PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Bayesian modelling”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 145 records · Page 8Linked to original sources
PubMed1987

Disposition of phenytoin in critically ill trauma patients.

Estimates of phenytoin pharmacokinetic variables and protein binding were determined in 10 adult critically ill trauma patients. Each study subject received phenytoin sodium as an intravenous loading dose of 15 mg/kg, followed by an initial intravenous maintenance dose of 6 mg/kg/day. Serial blood samples were obtained throughout the seven-day study period and analyzed for total and unbound serum phenytoin concentrations. The concentration data for each patients were fitted to a one-compartment model with elimination defined by the Michaelis-Menten constant Km and the maximum rate of metabolism (Vmax) and to a one-compartment model with first-order elimination. The Michaelis-Menten model used Bayesian parameter estimation while the linear model used weighted non-linear least-squares regression analysis. Unbound phenytoin fraction ranged from 0.073 to 0.25. Free fraction increased 7% to 108% in 9 of 10 patients (median increase 29%) from day 1 to day 7 of therapy. Variable estimates using the Michaelis-Menten model were as follows: volume of distribution, 0.76 +/- 0.15 L/kg (0.58-1.01 L/kg); Vmax, 568 +/- 197 mg/day (350-937 mg/day); and Km, 4.5 +/- 1.8 mg/L (1.8-6.2 mg/L). These estimates fell within the wide range of values obtained in studies using stable patients or healthy volunteers. The Michaelis-Menten model was significantly less biased and more precise than the linear model. Three of four patients who continued to receive their study maintenance dose had substantially lower measured total serum concentrations of phenytoin than predicted using the study variable estimates.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMedDate not supplied

Lansoprazole compared with histamine2-receptor antagonists in healing gastric ulcers: a meta-analysis.

To compare the gastric ulcer healing rates of lansoprazole with histamine2-receptor antagonists (H2RAs) (ranitidine, famotidine, cimetidine, and roxatidine), a meta-analysis was performed using data from five published and eight unpublished randomized controlled trials. Analyses were performed using (1) both evaluable patients (n = 1527) and all randomized patients (n = 1655) (assuming that patients lost to follow-up were treatment failures); (2) all studies and a subset of studies that received high methodologic quality scores; and (3) fixed-effects, random-effects, and Bayesian statistical models. In all cases, lansoprazole was associated with a significantly higher rate of endoscopic healing at both 4 and 8 weeks compared with the H2RAs. When the most conservative Bayesian statistical model and intent-to-treat analysis were used, lansoprazole was associated with a 33% higher healing rate at 4 weeks (risk ratio = 1.33; 95% confidence interval [CI] = 1.19 to 1.49) and a 12% higher healing rate at 8 weeks (risk ratio = 1.12; 95% CI = 1.06 to 1.19) than were the H2RA agents. Similar results were obtained when the meta-analysis was performed on evaluable rather than all randomized patients and using the three different analytical techniques noted above. Slightly lower, though still highly significant, improvement in ulcer healing rates was obtained when the meta-analysis was performed using a subset of six studies that received high methodologic quality scores. These results support the conclusion that lansoprazole heals ulcers more quickly than do the H2RAs and also achieves higher overall rates of healing. The eradication of Helicobacter pylori associated with gastric ulcers was not assessed in individual studies.

2-Pyridinylmethylsulfinylbenzimidazoles↗
PubMed1999

Bayesian inference on biopolymer models.

MOTIVATION: Most existing bioinformatics methods are limited to making point estimates of one variable, e.g. the optimal alignment, with fixed input values for all other variables, e.g. gap penalties and scoring matrices. While the requirement to specify parameters remains one of the more vexing issues in bioinformatics, it is a reflection of a larger issue: the need to broaden the view on statistical inference in bioinformatics. RESULTS: The assignment of probabilities for all possible values of all unknown variables in a problem in the form of a posterior distribution is the goal of Bayesian inference. Here we show how this goal can be achieved for most bioinformatics methods that use dynamic programming. Specifically, a tutorial style description of a Bayesian inference procedure for segmentation of a sequence based on the heterogeneity in its composition is given. In addition, full Bayesian inference algorithms for sequence alignment are described. AVAILABILITY: Software and a set of transparencies for a tutorial describing these ideas are available at http://www.wadsworth.org/res&res/bioinfo/

Bayes Theorem↗
PubMed1998

Bayesian decision procedures based on logistic regression models for dose-finding studies.

Early-phase clinical trials, conducted to determine the appropriate dose of an experimental drug to take forward to later trials, are considered. The objective is to find the dose associated with some low probability of an adverse event. A Bayesian model is presented, and a decision-theoretic procedure for finding the optimal doses for each of a series of cohorts of subjects is derived. The procedure is flexible and can easily be conducted using standard statistical software. The results of simulations investigating the properties of the procedure are presented.

Bayes Theorem↗
PubMed1996

A comparison of three techniques for rapid model development: an application in patient risk-stratification.

Accurately risk-stratifying patients is a key component of health care outcomes assessment. And, many health care organizations increasingly are relying upon automated means for assistance in making patient risk-stratification decisions. Unfortunately, the process of outcome model development, as it is currently practiced, is both time consuming and difficult. We investigated the relative abilities of three modeling techniques (logistic regression, artificial neural network (ANN), and Bayesian) to rapidly develop models for risk-stratifying patients. Our results demonstrated that all three modeling techniques perform equally well in certain situations. However, the Bayesian model with conditional independence had the best overall performance. Unfortunately, none of the models were able to achieve the degree of accuracy which would be required in a medical setting.

APACHE↗
PubMed1998

Issues in Bayesian Analysis of Neural Network Models

Stemming from work by Buntine and Weigend (1991) and MacKay (1992), there is a growing interest in Bayesian analysis of neural network models. Although conceptually simple, this problem is computationally involved. We suggest a very efficient Markov chain Monte Carlo scheme for inference and prediction with fixed&hyphenarchitecture feedforward neural networks. The scheme is then extended to the variable architecture case, providing a data&hyphendriven procedure to identify sensible architectures.

Journal Article↗
PubMed1996

Validity of Bayesian forecasting programme in therapeutic drug monitoring of vancomycin in a surgical intensive care unit: a prospective evaluation.

The objectives of this study were: (a) to assess the predictive performance of a 2-compartment Bayesian vancomycin forecasting programme in critically ill patients in the surgical intensive care unit, and (b) to show the applicability of the programme, which is based on parameters derived from Western population, in our local Asian population. Twenty critically ill patients were enrolled into the study programme. All patients received multiple-dose vancomycin for infections due to methicillin-resistant staphylococcus aureus (MRSA). The patients dosage regimen were optimised by entering a set of peak and trough vancomycin serum levels into a clinical computer program (MB; USC*PACK PC collection; University of Southern California, USA) by utilising a 2-compartment Bayesian population model. The correlation between observed and predicted serum peak and trough concentrations were evaluated for both the non-fitted and fitted models by linear regression analysis. There was a significant correlation between observed and predicted concentrations using the fitted model (r = 0.97, P < 0.05). There was no significant correlation of these concentrations in the non-fitted model (r = 0.8). This study shows that the Bayesian programme is able to accurately predict future vancomycin concentrations in our local Asian population. It is possible to optimise patients dosage regimens with the knowledge of two concentrations of vancomycin in order to achieve targeted therapeutic goals.

Aged↗
PubMed1985

A Bayesian approach to nonlinear random effects models.

Nonlinear random effects models are considered from the Bayesian point of view. The method of analysis follows closely that of Lindley and Smith (1972, Journal of the Royal Statistical Society, Series B 34, 1-42). The numerical method is related to the EM algorithm.

Analysis of Variance↗
PubMed1997

A flexible approach to time-varying coefficients in the Cox regression setting.

Research on methods for studying time-to-event data (survival analysis) has been extensive in recent years. The basic model in use today represents the hazard function for an individual through a proportional hazards model (Cox, 1972). Typically, it is assumed that a covariate's effect on the hazard function is constant throughout the course of the study. In this paper we propose a method to allow for possible deviations from the standard Cox model, by allowing the effect of a covariate to vary over time. This method is based on a dynamic linear model. We present our method in terms of a Bayesian hierarchical model. We fit the model to the data using Markov chain Monte Carlo methods. Finally, we illustrate the approach with several examples.

Bayes Theorem↗
PubMed1997

Large sample Bayesian inference on the parameters of the proportional hazard models.

This paper considers large sample Bayesian analysis of the proportional hazards model when interest is in inference on the parameters and estimation of the log relative risk for specified covariate vectors rather than on prediction of the survival function. We use a normal prior distribution for the parameters and make inferences based on the derived posterior distribution. The suggested approach is much simpler than alternative Bayesian analyses previously suggested for the proportional hazards models. Using simulated data we compare estimates obtained from the Bayesian analysis with those obtained from the full proportional hazards model and the reduced model after backwards elimination. We show that under a wider range of assumptions, the Bayesian analysis provides reduced estimation errors and improved rejection of noise variables. Finally, we illustrate the methodology using data from a large study of prognostic markers in breast cancer.

Bayes Theorem↗
PubMed1994

The hierarchical Bayesian approach to population pharmacokinetic modelling.

Compartmental models are widely used to model the profile of drug concentrations versus time from administration in an individual subject. Observed concentrations are then modelled as noisy departures from the underlying profile, the latter characterised for each individual by a small number of 'individual parameters'. When a population of individuals is studied, inter-individual variation is modelled by assuming that the individual profile parameters are drawn from a population distribution, the latter characterised by 'population parameters' describing, in effect, a mean population profile and individual variation around it. From a Bayesian statistical perspective, such models fit exactly into the so-called hierarchical modelling framework, which provides a coherent basis for individual and population inferences and prediction, as well as for decision-making (for example, the design of dosage regimens). This paper outlines the hierarchical model framework and describes how the required computations can be carried out in a straightforward manner by a Markov chain Monte Carlo technique known as Gibbs sampling, even when models involve mean-variance relationships and outliers.

Bayes Theorem↗
PubMed1996

A Bayesian approach to Weibull survival models--application to a cancer clinical trial.

In this paper we outline a class of fully parametric proportional hazards models, in which the baseline hazard is assumed to be a power transform of the time scale, corresponding to assuming that survival times follow a Weibull distribution. Such a class of models allows for the possibility of time varying hazard rates, but assumes a constant hazard ratio. We outline how Bayesian inference proceeds for such a class of models using asymptotic approximations which require only the ability to maximize the joint log posterior density. We apply these models to a clinical trial to assess the efficacy of neutron therapy compared to conventional treatment for patients with tumours of the pelvic region. In this trial there was prior information about the log hazard ratio both in terms of elicited clinical beliefs and the results of previous studies. Finally, we consider a number of extensions to this class of models, in particular the use of alternative baseline functions, and the extension to multi-state data.

Bayes Theorem↗
PubMed1997

Late potential recognition by artificial neural networks.

Ventricular late potentials (LP's) are high-frequency low-amplitude signals obtained from signal-averaged electrocardiograms (ECG's) [SAECG's]. LP's are useful in identifying patients prone to ventricular tachycardia (VT), spontaneous or inducible during electrophysiology testing. A combination of self-organizing and supervised artificial neural network (ANN) models was developed to identify patients with a positive electrophysiology (PEP) test for inducible ventricular tachycardia from patients with a negative electrophysiology (NEP) test using LP's. We have added morphology information of vector magnitude waveform to original set of three time-domain features of LP's, which are total QRS duration (TQRSD), high-frequency low-amplitude signal duration (HFLAD), and root-mean-square voltage (RMSV). Pattern recognition results from an ANN model with this combination feature set are superior to the results from Bayesian classification model based on conventional three time-domain features of SAECG. In order to increase the robustness of the recognition, a filtered QRS offset point is randomly shifted +/- 8 ms to form a fuzzy training set, which was to simulate the possible error in detecting QRS offset point of filtered SAECG. We also found that nonlinear transformation through the hidden layer of developed ANN model could increase Euclidean distance between PEP and NEP patterns.

Algorithms↗
PubMed2026

Modelling the effects of biological intervention in a dynamical gene network.

Cellular response to environmental and internal signals can be modeled by dynamical gene regulatory networks (GRN). In the literature, three main classes of gene network models can be distinguished: (1) non-quantitative (or data-based) models which do not describe the probability distribution of gene expressions; (2) quantitative models which fully describe the probability distribution of all genes co-expression; and (3) mechanistic models which allow for a causal interpretation of gene interactions. We propose two rigorous frameworks to model gene alteration in a dynamical GRN, depending on whether the network model is quantitative or mechanistic. We explain how these models can be used for design of experiment, or, if additional alteration data are available, for validation purposes or to improve the parameter estimation of the original model. We apply these methods to the Gaussian graphical model, which is quantitative but non-mechanistic, and to mechanistic models of Bayesian networks and penalized linear regression.

Gene Regulatory Networks↗
PubMed1996

Modelling techniques and their application for monitoring in high dependency environments--learning models.

This paper reviews the use of learning models including Bayesian classifiers and artificial neural networks in monitoring and interpreting biosignals. Generally learning models applied for analysis of biosignals are "black-box' types trained on the basis of measured signals. It is illustrated that the training and application of learning models more or less follow the same sequences. The main focus is the interpretation of electrical signals from the brain (electroencephalogram (EEG) and evoked potentials (EP)). Current analysis of these signals often reveals sudden changes in the EEG or evoked potentials to be the earliest discernible signs of inadequate perfusion of the brain. They may reflect problems such as systemic arterial oxygen desaturation or hypotension arising from other body system failures during critical illness. It is suggested that these brain signals should be recorded in the critical care unit, and that they should form part of the annotated database of biosignals established during the IMPROVE project. This would allow for the development of new methods for on-line warning of impending damage to the central nervous system, such that corrective actions could be taken before permanent damage occurred.

Auscultation↗
PubMed2026

BaGGLS: a Bayesian shrinkage framework for interpretable modeling of interactions in high-dimensional biological data.

MOTIVATION: Biological data is often high dimensional, noisy, and governed by complex interactions among sparse signals. This poses major challenges for interpretability and reliable feature selection. Tasks such as identifying motif interactions in genomics exemplify these difficulties, as only a small subset of biologically relevant features (e.g. motifs) are typically active, and their effects are often non-linear and context-dependent. While statistical approaches often result in more interpretable models, deep learning models have proven effective in modeling complex interactions and prediction accuracy, yet their black-box nature limits interpretability. RESULTS: We introduce BaGGLS, a flexible and interpretable probabilistic binary regression model designed for high-dimensional biological inference involving feature interactions. BaGGLS incorporates a Bayesian group global-local shrinkage prior, aligned with the group structure introduced by interaction terms. This prior encourages sparsity while retaining interpretability, helping to isolate meaningful signals and suppress noise. To enable scalable inference, we employ a partially factorized variational approximation that captures posterior skewness and supports efficient learning even in large feature spaces. In extensive simulations, we compare BaGGLS to frequentist probit regressions (unconstrained and with L1-penalty) as well as a probit model with Markov Chain Monte Carlo (MCMC) sampling under a horseshoe prior. We can show that BaGGLS outperforms the other methods with regard to interaction detection and is many times faster than MCMC sampling under the horseshoe prior. We also demonstrate the usefulness of BaGGLS in the context of interaction discovery from motif scanner outputs (e.g. Find Individual Motif Occurrences (FIMO)) and noisy attribution scores from deep learning models. This shows that BaGGLS is a promising approach for uncovering biologically relevant interaction patterns, with potential applicability across a range of high-dimensional tasks in computational biology. AVAILABILITY: Code is available at gitlab.com/dacs-hpi/baggls.

Bayes Theorem↗