			Home About us Contact

Multivariate Modelling (multivariate + modelling)

Distribution by Scientific Domains

Medical Sciences	50%

Selected Abstracts

A hybrid Bayesian back-propagation neural network approach to multivariate modelling

INTERNATIONAL JOURNAL FOR NUMERICAL AND ANALYTICAL METHODS IN GEOMECHANICS, Issue 8 2003
C. G. Chua
Abstract There is growing interest in the use of back-propagation neural networks to model non-linear multivariate problems in geotehnical engineering. To overcome the shortcomings of the conventional back-propagation neural network, such as overfitting, where the neural network learns the spurious details and noise in the training examples, a hybrid back-propagation algorithm has been developed. The method utilizes the genetic algorithms search technique and the Bayesian neural network methodology. The genetic algorithms enhance the stochastic search to locate the global minima for the neural network model. The Bayesian inference procedures essentially provide better generalization and a statistical approach to deal with data uncertainty in comparison with the conventional back-propagation. The uncertainty of data can be indicated using error bars. Two examples are presented to demonstrate the convergence and generalization capabilities of this hybrid algorithm. Copyright © 2003 John Wiley & Sons, Ltd. [source]

Simultaneous modelling of multiple traffic safety performance indicators by using a multivariate generalized linear mixed model

JOURNAL OF THE ROYAL STATISTICAL SOCIETY: SERIES A (STATISTICS IN SOCIETY), Issue 3 2004
Trevor C. Bailey
Summary., Traffic safety in the UK is one of the increasing number of areas where central government sets targets based on ,outcome-focused' performance indicators (PIs). Judgments about such PIs are often based solely on rankings of raw indicators and simple league tables dominate centrally published analyses. There is a considerable statistical literature examining health and education issues which has tended to use the generalized linear mixed model (GLMM) to address variability in the data when drawing inferences about relative performance from headline PIs. This methodology could obviously be applied in contexts such as traffic safety. However, when such models are applied to the fairly crude data sets that are currently available, the interval estimates generated, e.g. in respect of rankings, are often too broad to allow much real differentiation between the traffic safety performance of the units that are being considered. Such results sit uncomfortably with the ethos of ,performance management' and raise the question of whether the inference from such data sets about relative performance can be improved in some way. Motivated by consideration of a set of nine road safety performance indicators measured on English local authorities in the year 2000, the paper considers methods to strengthen the weak inference that is obtained from GLMMs of individual indicators by simultaneous, multivariate modelling of a range of related indicators. The correlation structure between indicators is used to reduce the uncertainty that is associated with rankings of any one of the individual indicators. The results demonstrate that credible intervals can be substantially narrowed by the use of the multivariate GLMM approach and that multivariate modelling of multiple PIs may therefore have considerable potential for introducing more robust and realistic assessments of differential performance in some contexts. [source]

Measurement of ischaemia,reperfusion in patients with intermittent claudication using NMR-based metabonomics

NMR IN BIOMEDICINE, Issue 7 2008
Stefan A. Coolen
Abstract Intermittent claudication has proved to be a good in vivo model for ischaemia,reperfusion. For assessment of ischaemia,reperfusion damage, the known biochemical markers all have disadvantages with respect to sensitivity and interference with other physiological events. In this work, we studied the metabolic effects of ischaemia,reperfusion in patients with intermittent claudication, and the effects of vitamin C and E intervention, using both traditional biochemical measurements and 1H-NMR-based metabonomics on urine and plasma. The 1H-NMR spectra were subjected to multivariate modelling using principal components discriminant analysis, and the observed clusters were validated using joint deployment of univariate analysis of variance and Tukey,Kramer honestly significant difference (HSD) testing. The study involved 14 patients with intermittent claudication and three healthy volunteers, who were monitored during a walking test, before and after a vitamin C/E intervention, and after a washout period. The effect of exercise was only observable for a limited number of biochemical markers, whereas 1H NMR revealed an effect in line with anaerobic ATP production via glycolysis in exercising (ischaemic) muscle of the claudicants. Thus, the beneficial effect of vitamins C and E in claudicants was more pronounced when observed by metabonomics than by traditional biochemical markers. The main effect was more rapid recovery from exercise to resting state metabolism. Furthermore, after intervention, claudicants tended to have lower concentrations of lactate and glucose and several other citric acid cycle metabolites, whereas acetoacetate was increased. The observed metabolic changes in the plasma suggest that intake of vitamin C/E leads to increased muscle oxidative metabolism. Copyright © 2008 John Wiley & Sons, Ltd. [source]

Effects of TNF-alpha antagonism on E-selectin in obese subjects with metabolic dysregulation

CLINICAL ENDOCRINOLOGY, Issue 1 2010
Markella V. Zanni
Summary Objective, Endothelial adhesion molecules like E-selectin play an important role in leukocyte recruitment and development of atherosclerotic plaque. E-selectin is increased in obesity, yet little is known regarding the specific factors contributing to elevated E-selectin in obesity and whether tumour necrosis factor alpha (TNF-alpha) increases E-selectin in vivo in this population. The objectives of this study were to: (1) determine the body composition, metabolic and inflammatory factors associated with increased E-selectin and (2) determine the role of TNF-alpha in the physiological regulation of E-selectin by antagonism of TNF-alpha with etanercept among obese subjects. Methods, E-selectin levels, body composition, metabolic parameters and inflammatory cytokines were assessed in 51 obese subjects and 37 non-obese healthy controls. Obese subjects were randomized to etanercept 50 mg weekly or placebo for 4 weeks. Changes in E-selectin were compared between treatment groups. Results, Obese subjects had higher E-selectin than non-obese controls (47·4 [32·7,58·8] vs. 27·2 [20·3,42·1] ng/ml, obese vs. non-obese, P < 0·0001). E-selectin was significantly associated with multiple body composition measures and metabolic parameters, along with specific measures of TNF-alpha activation, including soluble tumour necrosis factor receptors 1 (P = 0·03) and 2 (P = 0·02). In multivariate modelling, visceral adipose tissue, but not other measures of body composition, remained significantly associated with E-selectin. Among obese subjects, treatment with etanercept significantly decreased E-selectin (,5·7 ± 8·7 vs. 0·5 ± 6·0 ng/ml, etanercept vs. placebo, P = 0·005). Conclusions, E-selectin is increased in obesity, in relationship to increased visceral adiposity and markers of TNF-alpha activation. TNF-alpha antagonism with etanercept reduces E-selectin in obese subjects, providing evidence that the systemic circulatory release of E-selectin is regulated at least in part by TNF-alpha in obesity. [source]