PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Uncertainty 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 433 records · Page 24Linked to original sources

Bayesian estimation of cost-effectiveness: an importance-sampling approach.

We describe a method for estimating the cost-effectiveness of a new treatment compared to a standard, using data from a comparative clinical trial. We quantify the clinical effectiveness as a binary variable indicating success or failure. The underlying statistical model assumes that costs are uncensored and follow separate gamma distributions in each of the groups defined by the four possible combinations of treatment arm and effectiveness outcome. The method is subjectivist, in that it represents prior uncertainty about model parameters with a probability distribution, which we update via Bayes's theorem to produce a posterior distribution. We approximate the posterior by importance sampling, a straightforward simulation method. We illustrate the method with an analysis of cost (derived from resource usage data) and effectiveness (measured by one-year survival) in a clinical trial in heart disease. The example demonstrates that the method is practical and provides for a flexible data analysis.

Bayes Theorem↗

Influences of parameter uncertainties within the ICRP-66 respiratory tract model: regional tissue doses for 239PuO2 and 238UO2/238U3O8.

This paper extends an examination of the influence of parameter uncertainties on regional doses to respiratory tract tissues for short-ranged alpha particles using the ICRP-66 respiratory tract model. Previous papers examined uncertainties in the deposition and clearance aspects of the model. The critical parameters examined in this study included target tissue depths, thicknesses, and masses, particularly within the thoracic or lung regions of the respiratory tract. Probability density functions were assigned for the parameters based on published data. The probabilistic computer code LUDUC (Lung Dose Uncertainty Code) was used to assess regional and total lung doses from inhaled aerosols of 239PuO2 and 238UO2/238U3O8. Dose uncertainty was noted to depend on the particle aerodynamic diameter. Additionally, dose distributions were found to follow a lognormal distribution pattern. For 239PuO2 and 238UO2/238U3O8, this study showed that the uncertainty in lung dose increases by factors of approximately 50 and approximately 70 for plutonium and uranium oxides, respectively, over the particle size range from 0.1 to 20 microm. For typical exposure scenarios involving both radionuclides, the ratio of the 95% dose fractile to the 5% dose fractile ranged from approximately 8-10 (corresponding to a geometric standard deviation, or GSD, of about 1.7-2) for particle diameters of 0.1 to 1 microm. This ratio increased to about 370 for plutonium oxide (GSD approximately 4.5) and to about 600 for uranium oxide (GSD approximately 5) as the particle diameter approached 20 microm. However, thoracic tissue doses were quite low at larger particle sizes because most of the deposition occurred in the extrathoracic airways. For 239PuO2, median doses from LUDUC were found be in general agreement with those for Reference Man (via deterministic LUDEP 2.0 calculations) in the particle range of 0.1 to 5 microm. However, median doses to the basal cell nuclei of the bronchial airways (BB(bas)) calculated by LUDUC were found to be approximately 6 times higher than LUDEP reference doses. The higher BB(bas) doses were directly attributed to discrepancies between the ICRP default thickness for the bronchial epithelium (55 microm) and the probability density function assumed within LUDUC (uniform distribution from 20 to 60 microm based upon detailed literature reviews).

Adolescent↗

Can dose-response models predict reliable normal tissue complication probabilities in radical radiotherapy of urinary bladder cancer? The impact of alternative radiation tolerance models and parameters.

PURPOSE: To analyze the consequences of selecting alternative normal tissue complication probability (NTCP) models and parameters for evaluation of radiotherapy of urinary bladder cancer. METHODS AND MATERIALS: Treatment plans of 24 bladder cancer patients referred to radical 4-field conformal radiotherapy were analyzed. Small intestinal and rectal NTCPs were determined using both the probit and relative seriality models with several sets of published radiation tolerance parameters. Various combinations of NTCP models and parameters were applied to find the prescription dose in individual patients as well as to estimate the benefit of the conformal radiotherapy setup. RESULTS: Different risk estimates were predicted from the two NTCP models, even when the same clinical radiation tolerance doses were fitted into the two models. The demonstrated variability translated into significant deviations (7-10 Gy) in the recommended prescription doses. Even if it was possible to discriminate between a 2-field plan and the 4-field conformal setup using a given complication model and set of tolerance parameters, the estimated benefit of the conformal treatment in terms of permitted dose escalation varied with as much as 10-12 Gy between the different NTCP models/parameters used. CONCLUSION: Different NTCP models and tolerance parameters might propose different answers to important clinical questions in radiotherapy treatment of bladder cancer, such as dose prescription and scoring of rival treatment plans. We therefore recommend that the variability caused by tolerance parameter uncertainty and model selection should be taken into account in dose-response modeling of radiotherapy treatment.

Aged↗

Enhanced coagulation for satisfying the arsenic maximum contaminant level under variable and uncertain conditions.

This study evaluated the effects of influent variability and model parameter uncertainty when utilizing enhanced coagulation modification to bring existing treatment plants into compliance with a stricter arsenic regulation. Enhanced coagulation modification options include: (1) increased ferric chloride dose, (2) addition of an acid dose, and (3) a combination of the individual options. Arsenic removal is described by adsorption to hydrous ferric oxide with a surface complexation model and subsequent removal through sedimentation and filtration. The least-cost modification for reliably satisfying the arsenic regulation is determined using an optimization algorithm that explicitly includes variability and uncertainty. The ferric chloride only modification is always the least-cost treatment modification. The ferric chloride and acid modification could be the least-cost option when considering waste handling processes due to a tradeoff between modification cost and sludge production. By inclusion of variability and uncertainty, the relative importance of individual parameter distributions for determining whether the arsenic regulation is reliably satisfied is assessed. Influent arsenic concentration variability is always critical, while variability in the influent pH and sulfate concentrations and uncertainty in the filter removal efficiency and equilibrium adsorption constant for the triple bond Fe(s)OHCa2+ surface species are critical or important, depending on influent conditions.

Adsorption↗

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

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

Adult↗

Electrical properties of lens material at microwave frequencies.

Values of the complex permittivity of human and rabbit lens material in the frequency range 100--6000 MHz are reported. The data were obtained by using an on-line computer-based time-domain spectrometer. The lenses were divided into an outer (cortical) zone and an inner (nuclear) zone and the dielectric properties of each zone were measured separately for both species of lens. The results are analysed in terms of the aqueous and protein constituents, assuming a molecular model whereby the hydrated protein molecule is represented by a spherical particle embedded in an aqueous continuum. It is shown that for lens material taken from the nuclear zone the particle can be separated into protein and bound water (water of hydration). For the cortical zone the amount of water of hydration is smaller and, within the limitations of the model and uncertainties due to experimental error, cannot be distinguished from zero.

Animals↗

Effects of aluminum on rat bone cell populations.

Aluminum (Al) loading is associated with reduced bone formation and osteomalacia in human and certain animal models. However, uncertainty exists as to the cellular effect(s) of Al as both inhibition and stimulation of osteoblast proliferation have been reported. Furthermore, the extent to which Al affects osteoprogenitor cell populations is unknown. To determine the cellular effects of Al in the rat, an animal model in which Al bone disease has been produced, we compared the in vitro effect of 10-50 microns Al on the proliferation and hydroxyproline collagen formation of marrow osteoprogenitor stromal cell populations and perinatal rat calvarial osteoblasts. In subconfluent cultures, Al suppressed proliferation of both marrow fibroblast-like stromal cells and calvarial osteoblasts. In confluent cultures, however, Al selectively stimulated periosteal fibroblast and osteoblast DNA synthesis and collagen (hydroxyproline) production, both in the presence or absence of 1,25-dihydroxyvitamin D. Osteocalcin was not detected in osteoblast-conditioned media or extracellular matrix. These observations suggest that the bone formation defect associated with Al toxicity in growing rats may be a function of impaired patterns of osteoprogenitor/osteoblast proliferation. Furthermore, the Al-stimulated increase in collagen formation is consistent with the development of osteomalacia in Al-toxic humans and animals. The mechanism by which Al stimulated DNA synthesis and collagen production in more mature cultures awaits further study.

Aluminum↗

Greener by design.

Process simulation models and other design tools allow engineers to design, simulate, and optimize chemical processes. However, there is a critical need to incorporate green engineering into the design of these processes. This calls for extending the breadth of the design process. This paper presents an integrated framework for greener design. The framework starts the decision-making as early as the chemical and material selection stage and also includes management and planning decisions. The design goal is not restricted to profitability, but environmental and ecological objectives are also added. However, this integration poses challenging problem of discrete and continuous decisions, nonlinear models, and uncertainties. Furthermore, there are multiple and conflicting objectives to be considered. Therefore, the core of this integrated framework is the efficient algorithmic framework for multiobjective optimization under uncertainty. Two real world case studies are presented that illustrate the promise of such a framework.

Algorithms↗

This is more difficult than we thought! The responsibility of scientists, managers and stakeholders to mitigate the unsustainability of marine fisheries.

The management of marine fisheries needs to undergo dramatic change in the new millennium, in response to the well-documented evidence of global overfishing and the general depletion of commercial fish stocks. The axioms of sustainable development and equilibrium productivity of wild ecosystems are identified as misleading concepts, which nonetheless underlie current approaches to the management of living marine resources. Current trends in marine fisheries landings worldwide provide little evidence of sustainability of marine resources under current management paradigms, where biological, economic and social aspects of fisheries are usually treated as different disciplines. While open-access conditions are less widespread than formerly, except for many straddling and highly migratory resources, fishers usually have access to the resource year-round throughout its range. Despite quotas, the nominal control of capacity and technical measures protecting juveniles, top-down management has generally been unable to prevent stock depletion, particularly of the older spawners that for demersal stocks often support recruitment. An integrated solution to the complexity of managing wild resources seems not to have been achieved. Any new paradigm should assert the basic unpredictability of fisheries at the system level and require a broader range of performance indicators to be incorporated into the decisional framework. This must reflect the non-equilibrium nature of marine systems, and give greater importance to resource (as opposed to harvest) continuity in the face of regime shifts, and promote habitat restoration and conservation of genetic resources. The new management framework requires co-management and collective decision-making to be incorporated within a precautionary and pre-negotiated management framework. This must explicitly recognize that decision-making occurs in conditions of model-based uncertainty and precautionary approaches should be incorporated at all levels, not least of which is to avoid the assumption that all resources can be harvested in a sustainable fashion through time. Redundancy in data inputs to management are needed to avoid the surprises that model-based sampling occasionally leads to, for example, when regime changes reduce productivity in response to climatic fluctuations. Emergency frameworks imposing non-discretionary rules must be invoked when overfishing and/or regime change trigger reference points indicating stock depletion. Non-discretionary recovery plans should then override rights-based systems and persist until fish populations recover to pre-established healthy levels, which may in turn need to await the return of a favourable regime. In fact, some stocks may require periodic rebuilding after regime-induced collapses or because of a combination of ecological or economic impacts, hence a constant harvest policy may not always be possible. It will probably also be necessary to discard the axiom that a stock should be available to harvesting throughout its range and seasonal cycle. Technological advances mean that time- and area-specific access rights are now practical options, through satellite monitoring of vessel operations, even offshore. More fundamentally, the basic axiom of "enlightened self interest" underlying current methods of management will need to be tempered by an increased ethical concern for the fragility of natural ecosystems.

Animals↗

Human alimentary tract model for radiological protection. ICRP Publication 100. A report of The International Commission on Radiological Protection.

In this report, the ICRP provides a new biokinetic and dosimetric model of the human alimentary tract to replace the Publication 30 (ICRP, 1979) model. The new human alimentary tract model (HATM) will be used together with the human respiratory tract model (HRTM; ICRP, 1994a,b) in future ICRP publications on doses from ingested and inhaled radionuclides. The HATM is applicable to all situations of radionuclide intake by children and adults. It provides age-dependent parameter values for the dimensions of the alimentary tract regions, and associated transit times for the movement of materials through these regions. For adults, gender-dependent parameter values are given for dimensions and transit times. The default assumption is that radionuclide absorption takes place in the small intestine, but the model allows for absorption in other regions and for retention in or on tissues within the alimentary tract when information is available. Doses are calculated to target cells for cancer induction in the oral cavity, oesophagus, stomach, small intestine, and colon. This report provides reviews of information on the transit of materials through the alimentary tract and on radionuclide retention and absorption. It considers data on health effects, principally in order to specify the target cells for cancer induction within the mucosal lining of the tract and to justify approaches taken to dose averaging within regions. Comparisons are made between doses calculated using the HATM and the Publication 30 model for examples of radionuclide ingestion for which absorption is assumed to occur in the small intestine alone. Examples are also given of the effects on doses of considering absorption from other regions and the effect of possible retention in the alimentary tract. This report also considers uncertainties in model assumptions and their effect on doses, including alimentary tract dimensions, transit times, radionuclide absorption values, and the location of targets for cancer induction.

Digestive System Physiological Phenomena↗

Evaluation of the relative cost-effectiveness of treatments for infertility in the UK.

This paper aims to complement existing clinical guidelines by providing evidence of the relative cost-effectiveness of treatments for infertility in the UK. A series of decision-analytical models have been developed to reflect current diagnostic and treatment pathways for the five main causes of infertility. Data to populate the models are derived from a systematic review and routine National Health Service activity data, and are augmented with expert opinion. Costs are derived from an analysis of extra-contractual referral tariffs and private sector data. Sensitivity analysis has been carried out to take account of the uncertainty of model parameters and to allow results to be interpreted in the light of local circumstances. Results of the modelling exercise suggest in-vitro fertilization is the most cost-effective treatment option for severe tubal factors and endometriosis, with surgery the most cost-effective in the case of mild or moderate disease. Ovulatory factors should be treated medically with the addition of laparoscopic ovarian diathermy in the presence of polycystic ovarian syndrome. For other causes, stimulated intrauterine insemination (unexplained and moderate male factor) and stimulated donor intrauterine insemination (severe male) are cost-effective.

Cost-Benefit Analysis↗

An AIDS model with distributed incubation and variable infectiousness: applications to i.v. drug users in Latium, Italy.

An AIDS model with distributed incubation and variable infectiousness is considered and simulated via a second-order numerical method. The method is applied to the HIV epidemic among IV drug users in the Latium region of Italy, using available data on the length of the incubation period before the onset of AIDS, on the infectivity of infected individuals during that period, and on the demography of drug users. The contact rate is adjusted to match the actual number of AIDS cases. The sensitivity of the model to uncertainties in the parameters is finally investigated, by performing several simulations.

Acquired Immunodeficiency Syndrome↗

Uncertainty in estimating exposure using a toxicokinetic model. The example of 2,3,7,8-tetrachlorodibenzo-p-dioxin.

This paper deals with sources of uncertainty in the use of a minimal physiological toxicokinetic model to obtain dose estimates for a dose-response analysis of cancer in an occupational cohort. Toxicokinetic models make it possible to construct exposure parameters that are more closely related to the individual dose than traditional measures of exposures to toxic agents. However, the process introduces a wide array of sources of uncertainty. Selecting a model structure to describe the kinetics of a toxic agent implies necessarily making simplifications and assumptions that influence the range of applicability of the model. Once a model has been selected, the value of certain model parameters (constants) must be assigned, for example, from anthropometric data. The question then arises of how sensitive the model predictions are to variations in the values of these constants. Other model parameters, typically those describing the kinetics of the agent, are next estimated from actual data. There may be limitations in the data concerning, for example, sparseness (too few observations per subject) or missing values. The methods used for parameter estimation carry their own set of assumptions that need to be appropriate to the situation at hand. In summary, the dioxin example is used to characterize the sources of uncertainty at different levels, such as model structure, methods and data used for parameter estimation, estimation of occupational exposure, and imputation of missing values in exposure indices derived from the kinetic model.

Environmental Exposure↗

Variational Bayes for continuous hidden Markov models and its application to active learning.

In this paper, we present a varitional Bayes (VB) framework for learning continuous hidden Markov models (CHMMs), and we examine the VB framework within active learning. Unlike a maximum likelihood or maximum a posteriori training procedure, which yield a point estimate of the CHMM parameters, VB-based training yields an estimate of the full posterior of the model parameters. This is particularly important for small training sets since it gives a measure of confidence in the accuracy of the learned model. This is utilized within the context of active learning, for which we acquire labels for those feature vectors for which knowledge of the associated label would be most informative for reducing model-parameter uncertainty. Three active learning algorithms are considered in this paper: 1) query by committee (QBC), with the goal of selecting data for labeling that minimize the classification variance, 2) a maximum expected information gain method that seeks to label data with the goal of reducing the entropy of the model parameters, and 3) an error-reduction-based procedure that attempts to minimize classification error over the test data. The experimental results are presented for synthetic and measured data. We demonstrate that all of these active learning methods can significantly reduce the amount of required labeling, compared to random selection of samples for labeling.

Algorithms↗

The effects of nucleotide substitution model assumptions on estimates of nonparametric bootstrap support.

The use of parameter-rich substitution models in molecular phylogenetics has been criticized on the basis that these models can cause a reduction both in accuracy and in the ability to discriminate among competing topologies. We have explored the relationship between nucleotide substitution model complexity and nonparametric bootstrap support under maximum likelihood (ML) for six data sets for which the true relationships are known with a high degree of certainty. We also performed equally weighted maximum parsimony analyses in order to assess the effects of ignoring branch length information during tree selection. We observed that maximum parsimony gave the lowest mean estimate of bootstrap support for the correct set of nodes relative to the ML models for every data set except one. For several data sets, we established that the exact distribution used to model among-site rate variation was critical for a successful phylogenetic analysis. Site-specific rate models were shown to perform very poorly relative to gamma and invariable sites models for several of the data sets most likely because of the gross underestimation of branch lengths. The invariable sites model also performed poorly for several data sets where this model had a poor fit to the data, suggesting that addition of the gamma distribution can be critical. Estimates of bootstrap support for the correct nodes often increased under gamma and invariable sites models relative to equal rates models. Our observations are contrary to the prediction that such models cause reduced confidence in phylogenetic hypotheses. Our results raise several issues regarding the process of model selection, and we briefly discuss model selection uncertainty and the role of sensitivity analyses in molecular phylogenetics.

Animals↗

Upscaling of hydraulic conductivity and telescopic mesh refinement.

Performance assessments of repositories for the underground disposal of nuclear fuel and waste include models of ground water flow and transport in the host rocks. Estimates of hydraulic conductivity, K, based on field measurements may require adjustment (upscaling) for use in numerical models, but the choice of upscaling approach can be complicated by the use of nested modeling, large-scale fracture zones, and a high degree of heterogeneity. Four approaches to upscaling K are examined using a reference case based on exhaustive site data and an application of nested modeling to evaluate performance assessment of a waste repository. The upscaling approaches are evaluated for their effects on the flow balance between nested modeling domains and on simple measures of repository performance. Of the upscaling approaches examined in this study, the greatest consistency of boundary flows was achieved using the observed scale dependence for the rock domains, measured values from the large-scale interference test for the conductor domain, and a semivariogram regularization based on the Moye model for packer test interpretation. Making the assumption that large fracture zones are two-dimensional media results in the greatest changes to the median of travel time and improves the flow balance between the nested models. The uncertainty of upscaling methods apparently has a small impact on median performance measures, but a significant impact on the variances and earliest arrival times.

Models, Theoretical↗

Uncertainty in eddy covariance measurements and its application to physiological models.

Flux data are noisy, and this uncertainty is largely due to random measurement error. Knowledge of uncertainty is essential for the statistical evaluation of modeled and measured fluxes, for comparison of parameters derived by fitting models to measured fluxes and in formal data-assimilation efforts. We used the difference between simultaneous measurements from two towers located less than 1 km apart to quantify the distributional characteristics of the measurement error in fluxes of carbon dioxide (CO2) and sensible and latent heat (H and LE, respectively). Flux measurement error more closely follows a double exponential than a normal distribution. The CO2 flux uncertainty is negatively correlated with mean wind speed, whereas uncertainty in H and LE is positively correlated with net radiation flux. Measurements from a single tower made 24 h apart under similar environmental conditions can also be used to characterize flux uncertainty. Uncertainty calculated by this method is somewhat higher than that derived from the two-tower approach. We demonstrate the use of flux uncertainty in maximum likelihood parameter estimates for simple physiological models of daytime net carbon exchange. We show that inferred model parameters are highly correlated, and that hypothesis testing is therefore possible only when the joint distribution of the model parameters is taken into account.

Carbon Dioxide↗

Indicators for hydraulic and pollution retention assessment of stormwater infiltration basins.

Infiltration basins are frequently used for stormwater management even though their long-term evolution is not well understood nor controlled. The two main problems encountered are clogging which compromises the hydraulic capacity of the basin and possible contamination of underlying soil and groundwater. This paper defines a framework for evaluating the hydraulic and pollution retention performance of infiltration basins in the long-term. Sets of context and performance indicators are proposed, along with two complementary modes of evaluation. Context indicators are identified in order to define the clogging and contamination states of the basins. Performance indicators are developed to assess several aspects of basin performance: drainage duration, overflow frequency, predictive life period, particle filtration and pollution trapping. Modes of evaluation include field investigation and long-term simulation modeling. Indicators are tested on five infiltration basins in suburban Lyon (France). Both context indicators and hydraulic performance indicators are reliable and their evaluation is representative of basin behavior. This is not the case for pollution retention performance indicators. Their assessment is difficult because of data quality. Field data has high uncertainties. The model is satisfactory for the hydraulic simulation and the evolution of clogging. Improvements are necessary for pollution flow simulation and the acquisition of better quality data is required.

Conservation of Natural Resources↗