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 415 records · Page 23Linked to original sources
PubMed2002

Predictive non-linear modeling of complex data by artificial neural networks.

An artificial neural network (ANN) is an artificial intelligence tool that identifies arbitrary nonlinear multiparametric discriminant functions directly from experimental data. The use of ANNs has gained increasing popularity for applications where a mechanistic description of the dependency between dependent and independent variables is either unknown or very complex. This machine learning technique can be roughly described as a universal algebraic function that will distinguish signal from noise directly from experimental data. The application of ANNs to complex relationships makes them highly attractive for the study of biological systems. Recent applications include the analysis of expression profiles and genomic and proteomic sequences.

Biochemical Phenomena↗
PubMed2003

Normal linear models with genetically structured residual variance heterogeneity: a case study.

Normal mixed models with different levels of heterogeneity in the residual variance are fitted to pig litter size data. Exploratory analysis and model assessment is based on examination of various posterior predictive distributions. Comparisons based on Bayes factors and related criteria favour models with a genetically structured residual variance heterogeneity. There is, moreover, strong evidence of a negative correlation between the additive genetic values affecting litter size and those affecting residual variance. The models are also compared according to the purposes for which they might be used, such as prediction of 'future' data, inference about response to selection and ranking candidates for selection. A brief discussion is given of some implications for selection of the genetically structured residual variance model.

Analysis of Variance↗
PubMed1984

Estimating levels of disturbed behaviour among psychiatric in-patients using a general linear model.

Disturbed behaviour was studied in relation to two contrasting ward environments, one representing the 'medical model' of psychiatry and the other being a modified form of 'therapeutic community'. A mathematical model was used in order to identify the characteristics of disturbed patients. The results suggested that behavioural disturbance of a patient in our sample was related to diagnosis, to the ward of admission and to the length of stay in hospital. Patients in all diagnostic categories tended to be more disturbed in the 'therapeutic community' than in the 'medical model' ward.

Adult↗
PubMed2004

Identification of a non-linear model as a new method to detect expiratory airflow limitation in mechanically ventilated patients.

Expiratory flow limitation (EFL) can occur in mechanically ventilated patients with chronic obstructive pulmonary disease and other disorders. It leads to dynamic hyperinflation with ensuing deleterious consequences. Detecting EFL is thus clinically relevant. Easily applicable methods however lack this detection being routinely made in intensive care. Using a simple mathematical model, we propose a new method to detect EFL that does not require any intervention or modification of the ongoing therapeutic. The model consists in a monoalveolar representation of the respiratory system, including a collapsible airway that is submitted to periodic changes in pressure at the airway opening: EFL provokes a sharp expiratory increase in the resistance Rc of the collapsible airway. The model parameters were identified via the Levenberg-Marquardt method by fitting simulated data on the airway pressure and the flow signals recorded in 10 mechanically ventilated patients. A sensitivity study demonstrated that only 8/11 parameters needed to be identified, the remaining three being given reasonable physiological values. Flow-volume curves built at different levels of positive expiratory pressure, PEEP, during "PEEP trials" (stepwise increases in positive end-expiratory pressure to optimize ventilator settings) have shown evidence of EFL in three cases. This was concordant with parameter identification (high Rc during expiration for EFL patients). We conclude from these preliminary results that our model is a potential tool for the non-invasive detection of EFL in mechanically ventilated patients.

Adult↗
PubMed2003

Longitudinal hierarchical linear models of the memory functioning questionnaire.

Three hypotheses about the nature of self-rated memory as measured by the Memory Functioning Questionnaire (MFQ; M. J. Gilewski, E. M. Zelinski, & K. W. Schaie, 1990) were tested: that ratings reflect memory performance, that personality traits underlie ratings, and that ratings reflect implicit theories of memory change. Baseline scores and 19 year change slopes for the 4 MFQ factor ratings of a sample of 97 participants aged 30-81 were investigated. There were significant mean declines for all MFQ ratings except Frequency of Forgetting and significant individual differences in slopes for Frequency, Retrospective Functioning, and Mnemonics. Personality predicted baseline Frequency and Seriousness ratings and list and text recall slopes predicted Mnemonics slopes. Different mechanisms may underlie baseline ratings and changes in ratings for different factors.

Adult↗
PubMed1999

Non-linear model of cancer growth and metastasis: a limiting nutrient as a major determinant of tumor shape and diffusion.

A new approach for modelling the spatio-temporal evolution of tumors is presented. To test its validity, a very basic model is considered, which, in spite of its simplicity, is capable of generating a multiplicity of morphologies and growth and migration rates. From an in-vivo scenario of basic life processes, cancer cell proliferation is described as a competition for basic nutrients. The chosen mathematical treatment and simulation techniques permit a direct implementation of the local nonlinear couplings existing between the various cell populations and the free and bound nutrient concentration. A discussion of the results and proposed improvements and applications of the model is also presented.

Cell Division↗
PubMed1991

Adjustment of provisional mortality series: the dynamic linear model with structured measurement errors.

The authors "consider the problem of adjusting provisional time series using a bivariate structural model with correlated measurement errors. Maximum likelihood estimators and a minimum mean squared error adjustment procedure are derived for a provisional and final series containing common trend and seasonal components. The model also includes measurement errors common to both series and errors that are specific to the provisional series. [The authors] illustrate the technique by using provisional data to forecast ischemic heart disease mortality."

Cause of Death↗
PubMed2002

Non-linear modeling of bioconcentration using partition coefficients for narcotic chemicals.

Bioconcentration factors (BCFs) have traditionally been used to describe the tendency of chemicals to concentrate in aquatic organisms. A reexamination of the log-log QSAR between the BCF and Kow for non-congener narcotic chemicals is presented on the basis of recommended data for fish. The model is extended to give a simple correlation between BCF and the toxicity of highly, moderately and weakly hydrophilic chemicals. For the first time, in this study an equation for calculating BCF was applied in a QSAR model for predicting the acute toxicity of chemicals to aquatic organisms.

Animals↗
PubMed1996

Backbone--side-chain interactions in serine. Synthesis, crystal structure and solution conformation of a linear model peptide N-Boc-L-Ser-L-Phe-OCH3.

The peptide Boc-Ser-Phe-OCH3 was synthesised by a solution-phase method using the usual workup procedure. The peptide was crystallized from a 70:30 (v/v) methanol-water mixture. The crystals are monoclinic, space group P21 with a = 5.128(2), b = 17.873(2), c = 11.386(2) A, and beta = 98.03(3) degrees. The structure was determined by direct methods and refined by structure factor least-squares procedure. The final R-value for 1499 observed reflections was 0.041. The structure contains one peptide and one solvent water molecule. The peptide adopts a beta-strand-like conformation with phi 1 = -100.3(5), psi 1 = 99.9(5), phi 2 = -122.2(5), psi T2 = -172.5(6) degrees. The Ser side-chain assumes an extended conformation with chi 11 = -177.0(4) degrees. The O gamma H group of serine acts as a proton donor in an intramolecular weak hydrogen bond with (Ser) O'1 [O gamma 1 - H gamma 1 ... O'1 = 3.253(6) A]. The Phe side-chain adopts a staggered conformation with chi 1(2) = -70.9(6), chi 2,1 (2) = 88.4(7) degrees, chi 2,2(2) = -89.2(6) degrees. The water molecule generates a loop through two hydrogen bonds with O gamma 1 [OW ... O gamma 1 = 2.893(5) A] and O'2 [OW ... O'2 = 2.962(7) A] atoms. The unit-translated peptide molecules along the a-axis are held by hydrogen bonds: N1 - H1 ... O2 (chi - 1, y, z) = 2.954(4) A and N2 - H2 ... O'1 (chi + 1, y, z) = 2.897(6) A in a manner similar to those observed in parallel beta-pleated sheet structures. There is an additional interaction involving O gamma 1 and the water molecule [OW ... O gamma 1 (chi + 1, y, z) = 2.789 (4) A]. The strong NOE peak of Ci(H) ... Ni + 1 (H) and a simultaneous weak NOE peak of Ni(H) ... Ni + 1 (H) in the ROESY spectra of two-dimensional NMR in dimethyl sulfoxide indicate a beta-strand-like conformation for the peptide in solution.

Models, Molecular↗
PubMed1983

Abuse of statistical packages: the case of the general linear model.

Since their introduction in the 1960s, packaged computer programs have freed researchers from much drudgery and painstaking computational labor. Before the distribution of these packages, many meaningful projects could not be attempted, simply because the data analysis phase would have been too labor intensive. However, the very ease of use of these packages can lead to their abuse. In particular, performing all possible t tests among numerous groups, calculating many correlations and circling the ones with P values less than 0.05, and performing stepwise regression all can easily lead to spurious statistical conclusions. The danger of these three practices is illustrated with three simple computer runs using random data. In each run statistically significant results were found for the random data.

Animals↗
PubMed1989

[Positron emission tomography following brief infusion of 5-[18F]uracil: linear model for the kinetics of 18F radioactivity in tumors].

Patients with a colorectal malignancy were examined with a total body positron emission tomography, immediately following a brief infusion of 5-[18F]uracil. The radioactivity in normal liver tissue and in liver metastasis was monitored for two hours after infusion of the radiotracer. A kinetic model is described which permits calculation of the probable tissue concentration of 5-fluorouracil (FU) and its metabolites. Prerequisite for the model is that the time dependent plasma concentration of FU is known. It is hoped that such analysis will help predict therapeutic response, and permit evaluation of the kinetic consequences of different methods of drug application.

Colonic Neoplasms↗
PubMed1983

Estimating the size of subpopulations of heroin users: applications of log-linear models to capture/recapture sampling.

This article reviews two of the major methodologies applied to estimation of the number of heroin abusers: survey research methods and the capture/recapture technique. The main focus of the paper is to show the flexibility of the capture/recapture approach in handling not only the dependence of samples of heroin users but also the nonhomogeneity of sampling probabilities, allowing estimation in populations which are mixtures of qualitatively different heroin user types. Models with these features are illustrated using both simulated and real heroin abuse data.

Data Collection↗