China
Africa
Explore chapters and articles related to this topic
Characterization of Uncertainty
Published in Samuel C. Morris, Cancer Risk Assessment, 2020
Uncertainty analysis is the determination of the variation or imprecision in quantitative conclusion of a risk assessment model (e.g., the predicted exposure or effects) that results from the collective variation associated with the variables in the model (Iman and Helton, 1988). Methods used for uncertainty analysis in computerized risk assessment models, as described by Iman and Helton (1988) include: Response surface replacement, wherein the model is evaluated at many selected sets of input-parameter values and the results fitted in a general linear model which is then used as a replacement for the computer model. The time or cost involved in the large number of computer runs required to obtain sufficient data to fit the linear model limit the applicability of this techniqueModified Monte Carlo or latin hypercube sampling, wherein the range of each variable is divided into nonoverlapping intervals with equal probability and one value is selected at random from each interval. From these values, sets of input parameters are selected randomly to form the Latin Hypercube sample that is then used for the uncertainty analysisDifferential analysis, wherein a Taylor series approximation is used in conjunction with Monte Carlo analysis to estimate distribution functions
Radionuclide Transport Processes and Modeling
Published in Michael Pöschl, Leo M. L. Nollet, Radionuclide Concentrations in Food and the Environment, 2006
An estimation of the uncertainty model predictions can be obtained by different methods, including variance propagation, moment matching, and numerical methods known as Monte Carlo methods (e.g., simple random sampling or Latin hypercube sampling). Monte Carlo methods are widely used in environmental modeling, especially for complex models.
Patient-specific pre-operative simulation of the surgically assisted rapid maxillary expansion using finite element method and Latin hypercube sampling: workflow and first clinical results
Published in Computer Methods in Biomechanics and Biomedical Engineering, 2023
L. Bonitz, A. Volf, S. Hassfeld, A. Pugachev, B. Ludwig, S. Chhatwani, A. Bicsák
The sampling and sensitivity analysis was performed using the optiSLang 7.2.0 (Dynardo GmbH, Weimar, Germany). The lengths of the four horizontal cuts (Figure 5) were used as design variables. The design variables were constrained by the side constraints defined by the surgeon. The Latin hypercube sampling method (LHS) was used to generate a large number of designs for evaluation. The LHS is a statistical method for generating a near-random sample of parametric values from a multidimensional distribution. The sampling method is often used to construct computer experiments or Monte Carlo integration. Each design was transferred to ANSYS and simulated using the finite element model described above. The optimization environment was based on the automatic interaction between ANSYS and optiSLang.
Influence of population variability in ligament material properties on the mechanical behavior of ankle: a computational investigation
Published in Computer Methods in Biomechanics and Biomedical Engineering, 2020
Yuanjie Liu, Qing Zhou, Shun Gan, Bingbing Nie
Probabilistic analysis was conducted using Latin Hypercube (LH) as the sampling method (Mckay et al. 2000). LH sampling is a stratified sampling method, which can cover the whole sampling space with reduced sampling points. The sampling principle is briefly explained as follows. Assume that
Modeling gold nanoparticle biodistribution after arterial infusion into perfused tissue: effects of surface coating, size and protein corona
Published in Nanotoxicology, 2018
Jim E. Riviere, Majid Jaberi-Douraki, James Lillich, Tahmineh Azizi, Hyun Joo, Kyoungju Choi, Ravi Thakkar, Nancy A. Monteiro-Riviere
An important characteristic of modeling biological systems is determining how sensitive (or unreliable) different components of the model are to disturbances and noises. The Latin Hypercube Sampling (LHS) method in combination with Partial Rank Correlation Coefficient (PRCC) (Marino et al. 2008; Jaberi-Douraki, Pietropaolo, and Khadra 2015, Jaberi-Douraki, Schnell, et al. 2015) in a four-dimensional parameter space, Figure 2. These methods evaluate the robustness of the parameter ranges in terms of bias to initial estimates, sensitivity, and accuracy. For simulation purposes, we applied 1995; Mickens 2005; Jaberi-Douraki, Schnell, et al. 2015). The correlations between inputs (parameters) and outputs (simulations) of the model are then measured based on the rank transforms