Individual Parameter Estimates Research Articles

Design optimisation for pharmacokinetic modeling of a cocktail of phenotyping drugs.

Our paper proposes a methodological strategy to select optimal sampling designs for phenotyping studies including a cocktail of drugs. A cocktail approach is of high interest to determine the simultaneous activity of enzymes responsible for drug metabolism and pharmacokinetics, therefore useful in anticipating drug-drug interactions and in personalized medicine. Phenotyping indexes, which are area under the concentration-time curves, can be derived from a few samples using nonlinear mixed effect models and maximum a posteriori estimation. Because of clinical constraints in phenotyping studies, the number of samples that can be collected in individuals is limited and the sampling times must be as flexible as possible. Therefore to optimize joint design for several drugs (i.e., to determine a compromise between informative times that best characterize each drug's kinetics), we proposed to use a compound optimality criterion based on the expected population Fisher information matrix in nonlinear mixed effect models. This criterion allows weighting different models, which might be useful to take into account the importance accorded to each target in a phenotyping test. We also computed windows around the optimal times based on recursive random sampling and Monte-Carlo simulation while maintaining a reasonable level of efficiency for parameter estimation. We illustrated this strategy for two drugs often included in phenotyping cocktails, midazolam (probe for CYP3A) and digoxin (P-glycoprotein), based on the data of a previous study, and were able to find a sparse and flexible design. The obtained design was evaluated by clinical trial simulations and shown to be efficient for the estimation of population and individual parameters.

Combining Samples from Multiple Gears Helps to Avoid Fishy Growth Curves

AbstractSize‐at‐age information is critical in estimating growth parameters (e.g., the von Bertalanffy growth function [VBGF]) that are used to assess fish populations. Due to gear selectivity, single sampling methods rarely sample all ages or all sizes equally well. Most growth estimates rely on samples from a single gear or a haphazard combination of gears, potentially leading to biased and imprecise growth parameter estimates. We evaluated the efficacy of combining samples from two gears with different size selectivity to estimate VBGF parameters; we then applied that approach to a case study on the Lochloosa Lake (Florida) population of Black Crappies Pomoxis nigromaculatus. Simulated age‐ and size‐structured populations were randomly sampled with two gears characterized by different size‐selectivity curves (one gear was selective for smaller fish; the other was selective for larger fish). Maximum likelihood VBGF estimates obtained for each gear separately were compared with estimates from a combined VBGF fitted to data from both gears. In every simulated scenario, a combined‐gear approach reduced bias and increased precision for estimating the VBGF, but the gear‐specific proportions that improved VBGF estimates depended on size selectivity. The VBGF estimates for the Black Crappie population showed that the combined‐gear method yielded intermediate parameter values relative to single‐gear approaches based on (1) trawl sampling (fishery‐independent survey) and (2) angler harvest (as determined from carcass collections; fishery‐dependent data). Furthermore, the combined‐gear approach had greater precision in individual parameter estimates and much less variance than single‐gear approaches when estimating the VBGF. Combining data from two gears can increase sample representativeness, leading to improvements in VBGF estimation. Such approaches can reduce uncertainty in VBGF estimation and can provide insight into key demographic processes occurring in fish populations for which ontogeny and gear selectivity lead to imperfect sampling.Received January 22, 2015; accepted July 30, 2015

The temporal stability of factors affecting driver-injury severities in single-vehicle crashes: Some empirical evidence

This study explores the temporal stability of factors affecting driver-injury severities in single-vehicle crashes. Using data for single-vehicle crashes in Chicago, Illinois from a nine-year period from January 1, 2004 to December 31, 2012, separate annual models of driver-injury severities (with possible outcomes of severe injury, minor injury, and no injury) were estimated using a mixed logit model to capture potential unobserved heterogeneity. Likelihood ratio tests were conducted to examine the overall stability of model estimates across time periods and marginal effects of each explanatory variable were also considered to investigate the temporal stability of the effect of individual parameter estimates on injury-severity probabilities. A wide range of variables potentially affecting injury severities was considered including driver-contributing factors, location and time of day, crash-specific factors, driver attributes, roadway characteristics, environmental conditions, and vehicle characteristics. The results indicated that, although data from different years share some common features, the model specifications and estimated parameters are not temporally stable. In addition, complex temporal stability behaviors were observed for individual parameter estimates such as driver gender, apparent physical condition of driver, type of vehicle, vehicle occupancy, road surface, weather, and light conditions. It is speculated that this temporal instability could be a function of the urban nature of the data, possible variations in police-reporting of crash determinants over time, the impact of continuing improvements in vehicle safety features and drivers’ response to them, and/or the effects of macroeconomic instability that was present over the time period considered in this study. Although the source of temporal stability is not clearly known, the general subject of temporal instability warrants substantial attention in future research. The possible presence of temporal instability in injury-severity models can have significant consequences in highway-safety practice where accurate forecasting of the impacts of alternative safety countermeasures is sought.

Abstract 1496: Quantification of tumor blood perfusion of an orthotopic mouse model of neuroblastoma using nonlinear contrast-enhanced ultrasound imaging

Abstract This study quantifies tumor perfusion in individual tumors to estimate blood flow and blood volume parameters of an individualized tumor compartment of a comprehensive physiologically-based pharmacokinetic model of topotecan using an orthotopic xenograft model of pediatric neuroblastoma. We non-invasively imaged perfusion in orthotopic neuroblastoma (NB5) xenograft tumors (n = 3 CD1 nude mice/time point) using nonlinear contrast enhanced ultrasound technique (CEUS). Tumor tissue and organs from the mice were harvested at predefined time-points. We used a programmable syringe pump to inject MicroMarker® microbubbles via tail vein catheter and acquired images using VisualSonics VEVO 2100 imaging system. We used the burst-replenishment technique to image tumor perfusion, which requires a constant concentration of microbubbles in blood during acquisition. To maintain a steady concentration of microbubbles, we programmed the pump to inject a small bolus followed by constant infusion. Our preliminary analysis showed that healthy kidneys rapidly reach a steady state in less than 1 min, significantly shorter than the commonly used constant infusion without an initial bolus. The nonlinear CEUS signal intensities of kidney cortex showed less than 20% variation between mice. We used a custom program to acquire the CEUS perfusion images over a 3D volume that included the tumor and a kidney. We used the kidney as a reference organ to normalize whole tumor perfusion data. We fitted the log-normal perfusion model to estimate perfusion parameters for individual tumors. Our perfusion quantification over the entire tumor volume represents tumor perfusion more accurately than the commonly used methods based on a single 2D plane without a reference organ. Our approach provides population estimates of blood perfusion based on properly normalized estimates of individual blood perfusion parameters. Citation Format: Suresh Thiagarajan, Abbas Shirinifard, Megan O. Jacus, Abigail D. Davis, Yogesh T. Patel, Stacy L. Throm, Vinay Daryani, Clinton F. Stewart, András Sablauer. Quantification of tumor blood perfusion of an orthotopic mouse model of neuroblastoma using nonlinear contrast-enhanced ultrasound imaging. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1496. doi:10.1158/1538-7445.AM2015-1496

Validation of a pediatric population pharmacokinetic model for vancomycin.

Vancomycin is often required to treat serious infections in children, including methicillin-resistant Staphylococcus aureus infections. The pharmacodynamic index that best predicts efficacy with vancomycin use for methicillin-resistant S. aureus infection in adults is the 24-hour area under the curve (AUC) over the minimum inhibitory concentration. Although multiple pediatric population pharmacokinetic (PK) vancomycin models have been published, few use Bayesian optimization. The current standard of care remains measuring vancomycin serum trough concentrations as a surrogate marker of AUC. A prospective validation of a pediatric population PK model of vancomycin was performed. Hospitalized children younger than 18 years receiving vancomycin at Cincinnati Children's Hospital Medical Center were invited to participate. Peak, trough, and random vancomycin serum concentrations were used for Bayesian estimation of individual PK parameters. Model covariates included age, weight, and serum creatinine. To evaluate the predictive performance of the model, precision and bias were measured and compared using the 95% confidence interval. Fifteen subjects were enrolled; 13 subjects had vancomycin serum concentrations drawn per protocol. Of those 13 subjects, the median age was 6 years and 54% were male. Significant medical conditions included cancer (54%), lung disease (23%), neurologic disorders (23%), and prior transplantation (15%). The initial serum creatinine was normal (median, 0.33; interquartile range, 0.23-0.4 mg/dL), and none had underlying renal dysfunction. Equivalence of bias and precision between the original model validation and the Cincinnati Children's Hospital Medical Center validation were found. Pediatric population PK models for vancomycin with Bayesian estimation can be used to reliably predict vancomycin exposure in children. Using AUC instead of trough serum concentrations alone can provide an opportunity to maximally optimize vancomycin administration in children.

Effort, success, and nonuse determine arm choice.

How do humans choose one arm or the other to reach single targets in front of the body? Current theories of reward-driven decisionmaking predict that choice results from a comparison of "action values," which are the expected rewards for possible actions in a given state. In addition, current theories of motor control predict that in planning arm movements, humans minimize an expected motor cost that balances motor effort and endpoint accuracy. Here, we test the hypotheses that arm choice is determined by comparison of action values comprising expected effort and expected task success for each arm, as well as a handedness bias. Right-handed subjects, in either a large or small target condition, were first instructed to use each hand in turn to shoot through an array of targets and then to choose either hand to shoot through the same targets. Effort was estimated via inverse kinematics and dynamics. A mixed-effects logistic-regression analysis showed that, as predicted, both expected effort and expected success predicted choice, as did arm use in the preceding trial. Finally, individual parameter estimation showed that the handedness bias correlated with mean difference between right- and left-arm success, leading to overall lower use of the left arm. We discuss our results in light of arm nonuse in individuals' poststroke.

An improved cognitive model of the Iowa and Soochow Gambling Tasks with regard to model fitting performance and tests of parameter consistency.

The Iowa Gambling Task (IGT) and the Soochow Gambling Task (SGT) are two experience-based risky decision-making tasks for examining decision-making deficits in clinical populations. Several cognitive models, including the expectancy-valence learning (EVL) model and the prospect valence learning (PVL) model, have been developed to disentangle the motivational, cognitive, and response processes underlying the explicit choices in these tasks. The purpose of the current study was to develop an improved model that can fit empirical data better than the EVL and PVL models and, in addition, produce more consistent parameter estimates across the IGT and SGT. Twenty-six opiate users (mean age 34.23; SD 8.79) and 27 control participants (mean age 35; SD 10.44) completed both tasks. Eighteen cognitive models varying in evaluation, updating, and choice rules were fit to individual data and their performances were compared to that of a statistical baseline model to find a best fitting model. The results showed that the model combining the prospect utility function treating gains and losses separately, the decay-reinforcement updating rule, and the trial-independent choice rule performed the best in both tasks. Furthermore, the winning model produced more consistent individual parameter estimates across the two tasks than any of the other models.

Summary statistics from training images as prior information in probabilistic inversion

AbstractA strategy is presented to incorporate prior information from conceptual geological models in probabilistic inversion of geophysical data. The conceptual geological models are represented by multiple-point statistics training images (TIs) featuring the expected lithological units and structural patterns. Information from an ensemble of TI realizations is used in two different ways. First, dominant modes are identified by analysis of the frequency content in the realizations, which drastically reduces the model parameter space in the frequency-amplitude domain. Second, the distributions of global, summary metrics (e.g. model roughness) are used to formulate a prior probability density function. The inverse problem is formulated in a Bayesian framework and the posterior pdf is sampled using Markov chain Monte Carlo simulation. The usefulness and applicability of this method is demonstrated on two case studies in which synthetic crosshole ground-penetrating radar traveltime data are inverted to recover 2-D porosity fields. The use of prior information from TIs significantly enhances the reliability of the posterior models by removing inversion artefacts and improving individual parameter estimates. The proposed methodology reduces the ambiguity inherent in the inversion of high-dimensional parameter spaces, accommodates a wide range of summary statistics and geophysical forward problems.

Evaluation and improvement of the CLIGEN model for storm and rainfall erosivity generation in Central Chile

CLIGEN (CLImate GENerator) is a stochastic weather generator that produces daily estimates of precipitation and individual storm parameters, including time to peak, peak intensity and storm duration. These parameters are typically used as inputs for other models, such as the Water Erosion Prediction Project (WEPP) model. Although CLIGEN has been proven to be effective for predicting daily estimates, some discrepancies have been observed when generating storm parameters, such as the storm duration. Therefore, a study was conducted to evaluate and improve CLIGEN for storm generation. Individual rainfall events were identified from 1-h pluviograph records that were collected from 30 sites in Central Chile. In this study, 415years of data were used; 18,012 storms were analyzed. In addition, rainfall erosivity was computed for all storms using the prescribed method to compare the energy provided by the measured and generated rainfall events. Using measured rainfall data, a procedure was developed to improve the CLIGEN estimates by calibrating the input parameter that controls the storm durations. This procedure in turn improved the rainfall intensities and erosivities. The model was tested before and after calibration with the measured rainfall data from the 30 sites in both the wet and dry seasons. Based on a monthly rainfall analysis, the results demonstrated that the number of storms and rainfall amounts, which are not affected by the calibration process, were accurately estimated with CLIGEN. However, before the calibration, especially in the wet season, the storm durations and maximum intensities were consistently overestimated and underestimated, respectively, at most of the sites and for most months. Therefore, the annual rainfall erosivities were underestimated with CLIGEN at 19 of the 30 sites. After performing the calibration, the R2 value for the CLIGEN-generated storm durations increased from 0.41 to 0.65. The maximum intensities also exhibited an improvement; the R2 value increased from 0.31 to 0.60. Consequently, annual rainfall erosivities were generated with an R2 value of 0.89; these erosivities were accurately estimated at 29 of the 30 sites. Therefore, this calibration procedure proved to be an effective alternative for generating more reliable storm patterns. This paper explains the procedure in detail and analyzes the parameters related to the individual storm generation process.

スキルの習熟度モデルの構築と習熟過程の考察

To facilitate optimal skill acquisition process, modeling and estimation of individual learning processes will be of key importance. We have proposed a model of skill acquisition based on control theory which enable us to model individual learning process with three parameters; system gain K, time constant T, and dead time L. We measured individual performance time of fault finding task on software programming and successfully obtained the three parameters. These results indicate the feasibility of online estimation of individual learning parameters.

Fractionally Integrated Panel Data Systems

We consider large n, T panel data models with fixed effects, persistent common factors allowing for cross-section dependence, and possibly correlated persistent innovations that are assumed fractionally integrated. In the model, a) persistence in the innovations and the common factor allows for cointegrating relationships in the unobserved idiosyncratic components; b) the vector of innovations, whose elements can be contemporaneously correlated, can also exhibit short-memory dynamics in the form of a finite-order vector autoregressive process; c) deterministic linear and nonlinear trends can be nested. We perform estimations on the defactored observed series where the projections are based on the sample averages of fractionally differenced data. In the estimation, we use a computationally convenient equation-by-equation conditional-sum-of-squares (CSS) criterion, leading to GLS-type estimates for slope parameters. The CSS estimates of individual slope and long-range dependence parameters are root-T consistent while the mean-group slope estimate is root-n consistent, all with centered asymptotic normal distributions, irrespective of cointegration. A study of small-sample performance is carried out showing desirable properties for our estimation method. Finally, an empirical application to the long-run relationship between real GDP and debt is included.

Bayesian Information for Sensors

An engineering company manufacturing high‐precision sensors had accumulated huge historical databases of information on a type of sensors, which had been tested. The aim of the company was not to use this historical data to improve estimation of future individual sensor parameters, but rather to use it to reduce the number of measurements needed per sensor, guaranteeing a required level of accuracy. In the paper, we show how this can be performed, using Bayesian ideas, and introduce the novel theory for linear regression models, which determines how the reduction in individual sensor measurements can be achieved.Specifically, for estimating parameters of closely related sensors, an estimate can be thought of as comprising a global component, that is, the mean of all the sensors, and a local component, which is a shift from the mean. The historical data can, in a Bayesian framework, provide the global component and hence all that is needed from an individual sensor is the local component. In non‐Bayesian estimation methods, both components are required, and hence, many measurements are needed. On the other hand, with Bayesian methods, only the local fit is needed, and hence, fewer measurements per sensor are required. We provide the supporting theory and demonstrate on a real‐life application with real data. Copyright © 2014 John Wiley & Sons, Ltd.

Comparing Consumer Preferences for Livestock Production Process Attributes Across Products, Species, and Modeling Methods

Consumer preferences for four livestock products were investigated to determine consumer willingness to pay (WTP) for livestock production process attributes. We use an inferred method of attribute nonattendance (ANA) using the coefficient of variation on individual specific parameter estimates to assess the variability of preference intensity for various product characteristics. We find that accounting for ANA did not significantly impact mean estimates of WT P. Implications of our findings on the reliability of existing work in the area of consumer preferences for animal welfare attributes are discussed.

A Factor-Analytic Model for Assessing Individual Differences in Response Scale Usage

This article proposes a factor-analytic model, intended for graded-response or continuous-response personality and attitude items, which includes an additional multiplicative person parameter that models the individual's response mapping process. The model, which is a modification of Spearman's (1904) factor analysis (FA) model, is parameterized as both an FA model and an item response theory (IRT) model and is fully developed to the extent that it can be used in applications. Procedures for (a) calibrating the items and assessing data fit, (b) obtaining individual estimates of both person parameters, (c) determining measurement precision, and (d) assessing differential predictability are proposed and discussed. The potential advantages of the proposal, its practical relevance, and its relations with other approaches are also discussed. Its functioning is assessed with a simulation study and 3 empirical examples in the personality domain.

Designing driver assistance systems with crossmodal signals: multisensory integration rules for saccadic reaction times apply.

Modern driver assistance systems make increasing use of auditory and tactile signals in order to reduce the driver's visual information load. This entails potential crossmodal interaction effects that need to be taken into account in designing an optimal system. Here we show that saccadic reaction times to visual targets (cockpit or outside mirror), presented in a driving simulator environment and accompanied by auditory or tactile accessories, follow some well-known spatiotemporal rules of multisensory integration, usually found under confined laboratory conditions. Auditory nontargets speed up reaction time by about 80 ms. The effect tends to be maximal when the nontarget is presented 50 ms before the target and when target and nontarget are spatially coincident. The effect of a tactile nontarget (vibrating steering wheel) was less pronounced and not spatially specific. It is shown that the average reaction times are well-described by the stochastic “time window of integration” model for multisensory integration developed by the authors. This two-stage model postulates that crossmodal interaction occurs only if the peripheral processes from the different sensory modalities terminate within a fixed temporal interval, and that the amount of crossmodal interaction manifests itself in an increase or decrease of second stage processing time. A qualitative test is consistent with the model prediction that the probability of interaction, but not the amount of crossmodal interaction, depends on target–nontarget onset asynchrony. A quantitative model fit yields estimates of individual participants' parameters, including the size of the time window. Some consequences for the design of driver assistance systems are discussed.

Parameter‐based assessment of disturbed and intact components of visual attention in children with developmental dyslexia

People with developmental dyslexia (DD) have been shown to be impaired in tasks that require the processing of multiple visual elements in parallel. It has been suggested that this deficit originates from disturbed visual attentional functions. The parameter-based assessment of visual attention based on Bundesen's () theory of visual attention allows one to identify and quantify the underlying deficits. The present study provides the first groupwise comparison between children with DD (n = 12; mean age 9.84 years) and typically developing children (n = 12; mean age 9.87 years) with regard to intact and disturbed components of visual attention. From the performance on whole- and partial-report tasks, we derived individual parameter estimates of four different parameters: visual processing speed, storage capacity of visual short-term memory (VSTM), laterality of attentional weighting and efficiency of top-down control. Groupwise comparisons revealed that general attentional resources, processing speed and VSTM storage capacity, are impaired in children with DD compared to typically developing children. In contrast, selectivity parameters, laterality of attentional selection and attentional top-down control did not differ between these groups. Relating the current findings to previous results, obtained in highly comparable methodological settings, from single cases of children with DD, and from a group of adults with DD, we conclude that slowed perceptual processing speed is a primary visual attentional deficit in DD. Furthermore, reduced VSTM storage capacity seems to modulate the difficulties in written language processing imposed by the disorder.

P302 Utility of “trough levels” determination and anti-infliximab antibodies in patients with inflammatory bowel disease. Estimation of individual pharmacokinetic parameters (PK) through population pharmacokinetic model

P302 Utility of “trough levels” determination and anti-infliximab antibodies in patients with inflammatory bowel disease. Estimation of individual pharmacokinetic parameters (PK) through population pharmacokinetic model G. Juan1,2 *, A. Alvarino3,4, L. Oltra4, N. Maroto4, N. Cano2, I. Ferrer2, J. Hinojosa2,4. 1Hospital de Manises, Farmacia Hospitalaria, Valencia, Spain, 2Hospital de Manises, Digestivo, Manises, Spain, 3Hospital de Manises, LABCO, Valencia, Spain, 4Hospital de Manises, Gastroenterology, Valencia, Spain

3D Galactic dust extinction mapping with multiband photometry

We present a method to simultaneously infer the interstellar extinction parameters A0 and R0, stellar effective temperature Teff and distance modulus μ in a Bayesian framework. Using multiband photometry from Sloan Digital Sky Survey and UKIRT Infrared Deep Sky Survey, we train a forward model to emulate the colour change due to physical properties of stars and the interstellar medium for temperatures from 4000 to 9000 K and extinctions from 0 to 5 mag. We introduce a Hertzsprung–Russell diagram prior to account for physical constraints on the distribution of stars in the temperature–absolute magnitude plane. This allows us to infer distances probabilistically. Influences of colour information, priors and model parameters are explored. Residual mean absolute errors (MAEs) on a set of objects for extinction and temperature are 0.2 mag and 300 K, respectively, for R0 fixed to 3.1. For variable R0, we obtain MAEs of 0.37 mag, 412.9 K and 0.74 for A0, Teff and R0, respectively. Distance moduli are accurate to approximately 2 mag. Quantifying the precisions of individual parameter estimates with 68 per cent confidence interval of the posterior distribution, we obtain 0.05 mag, 66 K, 2 mag and 0.07 for A0, Teff, μ and R0, respectively, although we find that these underestimate the accuracy of the model. We produce two-dimensional maps in extinction and R0 that are compared to previous work. Furthermore, we incorporate the inferred distance information to compute fully probabilistic distance profiles for individual lines of sight. The individual stellar astrophysical parameter (AP) estimates, combined with inferred 3D information, will make possible many Galactic science and modelling applications. Adapting our method to work with other surveys, such as Pan-STARRS and Gaia, will allow us to probe other regions of the Galaxy.

Individual Column State and Parameter Estimation in the Simulated Moving Bed Process: an Optimization-based Method

The aim of this work is to propose a method for the simultaneous state and parameter estimation of the individual columns in the Simulated Moving Bed chromatographic process based on concentration measurements only in the switched product outlets. The method exploits the fact that the measurements switch from the inlets of the columns to their outlets sequentially. The estimation problem is cast as a dynamic optimization problem where the prediction error of the output profile is minimized. The initial concentrations in the individual columns and the parameters of the columns are treated as degrees of freedom of the optimization problem. As at each outlet the concentrations of one component are very small and therefore are subject to large relative measurement errors for given expected absolute errors of the concentration measurements, measurements for two different points in time, when the extract port and when the raffinate port is at the outlet of the column under consideration are taken into account simultaneously. The effectiveness of the method is demonstrated for the separation of an isopropanolol racemic mixture.

Developing Population Pharmacokinetic Parameters for High-Dose Methotrexate Therapy: Implication of Correlations among Developed Parameters for Individual Parameter Estimation Using the Bayesian Least-Squares Method

Bayesian estimation enables the individual pharmacokinetic parameters of the medication administrated to be estimated using only a few blood concentrations. Due to wide inter-individual variability in the pharmacokinetics of methotrexate (MTX), the concentration of MTX needs to be frequently determined during high-dose MTX therapy in order to prevent toxic adverse events. To apply the benefits of Bayesian estimation to cases treated with this therapy, we attempted to develop an estimation method using the Bayesian least-squares method, which is commonly used for therapeutic monitoring in a clinical setting. Because this method hypothesizes independency among population pharmacokinetic parameters, we focused on correlations among population pharmacokinetic parameters used to estimate individual parameters. A two-compartment model adequately described the observed concentration of MTX. The individual pharmacokinetic parameters of MTX were estimated in 57 cases using the maximum likelihood method. Among the available parameters accounting for a 2-compartment model, V1, k10, k12, and k21 were found to be the combination showing the weakest correlations, which indicated that this combination was best suited to the Bayesian least-squares method. Using this combination of population pharmacokinetic parameters, Bayesian estimation provided an accurate estimation of individual parameters. In addition, we demonstrated that the degree of correlation among population pharmacokinetic parameters used in the estimation affected the precision of the estimates. This result highlights the necessity of assessing correlations among the population pharmacokinetic parameters used in the Bayesian least-squares method.