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 1,423 records · Page 79Linked to original sources
PubMedDate not supplied

Optimisation of the predictive ability of artificial neural network (ANN) models: a comparison of three ANN programs and four classes of training algorithm.

The purpose of this study was to determine whether artificial neural network (ANN) programs implementing different backpropagation algorithms and default settings are capable of generating equivalent highly predictive models. Three ANN packages were used: INForm, CAD/Chem and MATLAB. Twenty variants of gradient descent, conjugate gradient, quasi-Newton and Bayesian regularization algorithms were used to train networks containing a single hidden layer of 3-12 nodes. All INForm and CAD/Chem models trained satisfactorily for tensile strength, disintegration time and percentage dissolution at 15, 30, 45 and 60 min. Similarly, acceptable training was obtained for MATLAB models using Bayesian regularization. Training of MATLAB models with other algorithms was erratic. This effect was attributed to a tendency for the MATLAB implementation of the algorithms to attenuate training in local minima of the error surface. Predictive models for tablet capping and friability could not be generated. The most predictive models from each ANN package varied with respect to the optimum network architecture and training algorithm. No significant differences were found in the predictive ability of these models. It is concluded that comparable models are obtainable from different ANN programs provided that both the network architecture and training algorithm are optimised. A broad strategy for optimisation of the predictive ability of an ANN model is proposed.

Algorithms↗
PubMed2006

Simultaneous fine mapping of multiple closely linked quantitative trait Loci using combined linkage disequilibrium and linkage with a general pedigree.

Within a small region (e.g., <10 cM), there can be multiple quantitative trait loci (QTL) underlying phenotypes of a trait. Simultaneous fine mapping of closely linked QTL needs an efficient tool to remove confounded shade effects among QTL within such a small region. We propose a variance component method using combined linkage disequilibrium (LD) and linkage information and a reversible jump Markov chain Monte Carlo (MCMC) sampling for model selection. QTL identity-by-descent (IBD) coefficients between individuals are estimated by a hybrid MCMC combining the random walk and the meiosis Gibbs sampler. These coefficients are used in a mixed linear model and an empirical Bayesian procedure combines residual maximum likelihood (REML) to estimate QTL effects and a reversible jump MCMC that samples the number of QTL and the posterior QTL intensities across the tested region. Note that two MCMC processes are used, i.e., an (internal) MCMC for IBD estimation and an (external) MCMC for model selection. In a simulation study, the use of the multiple-QTL model clearly removes the shade effects between three closely linked QTL located at 1.125, 3.875, and 7.875 cM across the region of 10 cM, using 40 markers at 0.25-cM intervals. It is shown that the use of combined LD and linkage information gives much more useful information compared to using linkage information alone for both single- and multiple-QTL analyses. When using a lower marker density (11 markers at 1-cM intervals), the signal of the second QTL can disappear. Extreme values of past effective size (resulting in extreme levels of LD) decrease the mapping accuracy.

Chromosome Mapping↗
PubMed1979

Regression models of emergency medical service demand for different types of emergencies.

Second-order statistical regression models are developed for the rates of demand for different types of emergency medical services (EMS) as they relate to various socioeconomic, demographic, and other characteristics of a service area. The model parameters are estimated by the recent technique of ridge regression, which is shown to provide superior estimates to those of the ordinary least squares regression method, because of the presence of substantial multicollinearity (nonorthogonality) in the exogenous data set. The ridge results are also compared with those derived from a related Bayesian approach. The resulting models provide substantial fits to the empirical data for the EMS system of the city of Atlanta, GA.

Catchment Area, Health↗
PubMed2005

Comparative phylogenomics of the food-borne pathogen Campylobacter jejuni reveals genetic markers predictive of infection source.

Campylobacter jejuni is the predominant cause of bacterial gastroenteritis worldwide, but traditional typing methods are unable to discriminate strains from different sources that cause disease in humans. We report the use of genomotyping (whole-genome comparisons of microbes using DNA microarrays) combined with Bayesian-based algorithms to model the phylogeny of this major food-borne pathogen. In this study 111 C. jejuni strains were examined by genomotyping isolates from humans with a spectrum of C. jejuni-associated disease (70 strains), chickens (17 strains), bovines (13 strains), ovines (5 strains), and the environment (6 strains). From these data, the Bayesian phylogeny of the isolates revealed two distinct clades unequivocally supported by Bayesian probabilities (P = 1); a livestock clade comprising 31/35 (88.6%) of the livestock isolates and a "nonlivestock" clade comprising further clades of environmental isolates. Several genes were identified as characteristic of strains in the livestock clade. The most prominent was a cluster of six genes (cj1321 to cj1326) within the flagellin glycosylation locus, which were confirmed by PCR analysis as genetic markers in six additional chicken-associated strains. Surprisingly these studies show that the majority (39/70, 55.7%) of C. jejuni human isolates were found in the nonlivestock clade, suggesting that most C. jejuni infections may be from nonlivestock (and possibly nonagricultural) sources. This study has provided insight into a previously unidentified reservoir of C. jejuni infection that may have implications in disease-control strategies. The comparative phylogenomics approach described provides a robust methodological prototype that should be applicable to other microbes.

Animals↗
PubMed2006

Estimating measures of diagnostic accuracy when some covariate information is missing.

Many biomedical data sets are concerned with relating the result of screening procedure(s) for a clinical event to the occurrence of that event. The effect of risk factors on measures of accuracy such as positive predictive value and negative predictive value is of great interest for clinicians. In this paper we propose a generic approach to estimate these measures of accuracy in the setting where an explanatory model has been fitted to the joint screening and event outcome data but information on one or more risk factors in the model is not available. We refer to these as conditional rates, i.e. rates conditioned on only a subset of risk factors. We argue that, based upon the joint distribution of the event outcome, the screening result and the risk factor occurrence, a formal expression for such a rate can be obtained. This expression is a function of model parameters and thus can be estimated once the model has been fitted. Inference within the Bayesian framework is particularly attractive since simulation based model fitting straightforwardly yields samples from the posterior distribution of any conditional rate of interest. We perform a simulation study to compare these estimated conditional rates with frequently used ad hoc estimates. Differences can be substantial. We also illustrate the proposed methodology to compute conditional positive predictive value for a screening mammography data set. The proposed approach is also applicable when there are multiple diagnostic screening test outcomes.

Bayes Theorem↗
PubMed2004

Population-based pharmacokinetics of the soluble TNFr etanercept: a clinical study in 43 patients with ankylosing spondylitis compared with post hoc data from patients with rheumatoid arthritis.

OBJECTIVE: The purpose of this study was to evaluate the pharmacokinetics of etanercept in patients with ankylosing spondylitis (AS) in a phase 3 study. METHODS: Serum etanercept concentrations were analyzed from samples obtained at weeks 4 and 12 from 43 patients with AS (median age: 45 years; median body weight: 75 kg; white/non-white: 40/3; male/female: 34/9) receiving 25 mg subcutaneously twice weekly for 12 weeks. A population pharmacokinetics analysis using NONMEM was conducted to estimate individual etanercept pharmacokinetic parameters. Initially, appropriate base and covariate population pharmacokinetic models were built based on data from 10 prior clinical studies of etanercept administered subcutaneously or intravenously to healthy subjects (n = 53) and to patients with rheumatoid arthritis (RA) (n = 212). The influence of demographic characteristics on the pharmacokinetics of etanercept was thoroughly evaluated. The stability of the final model was evaluated using both internal (bootstrapping) and external (data splitting) validation approaches. Finally, the selected final population covariate model was used to estimate the Bayesian pharmacokinetic parameters for the patients with AS. RESULTS: The data from the 10 prior clinical studies were optimally fitted to a 2-compartment linear population covariate model. Both age (< 17 years) and body weight (< 60 kg) were found to be important covariates on clearance. Both bootstrapping and data splitting validated the population model. The mean Bayesian-predicted etanercept clearance and steady-state trough concentration were 0.072 l/h and 2,004 ng/ml, respectively. The pharmacokinetic parameters of etanercept in the patients with AS were similar to those observed in the patients with RA. CONCLUSIONS: The pharmacokinetics of etanercept in patients with AS were similar to those in patients with RA. The AS disease state does not appear to alter the disposition of etanercept.

Adult↗
PubMed2003

Physiologically-based pharmacokinetics and molecular pharmacodynamics of 17-(allylamino)-17-demethoxygeldanamycin and its active metabolite in tumor-bearing mice.

A whole-body physiologically-based model was developed to describe the pharmacokinetics of the ansamycin benzoquinone antibiotic 17-(allylamino)-17-demethoxygeldanamycin (17AAG) and its active metabolite 17-(amino-)-17-demethoxygeldanamycin (17AG) in blood, normal organs (lung, brain, heart, spleen, liver, kidney, skeletal muscle) and implanted human tumor xenograft in nude mice. The distribution of 17 AAG in all organs was described by diffusion-limited exchange models, while that of 17 AG was described by perfusion-limited models. The intrinsic clearances of 17AAG and 17AG in the liver were uniquely identified using local models and were estimated to be 4.93 ml/hr and 3.34 ml/hr. It was also estimated that the formation of 17AG in liver accounted for 40% of the 17AAG intrinsic clearance. The model for the distribution of both 17AAG and 17AG in the human breast cancer tumor xenograft included vascular, interstitial and intracellular compartments, which yielded the predicted cellular concentrations of 17AAG and 17AG two to three times higher than the corresponding whole tissue measurements at steady state. Estimates of the vascular-interstitial permeability surface-area product were similar for 17AAG and 17AG (0.23 ml/hr and 0.26 ml/hr). However, the interstitial to cellular transport rate of 17AG was three-fold greater than that of 17AAG, which resulted in the preferential uptake of 17AG over 17AAG in tumor. Indirect response models were developed to describe the combined action of 17AAG and 17AG on the onco-proteins Raf-1 and p185erbB2 in tumor. The half-life of endogenous protein turnover was estimated to be 22.6 hr for Raf-1 and 8.6 hr for p185erbB2, and both were comparable to corresponding values measured in vitro. A model for the molecular chaperon heat shock proteins HSP70 and HSP90 was developed based on the molecular mechanism of heat shock auto-regulation and the action of 17AAG and 17AG on these proteins. The model provided in vivo estimates of endogenous HSP70 and HSP90 turnover. In modeling pharmacokmetics and pharmacodynamics, Bayesian inference was employed to estimate the kinetic, physiological and molecular parameters when prior information was available.

Animals↗
PubMed2001

Application of a gamma model of absorption to oral cyclosporin.

BACKGROUND: Some drugs, such as cyclosporin, exhibit flat and delayed absorption profiles, with a correlation between the delay and the peak width. Such profiles can be described by an absorption model in which the absorption rate is derived from a gamma distribution (of which the classical first-order absorption model is a special case). OBJECTIVE: To develop a model for the pharmacokinetics of extravascular administration of cyclosporin and apply it to a study of the pharmacokinetics of cyclosporin microemulsion in stable renal transplant recipients. PATIENTS AND PARTICIPANTS: 21 renal transplant patients receiving oral cyclosporin microemulsion 75 to 175 mg twice daily. METHODS: The equation of the plasma concentration-time curve after oral administration was expressed as a convolution product between the absorption rate and a multi-exponential impulse response. The convolution integral was computed analytically and expressed in terms of the incomplete gamma function. Cyclosporin was assayed by liquid chromatography/mass spectrophotometry. The model was fitted by nonlinear regression, using a specially developed program. RESULTS: The gamma model yielded a good fit in all of the 21 patients studied. Attempts to fit the same data by a classical exponential with lag-time model failed in most patients. CONCLUSIONS: This model could simplify the Bayesian monitoring of cyclosporin therapy.

Area Under Curve↗
PubMed2006

How accurately can the parameters from a model of anisotropic 3He gas diffusion in lung acinar airways be estimated? Bayesian view.

In the framework of a recently proposed method for in vivo lung morphometry, acinar lung airways are considered as a set of randomly oriented cylinders covered by alveolar sleeves. Diffusion of (3)He in each airway is anisotropic and can be described by distinct longitudinal and transverse diffusion coefficients. This macroscopically isotropic but microscopically anisotropic model allows estimation of these diffusion coefficients from multi b-value MR experiments despite the airways being too small to be resolved by direct imaging. Herein a Bayesian approach is used for analyzing the uncertainties in the model parameter estimates. The approach allows evaluation of relative errors of the parameter estimates as functions of the "true" values of the parameters, the signal-to-noise ratio, the maximum b-value and the total number of b-values used in the experiment. For a given set of the "true" diffusion parameters, the uncertainty in the estimated diffusion coefficients has a minimum as a function of maximum b-value and total number of data points. Choosing the MR pulse sequence parameters corresponding to this minimum optimizes the diffusion MR experiment and gives the best possible estimates of the diffusion coefficients. The mathematical approach presented can be generalized for models containing arbitrary numbers of estimated parameters.

Anisotropy↗
PubMed2005

Bayesian projections: what are the effects of excluding data from younger age groups?

Bayesian age-period-cohort models are used increasingly to project cancer incidence and mortality rates. Data for younger age groups for which rates are low are often discarded from the analysis. The authors explored the effect of excluding these data, in terms of the precision and accuracy of projections, for selected cancer mortality data sets. Projections were made by using a generalized Bayesian age-period-cohort model. Smoothing was applied to each time scale to reduce random variation between adjacent parameter estimates. The sum of squared standardized residuals was used to assess the accuracy of projections, and 90% credible intervals were calculated to assess precision. For the data sets considered, inclusion of all age groups in the analysis provided more precise age-standardized and age-specific projections as well as more accurate age-specific projections for younger age groups. An overall improvement in the accuracy of age-standardized rates was demonstrated for males but not females, which may suggest that analysis of the full data set is beneficial when projecting cancer rates with strong cohort effects.

Adolescent↗
PubMed2006

Inferring gene regulatory networks from time series data using the minimum description length principle.

MOTIVATION: A central question in reverse engineering of genetic networks consists in determining the dependencies and regulating relationships among genes. This paper addresses the problem of inferring genetic regulatory networks from time-series gene-expression profiles. By adopting a probabilistic modeling framework compatible with the family of models represented by dynamic Bayesian networks and probabilistic Boolean networks, this paper proposes a network inference algorithm to recover not only the direct gene connectivity but also the regulating orientations. RESULTS: Based on the minimum description length principle, a novel network inference algorithm is proposed that greatly shrinks the search space for graphical solutions and achieves a good trade-off between modeling complexity and data fitting. Simulation results show that the algorithm achieves good performance in the case of synthetic networks. Compared with existing state-of-the-art results in the literature, the proposed algorithm exceptionally excels in efficiency, accuracy, robustness and scalability. Given a time-series dataset for Drosophila melanogaster, the paper proposes a genetic regulatory network involved in Drosophila's muscle development. AVAILABILITY: Available from the authors upon request.

Algorithms↗
PubMed2003

Bayesian approaches to multiple sources of evidence and uncertainty in complex cost-effectiveness modelling.

Increasingly complex models are being used to evaluate the cost-effectiveness of medical interventions. We describe the multiple sources of uncertainty that are relevant to such models, and their relation to either probabilistic or deterministic sensitivity analysis. A Bayesian approach appears natural in this context. We explore how sensitivity analysis to patient heterogeneity and parameter uncertainty can be simultaneously investigated, and illustrate the necessary computation when expected costs and benefits can be calculated in closed form, such as in discrete-time discrete-state Markov models. Information about parameters can either be expressed as a prior distribution, or derived as a posterior distribution given a generalized synthesis of available data in which multiple sources of evidence can be differentially weighted according to their assumed quality. The resulting joint posterior distributions on costs and benefits can then provide inferences on incremental cost-effectiveness, best presented as posterior distributions over net-benefit and cost-effectiveness acceptability curves. These ideas are illustrated with a detailed running example concerning the cost-effectiveness of hip prostheses in different age-sex subgroups. All computations are carried out using freely available software for conducting Markov chain Monte Carlo analysis.

Adult↗
PubMed2001

Population pharmacokinetics of carbamazepine in Singapore epileptic patients.

AIMS: To document the population pharmacokinetics of carbamazepine in patients with epilepsy living in Singapore, the majority of whom are of Chinese origin and others of minority races. METHODS: Steady-state plasma carbamazepine concentration data were gathered during routine care from various hospitals in Singapore for patients with epilepsy. Age, body weight, gender, race, formulation and concurrent medication (for other illnesses) were the fixed effects (covariates) tested simultaneously for their influence on the population mean of carbamazepine clearance, using the nonlinear mixed-effects model, in the NONMEM program. RESULTS: No age, gender, race, or formulation-related effect was found. Body weight (W), age (A) and concurrent medication with phenobarbitone (PB) emerged as the determinants of carbamazepine clearance (CL). The final regression model for carbamazepine clearance found best to describe the data was CL = 40.7 x A(0.494) x W(-1.17) x 1.44PB where CL is in l day(-1) kg(-1), A is in years, W is in kg and PB = 0 for a patient on carbamazepine only and PB = 1 for a patient on concomitant PB. The corresponding interindividual variability (CV%) in CL, described by using an exponential model, was 21.4%, and the residual error, described by using an exponential error model, was 18.2%. Predictive performance of this population covariate model was evaluated by Bayesian forecasting in a similar, but independent cohort of patients. There was no statistically significant bias between predicted and measured plasma carbamazepine concentrations. The population mean value of carbamazepine clearance obtained was similar to that previously reported for patients with a very different ethnic (Caucasians and Blacks) or geographical background (South Africa, Europe and USA). CONCLUSIONS: The derived covariate regression model reasonably predicted concentrations in the separate validation Singapore patient data set. The correlation between carbamazepine clearance and patient-specific characteristics may thus allow dosage adjustment to be made to achieve target steady-state plasma concentrations.

Adolescent↗
PubMed2000

Validation of a population pharmacokinetic model for adjunctive lamotrigine therapy in children.

AIMS: This analysis was performed to validate a previously developed population pharmacokinetic model for lamotrigine in order to establish a basis for dosage recommendations for children. METHODS: (a) The importance of the covariates in the final model was confirmed using the model validation dataset. Population and individual (Bayesian estimate) pharmacokinetic parameters were estimated using both the initial model, which included none of the covariates, and the final model. Accuracy and precision of parameter estimation and of concentration prediction were compared between the two models. (b) The performance in predicting the validation concentrations by the final model parameters obtained previously from the model development dataset was assessed. (c) The parameters of the final model were refined using a dataset combining both the development and validation data. RESULTS: Prediction performance of the final pharmacostatistical model was superior to that of the initial model. The results of the validation confirmed that concomitant antiepileptic drugs that increased or reduced lamotrigine clearance in adults had similar effects in children. The validation also verified the linear relationship between weight and clearance. The previously seen small sex effect on clearance was found statistically insignificant. CONCLUSIONS: The current analysis confirmed the previous findings. To achieve the same concentrations, children receiving enzyme-inducing antiepileptic drugs without valproate require higher doses than those receiving valproate; and heavier children require higher doses.

Adolescent↗
PubMed2002

Field triage of trauma patients: improving on the Prehospital Index.

The purpose of the present study was to evaluate the predictive ability of the Prehospital Index (PHI) in identifying injury severity and to develop a trauma triage scale that incorporates, along with the PHI, a subset of time independent variables to improve the predictive ability of the PHI-based triage instrument. This study included 1,291 trauma patients treated in Montreal, Canada. The developed trauma triage protocol was based on logistic regression analysis, in which the model that predicts the data best was selected by using Bayesian information criterion. The selected regression model included the variables age, body region injured, mechanism of injury, comorbidity, and PHI. This algorithm was a substantial improvement in detecting major versus non-major injuries (major injury defined based on death, intensive care unit admission, and surgery intervention) over the PHI alone (area under the receiver operating characteristic curve: 0.76 v 0.66, P <.05). Considering time independent variables could lead to better injury triage decisions.

Adolescent↗
PubMed2006

Strength of evidence for density dependence in abundance time series of 1198 species.

Population limitation is a fundamental tenet of ecology, but the relative roles of exogenous and endogenous mechanisms remain unquantified for most species. Here we used multi-model inference (MMI), a form of model averaging, based on information theory (Akaike's Information Criterion) to evaluate the relative strength of evidence for density-dependent and density-independent population dynamical models in long-term abundance time series of 1198 species. We also compared the MMI results to more classic methods for detecting density dependence: Neyman-Pearson hypothesis-testing and best-model selection using the Bayesian Information Criterion or cross-validation. Using MMI on our large database, we show that density dependence is a pervasive feature of population dynamics (median MMI support for density dependence = 74.7-92.2%), and that this holds across widely different taxa. The weight of evidence for density dependence varied among species but increased consistently, with the number of generations monitored. Best-model selection methods yielded similar results to MMI (a density-dependent model was favored in 66.2-93.9% of species time series), while the hypothesis-testing methods detected density dependence less frequently (32.6-49.8%). There were no obvious differences in the prevalence of density dependence across major taxonomic groups under any of the statistical methods used. These results underscore the value of using multiple modes of analysis to quantify the relative empirical support for a set of working hypotheses that encompass a range of realistic population dynamical behaviors.

Ecosystem↗
PubMed2002

Measuring uncertainty in complex decision analysis models.

Prediction models used in support of clinical and health policy decision making often need to consider the course of a disease over an extended period of time, and draw evidence from a broad knowledge base, including epidemiologic cohort and case control studies, randomized clinical trials, expert opinions, and more. This paper is a brief introduction to these complex decision models, their relation to Bayesian decision theory, and the tools typically used to describe the uncertainties involved. Concepts are illustrated throughout via a simplified tutorial.

Bayes Theorem↗
PubMed1998

A population pharmacokinetic model of cyclosporine in the early postoperative phase in patients with liver transplants, and its predictive performance with Bayesian fitting.

The availability of personal computer programs to individualize drug regimens has stimulated interest in modeling population pharmacokinetics. This study used the NPEM2 software to determine cyclosporine population pharmacokinetic parameter values and distributions in a first group of 25 recipients of liver transplants during their first postoperative week. On a second group of 25 patients, the authors used these values to evaluate Bayesian predictive performance of cyclosporine blood concentrations with the USC*PACK PC program. During the study period, all the patients have been treated by continuous intravenous infusion. The one-compartment model pharmacokinetic parameter-the slope of volume to body weight (Vs) and the elimination rate constant (Kel) values found (mean values: Vs = 2.177 l/kg, Kel = 0.235 h(-1); median values: Vs = 1.559 l/kg, Kel = 0.163 h(-1); the percent coefficient of variation (Vs = 92%, Kel = 79%) appear reasonable and show the ability of NPEM2 to deal with sparse data. When the predictions were studied with day 1, day 2, or day 3 concentrations, predictive bias was respectively -0.030, -0.013, and 0.013 microg/ml, suggesting a greater clearance of cyclosporine immediately after surgery, the clearance decreasing in the days after. With the first three blood levels and the Bayesian fitting procedure, it was possible to predict at least half the subsequent measured blood levels of each patient accurately (within 20%) in more than three-quarters (76%) of the second group of recipients of transplants, and for 40% of patients the authors obtained accurate predictions in 100% of the subsequent blood levels. For a few patients (12%) they found quite poor predictions. The reason for this is unclear. The results suggest that this population model and the Bayesian fitting procedure using two or three blood levels can be reasonably and carefully used to control, in real time, cyclosporine blood levels in a majority of new patients with liver transplants.

Adolescent↗