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 1,153 records · Page 64Linked to original sources

PubMed2003

Complementary DNA macroarray analyses of differential gene expression in porcine fetal and postnatal muscle.

To study differential gene expression in porcine skeletal muscle, a porcine complementary DNA (cDNA) macroarray was produced that contained 327 expressed sequence tags (EST) derived from whole embryo and adult skeletal muscle, and differential display PCR products from fetal and postnatal muscle. Total RNA from four muscle samples, 75- and 105-d fetal hind limb muscles, and 1- and 7-wk postnatal semitendinosus muscle was used to make radiolabeled targets for duplicate hybridization to the macroarray membranes in an initial screen for expression. All EST that gave clear signals (n = 238) were then re-arrayed, and hybridization was conducted with additional biological replication of samples in the 75-d and 1-wk ages. Signal intensity for each gene was normalized to signal intensity measured at control spots on each membrane, which consisted of total cDNA from liver, lung, spleen, and skeletal muscle. Both normalized ratio levels and a mixed linear model analyses were used to identify genes differentially expressed among the muscle samples. Results showed 28 genes had differences in expression level greater than twofold between the 75-d fetal and 1-wk muscle RNA samples. All 28 genes were also identified as genes with significantly different (P < 0.01) expression using a mixed linear model analysis. Nineteen of these 28 genes had significant matches (basic local alignment search tool [BLAST] score > 100; P < 0.01) to known genes, two matched genes encoding human hypothetical proteins, and seven had no significant matches to Genbank nonredundant and dbEST (database of expressed sequence tags) entries. These results were confirmed for representative genes with RNA blot analysis of seven developmental time points, including RNA from the same muscle samples tested previously in the macroarray. The RNA blot results confirmed the macroarray results for all selected genes, demonstrating that the macroarray technique used in this study is accurate and reproducible. An unknown muscle clone (M218) with a slightly less than twofold increase in expression from the 75-d to the 1-wk age (1 wk/75 d = 1.94; P = 0.0114) was also shown to differ between these two ages using RNA blot analysis, demonstrating the methods used to identify differentially expressed genes may be conservative. The association between expression patterns of vimentin and desmin was also investigated. Results indicate the switch in intermediate filament protein from vimentin to desmin occurs primarily at the level of transcription and/or RNA processing.

Age Factors↗

PubMed1980

Discrimination and allocation using a mixture of discrete and continuous variables with some empty states.

The likelihood ratio classification rule based on the location mode is estimated given: (1) data consist of both binary and continuous variables; (2) some states have either zero frequency or too few observations, the case that usually happens the practice. An iterative proportional fitting of convenient approximation of the log-linear models as well as a linear additive model are utilized in estimating the rule's parameters. Performance of the obtained rule is then assessed by estimated error rates.

Analysis of Variance↗

PubMed1989

Presence of three or more rib fractures as an indicator of need for interhospital transfer.

During a 42-month period, 3,282 patients with blunt trauma were admitted to our hospital. Group I (n = 3,143) was defined as patients sustaining zero to two rib fractures, Group II (n = 139) as patients sustaining three or more rib fractures. Groups I and II differed in age and mechanism of injury. Two other descriptive variables (sex and year of admission) showed statistically significant associations with outcome. To adjust for the joint effect of these four variables, a log-linear model (binary outcomes) or a linear regression model (continuous outcomes) was used. We defined need for interhospital transfer (NIT) based on the following: death, an Injury Severity Score of greater than or equal to 16, or direct admission to the operating room or intensive care unit. We conclude that the presence of three or more rib fractures identified a group of patients requiring transfer to a trauma center (positive predictive value = 92.8%). A population-based study is needed to verify this finding.

Adult↗

PubMed2001

Modelling cause-specific hazards with radial basis function artificial neural networks: application to 2233 breast cancer patients.

The clinical course of a disease is often characterized by the possible occurrence of different types of events acting in a competing way. From a statistical point of view this translates into the need of modelling the dependence of cause-specific hazards as a function of covariates. Generalized linear models with Poisson error have previously been adopted for the analysis of competing risks as a function of discrete covariates. In the present paper an artificial neural network extension for the flexible joint estimation of cause-specific hazards depending on both discrete and continuous covariates is proposed. This approach is based on radial basis function networks which have the advantage of allowing parameter estimation by the adoption of standard software for generalized linear models. We have applied this method to data from 2233 breast cancer patients to investigate the effects of age, tumour size, number of metastatic axillary nodes, histology and tumour site on cause-specific hazards for intra-breast tumour recurrences and distant metastases. The adoption of a radial basis function network made it possible to highlight effects that were not considered by previous analyses of the same data.

Adult↗

PubMed2003

[Social support: scale test-retest reliability in the Pro-Health Study].

This article describes the test-retest reliability of a scale comprising five dimensions of social support: material, emotional, informational, affective, and positive social interaction. In the study, a sample of 192 employees at a university in Rio de Janeiro filled out the same questionnaire on two occasions, 15 days apart. Measures of stability used were the intraclass correlation coefficient (ICC), weighted kappa statistic, and log-linear models. Internal consistency was evaluated using the Cronbach's alpha coefficient. Social support dimensions showed internal consistency ranging from 0.75 to 0.91 at test, and 0.86 to 0.93 at retest. The ICC ranged from 0.78 to 0.87 in the five dimensions of the scale, with no substantial differences by gender, age, or level of schooling. For most questions, the "agreement plus linear by linear association" and "quasi-association" log-linear models gave the best fit. According to these results, the reliability of the instrument was considered adequate, enabling it to be used in ongoing assessment of associations between social support and health-related outcomes in a cohort study (the Pro-Health Study) recently begun in Rio de Janeiro.

Adult↗

PubMed1998

Temporal firing patterns of Purkinje cells in the cerebellar ventral paraflocculus during ocular following responses in monkeys II. Complex spikes.

Many theories of cerebellar motor learning propose that complex spikes (CS) provide essential error signals for learning and modulate parallel fiber inputs that generate simple spikes (SS). These theories, however, do not satisfactorily specify what modality is represented by CS or how information is conveyed by the ultra-low CS firing rate (1 Hz). To further examine the function of CS and the relationship between CS and SS in the cerebellum, CS and SS were recorded in the ventral paraflocculus (VPFL) of awake monkeys during ocular following responses (OFR). In addition, a new statistical method using a generalized linear model of firing probability based on a binomial distribution of the spike count was developed for analysis of the ultra-low CS firing rate. The results of the present study showed that the spatial coordinates of CS were aligned with those of SS and the speed-tuning properties of CS and SS were more linear for eye movement than retinal slip velocity, indicating that CS contain a motor component in addition to the sensory component identified in previous studies. The generalized linear model to reproduce firing probability confirmed these results, demonstrating that CS conveyed high-frequency information with its ultra-low firing frequency and conveyed both sensory and motor information. Although the temporal patterns of the CS were similar to those of the SS when the sign was reversed and magnitude was amplified approximately 50 times, the velocity/acceleration coefficient ratio of the eye movement model, an aspect of the CS temporal firing profile, was less than that of the SS, suggesting that CS were more sensory in nature than SS. A cross-correlation analysis of SS that are triggered by CS revealed that short-term modulation, that is, the brief pause in SS caused by CS, does not account for the reciprocal modulation of SS and CS. The results also showed that three major aspects of the CS and SS individual cell firing characteristics were negatively correlated on a cell-to-cell basis: the preferred direction of stimulus motion, the mean percent change in firing rate induced by upward stimulus motion, and patterns of temporal firing probability. These results suggest that CS may contribute to long-term interactions between parallel and climbing fiber inputs, such as long-term depression and/or potentiation.

Action Potentials↗

PubMed1997

A multilevel analysis of school factors associated with health-related fitness.

School factors associated with health-related fitness, using the National Children and Youth Fitness Survey II (NCYFS II) data, have been examined by Pate and Ross (1987), but the hierarchical structure of the data was ignored in the data analyses. The purpose of this study was to reanalyze the NCYFS II data using the hierarchical linear model and reexamine the effects of school physical education (PE) and other factors on children's 1-mile run/walk performance. Only two of five "significant" school factors were confirmed by this study, suggesting that heightened probabilities of Type I errors might have occurred in the previous study. Two confirmed key characteristics of school PE programs that led to improvement of children's cardiovascular endurance were PE specialists and the administration of fitness tests. For a hierarchical data structure, both experimental and observational units should be considered in the data analysis. The hierarchical linear model, which not only provides more accurate individual prediction but also takes group effect into account, proved to be an appropriate analytical model in analyzing the hierarchical data.

Age Factors↗

PubMed1996

Chromatic content of spectral lights.

Using three novel formats, we compare four estimates of the spectral sensitivity of the opponent stage channels: a Linear Model, Jameson and Hurvich's [Journal of the Optical Society of America, 45, 546-552 (1955)] hue cancellation sensitivities, Gordon and Abramov's [Optical Society of America Technical Digest Series, 15, 12-15 (1987)] hue scaling, and hue matching. The three formats are: the spectrum locus in the opponent equiluminant plane, null lines in the CIE XYZ chart and response functions for unique hues. All sensitivities show departures from the Linear Model and from each other. Relative to the model, common features of all estimates are that violets are compressed; long-wavelength reds are amplified; the redness component of violet lights is greatly attenuated; and saturation of violet stimuli is underestimated.

Color Perception↗

PubMed1988

Incorporating prior information into the analysis of contingency tables.

Contingency tables are often analyzed using log-linear models and in some situations prior information on the value of parameters in the log-linear model is available. In this article we describe a prior-posterior procedure that incorporates prior information directly into the analysis through a multivariate normal prior for the log-linear parameters. The mode and curvature of the posterior density are proposed as summary statistics.

Alcohol Drinking↗

PubMed2006

A meta-analysis of genotypes and haplotypes of methylenetetrahydrofolate reductase gene polymorphisms in acute lymphoblastic leukemia.

A meta-analysis of case-control studies that investigated the association between the C677T and/or A1298C polymorphisms in the methylenetetrahydrofolate reductase (MTHFR) gene and acute lymphoblastic leukemia (ALL) was carried out. Pooled odds ratios (OR) of various genetic contrasts of each polymorphism were estimated using random (RE) and fixed effects (FE) models. Pooled ORs for combined genotypes and haplotypes were estimated after adjustment for study effect using a log-linear model and the expectation-maximization algorithm in combination with log-linear modeling, respectively. The recessive model for allele 1298C produced a rather marginal association: RE OR: 0.67; 95% confidence interval (CI): 0.46-0.99 and FE OR: 0.64; 95% CI: 0.49-0.84. In Caucasians, the results of the recessive model for allele 1298C was consisted with a protective effect of ALL development: FE OR: 0.63; 95% CI: 0.46-0.87. In childhood ALL, according to the results of the allele contrast and the recessive model for 677T allele it was conceivable that a protective effect exist: RE OR = 0.74; 95% CI: 0.57-0.96 and RE OR: 0.69; 95% CI: 0.51-0.94, respectively. The combined genotypes produced significant pooled OR for the 677CC/1298CC relative to 677CC/1298AA (OR: 0.54; 95% CI: 0.36-0.80). The haplotype 677C/1298C might be more protective to ALL relative to haplotype 677C/1298A (OR: 0.77; 95% CI: 0.61-0.97). When studies not in Hardy-Weinberg equilibrium (HWE) were corrected to account for departures from HWE, then, the pattern of results remained the same. Overall, there is high heterogeneity between the studies in both polymorphisms. A differential magnitude of effect in large versus small studies and alteration of early extremes effects existed.

Adolescent↗

PubMed1995

Development of factor-score-based models to explain and predict maximal box-lifting performance.

The objectives of the study were threefold: (1) to develop factor-score-based models to predict maximum mass on a box-lifting task using multiple regressions; (2) to compare predictive and explanatory powers of factor-score-based models to models derived from data-level variables; and (3) to apply these findings to ergonomic research and practical problem-solving situations. Forty-eight volunteers (25 women and 23 men) completed a maximal box-lifting task and a maximal isoinertial lifting test on an Incremental Lifting Machine (ILM). Dynamic data collected during isoinertial testing were summarized into 32 lift parameters, and then subjected to principal components analyses using the 'FACTOR PROCEDURE' from the Statistical Analysis System (SAS). Factor scores were calculated for each participant on each of the four factors comprising the final solution, and multiple regression equations for men, women and combined data were generated using the 'GENERAL LINEAR MODELS' procedure from SAS. Results revealed that prediction of box-lifting performance was optimized when regression equations were developed using numerous data-level variables as predictors, i.e., all 32 lift parameters and ILM mass. In comparison, explanation was enhanced but predictive capabilities were reduced when linear models were formed using ILM mass and the factor scores derived from analyses of isoinertial lifting. The use of variables loading on the factors gave slightly increased predictive power than did the factor-score-based models. Similar trends in predictive and explanatory powers appeared when the data were analysed according to gender. Ergonomic applications of factor-score-based models were discussed with regard to ongoing research as well as to practical problem-solving situations. It was concluded that the advantages and usefulness of factor-score-based models warranted their inclusion in future investigations of lifting performance.

Adult↗

PubMed1990

Comparison with radionuclide angiography of two new geometric and four nongeometric models for echocardiographic estimation of left ventricular ejection fraction using segmental wall motion scoring.

Two new echocardiographic methods for estimating left ventricular ejection fraction, both based on segmental wall motion scoring on a continuous scale, were developed and compared to radionuclide angiography in 50 patients. The first method used a geometric model in which the ventricle was represented by a tubular base section joined to a hemielliptical apex section, each containing 4 quadrants. Contraction was scored visually in 12 regions of the ventricle. The 12 scores were converted to dimensional shortenings, which were then used to calculate separate ejection fractions for each of the 8 quadrants. These 8 values were averaged to obtain the global ejection fraction. Assuming an elliptical end-systolic cross-section shape for each quadrant, the correlation between the echocardiographic and radionuclide ejection fractions was 0.917 with a regression slope of 0.93 and an intercept of 1.5. Results were identical assuming circular end-systolic quadrant shapes. The second approach used a linear combination of the mean wall motion score at the base, midleft ventricle and apex to predict ejection fraction. Coefficients for the 3 mean scores were optimized by least squares best fit with radionuclide ejection fraction in all 50 patients. The simplest linear model showed a correlation of 0.90 between the echocardiographic and radionuclide ejection fraction but underestimated low, and overestimated high, ejection fractions. However, these systematic errors could be removed by regression, which added a constant term of 8.1 and a slope factor of 0.87 to the linear model, improving the correlation with radionuclide ejection fraction to 0.91.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗

PubMed1987

Models for the pharmacokinetics and pharmacodynamics of insulin in alloxan-induced diabetic dogs.

A combined pharmacokinetic/pharmacodynamic model was proposed to describe the pharmacokinetics of intravenously administered regular insulin (0.55 units/kg) in alloxan-induced diabetic dogs. Serum insulin concentrations were described by either a one- or two-compartment open model, in which a hypothetical effect compartment was linked to the central pharmacokinetic compartment, or in which the effect compartment was linked to the peripheral compartment. Response, as measured by percent change in glucose concentration from adjusted basal plasma concentrations, was modeled using the sigmoidal Emax effect model, a linear effects model, a log-linear effects model, and a gamma-linear effects model, using the insulin pharmacokinetic parameters to describe the amount in the hypothetical effect compartment. The results indicated that insulin pharmacokinetics are usually described by a two-compartment open model. Response to insulin was predicted more accurately in half of the dogs using the gamma-linear effects model in which the effect compartment was linked to the central compartment. In the other half of the dogs the best model was the sigmoidal Emax model in which the effect compartment was linked to the central pharmacokinetic compartment. The parameters in the latter model were correlated with each other and the confidence limits of the parameter estimates were larger than the parameters of the gamma-linear effects model. These models should be further investigated, but may offer an alternative method for distinguishing rapid insulin metabolism from insulin resistance.

Animals↗

PubMed1997

Empirical analyses of BOLD fMRI statistics. I. Spatially unsmoothed data collected under null-hypothesis conditions.

Temporal autocorrelation, spatial coherency, and their effects on voxel-wise parametric statistics were examined in BOLD fMRI null-hypothesis, or "noise," datasets. Seventeen normal, young subjects were scanned using BOLD fMRI while not performing any time-locked experimental behavior. Temporal autocorrelation in these datasets was described well by a 1/frequency relationship. Voxel-wise statistical analysis of these noise datasets which assumed independence (i.e., ignored temporal autocorrelation) rejected the null hypothesis at a higher rate than specified by the nominal alpha. Temporal smoothing in conjunction with the use of a modified general linear model (Worsley and Friston, 1995, NeuroImage 2: 173-182) brought the false-positive rate closer to the nominal alpha. It was also found that the noise fMRI datasets contain spatially coherent time signals. This observed spatial coherence could not be fully explained by a continuously differentiable spatial autocovariance function and was much greater for lower temporal frequencies. Its presence made voxel-wise test statistics in a given noise dataset dependent, and thus shifted their distributions to the right or left of 0. Inclusion of a "global signal" covariate in the general linear model reduced this dependence and consequently stabilized (i.e., reduced the variance of) dataset false-positive rates.

Artifacts↗

PubMedDate not supplied

Continuity of care and clinical effectiveness: treatment of posttraumatic stress disorder in the Department of Veterans Affairs.

Evaluation of the quality of outpatient treatment for patients with severe psychiatric or addictive disorders has often focused on the assessment of continuity of care (COC) as measured with administrative data. However, there has been little empirical evaluation of the relationship of measures of COC and treatment outcomes. This study used hierarchical linear modeling to examine the relationship between 6 indicators of COC and 6 outcome measures in a multisite monitoring effort for veterans with war-related posttraumatic stress disorder. There were few consistently significant associations between COC and outcome measures. Although measures of COC at the level of individual patients were associated with reductions in substance abuse symptoms, when COC measures were averaged to the site level and examined with hierarchical linear modeling models, thereby reducing the impact of intrasite selection bias, they were not associated with any desired outcomes. COC measures, at least in the sample used for this study, are not consistently associated with desirable client outcomes and may therefore be less than ideal performance measures in outcome evaluations following inpatient treatment, except to the extent that COC is considered to be an intrinsic indicator of higher quality regardless of its relationship to outcomes.

Ambulatory Care↗

PubMedDate not supplied

Simultaneous determination of carbamazepine and carbamazepine 10,11-epoxide by using microcolumn HPLC: study of pharmacokinetics of carbamazepine in a volunteer.

A highly-sensitive microcolumn HPLC method for the simultaneous determination of carbamazepine and carbamazepine-10,11-epoxide in human serum and saliva is described. The method was successfully employed for the study of pharmacokinetics of carbamazepine in humans. After oral administration of 100 and 200 mg of carbamazepine to a volunteer, multiple peaks were observed on the kinetic curves. They were symbathic in the serum and saliva. This indicated the presence of multiple peaks which characterize both free and protein-bound fractions of the drug. The existence of multiple peaks on the kinetic curves implies that the kinetic of carbamazepine cannot be described with the one-compartment linear model. Nevertheless, each peak was treated within the range of a one-compartment linear model of absorption and the results obtained were compared with published data. For the elucidation of the nature of multiple peaks the graphical differentiation of ascending and descending branches of all peaks were carried out. On this basis of the dependence of the absorption and elimination rates on time was constrained. The analysis of experimental data resulted in the following conclusions: (a) the presence of multiple peaks on the kinetic curves is induced by the interrupted character of carbamazepine absorption that is caused by the very poor solubility of carbamazepine; (b) the elimination of the drug from blood serum occurs in two phases. Binding of carbamazepine with tissues takes place in the first phase, and biotransformation and excretion occurs in the second phase. It is possible that the presence of multiple peaks on the kinetic curves is partially caused also by redistribution of the drug from the comparatively easily accessible to the less accessible tissues. This requires further investigation.

Anticonvulsants↗

PubMed2006

Quantification of Lactobacillus in fermented milk by multivariate image analysis with least-squares support-vector machines.

This paper reports an approach for quantification of Lactobacillus in fermented milk, grown in a selective medium (MRS agar), by use of digital colour images of Petri plates easily obtained by use of a flatbed scanner. A one-dimensional data vector was formed to characterize each digital image on the basis of the frequency-distribution curves of the red (R), green (G), and blue (B) colour values, and quantities derived from them, for example lightness (L), relative red (RR), relative green (RG), and relative blue (RB). The frequency distributions of hue, saturation, and intensity (HSI) were also calculated and included in the data vector used to describe each image. Multivariate non-linear modelling using the least-squares support vector machine (LS-SVM) and a linear model based on PLS regression were developed to relate the microbiological count and the frequency vector. Feasibly models were developed using the LS-SVM and errors were below than 10% for Lactobacillus quantification, indicating the proposed approach can be used for automatic counting of colonies.

Animals↗

PubMed2003

Learning higher-order structures in natural images.

The theoretical principles that underlie the representation and computation of higher-order structure in natural images are poorly understood. Recently, there has been considerable interest in using information theoretic techniques, such as independent component analysis, to derive representations for natural images that are optimal in the sense of coding efficiency. Although these approaches have been successful in explaining properties of neural representations in the early visual pathway and visual cortex, because they are based on a linear model, the types of image structure that can be represented are very limited. Here, we present a hierarchical probabilistic model for learning higher-order statistical regularities in natural images. This non-linear model learns an efficient code that describes variations in the underlying probabilistic density. When applied to natural images the algorithm yields coarse-coded, sparse-distributed representations of abstract image properties such as object location, scale and texture. This model offers a novel description of higher-order image structure and could provide theoretical insight into the response properties and computational functions of lower level cortical visual areas.

Learning↗