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 55 records · Page 3Linked to original sources

PubMed1979

The Chatham Blood Pressure Study. An application of Bayesian growth curve models to a longitudinal study of blood pressure in children.

Recent developments in statistics have produced powerful methods that facilitate the analysis of longitudinal studies. These methods are illustrated by an analysis of a longitudinal study of blood pressure in children. The results of the study show a clear tendency for blood pressure to increase with age, and Asian children tend to have lower blood pressures than their Caucasian counterparts of the same age. There is evidence to support the hypothesis that blood pressures track.

Age Factors↗

PubMed1992

Suramin: rapid loading and weekly maintenance regimens for cancer patients.

PURPOSE: Suramin is an anticancer agent with a narrow therapeutic window and a terminal half-life of 45 to 55 days. These characteristics make it necessary to control accurately the serum concentrations of the drug. Therefore, the aim of the present study was to develop a rapid loading regimen, followed by weekly administration of suramin to maintain serum concentrations of between 150 and 300 micrograms/mL for 8 weeks. PATIENTS AND METHODS: Eligible patients were treated with five different loading regimens. Initially, weekly maintenance doses were estimated manually by the treating physician. Subsequently, computer-assisted dosing that used Bayesian pharmacokinetic modeling was used. RESULTS: Thirty-eight courses of suramin that were administered to 35 patients were studied. The optimal loading regimen consisted of a continuous infusion of 600 mg/m2 during a 24-hour period, which resulted in a mean serum concentration of 319 micrograms/mL. Potentially toxic concentrations that were observed with shorter infusions were avoided. Maintenance treatment, which used the weekly administration of suramin during a 6-hour period, seemed to be able to maintain mean suramin serum trough concentrations of 150 micrograms/mL, while preventing mean peak concentrations of more than 300 micrograms/mL. The use of Bayesian pharmacokinetics was superior to manual estimation in tailoring the optimal dose to the therapeutic window. CONCLUSIONS: Continuous infusion is the optimal way of delivering suramin during the loading phase. To maintain trough levels and peak levels within a narrower therapeutic window, suramin will have to be administered more frequently than once a week. Bayesian modeling based on individual serum levels and population pharmacokinetics allows accurate dosing to maintain suramin levels within the therapeutic window.

Adult↗

PubMed2026

Genome-wide association studies for feed efficiency, production and feeding behavior traits in Canadian purebred Duroc pigs.

This study aimed to identify potential genetic variants and candidate genes associated with feed efficiency (FE), production, and feeding behavior traits in Canadian purebred Duroc pigs. Genome-wide association studies (GWAS) were conducted using 8,861 individuals and an imputed Affymetrix PigGen Canada 50K panel v2.0 using a linear mixed model (LMM) and a Bayesian B model. This analysis used an adjusted P-value threshold (ranging from 6.6 × 10-5 to 1.3 × 10-4) using a false-discovery rate to determine significance. The number of significant SNPs identified for each trait was as follows: average daily gain (ADG, 48), daily feed intake (DFI, 85), feed conversion ratio (FCR, 101), residual feed intake (RFI, 37), residual gain (RG, 64), residual intake and gain (RIG, 55), backfat thickness (BF, 100), loin depth (LD, 6), Kleiber's ratio (KR, 0), total time spent eating per day (TPD, 7), and number of visits to the feeder per day (NVD, 6). Several traits (BF, DFI, FCR, RFI, RG, and RIG) showed strong overlapping signals on chromosomes 7 and 10 with 24 shared significant SNPs, indicating potential shared genetic mechanisms. These traits also had 71 overlapping candidate genes, such as PACSIN1, PTCH1, ADIPOR1, and ITPR3, associated with glucose, lipid, and cholesterol metabolism. Well-known candidate genes in literature associated with growth and fatness such as MC4R and CDH20 were also identified to be associated with ADG, BF, FCR, and DFI in this study. Gene ontology enrichment analysis revealed that a set of the candidate genes were involved in the gonadotropin-releasing hormone (GnRH) and the platelet-derived growth factor (PDGF) signaling pathways. Overall, this study contributed to understanding the genetic architecture and provided a biological foundation for improving FE, production, and feeding behavior traits in Canadian Duroc pigs, facilitating the selection of more efficient pigs.

Sus scrofa↗

PubMed1991

A Monte Carlo method for Bayesian inference in frailty models.

Many analyses in epidemiological and prognostic studies and in studies of event history data require methods that allow for unobserved covariates or "frailties." Clayton and Cuzick (1985, Journal of the Royal Statistical Society, Series A 148, 82-117) proposed a generalization of the proportional hazards model that implemented such random effects, but the proof of the asymptotic properties of the method remains elusive, and practical experience suggests that the likelihoods may be markedly nonquadratic. This paper sets out a Bayesian representation of the model in the spirit of Kalbfleisch (1978, Journal of the Royal Statistical Society, Series B 40, 214-221) and discusses inference using Monte Carlo methods.

Algorithms↗

PubMed1992

Advanced computer programs for drug dosing that combine pharmacokinetic and symbolic modeling of patients.

In this paper, we describe our design for advanced drug dosing programs that "reason" using a combination of Bayesian pharmacokinetic modeling and symbolic modeling of patient status and drug response. Our design is similar to the design of the Digitalis Therapy Advisor program, but extends this previous work by incorporating a Bayesian pharmacokinetic model, performing a "meta-level" analysis of drug concentrations to identify sampling errors and changes in pharmacokinetics, and including the results of this analysis in reasoning for dosing and therapeutic monitoring recommendations. The design has been implemented in a program for aminoglycoside antibiotics called Aminoglycoside Therapy Manager. The program is user-friendly and runs on low-cost general-purpose hardware. The initial validation study showed that the program was as accurate in predicting future drug concentrations as an expert using commercial Bayesian forecasting software and that its dosing recommendations were similar to those of an expert.

Aminoglycosides↗

PubMed2025

Uncertainty Modeling Outperforms Machine Learning for Microbiome Data Analysis.

Microbiome sequencing measures relative rather than absolute abundances, providing no direct information about total microbial load. Normalization methods attempt to compensate, but rely on strong, often untestable assumptions that can bias inference. Experimental measurements of load (e.g., qPCR, flow cytometry) offer a solution, but remain costly and uncommon. A recent high-profile study proposed that machine learning could bypass this limitation by predicting microbial load from sequencing data alone. To evaluate this claim, we assembled mutt, the largest public database of paired sequencing and load measurements, spanning 35 studies and over 15,000 samples. Using mutt, we show that published machine learning models fail to generalize: on average they perform worse than a naive baseline that always predicted the training set mean. These failures stem from covariate shift-limited shared taxa between studies, differences in community composition, and differences in preprocessing pipelines-that silently derail model inputs. In contrast, Bayesian partially identified models do not attempt to impute microbial load, but instead propagate scale uncertainty through downstream analyses. Across 30 benchmark datasets, Bayesian partially identified models consistently outperformed normalization and machine learning approaches, providing a principled and reproducible foundation for microbiome inference.

16S rRNA-seq↗

PubMed2025

Genomic prediction and genome-wide association study for liver abscesses in crossbred beef cattle.

Liver abscesses are a concern in feedlot cattle, and little is known about the role of genetics in their development. This study aimed to estimate genetic parameters and to identify single-nucleotide polymorphisms (SNPs) associated with liver abscesses. Crossbred cattle representing 18 breeds in the U.S. Meat Animal Research Center Germplasm Evaluation Program were phenotyped for liver abscesses at slaughter (n = 9,044). Seventeen percent of cattle had liver abscesses. These cattle had genotypes that were imputed to sequence variant genotypes. After filtering and quality control, 340,723 SNPs were used in the analysis. Liver abscess prevalence was modeled with a single-step genomic best linear unbiased prediction (ssGBLUP) threshold model using a Bayesian framework. The model included contemporary group (sex, treatment group, and slaughter date), additive genomic, and residual effects. Genomic heritability was 0.039 (95% highest posterior density = 0.005, 0.081), which was very small. To assess prediction quality, a 5-fold random cross-validation structure was used. Method Linear Regression was used to assess accuracy, bias, and dispersion by comparing estimated breeding values (EBV) from full and reduced analyses. Cross-validation metrics showed EBV based on genotypes had 0.05 reliability (SD < 0.01) with no bias relative to EBV based on genotypes and phenotypes. For the genome-wide association study, SNP effects were back calculated from the EBV solutions from ssGBLUP. No SNPs were associated with liver abscesses at a Benjamini-Hochberg adjusted 0.05 significance level. Although a large dataset was used, this result was because of the low genomic heritability and imprecise EBV used to calculate SNP effects. Based on these results, environmental factors contribute to most of the variation in liver abscesses. Genetic selection to reduce liver abscesses would be slow because of the low genomic heritability, measurement late in life, and inability to measure breeding animals. A faster approach would be finding additional environmental interventions that maintain animal performance.

Animals↗

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↗

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↗

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↗

PubMed2026

Oncotype DX-guided vs physician-directed chemotherapy and survival in HR+/HER2- breast cancer.

BACKGROUND: Oncotype DX testing guides adjuvant chemotherapy decisions in early-stage hormone receptor-positive/HER2-negative breast cancer, but testing is not universally performed, and outcomes associated with genomic-informed versus clinicopathologic-based chemotherapy decision pathways remain unclear. METHODS: Using the 2022 National Cancer Database Breast Participant User File, we identified women diagnosed from 2010 to 2022 with pathologic T1b-T2, node-negative, hormone receptor-positive/HER2-negative invasive breast cancer who received adjuvant chemotherapy and endocrine therapy. Patients were classified into an Oncotype-guided group, defined by Oncotype DX testing with a recurrence score of 26 or higher, and a physician-directed group, defined by receipt of chemotherapy without genomic testing. The primary outcome was overall survival. Analyses used multivariable Cox models, logistic-IPTW and MLP-IPTW, restricted mean survival time analysis, and a Bayesian latent confounding survival model. RESULTS: Among 56,625 women, 27,278 were in the Oncotype-guided group and 29,347 in the physician-directed group. Median ages were 59 and 56 years, respectively. The Oncotype-guided group had more favorable overall survival than the physician-directed group in multivariable Cox analysis (HR, 0.906; 95% CI, 0.856-0.959; P < 0.001), with similar findings in IPTW analyses. The association was concentrated among patients aged 56 years or older (HR, 0.866; 95% CI, 0.809-0.927; P < 0.001). The Bayesian model showed no strong residual confounding signal. CONCLUSIONS: Among chemotherapy-treated women, an Oncotype-guided pathway was associated with more favorable overall survival than a physician-directed pathway, particularly among older patients, which indicating prognostic heterogeneity selected using genomic versus conventional clinicopathologic information.

Humans↗

PubMed1992

On some applications of Bayesian methods in cancer clinical trials.

The NCCTG randomized controlled clinical trial for the treatment of advanced colorectal carcinoma is a wonderful case study of the dynamic interplay between scientific learning and statistical inference. Ethical concerns for minimizing the number of patients assigned to an inferior treatment and interest in identifying subsets of patients for whom a treatment is most likely efficacious pose challenging problems for the practice of statistics. In the first part of this paper, I comment on the applications of Bayesian methods to these problems in the NCCTG trial as presented by Freedman and Spieglehalter and Dixon and Simon, respectively. In the second part of this paper, I discuss and illustrate a Bayesian approach to model sensitivity analysis with a particular focus on model specification and criticism. The Bayesian approach provides a formal methodology to assess the sensitivity of inferences to the inputs into an analysis so that it is possible to investigate the consequences of the specification of the model. I apply these methods to the specification and criticism of a class of survival models for the analysis of survival times in the NCCTG trial.

Antineoplastic Combined Chemotherapy Protocols↗

PubMed1992

Empirical Bayes versus fully Bayesian analysis of geographical variation in disease risk.

This paper reviews methods for mapping geographical variation in disease incidence and mortality. Recent results in Bayesian hierarchical modelling of relative risk are discussed. Two approaches to relative risk estimation, along with the related computational procedures, are described and compared. The first is an empirical Bayes approach that uses a technique of penalized log-likelihood maximization; the second approach is fully Bayesian, and uses an innovative stochastic simulation technique called the Gibbs sampler. We chose to map geographical variation in breast cancer and Hodgkin's disease mortality as observed in all the health care districts of Sardinia, to illustrate relevant problems, methods and techniques.

Bayes Theorem↗

PubMedDate not supplied

Bayesian analysis of a dose-response experiment with serial sacrifices.

This paper presents analysis and comments which are believed to be appropriate for certain carcinogenesis studies where sacrifices are performed throughout the experiment. Estimates of the risk probability for each dose level and sacrifice time are found utilizing the sample likelihood as the posterior density. The dose-response relationship is investigated with these estimates as the response. In order to test if the dose is effective and to check the appropriateness of the time-to-incidence model a Bayesian multiple comparisons technique is introduced.

Animals↗

PubMedDate not supplied

How much quality control is enough? A cost-effectiveness model for clinical laboratory quality control procedures (illustrated by its application to a ligand-assay-based screening program).

Quality assurance testing represents a substantial proportion of the clinical laboratory budget, but current guidelines are based on criteria that pertain to analytic error rather than to optimization of the cost-effectiveness of patient care. A general Bayesian mathematical model for the cost-effectiveness of assay quality control has been developed, and is demonstrated using previously published data. The cost-effectiveness of quality assurance as defined here depends upon the prevalence of disease, the shapes of the distributions of test results observed in the non-diseased and diseased populations, the decision limit selected for labeling results positive or negative, the costs and benefits associated with each of the possible therapeutic outcomes, the magnitude of random and systematic analytical errors, the statistical power of the quality control test in use, the costs associated with delays due to re-assay, and the proportion of total test cost attributable to quality control procedures. Given current clinical laboratory practice, much of this information will not be routinely available. The model combines these factors into a simple equation with three terms: one for the cost of the original and any required repeat laboratory analyses, one for the cost of delay entailed by the rejection of an assay batch, and one for the change in total costs consequent to rejection of erroneous assay results.

Clinical Laboratory Techniques↗

PubMed2025

Generative model for the first cell fate bifurcation in mammalian development.

The first cell fate bifurcation in mammalian development directs cells toward either the trophectoderm (TE) or inner cell mass (ICM) compartments in pre-implantation embryos. This decision is regulated by the subcellular localization of a transcriptional co-activator YAP and takes place over several progressively asynchronous cleavage divisions. As a result of this asynchrony and variable arrangement of blastomeres, reconstructing the dynamics of the TE/ICM cell specification from fixed embryos is extremely challenging. To address this, we developed a live-imaging approach and applied it to measure pairwise dynamics of nuclear YAP and its direct target genes, CDX2 and SOX2, which are key transcription factors of the TE and ICM, respectively. Using these datasets, we constructed a generative model of the first cell fate bifurcation, which reveals the time-dependent statistics of the TE and ICM cell allocation. In addition to making testable predictions for the joint dynamics of the full YAP/CDX2/SOX2 motif, the model revealed the stochastic nature of the induction timing of the key cell fate determinants and identified the features of YAP dynamics that are necessary or sufficient for this induction. Notably, temporal heterogeneity was particularly prominent for SOX2 expression among ICM cells. As heterogeneities within the ICM have been linked to the initiation of the second cell fate decision in the embryo, understanding the origins of this variability is of key significance. The presented approach reveals the dynamics of the first cell fate choice and lays the groundwork for dissecting the next cell fate decisions in mouse development.

Animals↗

PubMed1991

Estimation of relative potency with sequential dilution errors in radioimmunoassay.

Sequential dilution is a very common procedure in radioimmunoassay, in which the dilution error will be accumulated from the highest to the lowest concentration. A simulated example in relative potency determination is used to demonstrate the potentially wrong conclusion that can be drawn, when the dilution error is not properly included in the model. A Bayesian method is used and an alternative approximation via maximum likelihood is proposed. An alternative experimental design is recommended to increase the precision of the inference.

Bayes Theorem↗

PubMed2026

Comparative effectiveness of game-based learning modalities in nursing and medical education: a systematic review and Bayesian network meta-analysis.

BACKGROUND: Game-based learning (GBL) is increasingly used in healthcare education, but educators must choose among diverse modalities (e.g., quiz platforms, apps, serious games and metaverse environments). Comparative evidence on which modalities perform best across learning domains (knowledge, attitudes, and practice) remains limited. AIM: To compare the effects of distinct GBL modalities on knowledge, attitudes, and practice outcomes in nursing and medical education and to explore whether comparative effects differ by learner group (pre-licensure students and in-service professionals). DESIGN: PRISMA-NMA-aligned systematic review and Bayesian network meta-analysis. METHODS: We searched eight databases and trial registries through September 2, 2024, for randomized controlled trials comparing GBL with traditional teaching (TT). Outcomes were transformed to a 0-100 scale and analysed as change from baseline in Bayesian consistency models; random-effects models were selected using deviance information criterion (DIC). Risk of bias was assessed using RoB 2. We report mean differences (MDs) with 95% credible intervals (CrIs) versus TT, ranking probabilities, and subgroup NMAs by learner group. RESULTS: Thirty-one RCTs (n = 3439) were included; 15 contributed complete data to the network. Risk of bias was low in 15 trials and raised some concerns in 16. The network was modest for knowledge (11 trials) and sparse for attitudes (3) and practice (4). Compared with TT, metaverse-based learning showed improved attitudes (MD 15; 95% CrI 12 to 18), based on a single trial. For knowledge and practice, Kahoot-based quizzes (MD 9.1; 95% CrI -8.9 to 27) and app-based learning (MD 4.6; 95% CrI -4.4 to 14) had the highest estimated mean improvements, but credible intervals were wide and included the null for most comparisons. Subgroup rankings differed by learner group, but several comparisons were imprecise and uncertainty was substantial, particularly in sparse networks. CONCLUSIONS: GBL modalities may improve learning outcomes compared with TT, but relative effects appear domain-specific and the certainty of rankings is limited by sparse evidence and imprecision. Future trials should prioritise head-to-head comparisons, robust outcome measurement, and longer-term retention and transfer outcomes in both student and in-service populations.

Humans↗