Biomedical subjects

Pulak Ghosh

Publications and source records attributed to Pulak Ghosh.

4 recordsLinked to original sources

PubMed2007

Random changepoint modelling of HIV immunologic responses.

We propose a changepoint model for the analysis of longitudinal CD4 T-cell counts for HIV infected subjects following highly active antiretroviral treatment. The profile of CD4 counts for each subject follows a simple, 'broken stick' changepoint model, with random subject-specific parameters, including the changepoint. The model accounts for baseline covariates. The longitudinal CD4 records are censored at the time of the subject going off-study-treatment. This is a potentially informative drop-out mechanism, which we address by modelling it jointly with the CD4 count outcome. The drop-out model incorporates terms from the CD4 model, including the changepoint. The estimation is done in a Bayesian framework, with implementation via Markov chain Monte Carlo methods in the WinBUGS software. Model selection using DIC indicates that the data support the complex random changepoint and informative censoring model.

Antiretroviral Therapy, Highly Active↗

PubMed2007

A semi-parametric Bayesian approach to average bioequivalence.

Bioequivalence assessment is an issue of great interest. Development of statistical methods for assessing bioequivalence is an important area of research for statisticians. Bioequivalence is usually determined based on the normal distribution. We relax this assumption and develop a semi-parametric mixed model for bioequivalence data. The proposed method is quite flexible and practically meaningful. Our proposed method is based on a mixture normal distribution and a non-parametric Bayesian approach using a Dirichlet process mixture prior. A numerical example illustrates the use of our procedure.

Bayes Theorem↗

PubMed2007

Bivariate random effect model using skew-normal distribution with application to HIV-RNA.

Correlated data arise in a longitudinal studies from epidemiological and clinical research. Random effects models are commonly used to model correlated data. Mostly in the longitudinal data setting we assume that the random effects and within subject errors are normally distributed. However, the normality assumption may not always give robust results, particularly if the data exhibit skewness. In this paper, we develop a Bayesian approach to bivariate mixed model and relax the normality assumption by using a multivariate skew-normal distribution. Specifically, we compare various potential models and illustrate the procedure using a real data set from HIV study.

Bayes Theorem↗

PubMed2003

Bayesian approach to average bioequivalence using Bayes' factor.

In recent years, bioavailability studies for assessment of bioequivalence between two or more drug formulations have become very popular in drug development. However the current practice for the assessment of bioequivalence suffers from certain serious drawbacks. For example, sometimes these tests fail to control the consumer's risk. In this article a new methodology based on the application of Bayes' factor for solving the average bioequivalence problem is proposed. We compare our approach with the existing methods by using real data sets from the Food and Drug Administration (FDA). Results are further explored using simulation studies.

Bayes Theorem↗