A sequential classification rule based on multiple quantitative tests in the absence of a gold standard

Jingyang Zhang, Ying Zhang, Kathryn Chaloner, Jack T. Stapleton

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In many medical applications, combining information from multiple biomarkers could yield a better diagnosis than any single one on its own. When there is a lack of a gold standard, an algorithm of classifying subjects into the case and non-case status is necessary for combining multiple markers. The aim of this paper is to develop a method to construct a composite test from multiple applicable tests and derive an optimal classification rule under the absence of a gold standard. Rather than combining the tests, we treat the tests as a sequence. This sequential composite test is based on a mixture of two multivariate normal latent models for the distribution of the test results in case and non-case groups, and the optimal classification rule is derived returning the greatest sensitivity at a given specificity. This method is applied to a real-data example and simulation studies have been carried out to assess the statistical properties and predictive accuracy of the proposed composite test. This method is also attainable to implement nonparametrically.

Original languageEnglish (US)
Pages (from-to)1359-1372
Number of pages14
JournalStatistics in Medicine
Volume35
Issue number8
DOIs
StatePublished - Apr 15 2016

Fingerprint

Classification Rules
Gold
Composite
Biomarkers
Medical Applications
Multivariate Normal
Standards
Statistical property
Specificity
Simulation Study
Necessary

Keywords

  • Diagnostic test
  • EM algorithm
  • Mixture model
  • Multivariate normal

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

A sequential classification rule based on multiple quantitative tests in the absence of a gold standard. / Zhang, Jingyang; Zhang, Ying; Chaloner, Kathryn; Stapleton, Jack T.

In: Statistics in Medicine, Vol. 35, No. 8, 15.04.2016, p. 1359-1372.

Research output: Contribution to journalArticle

Zhang, Jingyang ; Zhang, Ying ; Chaloner, Kathryn ; Stapleton, Jack T. / A sequential classification rule based on multiple quantitative tests in the absence of a gold standard. In: Statistics in Medicine. 2016 ; Vol. 35, No. 8. pp. 1359-1372.
@article{b27e9066eb58443990b6c021e6b52aa0,
title = "A sequential classification rule based on multiple quantitative tests in the absence of a gold standard",
abstract = "In many medical applications, combining information from multiple biomarkers could yield a better diagnosis than any single one on its own. When there is a lack of a gold standard, an algorithm of classifying subjects into the case and non-case status is necessary for combining multiple markers. The aim of this paper is to develop a method to construct a composite test from multiple applicable tests and derive an optimal classification rule under the absence of a gold standard. Rather than combining the tests, we treat the tests as a sequence. This sequential composite test is based on a mixture of two multivariate normal latent models for the distribution of the test results in case and non-case groups, and the optimal classification rule is derived returning the greatest sensitivity at a given specificity. This method is applied to a real-data example and simulation studies have been carried out to assess the statistical properties and predictive accuracy of the proposed composite test. This method is also attainable to implement nonparametrically.",
keywords = "Diagnostic test, EM algorithm, Mixture model, Multivariate normal",
author = "Jingyang Zhang and Ying Zhang and Kathryn Chaloner and Stapleton, {Jack T.}",
year = "2016",
month = "4",
day = "15",
doi = "10.1002/sim.6780",
language = "English (US)",
volume = "35",
pages = "1359--1372",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "8",

}

TY - JOUR

T1 - A sequential classification rule based on multiple quantitative tests in the absence of a gold standard

AU - Zhang, Jingyang

AU - Zhang, Ying

AU - Chaloner, Kathryn

AU - Stapleton, Jack T.

PY - 2016/4/15

Y1 - 2016/4/15

N2 - In many medical applications, combining information from multiple biomarkers could yield a better diagnosis than any single one on its own. When there is a lack of a gold standard, an algorithm of classifying subjects into the case and non-case status is necessary for combining multiple markers. The aim of this paper is to develop a method to construct a composite test from multiple applicable tests and derive an optimal classification rule under the absence of a gold standard. Rather than combining the tests, we treat the tests as a sequence. This sequential composite test is based on a mixture of two multivariate normal latent models for the distribution of the test results in case and non-case groups, and the optimal classification rule is derived returning the greatest sensitivity at a given specificity. This method is applied to a real-data example and simulation studies have been carried out to assess the statistical properties and predictive accuracy of the proposed composite test. This method is also attainable to implement nonparametrically.

AB - In many medical applications, combining information from multiple biomarkers could yield a better diagnosis than any single one on its own. When there is a lack of a gold standard, an algorithm of classifying subjects into the case and non-case status is necessary for combining multiple markers. The aim of this paper is to develop a method to construct a composite test from multiple applicable tests and derive an optimal classification rule under the absence of a gold standard. Rather than combining the tests, we treat the tests as a sequence. This sequential composite test is based on a mixture of two multivariate normal latent models for the distribution of the test results in case and non-case groups, and the optimal classification rule is derived returning the greatest sensitivity at a given specificity. This method is applied to a real-data example and simulation studies have been carried out to assess the statistical properties and predictive accuracy of the proposed composite test. This method is also attainable to implement nonparametrically.

KW - Diagnostic test

KW - EM algorithm

KW - Mixture model

KW - Multivariate normal

UR - http://www.scopus.com/inward/record.url?scp=84959555083&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959555083&partnerID=8YFLogxK

U2 - 10.1002/sim.6780

DO - 10.1002/sim.6780

M3 - Article

C2 - 26522690

AN - SCOPUS:84959555083

VL - 35

SP - 1359

EP - 1372

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 8

ER -