Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Bayesian inference”

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 163 records · Page 9Linked to original sources
PubMed2001

Bayesian technique for investigating linearity in event-related BOLD fMRI.

Event-related BOLD fMRI data is modeled as a linear time-invariant system. Together with Bayesian inference techniques, a statistical test is developed for rigorously detecting linearity/nonlinearity in the BOLD response system. The test is applied to data collected from eight subjects using an event-related paradigm with a switching checkerboard as the visual stimulus. Analyzed as a group, the results clearly find the response to be nonlinear. When each subject is analyzed individually, however, the results are predominantly nonlinear, but there is some evidence to suggest that there may be a crossover from a linear to a nonlinear regime and vice versa. This could be important when estimating physiological parameters for individuals. Additionally, estimates of the hemodynamic response function and corresponding response were obtained, but there was no consistent appearance of a poststimulus undershoot in the event-related BOLD response.

Adult↗
PubMed2003

Estimation of infection and recovery rates for highly polymorphic parasites when detectability is imperfect, using hidden Markov models.

A Bayesian hierarchical model is proposed for estimating parasitic infection dynamics for highly polymorphic parasites when detectability of the parasite using standard tests is imperfect. The parasite dynamics are modelled as a non-homogeneous hidden two-state Markov process, where the observed process is the detection or failure to detect a parasitic genotype. This is assumed to be conditionally independent given the hidden process, that is, the underlying true presence of the parasite, which evolves according to a first-order Markov chain. The model allows the transition probabilities of the hidden states as well as the detectability parameter of the test to depend on a number of covariates. Full Bayesian inference is implemented using Markov chain Monte Carlo simulation. The model is applied to a panel data set of malaria genotype data from a randomized controlled trial of bed nets in Tanzanian children aged 6-30 months, with the age of the host and bed net use as covariates. This analysis confirmed that the duration of infections with parasites belonging to the MSP-2 FC27 allelic family increased with age.

Algorithms↗
PubMed2003

Markov modelling of changes in HIV-specific cytotoxic T-lymphocyte responses with time in untreated HIV-1 infected patients.

HIV-specific cytotoxic CD8(+) T-lymphocytes (CTL) appear to be the cornerstone of the immune response to HIV infection. Recent studies show that CTL activity reflects patients' anti-HIV immune status and slows disease progression. However, the dynamics of the diversity of this response also appears as a key parameter for immune control but the dynamics of this diversity is largely undocumented. We modelled changes in CTL responses against the seven principal HIV proteins over time. We also studied the influence of plasma viral load on temporal changes in HIV protein recognition by memory CTL. The generic model we developed is based on a continuous time homogeneous Markov process with reversible states. Those states are defined by the number of proteins recognized by memory CTL in a given patient at a given time. This approach was developed within a Bayesian framework. Full Bayesian inference is implemented using Markov chain Monte Carlo simulations (MCMC). The Gibbs sampling algorithm was used to estimate the marginal posterior distributions of the transition intensities between stages of CTL responses. We applied our model to data of 152 HIV-infected patients included in the IMMUNOCO cohort. The model suggested that the diversity of HIV protein recognition by memory CTL in treatment-naive patients decreases as the disease progresses. Namely, the loss of T cytotoxic responses is globally faster than their acquisition. Indeed, these patients' T cytotoxic responses were characterized by marked individual turnover and a gradual loss of multiple protein recognition over time, this loss accelerating as viral load increased.

Adult↗
PubMed1989

Comparison of the information in two lung function experiments.

The amount of ventilation relative to perfusion (the ventilation-perfusion ratio) received by the lung is a useful indicator of the efficiency of lung function. Two alternative techniques for recovering the ventilation-perfusion ratio are outlined. While both techniques rely on the use of inert gases, one is well established and the other is only in a developmental stage. This paper focuses on a comparison of the amount of statistical information provided by these two techniques about the ventilation-perfusion ratio. The criterion applied here for measuring amount of information has roots in communication theory and uses ideas inherent to Bayesian inference.

Bayes Theorem↗
PubMed1999

The pattern of variation in centipede segment number as an example of developmental constraint in evolution

The range of animal morphologies observed in nature is partly determined by natural selection. However, there is no agreement yet regarding whether it is also partly determined by developmental constraint. Testing for the effects of constraint has been difficult due to the lack of both an appropriate null model and a sufficiently simple system capable of yielding unambiguous results regarding the model's plausibility. Here we examine the case of variation in segment number in geophilomorph centipedes. Curiously, while this ranges between 29 and 191, there are no species in which an even number of segments is observed, in contrast to about 1000 species with odd numbers of segments. It seems unlikely that this distribution of character values is determined by selection alone. Using an approach based on Bayesian inference, we attempt to quantify the probability of obtaining the observed distribution of values given a null model in which developmental constraint is absent. Since this probability is in the region of 10(-20), we conclude that constraint must be involved. We discuss various implications of this conclusion, and comment on the unexpected absence of neoteny and progenesis in centipede evolution. Copyright 1999 Academic Press.

Journal Article↗
PubMed2002

On the probability model for asthma attacks.

In environmental epidemiology, the impact of environmental agents on symptoms or health status is of interest. This influence is described quantitatively in the theory of Whittemore & Keller (1979). They formulated a logistic model for individuals that is useful in evaluation of panel studies in which each participant protocols whether he does or does not have a certain symptom each day. In the present paper an equation for the prevalence of symptoms in the study population that is defined as the fraction of symptomatic subjects is deduced from the model for individuals. The model for the aggregated quantity depends on the individuals' parameters in a nonlinear manner. The relationship between the individual-based model and the corresponding population-based model is illustrated by means of a simulated panel. Bayesian estimates of the parameters are calculated and compared for both approaches. Bayesian inference enables to apply the prevalence model to a population of non-identical individuals. For such a heterogeneous population, we observe an attenuation of environmental effects on the aggregated symptom prevalence in comparison to the individual-based approach. The presented theory is applicable not only to panel studies but also in time-series analysis of prevalences and incidences.

Asthma↗
PubMed2002

Bayesian estimation of dynamical systems: an application to fMRI.

This paper presents a method for estimating the conditional or posterior distribution of the parameters of deterministic dynamical systems. The procedure conforms to an EM implementation of a Gauss-Newton search for the maximum of the conditional or posterior density. The inclusion of priors in the estimation procedure ensures robust and rapid convergence and the resulting conditional densities enable Bayesian inference about the model parameters. The method is demonstrated using an input-state-output model of the hemodynamic coupling between experimentally designed causes or factors in fMRI studies and the ensuing BOLD response. This example represents a generalization of current fMRI analysis models that accommodates nonlinearities and in which the parameters have an explicit physical interpretation. Second, the approach extends classical inference, based on the likelihood of the data given a null hypothesis about the parameters, to more plausible inferences about the parameters of the model given the data. This inference provides for confidence intervals based on the conditional density.

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↗
PubMed1990

Probability and the patient state space.

This paper describes work to develop a model-based system to support clinical decision-making. In previous articles, we have developed (from 695 measurement sets obtained from 148 patients) a physiologic state classification based on a set of 11 cardiovascular and metabolic measurements. There is an R or reference state, for stable ICU patients. Patients under (operative, traumatic, or compensated septic) stress, or with (septic or hepatic) metabolic, respiratory, or cardiac insufficiency are in the A, B, C, or D states, respectively. We wished to make the state easier to measure and eventually available continuously, automatically, and noninvasively, as well as reflecting a wider group of bodily systems. The 5 centers define a 4 dimensional affine subspace, designated the cardiovascular state space. Using eigenvector analysis, we have found four new derived physiologic variables CV1, CV2, CV3, and CV4 that span the state space. We have fit sets of linear regression equations that allow the patient's position in the state space, and therefore his state, to be determined from more easily obtainable sets of measurements. Further, we selected 1966 measurement sets from 512 patients at two hospitals. We used the data from 250 of these patients to define 13 prototypical types, namely survivors and deaths from various combinations of sepsis, cardiogenic decompensation, cirrhosis, and pneumonitis, following trauma or general surgery. For any future patient, the statistical theory of Bayesian inference allows one to infer back from the measurements observed to the probability of his being of any of these types and of surviving or dying. We used this method to predict the outcome of the other 262 patients, prospectively. Statistically, the predictions of survival or death were not significantly different from the actual. For individual patients, the method predicts a clinical course that closely follows the actual episodes in their history. These results confirm and explain the validity of the concept of the patient state and make the state easier to compute. The patient state and the probability plot together help to stage, select, and evaluate therapy. They do not replace the clinician's judgement, but rather are tools that help the clinician to exercise judgement.

Adult↗
PubMed2003

Estimation of single-trial multicomponent ERPs: differentially variable component analysis (dVCA).

A Bayesian inference framework for estimating the parameters of single-trial, multicomponent, event-related potentials is presented. Single-trial recordings are modeled as the linear combination of ongoing activity and multicomponent waveforms that are relatively phase-locked to certain sensory or motor events. Each component is assumed to have a trial-invariant waveform with trial-dependent amplitude scaling factors and latency shifts. A Maximum a Posteriori solution of this model is implemented via an iterative algorithm from which the component's waveform, single-trial amplitude scaling factors and latency shifts are estimated. Multiple components can be derived from a single-channel recording based on their differential variability, an aspect in contrast with other component analysis techniques (e.g., independent component analysis) where the number of components estimated is equal to or smaller than the number of recording channels. Furthermore, we show that, by subtracting out the estimated single-trial components from each of the single-trial recordings, one can estimate the ongoing activity, thus providing additional information concerning task-related brain dynamics. We test this approach, which we name differentially variable component analysis (dVCA), on simulated data and apply it to an experimental dataset consisting of intracortically recorded local field potentials from monkeys performing a visuomotor pattern discrimination task.

Algorithms↗
PubMedDate not supplied

Medical expert systems based on causal probabilistic networks.

Causal probabilistic networks (CPNs) offer new methods by which you can build medical expert systems that can handle all types of medical reasoning within a uniform conceptual framework. Based on the experience from a commercially available system and a couple of large prototype systems, it appears that CPNs are now an attractive alternative to other methods. A CPN is an intensional model of a domain, and it is therefore conceptually much closer to qualitative reasoning systems and to simulation systems than to rule-based or logic-based systems. Recent progress in Bayesian inference in networks has yielded computationally efficient methods. The inference method used follows the fundamental axioms of probability theory, and gives a sound framework for causal and diagnostic (deductive and abductive) reasoning under uncertainty. Experience with the prototypes indicates that it may be possible to use decision theory as a rational approach to test planning and therapy planning. The way in which knowledge is acquired and represented in CPNs makes it easy to express 'deep knowledge' for example in the form of physiological models, and the facilities for learning make it possible to make a smooth transition from expert opinion to statistics based on empirical data.

Artificial Intelligence↗
PubMed1996

Non-medical influences on medical decision-making.

BACKGROUND: The influence of non-medical factors on physicians' decision-making has been documented in many observational studies, but rarely in an experimental setting capable of demonstrating cause and effect. We conducted a controlled factorial experiment to assess the influence of non-medical factors on the diagnostic and treatment decisions made by practitioners of internal medicine in two common medical situations. METHOD: One hundred and ninety-two white male internists individually viewed professionally produced video scenarios in which the actor-patient, presenting with either chest pain or dyspnea, possessed various balanced combinations of sex, race, age, socioeconomic status, and health insurance coverage. Physician subjects were randomly drawn from lists of internists in private practice, hospital-based practice, and HMO's, at two levels of experience. RESULTS: The most frequent diagnoses for both chest pain and dyspnea were psychogenic origin and cardiac problems. Smoking cessation was the most frequent treatment recommendation for both conditions. Younger patients (all other factors being the same) were significantly more likely to receive the psychogenic diagnosis. Older patients were more likely to receive the cardiac diagnosis for chest pain, particularly if they were insured. HMO-based physicians were more likely to recommend a follow-up visit for chest pain. Several interactions of patient and physician factors were significant in addition to the main effects. CONCLUSIONS: The variability in decision-making evidenced by physicians in this experiment was not entirely accounted for by strictly rational Bayesian inference (the common prescriptive model for medical decision-making), in-as-much as non-medical factors significantly affected the decisions that they made. There is a need to supplement idealized medical schemata with considerations of social behavior in any comprehensive theory of medical decision-making.

Adult↗
PubMed1993

Exploring the relationship between rationality and bounded rationality in medical knowledge-based systems.

If our goal in Artificial Intelligence in Medicine (AIM) is to engineer systems health-care providers will both use and, in the process, improve their performance, we must concentrate on the development of causal theories of knowledge and problem solving. One broad direction in pursuing this goal is understanding the relationships between existing models of rationality and bounded rationality for similar tasks. Models of rationality refer to those approaches in which the optimal properties of the models are deductively provable, i.e. in which the processing is rational. Representative models of rationality used in AIM are deductive logical models, statistical models such as Bayesian inference models, and decision-analytic models. Models of bounded rationality are those which do not guarantee such optimal properties nor yield to deductive correctness proofs. These models have their roots in cognitive psychology. In this article we show how explicating the relationship between models of rationality and bounded rationality might be done in the case of abductive tasks in medicine. This is done by positioning these modeling approaches within the same framework (an abstract computational model) and interpreting in this context both computational complexity results concerning the nature of the task and empirical results studies of human problem-solving behavior.

Artificial Intelligence↗
PubMed2003

Apicoplast genome of the coccidian Eimeria tenella.

Unicellular apicomplexans possess an algal-originated plastid referred to as an apicoplast. Although apicomplexan parasites are comprised of highly diverse protists, the complete apicoplast genome sequences have only been determined from the hematozoan Plasmodium falciparum and cyst-forming coccidian Toxoplasma gondii. Here, we report the third complete sequence of apicoplast genome from the intestinal coccidian Eimeria tenella that may serve as a new drug target against coccidiosis in the livestock. The AT-rich E. tenella plastid genome is a 35-kb circular element. Its gene organization resembles more closely that of T. gondii than P. falciparum. Although the E. tenella plastid genome contains an almost identical set of genes to that found in P. falciparum and T. gondii, its encoded genes share low or moderate homologies with their counterparts in the other two apicomplexans. With the addition of this coccidian plastid genome sequence, we attempted to reexamine the apicoplast genome evolution and performed phylogenetic reconstructions using maximum likelihood and Bayesian inference (BI) methods based on a concatenated dataset of plastid-encoded rpoB, rpoC1 and rpoC2 proteins. All resulting rpo protein trees placed apicoplast as a sister to Euglena within the green lineage. On the other hand, many recent studies based on the organization of plastid genes and some nuclear-encoded plastid proteins have supported a common red algal ancestry of apicomplexan and dinoflagellate plastids. If the apicoplast indeed originated from a red ancestor, the green relationship of apicomplexan genes would probably imply that the ancestral host that gave rise to the (red) apicoplast might have already contained some primary green plastid genes.

Animals↗
PubMed2000

Estimation of sensitivity and specificity of diagnostic tests and disease prevalence when the true disease state is unknown.

The performance of a new diagnostic test is frequently evaluated by comparison to a perfect reference test (i.e. a gold standard). In many instances, however, a reference test is less than perfect. In this paper, we review methods for estimation of the accuracy of a diagnostic test when an imperfect reference test with known classification errors is available. Furthermore, we focus our presentation on available methods of estimation of test characteristics when the sensitivity and specificity of both tests are unknown. We present some of the available statistical methods for estimation of the accuracy of diagnostic tests when a reference test does not exist (including maximum likelihood estimation and Bayesian inference). We illustrate the application of the described methods using data from an evaluation of a nested polymerase chain reaction and microscopic examination of kidney imprints for detection of Nucleospora salmonis in rainbow trout.

Algorithms↗
PubMed1998

Reducing the incidence of epileptic seizures in the Belgian Tervuren through selection.

There is growing evidence that idiopathic epilepsy in the Belgium Tervuren has a genetic foundation. Reducing the incidence of this disorder, which may afflict as much as 17% of the breed, will rely upon the wise selection of parents. Seizure data on 997 dogs from the American Belgian Tervuren Club were collected through questionnaires in which animals were classified into one of four mutually exclusive categories: 1) no seizures observed, 2) one seizure observed, 3) two to five seizures, and 4) more than five seizures. The analysis of this ordered data made use of a threshold model of Bayesian inference. Integration of posterior densities was accomplished through Gibbs sampling. Through this analysis we are able to predict that the offspring of the mating of two non-epileptic dogs has a probability of 0.99 of never suffering from a seizure. The offspring of the mating of two dogs who have each had 1 seizure has a predicted probability 0.58 of never suffering from a seizure. Prevention of this disease is best prescribed through the selection of non-epileptic dogs as parents of future generations.

Animals↗
PubMed1998

Mixed graphical models for simultaneous model identification and control applied to the glucose-insulin metabolism.

In this paper a method for model identification of biological systems described by stochastic linear differential equations using a new computational technique for statistical Bayesian inference, namely mixed graphical models in the sense of Lauritzen and Wermuth, is presented. The model is identified in terms of biological model parameters and noise parameters. This non-linear estimation problem is solved by means of an exact inference algorithm. The parameter estimates are given as a-posteriori distributions which can be interpreted as fuzzy possibility distributions. For model-based simulations of the underlying biological system the model parameters are represented as uncertain parameters with the distributions obtained from the estimation procedure. We apply the presented methods to a model for the glucose-insulin metabolism: the Karlsburg model for type I diabetes.

Bayes Theorem↗
PubMed2000

Statistical analysis of domino chemical accidents.

A set of chemical accidents is retrieved from the literature and classified with regard to the substance involved and whether domino effects are present. This set of accidents and each of the classes defined are statistically analyzed with respect to its severity and comparison is made between domino and non-domino accidents. The analysis reveals that each accident category shows characteristic patterns in terms of fatalities caused and domino effects likelihood. Moreover, chemical accidents severity frequencies are described by using a two-parameter, revised form of the Pareto probability density function. The range within which the values of the parameters lie is investigated using Bayesian inference.

Accidents, Occupational↗