Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “model selection”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 649 records · Page 36Linked to original sources
PubMed2003

Asymptotic behaviors of support vector machines with Gaussian kernel.

Support vector machines (SVMs) with the gaussian (RBF) kernel have been popular for practical use. Model selection in this class of SVMs involves two hyperparameters: the penalty parameter C and the kernel width sigma. This letter analyzes the behavior of the SVM classifier when these hyperparameters take very small or very large values. Our results help in understanding the hyperparameter space that leads to an efficient heuristic method of searching for hyperparameter values with small generalization errors. The analysis also indicates that if complete model selection using the gaussian kernel has been conducted, there is no need to consider linear SVM.

Models, Theoretical↗
PubMed1997

Predictors of survival in 171 patients with systemic sclerosis (scleroderma).

Predictors of survival were determined in 171 patients with systemic sclerosis by univariate analysis, and the Cox proportional hazards model using both cross sectional data at entry into the follow-up and time-dependent follow-up data. Clinical and laboratory data were evaluated from 1982 to the end of 1993. The presence of diffuse scleroderma, kidney and cardiac involvements were unfavourable prognostic signs in both the univariate analysis, and the Cox proportional hazards models. The Cox model, using the variables detected at study entry, indicated that pericarditis, and anaemia were bad prognostic signs. Analysis with time dependent data has not been reported in systemic sclerosis. The appearance of pigmentation disturbances, anaemia, and respiratory failure during the follow-up also caused a poor prognosis of the disease by the Cox model. In the stepwise selection models, diffuse scleroderma, internal organ manifestations including renal, and cardiac involvements were predominantly selected as the most unfavourable factors for survival. As to the extent of skin involvement and internal organ manifestations, the general behaviour of the disease seems to be similar throughout the world. The early appearance of pericarditis and pigmentation disturbances at study entry are bad prognostic signs.

Adult↗
PubMed2005

Deficient attention is hard to find: applying the perceptual load model of selective attention to attention deficit hyperactivity disorder subtypes.

BACKGROUND: Whether selective attention is a primary deficit in childhood Attention Deficit Hyperactivity Disorder (ADHD) remains in active debate. METHODS: We used the perceptual load paradigm to examine both early and late selective attention in children with the Primarily Inattentive (ADHD-I) and Combined subtypes (ADHD-C) of ADHD. RESULTS: No evidence emerged for selective attention deficits in either of the subtypes, but sluggish cognitive tempo was associated with abnormal early selection. CONCLUSIONS: At least some, and possibly most, children with DSM-IV ADHD have normal selective attention. Results support the move away from theories of attention dysfunction as primary in ADHD-C. In ADHD-I, this was one of the first formal tests of posterior attention network dysfunction, and results did not support that theory. However, ADHD children with sluggish cognitive tempo (SCT) warrant more study for possible early selective attention deficits.

Attention↗
PubMed2001

Value of the dorsal cutaneous guinea pig model in selecting topical antiviral formulations for the treatment of recurrent herpes simplex type 1 disease.

Recurrent herpes simplex labialis represents a disease still difficult to treat, despite the availability of many established antiviral drugs used in clinical research since 30 years ago. Although differences between the human disease and that obtained in experimental animal suggest caution in predicting an effective clinical response from the experimental results, some of the animal models seem to be useful in optimising the topical formulation of single antiviral drugs. In the present work the dorsal cutaneous guinea pig model was used to compare 5 different topical antiviral formulations with clinical promise (active molecule: 5% w/w micronized aciclovir, CAS 59277-89-3), using both roll-on and lipstick application systems. The aim being to evaluate which vehicle (water, oil, low melting and high melting fatty base) and application system (roll-on, lipstick) enhances the skin penetration and the antiviral activity of the drug, after an experimental intradermal infection with Herpes simplex virus type 1 (HSV-1). As reference, a commercial formulation (5% aciclovir ointment) was used. The cumulative results of this study showed that the formulation A, containing 5% aciclovir in an aqueous base in a roll-on application system, has the better antiviral efficacy in reducing the severity of cutaneous lesions and the viral titer; among the lipsticks preparations, the formulation D, containing 5% aciclovir in a low melting fatty base, demonstrates a very strong antiviral activity, though slightly less than formulation A. This experimental work confirms the validity of the dorsal cutaneous guinea pig model as a rapid and efficient method to compare the antiviral efficacy of new formulations, with clinical promise, to optimise the topical formulation of the active antiviral drugs.

Acyclovir↗
PubMed2006

Asymmetric hydrogenation of trisubstituted olefins with iridium-phosphine thiazole complexes: a further investigation of the ligand structure.

New chiral bidentate phoshine thiazoles have been prepared and successfully applied as ligands in the homogeneous iridium-catalyzed asymmetric hydrogenation of aryl alkenes and aryl alkene esters. The ligands are designed to be highly modular and have one common chiral intermediate, from which diversity can be introduced at a late stage in the synthetic pathway. It was found that a six-member-ring backbone of the rigid ligand structure was preferred over seven- or five-member rings. In this study it is shown that the substituent pattern of the ligands has a major influence on the stereochemical outcome of the products. By applying the selectivity model proposed in this study, it is possible to match different substrates against different catalysts. In this way, good to excellent enantioselectivity can be obtained for typically difficult substrates. Geometrically different derivatives of alpha- and beta-methyl cinnamic acid ethyl esters were hydrogenated, to demonstrate the validity of the selectivity model and to verify the importance of steric and electronic matching of the catalyst and the substrate.

Journal Article↗
PubMed1998

Analytical considerations in the use of capture-recapture to estimate prevalence: case studies of the estimation of opiate use in the metropolitan area of Barcelona, Spain.

Capture-recapture, an indirect method widely used to estimate undetected populations, has been criticized because it causes problems due to a lack of compliance with several important assumptions and model selection strategies. This paper expands on the problems encountered when applying this methodology to drug abuse estimations, specifically the prevalence of opiate use in the metropolitan area of Barcelona, Spain, in 1993. Three samples of opiate users (from hospital emergency rooms, treatment centers, and prisons) were available in the area studied; an additional sample (mortality data) was analyzed for the city of Barcelona. Log-linear models that provided a good fit were considered, to which further model selection strategies were applied. A total of 3,207 unique individuals aged 15-44 years were identified in the three samples from the greater Barcelona area; the mortality sample from the city of Barcelona contained an additional 83 individuals. Heterogeneity was observed in different age, sex, and residence area subgroups. Population estimates differed widely according to the log-linear model chosen. Minimum Akaike's information criterion model and saturated model estimates were used to produce population prevalence rates. The main problems the authors encountered in this study were related to population definition, source heterogeneity, and assessment of an adequate model, a problem associated with sample size.

Adolescent↗
PubMed1979

Computed tomography in a primate stroke model using selective balloon catheter arterial occlusion.

Ten heparinized nonhuman primates (rhesus monkeys and Papio maryumaya baboons) were subjected to temporary occlusion of an insular branch of the middle cerebral artery with a balloon catheter. An infarct was created. The lesions were monitored in vivo by computed tomography performed at intervals for 1 year. Animals were sacrificed and the lesions examined histologically. The technique obviates the disadvantages of surgical reaction, intravascular thrombosis, and gross cerebral edema observed in prior animal stroke models.

Animals↗
PubMed2006

Application of public-domain statistical analysis software for evaluation and comparison of comet assay data.

A novel approach for statistical analysis of comet assay data (i.e.: tail moment) is proposed, employing public-domain statistical software, the R system. The analytical strategy takes into account that the distribution of comet assay data, like the tail moment, is usually skewed and do not follow a normal distribution. Probability distributions used to model comet assay data included: the Weibull, the exponential, the logistic, the normal, the log normal and log-logistic distribution. In this approach it was also considered that heterogeneity observed among experimental units is a random feature of the comet assay data. This statistical model can be characterized with a location parameter m(ij), a scale parameter r and a between experimental units variability parameter theta. In the logarithmic scale, the parameter m(ij) depends additively on treatment and random effects, as follows: log(m(ij)) = a0 + a1x(ij) + b(i), where exp(a0) represents approximately the mean value of the control group, exp(a1) can be interpreted as the relative risk of damage with respect to the control group, x(ij) is an indicator of experimental group and exp(b(i)) is the individual risk effects assume to follows a Gamma distribution with mean 1 and variance theta. Model selection is based on Akaike's information criteria (AIC). Real data coming from comet analysis of blood samples taken from the flounder Paralichtys orbignyanus (Teleostei: Paralichtyidae) and from samples of cells suspension obtained from the estuarine polychaeta Laeonereis acuta (Nereididae) were employed. This statistical approach showed that the comet assay data should be analyzed under a modeling framework that take into account the important features of these measurements. Model selection and heterogeneity between experimental units play central points in the analysis of these data.

Animals↗
PubMed2005

An application of model-fitting procedures for marginal structural models.

Marginal structural models (MSMs) are being used more frequently to obtain causal effect estimates in observational studies. Although the principal estimator of MSM coefficients has been the inverse probability of treatment weight (IPTW) estimator, there are few published examples that illustrate how to apply IPTW or discuss the impact of model selection on effect estimates. The authors applied IPTW estimation of an MSM to observational data from the Fresno Asthmatic Children's Environment Study (2000-2002) to evaluate the effect of asthma rescue medication use on pulmonary function and compared their results with those obtained through traditional regression methods. Akaike's Information Criterion and cross-validation methods were used to fit the MSM. In this paper, the influence of model selection and evaluation of key assumptions such as the experimental treatment assignment assumption are discussed in detail. Traditional analyses suggested that medication use was not associated with an improvement in pulmonary function--a finding that is counterintuitive and probably due to confounding by symptoms and asthma severity. The final MSM estimated that medication use was causally related to a 7% improvement in pulmonary function. The authors present examples that should encourage investigators who use IPTW estimation to undertake and discuss the impact of model-fitting procedures to justify the choice of the final weights.

Air Pollutants↗
PubMed1994

Specific inhibition of hepatitis C virus expression by antisense oligodeoxynucleotides. In vitro model for selection of target sequence.

The effect of sense and antisense oligodeoxynucleotides (ODNs) on hepatitis C virus (HCV) gene expression was studied to determine the role of the highly conserved 5'-untranslated region in the life cycle of the virus. It was found that antisense ODNs complementary to nucleotides (nt) 38-65, 134-175, and 312-339 in the 5' noncoding region and 341-377 in the core open reading frame efficiently blocked HCV RNA translation. Overlapping ODNs that differed by only several nucleotides showed substantially different inhibition of HCV RNA translation. Fine sequence specificity testing at nt positions 351-377 revealed that ODNs as small as a 12-mer (nt 351-363) retained a high degree (80%) of inhibitory activity compared to ODNs of longer sequences. These results suggest that there are three highly specific domains in the 5' noncoding region and a sequence immediately downstream of the HCV core initiation codon that may be critical for translation of HCV RNA. This study also provides an experimental approach for the selection of target HCV RNA sequences susceptible to antisense effects, as well as for definition of functional regions of the genome necessary for viral replication.

Base Sequence↗
PubMed1998

HLA-DR/DQ transgenic, class II deficient mice as a novel model to select for HSP T cell epitopes with immunotherapeutic or preventative vaccine potential.

Protective immunity against mycobacteria is dependent on antigen/MHC class II specific, CD4+ Th1 cells. HLA-DR3-restricted Th1 cells respond to a subset of mycobacterial antigens, including the immunodominant hsp65, and recognize a single epitope in hsp65, notably p1-20. Altered peptide ligands (APL) of p1-20 can inhibit p1-20/hsp65-induced proliferation of DR3-restricted T cells in an allele specific manner in vitro. In order to develop a preclinical model in which p1-20 APL can be tested in vivo in the context of HLA, we have used murine class II deficient, HLA transgenic (Ab0) mice, in which all CD4+ T cells are restricted by the tg HLA molecule. BCG-immunized DR3.Ab0 and DQ8.Ab0 mice both responded well to hsp65. Furthermore, DR3.Ab0 mice recognized precisely the same p1-20 epitope as DR3-restricted human T cells, whereas DQ8.Ab0 mice responded to a different set of hsp65 peptides. This shows that (i) the same immunodominant protein and peptide epitope are recognized by T cells from DR3.Ab0 mice and DR3+ humans and (ii) indicates the major role of HLA-polymorphism in controlling the human T cell response to mycobacterial antigens. Thus, HLA-transgenic, Ab0 mice provide a novel, preclinical model system to analyze APL and vaccines in the context of HLA polymorphism.

Animals↗
PubMed2005

Alternative QSAR models for selected estradiol and cytochrome P450 ligands: comparison between classical, spectroscopic, CoMFA and GRID/GOLPE methods.

The performance of the spectroscopic EVA (eigenvalue) and EEVA (electronic eigenvalue) methods was tested with data sets applying coumarin 7-hydroxylation inhibitors (28 compounds) for cytochrome P450 mouse CYP2A5 and human CYP2A6 enzymes and 11ss-, 16a-, and 17a-substituted estradiol derivatives (30 compounds) for the lamb uterine estrogen receptor, and compared with the performance of the classical Hansch-type, CoMFA and GRID/GOLPE methods. Besides the internal predictability, the external predictability of the models was tested with several randomized training and test sets to ensure the validity and reliability of the models. Partial least squares (PLS) regression was employed as a general statistical tool with the EVA and EEVA methods. Some supplementary models were also built using only one PLS component with McGowan's volumes (MgVol and MgVol(2)) as additional descriptors and employing multiple linear regression (MLR) as the modelling tool. In general, both the internal and external performance of the EVA model, and more especially the EEVA model, with one PLS component and MgVol parameters was satisfactory, being either as good as or clearly better than that of the Hansch-type, CoMFA and GRID/GOLPE models.

Cytochrome P-450 Enzyme System↗
PubMed2004

Determinants of cardiac electrophysiological properties in mice.

INTRODUCTION: The transgenic mouse is a popular model for human inherited cardiac disease. Electrophysiology (EP) studies have recently been performed in transgenic mice to characterize the electrical phenotype of the heart. However, little is known regarding the impact of experimental conditions or model selection on the outcome of EP studies in mice. METHODS AND RESULTS: We investigated the effects of experimental conditions on mouse cardiac EP by (1) comparing the findings of transesophageal pacing with those of invasive intracardiac pacing, (2) elucidating the effects of commonly used anesthetic agents, and (3) determining the impact of changes in body temperature. We also investigated the effects of model selection by (1) studying the dependence on mouse strain, and (2) exploring the effects of age. We found that EP parameters derived by both transesophageal and intracardiac pacing/recordings methods were similar. On the other hand, the anesthetic mixture of ketamine, xylazine, and acepromazine had profound effects on cardiac EP compared to sodium pentobarbital or isoflurane. Meanwhile, compared to normal body temperature (97-99 F), low body temperature (92-94 F) prolonged most cardiac EP parameters, while high body temperature (102-104 F) had little effect. Heart rate was a sensitive indicator of changes in body temperature. Significant differences were observed in specialized conduction system properties among the mouse strains studied (FVB, C57, and DBA). Furthermore, atrial electrical remodeling was evidently associated with age, while ventricular electrical properties were virtually unaltered. In comparison with corresponding invasive EP parameters, we found that the QT interval was not a reliable EP index in the mouse. CONCLUSIONS: Cardiac EP variability may result from differences in experimental techniques including anesthesia and body temperature and from differences in mouse selection including strain and age. The influence of these factors should be considered when characterizing the electrical phenotype of transgenic mice in cardiovascular research.

Age Factors↗
PubMed2004

Fast exact leave-one-out cross-validation of sparse least-squares support vector machines.

Leave-one-out cross-validation has been shown to give an almost unbiased estimator of the generalisation properties of statistical models, and therefore provides a sensible criterion for model selection and comparison. In this paper we show that exact leave-one-out cross-validation of sparse Least-Squares Support Vector Machines (LS-SVMs) can be implemented with a computational complexity of only O(ln2) floating point operations, rather than the O(l2n2) operations of a naïve implementation, where l is the number of training patterns and n is the number of basis vectors. As a result, leave-one-out cross-validation becomes a practical proposition for model selection in large scale applications. For clarity the exposition concentrates on sparse least-squares support vector machines in the context of non-linear regression, but is equally applicable in a pattern recognition setting.

Algorithms↗
PubMed2004

Hierarchical modelling of small area and hospital variation in short-term prognosis after acute myocardial infarction. A longitudinal study of 35- to 74-year-old men in Denmark between 1978 and 1997.

Models for analysis of trends in hospital and small area variation in case fatality after acute myocardial infarction are presented. The data are from administrative registries in Denmark. Hierarchical modelling in a logistic regression with a Bayesian approach is used. Model selection is undertaken using the deviance and the Bayesian information criteria. There is a modest trend for hospital variation in case-fatality rates that coincides with the introduction of new treatment strategies. This hospital variation is considerably larger than the variation at the area level. There is no trend for variation of the case-fatality rates at the area level. Unstructured random effects slightly outperform spatially correlated random effects at the area level. Somewhat high correlations over time within hospitals and within areas were detected for the case-fatality rates. Heavy-tailed distributions (T-distributions) could be an alternative for the random effect distribution in data from administrative registries and compete in the model selection with the normal distribution in this study.

Adult↗
PubMed1999

Estimation theory and model parameter selection for therapeutic treatment plan optimization.

Treatment optimization is usually formulated as an inverse problem, which starts with a prescribed dose distribution and obtains an optimized solution under the guidance of an objective function. The solution is a compromise between the conflicting requirements of the target and sensitive structures. In this paper, the treatment plan optimization is formulated as an estimation problem of a discrete and possibly nonconvex system. The concept of preference function is introduced. Instead of prescribing a dose to a structure (or a set of voxels), the approach prioritizes the doses with different preference levels and reduces the problem into selecting a solution with a suitable estimator. The preference function provides a foundation for statistical analysis of the system and allows us to apply various techniques developed in statistical analysis to plan optimization. It is shown that an optimization based on a quadratic objective function is a special case of the formalism. A general two-step method for using a computer to determine the values of the model parameters is proposed. The approach provides an efficient way to include prior knowledge into the optimization process. The method is illustrated using a simplified two-pixel system as well as two clinical cases. The generality of the approach, coupled with promising demonstrations, indicates that the method has broad implications for radiotherapy treatment plan optimization.

Algorithms↗