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 721 records · Page 40Linked to original sources
PubMed2006

Prediction of milk/plasma drug concentration (M/P) ratio using support vector machine (SVM) method.

PURPOSE: Development of reliable computational models to predict/classify milk-to-plasma (M/P) drug concentration ratio remains a challenging object. Support vector machine (SVM) method, as a new algorithm, was constructed to distinguish the potential risk of drugs to nursing infants. METHODS: Each drug was represented by a large pool of descriptors, of which five were found to be most important for constructing the predictive models. Next, two classification models, linear discriminant analysis (LDA) and SVM, were developed with bootstrapping validation based on the selected molecular descriptors. RESULTS AND CONCLUSIONS: The classification accuracy of training set and test set for SVM was 90.63 and 90.00%, respectively. The total accuracy for SVM was 90.48%, which was higher than that of LDA (77.78%). Comparison of the two methods shows that the performance of SVM was better than that of LDA, which implies that the SVM method is an effective tool in evaluating the risk of drugs when experimental M/P ratios have not been investigated.

Algorithms↗
PubMed2006

Perceptions of dental esthetics of Asian orthodontists and laypersons.

INTRODUCTION: The aim of the study was to determine the impact of anterior dental crowding, overjet, overbite, dental spacing, and type of malocclusion on dental esthetic perceptions of Asian orthodontists. METHODS: Twenty-one Asian orthodontists (10 men, 11 women) and 158 adult laypersons (38 men, 120 women) rated the attractiveness of 50 pairs of black-and-white intraoral photographic prints of various types of malocclusions. Each pair included a frontal and a right buccal view of the dentition at maximum intercuspation. Raters used a visual analogue scale (VAS), with "least attractive" (0 mm) and "most attractive" (100 mm) as the 2 extremes, to score the dental esthetics of the photographs. A principal investigator objectively measured the dental traits from study models. Linear regression analysis was performed with VAS scores as the dependent variable and anterior dental crowding, overjet, and overbite measurements as the independent variables. Independent sample t tests were used to determine the effect of dental spacing on the VAS scores. The Kruskal-Wallis test was used to determine the effect of type of malocclusion (Class I, Class II Division 1, Class II Division 2, Class III) on VAS scores. RESULTS: Increased overjet (>6 mm) and severe maxillary dental crowding (>8 mm) were found to be significantly correlated with laypersons' VAS scores (P <.05). Reverse overjet was found to be significantly correlated with orthodontists' VAS scores (P <.05). No significant correlation was found between overbite, dental spacing, and VAS scores. Class III malocclusions were significantly perceived to be the most unesthetic by orthodontists. CONCLUSIONS: Overjet was the major occlusal trait that influenced the perceptions of dental esthetics of both laypersons and orthodontists.

Adult↗
PubMed2004

Cardiac surgery risk models: a position article.

Differences in medical outcomes may result from disease severity, treatment effectiveness, or chance. Because most outcome studies are observational rather than randomized, risk adjustment is necessary to account for case mix. This has usually been accomplished through the use of standard logistic regression models, although Bayesian models, hierarchical linear models, and machine-learning techniques such as neural networks have also been used. Many factors are essential to insuring the accuracy and usefulness of such models, including selection of an appropriate clinical database, inclusion of critical core variables, precise definitions for predictor variables and endpoints, proper model development, validation, and audit. Risk models may be used to assess the impact of specific predictors on outcome, to aid in patient counseling and treatment selection, to profile provider quality, and to serve as the basis of continuous quality improvement activities.

Bayes Theorem↗
PubMed1999

Reality testing in adult women who report childhood sexual and physical abuse.

OBJECTIVE: To better understand how neighborhood and individual factors are related to child maltreatment. METHOD: Using an ecological framework, a multi-level model (Hierarchical Linear Modeling) was used to analyze neighborhood structural conditions and individual risk factors for child abuse and neglect. Parents (n = 400) of children under the age of 18 were systematically selected from 20 randomly selected census-defined block groups with different risk profiles for child maltreatment report rates. Parents were administered the Neighborhood Environment for Children Rating Scales, the Child Abuse Potential Inventory, the Zimet measure of social support, and the Conflict Tactics Scales as a measure of childhood experience with violence. RESULTS: Neighborhood factors of improverishment and child care burden significantly affect child abuse potential after controlling for individual risk factors. However, neighborhood effects are weaker than they appear to be in aggregate studies of official child maltreatment reports. Variation in child abuse potential within neighborhoods is greater than between neighborhoods. However, adverse neighborhood conditions weakened the effects of known individual risk and protective factors, such as violence in the family of origin. CONCLUSIONS: If individual potential for child maltreatment is more evenly distributed across neighborhoods than reported maltreatment, then neighborhood and community play an important, if as yet unspecified, role in child maltreatment. Multi-level models are a promising research strategy for disentangling the complex interactions of individual and contextual factors in child maltreatment.

Adolescent↗
PubMed2001

Relation between confidence in yes-no and forced-choice tasks.

Yes-no and forced-choice tasks are common in psychology, but the empirical relation between reported confidence in the 2 tasks has been unclear. The authors examined this relation with 2 experiments. The general experimental method had participants first report confidence in the truth of each of many general knowledge statements (a yes-no task) then report confidence in them again when the statements were put into pairs where it was known that one statement was true and one was false (a forced-choice task). At issue was how confidence in the statements changed between the yes-no task and the forced-choice task. Two models, including the normative one, were ruled out as descriptive models. A linear model and a multiplicative model remain viable contenders.

Choice Behavior↗
PubMed2003

Family meal patterns: associations with sociodemographic characteristics and improved dietary intake among adolescents.

OBJECTIVE: To examine family meal patterns and associations with sociodemographic characteristics and dietary intake in adolescents. DESIGN: A population-based cross-sectional study design was employed. Adolescents completed the Project EAT (Eating Among Teens) survey and the Youth and Adolescent Food Frequency Questionnaire within their schools. Subjects/setting The study population included 4,746 middle and high school students from Minneapolis/St. Paul public schools with diverse racial and socioeconomic backgrounds. Statistical analyses Associations were examined using cross tabulations, log-linear modeling, and linear regressions. RESULTS: There was a wide distribution in the frequency of family meals during the previous week: never (14.0%), 1 or 2 times (19.1%), 3 to 6 times (40.1%), and 7 or more times (24.8%). Sociodemographic characteristics associated with more frequent family meals included gender (boys), school level (middle school), race (Asian American), mother's employment status (not employed), and socioeconomic status (high). Frequency of family meals was positively associated with intake of fruits, vegetables, grains, and calcium-rich foods and negatively associated with soft drink consumption. Positive associations were also seen between frequency of family meals and energy; protein (percentage of total calories); calcium; iron; folate; fiber; and vitamins A, C, E, and B-6. CONCLUSIONS: Family meals appear to play an important role in promoting positive dietary intake among adolescents. Feasible ways to increase the frequency of family meals should be explored with adolescents and their families.

Adolescent↗
PubMed2000

Factors that influence personal perceptions of the risk of an acute myocardial infarction.

Personal risk perceptions of acute myocardial infarction (AMI) affect people's preventive health behaviors as well as their beliefs during a heart attack episode. The authors investigated factors that are associated with personal risk perceptions of having an AMI. A random-digit-dial survey was conducted among 1294 respondents, aged 18 years or older, in 20 communities across the nation as part of the Rapid Early Action for Coronary Treatment (REACT) trial. Results of two mixed-model linear regression analyses suggested that worse perceived general health, more risk factors, and greater knowledge were associated with greater perception of AMI risk. The results also showed that women who answered, incorrectly, that heart disease is not the most common cause of death for women in the United States reported significantly lower risk perceptions than women who answered this question correctly. The findings in this study suggest that interventions need to target specific misconceptions regarding AMI risk.

Adolescent↗
PubMed2003

A preliminary study of the application of some predictive modeling techniques to assess atmospheric mercury emissions from terrestrial surfaces.

Predictive modeling techniques are applied to investigate their potential usefulness in providing first order estimates on atmospheric emission flux of gaseous soil mercury and in identifying those parameters most critical in controlling such emissions. Predicted data by simulation and statistical techniques are compared to previously published observational data. Results showed that simulation techniques using air/soil coupling may provide a plausible description of mercury flux trends with a RMSE of 24.4ngm(-2)h(-1) and a mean absolute error of 10.2ngm(-2)h(-1) or 11.9%. From the statistical models, two linear models showed the lowest predictive abilities (R2=0.76 and 0.84, respectively) while the Generalized Additive model showed the closest agreement between estimated and observational data (R2=0.93). Predicted values from a Neural Network model and the Locally Weighted Smoother model showed also very good agreement to measured values of mercury flux (R2=0.92). A Regression Tree model demonstrated also a satisfactory predictability with a value of R2=0.90. Sensitivities and statistical analyses showed that surface soil mercury concentrations, solar radiation and, to a lesser degree, temperature are important parameters in predicting airborne Hg flux from terrestrial soils. These findings are compatible with results from recent experimental studies. Considering the uncertainties associated with mercury cycling and natural emissions, it is concluded, that predictions based on simple modeling techniques seem quite appropriate at present; they can be useful tools in evaluating the role of terrestrial emission sources as part of mercury modeling in local and regional airsheds.

Air Pollutants↗
PubMed2005

Gender and information and communication technologies (ICT) anxiety: male self-assurance and female hesitation.

This article presents the results of a quantitative study (n = 1,058) of the gender divide in ICT attitudes. In general, females had more negative attitudes towards computers and the Internet than did men. Results indicate a positive relationship between ICT experience and ICT attitudes. This experience is measured by period of time using a computer and self-perceived computer and Internet experience. Further analyses on the impact of gender on this correlation of ICT experience and ICT attitudes were conducted by means of a multivariate model. General Linear Model (GLM) analysis revealed that there was a significant effect of gender, computer use, and self-perceived computer experience on computer anxiety attitudes, as well as several significant interaction effects. Males were found to have less computer anxiety than females; respondents who have used computers for a longer period of time and respondents with a higher self-perception of experience also show less computer anxiety. However, the GLM plot shows that the influence of computer experience works in different ways for males and females. Computer experience has a positive impact on decreasing computer anxiety for men, but a similar effect was not found for women. The model was also tested for computer liking and Internet-liking factors.

Adult↗
PubMed2001

Analytic approaches for assessing long-term treatment effects. Examples of empirical applications and findings.

Analytic approaches, including the structural equation model (autoregressive panel model), hierarchical linear model, latent growth curve model, survival/event history analysis, latent transition model, and time-series analysis (interrupted time series, multivariate time-series analysis) are discussed for their applicability to data of different structures and their utility in evaluating temporal effects of treatment. Methods are illustrated by presenting applications of the various approaches in previous studies examining temporal patterns of treatment effects. Recent advancements in these longitudinal modeling approaches and the accompanying computer software development offer tremendous flexibility in examining long-term treatment effects through longitudinal data with varying numbers and intervals of assessment and types of measures. A multimethod assessment will contribute to a more complete understanding of the complex phenomena of the long-term courses of substance use and its treatment.

Alcoholism↗
PubMedDate not supplied

Effectiveness of academic detailing on breast cancer screening among primary care physicians in an underserved community.

BACKGROUND: Urban minority groups, such as those living in northern Manhattan and the South Bronx, are generally underserved with regard to breast cancer prevention and screening practices. Primary care physicians are critical for the recommendation of mammography and clinical breast examinations to their patients. DESIGN: Two medically underserved communities were matched and block randomized. The aim of the study was to assess the efficacy of academic detailing in increasing recommendations for breast cancer screening in community-based primary care physicians. SETTING/PARTICIPANTS: Ninety-four primary care community-based (ie, not hospital-based) physicians in northern Manhattan were compared with 74 physicians in the South Bronx who received no intervention. INTERVENTION: INTERVENTION participants received multicomponent physician-directed education, academic detailing, using the American Cancer Society guidelines for the early detection of breast cancer. MAIN OUTCOME MEASURES: We administered interviews to ask about primary care physicians' recommendation of mammography and clinical breast examination. They were also queried about their knowledge of major risk factors and perceived barriers to breast cancer screening. We conducted medical audits of 710 medical charts 2 years before and after the intervention. RESULTS: Using a mixed models linear analysis, we found a statistically significant intervention effect on the recommendation of mammography and clinical breast examination (according to medical audit) by female patients age 40 and over. INTERVENTION group physicians correctly identified significantly more risk factors for breast cancer, and significantly fewer barriers to practice, than did comparison physicians. CONCLUSIONS: We found some evidence of improvement in breast cancer screening practices due to academic detailing among primary care physicians practicing in urban underserved communities.

Breast Neoplasms↗
PubMed1999

The relationship between the myocardial kinetics of meperidine and its effect on myocardial contractility: model-independent analysis and optimal regional model.

The myocardial kinetics of meperidine and the relationship between these kinetics and the effect of meperidine on myocardial contractility (maximum positive rate of change of left ventricular pressure) were examined by analysis of previously published data collected in sheep after the i.v. injection of 100 mg of meperidine over 1 s. There was significant hysteresis between reductions in myocardial contractility and the arterial concentrations of meperidine, but not the coronary sinus blood (effluent from the heart) or calculated myocardial concentrations. The peak reduction in contractility occurred after the peak arterial concentration, at the time of the peak myocardial concentration, but before the peak coronary sinus concentration, suggesting that the site of drug action in the heart was not in equilibrium with either arterial blood or effluent blood from the heart. The most appropriate form of a dynamic model (a linear model with a threshold) was determined, without the need to assume a kinetic model, by directly fitting the observed reductions in myocardial contractility to the calculated myocardial concentrations. To determine the optimal kinetic and combined kinetic-dynamic models, a variety of one-, two-, and three-compartment models of the myocardium were fitted to the coronary sinus concentrations by using hybrid modeling. These included "tank in series" models that accounted well for drug dispersion and "peripheral compartment" models that accounted well for deep distribution. The most appropriate model was a "compilation" model, which incorporated features of both these extremes and was a better fit to the observed data than either a traditional single flow-limited compartment or a traditional membrane-limited model.

Algorithms↗
PubMed1997

[Health and the mass media].

Since the end of last century, due to the diffusion of certain--mostly contagious--diseases, the problem of communicating health issues to the public has emerged as crucial within the context of health care and prevention. The development of mass communication technologies, together with the transformation occured in the sphere of health attitudes and behaviours, have made this problem even more urgent. The plurality of messages and communication agents present the different media and more or less directly related to health has become so high as to discourage the adoption of simplified communication models--linear, unidirectional models--traditionally employed to understand such phenomenon.

Health↗
PubMed1994

[General expression of linear pharmacokinetic model].

This paper reports the integration of pharmacokinetic models of mammilla, chain, physiology, intravenous injection, intravenous infusion and extravascular administration into a single mathematical model. We have given a general mathematical treatment of this model and a common expression of drug concentration-time relationship. The model fits in any single or multiple dosing linear pharmacokinetic process involving in physiology, nonphysiology, and any compartmental administration and excretion.

Drug Administration Routes↗
PubMed2002

Marginally specified generalized linear mixed models: a robust approach.

Longitudinal data modeling is complicated by the necessity to deal appropriately with the correlation between observations made on the same individual. Building on an earlier nonrobust version proposed by Heagerty (1999, Biometrics 55, 688-698), our robust marginally specified generalized linear mixed model (ROBMS-GLMM) provides an effective method for dealing with such data. This model is one of the first to allow both population-averaged and individual-specific inference. As well, it adopts the flexibility and interpretability of generalized linear mixed models for introducing dependence but builds a regression structure for the marginal mean, allowing valid application with time-dependent (exogenous) and time-independent covariates. These new estimators are obtained as solutions of a robustified likelihood equation involving Huber's least favorable distribution and a collection of weights. Huber's least favorable distribution produces estimates that are resistant to certain deviations from the random effects distributional assumptions. Innovative weighting strategies enable the ROBMS-GLMM to perform well when faced with outlying observations both in the response and covariates. We illustrate the methodology with an analysis of a prospective longitudinal study of laryngoscopic endotracheal intubation, a skill that numerous health-care professionals are expected to acquire. The principal goal of our research is to achieve robust inference in longitudinal analyses.

Biometry↗
PubMed1988

The dynamics of the cat retinal Y cell subunit.

1. The dynamics of the subunit mechanism of individual cat Y retinal ganglion cells are investigated. In order to isolate the response of the non-linear subunit mechanism, the visual stimuli were sine gratings of a spatial frequency sufficiently high so that contrast reversal of the grating elicited no fundamental response in any spatial phase. For study of the non-linear subunit mechanism, the contrast of the spatial spine grating was varied in time by a temporal modulation signal, consisting of either a square wave or a sum of sinusoids. 2. The responses of twenty-three Y ganglion cells (sixteen on-centre, seven off-centre) to these two stimulus types were measured at a range of contrasts. Responses to the sum-of-sinusoids signal were characterized by the second-order frequency kernel. The overall size of the second-order frequency kernel was approximately proportional to contrast. The deviation from proportionality suggested a power-law scaling, with a power in the range 0.8-0.9. 3. Square-wave responses, as characterized by the post-stimulus histogram, demonstrated identical responses at both reversals of the grating. A similar contrast dependence was observed in the overall size of the square-wave responses. 4. In order to attempt to predict the square-wave responses from the sum-of-sinusoids responses, the second-order frequency kernel measured at each contrast level was fitted with a lumped linear-static non-linear-linear model. In eighteen of twenty-three cells (eleven on-centre, seven off-centre), this model provided an adequate description of the response to the sum-of-sinusoids stimulus. In these cells, the linear-static non-linear-linear model accurately reproduced the square-wave response. 5. In the remaining five ganglion cells (all on-centre), the second-order frequency kernel could not be fitted by a linear-static non-linear-linear model. This diversity of dynamical properties among Y cells was not apparent from the responses of these Y cells to the square-wave temporal stimulus. 6. In the eighteen Y ganglion cells that were fitted well with the linear-static non-linear-linear model, substantial variation of the dynamical parameters was found. However, there were systematic differences between the dynamics of the typical on-centre and off-centre ganglion cells. These differences relate to both linear stages of the model, and are not merely consequences of the lower firing rate of the off-centre cells.(ABSTRACT TRUNCATED AT 400 WORDS)

Action Potentials↗
PubMed1999

Discrete approximation of the linear Boolean model of heterogeneous materials.

The Boolean model is a random set process in which random shapes are positioned according to outcomes of a Poisson process. Two- and three-dimensional versions of the model characterize structures of certain heterogeneous materials. Linear transects of the Boolean model produce a one-dimensional Boolean model that summarizes some material properties. Two functions from linear transects, clump-length and lineal path distributions, provide information on material phase connectivity. Computation of these distributions is notoriously difficult. We provide a discrete approximation to the one-dimensional convex-grain Boolean model that yields stable, linear-time, recursive algorithms to approximate these functions. Computer simulations demonstrate accuracy and speed.

Journal Article↗