enam ei rakendu
Lisa ostukorvi
Lisa soovinimekirja
See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.

Formaat: PDF+DRM
Sari: Oxford Statistical Science Series
Ilmumisaeg: 13-Mar-2003
Kirjastus: Oxford University Press
Keel: eng
ISBN-13: 9780191588617

Teised raamatud teemal:

DRM piirangud

Kopeerimine (copy/paste):

ei ole lubatud
Printimine:

ei ole lubatud
Kasutamine:

Digitaalõiguste kaitse (DRM)
Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

Vajalik tarkvara
Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

Seda e-raamatut ei saa lugeda Amazon Kindle's.

This is a paperbound reprint of a 2003 book. Describing statistical concepts and techniques for evaluating the accuracy of medical tests, this text covers estimation and comparison of measures of accuracy, regression frameworks, sample size calculations, and other issues pertinent to study design. Problems relating to missing and imperfect reference data are discussed, and procedures for combining the results of multiple tests to improve classification are presented. The book's audience includes quantitative researchers and practicing and academic statisticians. Pepe teaches biostatistics at the University of Washington. Annotation ©2004 Book News, Inc., Portland, OR (booknews.com)

This book describes statistical concepts and techniques for evaluating medical diagnostic tests and biomarkers for detecting disease. More generally, the techniques pertain to the statistical classification problem for predicting a dichotomous outcome. Measures for quantifying test accuracy are described including sensitivity, specificity, predictive values, diagnostic likelihood ratios and the Receiver Operating Characteristic Curve that is commonly used for continuous and ordinal valued tests. Statistical procedures are presented for estimating and comparing them. Regression frameworks for assessing factors that influence test accuracy and for comparing tests while adjusting for such factors are presented. This book presents many worked examples of real data and should be of interest to practicing statisticians or quantitative researchers involved in the development of tests for classification or prediction in medicine.

Arvustused

Statistical results are given a methodical treatment. In each chapter statistical results are motivated, stated and usually proven, and then illustrated on a variety of datasets based on actual trials. Many of the results were developed by Pepe and her colleagues, who have advanced the field in seminal ways. Chapters end with a concluding remarks section, a set of exercises, and proofs of more involved theoretical results, when needed. Concluding remarks sections nicely summarize results and discuss open research questions. References to key papers are given throughout. This structure allows the book to serve as both a classroom text and an excellent reference to the current literature. * Journal of Biopharmaceutical Studies * I very much recommend this book to a range of audiences, from those wanting an introduction to diagnostic test concepts and methods to active researchers in the area. This book is likely to stimulate considerable progress in development of new statistical methods for diagnostic tests, an area that relative to therapeutics has received little attention from biostatisticians. * Journal of Biopharmaceutical Studies *

Notation

Introduction

(13)

The medical test

(2)

Tests, classification and the broader context

(1)

Disease screening versus diagnosis

(1)

Criteria for a useful medical test

(1)

Elements of study design

(5)

Scale for the test result

(1)

Selection of study subjects

(1)

Comparing tests

(1)

Test integrity

(1)

Sources of bias

(2)

Examples and datasets

(3)

Overview

(1)

The CASS dataset

(2)

Pancreatic cancer serum biomarkers study

(1)

Hepatitis metastasis ultrasound study

(1)

CARET PSA biomarker study

(1)

Ovarian cancer gene expression study

(1)

Neonatal audiology data

(1)

St Louis prostate cancer screening study

(1)

Topics and organization

(1)

Exercises

(2)

Measures of accuracy for binary tests

(21)

Measures of accuracy

(7)

Notation

(1)

Disease-specific classification probabilities

(2)

Predictive values

(1)

Diagnostic likelihood ratios

(4)

Estimating accuracy with data

(6)

Data from a cohort study

(1)

Proportions: (FPF, TPF) and (PPV, NPV)

(2)

Ratios of proportions: DLRs

(1)

Estimation from a case-control study

(1)

Merits of case-control versus cohort studies

(1)

Quantifying the relative accuracy of tests

(6)

Comparing classification probabilities

(1)

Comparing predictive values

(1)

Comparing diagnostic likelihood ratios

(1)

Which test is better?

(2)

Concluding remarks

(1)

Exercises

(1)

Comparing binary tests and regression analysis

(31)

Study designs for comparing tests

(2)

Unpaired designs

(1)

Paired designs

(1)

Comparing accuracy with unpaired data

(4)

Empirical estimators of comparative measures

(1)

Large sample inference

(3)

Comparing accuracy with paired data

(7)

Sources of correlation

(1)

Estimation of comparative measures

(1)

Wide or long data representations

(1)

Large sample inference

(1)

Efficiency of paired versus unpaired designs

(1)

Small sample properties

(1)

The CASS study

(3)

The regression modeling framework

(3)

Factors potentially affecting test performance

(2)

Questions addressed by regression modeling

(1)

Notation and general set-up

(1)

Regression for true and false positive fractions

(7)

Binary marginal GLM models

(1)

Fitting marginal models to data

(2)

Illustration: factors affecting test accuracy

(2)

Comparing tests with regression analysis

(3)

Regression modeling of predictive values

(3)

Model formulation and fitting

(1)

Comparing tests

(1)

The incremental value of a test for prediction

(2)

Regression models for DLRs

(2)

The model form

(1)

Fitting the DLR model

(1)

Comparing DLRs of two tests

(1)

Relationships with other regression models

(1)

Concluding remarks

(1)

Exercises

(2)

The receiver operating characteristic curve

(30)

The context

(1)

Examples of non-binary tests

(1)

Dichotomizing the test result

(1)

The ROC curve for continuous tests

(9)

Definition of the ROC

(1)

Mathematical properties of the ROC curve

(3)

Attributes of and uses for the ROC curve

(4)

Restrictions and alternatives to the ROC curve

(1)

Summary indices

(5)

The area under the ROC curve (AUC)

(2)

The ROC(t0) and partial AUC

(1)

Other summary indices

(1)

Measures of distance between distributions

(1)

The binormal ROC curve

(4)

Functional form

(1)

The binormal AUC

(1)

The binormal assumption

(1)

The ROC for ordinal tests

(7)

Tests with ordered discrete results

(1)

The latent decision variable model

(1)

Identification of the latent variable ROC

(2)

Changes in accuracy versus thresholds

(1)

The discrete ROC curve

(3)

Summary measures for the discrete ROC curve

(1)

Concluding remarks

(2)

Exercises

(2)

Estimating the ROC curve

(34)

Introduction

(1)

Approaches

(1)

Notation and assumptions

(1)

Empirical estimation

(14)

The empirical estimator

(2)

Sampling variability at a threshold

(1)

Sampling variability of ROCe(t)

(4)

The empirical AUC and other indices

103

(1)

Variability in the empirical AUC

104

(3)

Comparing empirical ROC curves

107

(2)

Illustration: pancreatic cancer biomarkers

109

(1)

Discrete ordinal data ROC curves

110

(1)

Modeling the test result distributions

111

(3)

Fully parametric modeling

111

(1)

Semiparametric location-scale models

112

(2)

Arguments against modeling test results

114

(1)

Parametric distribution-free methods: ordinal tests

114

(5)

The binormal latent variable framework

115

(2)

Fitting the discrete binormal ROC function

117

(1)

Generalizations and comparisons

118

(1)

Parametric distribution-free methods: continuous tests

119

(6)

LABROC

119

(1)

The ROC--GLM estimator

120

(4)

Inference with parametric distribution-free methods

124

(1)

Concluding remarks

125

(2)

Exercises

127

(1)

Proofs of theoretical results

128

(2)

Covariate effects on continuous and ordinal tests

130

(38)

How and why?

130

(6)

Notation

130

(1)

Aspects to model

131

(1)

Omitting covariates/pooling data

132

(4)

Reference distributions

136

(8)

Non-diseased as the reference population

136

(1)

The homogenous population

137

(2)

Nonparametric regression quantiles

139

(1)

Parametric estimation of SD,Z

140

(1)

Semiparametric models

141

(1)

Application

141

(2)

Ordinal test results

143

(1)

Modeling covariate effects on test results

144

(7)

The basic idea

144

(1)

Induced ROC curves for continuous tests

144

(4)

Semiparametric location-scale families

148

(2)

Induced ROC curves for ordinal tests

150

(1)

Random effect models for test results

150

(1)

Modeling covariate effects on ROC curves

151

(13)

The ROC--GLM regression model

152

(2)

Fitting the model to data

154

(3)

Comparing ROC curves

157

(2)

Three examples

159

(5)

Approaches to ROC regression

164

(2)

Modeling ROC summary indices

164

(1)

A qualitative comparison

164

(2)

Concluding remarks

166

(1)

Exercises

167

(1)

Incomplete data and imperfect reference tests

168

(46)

Verification biased sampling

168

(12)

Context and definition

168

(2)

The missing at random assumption

170

(1)

Correcting for bias with Bayes' theorem

170

(1)

Inverse probability weighting/imputation

171

(1)

Sampling variability of corrected estimates

172

(3)

Adjustments for other biasing factors

175

(2)

A broader context

177

(2)

Non-binary tests

179

(1)

Verification restricted to screen positives

180

(14)

Extreme verification bias

180

(1)

Identifiable parameters for a single test

181

(2)

Comparing tests

183

(2)

Evaluating covariate effects on (DP, FP)

185

(2)

Evaluating covariate effects on (TPF, FPF) and on prevalence

187

(2)

Evaluating covariate effects on (rTPF, rFPF)

189

(4)

Alternative strategies

193

(1)

Imperfect reference tests

194

(13)

Examples

194

(1)

Effects on accuracy parameters

194

(3)

Classic latent class analysis

197

(3)

Relaxing the conditional independence assumption

200

(3)

A critique of latent class analysis

203

(2)

Discrepant resolution

205

(1)

Composite reference standards

206

(1)

Concluding remarks

207

(2)

Exercises

209

(1)

Proofs of theoretical results

210

(4)

Study design and Hypothesis testing

214

(39)

The phases of medical test development

214

(4)

Research as a process

214

(1)

Five phases for the development of a medical test

215

(3)

Sample sizes for phase 2 studies

218

(11)

Retrospective validation of a binary test

218

(2)

Retrospective validation of a continuous test

220

(4)

Sample size based on the AUC

224

(4)

Ordinal tests

228

(1)

Sample sizes for phase 3 studies

229

(10)

Comparing two binary tests---paired data

229

(4)

Comparing two binary tests---unpaired data

233

(1)

Evaluating population effects on test performance

233

(1)

Comparisons with continuous test results

234

(3)

Estimating the threshold for screen positivity

237

(1)

Remarks on phase 3 analyses

238

(1)

Sample sizes for phase 4 studies

239

(6)

Designs for inference about (FPF, TPF)

239

(2)

Designs for predictive values

241

(2)

Designs for (FP, DP)

243

(1)

Selected verification of screen negatives

244

(1)

Phase 5

245

(1)

Matching and stratification

246

(2)

Stratification

246

(1)

Matching

247

(1)

Concluding remarks

248

(3)

Exercises

251

(2)

E-raamat: Statistical Evaluation of Medical Tests for Classification and Prediction

DRM piirangud

Kopeerimine (copy/paste):

Printimine:

Kasutamine:

Arvustused

Konto & seaded

Otsing

Otsingu andmebaas

Filtreeri tulemusi

Teemad E-raamatute teemad

Vali ostukorv