SEARCH · PubMed Health

Results for “Bayesian analysis”

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 667 records · Page 37Linked to original sources

PubMed2005

Long-oligomer microarray profiling in Neurospora crassa reveals the transcriptional program underlying biochemical and physiological events of conidial germination.

To test the inferences of spotted microarray technology against a biochemically well-studied process, we performed transcriptional profiling of conidial germination in the filamentous fungus, Neurospora crassa. We first constructed a 70 base oligomer microarray that assays 3366 predicted genes. To estimate the relative gene expression levels and changes in gene expression during conidial germination, we analyzed a circuit design of competitive hybridizations throughout a time course using a Bayesian analysis of gene expression level. Remarkable consistency of mRNA profiles with previously published northern data was observed. Genes were hierarchically clustered into groups with respect to their expression profiles over the time course of conidial germination. A functional classification database was employed to characterize the global picture of gene expression. Consensus motif searches identified a putative regulatory component associated with genes involved in ribosomal biogenesis. Our transcriptional profiling data correlate well with biochemical and physiological processes associated with conidial germination and will facilitate functional predictions of novel genes in N.crassa and other filamentous ascomycete species. Furthermore, our dataset on conidial germination allowed comparisons to transcriptional mechanisms associated with germination processes of diverse propagules, such as teliospores of the phytopathogenic fungus Ustilago maydis and spores of the social amoeba Dictyostelium discoideum.

Animals↗

PubMed2005

Variance components and selection response for feather-pecking behavior in laying hens.

Variance components and selection response for feather pecking behavior were studied by analyzing the data from a divergent selection experiment. An investigation indicated that a Box-Cox transformation with power lambda = -0.2 made the data approximately normally distributed and gave the best fit for the model. Variance components and selection response were estimated using Bayesian analysis with Gibbs sampling technique. The total variation was rather large for the investigated traits in both the low feather-pecking line (LP) and the high feather-pecking line (HP). Based on the mean of marginal posterior distribution, in the Box-Cox transformed scale, heritability for number of feather pecking bouts (FP bouts) was 0.174 in line LP and 0.139 in line HP. For number of feather-pecking pecks (FP pecks), heritability was 0.139 in line LP and 0.105 in line HP. No full-sib group effect and observation pen effect were found in the 2 traits. After 4 generations of selection, the total response for number of FP bouts in the transformed scale was 58 and 74% of the mean of the first generation in line LP and line HP, respectively. The total response for number of FP pecks was 47 and 46% of the mean of the first generation in line LP and line HP, respectively. The variance components and the realized selection response together suggest that genetic selection can be effective in minimizing FP behavior. This would be expected to reduce one of the major welfare problems in laying hens.

Animals↗

PubMed2007

Bayesian reconstruction of nanodosimetric cluster distributions at 100% detection efficiency.

Ionisation spectra in nanometric volumes at a given distance from a charged particle track are obtained by using electron (or ion) gas detectors, having non-uniformly distributed detection efficiency. Therefore, such spectra should be properly processed in order to reconstruct the frequency distribution of clusters really produced in the detector gas. A Bayesian unfolding has been applied to ionisation distributions due to 5.4 MeV alpha particles in a 20-nm site obtained by Monte Carlo simulations, taking into account different detection efficiency conditions. It will be shown that Bayesian analysis provides a valid tool for reconstructing the true ionisation distributions, well beyond the maximum measured cluster size.

Bayes Theorem↗

PubMed1997

Adequacy and efficacy of lateral cervical spine radiography in alert, high-risk blunt trauma patient.

OBJECTIVE: The purpose of this study was to determine the adequacy and accuracy of lateral cervical spine radiographs in the initial evaluation of alert, high-risk trauma patients evaluated at a Level I trauma center. METHODS: Data were obtained retrospectively through review of trauma service admissions from January 1, 1994, to July 31, 1995. Included were all patients triaged to a trauma response team with age > 15 years, Glasgow Coma Scale score > 13, and blunt mechanism of injury. Lateral cervical spine radiograms were obtained routinely before secondary survey and were reviewed for technical adequacy (all seven cervical vertebrae, C7/T1 interspace). The presence of cervical symptoms (pain, tenderness, neurologic deficits) was recorded. Sensitivity and specificity were calculated for lateral cervical spine radiography and cervical symptoms in predicting the presence of cervical spine injury. Bayesian analysis, which allows for the current probability of occurrence to be factored by previously reported probabilities of occurrence, was used to determine the negative predictive probability of lateral cervical spine radiography and absence of cervical symptoms to predict the absence of injury to the cervical spine. RESULTS: Three hundred fifty-three patients received lateral cervical spine radiograms, of which 223 (63%) were determined to be adequate for interpretation. Cervical symptoms were present in 77 patients (20%). Only 32 (42%) of this group's lateral cervical spine radiograms were adequate. Nine patients (2.4%) had acutely fractured cervical vertebrae or ligamentous disruption. Lateral cervical spine radiography showed the injury in only six of these patients. The sensitivity, specificity, and negative predictive probability for lateral cervical spine radiography were 67, 58, and 1.4%, respectively, and for absence of cervical symptoms, 89, 81, and 0.32%, respectively. CONCLUSION: The higher accuracy and lower negative predictive probability make the absence of cervical symptoms in the alert, high-risk, blunt trauma patient a better screening test than lateral cervical spine radiography. We suggest that lateral cervical spine radiography is not needed in the initial evaluation of alert patients who have sustained blunt trauma.

Adult↗

PubMed1993

Bayesian predictive inference for units with small sample sizes. The case of binary random variables.

The National Health Interview Survey is designed to produce precise estimates for the entire United States but not for individual states. In this study, Bayesian predictive inference is used to provide point estimates and measures of variability for the desired finite population quantities. The investigation reported here concerns binary random variables such as the occurrence of at least one doctor visit within the past 12 months. The specification is hierarchic. First, for each cluster, there is a separate logistic regression relating a patient's probability of a doctor visit with his or her characteristics. Second, there is a multivariate linear regression linking the (cluster) regression parameters to covariates measured at the cluster level. A fully Bayesian analysis is carried out; this technique provides gains over synthetic estimation and conventional randomization-based analysis. The reported approach is potentially useful for any situation when the sample size associated with a unit of interest (e.g., a hospital or small geographic area) is too small to permit satisfactory inference using only the data from that unit.

Bayes Theorem↗

PubMed1986

Evaluation of abdominal pain: clinicians' performance compared with three protocols.

Clinical findings in ambulatory patients with abdominal pain, presenting themselves to general medical practitioners, were analyzed by three different protocols: an algorithm using a branched-chain logic, a linear discriminant rule, and bayesian analysis. All methods missed some serious diagnoses. None was as effective as the clinician at distinguishing nonspecific abdominal pain (NSAP) from more serious conditions.

Abdomen↗

PubMed2002

Development of population pharmacokinetic models and optimal sampling times for ibuprofen tablet and suspension formulations in children with cystic fibrosis.

High-dose ibuprofen therapy has demonstrated to slow deterioration in pulmonary function in children with cystic fibrosis with mild lung disease. Therapeutic drug monitoring has been recommended to maintain peak concentrations within the range of 50 to 100 mg/L to ensure efficacy. Current methods for dosage individualization are based on dose proportionality using visual inspection of the peak concentration; however, because of interpatient variability in the absorption of the various formulations this method may result in incorrect assessments of the peak concentration achieved. Maximum a posteriori Bayesian analysis (MAP-B) has proven to be a useful and precise method of individualizing the dose of aminoglycosides but requires a description of the structural model. In this study we performed parametric population modeling analysis on plasma concentrations of ibuprofen after single doses of 20 to 30-mg/kg tablet or suspension in children with cystic fibrosis. Patients evaluated in this study were part of a single dose pharmacokinetic study that has been published previously. A one-compartment model with first order absorption and a lag time best described the data. The pharmacokinetic parameters differed significantly depending on the formulation administered. D-optimal sampling times for the suspension and tablet formulations are 0, 0.25 to 0.5, 1, and 3 to 4 hours and 0, 0.25 to 0.5, 1 to 1.5, and 5 hours respectively. Use of MAP-B analysis performed with the 4 d-optimal sampling strategy resulted in accurate and precise estimates of the pharmacokinetic parameters when compared with maximum likelihood analysis using the complete plasma concentrations data set. Further studies are needed to evaluate the performance of these models and the impact on patient outcomes.

Administration, Oral↗

PubMed1995

Physiologic state severity classification as an indicator of posttrauma cytokine response.

A Physiologic State Severity Classification (PSSC) derived from clustering of 17 cardiorespiratory variables was used to predict cytokine response in critically ill posttrauma patients. The PSSC defined physiologic states: A-State (A), normal stress response; B-State (B), metabolic insufficiency; C2-State (C), respiratory insufficiency. Bayesian analysis of these states defined a probability of death (Pdeath). 416 studies from 60 newly studied multiple trauma patients (70% males, Injury Severity Score = 27.5) were analyzed; 45 (75%) had sepsis (s), 28 (47%) had sepsis-adult respiratory distress syndrome (s-ARDS). Of 35 survivors (66% s, 37% s-ARDS, mean Pdeath = .42) 23% were predominantly A, 66% B, and 11% C. Of 25 deaths (88% s, 60% s-ARDS, mean Pdeath = .64) 0% were A, 44% B, and 56% C. PSSC States were correlated with incidence and mean plasma levels (pl) in picograms/mL of cytokines. 23 samples from recovering nonseptic trauma patients were used as controls.

Adolescent↗

PubMed2006

Population pharmacokinetics and pharmacodynamics of nelfinavir and its active metabolite M8 in HIV-1-infected children.

BACKGROUND: The objectives of this study are to develop and validate a population pharmacokinetic model that adequately describes the pharmacokinetics of nelfinavir and its active metabolite M8 in HIV-1-infected children; to define factors involved in the pharmacokinetic variability, which could aid in defining dosing strategies; and to correlate the pharmacokinetics to the treatment response. METHODS: Protease inhibitor-naive, HIV-1-infected children were included. A population pharmacokinetic model of nelfinavir and M8 was developed using NONMEM. Bayesian analysis was used to estimate pharmacokinetic values. A pharmacokinetic-pharmacodynamic analysis was performed to study relationships between these values and the virologic response to therapy. RESULTS: From 38 children, 724 nelfinavir and 636 M8 plasma concentrations were available. The pharmacokinetics of both compounds were described simultaneously with a one-compartment model with first-order elimination. Clearance (CL/F) and volume of distribution (V/F) were 32.6 L/h (interindividual variability [IIV]: 31.6%) and 281 L/h (IIV: 29.7%) for nelfinavir and 86.2 L/h (IIV: 43.1%) and 42.3 L/h for M8. No factors could be defined that affected the pharmacokinetics of nelfinavir or M8. The overall virologic response was 78% (HIV-1 RNA <500 copies/mL, on-treatment analysis). No differences in exposure to nelfinavir and M8 were observed between responders and nonresponders. The only factor distinguishing the two groups was a higher baseline HIV-1 RNA concentration in nonresponders. CONCLUSION: A model was developed and validated that adequately described the population pharmacokinetics of nelfinavir and M8 in a childhood population. No factors affecting dosing strategies were identified, and no correlation could be demonstrated between the exposure to nelfinavir and M8 and the virologic treatment response.

Adolescent↗

PubMed2005

Up hill, down dale: quantitative genetics of curvaceous traits.

'Repeated' measurements for a trait and individual, taken along some continuous scale such as time, can be thought of as representing points on a curve, where both means and covariances along the trajectory can change, gradually and continually. Such traits are commonly referred to as 'function-valued' (FV) traits. This review shows that standard quantitative genetic concepts extend readily to FV traits, with individual statistics, such as estimated breeding values and selection response, replaced by corresponding curves, modelled by respective functions. Covariance functions are introduced as the FV equivalent to matrices of covariances. Considering the class of functions represented by a regression on the continuous covariable, FV traits can be analysed within the linear mixed model framework commonly employed in quantitative genetics, giving rise to the so-called random regression model. Estimation of covariance functions, either indirectly from estimated covariances or directly from the data using restricted maximum likelihood or Bayesian analysis, is considered. It is shown that direct estimation of the leading principal components of covariance functions is feasible and advantageous. Extensions to multi-dimensional analyses are discussed.

Analysis of Variance↗

PubMed2005

Bayesian methods for the conformational classification of eight-membered rings.

Two methods for the classification of eight-membered rings based on a Bayesian analysis are presented. The two methods share the same probabilistic model for the measurement of torsion angles, but while the first method uses the canonical forms of cyclooctane and, given an empirical sequence of eight torsion angles, yields the probability that the associated structure corresponds to each of the ten canonical conformations, the second method does not assume previous knowledge of existing conformations and yields a clustering classification of a data set, allowing new conformations to be detected. Both methods have been tested using the conformational classification of Csp3 eight-membered rings described in the literature. The methods have also been employed to classify the solid-state conformation in Csp3 eight-membered rings using data retrieved from an updated version of the Cambridge Structural Database (CSD).

Algorithms↗

PubMed1998

Bayesian estimates of error bounds for EEG source imaging.

Given a set of electrical potential measurements at the surface of the head, localizing the sources of the electrical activity is an inherently ill-posed problem. Bayesian methods can be used to specify prior information to constrain the possible source solutions. We show that Bayesian analysis can also provide a means for characterizing system noise levels, estimating the "error bars" surrounding source localization results, and estimating the information about brain processes conveyed by dense sensor array electroencephalographic (EEG) recordings. This method is, in principal, applicable to any linear model of EEG or magnetoencephalographic (MEG) processes. A series of simulations demonstrated the internal consistency of our method, the robustness to noise levels, and the limitations of accurate source localization with large numbers of sources.

Artifacts↗

PubMed2004

A unified framework for subspace face recognition.

PCA, LDA, and Bayesian analysis are the three most representative subspace face recognition approaches. In this paper, we show that they can be unified under the same framework. We first model face difference with three components: intrinsic difference, transformation difference, and noise. A unified framework is then constructed by using this face difference model and a detailed subspace analysis on the three components. We explain the inherent relationship among different subspace methods and their unique contributions to the extraction of discriminating information from the face difference. Based on the framework, a unified subspace analysis method is developed using PCA, Bayes, and LDA as three steps. A 3D parameter space is constructed using the three subspace dimensions as axes. Searching through this parameter space, we achieve better recognition performance than standard subspace methods.

Algorithms↗

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↗

PubMed2003

On the precision of the conditionally autoregressive prior in spatial models.

Bayesian analyses of spatial data often use a conditionally autoregressive (CAR) prior, which can be written as the kernel of an improper density that depends on a precision parameter tau that is typically unknown. To include tau in the Bayesian analysis, the kernel must be multiplied by tau(k) for some k. This article rigorously derives k = (n - I)/2 for the L2 norm CAR prior (also called a Gaussian Markov random field model) and k = n - I for the L1 norm CAR prior, where n is the number of regions and I the number of "islands" (disconnected groups of regions) in the spatial map. Since I = 1 for a spatial structure defining a connected graph, this supports Knorr-Held's (2002, in Highly Structured Stochastic Systems, 260-264) suggestion that k = (n - 1)/2 in the L2 norm case, instead of the more common k = n/2. We illustrate the practical significance of our results using a periodontal example.

Bayes Theorem↗

PubMed2000

A mixture model for longitudinal data with application to assessment of noncompliance.

In clinical trials of a self-administered drug, repeated measures of a laboratory marker, which is affected by study medication and collected in all treatment arms, can provide valuable information on population and individual summaries of compliance. In this paper, we introduce a general finite mixture of nonlinear hierarchical models that allows estimates of component membership probabilities and random effect distributions for longitudinal data arising from multiple subpopulations, such as from noncomplying and complying subgroups in clinical trials. We outline a sampling strategy for fitting these models, which consists of a sequence of Gibbs, Metropolis-Hastings, and reversible jump steps, where the latter is required for switching between component models of different dimensions. Our model is applied to identify noncomplying subjects in the placebo arm of a clinical trial assessing the effectiveness of zidovudine (AZT) in the treatment of patients with HIV, where noncompliance was defined as initiation of AZT during the trial without the investigators' knowledge. We fit a hierarchical nonlinear change-point model for increases in the marker MCV (mean corpuscular volume of erythrocytes) for subjects who noncomply and a constant mean random effects model for those who comply. As part of our fully Bayesian analysis, we assess the sensitivity of conclusions to prior and modeling assumptions and demonstrate how external information and covariates can be incorporated to distinguish subgroups.

Anti-HIV Agents↗

PubMed2004

Bayesian experimental design for nonlinear mixed-effects models with application to HIV dynamics.

Bayesian experimental design is investigated for Bayesian analysis of nonlinear mixed-effects models. Existence of the posterior risk for parameter estimation is shown. When the same prior distribution is used for both design and inference, existence of the preposterior risk for design is also proven. If the prior distribution used in design is different from that used for inference, sufficient conditions are established for existence of the preposterior risk for design. A case study of design for an experiment in population HIV dynamics is provided.

Bayes Theorem↗

PubMed2004

Speciation and diversity on tropical rocky shores: a global phylogeny of snails of the genus Echinolittorina.

A phylogenetic approach to the origin and maintenance of species diversity ideally requires the sampling of all species within a clade, confirmation that they are evolutionarily distinct entities, and knowledge of their geographical distributions. In the marine tropics such studies have mostly been of fish and reef-associated organisms, usually with high dispersal. In contrast, snails of the genus Echinolittorina (Littorinidae) are restricted to rocky shores, have a four-week pelagic development (and recorded dispersal up to 1400 km), and show different evolutionary patterns. We present a complete molecular phylogeny of Echinolittorina, derived from Bayesian analysis of sequences from nuclear 28S rRNA and mitochondrial 12S rRNA and COI genes (nodal support indicated by posterior probabilities, maximum likelihood, and neighbor-joining bootstrap). This consists of 59 evolutionarily significant units (ESUs), including all 50 known taxonomic species. The 26 ESUs found in the Indo-West Pacific region form a single clade, whereas the eastern Pacific and Atlantic species are basal. The earliest fossil occurred in the Tethys during the middle Eocene and we suggest that the Indo-West Pacific clade has been isolated since closure of the Tethyan seaway in the early Miocene. The geographical distributions of all species (based on more than 3700 locality records) appear to be circumscribed by barriers of low temperature, unsuitable sedimentary habitat, stretches of open water exceeding about 1400 km, and differences in oceanographic conditions on the continuum between oceanic and continental. The geographical ranges of sister species show little or no overlap, indicating that the speciation mode is predominantly allopatric. Furthermore, range expansion following speciation appears to have been limited, because a high degree of allopatry is maintained through three to five branching points of the phylogeny. This may be explained by infrequent long-distance colonization, habitat specialization on the oceanic/continental gradient, and perhaps by interspecific competition. In the eastern Pacific plus Atlantic we identify five cases of divergence on either side of the Isthmus of Panama, but our estimates of their ages pre-date the emergence of the Isthmus. There are three examples of sister relationships between species in the western Atlantic and eastern Atlantic, all resulting from dispersal to the east. Within the Indo-West Pacific, we find no geographical pattern of speciation events; narrowly endemic species of recent origin are present in both peripheral and central parts of the region. Evidence from estimated divergence times of sister species, and from a plot of the number of lineages over time, suggest that there has been no acceleration of diversification during the glacio-eustatic cycles of the Plio-Pleistocene. In comparison with reefal organisms, species of Echinolittorina on rocky shores may be less susceptible to extinction or isolation during sea-level fluctuations. The species richness of Echinolittorina in the classical biogeographic provinces conforms to the common pattern of highest diversity (11 species) in the central "East Indies Triangle" of the Indo-West Pacific, with a subsidiary focus in the eastern Pacific and western Atlantic, and lowest diversity in the eastern Atlantic. The diversity focus in the East Indies Triangle is produced by a mosaic of restricted allopatric species and overlap of a few widespread ones, and is the result of habitat specialization rather than historical vicariance. This study emphasizes the plurality of biogeographic histories and speciation patterns in the marine tropics.

Animals↗