Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “linear mixed model”

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 1,513 records · Page 84Linked to original sources
PubMed2000

Mixed dentition space analysis in a Thai population.

This study produced simple linear regression equations to be used for mixed dentition space analysis for males and females, and sexes pooled in a population living in northeastern Thailand. Measurements of teeth were made to within 0.01 mm on the dental casts of 215 boys and 215 girls (mean age 15.7 years). All dentitions were required to be free of any signs of dental pathology or anomalies. It was found that males had significantly larger teeth than females as represented by summations of mandibular incisor, canine, and premolar widths. ANOVA of regression indicated a close relationship between mandibular incisor summation and corresponding summations of canine and premolars. The low coefficients of determination (r2) of the regressions ranged between 0.29 and 0.42, and were higher for females than males, which might be attributable to the ethnic diversity of the sampled population. The regression equations produced predictions of mesio-distal width summations for maxillary and mandibular canine, and premolar arch segments that were slightly different from other reported Asian studies. Moyers' prediction tables at the 50th percentile were found to under-estimate tooth size summation compared with the present investigation. The predictions from simplified regression equations matched well with those of this study for sexes pooled, and for males and females separately.

Adolescent↗
PubMed2006

Image denoising using total least squares.

In this paper, we present a method for removing noise from digital images corrupted with additive, multiplicative, and mixed noise. An image patch from an ideal image is modeled as a linear combination of image patches from the noisy image. We propose to fit this model to the real-world image data in the total least square (TLS) sense, because the TLS formulation allows us to take into account the uncertainties in the measured data. We develop a method to reduce the contribution from the irrelevant image patches, which will sharpen the edges and reduce edge artifacts at the same time. Although the proposed algorithm is computationally demanding, the image quality of the output image demonstrates the effectiveness of the TLS algorithms.

Algorithms↗
PubMed2004

Population pharmacokinetics of total and unbound plasma cisplatin in adult patients.

AIMS: To investigate the pharmacokinetics of unbound (ultrafilterable) and total plasma platinum using a population approach and to identify patient characteristics that may influence the disposition of the drug. METHODS: Pharmacokinetic and demographic data were collected from adult patients treated with 30-min daily infusions of cisplatin for various malignancies. Unbound and total platinum concentration-time data were analysed using a nonlinear mixed effects model. RESULTS: Data from 43 patients were available for analysis. A linear two-compartment model best described total and unbound platinum plasma concentration-time data. The mean population estimates for total and unbound drug were, respectively, 0.68 and 35.5 l h(-1) for clearance and 21.1 and 23.4 l for central distribution volume (V(1)). Unbound clearance (CL) was dependent on body surface area (BSA) and creatinine clearance, and V(1) was dependent on BSA. The elimination rate constant for plasma-bound platinum (modelled as metabolite formation) was 0.014 h(-1). The pharmacokinetic parameter, f(m)/V(m), a measure of the clearance of unbound platinum due to irreversible plasma binding, was related to serum protein concentration and to the inverse of dose per m(2). The covariate modelling of CL, V(1) and f(m)/V(m) improved the intersubject variabilities associated with these parameters. The final pharmacokinetic models were validated using 200 bootstrap samples from the original datasets. CONCLUSIONS: The results support the conventional dose adjustment of cisplatin based on BSA. They also support the need for a dose reduction in case of renal insufficiency.

Adult↗
PubMed2004

A method for matching the refractive index and kinematic viscosity of a blood analog for flow visualization in hydraulic cardiovascular models.

In this work, we propose a simple method to simultaneously match the refractive index and kinematic viscosity of a circulating blood analog in hydraulic models for optical flow measurement techniques (PIV, PMFV, LDA, and LIF). The method is based on the determination of the volumetric proportions and temperature at which two transparent miscible liquids should be mixed to reproduce the targeted fluid characteristics. The temperature dependence models are a linear relation for the refractive index and an Arrhenius relation for the dynamic viscosity of each liquid. Then the dynamic viscosity of the mixture is represented with a Grunberg-Nissan model of type 1. Experimental tests for acrylic and blood viscosity were found to be in very good agreement with the targeted values (measured refractive index of 1.486 and kinematic viscosity of 3.454 milli-m2/s with targeted values of 1.47 and 3.300 milli-m2/s).

Biomechanical Phenomena↗
PubMed1999

Caffeine overdose in a premature infant: clinical course and pharmacokinetics.

The elimination of caffeine was investigated in a 1860 g, 31 week gestation neonate, following the accidental administration of a 160 mg.kg-1 dose. The first serum concentration measured was 217.5 mg.l-1 at 36.5 h after dosing. Fitting of time-concentration data was performed using non-linear regression with MKMODEL. A first order elimination model was superior to a mixed order model. Parameter estimates were: clearance 0.01 l.h-1, volume of distribution 1.17 litres, elimination half-life 81 h. Toxic manifestations included hypertonia, sweating, tachycardia, cardiac failure, pulmonary oedema and metabolic disturbances (metabolic acidosis, hyperglycaemia and creatine kinase elevation). An unusual feature of this infant's illness course was gastric dilatation. These signs resolved by day 7 at a serum concentration of 60-70 mg.l-1. Caffeine clearance has traditionally been reported as either an absolute value or as directly proportional to body weight. The per kilogram model gives an erroneous impression that clearance is greatest in early childhood and then decreases with age until adult rates are reached in late adolescence. Age-related clearance values reported in the literature were reviewed using an allometric 3/4 power model. This size model demonstrates that clearance increases in infancy and reaches adult rates within the first three months of life.

Age Factors↗
PubMed2002

Population pharmacokinetics model and limited sampling strategy for intravenous vinorelbine derived from phase I clinical trials.

AIMS: a) To characterize the pharmacokinetics of intravenous vinorelbine, b) to use a population analysis for the identification of patient covariates that might appreciably influence its disposition and c) to define a limited sampling strategy for further Bayesian estimation of individual pharmacokinetic parameters. METHODS: All data were collected from 64 patients (99 courses) entered in three different phase I trials that have been previously reported. All patients received vinorelbine as a 20 min infusion with dose levels ranging from 20-45 mg m-2. The population pharmacokinetic model was built in a sequential manner on a subset of two-thirds of the data, starting with a covariate-free model then progressing to a covariate model using the nonlinear-mixed effect methodology. The remaining one-third of the data were used to validate several sparse sampling designs. RESULTS: A linear three-compartment model characterized vinorelbine blood concentrations (n=1228). Two primary pharmacokinetic parameters (total clearance and volume of distribution) were related to various combinations of covariates. The relationship for total clearance (CLtotal (l h-1)=29.2xBSAx(1-0.0090 Plt)+6.7xWt/Crs) was dependent on the patient's body surface area (BSA), weight (Wt), serum creatinine (Crs) and platelet count before administration (Plt). The optimal limited sampling strategy consisted of a combination of three measured blood concentrations; the first immediately before the end of infusion or 20 min later, the second at either 1 h, 3 h or 6 h and the third at 24 h after drug administration. CONCLUSIONS: A population pharmacokinetic model and a limited sampling strategy for intravenous vinorelbine have been developed. This is the first population analysis performed on the basis of a large phase I database that has identified clinical covariates influencing the disposition of i.v. vinorelbine. The model can be used to obtain accurate Bayesian estimates of pharmacokinetic parameters in situations where extensive pharmacokinetic sampling is not feasable.

Adult↗
PubMed2000

Zero-inflated Poisson and binomial regression with random effects: a case study.

In a 1992 Technometrics paper, Lambert (1992, 34, 1-14) described zero-inflated Poisson (ZIP) regression, a class of models for count data with excess zeros. In a ZIP model, a count response variable is assumed to be distributed as a mixture of a Poisson(lambda) distribution and a distribution with point mass of one at zero, with mixing probability p. Both p and lambda are allowed to depend on covariates through canonical link generalized linear models. In this paper, we adapt Lambert's methodology to an upper bounded count situation, thereby obtaining a zero-inflated binomial (ZIB) model. In addition, we add to the flexibility of these fixed effects models by incorporating random effects so that, e.g., the within-subject correlation and between-subject heterogeneity typical of repeated measures data can be accommodated. We motivate, develop, and illustrate the methods described here with an example from horticulture, where both upper bounded count (binomial-type) and unbounded count (Poisson-type) data with excess zeros were collected in a repeated measures designed experiment.

Animals↗
PubMed2005

Modeling solubilization of oil mixtures in anionic microemulsions II. Mixtures of polar and non-polar oils.

Polar/amphiphilic oils, called lipophilic linkers, are sometimes added to oil-water-ionic surfactant microemulsions in order to increase the solubilization of hydrophobic oils. The solubilization increase has been well documented for a number of systems. However, mathematical models to calculate the solubilization increase have been proposed only for optimum microemulsions (i.e., middle phase microemulsions solubilizing equal volumes of oil and water). In this paper we propose a model, which predicts solubilization enhancement for non-optimum microemulsion systems as well. The model is an extension of the net-average curvature model of microemulsion. The net-average curvature model is combined with a surface activity model to account for the increased palisade layer solubilization due to the presence of the polar/amphiphilic oil component. New non-linear mixing rules are also incorporated to account for the optimum salinity and the characteristic length variation of the anionic surfactant microemulsion as a function of the lipophilic linker concentration. The model predicts the effect of the lipophilic linker and the electrolyte concentration on the oil solubilization in accordance with the experimental results.

Journal Article↗
PubMed1986

Excretion-retention data of steady state gas exchange in tidal breathing. II. Dependency on the diffusion coefficient.

The steady state gas transfer properties of the lung can be derived from excretion-retention (E-R) data of inert tracer gases that are infused intravenously. E = PE/Pv and R = Pa/Pv, where PE, Pa and Pv represent the partial pressures of the tracer gases in mixed expired gas, arterial blood and mixed venous blood, respectively. In this paper, we investigate the influence of diffusive gas mixing in the lung on E and R. To that end, E-R data sets were simulated with a lung model that takes into account tidal breathing, the morphometric geometry of the airways, diffusion limited gas mixing in the alveolar space and gas dissolved in superficial lung tissue. The results show a linear relationship between R/E and D-0.5, where D represents the diffusion coefficient of the tracer gases in the alveolar gas mixture. This is in contrast to the results of simulations with a lung model that describes the different gas transport mechanisms, including diffusion, as a constant rate process, where a linear relationship between R/E and D-1 is predicted. It is further shown that E-R data of helium and sulphur hexafluoride cannot be used to demonstrate diffusion limited gas mixing in the lung, in particular, in the presence of a real shunt fraction. For that purpose, excretion data of pairs of tracer gases with different D but equal, medium blood-gas partition coefficients (1 less than lambda less than 30) should be used. For such pairs of tracer gases, the E values may differ by more than 10% when the D values for the two gases are 0.22 and 0.1 cm2s-1, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Blood Gas Analysis↗
PubMed2003

Once-a-day controlled-release dosage form of divalproex sodium II: development of a predictive in vitro drug release method.

During formulation design of a once-daily controlled release matrix system of divalproex sodium, the in vitro dissolution test (USP II, 100 rpm, pH 6.8 buffer) was found to result in release rates that were slower than in vivo absorption. The test method also did not sufficiently discriminate formulations with different in vivo absorption rates. To develop an in vitro method that is directly correlated with in vivo absorption, statistically designed studies were carried out to investigate the effects of various in vitro testing variables on drug release using USP dissolution apparatuses. The variables studied included agitation intensity, apparatus, pH, surfactant and ionic strength of the dissolution medium. Experimental data were analyzed using ANOVA. In vitro/in vivo correlation was tested based on the hypothesis that the same linear regression equation holds for three formulations with different release rates. A mixed effects model was used in which the dependence among observations from the same subject was taken into account. Factorial studies indicated that higher pH, addition of sodium lauryl sulphate (SLS) to the dissolution medium, and higher agitation intensity increased the release rate from the matrix tablet. Use of SLS not only lead to increased release rates that are more comparable to in vivo absorption rates, but also improved differentiation among formulations with varying release rates. Furthermore, drug release was also affected by interactions among the variables studied. Statistical analysis indicated that a combination of higher SLS concentration and lower pH provided enhanced differentiation between release profiles of the fast and slow releasing formulations. Based on the above findings, a new set of testing conditions was identified and demonstrated to be predictive of in vivo drug absorption for various controlled release formulations of divalproex sodium. The new method uses USP Apparatus II operating at 100 rpm in 500 mL of 0.1 N HCl for 45 min followed by 900 mL of 0.05 M phosphate buffer containing 75 mM SLS, pH 5.5, 37 +/- 0.5 degrees C. In conclusion, adjusting dissolution testing conditions to match the behavior of the formulations in vitro with that in vivo is a useful approach in identifying a predictive method in development of in vitro-in vivo correlation.

Anticonvulsants↗
PubMed1996

A limited sampling method to estimate methotrexate pharmacokinetics in patients with rheumatoid arthritis using a Bayesian approach and the population data modeling program P-PHARM.

This paper describes a methodology to calculate methotrexate (MTX) pharmacokinetic parameters after intramuscular administration using two samples and the population parameters. Total and free MTX were measured over a 36-h period in 56 rheumatoid arthritis patients; 14 patients were studied after a two-dose scheme at 15-day intervals. The Hill equation was used to relate the free MTX to the total MTX changes in plasma concentrations, and a two-compartment open model was used to fit the total MTX plasma concentrations. A non-linear mixed effect procedure was used to estimate the population parameters and to explore the interindividual variability in relation to the following covariables: age, weight, height, haemoglobin, erythrocyte sedimentation rate, platelet count, creatinine clearance, rheumatoid factor, C-reactive protein, swelling joint count, and Ritchie's articular index. Population parameters were evaluated for 40 patients using a three-step approach. The population average parameters and the interindividual variabilities expressed as coefficients of variation (CV%) were: CL, 6.94 l center dot h-1 (20.5%); V, 34.8 l (32.2%); k12, 0.0838 h-1 (47.7%); k21, 0.0769 h-1 (61.6%); ka, 4.31 h-1 (58%); Emax, 1.12 mu mol center dot l-1 (19.7%); gamma, 0.932 (12.3%); and EC50, 2.14 mu mol center dot l-1 (27.3%). Thirty additional data sets (16 new patients and 14 patients of the previous population but treated on a separate occasion) were used to evaluate the predictive performance of the population parameters. Twelve blood samples were collected from each individual in order to calculate individual parameters using standard fitting procedures. These values were compared to the ones estimated using a Bayesian approach with population parameters as a priori information together with two samples, selected from the individual observations. The results show that the bias was not statistically different from zero and the precision of these parameters was excellent.

Adult↗
PubMed1993

Influences on control in diabetes mellitus: patient, doctor, practice, or delivery of care?

OBJECTIVE: To assess patient, doctor, practice, and process of care variables for their effect on glycaemic control in diabetes mellitus, and to quantify their relative effects. DESIGN: Search of general practice medical records, patient questionnaires and examination, doctor questionnaire, videotaping and analysis of consultations, and practice questionnaire. SETTING: 12 practices with 32 participating general practitioners in Nottinghamshire. SUBJECTS: 318 patients randomly selected from those with diabetes in each practice, 10 for each participating doctor. MAIN OUTCOME MEASURE: Glycaemic control as measured by random glycated haemoglobin A1c estimation (random haemoglobin A1 measurement). RESULTS: Glycaemic control was significantly related to the disease process as measured by years since diagnosis, treatment group, and number of diabetes related clinical events. Females had significantly worse control than males. Other patient factors, such as age, social class, lifestyle, attitudes, satisfaction, and knowledge, had no association with glycaemic control. Of all the doctor factors examined, only doctors who professed a special interest in diabetes achieved significantly better glycaemic control. Bigger and better equipped practices and those with a diabetic miniclinic had patients with significantly better glycaemic control, as did those with access to dietetic advice. Patients attending hospital clinics had worse glycaemic control, but this seemed to be attributable to the case mix and practice characteristics. Shared care did not contribute to the multiple linear regression model. CONCLUSION: Glycaemic control among diabetic patients in the community is related to such factors as treatment group, sex, and years since diagnosis; it is also related to the organisation and process of care. The findings support concentrating diabetic care on partners with special interests in diabetes in well equipped practices with adequate dietetic support.

Attitude of Health Personnel↗
PubMed1999

On the approximation rate of hierarchical mixtures-of-experts for generalized linear models.

We investigate a class of hierarchical mixtures-of-experts (HME) models where generalized linear models with nonlinear mean functions of the form psi (alpha + xT beta) are mixed. Here psi (.) is the inverse link function. It is shown that mixtures of such mean functions can approximate a class of smooth functions of the form psi (h(x)), where h(.) epsilon W2;K infinity (a Sobolev class over [0, 1]s), as the number of experts m in the network increases. An upper bound of the approximation rate is given as O(m-2/s) in Lp norm. This rate can be achieved within the family of HME structures with no more than s-layers, where s is the dimension of the predictor x.

Algorithms↗
PubMed1989

Analysis of the binding sites for the cardiotonic phosphodiesterase inhibitor [3H]LY186126 in ventricular myocardium.

The positive inotropic action of the newer cardiotonic phosphodiesterase inhibitors such as indolidan, milrinone, and imazodan has been previously attributed to selective inhibition of cGMP-inhibitable Type IV (high affinity) cAMP phosphodiesterase activity. However, the subcellular binding site(s) for this class of compounds has not been defined. We have characterized the binding of [3H]LY186126, an analogue of indolidan, in subcellular fractions prepared from rabbit and sheep ventricular myocardium. Binding required magnesium ion and exhibited rapid association and dissociation kinetics. Specific binding (defined by ligand displacement with 5 microM indolidan) to enriched rabbit sarcoplasmic reticulum (SR) membrane vesicles was saturable (Bmax = 714 +/- 77 fmol/mg of protein) and of high affinity (Kd = 6.2 +/- 1.4 nM). Linear and nonlinear analyses of the binding isotherms fit a single-site model. Mixed SR preparations from sheep myocardium exhibited binding characteristics (Bmax = 944 +/- 115 fmol/mg; Kd = 8.5 +/- 2.3 nM) comparable to those of rabbit cardiac SR. Further subfractionation of sheep SR indicated that the binding sites were equally distributed between free (Bmax = 630 fmol/mg; Kd = 4.4 nM) and junctional SR (Bmax = 569 fmol/mg; Kd = 10.9 nM). Specific binding of [3H]LY186126 was also demonstrated in the cytosolic subfraction of rabbit myocardium that contained Type IV phosphodiesterase activity (Peak III from anion exchange chromatography). Competition for [3H] LY186126 binding studied in rabbit SR showed that, of the compounds tested, lixazinone (RS 82856) competed most effectively (IC50 = 0.030 +/- 0.008 nM), followed by indolidan (0.14 +/- 0.05 nM), cGMP (17.8 +/- 2.6 nM), milrinone (39.3 +/- 13.2 nM), and imazodan (192 +/- 73 nM). In contrast, rolipram, which does not inhibit SR-associated Type IV phosphodiesterase activity, was not effective at competing for [3H]LY186126 binding (IC50 greater than 30 microM). These results indicate that [3H]LY186126 has specific binding sites in myocardial subcellular fractions that contain cGMP-inhibitable Type IV (high affinity) cAMP phosphodiesterase activity.

Animals↗
PubMedDate not supplied

Mixed effect models of longitudinal Alzheimer's disease data: a cautionary note.

Longitudinal studies of cognitive function in Alzheimer's disease (AD) patients are powerful tools to better understand the biology and natural history of the disease, but the attributes of the studies that make them valuable also pose special challenges to analysts. A fundamental problem is the accurate measure of time at which cognitive decline begins. Investigators typically use the date of AD diagnosis or the date of enrollment in an AD study. If the rate of cognitive decline is non-linear, variables associated with the time of diagnosis or enrollment might artificially be associated with the rate of decline. Unlike the mixed effects models typically used to analyse cognitive decline, summary measure analyses do not directly compare the rate of decline with time since decline began, and, therefore, are less sensitive to biased measures of time of decline. We simulated trajectories of cognitive decline using the multivariate normal random effect model and tested the ability of the two analytic techniques to discriminate between true and spurious associations. Our analyses suggest summary measure models are less likely to detect spurious associations generated by biased measures of time at which decline begins, and more likely to detect true associations concealed by biased time measurement.

Aged↗
PubMed2001

Accuracy of mandibular cross-sectional imaging with tuned- aperture computed tomography (TACT), iteratively reconstructed TACT, and multidirectional, linear, and transverse panoramic tomography.

OBJECTIVE: This study was designed to compare 5 modalities with respect to accuracy in mandibular cross-sectional imaging. The modalities tested were tuned-aperture computed tomography (TACT), iteratively reconstructed TACT, multidirectional tomography, linear tomography, and transverse panoramic tomography. STUDY DESIGN: Twenty sites were selected from 3 dry human mandibles, and cross-sectional views were imaged through use of each of the 5 modalities. A quantitative analysis included measurements of 2 linear distances; a qualitative study included image evaluation by 6 observers. A nested mixed analysis of variance model was used to control for mandibles and locations within mandibles for the quantitative analysis; the Cochran-Mantel-Haenszel test was used for the qualitative analysis. RESULTS: There was a significant difference in measurement error for maximum height but not for width. There was also a significant difference in qualitative image evaluation results. CONCLUSIONS: Of the 5 modalities tested, the narrow-layer multidirectional tomographic technique produced the greatest diagnostic accuracy and quality in cross-sectional imaging. The transverse panoramic tomographic technique produced the least diagnostic accuracy and quality. Linear tomography, TACT, and iteratively reconstructed TACT were intermediate in accuracy and quality.

Analysis of Variance↗
PubMed2006

Rheology of sludge in a complete retention membrane bioreactor.

A rheological characterization of the sludge sampled from a complete retention membrane bioreactor was performed in order to correlate the apparent viscosity with the concentration of solids. The three most commonly adopted models were tested to select the one better fitting the experimental data. Ostwald model was choosen, and the relationship between the apparent viscosity and the shear rate was determined for mixed liquor suspended solids (MLSS) concentrations ranging between 9 and 25 g l(-1). Ostwald model parameters k and n were correlated with MLSS concentration, comparing linear, power, and exponential-power (only for k) laws in terms of R2 and Mean Root Square Error (MRSE). Both power and exponential-power functions provided good and comparable correlations for parameter k, while the linear relationship was much less accurate, especially at the highest solid concentrations. The parameter n was better modelled by a power function than by a linear one. Therefore two simulation models were proposed, both based on Ostwald's equation, where the two parameters k and n were expressed as functions of MLSS. Evaluation of energy consumption for mixing showed that the increase of solid concentration from 3 to 30 g l(-1) resulted in a limited increase in energy requirement (25-30 %). In real systems, where Reynolds numbers shift towards the turbulent regime, the increase of energy requirements for increasing solids concentration is even less pronounced. The thixotropy of membrane bioreactor sludge was also evaluated by measuring the rHa (reduced hysteresis area) and relating this parameter to the characteristics of the sludge growing within the reactor.

Bioreactors↗