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 1,333 records · Page 74Linked to original sources
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↗
PubMed2002

Lensed optical fiber sensors for on-line measurement of flow.

This paper describes a system using lensed optical fiber sensors that are arranged in the form of two orthogonal projections. The sensors are placed around a process vessel for upstream and downstream measurements. The purpose of the system is for on-line monitoring of particles and droplets being conveyed by a fluid. The lenses were constructed using a custom heating fixture. The fixture enables the lenses to be constructed with similar radii resulting in identical characteristics with minimum differences in transmitted intensity and emission angle. By collimating radiation from two halogen bulbs, radiation can be obtained by the sensors with radiation intensity related to the nature of the media. Each sensor interrogates a finite section of the measurement section. Each sensor provides a view. Parallel sensors provide a projection. Signal processing is carried out on the measured data in the time and frequency domains to investigate the latent information present in the flow signals.

Algorithms↗
PubMedDate not supplied

HOUDINI: a new approach to computer-based structure generation.

A new method of structure generation called convergent structure generation has been developed to address limitations of earlier methods. The features of the program (HOUDINI) based on this method include the following: a single integrated representation of the collective substructural information; the use of parallel atom groups for efficient processing of families of alternative substructural inferences; and a managed structure generation procedure designed to build required structural features early in the process.

Algorithms↗
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↗
PubMed2003

Stochastic modeling of RNA pseudoknotted structures: a grammatical approach.

MOTIVATION: Modeling RNA pseudoknotted structures remains challenging. Methods have previously been developed to model RNA stem-loops successfully using stochastic context-free grammars (SCFG) adapted from computational linguistics; however, the additional complexity of pseudoknots has made modeling them more difficult. Formally a context-sensitive grammar is required, which would impose a large increase in complexity. RESULTS: We introduce a new grammar modeling approach for RNA pseudoknotted structures based on parallel communicating grammar systems (PCGS). Our new approach can specify pseudoknotted structures, while avoiding context-sensitive rules, using a single CFG synchronized with a number of regular grammars. Technically, the stochastic version of the grammar model can be as simple as an SCFG. As with SCFG, the new approach permits automatic generation of a single-RNA structure prediction algorithm for each specified pseudoknotted structure model. This approach also makes it possible to develop full probabilistic models of pseudoknotted structures to allow the prediction of consensus structures by comparative analysis and structural homology recognition in database searches.

Algorithms↗
PubMed2003

Assessing stage of change for physical activity: how congruent are parallel methods?

The single-question self-classification Stages of Change scales (SAS) for two modes of physical activity were compared with parallel staging methods. In Study 1, the participants (N = 50) completed SAS in a questionnaire and were then personally interviewed on their physical activity. In four fifths of the cases, SAS indicated the same stage as the interviewer's judgment. In Study 2, a representative survey sample (N = 600) completed both SAS and, in another questionnaire, a three-question algorithm staging instrument (TSQ) constructed for the same target behaviors. About 50% of all participants were placed in the same stage with both instruments. The compatibility rate rose to 80% when the number of stages was reduced from the original eight to five. However, TSQ also accumulated a higher share of cases in the stages with regular action. In both studies, the most obvious sources for incompatible staging were the regularity and time frame of the targeted behavior. Thus, neither SAS nor TSQ is on its own a sufficiently accurate instrument for use in personalized stage-based interventions. TSQ shows no obvious advantages over SAS. In counseling, SAS seems useful in combination with a personal interview.

Adult↗
PubMed2005

Using geometric constraints through parallelepipeds for calibration and 3D modeling.

This paper concerns the incorporation of geometric information in camera calibration and 3D modeling. Using geometric constraints enables more stable results and allows us to perform tasks with fewer images. Our approach is motivated and developed within a framework of semi-automatic 3D modeling, where the user defines geometric primitives and constraints between them. It is based on the observation that constraints, such as coplanarity, parallelism, or orthogonality, are often embedded intuitively in parallelepipeds. Moreover, parallelepipeds are easy to delineate by a user and are well adapted to model the main structure of, e.g., architectural scenes. In this paper, first a duality that exists between the shape parameters of a parallelepiped and the intrinsic parameters of a camera is described. Then, a factorization-based algorithm exploiting this relation is developed. Using images of parallelepipeds, it allows us to simultaneously calibrate cameras, recover shapes of parallelepipeds, and estimate the relative pose of all entities. Besides geometric constraints expressed via parallelepipeds, our approach simultaneously takes into account the usual self-calibration constraints on cameras. The proposed algorithm is completed by a study of the singular cases of the calibration method. A complete method for the reconstruction of scene primitives that are not modeled by parallelepipeds is also briefly described. The proposed methods are validated by various experiments with real and simulated data, for single-view as well as multiview cases.

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↗
PubMed2005

A novel extension of the parallel-beam projection-slice theorem to divergent fan-beam and cone-beam projections.

The general goal of this paper is to extend the parallel-beam projection-slice theorem to divergent fan-beam and cone-beam projections without rebinning the divergent fan-beam and cone-beam projections into parallel-beam projections directly. The basic idea is to establish a novel link between the local Fourier transform of the projection data and the Fourier transform of the image object. Analogous to the two- and three-dimensional parallel-beam cases, the measured projection data are backprojected along the projection direction and then a local Fourier transform is taken for the backprojected data array. However, due to the loss of the shift invariance of the image object in a single view of the divergent-beam projections, the measured projection data is weighted by a distance dependent weight w(r) before the local Fourier transform is performed. The variable r in the weighting function w(r) is the distance from the backprojected point to the x-ray source position. It is shown that a special choice of the weighting function, w(r)=1/r, will facilitate the calculations and a simple relation can be established between the Fourier transform of the image function and the local Fourier transform of the 1/r-weighted backprojection data array. Unlike the parallel-beam cases, a one-to-one correspondence does not exist for a local Fourier transform of the backprojected data array and a single line in the two-dimensional (2D) case or a single slice in the 3D case of the Fourier transform of the image function. However, the Fourier space of the image object can be built up after the local Fourier transforms of the 1/r-weighted backprojection data arrays are shifted and then summed in a laboratory frame. Thus the established relations Eq. (27) and Eq. (29) between the Fourier space of the image object and the Fourier transforms of the backprojected data arrays can be viewed as a generalized projection-slice theorem for divergent fan-beam and cone-beam projections. Once the Fourier space of the image function is built up, an inverse Fourier transform could be performed to reconstruct tomographic images from the divergent beam projections. Due to the linearity of the Fourier transform, an image reconstruction step can be performed either when the complete Fourier space is available or in parallel with the building of the Fourier space. Numerical simulations are performed to verify the generalized projection-slice theorem by using a disc phantom in the fan-beam case.

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↗
PubMed2004

SS-Wrapper: a package of wrapper applications for similarity searches on Linux clusters.

BACKGROUND: Large-scale sequence comparison is a powerful tool for biological inference in modern molecular biology. Comparing new sequences to those in annotated databases is a useful source of functional and structural information about these sequences. Using software such as the basic local alignment search tool (BLAST) or HMMPFAM to identify statistically significant matches between newly sequenced segments of genetic material and those in databases is an important task for most molecular biologists. Searching algorithms are intrinsically slow and data-intensive, especially in light of the rapid growth of biological sequence databases due to the emergence of high throughput DNA sequencing techniques. Thus, traditional bioinformatics tools are impractical on PCs and even on dedicated UNIX servers. To take advantage of larger databases and more reliable methods, high performance computation becomes necessary. RESULTS: We describe the implementation of SS-Wrapper (Similarity Search Wrapper), a package of wrapper applications that can parallelize similarity search applications on a Linux cluster. Our wrapper utilizes a query segmentation-search (QS-search) approach to parallelize sequence database search applications. It takes into consideration load balancing between each node on the cluster to maximize resource usage. QS-search is designed to wrap many different search tools, such as BLAST and HMMPFAM using the same interface. This implementation does not alter the original program, so newly obtained programs and program updates should be accommodated easily. Benchmark experiments using QS-search to optimize BLAST and HMMPFAM showed that QS-search accelerated the performance of these programs almost linearly in proportion to the number of CPUs used. We have also implemented a wrapper that utilizes a database segmentation approach (DS-BLAST) that provides a complementary solution for BLAST searches when the database is too large to fit into the memory of a single node. CONCLUSIONS: Used together, QS-search and DS-BLAST provide a flexible solution to adapt sequential similarity searching applications in high performance computing environments. Their ease of use and their ability to wrap a variety of database search programs provide an analytical architecture to assist both the seasoned bioinformaticist and the wet-bench biologist.

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↗