A novel gibbs maximum a posteriori (GMAP) approach on bayesian nonlinear mixed-effects population pharmacokinetics (PK) models

Seongho Kim, Stephen D. Hall, Lang Li

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

In this paper, various Bayesian Monte Carlo Markov chain (MCMC) methods and the proposed algorithm, the Gibbs maximum a posteriori (GMAP) algorithm, are compared for implementing the nonlinear mixed-effects model in pharmacokinetics (PK) studies. An intravenous two-compartmental PK model is adopted to fit the PK data from the midazolam (MDZ) studies, which recruited twenty-four individuals with nine different time points per subject. The three-stage hierarchical nonlinear mixed model is constructed. Data analysis and model performance comparisons show that GMAP converges the fastest and provides reliable results. At the mean time, data augmentation (DA) methods are used for the Random-walk Metropolis method. Data analysis shows that the speed of the convergence of Random-walk Metropolis can be improved by DA, but all of them are not as fast as GMAP. The performance of GMAP and various MCMC algorithms are compared through Midazolam data analysis and simulation.

Original languageEnglish (US)
Pages (from-to)700-720
Number of pages21
JournalJournal of biopharmaceutical statistics
Volume19
Issue number4
DOIs
StatePublished - Jul 1 2009

Fingerprint

Mixed Effects
Bayes Theorem
Maximum a Posteriori
Pharmacokinetics
Nonlinear Effects
Markov Chains
Midazolam
Data Augmentation
Data analysis
Population
Random walk
Nonlinear Dynamics
Nonlinear Mixed Model
Nonlinear Mixed Effects Model
Model Comparison
Markov Chain Monte Carlo Algorithms
Markov Chain Monte Carlo Methods
Performance Comparison
Model
Data Model

Keywords

  • Bayesian model
  • Gibbs Maximum a Posteriori (GMAP)
  • Monte Carlo Markov Chain (MCMC)
  • Pharmacokinetics (PK)

ASJC Scopus subject areas

  • Pharmacology (medical)
  • Pharmacology
  • Statistics and Probability

Cite this

A novel gibbs maximum a posteriori (GMAP) approach on bayesian nonlinear mixed-effects population pharmacokinetics (PK) models. / Kim, Seongho; Hall, Stephen D.; Li, Lang.

In: Journal of biopharmaceutical statistics, Vol. 19, No. 4, 01.07.2009, p. 700-720.

Research output: Contribution to journalArticle

@article{a848557a94694b4e98a4a4820e86f692,
title = "A novel gibbs maximum a posteriori (GMAP) approach on bayesian nonlinear mixed-effects population pharmacokinetics (PK) models",
abstract = "In this paper, various Bayesian Monte Carlo Markov chain (MCMC) methods and the proposed algorithm, the Gibbs maximum a posteriori (GMAP) algorithm, are compared for implementing the nonlinear mixed-effects model in pharmacokinetics (PK) studies. An intravenous two-compartmental PK model is adopted to fit the PK data from the midazolam (MDZ) studies, which recruited twenty-four individuals with nine different time points per subject. The three-stage hierarchical nonlinear mixed model is constructed. Data analysis and model performance comparisons show that GMAP converges the fastest and provides reliable results. At the mean time, data augmentation (DA) methods are used for the Random-walk Metropolis method. Data analysis shows that the speed of the convergence of Random-walk Metropolis can be improved by DA, but all of them are not as fast as GMAP. The performance of GMAP and various MCMC algorithms are compared through Midazolam data analysis and simulation.",
keywords = "Bayesian model, Gibbs Maximum a Posteriori (GMAP), Monte Carlo Markov Chain (MCMC), Pharmacokinetics (PK)",
author = "Seongho Kim and Hall, {Stephen D.} and Lang Li",
year = "2009",
month = "7",
day = "1",
doi = "10.1080/10543400902964159",
language = "English (US)",
volume = "19",
pages = "700--720",
journal = "Journal of Biopharmaceutical Statistics",
issn = "1054-3406",
publisher = "Taylor and Francis Ltd.",
number = "4",

}

TY - JOUR

T1 - A novel gibbs maximum a posteriori (GMAP) approach on bayesian nonlinear mixed-effects population pharmacokinetics (PK) models

AU - Kim, Seongho

AU - Hall, Stephen D.

AU - Li, Lang

PY - 2009/7/1

Y1 - 2009/7/1

N2 - In this paper, various Bayesian Monte Carlo Markov chain (MCMC) methods and the proposed algorithm, the Gibbs maximum a posteriori (GMAP) algorithm, are compared for implementing the nonlinear mixed-effects model in pharmacokinetics (PK) studies. An intravenous two-compartmental PK model is adopted to fit the PK data from the midazolam (MDZ) studies, which recruited twenty-four individuals with nine different time points per subject. The three-stage hierarchical nonlinear mixed model is constructed. Data analysis and model performance comparisons show that GMAP converges the fastest and provides reliable results. At the mean time, data augmentation (DA) methods are used for the Random-walk Metropolis method. Data analysis shows that the speed of the convergence of Random-walk Metropolis can be improved by DA, but all of them are not as fast as GMAP. The performance of GMAP and various MCMC algorithms are compared through Midazolam data analysis and simulation.

AB - In this paper, various Bayesian Monte Carlo Markov chain (MCMC) methods and the proposed algorithm, the Gibbs maximum a posteriori (GMAP) algorithm, are compared for implementing the nonlinear mixed-effects model in pharmacokinetics (PK) studies. An intravenous two-compartmental PK model is adopted to fit the PK data from the midazolam (MDZ) studies, which recruited twenty-four individuals with nine different time points per subject. The three-stage hierarchical nonlinear mixed model is constructed. Data analysis and model performance comparisons show that GMAP converges the fastest and provides reliable results. At the mean time, data augmentation (DA) methods are used for the Random-walk Metropolis method. Data analysis shows that the speed of the convergence of Random-walk Metropolis can be improved by DA, but all of them are not as fast as GMAP. The performance of GMAP and various MCMC algorithms are compared through Midazolam data analysis and simulation.

KW - Bayesian model

KW - Gibbs Maximum a Posteriori (GMAP)

KW - Monte Carlo Markov Chain (MCMC)

KW - Pharmacokinetics (PK)

UR - http://www.scopus.com/inward/record.url?scp=70449635585&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449635585&partnerID=8YFLogxK

U2 - 10.1080/10543400902964159

DO - 10.1080/10543400902964159

M3 - Article

C2 - 20183435

AN - SCOPUS:70449635585

VL - 19

SP - 700

EP - 720

JO - Journal of Biopharmaceutical Statistics

JF - Journal of Biopharmaceutical Statistics

SN - 1054-3406

IS - 4

ER -