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 631 records · Page 35Linked to original sources
PubMed2001

The human macrophage cell line U937 as an in vitro model for selective evaluation of mycobacterial antigen-specific cytotoxic T-cell function.

Despite strong evidence for CD8+ T-cell function in murine mycobacterial infections, their corresponding role in human tuberculosis has proven more difficult to demonstrate. We have evaluated the human macrophage (Mphi) cell line U937 as an in vitro model for human leucocyte antigen (HLA) class I-restricted presentation of mycobacterial antigens, as HLA class I is constitutively expressed at high levels by U937 cells in the absence of detectable HLA class II or CD1 molecules. U937 cells were evaluated for their ability to phagocytose Mycobacterium tuberculosis and for their ability to present mycobacterial antigens to human HLA class I-matched cytotoxic T lymphocytes (CTLs). Differentiated U937 cells were capable of efficient phagocytosis of M. tuberculosis but did not generate a subsequent respiratory burst response, and were permissive for intracellular growth of both bacillus Calmette-Guérin (BCG) and the virulent M. tuberculosis H37Rv strain. CTL activity was restricted to live mycobacterial organisms and was shown to be mediated by M. tuberculosis-specific, HLA class I-matched, purified CD8+ CTL lines and CD8+ T-cell clones. Furthermore, M. tuberculosis-infected U937 targets were more rapidly and strongly lysed by CD8+ CTLs than were infected autologous Mphi. Finally, M. tuberculosis-infected U937 cells simultaneously provided a sensitive indicator for detection of mycobacterial-specific, HLA-unrestricted gammadelta+ CTL activity.

Antigen-Presenting Cells↗
PubMed1999

Prognostic factors for the outcome of chemotherapy in advanced soft tissue sarcoma: an analysis of 2,185 patients treated with anthracycline-containing first-line regimens--a European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group Study.

PURPOSE: A total of 2,185 patients with advanced soft tissue sarcomas who had been treated in seven clinical trials investigating the use of doxorubicin- or epirubicin-containing regimens as first-line chemotherapy were studied in this prognostic-factor analysis. PATIENTS AND METHODS: Overall survival time (median, 51 weeks) and response to chemotherapy (26% complete response or partial response) were the two end points. The cofactors were sex; age; performance status; prior therapies; the presence of locoregional or recurrent disease; lung, liver, and bone metastases at the time of entry onto the trial; long time period between the initial diagnosis of sarcoma and entry onto the study; and histologic type and grade. RESULTS: Univariate analyses showed (a) a significant, favorable influence of good performance status, young age, and absence of liver metastases on both survival time and response rate, (b) a significant, favorable influence of low histopathologic disease grade on survival time, despite a significantly lower response rate, (c) increased survival time for patients with a long time period between the initial diagnosis of sarcoma and entry onto the study, despite equivalent response rates, and (d) increased survival time with liposarcoma or synovial sarcoma, a decreased survival time with malignant fibrous histiocytoma, a lower response rate with leiomyosarcoma, and a higher response rate with liposarcoma (P < .05 for all log-rank and chi2 tests). The Cox model selected good performance status (P < .0001), absence of liver metastases (P = .0001), low histopathologic grade (P = .0002), long time lapse since initial diagnosis (P = .0004), and young age (P = .0045) as favorable prognostic factors of survival time. The logistic model selected absence of liver metastases (P < .0001), young age (P = .0024), high histopathologic grade (P = .0051), and liposarcoma (P = .0065) as favorable prognostic factors of response rate. CONCLUSION: This analysis demonstrates that for advanced soft tissue sarcoma, response to chemotherapy is not predicted by the same factors as is overall survival time. This needs to be taken into account in the interpretation of trials assessing the value of new agents for this disease on the basis of response to treatment.

Adult↗
PubMed2003

Computer modeling of selective regions in the active site of nitric oxide synthases: implication for the design of isoform-selective inhibitors.

Selective inhibition of nitric oxide synthase (NOS) isoforms has great therapeutic potential in the treatment of certain disease states arising from the pathological overproduction of nitric oxide. In this study three structures of each NOS isoform were employed to examine selective regions in the active site using the GRID/CPCA approach. In the GRID calculations, 10 probes covering hydrophobic, steric, and hydrogen-bond-acceptor and -donor interactions were used to calculate the molecular interaction fields (MIFs) in the active site. The side chain flexibility of the residues and the grid spacings were considered at the same time. Consensus principal component analysis (CPCA) was applied to analyze the MIFs differences in the active site between the NOS isoforms. By combining the cutout tool with GRID/CPCA pseudofield differential plots, several selective regions in the active site were identified. The selectivity analysis showed that the most important determinants for NOS inhibitor selectivity are hydrophobic and charge-charge interactions. Twenty-five inhibitors of NOS were then docked into the active site using the program AutoDock3.0. The regions identified as being important for selectivity by this method are in excellent agreement with inhibitor structure-activity relationships. A rational usage of the selective region described in this work should make it possible to develop NOS isoform-selective inhibitors.

Binding Sites↗
PubMed2006

Cannabinoid CB2/CB1 selectivity. Receptor modeling and automated docking analysis.

Three-dimensional models of the CB1 and CB2 cannabinoid receptors were constructed by means of a molecular modeling procedure, using the X-ray structure of bovine rhodopsin as the initial template, and taking into account the available site-directed mutagenesis data. The cannabinoid system was studied by means of docking techniques. An analysis of the interaction of WIN55212-2 with both receptors showed that CB2/CB1 selectivity is mainly determined by the interaction in the CB2 with the nonconserved residues S3.31 and F5.46, whose importance was suggested by site-directed mutagenesis data. We also carried out an automated docking of several ligands into the CB2 model, using the AUTODOCK 3.0 program; the good correlation obtained between the estimated free energy binding and the experimental binding data confirmed our binding hypothesis and the reliability of the model.

Amino Acid Sequence↗
PubMed1987

Simulation of genetic control of reproduction in beef cows. III. Within-herd breeding value estimation with known breeding dates.

Procedures for breeding value estimation for reproductive traits with known breeding dates were developed and tested using results of a computer simulation model of genetic control of bovine reproduction. The model generated realized reproductive outputs as a function of underlying genetic variation in two independent traits: conception rate (CR), which was indicative of the ability to conceive given estrus, and PPI, the postpartum interval from calving to first estrus. Two scenarios were considered. In the first, all cows were assumed to be cycling at the start of breeding and to be bred artificially. For this scenario, breeding values for CR could be estimated from information on observed breeding and calving dates by using a categorical trait, multi-stage selection model. Breeding value estimation for PPI, however, required actual measurement of PPI because if PPI and CR are genetically independent and if all cows are cycling at the beginning of breeding, subsequent breeding and calving dates are independent of PPI. The second scenario recognized that not all females would be cycling at the start of breeding. For this scenario, the categorical trait, multi-stage selection model could still be applied for breeding value estimation for CR, but accumulation of data across years was complicated by a need to consider the lifetime reproductive pattern of the individual rather than just the sum of each year's performances. Breeding value estimates for PPI could be obtained from observed breeding and calving dates for this scenario, but required consideration of the distributional properties of PPI.

Animals↗
PubMed2004

Assessing repeatability and validity of a visual analogue scale questionnaire for use in assessing pain and lameness in dogs.

OBJECTIVE: To develop a visual analogue scale (VAS) questionnaire that is repeatable and valid for use in assessing pain and lameness in dogs. SAMPLE POPULATION: 48 client-owned dogs with mild to moderate lameness. PROCEDURE: The dogs were from 3 studies conducted during a 3-year period. Of the 48 dogs, 19 were used in repeatability assessment, 48 were used in principal component analysis, and 44 were used in model selection procedures and validity testing. A test-retest measure of repeatability was conducted on dogs with a change of < 10% in vertical peak force. A force platform was used as the criterion-referenced standard for detecting lameness. Principal component analysis was used to describe dimensionality of the data. Repeatable questions were used as explanatory variables in multiple regression models to predict force plate measurements. Peak vertical, craniocaudal, and associated impulses were the forces used to quantify lameness. The regression models were used to test the criterion validity of the questionnaire. RESULTS: 19 of 39 questions were found to be repeatable on the basis of a Spearman rank-correlation cut point of > 0.6. Model selection procedures resulted in 3 overlapping subsets of questions that were considered valid representations of the forces measured (vertical peak, vertical impulse, and propulsion peak). Each reduced model fit the data as well as the full model. CONCLUSIONS AND CLINICAL RELEVANCE: The VAS questionnaire was repeatable and valid for use in assessing the degree of mild to moderate lameness in dogs.

Animals↗
PubMed2005

Pattern formation induced by ion-selective surfaces: models and simulations.

Simple inorganic reactions in gels, such as NaOH + CuCl(2), NaOH + AgNO(3), and CuCl(2) + K(3)[Fe(CN)(6)], can yield to various precipitation patterns. The first compound penetrates in a hydrogel by diffusion, and reacts with the second compound homogenized in the gel. The precipitate patterns formed in these reactions have got two kinds of bordering surfaces. Recent experimental results suggested that one of these surfaces has an ion-selective (semipermeable) character: It restrains the diffusion of the reacting ion contained by the reactant that diffuses into the gel. In this paper, we built the above experimental observation into a reaction-diffusion cellular-automata model of the pattern formation. Computer simulations showed that the model is able to reproduce the basic building elements of the patterns.

Journal Article↗
PubMed2005

Evolutionary optimization of radial basis function classifiers for data mining applications.

In many data mining applications that address classification problems, feature and model selection are considered as key tasks. That is, appropriate input features of the classifier must be selected from a given (and often large) set of possible features and structure parameters of the classifier must be adapted with respect to these features and a given data set. This paper describes an evolutionary algorithm (EA) that performs feature and model selection simultaneously for radial basis function (RBF) classifiers. In order to reduce the optimization effort, various techniques are integrated that accelerate and improve the EA significantly: hybrid training of RBF networks, lazy evaluation, consideration of soft constraints by means of penalty terms, and temperature-based adaptive control of the EA. The feasibility and the benefits of the approach are demonstrated by means of four data mining problems: intrusion detection in computer networks, biometric signature verification, customer acquisition with direct marketing methods, and optimization of chemical production processes. It is shown that, compared to earlier EA-based RBF optimization techniques, the runtime is reduced by up to 99% while error rates are lowered by up to 86%, depending on the application. The algorithm is independent of specific applications so that many ideas and solutions can be transferred to other classifier paradigms.

Algorithms↗
PubMed1991

Asymptotic behavior of nonlinear semigroup describing a model of selective cell growth regulation.

A new scheme of regulation of cell population growth is considered, called the selective growth regulation. The principle is that cells are withdrawn from proliferation depending on their contents of certain biochemical species. The dynamics of the cell population structured by the contents of this species is described by the functional integral equation model, previously introduced by the authors. The solutions of the model equations generate a semigroup of nonlinear positive operators. The main problem solved in this paper concerns stability of the equilibria of the model. This requires stating and proving of an original abstract result on the spectral radius of a perturbation of a semigroup of positive linear operators. Biological applications are discussed.

Cell Division↗
PubMed1996

Assessing the congruence of nursing models with organizational culture: a quality improvement perspective.

Model-based practice was identified by the Nursing Department at Riverview Health Centre, a 320-bed long-term care facility located in Winnipeg, Manitoba, as having the potential to enhance care quality significantly. To achieve real impact in the clinical setting, however, the model selected would need to reflect closely the culture and values of the department. It was decided to explore these phenomena using cross-method triangulation involving a cultural assessment survey (the Nursing Unit Cultural Assessment Tool) and focus groups. Patient comfort and empathy emerged consistently as core values for staff. Greater appreciation of the depth and complexity of the nursing department culture and values has provided invaluable direction vis-à-vis conceptual model selection.

Chi-Square Distribution↗
PubMed1978

Selected animal models for systematic antiatherosclerotic drug development.

We have developed an integrated system for antiatherosclerosis drug development utilizing rats, SEA quail, and cynomolgus monkeys as animal models. In general, the way the system is presently functioning is that thousands of compounds per year are randomly screened for hypobeta- or hyperalphalipoproteinemic activity in rats, and hundreds of compounds per year are screened in SEA quail for antiatherosclerotic and hypocholesterolaric activity. A few selected compounds that have activity in both rats and quail are then tested for lipoprotein modifying activity in cynomolgus monkeys. Nontoxic compounds having very good lipoprotein modifying activity in the monkey will then be recommended for clinical trials in man. We do not anticipate that this battery of testing will guarantee activity in man, but are hoping that it will at least increase the probability for finding a truly effective antiatherosclerotic drug to control the human disease.

Animals↗
PubMed2003

Prediction of cyclosporine dosage in patients after kidney transplantation using neural networks.

This paper proposes the use of neural networks for individualizing the dosage of cyclosporine A (CyA) in patients who have undergone kidney transplantation. Since the dosing of CyA usually requires intensive therapeutic drug monitoring, the accurate prediction of CyA blood concentrations would decrease the monitoring frequency and, thus, improve clinical outcomes. Thirty-two patients and different factors were studied to obtain the models. Three kinds of networks (multilayer perceptron, finite impulse response (FIR) network, and Elman recurrent network) and the formation of neural-network ensembles are used in a scheme of two chained models where the blood concentration predicted by the first model constitutes an input to the dosage prediction model. This approach is designed to aid in the process of clinical decision making. The FIR network, yielding root-mean-square errors (RMSEs) of 52.80 ng/mL and mean errors (MEs) of 0.18 ng/mL in validation (10 patients) showed the best blood concentration predictions and a committee of trained networks improved the results (RMSE = 46.97 ng/mL, ME = 0.091 ng/mL). The Elman network was the selected model for dosage prediction (RMSE = 0.27 mg/Kg/d, ME = 0.07 mg/Kg/d). However, in both cases, no statistical differences on the accuracy of neural methods were found. The models' robustness is also analyzed by evaluating their performance when noise is introduced at input nodes, and it results in a helpful test for models' selection. We conclude that neural networks can be used to predict both dose and blood concentrations of cyclosporine in steady-state. This novel approach has produced accurate and validated models to be used as decision-aid tools.

Administration, Oral↗
PubMed1995

An emergent model of orientation selectivity in cat visual cortical simple cells.

It is well known that visual cortical neurons respond vigorously to a limited range of stimulus orientations, while their primary afferent inputs, neurons in the lateral geniculate nucleus (LGN), respond well to all orientations. Mechanisms based on intracortical inhibition and/or converging thalamocortical afferents have previously been suggested to underlie the generation of cortical orientation selectivity; however, these models conflict with experimental data. Here, a 1:4 scale model of a 1700 microns by 200 microms region of layer IV of cat primary visual cortex (area 17) is presented to demonstrate that local intracortical excitation may provide the dominant source of orientation-selective input. In agreement with experiment, model cortical cells exhibit sharp orientation selectivity despite receiving strong iso-orientation inhibition, weak cross-orientation inhibition, no shunting inhibition, and weakly tuned thalamocortical excitation. Sharp tuning is provided by recurrent cortical excitation. As this tuning signal arises from the same pool of neurons that it excites, orientation selectivity in the model is shown to be an emergent property of the cortical feedback circuitry. In the model, as in experiment, sharpness of orientation tuning is independent of stimulus contrast and persists with silencing of ON-type subfields. The model also provides a unified account of intracellular and extracellular inhibitory blockade experiments that had previously appeared to conflict over the role of inhibition. It is suggested that intracortical inhibition acts nonspecifically and indirectly to maintain the selectivity of individual neurons by balancing strong intracortical excitation at the columnar level.

4,4'-Diisothiocyanostilbene-2,2'-Disulfonic Acid↗
PubMedDate not supplied

A model for selecting a clinical laboratory information system. A four-phase process.

Laboratory information systems (LISs) have become an essential part of an efficient and effective laboratory. In the past, selecting an LIS was a relatively simple procedure because there were only a few candidates to select from. Now, however, selecting an LIS from the myriad available has become a time-consuming and complex process. This article presents a multi-attribute utility (MAU) method for selecting an LIS. This MAU method has four phases: identifying LIS vendors and disseminating requests for proposals (RFPs), analyzing RFP responses and selecting the top three vendors, validating responses of the top three vendors, and selecting a primary vendor and preparing a formal recommendation. By following this four-phase process, laboratories will simplify the complex LIS selection process and increase their chances of selecting the best LIS for their needs.

Clinical Laboratory Information Systems↗
PubMed1993

A model of selective synapse formation in sympathetic ganglia.

In the sympathetic system, neurons from several spinal segments are mapped onto targets in the periphery in a topographically ordered way by means of selective synaptic connections in the superior cervical ganglion. Experimental evidence points to a crucial role for chemoaffinity in establishing this topographic map. Furthermore, rearrangements of synapses after surgical manipulations indicate that this chemoaffinity is not based on rigid "key-and-lock" markers. Our model is used to study how such nonrigid markers may interact with other regulatory factors, including growth-regulating signals and the growth potential of individual neurons. In the model, these latter factors are limiting, so that an increasing number of synaptic contacts decreases the likelihood of further synapse formation. These factors are combined with chemoaffinity using a linear threshold model. The model is robust to parameter changes and reproduces experimental observations with reasonable detail. Simulation results are used to discuss characteristic experimental results, such as the substantial plasticity of the connections seen after partial denervation. A surprisingly small effect of transient hyperinnervation in the model may help explain why final connectivities are similar in two real situations with high and low degrees of transient hyperinnervation (development and adult reinnervation). It is shown that spatial restrictions on post-synaptic neurons (dendrites) may contribute significantly to the segmentally broad innervation of each ganglion cell. Finally, we discuss potential effects of presynaptic neuronal death in systems with a high degree of plasticity.

Animals↗
PubMed2001

Efficient regression calibration for logistic regression in main study/internal validation study designs with an imperfect reference instrument.

An extension to the version of the regression calibration estimator proposed by Rosner et al. for logistic and other generalized linear regression models is given for main study/internal validation study designs. This estimator combines the information about the parameter of interest contained in the internal validation study with Rosner et al.'s regression calibration estimate, using a generalized inverse-variance weighted average. It is shown that the validation study selection model can be ignored as long as this model is jointly independent of the outcome and the incompletely observed covariates, conditional, at most, upon the surrogates and other completely observed covariates. In an extensive simulation study designed to follow a complex, multivariate setting in nutritional epidemiology, it is shown that with validation study sizes of 340 or more, this estimator appears to be asymptotically optimal in the sense that it is nearly unbiased and nearly as efficient as a properly specified maximum likelihood estimator. A modification to the regression calibration variance estimator which replaces the standard uncorrected logistic regression coefficient variance with the sandwich estimator to account for the possible misspecification of the logistic regression fit to the surrogate covariates in the main study, was also studied in this same simulation experiment. In this study, the alternative variance formula yielded results virtually identical to the original formula. A version of the proposed estimator is also derived for the case where the reference instrument, available only in the validation study, is imperfect but unbiased at the individual level and contains error that is uncorrelated with other covariates and with error in the surrogate instrument. Replicate measures are obtained in a subset of study participants. In this case it is shown that the validation study selection model can be ignored when sampling into the validation study depends, at most, only upon perfectly measured covariates. Two data sets, a study of fever in relation to occupational exposure to antineoplastics among hospital pharmacists and a study of breast cancer incidence in relation to dietary intakes of alcohol and vitamin A, adjusted for total energy intake, from the Nurses' Health Study, were analysed using these new methods. In these data, because the validation studies contained less than 200 observations and the events of interest were relatively rare, as is typical, the potential improvements offered by this new estimator were not apparent.

Adult↗