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 991 records · Page 55Linked to original sources

PubMed1993

Accident rates for heavy truck-tractors in Michigan.

Accident rates of heavy truck-tractors are modelled using log-linear methods. The accident data used are a census of truck-tractor involvements in Michigan from May 1987 to April 1988. Travel data used to calculate the rates were produced by a survey of truck-tractors in Michigan covering the same time period. Both the accident and travel data were limited to Michigan-registered tractors operating in Michigan. Log-linear models of casualty and property-damage-only accident rates were developed using number of trailers, road type, area type, and time of day as predictor variables. Overall, differences between tractors with one and two trailers were not significant. Tractors with no trailers (bobtails) have significantly higher accident rates. Characteristics of the operating environment were found to have larger effects on the accident rate than tractor configuration (except for the bobtail). Rates varied by a factor of up to 6.8, depending on the road type. Casualty accident risk at night was 1.4 times the risk during the day. The risk of a casualty accident in rural areas was 1.6 times that of urban areas.

Accidents, Traffic↗

PubMed1999

Testing non-linearity and directedness of interactions between neural groups in the macaque inferotemporal cortex.

Information processing in the visual cortex depends on complex and context sensitive patterns of interactions between neuronal groups in many different cortical areas. Methods used to date for disentangling this functional connectivity presuppose either linearity or instantaneous interactions, assumptions that are not necessarily valid. In this paper a general framework that encompasses both linear and non-linear modelling of neurophysiological time series data by means of Local Linear Non-linear Autoregressive models (LLNAR) is described. Within this framework a new test for non-linearity of time series and for non-linearity of directedness of neural interactions based on LLNAR is presented. These tests assess the relative goodness of fit of linear versus non-linear models via the bootstrap technique. Additionally, a generalised definition of Granger causality is presented based on LLNAR that is valid for both linear and non-linear systems. Finally, the use of LLNAR for measuring non-linearity and directional influences is illustrated using artificial data, reference data as well as local field potentials (LFPs) from macaque area TE. LFP data is well described by the linear variant of LLNAR. Models of this sort, including lagged values of the preceding 25 to 60 ms, revealed the existence of both uni- and bi-directional influences between recording sites.

Action Potentials↗

PubMed2005

Diversity and productivity of plant communities across the Inland Northwest, USA.

No definitive explanation for the form of the relationship between species diversity and ecosystem productivity exists nor is there agreement on the mechanisms linking diversity and productivity across scales. Here, we examine changes in the form of the diversity-productivity relationship within and across the plant communities at three observational scales: plots, alliances, and physiognomic vegetation types (PVTs). Vascular plant richness data are from 4,760 20 m2 vegetation field plots. Productivity estimates in grams carbon per square meter are from annual net primary productivity (ANPP) models. Analyses with generalized linear models confirm scale dependence in the species diversity-productivity relationship. At the plot focus, the observed diversity-productivity relationship was weak. When plot data were aggregated to a focus of vegetation alliances, a hump-shaped relationship was observed. Species turnover among plots cannot explain the observed hump-shaped relationship at the alliance focus because we used mean plot richness across plots as our index of species richness for alliances and PVTs. The sorting of alliances along the productivity gradient appears to follow regional patterns of moisture availability, with alliances that occupy dry environments occurring within the increasing phase of the hump-shaped pattern, alliances that occupy mesic to hydric environments occurring near the top or in the decreasing phase of the curve, and alliances that occupy the wettest environments having the fewest species and the highest ANPP. This pattern is consistent with the intermediate productivity theory but appears to be inconsistent with the predictions of water-energy theory.

Biodiversity↗

PubMed1998

Linear extrapolation models of lung cancer risk associated with exposure to environmental tobacco smoke.

This paper presents a model to estimate the number of lung cancer deaths due to ETS exposure among the 1992 U.S. never-smoking population, based on downward linear extrapolation from the estimated risks of active smokers. The model uses several recently available data sources including an extensive review of the published literature on indoor concentration of ETS constituents measured under real-world conditions and data from the National Mortality Followback Survey and the National Health Interview Survey which furnish nationally representative estimates of the distribution of the U.S. population and the persons who died from lung cancer by sex, age, and smoking status. The linear extrapolation model estimates that five male and six female excess lung cancer deaths due to ETS exposure would be expected in the 1992 U.S. population of over 52 million never smokers age 35 and over. Explanations for differences between the results of our downward extrapolation model and those of others are presented.

Age Distribution↗

PubMed1999

Approximate Monte Carlo conditional inference in exponential families.

This article presents an algorithm for approximate frequentist conditional inference on two or more parameters for any regression model in the Generalized Linear Model (GLIM) family. We thereby extend highly accurate inference beyond the cases of logistic regression and contingency tables implimented in commercially available software. The method makes use of the double saddlepoint approximations of Skovgaard (1987, Journal of Applied Probability 24, 875-887) and Jensen (1992, Biometrika 79, 693-703) to the conditional cumulative distribution function of a sufficient statistic given the remaining sufficient statistics. This approximation is then used in conjunction with noniterative Monte Carlo methods to generate a sample from a distribution that approximates the joint distribution of the sufficient statistics associated with the parameters of interest conditional on the observed values of the sufficient statistics associated with the nuisance parameters. This algorithm is an alternate approach to that presented by Kolassa and Tanner (1994, Journal of the American Statistical Association 89, 697-702), in which a Markov chain is generated whose equilibrium distribution under certain regularity conditions approximates the joint distribution of interest. In Kolassa and Tanner (1994), the Gibbs sampler was used in conjunction with these univariate conditional distribution function approximations. The method of this paper does not require the construction and simulation of a Markov chain, thus avoiding the need to develop regularity conditions under which the algorithm converges and the need for the data analyst to check convergence of the particular chain. Examples involving logistic and truncated Poisson regression are presented.

Algorithms↗

PubMed1997

Single point dilution method for the quantitative analysis of antibodies to the gag24 protein of HIV-1.

In the present work a concept proposed in 1992 by Dopotka and Giesendorf was applied to the quantitative analysis of antibodies to the p24 protein of HIV-1 in infected asymptomatic individuals and AIDS patients. Two approaches were analyzed, a linear model OD = b0 + b1.log(titer) and a nonlinear log(titer) = alpha.OD beta, similar to the Dopotka-Giesendorf's model. The above two proposed models adequately fit the dependence of the optical density values at a single point dilution, and titers achieved by the end point dilution method (EPDM). Nevertheless, the nonlinear model better fits the experimental data, according to residuals analysis. Classical EPDM was compared with the new single point dilution method (SPDM) using both models. The best correlation between titers calculated using both models and titers achieved by EPDM was obtained with the nonlinear model. The correlation coefficients for the nonlinear and linear models were r = 0.85 and r = 0.77, respectively. A new correction factor was introduced into the nonlinear model and this reduced the day-to-day variation of titer values. In general, SPDM saves time, reagents and is more precise and sensitive to changes in antibody levels, and therefore has a higher resolution than EPDM.

Enzyme-Linked Immunosorbent Assay↗

PubMed2004

Environmental tobacco smoke and pregnancy outcome.

BACKGROUND: Recent reviews conclude that environmental tobacco smoke (ETS) leads to diminished birth weight. However, the threshold and magnitude of that effect is uncertain. We aimed to determine the magnitude and shape of the relations between ETS and various adverse pregnancy outcomes using a highly sensitive biochemical assay. METHODS: Maternal serum specimens were collected from more than 3000 women enrolled in California's prenatal screening program in 1992 and analyzed for cotinine. Information on pregnancy outcomes was obtained from live birth/fetal death records and hospital questionnaires. We conducted analyses on 2777 woman-live birth pairs and 19 woman-fetal death pairs in which the mother was presumed to be a nonsmoker (midtrimester cotinine levels < or =10 ng/mL). RESULTS: In multiple logistic regression analyses, the odds ratios of fetal death, preterm delivery, and term-low birth weight were 3.4, 1.8, and 1.8, respectively, in the highest cotinine quintile (0.236-10 ng/mL), compared with the lowest quintile (<0.026 ng/mL). In adjusted linear models, there was a linear dose-dependent effect of log cotinine on mean birth weight (-109 g) and mean infant length (-0.84 cm) over the range of cotinine values. Linear relations were not found with respect to infant head circumference or the ratio of brain weight to body weight. Infant's body mass index declined with exposures above approximately 0.5 ng/mL cotinine. We estimated that ETS levels at or above 0.05 ng/mL (experienced by 62% of the study population) accounted for 12% of all adverse outcomes. CONCLUSIONS: ETS exposure in pregnant women adversely affects pregnancy by increasing fetal mortality and preterm delivery at higher exposure levels and slowing fetal growth across all levels of ETS exposure.

Adult↗

PubMed1989

Academic middle managers for baccalaureate nursing: work motivation and satisfaction.

Seventy-nine percent of academic middle managers for baccalaureate nursing reported that they did not plan to continue in their current management positions, or advance in academic leadership positions (George, 1981). This study examined the relationships between the job characteristics, a mediating variable "growth need strength", and the general job satisfaction and work motivation of academic middle managers for baccalaureate nursing. The sample was drawn from the population of academic middle managers for baccalaureate nursing in 126 colleges and universities across the United States that offer both baccalaureate and higher nursing degree programs. Sixty seven percent returned the Job Diagnostic Survey Short Form (JDSSF). The findings were interpreted through an examination of three analytical techniques: zero-order correlations, multiple regression, and multiple regression with interaction. Findings indicate that the dependent variable, "internal work motivation," can be measured in the population, but it cannot be predicted using the seven job characteristics measured by the JDSSF. The job characteristic autonomy was significant in predicting "general job satisfaction." General job satisfaction also proved to be some combination of security satisfaction and growth need satisfaction. both independent variables "autonomy" and "feedback from the job itself" were significant in predicting growth need satisfaction. As a mediating variable, "growth need strength" was not significant in predicting general job satisfaction in a linear model, nor was it significant when entered as an interactive term. However, the multiplicative model did increase, by four percent above the linear model, the amount of variance predictable in general job satisfaction.(ABSTRACT TRUNCATED AT 250 WORDS)

Administrative Personnel↗

PubMed2005

Linear parameter haplotype models with stratification.

OBJECTIVES: The question of interest is estimating the relationship between haplotypes and an outcome measure, based upon unphased genotypes. The outcome of interest might be predicting the presence of disease in a logistic model, predicting a numeric drug response in a linear model, or predicting survival time in a parametric survival model with censoring. Explanatory variables may include phased haplotype design variables, environmental variables, or interactions between them. METHODS: We extend existing generalized linear haplotype models to parametric survival outcomes. To improve the stability of model variance estimates, a profile likelihood solution is proposed. An adjustment for population stratification is also considered. Here we investigate data sampled from known 'strata' (e.g., gender or ethnicity) that influence haplotype prior probabilities and thus the regression model weights. Differing linear model variance estimates, and the effect of stratification and departures from Hardy-Weinberg Equilibrium (HWE) on parameter estimates, are compared and contrasted via simulation. RESULTS: From simulations, we observed an improvement in statistical power when using a solution to profile likelihood equations. We also saw that stratification had little impact on estimates. Haplotypes that are not in HWE had a negative impact on power to test hypotheses. Finally, profile likelihood solutions for haplotypes deviating from HWE had improved power and confidence interval coverage of regression model coefficients.

Carcinoma, Squamous Cell↗

PubMed1998

The duration of predicting trials influences time to fatigue at critical power.

The present study compared times to fatigue at CP which had been calculated from relatively long duration predicting trials. With eight recreationally active males (mean age +/- SE = 18.6 +/- 2.1 years) having first cycled to fatigue on five occasions at different fixed work rates, CP was calculated in three ways: 1. using data from the three lowest exercise intensities (CPl); 2. using data from all five exercise intensities (CPa); and 3. using data from the highest three exercise intensities (CPh). Although CP was calculated using a linear and a three-parameter non-linear model, there were insufficient suitable data to complete the latter analysis. After three days and over an eight day period, the subjects cycled for up to 60 minutes at each of the three CPs calculated using the linear model. Analysis revealed that despite high linearity with the five-point work-time regression (average r2=0.996), CPl, CPa and CPh significantly differed to each other (268 +/- 17.5W; 285 +/- 12.1W; 321 +/- 8.8W respectively; p<0.05). Significant differences were also found between times to fatigue at CPl, CPa and CPh (42.9 +/- 3.9, 39.9 +/- 4.6 and 34.4 +/- 2.7 minutes respectively; p<0.05). The data show that when CP is calculated using a linear work-time regression, time to fatigue at CP is significantly influenced by the duration of predicting trials. Moreover, exercise at CP could only be maintained for 43 minutes despite CP being calculated from predicting trials averaging between 10 and 25 minutes.

Adolescent↗

PubMed2026

Non-linear predictive modeling and comprehensive meta-analysis of rectal temperature in Santa Inês sheep: a systematic review of thermal challenges and biometerological trends.

A systematic and bibliometric review, combined with a meta-analysis, was used to adjust an equation for estimating the physiological responses of Santa Inês sheep subjected to different thermal challenges. The systematic review compiled data on physiological responses and the thermal environment, which were then used in the meta-analysis to adjust regression models. The bibliometric analysis mapped the relationships among studies, highlighting their usefulness in interpreting research findings and biases. Addressing prior methodological critiques, the core of this study involves replacing the linear approach with a non-linear segmented regression model to accurately define the Thermal Neutral Zone (TNZ). The Segmented Regression Model was crucial, establishing the upper limit of the Thermal Neutral Zone (TNZ) at an air temperature (tair) of 34.64 °C, where trectal begins to increase abruptly. The model, while identifying a biologically significant breakpoint, exhibited a moderate Multiple R-squared of 0.3529, highlighting the high heterogeneity and methodological variability in the current Santa Inês literature. This non-linear approach offers a biologically superior tool for identifying the onset of thermal distress.

Animals↗

PubMed2003

Investigating the dose-response relation between air pollution and total mortality in the APHEA-2 multicity project.

BACKGROUND: Several recent studies have reported significant health effects of air pollution even at low levels of air pollutants, but in most of these studies linear non-threshold relations were assumed. AIMS: To investigate the NO2 mortality dose-response association in nine cities participating in the APHEA-2 project using two different methods: the meta-smooth and the cubic spline method. METHODS: The meta-smooth method developed by Schwartz and Zanobetti is based on combining individual city non-parametric smooth curves; the cubic spline method developed within the APHEA-2 project combines individual city estimates of cubic spline shaped dose-response relations. The meta-smooth method is easier and faster to implement, but the cubic spline method is more flexible for further investigation of possible heterogeneity in the dose-response curves among cities. RESULTS: In the range of the pollutant common to all cities the two methods gave similar and comparable curves. Using the cubic spline method it was found that smoking prevalence acts as an effect modifier with larger NO2 effects on mortality at lower smoking prevalence. CONCLUSIONS: The NO2-mortality association in the cities included in the present analysis, could be adequately estimated using the linear model. However, investigation of the city specific dose-response curves should precede the application of linear models.

Air Pollutants↗

PubMed2007

Genetic analysis of somatic cell scores in US Holsteins with a Bayesian mixture model.

The objective of this study was to apply finite mixture models to field data for somatic cell scores (SCS) for estimation of genetic parameters. Data were approximately 170,000 test-day records for SCS from first-parity Holstein cows in Wisconsin. Five different models of increasing level of complexity were fitted. Model 1 was the standard single-component model, and the others were 2-component Gaussian mixtures consisting of similar but distinct linear models. All mixture models (i.e., 2 to 5) included separate means for the 2 components. Model 2 assumed entirely homogeneous variances for both components. Models 3 and 4 assumed heterogeneous variances for either residual (model 3) or genetic and permanent environmental variances (model 4). Model 5 was the most complex, in which variances of all random effects were allowed to vary across components. A Bayesian approach was applied and Gibbs sampling was used to obtain posterior estimates. Five chains of 205,000 cycles were generated for each model. Estimates of variance components were based on posterior means. Models were compared by use of the deviance information criterion. Based on the deviance information criterion, all mixture models were superior to the linear model for analysis of SCS. The best model was one in which genetic and PE variances were heterogeneous, but residual variances were homogeneous. The genetic analysis suggested that SCS in healthy and infected cattle are different traits, because the genetic correlation between SCS in the 2 components of 0.13 was significantly different from unity.

Animals↗

PubMed1994

Modelling of human acute toxicity from physicochemical properties and non-vertebrate acute toxicity of the 38 organic chemicals of the MEIC priority list by PLS regression and neural network.

Linear and non-linear modelling of human acute toxicity (as human lethal concentrations; HLCs) of the 38 organic chemicals from the 50 priority compounds of the Multicentre Evaluation of In Vitro Cytotoxicity (MEIC) programme was investigated. The models obtained were derived either from a set of 23 physicochemical properties of the compounds or from their acute toxicities to five aquatic non-vertebrates together with the physicochemical properties. For the linear type, modelling was performed using a partial least square projection to latent structures (PLS) regression method; for the non-linear models, both PLS regression and neural network were utilized. A neural network using a combination of backpropagation and cascade-correlation algorithms was applied in this study. The results generally reveal a slightly better predictive performance of the models obtained from PLS regression than those obtained from neural networks. However, the model composed of physicochemical properties (PC-model) from the trained neural network using a back propagation algorithm with pruning technique proved superior to that trained with a combination of backpropagation and cascade-correlation algorithms after leave-one-out cross-validation. The predictive power of the PC-models, whether linear or non-linear, was comparable with that of the corresponding models consisting of both structural descriptors and the ecotoxicological tests (ECOPC-models), except for the battery (ECOPC-model) from the neural network. The composition of the 'best' PLS and neural network models points to the importance of the combination of physicochemical properties reflecting lipophilicity, size, volume, intermolecular binding forces and electronic properties of the molecule. All the aquatic non-vertebrate tests are shown to be essential in explaining human acute toxicity. However, the degree of contribution differed, with the crustacean (Artemia salina) and the bacterial (Microtox) bioassays being more important to the linear and non-linear PLS models, whereas the crustacean (Artemia salina and Streptocephalus proboscideus) tests, and the rotifer (Brachionus calyciflorus) assay were important to the neural network models. The organochlorine (lindane) and bipyridinium (paraquat) pesticides were common outliers in all the models. Moreover, the latter two compounds and the organophosphate (malathion) pesticide were also common outliers in all ECOPC-models. Other types of pesticides, however, fit the models. The predicted HLCs of a number of non-pesticides, including some chlorinated compounds, also deviated from the observed HLCs by more than one order of magnitude.

Algorithms↗

PubMed2000

Effective restoration of the U(A)(1) symmetry in the SU(3) linear sigma model

The effective restoration of the chiral U(A)(1) symmetry in strong interactions is studied using the linear chiral SU(3)xSU(3) model at finite temperatures. We find that the disappearance of the chiral anomaly causes a considerable change in the meson mass spectrum. We propose several signals for detecting this chiral phase in ultrarelativistic heavy-ion collisions: The eta/pi(0) ratio is enhanced by an order of magnitude, the a(0) is suppressed in the K&Kmacr; mass spectrum, and the scalar kappa meson appears as a peak just below the K(*)(892) in the invariant piK mass spectrum.

Journal Article↗

PubMed1988

Some implications of the Linear Quadratic model for tumor control probability.

To define an optimal radiation therapy strategy, the dependence of the probability of cure and of significant complications, on the parameters controlled by the radiotherapist must be determined. The recent success of the Linear Quadratic (LQ) model in constructing isoeffect relations for normal tissue damage and in describing in vitro cell survival curves, indicate that this model could be used to determine this dependence. The problem of tumor control is addressed. Using LQ model parameters obtained from human tumor cell lines, the sigmoid dose-response curves for controlling tumors of fixed size but of several histologies, with a fractionated course of radiotherapy is obtained. Except for squamous cell carcinoma, the calculated average tumor control doses (TCD37 or TCD50) are unrealistically low, but the model can be made more realistic by including inhomogeneities in the spatial dose distribution and heterogeneous tumor cell populations. The slope of the dose response curves are determined and the significance of the "relative slope" parameter rho as a measure of the number of cells in a tumor's most radioresistant clone is noted. The relation of the model's predictions to qualitative features of the experimental animal data for both tumor control and for normal tissue damage is discussed. Experiments to test the validity of this type of model are suggested.

Cell Line↗

PubMed1995

Characterizing dynamic brain responses with fMRI: a multivariate approach.

In this paper we present a multivariate analysis of evoked hemodynamic responses and their spatiotemporal dynamics as measured with fast fMRI. This analysis uses standard multivariate statistics (MANCOVA) and the general linear model to make inferences about effects of interest and canonical variates analysis (CVA) to describe the important features of these effects. We have used these techniques to characterize the form of hemodynamic transients that are evoked during a cognitive or sensorimotor task. In particular we do not assume that the neural or hemodynamic response reaches some "steady state" but acknowledge that these physiological changes could show profound task-dependent adaptation and time-dependent changes during the task. To address this issue we have modeled hemodynamic responses using appropriate temporal basis functions and estimated their exact form within the general linear model using MANCOVA. We do not propose that this analysis is a particularly powerful way to make inferences about functional specialization (or more generally functional anatomy) because it only provides statistical inferences about the distributed (whole brain) responses evoked by different conditions. However, its application to characterizing the temporal aspects of evoked hemodynamic responses reveals some compelling and somewhat unexpected perspectives on transient but stereotyped responses to changes in cognitive or sensorimotor processing. The most remarkable observation is that these responses can be biphasic and show profound differences in their form depending on the extant task or condition. Furthermore these differences can be seen in the absence of changes in mean signal.

Arousal↗

PubMed2004

Artificial neural network modeling to predict the plasma concentration of aminoglycosides in burn patients.

The goal was to use an artificial neural network model to predict the plasma concentration of aminoglycosides in burn patients and identify patients whose plasma antibiotic concentration would be sub-therapeutic based on the patients' physiological data and taking into account burn severity. Physiological data and some indicators of burn severity were collected from 30 burn patients who received arbekacin. A three-layer artificial neural network with five neurons in the hidden layer was used to predict the plasma concentration of arbekacin. Linear modeling for prediction of plasma concentration and logistic regression modeling for the classification of patients were also used and the predictive performance was compared to results from the artificial neural network model. Dose, body mass index, serum creatinine concentration and amount of parenteral fluid were selected as covariates for the plasma concentration of arbekacin. Area of burn after skin graft was a good covariate for indicating burn severity. Predictive performance of the artificial neural network model including burn severity was much better than linear modeling and logistic regression analysis. An artificial neural network model should be helpful for the prediction of plasma concentration using patients' physiological data, and burn severity should be included for improved prediction in burn patients. Because the relationship between burn severity and plasma concentration of aminoglycosides is thought to be nonlinear, it is not surprising that the artificial neural network model showed better predictive performance compared to the linear or logistic regression models.

Adult↗