PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Linear Models”

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 253 records · Page 14Linked to original sources
PubMed1997

A simple non-linear model in incidence prediction.

A simple model is proposed for incidence prediction. The model is non-linear in parameters but linear in time, following models in environmental cancer epidemiology. Assuming a Poisson distribution for the age and period specific numbers of incident cases approximate confidence and prediction intervals are calculated. The major advantage of this model over current models is that age-specific predictions can be made with greater accuracy. The model also preserves in the period of prediction the age pattern of incidence rates existing in the data. It may be fitted with any package which includes an iteratively reweighted least squares algorithm, for example GLIM. Cancer incidence predictions for the Stockholm-Gotland Oncological Region in Sweden are presented as an example.

Adult↗
PubMedDate not supplied

A linear model for the management of a developmental planning project.

A developmental planning model has been presented and its use is described. The theoretical premises of the model are noted. Action planning using the Program Evaluation Review Technique (PERT) has a dominant role in the effective use of the model. A developmental planning project requires a managerial response that is different from operational planning. The manager should adopt and use a formal planning model and planning techniques for risky planning problems that are characterized by complexity or criticality. The Linear Model for the Management of a Developmental Planning Project is proposed for risky planning problems in hospital pharmacy. The model can be an effective aid to management for meaningful progress assessment during implementation and for quality assurance of a planning project.

Models, Theoretical↗
PubMed2005

Proton conduction along a chain of water molecules. Development of a linear model and quantum dynamical investigations using the multiconfiguration time-dependent Hartree method.

Proton transfer along a chain of water molecules is discussed. A linear model for such a chain is developed and its parameters are determined by comparison to quantum chemistry calculations. Fully quantum mechanical dynamical simulations on the translocation process are performed for different chain lengths, with up to five water molecules. We found that tunneling is important for the proton-transfer process. Furthermore, translocation is accomplished through a strongly correlated motion involving both hydrogen and oxygen atoms. An approximate treatment, which limits or even neglects this correlation, may lead to severely incorrect results.

Journal Article↗
PubMed2000

On power and sample size calculations for likelihood ratio tests in generalized linear models.

A direct extension of the approach described in Self, Mauritsen, and Ohara (1992, Biometrics 48, 31-39) for power and sample size calculations in generalized linear models is presented. The major feature of the proposed approach is that the modification accommodates both a finite and an infinite number of covariate configurations. Furthermore, for the approximation of the noncentrality of the noncentral chi-square distribution for the likelihood ratio statistic, a simplification is provided that not only reduces substantial computation but also maintains the accuracy. Simulation studies are conducted to assess the accuracy for various model configurations and covariate distributions.

Biometry↗
PubMed2003

Finite-element time-domain algorithms for modeling linear Debye and Lorentz dielectric dispersions at low frequencies.

We present what we believe to be the first algorithms that use a simple scalar-potential formulation to model linear Debye and Lorentz dielectric dispersions at low frequencies in the context of finite-element time-domain (FETD) numerical solutions of electric potential. The new algorithms, which permit treatment of multiple-pole dielectric relaxations, are based on the auxiliary differential equation method and are unconditionally stable. We validate the algorithms by comparison with the results of a previously reported method based on the Fourier transform. The new algorithms should be useful in calculating the transient response of biological materials subject to impulsive excitation. Potential applications include FETD modeling of electromyography, functional electrical stimulation, defibrillation, and effects of lightning and impulsive electric shock.

Algorithms↗
PubMed1979

[National Research Program 1: participation in the base examination, analysis by means of a logit-linear model].

In each of four swiss cities participation to baseline screening for the prevention of cardio-vascular diseases is analyzed within a stratified random sample using a logit-linear model. Stratification was chosen along sex, age and time of residence for persons living alone, and mean age of parents, number of children and time of residence for persons living as a family.

Adolescent↗
PubMed1986

Log-linear model analysis of allelic associations.

An approach is outlined for the analysis of nonrandom allelic associations in multilocus systems in a diploid population. The concept of composite link functions in generalised linear model analysis is used to handle the problem of incomplete identification of constituent gametes often encountered in genotypic data for two or more marker loci.

Alleles↗
PubMed1980

The analysis of rates and of survivorship using log-linear models.

Models are considered in which the underlying rate at which events occur has a log-linear relationship with covariates. It is shown that the estimation of parameters involves the solution of identical systems of equations for data from either a Poisson process, an exponential distribution, a survival model or a generalized log-linear model. This enables one to use algorithms for fitting log-linear models, such as iterative proportional fitting (IPF), for the analysis of rates or survivorship.

Epidemiologic Methods↗
PubMed1997

Regression analysis of interval-censored survival data with covariates using log-linear models.

We considered the regression analysis of the event time data with left-, right-, or interval-censored observations. We extended life-table techniques for censored survival data using log-linear models to incorporate interval-censored failures. The EM algorithm was used to calculate maximum likelihood estimates for the parameters. We assumed that the hazard function was a stepwise function over disjoint intervals of time; thus, the nonparametric model, the parametric exponential model, and the semiparametric Cox proportional hazard model were easily implemented as special cases. We adapted the restricted EM algorithm to test hypotheses and to construct confidence intervals for the parameters. These methods were applied in an analysis of the recurrence time for treated melanoma patients.

Adult↗
PubMed1995

Correction for covariate measurement error in generalized linear models--a bootstrap approach.

A two-phase bootstrap method is proposed for correcting covariate measurement error. Two data sets are needed: validation data for approximating the measurement model and data with a response variable. Bootstrap samples from both the data sets validation data are taken. Parameter estimates of the generalized linear model are calculated using expectations of the measurement model from the validation data as explanatory variables. The method is compared through simulation in logistic regression with the correction method proposed by Rosner, Willet, and Spiegelman (1991, Statistics in Medicine 8, 1051-1069). A real data example is also presented.

Age Factors↗
PubMed1987

Auditory evoked potentials to continuous amplitude-modulated sounds: can they be described by linear models?

The responses from the round window of the cochlea and the surface of the cochlear nucleus to continuous tones and noise that were amplitude-modulated with pseudorandom noise were studied. The responses were averaged with the averager locked to the periodicity of the pseudorandom noise, and the cross-correlation between the averaged responses and one period of the pseudorandom noise was computed. The degrees of non-linearities in the responses were estimated by comparing the response of a linear model that had this cross-correlogram as its impulse response and the physiological response. The root mean square (RMS) value of the histograms increases as a function of the stimulus intensity and reaches a peak at 40-60 dB above threshold, above which intensity it decreases. The non-linear component of the response increases monotonically with sound intensity in the range from threshold to 60-70 dB above threshold. The pseudorandom noise was of the inverse-repeat type, and when even-order non-linearities were canceled by subtracting the latter half of the responses to one period of the pseudorandom noise from the first half a much closer agreement with the model responses was obtained, indicating that the non-linearities were mainly of even order. It was concluded that the non-linearities of the responses may have been caused by an unequal response to increases and decreases in stimulus intensity.

Acoustic Stimulation↗
PubMed1979

The use of linear models to investigate the "centre effect" on graft survival.

First cadaver graft survival at 90 days in six UK centres was analysed, and found to differ widely between the centres (p less than 0.0005). Linear models were used to test whether these differences could be explained by other factors known to influence graft survival, such as age of recipients, tissue typing, blood group matching or year of graft. Adjusting for these factors singly and in various combinations did not reduce the significance of the "centre effect". These results held good both when deaths with a functioning graft within 90 days were treated as exclusions and also when they were treated as graft failures.

Aging↗
PubMed1991

Linearized model for the initiation of factor Va, and thrombin generation.

A simple model of the initiation of thrombin formation in plasma as a response to factor Xa generation was constructed. In this model factor Xa is considered as an input with a constant concentration. Substrate depletion and inactivation by activated protein C are neglected. The resulting linear model allows a closed form solution by standard methods. With values of the reaction rate constants, as determined in purified systems, this model predicts a highly explosive and complete activation of factor V and prothrombin as a response to any given (steady state) factor Xa concentration even in situations where prothrombinase and(/or) thrombin are rapidly inactivated. However, the time delay to rapid thrombin production becomes longer at lower factor Xa concentrations. Analysis of this time delay as a function of the factor Xa concentration indicates that the gain of the feedback loop of factor V activation by thrombin is so high that the contribution of factor V activation by factor Xa is relatively unimportant for factor Xa concentrations in the nanomolar range. It appears that the time lag is mainly determined by the gain of this feedback loop: similar proportional reductions of each of these reaction rates causes a similar effect. The effects of moderately enhanced inhibition rates of thrombin and prothrombinase on the time delay depend strongly on factor Xa concentration. Only a minor prolongation of the delay is predicted for factor Xa concentrations in the nanomolar range, but for factor Xa concentrations in the 1-10 pM range, the enhanced decay will cause considerable delays. Simultaneous reduction of the turnover rate of prothrombinase results in much larger delays for the entire range of factor Xa concentrations.

Computer Simulation↗
PubMed2025

Simultaneous inference for generalized linear models with unmeasured confounders.

Tens of thousands of simultaneous hypothesis tests are routinely performed in genomic studies to identify differentially expressed genes. However, due to unmeasured confounders, many standard statistical approaches may be substantially biased. This paper investigates the large-scale hypothesis testing problem for multivariate generalized linear models in the presence of confounding effects. Under arbitrary confounding mechanisms, we propose a unified statistical estimation and inference framework that harnesses orthogonal structures and integrates linear projections into three key stages. It begins by disentangling marginal and uncorrelated confounding effects to recover the latent coefficients. Subsequently, latent factors and primary effects are jointly estimated through lasso-type optimization. Finally, we incorporate projected and weighted bias-correction steps for hypothesis testing. Theoretically, we establish the identification conditions of various effects and non-asymptotic error bounds. We show effective Type-I error control of asymptotic-tests as sample and response sizes approach infinity. Numerical experiments demonstrate that the proposed method controls the false discovery rate by the Benjamini-Hochberg procedure and is more powerful than alternative methods. By comparing single-cell RNA-seq counts from two groups of samples, we demonstrate the suitability of adjusting confounding effects when significant covariates are absent from the model.

Hidden variables↗
PubMed2002

Non-linear model for the kinetics of 10B in blood after BPA-fructose complex infusion.

A numerical model with a memory effect was created to describe the kinetics of 10B in blood after a single 4-dihydroxyborylphenylalanine-fructose complex (BPA-F) infusion in boron neutron capture therapy (BNCT). The model formulation was based on the averaged data from 10 glioma patients from the Brookhaven National Laboratory (BNL) BNCT-trials. These patients received a 2 h i.v. infusion of a BPA-fructose complex that delivered 290 mg BPA/kg body weight. The model was validated by fitting the original BNL patient data and new patient data from the Finnish BNCT-trials. The new 3-parameter non-linear model provided mean absolute differences between the measured and estimated 10B concentrations in blood that were less than 3.9% when used to simulate actual patient irradiations that comprised two irradiation fields separated by a break to reposition the patient. The flexibility of the model was successfully tested with two different infusion protocols. The patient data were modelled with a two-compartment model and a bi-exponential fit for comparison. The 3-parameter model is better than previously described models in predicting the time course of blood 10B concentration after cessation of intravenous infusion of BPA-fructose.

Boron↗
PubMed1982

Pharmacokinetics of inhalation anesthetics: a three-compartment linear model.

The evolution of mathematical models of the uptake of the inhaled anesthetic agents has produced increasingly complex models in which researchers have attempted to incorporate more and more data on the effects of anesthetics on the processes of respiration, circulation, and metabolism. One result of this evolution has been to limit the application of these models due to the large amount of data required by the model and the need for a large digital computer to generate a solution. The purpose of this study is to show that a three-compartment linear model, using only the solubility of an anesthetic in water and oil, may br used to predict the uptake of a volatile anesthetic with sufficient accuracy for practical purposes. Only a programmable hand calculator is needed for the solution. Due to the simplicity of this model, compared with previously described models, it should prove useful in understanding the kinetics of gas uptake by the body.

Anesthesia, Inhalation↗
PubMed2004

Simple linear model provides highly accurate genotypic predictions of HIV-1 drug resistance.

Drug resistance is a major obstacle to the successful treatment of HIV-1 infection. Genotypic assays are used widely to provide indirect evidence of drug resistance, but the performance of these assays has been mixed. We used standard stepwise linear regression to construct drug resistance models for seven protease inhibitors and 10 reverse transcriptase inhibitors using data obtained from the Stanford HIV drug resistance database. We evaluated these models by hold-one-out experiments and by tests on an independent dataset. Our linear model outperformed other publicly available genotypic interpretation algorithms, including decision tree, support vector machine and four rules-based algorithms (HIVdb, VGI, ANRS and Rega) under both tests. Interestingly, our model did well despite the absence of any terms for interactions between different residues in protease or reverse transcriptase. The resulting linear models are easy to understand and can potentially assist in choosing combination therapy regimens.

Algorithms↗