Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Ensemble methods”

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 523 records · Page 29Linked to original sources
PubMed1999

Wavelet-based enhancement of signal-averaged electrocardiograms for late potential detection.

An optimal wavelet filter to improve the signal-to-noise ratio (SNR) of the signal-averaged electrocardiogram is described. As the averaging technique leads to the best unbiased estimator, the challenge is to attenuate the noise while preserving the low amplitude signals that are usually embedded in it. An optimal, in the mean-square sense, wavelet-based filter has been derived from the model of the signal. However, such a filter needs exact knowledge of the noise statistic and the noise-free signal. Hence, to implement such a filter, a method based on successive sub-averaging and wavelet filtering is proposed. Its performance was evaluated using simulated and real ECGs. An improvement in SNR of between 6 and 10 dB can be achieved compared to a classical averaging technique which uses an ensemble of 64 simulated ECG beats. Tests on real ECGs demonstrate the utility of the method as it has been shown that by using fewer beats in the filtered ensemble average, one can achieve the same noise reduction. Clinical use of this technique would reduce the ensemble needed for averaging while obtaining the same diagnostic result.

Electrocardiography↗
PubMed2006

Improved diffusion Monte Carlo propagators for bosonic systems using Itô calculus.

The construction of importance sampled diffusion Monte Carlo (DMC) schemes accurate to second order in the time step is discussed. A central aspect in obtaining efficient second order schemes is the numerical solution of the stochastic differential equation (SDE) associated with the Fokker-Plank equation responsible for the importance sampling procedure. In this work, stochastic predictor-corrector schemes solving the SDE and consistent with Itô calculus are used in DMC simulations of helium clusters. These schemes are numerically compared with alternative algorithms obtained by splitting the Fokker-Plank operator, an approach that we analyze using the analytical tools provided by Ito; calculus. The numerical results show that predictor-corrector methods are indeed accurate to second order in the time step and that they present a smaller time step bias and a better efficiency than second order split-operator derived schemes when computing ensemble averages for bosonic systems. The possible extension of the predictor-corrector methods to higher orders is also discussed.

Journal Article↗
PubMed2004

An artificial neural network ensemble to predict disposition and length of stay in children presenting with bronchiolitis.

BACKGROUND: Artificial neural networks apply complex non-linear functions to pattern recognition problems. An ensemble is a 'committee' of neural networks that usually outperforms single neural networks. Bronchiolitis is a common manifestation of viral lower respiratory tract infection in infants and toddlers. OBJECTIVE: To train artificial neural network ensembles to predict the disposition and length of stay in children presenting to the Emergency Department with bronchiolitis. METHODS: A specifically constructed database of 119 episodes of bronchiolitis was used to train, validate, and test a neural network ensemble. We used EasyNN 7.0 on a 200 Mhz pentium PC with a maths co-processor. The ensemble of neural networks constructed was subjected to fivefold validation. Comparison with actual and predicted dispositions was measured using the kappa statistic for disposition and the Kaplan-Meier estimations and log rank test for predictions of length of stay. RESULTS: The neural network ensembles correctly predicted disposition in 81% (range 75-90%) of test cases. When compared with actual disposition the neural network performed similarly to a logistic regression model and significantly better than various 'dumb machine' strategies with which we compared it. The prediction of length of stay was poorer, 65% (range 60-80%), but the difference between observed and predicted lengths of stay were not significantly different. CONCLUSION: Artificial neural network ensembles can predict disposition for infants and toddlers with bronchiolitis; however, the prediction of length of hospital stay is not as good.

Bronchiolitis↗
PubMed2007

Robust fluoroscopic respiratory gating for lung cancer radiotherapy without implanted fiducial markers.

For gated lung cancer radiotherapy, it is difficult to generate accurate gating signals due to the large uncertainties when using external surrogates and the risk of pneumothorax when using implanted fiducial markers. We have previously investigated and demonstrated the feasibility of generating gating signals using the correlation scores between the reference template image and the fluoroscopic images acquired during the treatment. In this paper, we present an in-depth study, aiming at the improvement of robustness of the algorithm and its validation using multiple sets of patient data. Three different template generating and matching methods have been developed and evaluated: (1) single template method, (2) multiple template method, and (3) template clustering method. Using the fluoroscopic data acquired during patient setup before each fraction of treatment, reference templates are built that represent the tumour position and shape in the gating window, which is assumed to be at the end-of-exhale phase. For the single template method, all the setup images within the gating window are averaged to generate a composite template. For the multiple template method, each setup image in the gating window is considered as a reference template and used to generate an ensemble of correlation scores. All the scores are then combined to generate the gating signal. For the template clustering method, clustering (grouping of similar objects together) is performed to reduce the large number of reference templates into a few representative ones. Each of these methods has been evaluated against the reference gating signal as manually determined by a radiation oncologist. Five patient datasets were used for evaluation. In each case, gated treatments were simulated at both 35% and 50% duty cycles. False positive, negative and total error rates were computed. Experiments show that the single template method is sensitive to noise; the multiple template and clustering methods are more robust to noise due to the smoothing effect of aggregation of correlation scores; and the clustering method results in the best performance in terms of computational efficiency and accuracy.

Biophysical Phenomena↗
PubMed2006

Comparison of different torsion angle approaches for NMR structure determination.

A new procedure for NMR structure determination, based on the Internal Coordinate Molecular Dynamics (ICMD) approach, is presented. The method finds biopolymer conformations that satisfy usual NMR-derived restraints by using high temperature dynamics in torsion angle space. A variable target function algorithm gradually increases the number of NOE-based restraints applied, with the treatment of ambiguous and floating restraints included. This soft procedure allows combining artificially high temperature with a general purpose force-field including Coulombic and Lennard-Jones non-bonded interactions, which improves the quality of the ensemble of conformations obtained in the gas-phase. The new method is compared to existing algorithms by using the structures of eight ribosomal proteins earlier obtained with state-of-the-art procedures and included into the RECOORD database [Nederveen, A., Doreleijers, J., Vranken, W., Miller, Z., Spronk, C., Nabuurs, S., Guntert, P., Livny, M., Markley, M., Nilges, M., Ulrich, E., Kaptein, R. and Bonvin, A.M. (2005) Proteins, 59, 662-672]. For the majority of tested proteins, the ICMD algorithm shows similar convergence and somewhat better quality Z scores for the phi, psi distributions. The new method is more computationally demanding although the overall load is reasonable.

Algorithms↗
PubMed2005

Direct evaluation of multicomponent phase equilibria using flat-histogram methods.

We present a method for directly locating density-driven phase transitions in multicomponent systems. Phase coexistence conditions are determined through manipulation of a total density probability distribution evaluated over a density range that includes both coexisting phases. Saturation quantities are determined through appropriate averaging of density-dependent mean values of a given property of interest. We discuss how to implement the method in both the grand-canonical and isothermal-isobaric semigrand ensembles. Calculations can be conducted using any of the recently introduced flat-histogram techniques. Here, we combine the general algorithm with a transition-matrix approach to produce an efficient self-adaptive technique for determining multicomponent phase equilibrium properties. To assess the performance of the new method, we generate phase diagrams for a number of binary and ternary Lennard-Jones mixtures.

Journal Article↗
PubMed2000

Intermodal selective attention in monkeys. II: physiological mechanisms of modulation.

Of all areas studied in the accompanying study, attention effects were most consistent and well resolved in V4. In this study, to define some of the anatomical circuits and neural processes underlying the influence of attention, we examined the laminar distribution and physiology of attention effects in V4 and in two lower areas, V1 and V2. Laminar event-related potential (ERP), current source density (CSD) and multiunit activity (MUA) profiles allowed identification of processes occurring in the local ensembles, as well as their sequence and laminar distribution. These methods also permitted us to analyze the brain processes reflected in attention-sensitive components of the surface ERP. As outlined in the previous study, the first robust modulation by attention occurred in V4 during the 100-300 ms poststimulus interval. This is the time frame of the net refractoriness which follows the net local excitatory response to luminance increment. Over this interval, attention reduced CSD amplitudes and increased action potential firing rates, findings consistent with disinhibition as a mechanism for attention in V4. Similar effects were observed during the 100-300 ms time frame in V2. In V4, attention had no effect on the initial excitatory response at the depth of lamina 4, but it did produce large modulations in supragranular and deep laminae, origins of both feedforward and feedback projections. Attentional modulation in V2 was later than in V4 and concentrated in extragranular laminae, with no modulation of the initial layer 4 response. Attentional modulation in V1 was smaller and still later than that in V2 and was focused in the supragranular laminae. In this paradigm, attention did not modulate either the response in lateral geniculate nucleus (LGN) or the initial excitation in lamina 4C of V1. The timing of effects across areas and the laminar distribution of effects within areas indicate that attention effects are mediated by feedback projections. Moreover, our findings suggest that attention may increase the perceptual salience of stimuli by reducing stimulus-evoked refractoriness and/or inhibition in cortical ensembles. Finally, attentional modulation of transmembrane current flow in V4 produced a sustained negative deflection in the laminar ERP profile, that was manifested in the ERP over the occipital surface. This posits a mechanism for the 'selection negativity', a scalp ERP effect noted under similar experimental conditions in human subjects.

Animals↗
PubMed2003

Comparative gene prediction in human and mouse.

The completion of the sequencing of the mouse genome promises to help predict human genes with greater accuracy. While current ab initio gene prediction programs are remarkably sensitive (i.e., they predict at least a fragment of most genes), their specificity is often low, predicting a large number of false-positive genes in the human genome. Sequence conservation at the protein level with the mouse genome can help eliminate some of those false positives. Here we describe SGP2, a gene prediction program that combines ab initio gene prediction with TBLASTX searches between two genome sequences to provide both sensitive and specific gene predictions. The accuracy of SGP2 when used to predict genes by comparing the human and mouse genomes is assessed on a number of data sets, including single-gene data sets, the highly curated human chromosome 22 predictions, and entire genome predictions from ENSEMBL. Results indicate that SGP2 outperforms purely ab initio gene prediction methods. Results also indicate that SGP2 works about as well with 3x shotgun data as it does with fully assembled genomes. SGP2 provides a high enough specificity that its predictions can be experimentally verified at a reasonable cost. SGP2 was used to generate a complete set of gene predictions on both the human and mouse by comparing the genomes of these two species. Our results suggest that another few thousand human and mouse genes currently not in ENSEMBL are worth verifying experimentally.

Animals↗
PubMed1992

Assembly of polypeptide and protein backbone conformations from low energy ensembles of short fragments: development of strategies and construction of models for myoglobin, lysozyme, and thymosin beta 4.

Recently we developed methods for the construction of knowledge-based mean fields from a data base of known protein structures. As shown previously, this approach can be used to calculate ensembles of probable conformations for short fragments of polypeptide chains. Here we develop procedures for the assembly of short fragments to complete three-dimensional models of polypeptide chains. The amino acid sequence of a given protein is decomposed into all possible overlapping fragments of a given length, and an ensemble of probable conformations is calculated for each fragment. The fragments are assembled to a complete model by choosing appropriate conformations from the individual ensembles and by averaging over equivalent angles. Finally a consistent model is obtained by rebuilding the conformation from the average angles. From the average angles the local variability of the structure can be calculated, which is a useful criterion for the reliability of the model. The procedure is applied to the calculation of the local backbone conformations of myoglobin and lysozyme whose structures have been solved by X-ray analysis and thymosin beta 4, a polypeptide of 43 amino acid residues whose structure was recently investigated by NMR spectroscopy. We demonstrate that substantial fractions of the calculated local backbone conformations are similar to the experimentally determined structures.

Amino Acid Sequence↗
PubMed2004

The UniMarker (UM) method for synteny mapping of large genomes.

MOTIVATION: Synteny mapping, or detecting regions that are orthologous between two genomes, is a key step in studies of comparative genomics. For completely sequenced genomes, this is increasingly accomplished by whole-genome sequence alignment. However, such methods are computationally expensive, especially for large genomes, and require rather complicated post-processing procedures to filter out non-orthologous sequence matches. RESULTS: We have developed a novel method that does not require sequence alignment for synteny mapping of two large genomes, such as the human and mouse. In this method, the occurrence spectra of genome-wide unique 16mer sequences present in both the human and mouse genome are used to directly detect orthologous genomic segments. Being sequence alignment-free, the method is very fast and able to map the two mammalian genomes in one day of computing time on a single Pentium IV personal computer. The resulting human-mouse synteny map was shown to be in excellent agreement with those produced by the Mouse Genome Sequencing Consortium (MGSC) and by the Ensembl team; furthermore, the syntenic relationship of segments found only by our method was supported by BLASTZ sequence alignment.

Algorithms↗
PubMed2001

Constructing, verifying, and dissecting the folding transition state of chymotrypsin inhibitor 2 with all-atom simulations.

Experimentally, protein engineering and phi-value analysis is the method of choice to characterize the structure in folding transition state ensemble (TSE) of any protein. Combining experimental phi values and computer simulations has led to a deeper understanding of how proteins fold. In this report, we construct the TSE of chymotrypsin inhibitor 2 from published phi values. Importantly, we verify, by means of multiple independent simulations, that the conformations in the TSE have a probability of approximately 0.5 to reach the native state rapidly, so the TSE consists of true transition states. This finding validates the use of transition state theory underlying all phi-value analyses. Also, we present a method to dissect and study the TSE by generating conformations that have a disrupted alpha-helix (alpha-disrupted states) or disordered beta-strands 3 and 4 (beta-disrupted states). Surprisingly, the alpha-disrupted states have a stronger tendency to fold than the beta-disrupted states, despite the higher phi values for the alpha-helix in the TSE. We give a plausible explanation for this result and discuss its implications on protein folding and design. Our study shows that, by using both experiments and computer simulations, we can gain many insights into protein folding.

Peptides↗
PubMed1998

Ensembles of radial basis function networks for spectroscopic detection of cervical precancer.

The mortality related to cervical cancer can be substantially reduced through early detection and treatment. However, current detection techniques, such as Pap smear and colposcopy, fail to achieve a concurrently high sensitivity and specificity. In vivo fluorescence spectroscopy is a technique which quickly, noninvasively and quantitatively probes the biochemical and morphological changes that occur in precancerous tissue. A multivariate statistical algorithm was used to extract clinically useful information from tissue spectra acquired from 361 cervical sites from 95 patients at 337-, 380-, and 460-nm excitation wavelengths. The multivariate statistical analysis was also employed to reduce the number of fluorescence excitation-emission wavelength pairs required to discriminate healthy tissue samples from precancerous tissue samples. The use of connectionist methods such as multilayered perceptrons, radial basis function (RBF) networks, and ensembles of such networks was investigated. RBF ensemble algorithms based on fluorescence spectra potentially provide automated and near real-time implementation of precancer detection in the hands of nonexperts. The results are more reliable, direct, and accurate than those achieved by either human experts or multivariate statistical algorithms.

Algorithms↗
PubMed2003

Lower receiving frequencies than transmitting frequencies could yield improved results for contrast imaging: an in vivo study in closed chest canines.

BACKGROUND: Ultrasonic imaging methods of receiving at higher frequencies, which are multiples of the transmitting frequencies (harmonic imaging), are well established as a means of improving myocardial visualization in association with intravenous contrast administration. This exploratory study examined the effect of using receive frequencies that were lower than the transmit frequencies while imaging closed chest dogs with an Ensemble wideband, phase inversion contrast program on a modified Siemens Elegra scanner. METHODS: Intravenous bolus injections of 0.75 mL Definity and 1 mL QW7437 were administered to six anesthetized dogs. Intermittent imaging for contrast visualization was performed using either a broadband array, transmitting at 1.4 MHz and receiving at 2.6-3.2 MHz or a broadband 4-7.5 MHz transducer transmitting at 6.0 MHz and receiving at 4.2-4.5 MHz. Contrast enhancement was measured by videodensitometry, sampling mid-cavity and within the myocardium before and after injection. The changes in videodensity from control to after injection were calculated for each method. RESULTS: There was no significant difference in the change in intracavity videodensity between the two imaging strategies although there was near full intracavity saturation in all cases. However, the change in myocardial density was significantly greater for both contrast agents when using receiving frequencies lower than transmitting frequencies (P = 0.02 and 0.03). The difference in duration of the myocardial blush did not reach statistical significance but it tended to persist for longer with the lower receiving frequencies. CONCLUSION: Delivering sound energy at a slightly higher frequency and receiving at lower than the transmit frequency may be an advantageous method of enhancing myocardial perfusion signals during intravenous contrast echocardiography.

Animals↗
PubMedDate not supplied

Thermal characteristics of clothing ensembles for use in heat stress analysis.

The Heat Stress Index was an early model for the assessment of heat stress. The International Organization for Standardization (ISO) standard for required sweat rate is the current generation of heat balance methods for occupational heat stress. The method assumes cotton clothing and works adequately for cotton/polyester blends. To extend the usefulness of the model, the thermal characteristics of a variety of commercially available and prototype protective clothing ensembles have been determined for application in the ISO method. The fundamental principle for assessing thermal characteristics of work clothing is establishing the critical environmental conditions in which test subjects were just able to maintain thermal equilibrium. Critical conditions were found for warm, humid conditions; hot, dry conditions; intermediate conditions of temperature and humidity; and/or moderate conditions in which metabolic rate was increased to a limiting thermal load. Typically, five subjects at each condition for each ensemble were used. Metabolic rate, average skin temperature, and the environmental conditions (air temperature and vapor pressure) were noted at the critical conditions, and the total insulation was estimated for each ensemble. From these values, the total evaporative resistance, the clothing factor for dry heat exchange (CFcl), and the clothing factor for evaporative cooling (CFpcl) were determined. When compared with reports of others on thermal characteristics the results agreed when pumping factors and clothing wetness were considered. The result was higher than expected values for CFcl and lower values for CFpcl.

Adult↗
PubMed2005

A skewed-momenta method to efficiently generate conformational-transition trajectories.

We present a novel computational method, the skewed-momenta method (Skew'M), which applies a bias to the Maxwell distribution of initial momenta used to generate ensembles of trajectories. As a result, conformational transitions are accentuated and kinetic properties are calculated more effectively. The connection to the related puddle jumping method is discussed. A reweighting scheme permits the exact calculation of kinetic properties. Applications are presented for the rapid calculation of rate constants for molecular isomerization, and for the efficient reconstruction of free-energy profiles using a straightforward modification of the Jarzynski identity.

Chemistry, Physical↗
PubMed2004

Free energy surfaces of beta-hairpin and alpha-helical peptides generated by replica exchange molecular dynamics with the AGBNP implicit solvent model.

We have studied the potential of mean force of two peptides, one known to adopt a beta-hairpin and the other an alpha-helical conformation in solution. These peptides are, respectively, residues 41-56 of the C-terminus (GEWTYDDATKTFTVTE) of the B1 domain of protein G and the 13 residue C-peptide (KETAAAKFERQHM) of ribonuclease A. Extensive canonical ensemble sampling has been performed using a parallel replica exchange method. The effective potential employed in this work consists of the OPLS all-atom force field (OPLS-AA) and an analytical generalized Born (AGB) implicit solvent model including a novel nonpolar solvation free energy estimator (NP). An additional dielectric screening parameter has been incorporated into the AGBNP model. In the case of the beta-hairpin, the nonpolar solvation free energy estimator provides the necessary effective interactions for the collapse of the hydrophobic core (W43, Y45, F52, and V54), which the more commonly used surface-area-dependent nonpolar model does not provide. For both the beta-hairpin and the alpha-helix, increased dielectric screening reduces the stability of incorrectly formed salt bridges, which tend to disrupt the formation of the hairpin and helix, respectively. The fraction of beta-hairpin and alpha-helix content we obtained using the AGBNP model agrees well with experimental results. The thermodynamic stability of the beta-hairpin from protein G and the alpha-helical C-peptide from ribonuclease A as modeled with the OPLS-AA/AGBNP effective potential reflects the balance between the nonpolar effective potential terms, which drive compaction, and the polar and hydrogen bonding terms, which promote secondary structure formation.

Amino Acid Sequence↗
PubMed1990

Electrostatic contributions to the binding of Ca2+ in calbindin mutants. A Monte Carlo study.

Monte Carlo simulation is used to calculate the free energy of binding of calcium ions to the native and several mutant forms of bovine calbindin D(9K) in salt solution. The simulations are performed in the canonical ensemble wherein free energies are calculated with a modified Widom method. The protein is modelled as a set of fixed hard spheres of fractional or unit charge with the surrounding solution as a dielectric continuum containing counterions and added salt particles. The interior of the protein is assumed to have the same dielectric permittivity as the solvent, which turns out to be an excellent approximation. Indeed, this simple model is able to predict accurately experimentally measured shifts in the calcium binding constants of up to five orders of magnitude, due to mutations and added salt.

Journal Article↗
PubMed2007

Diagnosing scrapie in sheep: a classification experiment.

Scrapie is a neuro-degenerative disease in small ruminants. A data set of 3113 records of sheep reported to the Scrapie Notifications Database in Great Britain has been studied. Clinical signs were recorded as present/absent in each animal by veterinary officials (VO) and a post-mortem diagnosis was made. In an attempt to detect healthy animals within the set of suspects using only the clinical signs, 18 classification methods were applied ranging from simple linear classifiers to classifier ensembles such as Bagging, AdaBoost and Random Forests. The results suggest that the clinical classification by the VO was adequate as no further differentiation within the set of suspects was feasible.

Animals↗