Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Bayesian modelling”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,315 records · Page 73Linked to original sources
PubMed2005

Creative and innovative statistics in clinical research and development.

OBJECTIVES: The aim of this paper is to show that even in a highly regulated area such as clinical research and development in pharmaceutical industry, there are needs and ample opportunities for statisticians and other medical informatics professionals to further creatively develop and implement methods in order to support the collection, analysis and interpretation of clinical data. METHODS: The recently published "Critical Path" initiative of the US Food and Drug Administration discusses the decline in new drug submissions in the last decade and illustrates potential causes in the present clinical development process. Areas where statisticians can and have begun to look for new innovative ways to overcome these shortcomings are presented and examples of such novel approaches that have been developed by statistical methodologists in the pharmaceutical industry together with statisticians in academia are given. RESULTS: In Early Development, i.e., in the first studies in man with a new compound, a combination of Bayesian methods and modeling approaches is particularly promising to increase the efficiency of decision making whereas in later phases (IIb and III) a marriage of modeling and classical frequentist approaches together with novel adaptive designs is expected help to chose the right dose regimen and to perform the trials more efficiently in reduced time. CONCLUSIONS: The combination of known statistical methods and thinking and the development of new approaches are in line with the present paradigm of "learning and confirming" in regulated clinical development while increasing the efficiency of both.

Bayes Theorem↗
PubMed2005

Probability distribution on the parse trees.

Paper describes probability distribution on the parse trees of natural language by using Bayesian networks. First parts of the paper describes probabilistic context-free grammar and parse trees. In the second part of the paper, Bayesian network was modelled and joint probability distribution on their vertex. On these theoretical ideas, in the third part, we describe our model tested on Database of grammatical sentences of Croatian language (http:/ /infoz.ffzg.hr / tepes /). At the end was presented a backward procedure and evaluation of our results.

Anthropology↗
PubMedDate not supplied

Evaluation of theophylline clearance in children with bronchial asthma.

The total body clearance of theophylline was examined in 95 children aged 6 months to 15 years with chronic bronchial asthma. Each patient received sustained-release theophylline preparations and was evaluated at steady state. Blood samples were obtained 1-8 hours after a morning dose, and individual total body clearance was determined with a Bayesian forecasting computer model. Changes in theophylline clearance were evaluated within age groups, with the age determined at the time of blood sampling. Total body clearance tended to increase with age up to 10 years. In children over 10 years old, total body clearance significantly decreased with age. The interpatient variability of theophylline clearance was most noticeable at 2-4 years of age. Changes in volume of distribution were similar to those in clearance. In conclusion, there are pronounced age-dependent differences in theophylline clearance that require individual adjustment of dosage regimens on the basis of serum concentrations and clinical response in children. Individualization of the oral dose based on frequent serum measurements is necessary to maintain theophylline levels in the therapeutic range.

Adolescent↗
PubMed1995

Pharmacokinetic development of quinolone antibiotics.

A prerequisite for the pharmacokinetic development of quinolone antibiotics is a sensitive and accurate method for the quantification of the drug in biological fluids. Both, a drug specific (e.g. HPLC) and a drug non-specific but effect related assay (e.g. bioassay) should be used during early clinical development to detect major active metabolites. The basic pharmacokinetic behavior of the drug is investigated as part of the early phase I program, where single and multiple ascending dose studies are performed to characterize the safety and tolerability of the quinolone in healthy volunteers. Further pharmacokinetic studies are performed to describe the absolute bioavailability, dose proportionality, pharmacokinetics in young and elderly, male and female volunteers. The suitability of the clinical dosage from must be evaluated in comparison to an oral solution and by quantification of the effect of food on bioavailability. The characterization of the absorption in different parts of the gastrointestinal tract may be valuable for dosage form optimization. In order to start phase IIb clinical trials, the potential of possible drug-drug interactions with antacids, cimetidine, theophylline and warfarin has to be evaluated. This can be done by in vitro and in vivo preclinical experiments, before formal clinical-pharmacology studies are performed. Further pharmacokinetic characterization (e.g. studies in special subpopulation, extended interaction studies, total recovery using 14C-labelled compound, blister fluid penetration) will be done parallel to the phase II/III development program. During these efficacy and safety trials blood samples should be obtained and PK-parameters can be calculated using sparse data analysis methods like non-linear mixed effect modeling (NONMEM) or Bayesian methods to characterize the pharmacokinetics in the target population.

4-Quinolones↗
PubMed2006

A Bayesian toolkit for genetic association studies.

We present a range of modelling components designed to facilitate Bayesian analysis of genetic-association-study data. A key feature of our approach is the ability to combine different submodels together, almost arbitrarily, for dealing with the complexities of real data. In particular, we propose various techniques for selecting the "best" subset of genetic predictors for a specific phenotype (or set of phenotypes). At the same time, we may control for complex, non-linear relationships between phenotypes and additional (non-genetic) covariates as well as accounting for any residual correlation that exists among multiple phenotypes. Both of these additional modelling components are shown to potentially aid in detecting the underlying genetic signal. We may also account for uncertainty regarding missing genotype data. Indeed, at the heart of our approach is a novel method for reconstructing unobserved haplotypes and/or inferring the values of missing genotypes. This can be deployed independently or, alternatively, it can be fully integrated into arbitrary genotype- or haplotype-based association models such that the missing data and the association model are "estimated" simultaneously. The impact of such simultaneous analysis on inferences drawn from the association model is shown to be potentially significant. Our modelling components are packaged as an "add-on" interface to the widely used WinBUGS software, which allows Markov chain Monte Carlo analysis of a wide range of statistical models. We illustrate their use with a series of increasingly complex analyses conducted on simulated data based on a real pharmacogenetic example.

Bayes Theorem↗
PubMed2006

Disease cluster detection: a critique and a Bayesian proposal.

This paper reviews issues in the analysis of non-focussed clustering, and proposes a novel approach to cluster modelling that can be used in a surveillance context. The novel approach involves the use of local likelihood models for the analysis of clustering in small area health data. Local likelihood is used when interdependence between data events at locations is modelled directly, as opposed to the modelling of a hidden process of cluster centres. This approach allows the use of conventional posterior sampling. It also allows a less parameterized approach to the form of clusters detected. The idea of a spatially dependent lasso which provides the local maxima for the aggregation of locations is considered as an approximation. The methods are applied to a well known data set and compared with Satscan, and a conditional logistic Bayesian model.

Air Pollutants↗
PubMed2006

Phylogeny of sipunculan worms: A combined analysis of four gene regions and morphology.

The intra-phyletic relationships of sipunculan worms were analyzed based on DNA sequence data from four gene regions and 58 morphological characters. Initially we analyzed the data under direct optimization using parsimony as optimality criterion. An implied alignment resulting from the direct optimization analysis was subsequently utilized to perform a Bayesian analysis with mixed models for the different data partitions. For this we applied a doublet model for the stem regions of the 18S rRNA. Both analyses support monophyly of Sipuncula and most of the same clades within the phylum. The analyses differ with respect to the relationships among the major groups but whereas the deep nodes in the direct optimization analysis generally show low jackknife support, they are supported by 100% posterior probability in the Bayesian analysis. Direct optimization has been useful for handling sequences of unequal length and generating conservative phylogenetic hypotheses whereas the Bayesian analysis under mixed models provided high resolution in the basal nodes of the tree.

Animals↗
PubMed1994

Suramin: development of a population pharmacokinetic model and its use with intermittent short infusions to control plasma drug concentration in patients with prostate cancer.

PURPOSE: This study aimed to (1) develop a population pharmacokinetic model for suramin; (2) use Bayesian methods to assess suramin pharmacokinetics in individual patients; (3) use individual patients' pharmacokinetic parameter estimates to individualize suramin dose and schedule and maintain plasma suramin concentrations within predetermined target ranges; and (4) assess the feasibility of outpatient administration of suramin by intermittent, short infusions. METHODS: Plasma suramin concentrations were measured by high-performance liquid chromatography (HPLC), and compartmental pharmacokinetic models were fit using a Bayesian algorithm. Population pharmacokinetic models were developed using an iterative two-stage approach. Estimates of each patient's central-compartment volume were used to calculate suramin dosage. Simulation of that patient's suramin clearance was used to predict the time of his next dose. Using this approach, plasma suramin concentration was maintained at between 200 and 300, 175 and 275, 150 and 250, or 100 and 200 microgram/mL in four sequential patient cohorts. The ability of two- and three-compartment, open, linear models to fit the pharmacokinetic data was compared. Population pharmacokinetic parameters were estimated, using both two- and three-compartment structural models in 69 hormone-refractory prostate cancer patients. RESULTS: Target plasma suramin concentrations in individual patients were rapidly achieved. Concentrations were maintained within desired ranges for > or = 85% of treatment duration in all cohorts. A three-compartment, open, linear model described suramin pharmacokinetics better than did a two-compartment, open, linear model. Population pharmacokinetic estimates generated for two- and three-compartment pharmacokinetic models demonstrated modest interpatient pharmacokinetic variability and the long terminal half-life of suramin. CONCLUSION: Suramin can be administered by intermittent short infusion. Adaptive-control-with-feedback dosing facilitated precise control of plasma suramin concentrations and allowed a number of different concentration ranges to be studied. This approach is expensive and labor-intensive. Although we have demonstrated the ability to control drug exposure, simpler dosing schedules require critical evaluation. Population pharmacokinetic parameters generated in men with hormone-refractory prostate cancer will facilitate rational design of such schedules.

Adult↗
PubMed2004

Molecular evolution and mosaicism of leptospiral outer membrane proteins involves horizontal DNA transfer.

Leptospires belong to a genus of parasitic bacterial spirochetes that have adapted to a broad range of mammalian hosts. Mechanisms of leptospiral molecular evolution were explored by sequence analysis of four genes shared by 38 strains belonging to the core group of pathogenic Leptospira species: L. interrogans, L. kirschneri, L. noguchii, L. borgpetersenii, L. santarosai, and L. weilii. The 16S rRNA and lipL32 genes were highly conserved, and the lipL41 and ompL1 genes were significantly more variable. Synonymous substitutions are distributed throughout the ompL1 gene, whereas nonsynonymous substitutions are clustered in four variable regions encoding surface loops. While phylogenetic trees for the 16S, lipL32, and lipL41 genes were relatively stable, 8 of 38 (20%) ompL1 sequences had mosaic compositions consistent with horizontal transfer of DNA between related bacterial species. A novel Bayesian multiple change point model was used to identify the most likely sites of recombination and to determine the phylogenetic relatedness of the segments of the mosaic ompL1 genes. Segments of the mosaic ompL1 genes encoding two of the surface-exposed loops were likely acquired by horizontal transfer from a peregrine allele of unknown ancestry. Identification of the most likely sites of recombination with the Bayesian multiple change point model, an approach which has not previously been applied to prokaryotic gene sequence analysis, serves as a model for future studies of recombination in molecular evolution of genes.

Amino Acid Sequence↗
PubMed2006

The Bayesian reader: explaining word recognition as an optimal Bayesian decision process.

This article presents a theory of visual word recognition that assumes that, in the tasks of word identification, lexical decision, and semantic categorization, human readers behave as optimal Bayesian decision makers. This leads to the development of a computational model of word recognition, the Bayesian reader. The Bayesian reader successfully simulates some of the most significant data on human reading. The model accounts for the nature of the function relating word frequency to reaction time and identification threshold, the effects of neighborhood density and its interaction with frequency, and the variation in the pattern of neighborhood density effects seen in different experimental tasks. Both the general behavior of the model and the way the model predicts different patterns of results in different tasks follow entirely from the assumption that human readers approximate optimal Bayesian decision makers.

Attention↗
PubMed2004

A Bayesian approach to tracking patients having changing pharmacokinetic parameters.

This paper considers the updating of Bayesian posterior densities for pharmacokinetic models associated with patients having changing parameter values. For estimation purposes it is proposed to use the Interacting Multiple Model (IMM) estimation algorithm, which is currently a popular algorithm in the aerospace community for tracking maneuvering targets. The IMM algorithm is described, and compared to the multiple model (MM) and Maximum A-Posteriori (MAP) Bayesian estimation methods, which are presently used for posterior updating when pharmacokinetic parameters do not change. Both the MM and MAP Bayesian estimation methods are used in their sequential forms, to facilitate tracking of changing parameters. Results indicate that the IMM algorithm is well suited for tracking time-varying pharmacokinetic parameters in acutely ill and unstable patients, incurring only about half of the integrated error compared to the sequential MM and MAP methods on the same example.

Algorithms↗
PubMed2005

Modeling of farnesyltransferase inhibition by some thiol and non-thiol peptidomimetic inhibitors using genetic neural networks and RDF approaches.

Inhibition of farnesyltransferase (FT) enzyme by a set of 78 thiol and non-thiol peptidomimetic inhibitors was successfully modeled by a genetic neural network (GNN) approach, using radial distribution function descriptors. A linear model was unable to successfully fit the whole data set; however, the optimum Bayesian regularized neural network model described about 87% inhibitory activity variance with a relevant predictive power measured by q2 values of leave-one-out and leave-group-out cross-validations of about 0.7. According to their activity levels, thiol and non-thiol inhibitors were well-distributed in a topological map, built with the inputs of the optimum non-linear predictor. Furthermore, descriptors in the GNN model suggested the occurrence of a strong dependence of FT inhibition on the molecular shape and size rather than on electronegativity or polarizability characteristics of the studied compounds.

Enzyme Inhibitors↗
PubMed2002

Semiparametric regression modeling with mixtures of Berkson and classical error, with application to fallout from the Nevada test site.

We construct Bayesian methods for semiparametric modeling of a monotonic regression function when the predictors are measured with classical error. Berkson error, or a mixture of the two. Such methods require a distribution for the unobserved (latent) predictor, a distribution we also model semiparametrically. Such combinations of semiparametric methods for the dose response as well as the latent variable distribution have not been considered in the measurement error literature for any form of measurement error. In addition, our methods represent a new approach to those problems where the measurement error combines Berkson and classical components. While the methods are general, we develop them around a specific application, namely, the study of thyroid disease in relation to radiation fallout from the Nevada test site. We use this data to illustrate our methods, which suggest a point estimate (posterior mean) of relative risk at high doses nearly double that of previous analyses but that also suggest much greater uncertainty in the relative risk.

Bayes Theorem↗
PubMed2005

Active concept learning in image databases.

Concept learning in content-based image retrieval systems is a challenging task. This paper presents an active concept learning approach based on the mixture model to deal with the two basic aspects of a database system: the changing (image insertion or removal) nature of a database and user queries. To achieve concept learning, we a) propose a new user directed semi-supervised expectation-maximization algorithm for mixture parameter estimation, and b) develop a novel model selection method based on Bayesian analysis that evaluates the consistency of hypothesized models with the available information. The analysis of exploitation versus exploration in the search space helps to find the optimal model efficiently. Our concept knowledge transduction approach is able to deal with the cases of image insertion and query images being outside the database. The system handles the situation where users may mislabel images during relevance feedback. Experimental results on Corel database show the efficacy of our active concept learning approach and the improvement in retrieval performance by concept transduction.

Algorithms↗
PubMed2006

Spatial analysis and mapping of malaria risk in Malawi using point-referenced prevalence of infection data.

BACKGROUND: Current malaria control initiatives aim at reducing malaria burden by half by the year 2010. Effective control requires evidence-based utilisation of resources. Characterizing spatial patterns of risk, through maps, is an important tool to guide control programmes. To this end an analysis was carried out to predict and map malaria risk in Malawi using empirical data with the aim of identifying areas where greatest effort should be focussed. METHODS: Point-referenced prevalence of infection data for children aged 1-10 years were collected from published and grey literature and geo-referenced. The model-based geostatistical methods were applied to analyze and predict malaria risk in areas where data were not observed. Topographical and climatic covariates were added in the model for risk assessment and improved prediction. A Bayesian approach was used for model fitting and prediction. RESULTS: Bivariate models showed a significant association of malaria risk with elevation, annual maximum temperature, rainfall and potential evapotranspiration (PET). However in the prediction model, the spatial distribution of malaria risk was associated with elevation, and marginally with maximum temperature and PET. The resulting map broadly agreed with expert opinion about the variation of risk in the country, and further showed marked variation even at local level. High risk areas were in the low-lying lake shore regions, while low risk was along the highlands in the country. CONCLUSION: The map provided an initial description of the geographic variation of malaria risk in Malawi, and might help in the choice and design of interventions, which is crucial for reducing the burden of malaria in Malawi.

Cluster Analysis↗
PubMed2006

Building statistical models to analyze species distributions.

Models of the geographic distributions of species have wide application in ecology. But the nonspatial, single-level, regression models that ecologists have often employed do not deal with problems of irregular sampling intensity or spatial dependence, and do not adequately quantify uncertainty. We show here how to build statistical models that can handle these features of spatial prediction and provide richer, more powerful inference about species niche relations, distributions, and the effects of human disturbance. We begin with a familiar generalized linear model and build in additional features, including spatial random effects and hierarchical levels. Since these models are fully specified statistical models, we show that it is possible to add complexity without sacrificing interpretability. This step-by-step approach, together with attached code that implements a simple, spatially explicit, regression model, is structured to facilitate self-teaching. All models are developed in a Bayesian framework. We assess the performance of the models by using them to predict the distributions of two plant species (Proteaceae) from South Africa's Cape Floristic Region. We demonstrate that making distribution models spatially explicit can be essential for accurately characterizing the environmental response of species, predicting their probability of occurrence, and assessing uncertainty in the model results. Adding hierarchical levels to the models has further advantages in allowing human transformation of the landscape to be taken into account, as well as additional features of the sampling process.

Bayes Theorem↗
PubMed2004

Estimating the demand for health care with panel data: a semiparametric Bayesian approach.

This paper is concerned with the problem of estimating the demand for health care with panel data. A random effects model is specified within a semiparametric Bayesian approach using a Dirichlet process prior. This results in a very flexible distribution for both the random effects and the count variable. In particular, the model can be seen as a mixture distribution with a random number of components, and is therefore a natural extension of prevailing latent class models. A full Bayesian analysis using Markov chain Monte Carlo simulation methods is proposed. The methodology is illustrated with an application using data from Germany.

Aged↗
PubMed2002

Physiologically based modeling of the inhalation kinetics of styrene in humans using a bayesian population approach.

Animal studies have implicated styrene as toxic to the central nervous system and its major metabolite styrene-7,8-oxide as a carcinogen. Therefore, a reliable estimate of the metabolic capacity for styrene in humans is of interest. However, the available models describing styrene kinetics in humans lack rigorous statistical validation and also ignore the population variability in metabolism. The population variability may be estimated by the use of population models. Furthermore, the statistical validation of pharmacokinetic models may be improved by use of Bayesian methods. These two approaches may be combined and recently have been gaining interest in the toxicology literature. A population-based physiologically based pharmacokinetic (PBPK) model for styrene was developed. The model was calibrated to extensive human toxicokinetic data from three previous studies in which 24 volunteers were exposed to 50-386 ppm of styrene at rest and various levels of exercise. Model fitting was performed in a Bayesian framework using Markov chain Monte Carlo simulation. The uncertainty around the partition coefficients and metabolic parameters for styrene was reduced. The metabolic capacity for styrene in humans was estimated to be 0.92 micromol/l kg(-1), with a lognormal standard deviation of 1.66. The estimated Vmax is 40% higher than previously estimated, whereas the population standard deviation is estimated for the first time.

Adolescent↗