Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Parallel Algorithms”

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 883 records · Page 49Linked to original sources
PubMed2002

[A perturbation solution of pulsatile Casson flow in the parallel-plate flow chamber].

The parallel-plate flow chamber (PPFC) with rectangular shape, of which the height is far smaller than its own length and width, is one of the main apparatus for in vitro studies of the mechanical behavior of the cells. Considering that the Reynolds numbers of flows in the usually used PPFCs are small, a perturbation solution of laminar pulsatile Casson fluid in the PPFC is presented using Reynolds number as perturbation parameter. Furthermore, the velocities and shear stresses in the PPFC are given. The numerical results demonstrate that under the same pressure gradient the shear stresses are almost identical between Casson fluid and Newtonian fluid, whereas under the same flux the shear stresses are obviously different between Casson fluid and Newtonian fluid. The results in this paper provide a theoretical way to determine the shear stresses in the parallel-plate flow chamber under pulsatile Casson fluid.

Algorithms↗
PubMed2005

Benchmarking analytical calculations of proton doses in heterogeneous matter.

A proton dose computational algorithm, performing an analytical superposition of infinitely narrow proton beamlets (ASPB) is introduced. The algorithm uses the standard pencil beam technique of laterally distributing the central axis broad beam doses according to the Moliere scattering theory extended to slablike varying density media. The purpose of this study was to determine the accuracy of our computational tool by comparing it with experimental and Monte Carlo (MC) simulation data as benchmarks. In the tests, parallel wide beams of protons were scattered in water phantoms containing embedded air and bone materials with simple geometrical forms and spatial dimensions of a few centimeters. For homogeneous water and bone phantoms, the proton doses we calculated with the ASPB algorithm were found very comparable to experimental and MC data. For layered bone slab inhomogeneity in water, the comparison between our analytical calculation and the MC simulation showed reasonable agreement, even when the inhomogeneity was placed at the Bragg peak depth. There also was reasonable agreement for the parallelepiped bone block inhomogeneity placed at various depths, except for cases in which the bone was located in the region of the Bragg peak, when discrepancies were as large as more than 10%. When the inhomogeneity was in the form of abutting air-bone slabs, discrepancies of as much as 8% occurred in the lateral dose profiles on the air cavity side of the phantom. Additionally, the analytical depth-dose calculations disagreed with the MC calculations within 3% of the Bragg peak dose, at the entry and midway depths in the phantom. The distal depth-dose 20%-80% fall-off widths and ranges calculated with our algorithm and the MC simulation were generally within 0.1 cm of agreement. The analytical lateral-dose profile calculations showed smaller (by less than 0.1 cm) 20%-80% penumbra widths and shorter fall-off tails than did those calculated by the MC simulations. Overall, this work validates the usefulness of our ASPB algorithm as a reasonably fast and accurate tool for quality assurance in planning wide beam proton therapy treatment of clinical sites either composed of homogeneous materials or containing laterally extended inhomogeneities that are comparable in density and located away from the Bragg peak depths.

Air↗
PubMed2005

Monte Carlo- versus pencil-beam-/collapsed-cone-dose calculation in a heterogeneous multi-layer phantom.

The aim of this work was to evaluate the accuracy of dose predicted in heterogeneous media by a pencil beam (PB), a collapsed cone (CC) and a Monte Carlo (MC) algorithm. For this purpose, a simple multi-layer phantom composed of Styrofoam and white polystyrene was irradiated with 10 x 10 cm2 as well as 20 x 20 cm2 open 6 MV photon fields. The beam axis was aligned parallel to the layers and various field offsets were applied. Thereby, the amount of lateral scatter was controlled. Dose measurements were performed with an ionization chamber positioned both in the central layer of white polystyrene and the adjacent layers of Styrofoam. It was found that, in white polystyrene, both MC and CC calculations agreed satisfactorily with the measurements whereas the PB algorithm calculated 12% higher doses on average. By studying off-axis dose profiles the observed differences in the calculation results increased dramatically for the three algorithms. In the regions of low density CC calculated 10% (8%) lower doses for the 10 x 10 cm2 (20 x 20 cm2) fields than MC. The MC data on the other hand agreed well with the measurements, presuming that proper replacement correction for the ionization chamber embedded in Styrofoam was performed. PB results evidently did not account for the scattering geometry and were therefore not really comparable. Our investigations showed that the PB algorithm generates very large errors for the dose in the vicinity of interfaces and within low-density regions. We also found that for the used CC algorithm large deviations for the absolute dose (dose/monitor unit) occur in regions of electronic disequilibrium. The performance might be improved by better adapted parameters. Therefore, we recommend a careful investigation of the accuracy for dose calculations in heterogeneous media for each beam data set and algorithm.

Algorithms↗
PubMedDate not supplied

AUDENS: a tool for automated peptide de novo sequencing.

We present AUDENS, a new platform-independent open source tool for automated de novo sequencing of peptides from MS/MS data. We implemented a dynamic programming algorithm and combined it with a flexible preprocessing module which is designed to distinguish between signal and other peaks. By applying a user-defined set of heuristics, AUDENS screens through the spectrum and assigns high relevance values to putative signal peaks. The algorithm constructs a sequence path through the MS/MS spectrum using the peak relevances to score each suggested sequence path, i.e., the corresponding amino acid sequence. At present, we consider AUDENS a prototype that unfolds its biggest potential if used in parallel with other de novo sequencing tools. AUDENS is available open source and can be downloaded with further documentation at http://www.ti.inf.ethz.ch/pw/software/audens/ .

Algorithms↗
PubMed2000

Distributed computing for membrane-based modeling of action potential propagation.

Action potential propagation simulations with physiologic membrane currents and macroscopic tissue dimensions are computationally expensive. We, therefore, analyzed distributed computing schemes to reduce execution time in workstation clusters by parallelizing solutions with message passing. Four schemes were considered in two-dimensional monodomain simulations with the Beeler-Reuter membrane equations. Parallel speedups measured with each scheme were compared to theoretical speedups, recognizing the relationship between speedup and code portions that executed serially. A data decomposition scheme based on total ionic current provided the best performance. Analysis of communication latencies in that scheme led to a load-balancing algorithm in which measured speedups at 89 +/- 2% and 75 +/- 8% of theoretical speedups were achieved in homogeneous and heterogeneous clusters of workstations. Speedups in this scheme with the Luo-Rudy dynamic membrane equations exceeded 3.0 with eight distributed workstations. Cluster speedups were comparable to those measured during parallel execution on a shared memory machine.

Action Potentials↗
PubMed2005

Calculating the 2D motion of lumbar vertebrae using splines.

In this study we investigate the use of splines and the ICP method [Besl, P., McKay, N., 1992. A method for registration of 3d shapes. IEEE Transactions on Pattern Analysis and Machine Intelligence 14, 239-256.] for calculating the transformation parameters for a rigid body undergoing planar motion parallel to the image plane. We demonstrate the efficacy of the method by estimating the finite centre of rotation and angle of rotation from lateral flexion/extension radiographs of the lumbar spine. In an in vitro error study, the method displayed an average error of rotation of 0.44 +/- 0.45 degrees, and an average error in FCR calculation of 7.6 +/- 8.5 mm. The method was shown to be superior to that of Crisco et al. [Two-dimensional rigid-body kinematics using image contour registration. Journal of Biomechanics 28(1), 119-124.] and Brinckmann et al. [Quantification of overload injuries of the thoracolumbar spine in persons exposed to heavy physical exertions or vibration at the workplace: Part i - the shape of vertebrae and intervertebral discs - study of a yound, healthy population and a middle-aged control group. Clinical Biomechanics Supplement 1, S5-S83.] for the tests performed here. In general, we believe the use of splines to represent planar shapes to be superior to using digitised curves or landmarks for several reasons. First, with appropriate software, splines require less effort to define and are a compact representation, with most vertebra outlines using less than 30 control points. Second, splines are inherently sub-pixel representations of curves, even if the control points are limited to pixel resolutions. Third, there is a well-defined method (the ICP algorithm) for registering shapes represented as splines. Finally, like digitised curves, splines are able to represent a large class of shapes with little effort, but reduce potential segmentation errors from two dimensions (parallel and perpendicular to the image gradient) to just one (parallel to the image gradient). We have developed an application for performing all the necessary computations which can be downloaded from http://www.claritysmart.com.

Algorithms↗
PubMed2006

Decomposed direct matrix inversion for fast non-cartesian SENSE reconstructions.

A new k-space direct matrix inversion (DMI) method is proposed here to accelerate non-Cartesian SENSE reconstructions. In this method a global k-space matrix equation is established on basic MRI principles, and the inverse of the global encoding matrix is found from a set of local matrix equations by taking advantage of the small extension of k-space coil maps. The DMI algorithm's efficiency is achieved by reloading the precalculated global inverse when the coil maps and trajectories remain unchanged, such as in dynamic studies. Phantom and human subject experiments were performed on a 1.5T scanner with a standard four-channel phased-array cardiac coil. Interleaved spiral trajectories were used to collect fully sampled and undersampled 3D raw data. The equivalence of the global k-space matrix equation to its image-space version, was verified via conjugate gradient (CG) iterative algorithms on a 2x undersampled phantom and numerical-model data sets. When applied to the 2x undersampled phantom and human-subject raw data, the decomposed DMI method produced images with small errors (< or = 3.9%) relative to the reference images obtained from the fully-sampled data, at a rate of 2 s per slice (excluding 4 min for precalculating the global inverse at an image size of 256 x 256). The DMI method may be useful for noise evaluations in parallel coil designs, dynamic MRI, and 3D sodium MRI with fixed coils and trajectories.

Algorithms↗
PubMed2002

A novel quantitative dual-isotope method for simultaneous ventilation and perfusion lung SPET.

A quantitative dual-isotope single-photon emission tomography (SPET) technique for the assessment of lung ventilation (V) and perfusion (Q) using, respectively, technetium-99m labelled Technegas (140 keV) and indium-113m labelled macro-aggregated albumin (392 keV), is presented, validated and clinically tested in a healthy volunteer. In order to assess V, Q and V/Q distributions in quantitative terms, algorithms which correct for down scattering, photon scattering and attenuation, as well as an organ outline algorithm, were implemented. Scatter and down-scatter correction were made in the spatial domain by pixel-wise image subtraction of projection-dependent global scattering factors obtained from the energy domain. The attenuation correction was based on an iterative projection/back-projection method. All studies were made on a three-headed SPET system (Trionix) with medium-energy parallel-hole collimators. The set of input data for quantification was based on SPET acquisition of emission data in four separate energy windows, the associated cumulative energy spectra and transmission data. The attenuation correction routine as well as the edge detection algorithm utilized data from (99m)Tc transmission tomography. Attenuation data for (113m)In were obtained by linear scaling of the (99m)Tc attenuation maps. The correction algorithms were experimentally validated with a stack phantom system and applied on a healthy volunteer. The mean difference between the corrected SPET data of the dense stack lung phantom and those obtained from the corresponding scatter- and attenuation-"free" version was only 1.9% for (99m)Tc and 0.9% for (113m)In. The estimated fractional V/Q distribution in the 3-D lung phantom volume had its peak at V/Q=1, with a width (FWHM) of 0.31 due to noise, particularly in the (113m)In images, and to partial volume effects. For a healthy volunteer, the corresponding values were 0.9 and 0.35, respectively. This method allows accurate assessment of radionuclide distribution on a regional basis. For basic lung physiology and clinical practice, the method allows assessment of the global frequency functions of the V, Q and V/Q distributions.

Algorithms↗
PubMed2005

Accurate technique for complete geometric calibration of cone-beam computed tomography systems.

Cone-beam computed tomography systems have been developed to provide in situ imaging for the purpose of guiding radiation therapy. Clinical systems have been constructed using this approach, a clinical linear accelerator (Elekta Synergy RP) and an iso-centric C-arm. Geometric calibration involves the estimation of a set of parameters that describes the geometry of such systems, and is essential for accurate image reconstruction. We have developed a general analytic algorithm and corresponding calibration phantom for estimating these geometric parameters in cone-beam computed tomography (CT) systems. The performance of the calibration algorithm is evaluated and its application is discussed. The algorithm makes use of a calibration phantom to estimate the geometric parameters of the system. The phantom consists of 24 steel ball bearings (BBs) in a known geometry. Twelve BBs are spaced evenly at 30 deg in two plane-parallel circles separated by a given distance along the tube axis. The detector (e.g., a flat panel detector) is assumed to have no spatial distortion. The method estimates geometric parameters including the position of the x-ray source, position, and rotation of the detector, and gantry angle, and can describe complex source-detector trajectories. The accuracy and sensitivity of the calibration algorithm was analyzed. The calibration algorithm estimates geometric parameters in a high level of accuracy such that the quality of CT reconstruction is not degraded by the error of estimation. Sensitivity analysis shows uncertainty of 0.01 degrees (around beam direction) to 0.3 degrees (normal to the beam direction) in rotation, and 0.2 mm (orthogonal to the beam direction) to 4.9 mm (beam direction) in position for the medical linear accelerator geometry. Experimental measurements using a laboratory bench Cone-beam CT system of known geometry demonstrate the sensitivity of the method in detecting small changes in the imaging geometry with an uncertainty of 0.1 mm in transverse and vertical (perpendicular to the beam direction) and 1.0 mm in the longitudinal (beam axis) directions. The calibration algorithm was compared to a previously reported method, which uses one ball bearing at the isocenter of the system, to investigate the impact of more precise calibration on the image quality of cone-beam CT reconstruction. A thin steel wire located inside the calibration phantom was imaged on the conebeam CT lab bench with and without perturbations in source and detector position during the scan. The described calibration method improved the quality of the image and the geometric accuracy of the object reconstructed, improving the full width at half maximum of the wire by 27.5% and increasing contrast of the wire by 52.8%. The proposed method is not limited to the geometric calibration of cone-beam CT systems but can be used for many other systems, which consist of one or more point sources and area detectors such as calibration of megavoltage (MV) treatment system (focal spot movement during the beam delivery, MV source trajectory versus gantry angle, the axis of collimator rotation, and couch motion), cross calibration between Kilovolt imaging and MV treatment system, and cross calibration between multiple imaging systems. Using the complete information of the system geometry, it was demonstrated that high image quality in CT reconstructions is possible even in systems with large geometric nonidealities.

Algorithms↗
PubMed1991

Parallel simulated annealing for emission tomography.

A method for implementing simulated annealing in parallel to speed up the execution of emission tomography (ET) image reconstruction is presented. A high degree of parallelism can be attained by using a parallel-acceptance partitioning strategy, in which perturbations to subsets of the estimate are evaluated in parallel. However because the point spread function in ET imaging systems is globally dependent, processors cannot update the current estimate independently. Consequently, processors must be synchronized each time a perturbation is accepted to avoid introducing error. This can produce excessive communications overhead, especially when the acceptance rate is high. In this paper an energy function is constructed to reduce the synchronization requirements by using a reformulation of the log-likelihood function from the expectation maximization (EM) algorithm. The approach is to change the global dependence in the energy function from the current estimate to the estimate generated during the last iteration. The synchronization requirements for guaranteed convergence are then significantly reduced from once per acceptance to once per iteration. This parallel implementation on 54 Inmos T800 transputers connected in a ring topology resulted in execution times that were almost 50 times faster than on a VAX 8600.

Algorithms↗
PubMed2004

Exact reconstruction in 2D dynamic CT: compensation of time-dependent affine deformations.

This work is dedicated to the reduction of reconstruction artefacts due to motion occurring during the acquisition of computerized tomographic projections. This problem has to be solved when imaging moving organs such as the lungs or the heart. The proposed method belongs to the class of motion compensation algorithms, where the model of motion is included in the reconstruction formula. We address two fundamental questions. First what conditions on the deformation are required for the reconstruction of the object from projections acquired sequentially during the deformation, and second how do we reconstruct the object from those projections. Here we answer these questions in the particular case of 2D general time-dependent affine deformations, assuming the motion parameters are known. We treat the problem of admissibility conditions on the deformation in the parallel-beam and fan-beam cases. Then we propose exact reconstruction methods based on rebinning or sequential FBP formulae for each of these geometries and present reconstructed images obtained with the fan-beam algorithm on simulated data.

Algorithms↗
PubMed2005

Algorithm for neuropathic pain treatment: an evidence based proposal.

New studies of the treatment of neuropathic pain have increased the need for an updated review of randomized, double-blind, placebo-controlled trials to support an evidence based algorithm to treat neuropathic pain conditions. Available studies were identified using a MEDLINE and EMBASE search. One hundred and five studies were included. Numbers needed to treat (NNT) and numbers needed to harm (NNH) were used to compare efficacy and safety of the treatments in different neuropathic pain syndromes. The quality of each trial was assessed. Tricyclic antidepressants and the anticonvulsants gabapentin and pregabalin were the most frequently studied drug classes. In peripheral neuropathic pain, the lowest NNT was for tricyclic antidepressants, followed by opioids and the anticonvulsants gabapentin and pregabalin. For central neuropathic pain there is limited data. NNT and NNH are currently the best way to assess relative efficacy and safety, but the need for dichotomous data, which may have to be estimated retrospectively for old trials, and the methodological complexity of pooling data from small cross-over and large parallel group trials, remain as limitations.

Algorithms↗
PubMed2004

Parallel processing across neural systems: implications for a multiple memory system hypothesis.

A common conceptualization of the organization of memory systems in brain is that different types of memory are mediated by distinct neural systems. Strong support for this view comes from studies that show double (or triple) dissociations between spatial, response, and emotional memories following selective lesions of hippocampus, striatum, and the amygdala. Here, we examine the extent to which hippocampal and striatal neural activity patterns support the multiple memory systems view. A comparison is made between hippocampal and striatal neural correlates with behavior during asymptotic performance of spatial and response maze tasks. Location- (or place), movement, and reward-specific firing patterns were found in both structures regardless of the task demands. Many, but not all, place fields of hippocampal and striatal neurons were similarly affected by changes in the visual and reward context regardless of the cognitive demands. Also, many, but not all, hippocampal and striatal movement-sensitive neurons showed significant changes in their behavioral correlates after a change in visual context, irrespective of cognitive strategy. Similar partial reorganization was observed following manipulations of the reward condition for cells recorded from both structures, again regardless of task. Assuming that representations that persist across context changes reflect learned information, we make the following conclusions. First, the consistent pattern of partial reorganization supports a view that the analysis of spatial, response, and reinforcement information is accomplished via an error-driven, or match-mismatch, algorithm across neural systems. Second, task-relevant processing occurs continuously within hippocampus and striatum regardless of the cognitive demands of the task. Third, given the high degree of parallel processing across allegedly different memory systems, we propose that different neural systems may effectively compete for control of a behavioral expression system. The strength of the influence of any one neural system on behavioral output is likely modulated by factors such as motivation, experience, or hormone status.

Animals↗
PubMed1996

A self-organizing cognitive network of antibody repertoire development.

A self-organizing cognitive network is mapped here onto the Id network model. The weight-vectors in this network represent some important topographical and biophysical parameters in the antibody-antigen affinity landscape. The Kohonen layers in the network correspond to affinity clones and the involved algorithm simulates the operations of clonal selection, hypermutation, differentiation, diversity, and affinity maturation. Two significant features of this model are: (i) a computationally feasible and biophysically informative representation of the para/epitopes, and (ii) the ability to perform simultaneous (parallel) and associative computations in a multidimensional shape-space. Computational experiments with real data have shown cognitive properties of this network. The results also indicate scope in quantitative characterization of the metadynamics of the above operations/weights in the adaptive development of the antibody repertoire.

Algorithms↗
PubMed1996

Investigating shape-from-shading illusions using solid objects.

Recent growth in the shape-from-shading psychophysics literature has been paralleled by an increasing availability of computer graphics hardware and software, to the extent that most psychophysical studies in this area now employ computer lighting algorithms. The most widely used of these algorithms in shape-from-shading psychophysics is the Phong lighting model. This model, and other shading models of its genre, produce readily interpretable images of three-dimensional scenes. However, such algorithms are only approximations of how light interacts with real objects in the natural environment. Nevertheless, the results from psychophysical experiments using these techniques have been used to infer the processes underlying the perception of shape-from-shading in natural environments. It is important to establish whether this substitution is ever valid. We report a series of experiments investigating whether two recently reported illusions seen in computer-generated, Phong shaded images occur for solid objects under real illuminants. The two illusions investigated are three-dimensional curvature contrast and the illuminant-position effect on perceived curvature. We show that both effects do occur for solid objects, and that the magnitude of these effects are equivalent regardless of whether subjects are presented with ray traced or solid objects.

Algorithms↗
PubMedDate not supplied

A distributed and interactive three-dimensional medical image system.

Three-dimensional (3D) arrays of digital data representing spatial volumes arise in many scientific applications, such as computed tomography (CT) and magnetic resonance imaging (MRI) created by imaging a series of cross sections of human bodies in medical applications. In this article, a software system architecture, called DISCOVER (a Distributed Interactive Scientific COmputing and Visualization EnviRonment), which can take advantage of the power of parallel processing, is proposed and implemented for interactive visualization and manipulation of the 3D digital data. The surface-rendering and the volume-rendering algorithms are implemented. The same software program can be executed on several different hardware platforms. We also propose a new rendering algorithm, called volume-surface rendering, for medical applications. The algorithm enables users to visualize the external and internal structures of medical objects simultaneously. The network version of the DISCOVER, as it stands today, is in practical use in the Hospital of National Cheng Kung University in Taiwan for real clinical applications.

Algorithms↗
PubMed2006

Recognition of external object features in gas media using ultrasound transmission tomography.

The paper presents and analyzes a new way of recognizing external object features (shape, size, location) in gas media using ultrasound transmission tomography (UTT) with parallel-ray-projection scanning geometry. The concept of UTT in a gas medium is close to classical tomography, however because of the nature of the measurement environment, visualizing the internal structure of solid objects is difficult; whereas it is possible to image their external features: their shape, size and spatial location. The paper presents the results of examining the shape, size and location of different objects in the air in the form of tomographic images, obtained in parallel-ray-projection geometry, using a specially elaborated research setup for UTT. Applying parallel-ray-projection geometry enabled us to investigate the influence of scanning resolution on image quality. In order to test the operation of the elaborated algorithm of tomographic image reconstruction on the basis of correct measurement data, special software was written for simulating the binary matrix of the measurements for a set of a dozen or so solid objects of different shapes and a few simulations were performed.

Journal Article↗
PubMed1997

Validation of the central-ray approximation for attenuated depth-dependent convolution in quantitative SPECT reconstruction.

In order to model photon attenuation and detector resolution variation as a depth-dependent convolution for efficient reconstruction of quantitative SPECT, a central-ray approximation is necessary. This work investigates the impact of the approximation upon reconstruction accuracy and computational efficiency. A patient chest CT image was acquired and converted into an object-specific attenuation map. From a segmentation of the map, an emission thorax phantom was constructed with a cardiac insert. To generate a system-specific resolution-variant kernal, a point source was measured at several depths from the surface of a low-energy, high-resolution, parallel-hole collimator of a SPECT system. Projections of parallel-beam geometry were simulated from the phantom, the map, and the kernel on an elliptical orbit. Reconstruction was performed by the ML-EM algorithm with and without the central-ray approximation. The approximation cuts down dramatically (more than 100 fold) the computing time with a negligible loss (less than 1%) of reconstruction accuracy.

Algorithms↗