SEARCH · PubMed Health

Results for “Algorithms”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,495 records · Page 83Linked to original sources

PubMed2002

Neural network based optimal routing algorithm for communication networks.

This paper presents the capability of the neural networks as a computational tool for solving constrained optimization problem, arising in routing algorithms for the present day communication networks. The application of neural networks in the optimum routing problem, in case of packet switched computer networks, where the goal is to minimize the average delays in the communication have been addressed. The effectiveness of neural network is shown by the results of simulation of a neural design to solve the shortest path problem. Simulation model of neural network is shown to be utilized in an optimum routing algorithm known asflow deviation algorithm. It is also shown that the model will enable the routing algorithm to be implemented in real time and also to be adaptive to changes in link costs and network topology.

Algorithms↗

PubMed2003

The functional localization of neural networks using genetic algorithms.

We presented an algorithm for extracting Boolean functions (propositions, rules) from the units in trained neural networks. The extracted Boolean functions make the hidden units understandable. However, in some cases, the extracted Boolean functions are complicated, and so are not understandable, which means that the hidden units are not functionally localized. This paper presents an algorithm for the functional localization of (the hidden units of) neural networks. When a hidden unit is well approximated to a low-order Boolean function, the unit can be regarded as functionally localized. The functional localization of a hidden unit is evaluated by the error between the hidden unit and the low-order Boolean function extracted from the hidden unit. The optimization is executed by genetic algorithms. We applied it to vote data, mushroom data and chess data. Experimental results show that the algorithm works well.

Agaricales↗

PubMed2003

A new algorithm for online structure and parameter adaptation of RBF networks.

This paper deals with the problem of online adaptation of radial basis function (RBF) neural networks. A new adaptive training method is presented, which is able to modify both the structure of the network (the number of nodes in the hidden layer) and the output weights, as the algorithm proceeds. These adaptation capabilities make the algorithm suitable for modeling dynamical time varying systems, where not only the dynamics but also the operating region changes with time. Therefore, the important issue of extrapolation is faced successfully, but at the same time the algorithm takes care of the size of the network, by deleting the hidden node centers that remain inactive for a long time. The selection of the network centers is based on a fuzzy partition of the input space, which defines a number of fuzzy subspaces. The algorithm considers the centers of the fuzzy subspaces as candidates for becoming hidden node centers and makes the selections, so that at least one center is close enough to each input example. The proposed technique is illustrated through the application to time varying dynamical systems and is compared to other adaptive training methods.

Adaptation, Biological↗

PubMedDate not supplied

On structure-exploiting trust-region regularized nonlinear least squares algorithms for neural-network learning.

This paper briefly introduces our numerical linear algebra approaches for solving structured nonlinear least squares problems arising from 'multiple-output' neural-network (NN) models. Our algorithms feature trust-region regularization, and exploit sparsity of either the 'block-angular' residual Jacobian matrix or the 'block-arrow' Gauss-Newton Hessian (or Fisher information matrix in statistical sense) depending on problem scale so as to render a large class of NN-learning algorithms 'efficient' in both memory and operation costs. Using a relatively large real-world nonlinear regression application, we shall explain algorithmic strengths and weaknesses, analyzing simulation results obtained by both direct and iterative trust-region algorithms with two distinct NN models: 'multilayer perceptrons' (MLP) and 'complementary mixtures of MLP-experts' (or neuro-fuzzy modular networks).

Algorithms↗

PubMed2000

Construction of an algorithm for quick detection of patients with low bone mineral density and its applicability in daily general practice.

OBJECTIVE: To construct a quick algorithm to detect patients with low bone mineral density (BMD) and osteoporosis and determine its applicability in daily general practice. DESIGN: Cross-sectional study in all 9107 postmenopausal women, aged 50-80, registered at 12 general practice centers. SUBJECTS AND MEASUREMENTS: All healthy women (5303) and 25% of the remaining group (943/3804) were invited to participate. Of 6246 invited women, 4725 (76%) participated. The women were questioned (state of health, medical history, family history, and food questionnaire) and examined [weight, height, body mass index (BMI), and BMD of the lumbar spine]. STATISTICS: Multivariable, stepwise backward and forward logistic regression analyses were performed, with BMD of the lumbar spine (L2-L4, cut-off points at 0.800 g/cm(2) for osteoporosis and 0.970 g/cm(2) for low BMD) as the dependent variable. An algorithm was constructed with those variables that correlated statistically significantly and clinically relevant with the presence of both osteoporosis and low BMD. RESULTS: The prevalence of osteoporosis was 23%, that of low BMD was 65%. Only three variables (age, BMI, and fractures) were statistically significant and clinically relevant correlated with the presence of both osteoporosis and low BMD. Age (OR 2.70 for osteoporosis and OR 1.77 for low BMD) and fractures during the past five years (OR 3.60 for osteoporosis and OR 2.85 for low BMD) were found to be the key predictors. From the algorithm the absolute risks varied from 9% to 51% for osteoporosis and from 48% to 84% for low BMD. The corresponding relative risks varied from 1.0 to 5.7 and from 1.0 to 1.8. CONCLUSIONS: Using an algorithm with age, BMI, and fracture history subgroups at high risk could be identified. However, in whatever combination, many women with osteoporosis could not be identified. Despite the differences in methods, we found predictors for osteoporosis which were comparable with the results of other cross-sectional studies, meaning that the first selection of patients at high risk for low BMD can be done adequately by both specialists and general practitioners.

Aged↗

PubMed2003

Development of a computer algorithm for defining an active drug list using an automated pharmacy database.

BACKGROUND AND OBJECTIVE: Increasingly, pharmacy databases are being used to assist in evaluating the appropriateness of drug therapy. Such determinations often require the creation of a drug regimen at a particular point in time. The objective of this study was to develop a computer algorithm for defining a cross-sectional active drug list. METHODS: Electronic pharmacy data were obtained as part of the Iowa Medicaid Pharmaceutical Case Management Program. The active drug lists generated by the computer algorithm were compared with active drug lists generated by independent pharmacist review of the pharmacy refill data. RESULTS: In a sample of 25 patients who received 379 potentially active medications, the interrater reliability between pharmacist reviewers was excellent (kappa=0.94). In a second sample of 100 patients who received 1476 potentially active medications, the computer algorithm had a sensitivity of 93.8% and specificity of 91.7%, using pharmacist review as the comparison standard. CONCLUSION: The computer algorithm was found to be a valid method of processing electronic pharmacy data to yield a characterization of drug exposure at a point in time. The potential benefits and limitations of using this approach are discussed.

Adult↗

PubMed1998

The equivalence of SF-36 summary health scores estimated using standard and country-specific algorithms in 10 countries: results from the IQOLA Project. International Quality of Life Assessment.

Data from general population surveys (n = 1771 to 9151) in nine European countries (Denmark, France, Germany, Italy, the Netherlands, Norway, Spain, Sweden, and the United Kingdom) were analyzed to test the algorithms used to score physical and mental component summary measures (PCS-36/MCS-36) based on the SF-36 Health Survey. Scoring coefficients for principal components were estimated independently in each country using identical methods of factor extraction and orthogonal rotation. PCS-36 and MCS-36 scores were also estimated using standard (U.S.-derived) scoring algorithms, and results were compared. Product-moment correlations between scores estimated from standard and country-specific scoring coefficients were very high (0.98 to 1.00) for both physical and mental health components in all countries. As hypothesized for orthogonal components, correlations between physical and mental components within each country were very low (0.00 to 0.12) for both estimation methods. Mean scores for PCS-36 differed by as much as 3.0 points across countries using standard scoring, and mean scores for MCS-36 differed across countries by as much as 6.4 points. In view of the high degree of equivalence observed within each country, using standard and country-specific algorithms, we recommend use of standard scoring algorithms for purposes of multinational studies involving these 10 countries.

Algorithms↗

PubMedDate not supplied

How blurry is that border? An investigation into algorithmic reproduction of skin lesion border cut-off.

This paper describes an approach to algorithmically measuring the second of the ABCD criteria [Stolz W, Braun-Falco O, Bilek P, Landthaler M, Cognetta AB. Color atlas of dermatoscopy. Oxford: Blackwell Science Ltd, 1994], border cut-off. The method measures the lightness gradient of the lesion at the boundary. Gradients are assigned to "blurry" or "sharp cut-off" on the basis of a threshold obtained by comparing dermatologists' opinions with the algorithm. The algorithm appears to correlate quite well with dermatologists' opinions, although several concerns are apparent, especially the distribution of the image set. This algorithm presents a first step at measuring border cut-off and eventually, reproducing the ABCD criteria [Stolz W, Braun-Falco O, Bilek P, Landthaler M, Cognetta AB. Color atlas of dermatoscopy. Oxford: Blackwell Science Ltd, 1994].

Algorithms↗

PubMedDate not supplied

Wavelet based multiresolution expectation maximization image reconstruction algorithm for positron emission tomography.

Maximum Likelihood (ML) estimation based Expectation Maximization (EM) [IEEE Trans Med Imag, MI-1 (2) (1982) 113] reconstruction algorithm has shown to provide good quality reconstruction for positron emission tomography (PET). Our previous work [IEEE Trans Med Imag, 7(4) (1988) 273; Proc IEEE EMBS Conf, 20(2/6) (1998) 759] introduced the multigrid (MG) and multiresolution (MR) concept for PET image reconstruction using EM. This work transforms the MGEM and MREM algorithm to a Wavelet based Multiresolution EM (WMREM) algorithm by extending the concept of switching resolutions in both image and data spaces. The MR data space is generated by performing a 2D-wavelet transform on the acquired tube data that is used to reconstruct images at different spatial resolutions. Wavelet transform is used for MR reconstruction as well as adapted in the criterion for switching resolution levels. The advantage of the wavelet transform is that it provides very good frequency and spatial (time) localization and allows the use of these coarse resolution data spaces in the EM estimation process. The MR algorithm recovers low-frequency components of the reconstructed image at coarser resolutions in fewer iterations, reducing the number of iterations required at finer resolution to recover high-frequency components. This paper also presents the design of customized biorthogonal wavelet filters using the lifting method that are used for data decomposition and image reconstruction and compares them to other commonly known wavelets.

Algorithms↗

PubMedDate not supplied

A comparative study of Powell's and Downhill Simplex algorithms for a fast multimodal surface matching in brain imaging.

Multimodal images registration can be very helpful for diagnostic applications. However, even if a lot of registration algorithms exist, only a few really work in clinical routines. We developed a method based on surface matching and compared two minimization algorithms: Powell's and Downhill Simplex. We studied the influence of some factors (chamfer map computation, number and order of parameters to determine, minimization criteria) on the final accuracy of the algorithm. Using this comparison, we improved some processing steps to allow a clinical use, and selected the simplex algorithm which presented the best results.

Algorithms↗

PubMed2000

A hierarchical neural network algorithm for robust and automatic windowing of MR images.

A novel hierarchical neural network based algorithm for automatic adjustment of display window width and center for a wide range of magnetic resonance (MR) images is presented in this paper. The algorithm consists of a feature generator utilizing both wavelet histogram and compact spatial statistical information computed from a MR image, a competitive layer based neural network for clustering MR images into different subclasses, two pairs of a radial basis function (RBF) network and a bi-modal linear estimator for each subclass, as well as a data fusion process using estimates from both estimators to compute the final display parameters. Both estimators can adapt to new kinds of MR images simply by training them with those images, which make the algorithm adaptive and extendable. The RBF based estimator performs very well for images that are similar to those in the training data set. The bi-modal linear estimator provides reasonable estimations for a wide range of images that may not be included in the training data set. The data fusion step makes the final estimation of the display parameters accurate for trained images and robust for the unknown images. The algorithm has been tested on a wide range of MR images and has shown satisfactory results.

Algorithms↗

PubMed1999

Feature selection for optimized skin tumor recognition using genetic algorithms.

In this paper, a new approach to computer supported diagnosis of skin tumors in dermatology is presented. High resolution skin surface profiles are analyzed to recognize malignant melanomas and nevocytic nevi (moles), automatically. In the first step, several types of features are extracted by 2D image analysis methods characterizing the structure of skin surface profiles: texture features based on cooccurrence matrices, Fourier features and fractal features. Then, feature selection algorithms are applied to determine suitable feature subsets for the recognition process. Feature selection is described as an optimization problem and several approaches including heuristic strategies, greedy and genetic algorithms are compared. As quality measure for feature subsets, the classification rate of the nearest neighbor classifier computed with the leaving-one-out method is used. Genetic algorithms show the best results. Finally, neural networks with error back-propagation as learning paradigm are trained using the selected feature sets. Different network topologies, learning parameters and pruning algorithms are investigated to optimize the classification performance of the neural classifiers. With the optimized recognition system a classification performance of 97.7% is achieved.

Algorithms↗

PubMed2001

[Analysis of the convergence relation of reconstructed algorithms disregarding local and global visual evaluations].

Tomographic reconstruction methods used in positron emission tomography are classified in two major groups: the traditionally and still widely applied filtered backprojection, and the iterative methods based on statistical models. This study focused on the objective comparison of different reconstruction algorithms, excluding criteria based on pure visual evaluation. The evaluation criteria were mathematically defined parameters, i.e., mean square error, standard deviation, signal-to-noise ratio and contrast recovery. The methods used for comparison were the classical filtered backprojection, the maximum likelihood expectation maximization algorithm, the maximum a-posteriori reconstruction model based on the Bayes Theorem, as well as the acceleration algorithms based on ordered subsets and high over-relaxation. These algorithms were evaluated by means of a mathematical brain phantom and of a physical spherical phantom. In terms of the applied parameters, the majority of the experiments showed a quantifiable superiority of the iterative methods compared the filtered backprojection.

Algorithms↗

PubMed2002

A simple exploratory algorithm for the accurate and fast detection of spontaneous synaptic events.

We have developed a program for the fast and accurate detection of spontaneous synaptic events. The algorithm identifies each event of which the slope and amplitude which meet criteria. The significant feature of this algorithm is its stepwise and exploratory search for the onset and the peak points. During the first step, the program employing the algorithm makes a rough estimate of the candidate for a synaptic event, and determines a 'temporary' onset data point. The next step is the detection of the true onset data point and 'temporary' peak data point, which probably exist several points after the temporary onset data point. The third step is a backward search to detect the true peak data point. The final step is to check whether the amplitude of the detected event exceeds the threshold. This stepwise and shuttlewise search allows for the accurate detection of the peak points. Using this program, we succeeded in detecting an increased frequency and amplitude of spontaneous excitatory postsynaptic currents in chick cerebral neurons following the application of 12-O-tetradecanoyl-phorbol-13-acetate (TPA). In addition, we demonstrated that the program employing the algorithm was able to be used for the detection of extracellular action potentials.

Action Potentials↗

PubMed1998

Computer search algorithms in protein modification and design.

The computer-aided design of protein sequences requires efficient search algorithms to handle the enormous combinatorial complexity involved. A variety of different algorithms have now been applied with some success. The choice of algorithm can influence the representation of the problem in several important ways--the discreteness of the configuration, the types of energy terms that can be used and the ability to find the global minimum energy configuration. The use of dead end elimination to design the complete sequence for a small protein motif and the use of genetic and mean-field algorithms to design hydrophobic cores for proteins represent the major themes of the past year.

Algorithms↗

PubMed1998

A universal algorithm for fast and automated charge state deconvolution of electrospray mass-to-charge ratio spectra.

This article describes a new algorithm for charge state determination and deconvolution of electrospray ionization (ESI) mass-to-charge ratio spectra. The algorithm (ZSCORE) is based on a charge scoring scheme that incorporates all above-threshold members of a family of charge states or isotopic components, and deconvolves both low- and high-resolution mass-to-charge ratio spectra, with or without a peak list (stick plot). A scoring weight factor, log (I/I0), in which I is the signal magnitude at a calculated mass-to-charge ratio, and I0 is the signal threshold near that mass-to-charge ratio, was used in most cases. For high-resolution mass-to-charge ratio spectra in which all isotopic peaks are resolved, the algorithm can deconvolve overlapped isotopic multiplets of the same or different charge state. Compared to other deconvolution techniques, the algorithm is robust, rapid, and fully automated (i.e., no user input during the deconvolution process). It eliminates artifact peaks without introducing peak distortions. Its performance is demonstrated for experimental ESI Fourier transform ion cyclotron resonance mass-to-charge ratio spectra (both low and high resolution). Charge state deconvolution to yield a "zero-charge" mass spectrum should prove particularly useful for interpreting spectra of complex mixtures, identifying contaminants, noncovalent adducts, fragments (N-terminal, C-terminal, internal), and chemical modifications of electrosprayed biomacromolecules.

Algorithms↗

PubMed1990

The effect of an airway algorithm on flight nurse behavior.

This study examined the effects of a clinical algorithm (based on Glasgow Coma Scale) on the airway management of patients transported from the scene of accident or illness by a helicopter emergency medical service staffed with two flight nurses. The year before institution of the algorithm was compared to the subsequent year. Patients from the two years were similar with regard to diagnoses and physiologic status. Patients transported after the introduction of the algorithm were over twice as likely to have had an airway maneuver attempted. An algorithm defining when airway maneuvers ought to be attempted in the field may assist flight nurses in more aggressively managing airways.

Aircraft↗

PubMed2003

Diffusion-weighted magnetic resonance imaging fibre tracking using a front evolution algorithm.

A novel technique is presented for estimating white matter connectivity in vivo using diffusion-weighted magnetic resonance imaging. The concept of a fibre orientation density function (ODF) is described, which characterises the uncertainty in the orientation of the underlying white matter fibres, given the set of diffusion-weighted signal intensities at the point of interest. The proposed algorithm is based on the evolution of a front from a seed region, using the information provided by the fibre ODF. Each point reached by the front is assigned an index of connectivity with the seed region. The algorithm was used to track various major white matter fibre pathways in two data sets acquired on the same healthy adult volunteer over separate occasions. Example tracks are shown to illustrate some of the properties of the algorithm, such as robustness to noise and branching capability. Finally, the dependence of the algorithm on the model used to derive the fibre ODF is discussed.

Adult↗