SEARCH · PubMed Health

Results for “regularized model”

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 181 records · Page 10Linked to original sources

PubMed2006

Stochastic complexities of general mixture models in variational Bayesian learning.

In this paper, we focus on variational Bayesian learning of general mixture models. Variational Bayesian learning was proposed as an approximation of Bayesian learning. While it has provided computational tractability and good generalization in many applications, little has been done to investigate its theoretical properties. The asymptotic form was obtained for the stochastic complexity, or the free energy in the variational Bayesian learning of a mixture of exponential-family distributions, which is the main contribution this paper makes. We reveal that the stochastic complexities become smaller than those of regular statistical models, which implies that the advantages of Bayesian learning are still retained in variational Bayesian learning. Moreover, the derived bounds indicate what influence the hyperparameters have on the learning process, and the accuracy of the variational Bayesian approach as an approximation of true Bayesian learning.

Animals↗

PubMed2001

Integrated exposure and dose modeling and analysis system. 3. Deposition of inhaled particles in the human respiratory tract.

Detailed information on the composition-resolved size distribution of particulate matter deposited along the human respiratory tract can help linking epidemiological, toxicological, and pathological studies and thus potentially improve the understanding of the origin of pulmonary disorders induced by respirable pathogens. For this purpose, a new mechanistic dosimetry model describing the dynamics of respirable particles in the human airways was developed. Model predictions of transport and fate of inhaled aerosols are based on solutions of the aerosol general dynamic equation, which describes changes in particle size and mass distributions resulting from processes such as nucleation, condensation, coagulation, gas phase chemical reaction, and deposition. To compensate for approximating the three-dimensional problem by considering only axial variations along the airways, boundary layer effects are introduced via appropriate dimensionless transport parameters. The architecture of the human lung is described by Weibel's simple regular dichotomous model. An important advantage of the present approach is that it allows testing the significance of intersubject lung morphology and ventilation variability for particle deposition and dose calculations. The model predicts the evolution of size and composition distributions of inhaled particles and the deposition profile along the human lower respiratory tract: in general, model predictions are in qualitative and quantitative agreement with tracheobronchial and alveolar deposition data.

Aerosols↗

PubMed2004

Biomechanical comparison of inside-outside screws, cables, and regular screws, using a sawbone model.

Our aim was to conduct a biomechanical comparison of the pull-out strengths of inside-outside (I/O) screws, cables, and bone screws to determine whether I/O screws provide greater pull-out resistance than cables or bone screws, and their effectiveness with the screw diameter. There is no remarkable biomechanical experimental study comparing the I/O technique with conventional spinal techniques. The diameter of the screw heads were also biomechanically tested to determine the optimal size that can be used. In this study, 45 blocks of 50x50x5 mm of "sawbone" (synthetic bone, model 1137, Pacific Research Laboratories, Vashon, WA, USA) were used as bone substitutes. Fifteen sets of 14-mm inside-outside Dynalok screws and nuts, 15 wire cables, and 15 bone screws were inserted into a separate sawbone block. An MTS Bionx materials testing machine was used to measure the load to failure of each implant. The mean values and standard deviations of each group were calculated and Student's t-test was used for comparison. The load to failure of the inside-outside screws was significantly greater than that of the cables (p<0.0000004) and the regular bone screws (p<0.000002). The results also revealed that increasing the diameter of the head of the screw also increases the resistance against the pull-out strengths. Thus, using a larger screw in occipitocervical stabilization provides safe and stable fixation of the occipital bone to the cervical spine. This study also proved that sawbone is a useful and reliable alternative to allogenic fresh cadaveric bone grafts or animal bones for certain biomechanical testing.

Atlanto-Occipital Joint↗

PubMed1985

Controlled trials in gastrodyspepsia: a methodological aspect.

Non-ulcer dyspepsia (NUD) is a poorly defined condition that is not very suitable for conventional randomised double-blind studies. A multi cross-over model (MCO-model) has been designed allowing identification of individual drug responders with a defined degree of certainty. The model involves regular interchanges between periods with active drug and placebo, and the evaluation is based on the number of times the active drug is associated with fewer symptoms than the preceding or following placebo period (X-score). A drug responder may be defined by a certain minimum value for the X-score. The risk of being wrong may then be easily calculated from the probability distribution of the X-score. The effect of cimetidine in patients with NUD has been studied using a variant of the MCO-model including 6 treatment periods of 2 or 4 days' duration. So far, the conclusion that the MCO-model is able to identify individual cimetidine responders among patients with NUD appears to be justified. The preliminary findings furthermore suggest that cimetidine responders among patients with NUD are characterised rather by symptoms suggestive of reflux esophagitis than by hypersecretion of acid.

Cimetidine↗

PubMed2004

Characterizing the evolutionary dynamics of language.

In a recent article Mitchener and Nowak present a model of the evolutionary dynamics of language. The model exhibits regular and chaotic oscillations in changes to the proportions of grammars spoken in a population over the course of evolution. These oscillations are within the purview of evolutionary game theory, but they suggest the lack of an evolutionarily stable strategy. Implications for self-organization across scales of adaptation are discussed.

Adaptation, Psychological↗

PubMed1997

Hydraulic optimalization of surgical flow drainage in physical model.

In the present study, the flow in deep, closed wound with various types and position of drain was studied. A physical model of regular geometrical wound was designed to investigate the events occurring in the wound from the standpoint of hydraulics. Results of the study showed that the commonly used method of drain application could be improved. The results may also prove helpful for improvement of the existing drains and their more favorable positioning.

Drainage↗

PubMed1999

Crystal structure of the two-RRM domain of hnRNP A1 (UP1) complexed with single-stranded telomeric DNA.

Human hnRNP A1 is a versatile single-stranded nucleic acid-binding protein that functions in various aspects of mRNA maturation and in telomere length regulation. The crystal structure of UP1, the amino-terminal domain of human hnRNP A1 containing two RNA-recognition motifs (RRMs), bound to a 12-nucleotide single-stranded telomeric DNA has been determined at 2.1 A resolution. The structure of the complex reveals the basis for sequence-specific recognition of the single-stranded overhangs of human telomeres by hnRNP A1. It also provides insights into the basis for high-affinity binding of hnRNP A1 to certain RNA sequences, and for nucleic acid binding and functional synergy between the RRMs. In the crystal structure, a UP1 dimer binds to two strands of DNA, and each strand contacts RRM1 of one monomer and RRM2 of the other. The two DNA strands are antiparallel, and regions of the protein flanking each RRM make important contacts with DNA. The extensive protein-protein interface seen in the crystal structure of the protein-DNA complex and the evolutionary conservation of the interface residues suggest the importance of specific protein-protein interactions for the sequence-specific recognition of single-stranded nucleic acids. Models for regular packaging of telomere 3' overhangs and for juxtaposition of alternative 5' splice sites are proposed.

Alternative Splicing↗

PubMed2001

[Establishment of rabbit model for peritoneal dialysis related peritonits].

OBJECTIVE: To establish a rabbit model corresponding with clinical characteristics of peritoneal dialysis related peritonitis. METHODS: Twelve New Zealand rabbits were randomly divided into peritonitis group and control group. They were infused standard living Staphylococcus aureus 3 x 10(9) colony forming unit (cfu) (peritonitis group) or 0.9% NaCl 1 ml (control group). After 48 hours, Laboratory rabbits in acute uremic model underwent regular dialysis for 6 hours. At different points of time, the concentration of glucose in dialysate and the dialysate/plasma (D/P) ratio of the concentration creatinine (Cr), total protein(TP), album(Alb) and IgG, the counts of leukocyte in dialysate and water ultrafiltrate were determined. RESULTS: D/D0 were significantly lower in the peritonitis group than that in the control group (P < 0.05). The counts of leukocyte and D/P ratio of Cr, TP, Alb and IgG were significantly increased in the peritonitis group (all P < 0.05). The drained volume was lower in the peritonitis group than that in the control group (but P > 0.05). CONCLUSION: In the rabbit model infused intraperoneally staphylococcus aureus, (1) transport peritoneum of low and large molecular weight solutes is increased, (2) the volume of ultrafiltrate is decreased, (3) the counts of leukocyte in dialysate is increased. These change in this model is basic all correspond with the clinical feature of patients of peritoneal dialysis related peritonitis. So this rabbit model of peritoneal dialysis related peritonitis is a successful experimental infection model.

Animals↗

PubMed2005

On the use of the Price equation.

This paper distinguishes two categories of questions that the Price equation can help us answer. The two different types of questions require two different disciplines that are related, but nonetheless move in opposite directions. These disciplines are probability theory on the one hand and statistical inference on the other. In the literature on the Price equation this distinction is not made. As a result of this, questions that require a probability model are regularly approached with statistical tools. In this paper, we examine the possibilities of the Price equation for answering questions of either type. By spending extra attention on mathematical formalities, we avoid the two disciplines to get mixed up. After that, we look at some examples, both from kin selection and from group selection, that show how the inappropriate use of statistical terminology can put us on the wrong track. Statements that are 'derived' with the help of the Price equation are, therefore, in many cases not the answers they seem to be. Going through the derivations in reverse can, however, be helpful as a guide how to build proper (probabilistic) models that do give answers.

Animals↗

PubMed2004

Determination of solid surface tension from particle-substrate pull-off forces measured with the atomic force microscope.

Atomic force microscopy (AFM) is capable of solid surface characterization at the microscopic and submicroscopic scales. It can also be used for the determination of surface tension of solids (gamma) from pull-off force (F) measurements, followed by analysis of the measured F values using contact mechanics theoretical models. Although a majority of the literature gamma results was obtained using either Johnson-Kendall-Roberts (JKR) or Derjaguin-Muller-Toporov (DMT) models, re-analysis of the published experimental data presented in this paper indicates that these models are regularly misused. Additional complication in determination of gamma values using the AFM technique is that the measured pull-off forces have poor reproducibility. Reproducible and meaningful F values can be obtained with strict control over AFM experimental conditions during the pull-off force measurements (low humidity level, controlled and known loads) for high quality substrates and probes (surfaces should be free of heterogeneity, roughness, and contamination). Any probe or substrate imperfections complicate the interpretation of experimental results and often reduce the quality of the generated data. In this review, surface imperfection in terms of roughness and heterogeneity that influence the pull-off force are analyzed based upon the contact mechanics models. Simple correlations are proposed that could guide in selection and preparation of AFM probes and substrates for gamma determination and selection of loading conditions during the pull-off force measurements. Finally, the possibility of AFM measurements of solid surface tension using materials with rough surfaces is discussed.

Journal Article↗

PubMed1996

Adsorption of globular proteins on locally planar surfaces: models for the effect of excluded surface area and aggregation of adsorbed protein on adsorption equilibria.

Equilibrium statistical-thermodynamic models are presented for the surface adsorption of proteins modeled as regular convex hard particles. The adsorbed phase is treated as a two-dimensional fluid, and the chemical potential of adsorbed protein is obtained from scaled particle theory. Adsorption isotherms are calculated for nonassociating and self-associating adsorbing proteins. Area exclusion broadens adsorption isotherms relative to the Langmuir isotherm (negative cooperativity), whereas self-association steepens them (positive cooperativity). The calculated isotherm for adsorption of hard spheres using scaled particle theory for hard discs agrees well with that calculated from the hard disc virial expansion. As the cross section of the adsorbing protein in the plane of the surface becomes less discoidal, the apparent negative cooperativity manifested in the isotherm becomes more pronounced. The model is extended to the case of simultaneous adsorption of a tracer protein at low saturation and a competitor protein with a different size and/or shape at arbitrary fractional saturation. Area exclusion by competitor for tracer (and vice versa) is shown to substantially enhance the displacement of tracer by competitor and to qualitatively invalidate the standard interpretation of ligand competition experiments, according to which the fractional displacement of tracer by competitor is equal to the fractional saturation by competitor.

Adsorption↗

PubMed2001

Virulence-associated genes in avian pathogenic Escherichia coli (APEC) isolated from internal organs of poultry having died from colibacillosis.

Escherichia coli infections are responsible for significant losses in the poultry industry in many parts of the world. The pathogenesis and the role of virulence factors are not yet totally elucidated. We, therefore, examined 150 E. coli strains isolated from visceral organs of poultry having died from colibacillosis for the presence of virulence-associated genes by PCR. The E. coli strains were investigated for the presence of a total of 17 virulence-associated genes described for diarrheagenic (stx1/2, eae, hlyEHEC, estl, eltI, astA, cdtb), septicemic (hlyA, papC, cnf1/2, fyuA, irp2) and avian pathogenic E. coli (APEC; iucD, tsh, fimC, and hlyE as well as stx2f). Seven genes were significantly distributed among APEC strains, while most of the other investigated genes could be demonstrated only sporadically or not at all. FimC (Type I fimbriae) was detected with the highest prevalence in 92.7% of the isolates. Most of the strains harboring iucD (88.7%) also gave positive results for tsh (85.3%). Genes fyuA (ferric yersiniabactin uptake) (66.0%) and irp2 (iron-repressible protein) (68.0%), necessary for Yersinia to acquire iron in the mouse infection model, were regularly detected in combination. Moreover, we found papC (pyelonephritis-associated pili) in 30.0% and astA (enteroaggregative heat stable toxin) in 17.3% of the field strains. A significant amount of strains (57.3%) harbored a combination of iucD, tsh, fimC, fyuA and irp2 virulence-associated genes, presumably rendering these strains particularly virulent. These findings provide novel insights into the presence and distribution of virulence-associated genes in avian pathogenic E. coli field strains, which will help to more comprehensively characterize APEC in future epidemiological studies. It is assumed that the existence of two iron acquisition systems points towards their important role in virulence. Furthermore, we suggest that characterization of the respective phenotypes in infection models will provide substantial information to better understand the pathogenesis of colibacillosis in poultry.

Animals↗

PubMed1998

Crystal structure of a colicin N fragment suggests a model for toxicity.

BACKGROUND: Pore-forming colicins are water-soluble bacteriocins capable of binding to and translocating through the Escherichia coli cell envelope. They then undergo a transition to a transmembrane ion channel in the cytoplasmic membrane leading to bacterial death. Colicin N is the smallest pore-forming colicin known to date (40 kDa instead of the more usual 60 kDa) and the crystal structure of its membrane receptor, the porin OmpF, is already known. Structural knowledge of colicin N is therefore important for a molecular understanding of colicin toxicity and is relevant to toxic mechanisms in general. RESULTS: The crystal structure of colicin N reveals a novel receptor-binding domain containing a six-stranded antiparallel beta sheet wrapped around the 63 A long N-terminal alpha helix of the pore-forming domain. The pore-forming domain adopts a ten alpha-helix bundle that has been observed previously in the pore-forming domains of colicin A, la and E1. The translocation domain, however, does not appear to adopt any regular structure. Models for receptor binding and translocation through the outer membrane are proposed based on the structure and biochemical data. CONCLUSIONS: The colicin N-ompF system is now the structurally best-defined translocation pathway. Knowledge of the colicin N structure, coupled with the structure of its receptor, OmpF, and previously published biochemical data, limits the numerous possibilities of translocation and leads to a model in which the translocation domain inserts itself through the porin pore, the receptor-binding domain stays outside and the pore-forming domain translocates along the outer wall of the trimeric porin channel.

Binding Sites↗

PubMed2001

Robust full Bayesian learning for radial basis networks.

We propose a hierarchical full Bayesian model for radial basis networks. This model treats the model dimension (number of neurons), model parameters, regularization parameters, and noise parameters as unknown random variables. We develop a reversible-jump Markov chain Monte Carlo (MCMC) method to perform the Bayesian computation. We find that the results obtained using this method are not only better than the ones reported previously, but also appear to be robust with respect to the prior specification. In addition, we propose a novel and computationally efficient reversible-jump MCMC simulated annealing algorithm to optimize neural networks. This algorithm enables us to maximize the joint posterior distribution of the network parameters and the number of basis function. It performs a global search in the joint space of the parameters and number of parameters, thereby surmounting the problem of local minima to a large extent. We show that by calibrating the full hierarchical Bayesian prior, we can obtain the classical Akaike information criterion, Bayesian information criterion, and minimum description length model selection criteria within a penalized likelihood framework. Finally, we present a geometric convergence theorem for the algorithm with homogeneous transition kernel and a convergence theorem for the reversible-jump MCMC simulated annealing method.

Journal Article↗

PubMed2005

Interacting damage models mapped onto Ising and percolation models.

We introduce a class of damage models on regular lattices with isotropic interactions between the broken cells of the lattice. Quasi-static fiber bundles are an example. The interactions are assumed to be weak, in the sense that the stress perturbation from a broken cell is much smaller than the mean stress in the system. The system starts intact with a surface-energy threshold required to break any cell sampled from an uncorrelated quenched-disorder distribution. The evolution of this heterogeneous system is ruled by Griffith's principle which states that a cell breaks when the release in potential (elastic) energy in the system exceeds the surface-energy barrier necessary to break the cell. By direct integration over all possible realizations of the quenched disorder, we obtain the probability distribution of each damage configuration at any level of the imposed external deformation. We demonstrate an isomorphism between the distributions so obtained and standard generalized Ising models, in which the coupling constants and effective temperature in the Ising model are functions of the nature of the quenched-disorder distribution and the extent of accumulated damage. In particular, we show that damage models with global load sharing are isomorphic to standard percolation theory and that damage models with a local load sharing rule are isomorphic to the standard Ising model, and draw consequences thereof for the universality class and behavior of the autocorrelation length of the breakdown transitions corresponding to these models. We also treat damage models having more general power-law interactions, and classify the breakdown process as a function of the power-law interaction exponent. Last, we also show that the probability distribution over configurations is a maximum of Shannon's entropy under some specific constraints related to the energetic balance of the fracture process, which firmly relates this type of quenched-disorder based damage model to standard statistical mechanics.

Journal Article↗

PubMed1997

Fully Bayesian estimation of Gibbs hyperparameters for emission computed tomography data.

In recent years, many investigators have proposed Gibbs prior models to regularize images reconstructed from emission computed tomography data. Unfortunately, hyperparameters used to specify Gibbs priors can greatly influence the degree of regularity imposed by such priors and, as a result, numerous procedures have been proposed to estimate hyperparameter values from observed image data. Many of these procedures attempt to maximize the joint posterior distribution on the image scene. To implement these methods, approximations to the joint posterior densities are required, because the dependence of the Gibbs partition function on the hyperparameter values is unknown. In this paper, we use recent results in Markov chain Monte Carlo (MCMC) sampling to estimate the relative values of Gibbs partition functions and using these values, sample from joint posterior distributions on image scenes. This allows for a fully Bayesian procedure which does not fix the hyperparameters at some estimated or specified value, but enables uncertainty about these values to be propagated through to the estimated intensities. We utilize realizations from the posterior distribution for determining credible regions for the intensity of the emission source. We consider two different Markov random field (MRF) models-the power model and a line-site model. As applications we estimate the posterior distribution of source intensities from computer simulated data as well as data collected from a physical single photon emission computed tomography (SPECT) phantom.

Algorithms↗

PubMed2004

Solving the neuroimaging puzzle: the multimodal integration of neuroelectromagnetic and functional magnetic resonance recordings.

In this chapter, advanced methods for the modeling of human cortical activity from combined high-resolution electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) data are reviewed. These methods include a subject's multicompartment head model (scalp, skull, dura mater, cortex) constructed from magnetic resonance images, multidipole source model, and regularized linear inverse source estimates. Determination of the priors in the resolution of the linear inverse problem was performed with the use of information from the hemodynamic responses of the cortical areas as revealed by block-designed (strength of activated voxels) and event-related (coupling of activated voxels) fMRI. As an example, these methods were applied to EEG (128 electrodes) and fMRI data, which were recorded in separate sessions while normal subjects executed voluntary right one-digit movements.

Brain Mapping↗

PubMed2004

Is B-type natriuretic peptide-guided heart failure management cost-effective?

BACKGROUND: Chronic heart failure (CHF) is a common medical problem and B-type natriuretic peptide (BNP)-guided heart failure management for outpatients with symptomatic CHF was found to reduce the readmission rate and mortality, but the costs of treatment may provoke concern in the current cost-conscious clinical setting. METHODS: We conducted a cost-effectiveness analysis using a Markov model of regular BNP measurement in the outpatient setting. The target population was symptomatic CHF patients aged 35-85 years recently discharged from the hospital. Intervention was BNP measurement once every 3 months (BNP group) or no BNP measurement (clinical group). Clinical and utility data were retrieved from published studies. Costs were based on published data in the US. Cost-effectiveness was measured by $ per quality-adjusted life year (QALY). RESULTS: The baseline analysis during the 9-month period after hospitalization showed the QALY to be longer for the BNP group (0.57 for the BNP group and 0.55 for the clinical group) and the costs were also lower for the BNP group ($9577 and 10,131). The dominance of the BNP group continued during the 1-year follow-up. The incremental costs would be $3491-7787 per QALY. In sensitivity analyses, two parameters with strong effects on the cost-effectiveness were the probability of the first readmission for CHF in the clinical group and the costs for inpatient CHF care. When these two parameters were simultaneously put on the simulation model, the incremental costs of the BNP group may exceed $50,000. CONCLUSIONS: Introduction of BNP measurement in heart failure management may be cost-effective.

Adult↗