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 397 records · Page 22Linked to original sources
PubMed2002

Fuzzy modelling for selecting headgear types.

The purpose of this study was to develop a computer-assisted inference model for selecting appropriate types of headgear appliance for orthodontic patients and to investigate its clinical versatility as a decision-making aid for inexperienced clinicians. Fuzzy rule bases were created for degrees of overjet, overbite, and mandibular plane angle variables, respectively, according to subjective criteria based on the clinical experience and knowledge of the authors. The rules were then transformed into membership functions and the geometric mean aggregation was performed to develop the inference model. The resultant fuzzy logic was then tested on 85 cases in which the patients had been diagnosed as requiring headgear appliances. Eight experienced orthodontists judged each of the cases, and decided if they 'agreed', 'accepted', or 'disagreed' with the recommendations of the computer system. Intra-examiner agreements were investigated using repeated judgements of a set of 30 orthodontic cases and the kappa statistic. All of the examiners exceeded a kappa score of 0.7, allowing them to participate in the test run of the validity of the proposed inference model. The examiners' agreement with the system's recommendations was evaluated statistically. The average satisfaction rate of the examiners was 95.6 per cent and, for 83 out of the 85 cases, 97.6 per cent. The majority of the examiners (i.e. six or more out of the eight) were satisfied with the recommendations of the system. Thus, the usefulness of the proposed inference logic was confirmed.

Adolescent↗
PubMed2005

Support vector machines committee classification method for computer-aided polyp detection in CT colonography.

RATIONALE AND OBJECTIVES: A new classification scheme for the computer-aided detection of colonic polyps in computed tomographic colonography is proposed. MATERIALS AND METHODS: The scheme involves an ensemble of support vector machines (SVMs) for classification, a smoothed leave-one-out (SLOO) cross-validation method for obtaining error estimates, and use of a bootstrap aggregation method for training and model selection. Our use of an ensemble of SVM classifiers with bagging (bootstrap aggregation), built on different feature subsets, is intended to improve classification performance compared with single SVMs and reduce the number of false-positive detections. The bootstrap-based model-selection technique is used for tuning SVM parameters. In our first experiment, two independent data sets were used: the first, for feature and model selection, and the second, for testing to evaluate the generalizability of our model. In the second experiment, the test set that contained higher resolution data was used for training and testing (using the SLOO method) to compare SVM committee and single SVM performance. RESULTS: The overall sensitivity on independent test set was 75%, with 1.5 false-positive detections/study, compared with 76%-78% sensitivity and 4.5 false-positive detections/study estimated using the SLOO method on the training set. The sensitivity of the SVM ensemble retrained on the former test set estimated using the SLOO method was 81%, which is 7%-10% greater than the sensitivity of a single SVM. The number of false-positive detections per study was 2.6, a 1.5 times reduction compared with a single SVM. CONCLUSION: Training an SVM ensemble on one data set and testing it on the independent data has shown that the SVM committee classification method has good generalizability and achieves high sensitivity and a low false-positive rate. The model selection and improved error estimation method are effective for computer-aided polyp detection.

Algorithms↗
PubMed2002

Sequencing, modeling, and selective inhibition of Trypanosoma brucei hexokinase.

For Trypanosoma brucei, a parasite responsible for African sleeping sickness, carbohydrate metabolism is the only source of ATP, and glycolytic enzymes are localized within membrane-bound organelles called glycosomes. Hexokinase, the first enzyme of the glycolytic pathway, was chosen as a target for selective drug design. We have cloned and sequenced the hexokinase gene of T. brucei. In parallel, we have synthesized several inhibitors. Kinetic analysis revealed differences in the binding mode of these compounds toward yeast and T. brucei hexokinases, while the m-bromophenyl glucosamide was found to be selective for T. brucei. The modeled structure of T. brucei hexokinase-inhibitor complex (using the crystal structure of the Schistosoma mansoni hexokinase as a template) allows us to propose a mode of action of this inhibitor for the trypanosome hexokinase and to account for the observed selectivity.

Adenosine Diphosphate↗
PubMed2004

Application of maximum-likelihood models to selection pressure analysis of group I nucleopolyhedrovirus genes.

Knowledge of virus genes under positive selection pressure can help identify molecular determinants of species-specific virulence or host range without prior knowledge of the mechanisms governing host range and virulence. Towards this end, codon-based models of substitution were used in a maximum-likelihood approach to analyse selection pressures acting on 83 genes of group I nucleopolyhedroviruses (NPVs). Evidence for positive selection was found for nine genes: ac38, ac66, arif-1, lef-7, lef-10, lef-12, odv-e18, odv-e56 and vp80. The baculovirus DNA helicase gene (dnahel) was not found to be positively selected using models that allowed the intensity of selection pressure to vary among codon sites. Further analysis with a method that allows selection pressure intensity to vary among lineages suggests that positive selection may have occurred in dnahel during the divergence of Bombyx mori NPV and the NPVs of Autographa californica and Rachiplusia ou. NPV genes that have undergone positive selection may modulate the ability of different NPVs to replicate efficiently in cells (lef-7, lef-10, lef-12) or to establish primary infection of the midgut (odv-e18, odv-e56) of different host species.

Animals↗
PubMed2004

Adjusting for publication bias: modelling the selection process.

UNLABELLED: RATIONALE, AIMS AND BACKGROUND: Systematic review with meta-analysis, a statistical technique for combining results of several studies, is progressively being used to guide decisions in medicine. Publication bias is acknowledged as a threat to the validity of systematic reviews and its existence may lead to inappropriate decisions about patient management or health policy. It is said to occur when the results of research available in the literature are not representative of the totality of all research. The selection mechanism that causes publication bias is complex, yet despite an extensive literature of empirical research identifying risk factors for publication, little work has been done to improve models of selection. Methods METHODS: that adjust combined meta-analytic estimates for publication bias are compared and applied to a systematic review of oral rehydration solution in the treatment of dehydration. Within a weighted distributions framework models of the selection process are considered and developed further. CONCLUSIONS: Weighted distributions offer a flexible approach that allows the potential to modify the selection function to incorporate other factors. Methods that adjust combined estimates should not be used to provide an alternative answer but to consider the robustness of the combined estimate to publication bias.

Child↗
PubMedDate not supplied

A model of selective experimental ischaemia in the primate thalamus.

A model for studying changes in local CBF and evoked potentials in selective thalamic ischaemia has been developed. The arterial supply to the posterior thalamus (mainly from the posterior choroidal arteries) was occluded in the baboon using a transorbital approach to the region of prepontine and ambient cisterns. Local CBF was measured by the hydrogen clearance method using electrodes introduced into the nucleus ventralis posterior lateralis of thalamus as well as cortex on both sides. The production of focal ischaemia was demonstrated by a significant decrease in thalamic CBF and confirmed by examination of the brain perfused with carbon particles.

Animals↗
PubMedDate not supplied

Guidelines for selecting a nursing model for practice.

Nursing models can provide direction to the practicing nurse. In this article, the process of selecting a nursing model is outlined. Noteworthy is the need for administrative support in this process in terms of providing release time, adequate coverage of the unit, scheduling, secretarial support, praise, and encouragement. It is suggested that nurses from various levels form an ad hoc committee so that there be representatives from both management and staff nurses as well as those who have been exposed to the use of models on the committee. Initially, the committee members should examine beliefs about nursing, the client, and the environment so that the model selected will reflect the beliefs and goals of the staff. Factors to consider in model selection are outlined so that the selected model will provide direction to each phase of the nursing process. Once a model is selected, it is recommended that individuals with experience in using it be approached to share their successes and difficulties. It is imperative that staff nurses be involved in the decision-making process.

Decision Making, Organizational↗
PubMed2005

Improving sequence-based fold recognition by using 3D model quality assessment.

MOTIVATION: The ability of a simple method (MODCHECK) to determine the sequence-structure compatibility of a set of structural models generated by fold recognition is tested in a thorough benchmark analysis. Four Model Quality Assessment Programs (MQAPs) were tested on 188 targets from the latest LiveBench-9 automated structure evaluation experiment. We systematically test and evaluate whether the MQAP methods can successfully detect native-like models. RESULTS: We show that compared with the other three methods tested MODCHECK is the most reliable method for consistently performing the best top model selection and for ranking the models. In addition, we show that the choice of model similarity score used to assess a model's similarity to the experimental structure can influence the overall performance of these tools. Although these MQAP methods fail to improve the model selection performance for methods that already incorporate protein three dimension (3D) structural information, an improvement is observed for methods that are purely sequence-based, including the best profile-profile methods. This suggests that even the best sequence-based fold recognition methods can still be improved by taking into account the 3D structural information. CONTACT: d.jones@cs.ucl.ac.uk

Algorithms↗
PubMed2004

A computational model of selection by consequences.

Darwinian selection by consequences was instantiated in a computational model that consisted of a repertoire of behaviors undergoing selection, reproduction, and mutation over many generations. The model in effect created a digital organism that emitted behavior continuously. The behavior of this digital organism was studied in three series of computational experiments that arranged reinforcement according to random-interval (RI) schedules. The quantitative features of the model were varied over wide ranges in these experiments, and many of the qualitative features of the model also were varied. The digital organism consistently showed a hyperbolic relation between response and reinforcement rates, and this hyperbolic description of the data was consistently better than the description provided by other, similar, function forms. In addition, the parameters of the hyperbola varied systematically with the quantitative, and some of the qualitative, properties of the model in ways that were consistent with findings from biological organisms. These results suggest that the material events responsible for an organism's responding on RI schedules are computationally equivalent to Darwinian selection by consequences. They also suggest that the computational model developed here is worth pursuing further as a possible dynamic account of behavior.

Algorithms↗
PubMed1990

A mathematical model of cancer chemotherapy with an optimal selection of parameters.

An optimal parameter selection model of cancer chemotherapy is presented which describes the treatment of a tumor over a fixed period of time by the repeated administration of a single drug. The drug is delivered at evenly spaced intervals over the treatment period at doses to be selected by the model. The model constructs a regimen that both minimizes the tumor population at the end of the treatment and satisfies constraints on the drug toxicity and intermediate tumor size. Numerical solutions show that an optimal regimen withholds the bulk of the doses until the end of the treatment period. When a drug used is of either moderate or low effectiveness, an optimal regimen is superior to a schedule that delivers all of the drug at the beginning of the treatment. This study questions whether the current method for the administration of chemotherapy is optimal and suggests that alternative regimens should be considered.

Antineoplastic Agents↗
PubMed2004

An empirical comparison of information-theoretic selection criteria for multivariate behavior genetic models.

Information theory provides an attractive basis for statistical inference and model selection. However, little is known about the relative performance of different information-theoretic criteria in covariance structure modeling, especially in behavioral genetic contexts. To explore these issues, information-theoretic fit criteria were compared with regard to their ability to discriminate between multivariate behavioral genetic models under various model, distribution, and sample size conditions. Results indicate that performance depends on sample size, model complexity, and distributional specification. The Bayesian Information Criterion (BIC) is more robust to distributional misspecification than Akaike's Information Criterion (AIC) under certain conditions, and outperforms AIC in larger samples and when comparing more complex models. An approximation to the Minimum Description Length (MDL; Rissanen, J. (1996). IEEE Transactions on Information Theory 42:40-47, Rissanen, J. (2001). IEEE Transactions on Information Theory 47:1712-1717) criterion, involving the empirical Fisher information matrix, exhibits variable patterns of performance due to the complexity of estimating Fisher information matrices. Results indicate that a relatively new information-theoretic criterion, Draper's Information Criterion (DIC; Draper, 1995), which shares features of the Bayesian and MDL criteria, performs similarly to or better than BIC. Results emphasize the importance of further research into theory and computation of information-theoretic criteria.

Bayes Theorem↗
PubMed2001

A comparison of total hospital costs for percutaneous coronary intervention patients receiving abciximab versus tirofiban.

The purpose of this study was to examine the total hospital costs associated with the receipt of abciximab versus tirofiban for percutaneous coronary intervention (PCI) patients. Hospital billing data for patients with a primary procedure of PCI was examined for the period of July 1998 to June 1999 from HCIA-Sach's Clinical Pathways Database. Data were analyzed for all patient discharges whose records indicated use of abciximab or tirofiban with a PCI. Results are reported for 3,967 patients. Multivariate analysis was used to control for a wide range of factors (GP IIb/IIIa selection, patient demographics, stent use, insurance type, health conditions, admission information, and hospital characteristics) that may influence the cost of hospitalization. A two-stage sample selection model was used to estimate total costs. The first stage of the analysis utilizes a probit regression to determine the factors associated with the likelihood of receiving abciximab versus tirofiban. The second stage of the analysis examines the factors associated with total hospital costs, while controlling for unobserved factors that may be correlated with the patient's likelihood of receiving abciximab. The mean unadjusted cost per hospitalization, including drug costs, was $10,762 (abciximab $10,813 and tirofiban $10,567). After controlling for high-risk indications and selection bias with the two-stage sample selection model, results indicate there was no significant difference in costs associated with the receipt of abciximab versus tirofiban. However, the results also indicate that the two-stage sample selection model may not be needed (lambda was not statistically significant) hence, the cost equation was reestimated using ordinary least-squares methodology (OLS). In the OLS analysis, receipt of abciximab versus tirofiban was associated with a significant reduction in costs ($470 reduction; P = 0.05). This study uses real-world data to examine the total hospital costs for PCI patients who receive abciximab versus tirofiban. Results of the two-stage sample selection model indicate there is no difference in total hospital costs (including drug costs) between abciximab- and tirofiban-treated patients. If the results of the OLS model are considered, a slight decrease in total hospital costs is observed in abciximab recipients. Cost-containment strategies that focus on component costs may not lead to intended overall cost savings.

Abciximab↗
PubMed2006

[Effect of phosphodiestrase 4 inhibitor (rolipram) on experimental allergic asthma-guinea pig model].

Selective phosphodiesterases (PDE) inhibitors are the new group of antiasthmatic drugs, which integrate antiinflammatory activity with bronchoconstriction counteraction. Selective inhibitors of phosphodiesterase type 4 are used as alternative or assist drugs in treatment of respiratory system diseases. So far glucocorticosteroids remain the most efficient and widely used medicine in the treatment of asthma. However application of glucocorticosteroid is greatly limited because of numerous side effects, what induce to permanent search for new antiasthmatic drugs. Examination new substances are executed on animal models. Guinea pig model is widely used to research course of asthmatic reaction. This model is especially convenient on the ground of that: lung is major shock organ, airway respond to histamine, animals demonstrated early asthmatic reaction (EAR) and late asthmatic reaction (LAR), eosinophils flow in bronchoalveolar space during LAR. In ovalbumin (OA) sensitized guinea pigs hypersensitivity reaction breaks out as a result of OA provocation. Aims of our experiments, execute on guinea pig model were to determine the influence of rolipram (PDE 4 inhibitor) on modulation experimental asthmatic reaction and comparison activity of rolipram versus dexamethasone in attribution to chosen parameters of allergic reaction such as: lung resistance, influx of protein and inflammatory cells in airways, and mastocytes degranulation. Experiments were made on guinea pigs sensitized and provoked with ovalbumin The obtain data indicate that rolipram was effective in reduction the rise of lung resistance during EAR, restricted influx of eosinophils to bronchoalveolar space between 1,5 and 24 hours after provocation, and reduced increase of histamine concentration in bronchoalveolar lavage fluid (BALf). Rolipram had no influence on number of neutrophils present in BALf. Dexamethasone in double dose of 1,2mg/kg effectively bordered the growth of lung resistance during EAR, and broke influx of eosinophils and neutrophils to bronchoalveolar space.

Administration, Inhalation↗
PubMed1996

Metabolic models of selection response.

Consequences of directional selection on metabolic flux are explored in models for which variation in flux among individuals is generated by segregation of allelic variants at enzyme activity loci. The pattern of selection response is strongly affected by the presence of genetic dominance and epistasis, which are automatically generated in metabolic systems. The expected magnitudes of dominance and epistasis effects on flux are evaluated. Small differences in enzyme activity generate little dominance, but a null allele will tend to be recessive for the pathway in which it occurs and for metabolically distant pathways. Epistasis is found to be greatest in short pathways in which large differences in enzyme activity occur. Under divergent artificial selection asymmetrical responses can occur due to the presence of directional dominance and epistasis, and lead to departures from the classic infinitesimal model of quantitative genetic variation. The effects of epistasis and dominance are in opposite directions, however, and partially cancel each other out in a diploid population.

Alleles↗
PubMed2005

Free knot splines for logistic models and threshold selection.

The logistic regression model has been in use in statistical analysis for many years. The paper introduces a spline model to remove the linear restriction on logit function. By considering knot locations as free variables, spline approximation of data is improved. The number of knots and the degree of the spline functions can still be determined by using a model selection procedure. Moreover, a knot, seen as a free parameter for a piecewise linear spline, represents a break point in the logit function which may be interpreted as a threshold value. This method is applied to a clinical trial for an in vitro fertilization program.

Clinical Trials as Topic↗
PubMed1999

Changes in microvessel endothelial cell gene expression in an in vitro human breast tumour endothelial cell model.

Selective targeting of tumour-endothelium has been proposed as a means of therapy. The successful exploitation of this approach will rely on the identification of suitable targets expressed specifically on the tumour-associated endothelium. In an attempt to identify novel tumour-endothelium associated targets we have used differential mRNA display to identify genes up-regulated in an in vitro breast tumour-endothelial cell culture model. Confluent monolayers of human mammary microvessel endothelial cells (HuMMEC) were incubated for 5 days with MDA-MB-231 breast adenocarcinoma cell-conditioned medium (TCM). mRNAs isolated from TCM-treated and control cells were amplified using 104 combinations of four 3(') anchored T(12)VN primers and 26 'random' 10mers by RT-PCR and the products examined on DNA sequencing gels. Seventy-four sequences were cloned and the differential expression of five genes was confirmed using dot-blots. These were identified as procollagen type-IV, Tie-2/Tek receptor tyrosine kinase, NADH dehydrogenase subunit-6, and ferritin heavy-chain, which were up-regulated, and insulin-like growth factor binding protein-5, which was down-regulated. Increased endothelial expression of basement membrane proteins and tyrosine kinase receptors is known to occur during angiogenesis. Our data support the use of this model for further in vitro investigation of tumour angiogenesis.

Journal Article↗
PubMed1986

Modeling of selection processes with age-dependent birth and death rates.

Two simple models for the competition and selection in age-dependent populations are developed and analyzed mathematically. Following Eigen, competition is introduced by the condition of constant overall-number of the population. In the first model this condition is satisfied by regulation of a dilution flux and in the second case by regulation of a food density. The calculation of maximal fitness is given explicitly for both situations. It is shown that fitness depends in a complicated way on the age-dependence of the birth and death rates. Therefore species have to develop special aging strategies in order to survive in a population under selection pressure. In general, early reproduction is of advantage and increases fitness.

Age Factors↗
PubMed1996

Predictive minimum description length criterion for time series modeling with neural networks.

Nonlinear time series modeling with a multilayer perceptron network is presented. An important aspect of this modeling is the model selection, i.e., the problem of determining the size as well as the complexity of the model. To overcome this problem we apply the predictive minimum description length (PMDL) principle as a minimization criterion. In the neural network scheme it means minimizing the number of input and hidden units. Three time series modeling experiments are used to examine the usefulness of the PMDL model selection scheme. A comparison with the widely used cross-validation technique is also presented. In our experiments the PMDL scheme and the cross-validation scheme yield similar results in terms of model complexity. However, the PMDL method was found to be two times faster to compute. This is significant improvement since model selection in general is very time consuming.

Algorithms↗