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 523 records · Page 29Linked to original sources
PubMed1999

Assessment of the uncertainties in dose delivery of a commercial system for linac-based stereotactic radiosurgery.

PURPOSE: Linac-based stereotactic radiosurgery (SRS) was introduced in our department in 1992, and since then, more than 200 patients have been treated with this method. An in-house-developed algorithm for target localization and dose calculation has recently been replaced with a commercially available system. In this study, both systems have been compared, and positional accuracy, as well as dose calculation, have been verified experimentally. METHODS AND MATERIALS: The in-house-developed software for target localization and dose calculation is an extension to George Sherouse's GRATIS(R) software for radiotherapy treatment planning, and has been replaced by a commercial (BrainSCAN version 3.1; BrainLAB, Germany) treatment planning system (TPS) for SRS. The positional accuracy for the entire SRS procedure (from image acquisition to treatment) has been investigated by treatment of simulated targets in the form of 0.2-cm lead beads inserted into an anthropomorphic phantom. Both dose calculation algorithms have been verified against manual calculations (based on basic beam data and CT data from phantom and patients), and measurements with the anthropomorphic phantom applying ionization chamber, thermoluminescent detectors, and radiographic film. This analysis has been performed on a variety of experimental situations, starting with static beams and simple one-arc treatments, to more complex and clinical relevant applications. Finally, 11 patients have been evaluated with both TPS in parallel for comparison and continuity of clinical experience. RESULTS: Phantom studies evaluating the entire SRS procedure have shown that a target, localized by CT, can be irradiated with a positional accuracy of 0.08 cm in any direction with 95% confidence. Neglecting the influence of dose perturbation when the beam passes through bone tissue or air cavities, the calculated dose values obtained from both TPSs agreed within 1% (SD 1%) for phantom and patient studies. The application of a one-dimensional path length correction for tissue heterogeneity influences the treatment prescription 4% on average (SD 1%), which is in compliance with theoretical predictions. The phantom measurements confirmed the predicted dose at isocenter within uncertainty for the different treatment schedules in this study. CONCLUSION: The full SRS procedure applied to an anthropomorphic phantom has been used as a comprehensive method to assess the uncertainties involved in dose delivery and target positioning. The results obtained with both TPSs are in agreement with AAPM Report 54, TG 42 and clinical continuity is assured. However, the use of a one-dimensional path length correction will result in an increase of 4% in dose prescription, which is slightly more than that predicted in the literature.

Algorithms↗
PubMed1993

Magnetic field distribution in models of trabecular bone.

A magnetostatic model consisting of a tetragonal lattice of struts of diamagnetic material, mimicking vertebral trabecular bone, was developed. The model allows estimation of the magnetic field histogram within the lattice's unit cell as a function of geometric parameters. The field was computed analytically from the induced magnetic surface charge density on the faces of the struts. The contribution from the induced magnetic field to the effective transverse relaxation rate, R2', was obtained as the mean decay rate of the Fourier transformed histograms, for both fixed and randomly oriented lattices. The model predicts the field distribution to increase with both strut thickness and density, paralleling material density. Finally, significant changes in R2' are predicted at constant material density, in that the field distribution widens with simultaneously increasing strut number density and decreasing strut thickness.

Algorithms↗
PubMed1997

New finite difference formulations for general inhomogeneous anisotropic bioelectric problems.

Due to its low computational complexity, finite difference modeling offers a viable tool for studying bioelectric problems, allowing the field behavior to be observed easily as different system parameters are varied. Previous finite difference formulations, however, have been limited mainly to systems in which the conductivity is orthotropic, i.e., a strictly diagonal conductivity tensor. This in turn has limited the effectiveness of the finite difference, technique in modeling complex anatomies with arbitrarily anisotropic conductivities, e.g., detailed fiber structures of muscles where the fiber can lie in any arbitrary direction. In this paper, we present both two-dimensional and three-dimensional finite difference formulations that are valid for structures with an inhomogeneous and nondiagonal conductivity tensor. A data parallel computer, the connection machine CM-5, is used in the finite difference implementation to provide the computational power and memory for solving large problems. The finite difference grid is mapped effectively to the CM-5 by associating a group of nodes with one processor. Details on the new approach and its data parallel implementation are presented together with validation and computational performance results. In addition, an application of the new formulation in providing the potential distribution inside a canine torso during electrical defibrillation is demonstrated.

Algorithms↗
PubMed1992

Fast computation of genetic likelihoods on human pedigree data.

Gene mapping and genetic epidemiology require large-scale computation of likelihoods based on human pedigree data. Although computation of such likelihoods has become increasingly sophisticated, fast calculations are still impeded by complex pedigree structures, by models with many underlying loci and by missing observations on key family members. The current paper 'introduces' a new method of array factorization that substantially accelerates linkage calculations with large numbers of markers. This method is not limited to nuclear families or to families with complete phenotyping. Vectorization and parallelization are two general-purpose hardware techniques for accelerating computations. These techniques can assist in the rapid calculation of genetic likelihoods. We describe our experience using both of these methods with the existing program MENDEL. A vectorized version of MENDEL was run on an IBM 3090 supercomputer. A parallelized version of MENDEL was run on parallel machines of different architectures and on a network of workstations. Applying these revised versions of MENDEL to two challenging linkage problems yields substantial improvements in computational speed.

Algorithms↗
PubMed1998

Algorithmic approach to high-throughput molecular screening for alpha interferon-resistant genotypes in hepatitis C patients.

This study was designed to analyze the feasibility and validity of using Cleavase Fragment Length Polymorphism (CFLP) analysis as an alternative to DNA sequencing for high-throughput screening of hepatitis C virus (HCV) genotypes in a high-volume molecular pathology laboratory setting. By using a 244-bp amplicon from the 5' untranslated region of the HCV genome, 61 clinical samples received for HCV reverse transcription-PCR (RT-PCR) were genotyped by this method. The genotype frequencies assigned by the CFLP method were 44.3% for type 1a, 26.2% for 1b, 13.1% for type 2b, and 5% type 3a. The results obtained by nucleotide sequence analysis provided 100% concordance with those obtained by CFLP analysis at the major genotype level, with resolvable differences as to subtype designations for five samples. CFLP analysis-derived HCV genotype frequencies also concurred with the national estimates (N. N. Zein et al., Ann. Intern. Med. 125:634-639, 1996). Reanalysis of 42 of these samples in parallel in a different research laboratory reproduced the CFLP fingerprints for 100% of the samples. Similarly, the major subtype designations for 19 samples subjected to different incubation temperature-time conditions were also 100% reproducible. Comparative cost analysis for genotyping of HCV by line probe assay, CFLP analysis, and automated DNA sequencing indicated that the average cost per amplicon was lowest for CFLP analysis, at $20 (direct costs). On the basis of these findings we propose that CFLP analysis is a robust, sensitive, specific, and an economical method for large-scale screening of HCV-infected patients for alpha interferon-resistant HCV genotypes. The paper describes an algorithm that uses as a reflex test the RT-PCR-based qualitative screening of samples for HCV detection and also addresses genotypes that are ambiguous.

Algorithms↗
PubMed1997

Quantitation of T2' anisotropic effects on magnetic resonance bone mineral density measurement.

In this paper, the authors quantitate the anisotropy of susceptibility effects in an uniaxial trabecular bone model and show its relevance to clinical MR bone mineral density measurements. A physical model is described that quantitates the anisotropic MR behavior of uniaxial trabecular bone. To test the model; a phantom of parallel polyethylene filaments was scanned every 15 degrees between 0 degrees and 90 degrees with respect to the system's main magnetic field (B0). The distal radial metaphysis of a healthy female volunteer was scanned in orthogonal projections. The signal from each phantom image and each radial image was separated in a pixel-wise fashion into R2 and R2' maps. As predicted, R2' relaxation showed anisotropic behavior and changed according to sin2 (theta), confirming that columnar structures parallel with B0 will cause no MR susceptibility effects. Scans of the distal radius showed that R2' relaxation was twice as great with the forearm perpendicular to B0 as when it was parallel to it, demonstrating different contributions from struts and columns. For both phantom and radial bone scans, R2 relaxation was isotropic and did not change with object orientation.

Adult↗
PubMed1996

Segmentation of multispectral magnetic resonance image using penalized fuzzy competitive learning network.

Segmentation (tissue classification) of the medical images obtained from Magnetic resonance (MR) images is a primary step in most applications of computer vision to medical image analysis. This paper describes a penalized fuzzy competitive learning network designed to segment multispectral MR spin echo images. The proposed approach is a new unsupervised and winner-takes-all scheme based on a neural network using the penalized fuzzy clustering technique. Its implementation consists of the combination of a competitive learning network and penalized fuzzy clustering methods in order to make parallel implementation feasible. The penalized fuzzy competitive learning network could provide an acceptable result for medical image segmentation in parallel processing using the hardware implementation. The experimental results show that a promising solution can be obtained using the penalized fuzzy competitive learning neural network based on least squares criteria.

Adult↗
PubMed1992

Near-infrared imaging in vivo (I): Image restoration technique applicable to the NIR projection images.

To enhance spatial resolution of NIR projection images in vivo, we performed refocusing of NIR projection images of human forearm of about 50 mm thickness. A volunteer's forearm was illuminated by parallel light beam-flux, and then projection images at 750 nm was measured with a Pertier-cooled CCD video camera and digitized. For refocusing computation, PSF of the light transmitted through the tissue was calculated by general equation proposed by van der Zee and Delpy (1988). Simple inverse, constrained least squares, and Wiener filters were tested as refocusing algorithms. Wiener filter gave the best result in terms of image quality and computation time. By applying Wiener filter to the image refocusing of NIR projection images of human forearm, we obtained enhanced spatial resolution.

Evaluation Studies as Topic↗
PubMed1993

Optimal test strategy in the case of two tests and one disease.

Decision-making, especially about test performance, is very complex in nature. Clinical decision analysis can provide tools for doctors which can be used in improving the ordering of laboratory tests. This article describes an approach which is relevant for medical practice and easy to understand, with the goal of obtaining better decisions rather than optimal solutions. The methodology enables a clear understanding of the possibilities and restrictions of test use and needs very little calculation. The cornerstone of the methodology is a graphical representation, by which the benefits of test use are evaluated. Furthermore, a simple algorithm has been developed that can be used to find the optimal solution in the case of two tests. In each step decision rules can be used. In a graphical representation the effect of combining tests can be easily evaluated. If a test combination is chosen one has to decide which sequence is optimal. Finally one has to choose between parallel and series testing. The gain in time of the parallel procedure (and possible gain in effectiveness of treatment) should be compared with the efficiency gain of series testing. The authors conclude that the developed methodology is closer to the intuitive decision-making process than the traditional decision-making techniques and therefore can be used in order to improve the rather intuitive decisions of doctors.

Clinical Laboratory Techniques↗
PubMed2026

The application of artificial intelligence in healthcare practice: A mapping review of systematic reviews.

Artificial intelligence (AI) is rapidly transforming healthcare practice, with growing evidence supporting its use in diagnosis, prognosis, treatment planning, and operational decision-making. The proliferation of systematic reviews in recent years underscores the need for an updated synthesis of the literature to inform research, policy, and practice. We searched PubMed, Web of Science, Scopus, IEEE Xplore, and CINAHL for systematic reviews and meta-analyses published between 2019 and February 2026. Eligible reviews focused on AI applications in healthcare practice, were peer-reviewed, and written in English. A total of 368 reviews met the inclusion criteria. Publication volume increased steadily, peaking in 2025. AI research was concentrated in high-density domains, such as radiology, oncology, and critical care. Across reviews, diagnostic imaging, electronic health record (EHR) data, and biomarkers/laboratory results accounted for 68% of training data sources, though newer data types, such as wearable device and sensor data, emerged from 2022 onward. Diagnosis, prognosis, and treatment comprised over 80% of AI applications, with novel uses emerging in recent years, such as AI-assisted clinical documentation (e.g., ambient documentation tools) and patient education. Ethical concerns were reported in 78.5% of reviews, with privacy, model accuracy, data and algorithmic bias, and explainability as recurrent themes. The proportion of reviews reporting ethical concerns increased from 2021 to 2025. AI applications in healthcare are expanding in scope, diversifying in data sources, and evolving toward novel clinical and operational uses. The human-centered AI or augmented intelligence paradigm, integrating computational precision with clinical expertise, holds significant promise but will require parallel advances in governance, regulatory frameworks, and ethical oversight to ensure safe adoption.

Artificial Intelligence↗
PubMed2000

Optimal asthma control, starting with high doses of inhaled budesonide.

The aim of this study was to determine whether outcomes in poorly controlled asthma can be further improved with a starting dose of inhaled budesonide higher than that recommended in international guidelines. The study had a parallel-group design and included 61 subjects with poorly controlled asthma, randomized to receive 3,200 microg or 1,600 microg budesonide daily by Turbuhaler for 8 weeks (double-blind), then 1,600 microg x day(-1) for 8 weeks (single-blind), followed by 14 months of open-label budesonide dose down-titration using a novel algorithm, with a written asthma crisis plan based on electronic peak expiratory flow monitoring. The primary outcome variable for weeks 1-16 was change in airway hyperresponsiveness (AHR), and, for the open-label phase, mean daily budesonide dose. By week 16, there were large changes from baseline in all outcomes, with no significant differences between the 3,200- and 1,600-microg x day(-1) starting dose groups (AHR increased by 3.2 versus 3.0 doubling doses, p=0.7; morning peak flow increased by 134 versus 127 L x min(-1), p=0.8). Subjects starting with 3,200 microg x day(-1) were 3.8 times more likely to achieve AHR within the normal range, as defined by a provocative dose of histamine causing a 20% fall in forced expiratory volume in one second (PD20) of > or = 3.92 micromol by week 16 (p=0.03) [corrected]. During dose titration, there was no significant difference in mean budesonide dose (1,327 versus 1,325 microg x day(-1), p>0.3). Optimal asthma control was achieved in the majority of subjects (at completion/withdrawal: median symptoms 0.0 days x week(-1), beta2-agonist use 0.2 occasions x day(-1), and PD20 2.4 micromol). In subjects with poorly controlled asthma, a starting dose of 1,600 microg x day(-1) budesonide was sufficient to lead to optimal control in most subjects. The high degree of control achieved, compared with previous studies, warrants further investigation.

Administration, Inhalation↗
PubMed1998

A linear model method for rank measures of association from longitudinal studies with fixed conditions (visits) for data collection and more than two groups.

Several statistical methods are available for the analysis of responses with ordinal categories or continuous distributions for the respective visits in longitudinal studies. This paper discusses an alternative nonparametric strategy for studies with more than two groups through Mann-Whitney rank measures of association for all pairs of groups. The proposed method is based on U-statistic theory, and it applies a linear or linear logistic model to the Mann-Whitney estimators for the probabilities of better response for each group relative to each of the others. In addition, the ways of adjusting for covariables and managing stratification factors are explained. Analysis of parallel dose-response relationships for two treatments is illustrated for the proposed method with data from a multicenter study with repeated measurements. A nonparametric estimator for relative potency is provided from the method.

Algorithms↗
PubMed1999

Signalling of static and dynamic features of muscle spindle input by external cuneate neurones in the cat.

1. The present experiments examined the capacity of external cuneate nucleus (ECN) neurones in the anaesthetized cat to respond to static and vibrotactile stretch of forearm extensor muscles. The aim was to compare their signalling capacities with the known properties of main cuneate neurones in order to determine whether there is differential processing of muscle spindle inputs at these parallel relay sites. 2. Static stretch (<= 2 mm in amplitude) and sinusoidal vibration were applied longitudinally to individual muscle tendons and responses recorded from single ECN neurones. The muscle-related ECN neurones that were sampled displayed a high sensitivity to both static and dynamic components of stretch, including muscle vibration at frequencies of 50-800 Hz, consistent with their dominant input being derived from primary spindle afferent fibres. 3. In response to ramp-and-hold muscle stretch, ECN neurones resembled their main cuneate counterparts in the pattern of their responses and in quantitative response measures. Their coefficients of variation in interspike intervals during steady stretch ranged from approximately 0.3 to 0.7, as they do in main cuneate responses, and their stimulus-response relations were graded as a function of stretch magnitude with low variability in responses at a fixed stretch amplitude. 4. In response to muscle vibration, ECN activity was tightly phase locked to the vibration waveform, in particular at frequencies of <= 150 Hz, where vector strength measures (R) were high (R >= 0.8) before declining as a function of frequency, with R values of approximately 0.6 at 300 Hz and <= 0.4 at 800 Hz. Both the qualitative and quantitative aspects of ECN responsiveness to the vibro-stretch disturbances were indistinguishable from those of the main cuneate neurones. 5. The results demonstrate a high transmission fidelity for muscle signals across the ECN and no evidence for differential synaptic transmission across the parallel main and external cuneate nuclei. Earlier limitations observed in the capacity of cerebellar Purkinje cells to respond to primary spindle inputs must therefore be imposed at synapses within the cerebellum.

Algorithms↗
PubMed1994

Two-dimensional random arrays for real time volumetric imaging.

Two-dimensional arrays are necessary for a variety of ultrasonic imaging techniques, including elevation focusing, 2-D phase aberration correction, and real time volumetric imaging. In order to reduce system cost and complexity, sparse 2-D arrays have been considered with element geometries selected ad hoc, by algorithm, or by random process. Two random sparse array geometries and a sparse array with a Mills cross receive pattern were simulated and compared to a fully sampled aperture with the same overall dimensions. The sparse arrays were designed to the constraints of the Duke University real time volumetric imaging system, which employs a wide transmit beam and receive mode parallel processing to increase image frame rate. Depth-of-field comparisons were made from simulated on-axis and off-axis beamplots at ranges from 30 to 160 mm for both coaxial and offset transmit and receive beams. A random array with Gaussian distribution of transmitters and uniform distribution of receivers was found to have better resolution and depth-of-field than both a Mills cross array and a random array with uniform distribution of both transmit and receive elements. The Gaussian random array was constructed and experimental system response measurements were made at several ranges. Comparisons of B-scan images of a tissue mimicking phantom show improvement in resolution and depth-of-field consistent with simulation results.

Ultrasonography↗
PubMed1992

Evidence of hyperplanes in the genetic learning of neural networks.

Genetic Algorithms have been successfully applied to the learning process of neural networks simulating artificial life. In previous research we compared mutation and crossover as genetic operators on neural networks directly encoded as real vectors (Manczer and Parisi 1990). With reference to crossover we were actually testing the building blocks hypothesis, as the effectiveness of recombination relies on the validity of such hypothesis. Even with the real genotype used, it was found that the average fitness of the population of neural networks is optimized much more quickly by crossover than it is by mutation. This indicated that the intrinsic parallelism of crossover is not reduced by the high cardinality, as seems reasonable and has indeed been suggested in GA theory (Antonisse 1989). In this paper we first summarize such findings and then propose an interpretation in terms of the spatial correlation of the fitness function with respect to the metric defined by the average steps of the genetic operators. Some numerical evidence of such interpretation is given, showing that the fitness surface appears smoother to crossover than it does to mutation. This confirms indirectly that crossover moves along privileged directions, and at the same time provides a geometric rationale for hyperplanes.

Algorithms↗
PubMedDate not supplied

FDTD simulations of Clini-Therm applicators on inhomogeneous planar tissue models.

A finite-difference time-domain (FDTD) algorithm was used to compute SAR distributions in planar fat-muscle phantom exposed to the Clini-Therm microwave applicators. The models consisted of a 30 X 30 X 7.5 cm phantom and a 15 X 15 cm, 10 X 10 cm or 7.5 X 7.5 cm aperture dielectric slab loaded applicator. The phantom was either filled with muscle material or with 1.0 cm fat on 6.5 cm muscle. A mineral oil bolus was placed on the fat-muscle model with its integrated water channels parallel to the electric or magnetic field. The FDTD resolution was 3 mm and the applicators were excited with a Gaussian pulse. The computations required 6000-8000 time steps to reach steady state, with 45-48 Mwords on a Cray Y-MP C-90 in 1000-1200 CPU seconds. The electric field components at 915 MHz were obtained by summing the Fourier coefficients at each grid point during each time step and SAR was determined. The results were qualitatively compared to existing and published thermographic heating patterns with good agreement. The computed electric field distributions had provided a three dimensional view into the problem space to investigate and understand wave propagation phenomena in complex inhomogeneous configurations that were not feasible with experimental models.

Adipose Tissue↗
PubMedDate not supplied

Automated texture-based segmentation of ultrasound images of the prostate.

Segmenting two-dimensional images of the prostate into prostate and nonprostate regions is required when forming a three-dimensional image of the prostate from a set of parallel two-dimensional images. The texture-based segmentation method presented here is a pixel classifier based on four texture energy measures associated with each pixel in the image. An automated clustering procedure is used to label each pixel in the image with the label of its most probable class. The segmented images produced as the result of applying the algorithm to an example image are presented and discussed. The automated segmentation algorithm has been found to hold promise as an automated segmentation method.

Adult↗
PubMed1986

Interactive display of three-dimensional radionuclide distributions.

Tomographic images of three-dimensional radionuclide distributions are usually presented as a set of parallel planar slices. A technique has been developed to enable the observer to perceive directly the three-dimensional morphology of the activity distribution by displaying iso-count surfaces. Computer graphics techniques have been used to present shaded surface images using a conventional nuclear medicine computer system. Because of the relatively coarse spatial sampling of radionuclide tomograms, a smooth shading algorithm based on a local polynomial fit procedure was developed. By using a simple solid model approach for data storage the program executes interactively, allowing the observer to view the reconstructed activity distribution from any aspect. The technique has been successfully applied to radionuclide tomograms of several organs, and promises to be particularly useful for the display or cardiac blood pool data.

Algorithms↗