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 1,243 records · Page 69Linked to original sources

PubMedDate not supplied

An inductive algorithm approach to knowledge acquisition for expert system development. A pilot study.

Knowledge acquisition, which consists of knowledge elicitation and knowledge representation, often is considered the weakest link in the design of expert systems. Systems frequently are built on the knowledge of one expert and require extensive use of knowledge engineering techniques to elicit this knowledge from the expert. Inductive algorithms are a potential alternative method of knowledge acquisition for expert system development. The aim of this pilot study was to examine the feasibility of applying machine learning techniques, specifically, inductive algorithms, to an existing research database as a method for knowledge elicitation and knowledge representation for expert system development. Two inductive algorithms (C4 and Classification and Regression Trees [CART]) that generate decision trees were selected for the analysis using a data set of 201 patients hospitalized for Pneumocystis carinii pneumonia. Neither C4 nor CART produced trees with an accuracy that was significantly better than the baseline accuracy (71.3%) for prediction of outcome in the data set. The mean accuracy of the C4 decision trees was below baseline and the mean accuracy of CART decision trees was 74.6%. The experts found both algorithms comprehensible, but not adequate, and identified important missing predictor variables. The study findings suggest that additional research is needed to examine the appropriate use of inductive algorithms in the transformation of nursing data and information into nursing knowledge.

Algorithms↗

PubMed1993

Do diagnostic algorithms always produce a uniform lung scan interpretation?

Several algorithms have been devised to assist in the interpretation of ventilation-perfusion (V/Q) scans performed to diagnose pulmonary embolism. The degree to which adherence to a single algorithm facilitates diagnostic homogeneity among different readers, however, has been little investigated. We evaluated the individual variability in V/Q lung scan interpretation in a large, academic nuclear medicine division to determine the degree of interpretive heterogeneity among a group of physicians all using the same image interpretation algorithm. Ventilation-perfusion scan interpretive patterns and the diagnostic accuracy of individual physicians were evaluated using quantitative parameters to establish group norms and to detect variations from these norms. The performance of each reader was tracked over a 4 yr period. There was a significant variation in V/Q interpretive patterns and diagnostic accuracy between readers despite the attempted use of a uniform diagnostic algorithm. Subgroups of interpretive styles could be defined based on the percentage of intermediate (including both indeterminate and intermediate categories) scans read. Although there was significant variation in diagnostic accuracy among readers, there was no obvious correlation between accuracy and reading style except that the most nonstandard diagnostic patterns were associated with the most variable diagnostic accuracy. These data show a measurable variation in interpretive patterns and accuracy among multiple readers of V/Q scans despite attempted group adherence to an established diagnostic algorithm.

Algorithms↗

PubMed1995

Validation of an algorithm for oral anticoagulant dosing and appointment scheduling.

Computer clinical decision-support systems require validation before clinical use. This study compared recommendations on warfarin dosage adjustment and timing of the next appointment made by an algorithm with those made by experienced and inexperienced clinicians. Data abstracted from the records of 125 patients seen regularly in the anticoagulant clinic were used. The algorithm recommended dose changes and next appointment for cases with INRs between 1.8 to 4.2 (therapeutic range 2.0-3.0) and between 2.3 to 5.3 (therapeutic range 3.0-4.5). Beyond these values the algorithm referred the cases to "see doctor'. Compared to experienced clinicians, the algorithm was better at "recognising' difficult patients than inexperienced clinicians (kappa = 0.43 and 0.32 respectively). There was no statistically significant difference between all decision makers in dosage recommendations for the non-difficult cases, but there was much more variation amongst the inexperienced clinicians. The interval recommendations were statistically different between and within the different decision-makers. The inexperienced clinicians tended to give relatively longer intervals for a given dose change. In conclusion, the algorithm performs better than inexperienced clinicians and as well as experienced clinicians for the non-difficult cases.

Algorithms↗

PubMed1996

A branch and bound algorithm for local multiple alignment.

A Branch and Bound Algorithm has been developed to find a set of window positions in a compilation of sequences with globally maximal information content. We have also developed an algorithm for brute force evaluation of solutions which is faster by a factor of the length of the windows than the naïve brute force algorithm. The combination of these two algorithms allows us to solve problems to optimality that were previously amenable only to heuristic algorithms.

Algorithms↗

PubMed1997

Closed-loop subcutaneous insulin infusion algorithm with a short-acting insulin analog for long-term clinical application of a wearable artificial endocrine pancreas.

Considering the management and safety of the insulin delivery route when a wearable artificial endocrine pancreas is applied to ambulatory diabetic patients on a long-term basis, we developed a s.c. insulin infusion algorithm by analyzing the dynamics of a s.c. injected short-acting insulin analog (Insulin Lispro) by a three-compartment model. Principally the insulin infusion algorithm was developed as a transfer function with the first-order delay in both proportional and derivative actions to blood glucose concentrations. The parameters for this algorithm were calculated to simulate a physiological plasma insulin profile as closely as possible. By applying this algorithm with regular insulin, diabetic patients showed a 2 h postprandial hyperglycemia and a delayed hyperinsulinemia, followed by hypoglycemic episodes 4-5 h after oral glucose load, just as observed in the computer simulation study. However, using Insulin Lispro, a near-physiological glycemic control (postprandial blood glucose of 153.1 +/- 8.3 mg/100 ml at 60 min and 90.3 +/- 7.1 mg/100 ml at 180 min, respectively) could be achieved without showing any delayed hyperinsulinemia or hypoglycemia. Daily glycemic excursions were also controlled near-physiologically and although the daily insulin requirement (731.7 +/- 160.5 mU/kg/day) was slightly higher, it was not significantly different from that with i.v. insulin infusion (622.3 +/- 142.6 mU/kg/day). These results indicate that the application of s.c. insulin infusion algorithm with Insulin Lispro is feasible for long-term glycemic control with a wearable artificial endocrine pancreas in ambulatory diabetic patients.

Adult↗

PubMed1998

Blind source separation and deconvolution: the dynamic component analysis algorithm.

We derive a novel family of unsupervised learning algorithms for blind separation of mixed and convolved sources. Our approach is based on formulating the separation problem as a learning task of a spatiotemporal generative model, whose parameters are adapted iteratively to minimize suitable error functions, thus ensuring stability of the algorithms. The resulting learning rules achieve separation by exploiting high-order spatiotemporal statistics of the mixture data. Different rules are obtained by learning generative models in the frequency and time domains, whereas a hybrid frequency-time model leads to the best performance. These algorithms generalize independent component analysis to the case of convolutive mixtures and exhibit superior performance on instantaneous mixtures. An extension of the relative-gradient concept to the spatiotemporal case leads to fast and efficient learning rules with equivariant properties. Our approach can incorporate information about the mixing situation when available, resulting in a "semiblind" separation method. The spatiotemporal redundancy reduction performed by our algorithms is shown to be equivalent to information-rate maximization through a simple network. We illustrate the performance of these algorithms by successfully separating instantaneous and convolutive mixtures of speech and noise signals.

Algorithms↗

PubMed1998

Effects of the attenuation map used in the Chang algorithm on quantitative SPECT results.

OBJECTIVE: This study examined the effects on SPECT quantitation caused by erroneous size and position of the attenuation map and inaccurate pixel size used in the Chang algorithm. METHODS: Projection data of a three-dimensional head phantom were simulated with a uniform attenuation coefficient of 0.15/cm for the inside of the phantom. Images were reconstructed using the filtered backprojection algorithm without attenuation compensation and the Chang algorithm with different attenuation maps. Quantitative comparison then was performed between the reconstructed images and the phantom. RESULTS: The pixel values obtained for noisy data by using the first-order Chang algorithm with an accurate attenuation map were less than 10% different from the true values and the left-right asymmetry was under 5%. Small errors in the geometric parameters of the attenuation map, however, caused considerable quantitative inaccuracy in the reconstructed image. For example, a 0.64-cm error in the size of the map caused 10% deviation from the true value and a 0.64-cm shift of the position of the map towards the left produced 10% left-right pixel value asymmetry. CONCLUSION: The accuracy of the Chang algorithm critically depends on the geometric parameters. For a uniform attenuator with symmetric geometry, such as the human brain, a true left-right symmetry in the pixel value can be altered significantly by a small error in the geometric parameters, while symmetry can be maintained with no attenuation compensation.

Algorithms↗

PubMed1998

Integral and shell-MIP display algorithms in MR and CT three-dimensional models of the brain surface.

BACKGROUND AND PURPOSE: Our purpose was to demonstrate the use of integral and shell maximum intensity projection (shell-MIP) display algorithms in the 3-D CT and MR depiction of cerebral gyral and surface venous anatomy and disorders. These new algorithms are compared against MIP and shaded-surface-display (SSD) algorithms. METHODS: Integral and shell-MIP displays were generated from a specified number of proximal surface voxel layers in a 3-D model. Algorithmic models were compared on nine contrast-enhanced spoiled gradient-recalled acquisition in a steady state (SPGR) MR venograms for brain surface anatomic identification and detail. Seven CT venograms were compared for conspicuity of filling defects. Twelve contrast-enhanced preoperative planning 3-D MR models were rated for neurosurgical utility. RESULTS: A shell-MIP score of 7.00 and an integral score of 6.78 represented the highest mean subjective MR gyral quality (1-10 scale) followed by an SSD score of 3.89 and an MIP score of 1.06. Mean confidence scores for MR central sulcus identification (1-10 scale) were shell-MIP, 7.67; integral, 7.00; SSD, 3.22; and MIP, 1.00. Mean superficial venous quality MR ratings (1-10 scale) were shell-MIP, 8.22; MIP, 7.39; integral, 7.00; and SSD, 3.72. The mean number of cortical veins draining into each side of the superior sagittal sinus on MR was as follows: MIP, 6.19; integral, 6.06; shell-MIP, 5.94; and SSD, 3.81. Mean confidence scores for filling defect identification on CT venograms (1-5 scale) revealed a shell-MIP score of 4.36 and an integral score of 4.29 to be superior to a MIP score of 3.00. In selected cases, 3-D presurgical planning, prior to tumor resection, was clinically useful. CONCLUSION: Integral and shell-MIP are useful 3-D display algorithms for simultaneous display of superficial cerebral veins and gyri on MR images and of thrombosis on CT venograms.

Algorithms↗

PubMed1998

Genetic algorithms for protein threading.

Despite many years of efforts, a direct prediction of protein structure from sequence is still not possible. As a result, in the last few years researchers have started to address the "inverse folding problem": Identifying and aligning a sequence to the fold with which it is most compatible, a process known as "threading". In two meetings in which protein folding predictions were objectively evaluated, it became clear that threading as a concept promises a real breakthrough, but that much improvement is still needed in the technique itself. Threading is a NP-hard problem, and thus no general polynomial solution can be expected. Still a practical approach with demonstrated ability to find optimal solutions in many cases, and acceptable solutions in other cases, is needed. We applied the technique of Genetic Algorithms in order to significantly improve the ability of threading algorithms to find the optimal alignment of a sequence to a structure, i.e. the alignment with the minimum free energy. A major progress reported here is the design of a representation of the threading alignment as a string of fixed length. With this representation validation of alignments and genetic operators are effectively implemented. Appropriate data structure and parameters have been selected. It is shown that Genetic Algorithm threading is effective and is able to find the optimal alignment in a few test cases. Furthermore, the described algorithm is shown to perform well even without pre-definition of core elements. Existing threading methods are dependent on such constraints to make their calculations feasible. But the concept of core elements is inherently arbitrary and should be avoided if possible. While a rigorous proof is hard to submit yet an, we present indications that indeed Genetic Algorithm threading is capable of finding consistently good solutions of full alignments in search spaces of size up to 10(70).

Algorithms↗

PubMed1998

Impact of reorientation algorithms on quantitative myocardial SPECT perfusion imaging.

UNLABELLED: In myocardial SPECT perfusion imaging, reorientation algorithms from transaxial image planes are used to generate short- and long-axis views of myocardial tracer uptake. We performed phantom experiments with 201Tl to delineate how image reorientation affects the results of quantitative image analysis. METHODS: Thirty consecutive patient studies were analyzed to characterize the distribution of the angle of reorientation in a clinical setting. Short-axis SPECT images of a cardiac phantom with and without a 180 degrees cold-spot insert were reconstructed with three different backprojection filters (ramp, Metz and Butterworth) and reoriented through different angles ranging from 45 degrees to 89 degrees. Four interpolation algorithms were used to calculate from the transaxial images the pixel values of the reoriented images: (a) a simple interpolator that averages the pixel values of the eight neighboring pixels of the transaxial image; (b) a three-dimensional linear interpolator; (c) a hybrid interpolator that combines a two-dimensional linear in-plane with a one-dimensional cubic across-plane interpolation; and (d) a three-dimensional cubic convolution interpolator. Images were reoriented twice with opposite angles so that the original and the reoriented images could be directly compared. Circumferential profile analysis was applied to determine the root mean square error of corresponding profiles and the difference of the extent and the severity of perfusion defects. Single and multivariate analyses of variance (ANOVA) were used to compare the effects of the reorientation angle, the backprojection filter and the interpolation algorithm. RESULTS: In the clinical studies, the angle between the transaxial and reoriented images was 75 degrees +/- 10 degrees (s.d.). In 48 phantom experiments, multivariate ANOVA demonstrated that the backprojection filter and the interpolation algorithm significantly affect the circumferential profiles and the extent and severity of a perfusion defect (p < 0.05). In contrast, the angle of reorientation was not a significant factor (p = ns). By univariate analysis, the three-dimensional cubic interpolator was associated with significantly (p < 0.05) less error than the simple and three-dimensional linear algorithms. Relative computation times (simple interpolator = 100%) were 119% for the three-dimensional linear, 136% for the hybrid and 243% for the three-dimensional cubic interpolator. CONCLUSION: For quantitative analysis of myocardial SPECT perfusion images, a Metz filter for filtered backprojection in combination with a three-dimensional cubic convolution interpolation for image reorientation appears to offer improved accuracy.

Algorithms↗

PubMed1980

Treatment of surgical emergencies with and without an algorithm.

A patient care algorithm was developed for resuscitation of patients entering the surgical emergency department with hypotension. The diagnostic workup, monitoring, and therapy were progressively excalated according to admission blood pressure and responses to therapeutic interventions. The branching-chain logic is ideally suited for rapid decision making in emergency conditions where the need is most urgent, the time constraints are most severe, and the potential improvements in terms of patient salvage are greatest. Preliminary results from these ongoing clinical trials indicate that (1) physicians can and will use an algorithm for emergency medical service resuscitation; (2) in a university hospital with a large emergency service and a commitment to emergency care, the physicians using the aigorithm performed as well as and in some instances better than those not using the algorithm; and (3) the use of the algorithm may prevent delays in resuscitation and lead to less morbidity and mortality. Thus, we conclude that the algorithm helps to organize emergency care, establish standards, and improve care.

Adult↗

PubMed1980

Evaluation of the S phase distribution of flow cytometric DNA histograms by autoradiography and computer algorithms.

Cell sorting and tritiated thymidine autoradiography were used to define the distribution of S phase cells in flow cytometric DNA histograms obtained from exponential mouse lymphoma cells (L5178Y). The numbers of labeled S phase cells, autoradiographically determined from cells sorted at 2-channel intervals in the G1/early S and late S/G2M regions of the histogram, were compared with the numbers of computed S phase cells in comparable 2-channel intervals as predicted by several computer algorithms used to extract cell cycle phase distributions from DNA histograms. Polynomial and multirectangle algorithms gave computed estimates of total %S in close agreement with the tritiated thymidine labeling index for the cell population, while multi-Gaussian algorithms underestimated %S. Interval autoradiographic and algorithm studies confirmed these results in that no significant differences were found between the autoradiographic S phase distribution and S phase distributions calculated by the polynomial and multirectangle models. However, S phase cells were significantly underestimated in G1/early S by a constrained multi-Gaussian model and in both G1/early S and late S/G2 by an unconstrained multi-Gaussian model. For the particular cell line (L5178Y), staining protocol (mithramycin following ethanol fixation) and instrumentation (Coulter TPS-2 cell sorter) used in this study, close agreement between computed %S and tritiated thymidine labeling index was found to be a reliable indicator of an algorithm's success in resolving S phase cells in the G1/S and S/G2 transition regions of the DNA histograms.

Animals↗

PubMed2003

An improved algorithm to locate critical points in a 3D scalar field as implemented in the program MORPHY.

A new algorithm for location of the critical points in general scalar fields is described. The new method has been developed as part of an on-going process to exploit the topologic analysis of general 3D scalar fields. Part of this process involves the use of topologic information to seed the critical point search algorithm. The continuing move away from topologic studies of just the electron density requires more general algorithms and the ability to easily "plug in" new functions, for example, the Laplacian of the electron density ( triangle down (2)rho), the Electron Localisation Function (ELF), the Localised Orbital Locator (LOL), the Lennard-Jones function (LJF), as well as any new functions that may be proposed in the future. Another important aspect of the current algorithm is the retention of all possible intermediate information, for example, the paths describing the connectivity of critical points, as well as an ability to restart searches, something that becomes increasingly important when analysing larger systems. This new algorithm represents a core part of a new local version of the MORPHY code. We distinguish nine universal types of gradient paths.

Journal Article↗

PubMed2002

A subspace time-domain algorithm for automated NMR spectral normalization.

Recently, two methods have been proposed for quantitatively comparing NMR spectra of control and treated samples, in order to examine the possible occurring variations in cell metabolism and/or structure in response to numerous physical, chemical, and biological agents. These methods are the maximum superposition normalization algorithm (MaSNAl) and the minimum rank normalization algorithm (MiRaNAl). In this paper a new subspace-based time-domain normalization algorithm, denoted by SuTdNAl (subspace time-domain normalization algorithm), is presented. By the determination of the intersection of the column spaces of two Hankel matrices, the common signal poles and further on the components having proportionally varying amplitudes are detected. The method has the advantage that it is computationally less intensive than the MaSNAl and the MiRaNAl. Furthermore, no approximate estimate of the normalization factor is required. The algorithm was tested by Monte Carlo simulations on a set of simulation signals. It was shown that the SuTdNAl has a statistical performance similar to that of the MiRaNAl, which itself is an improvement over the MaSNAl. Furthermore, two samples of known contents are compared with the MiRaNAl, the SuTdNAl, and an older method using a standard. Finally, the SuTdNAl is tested on a realistic simulation example derived from an in vitro measurement on cells.

Journal Article↗

PubMed1986

An algorithm for comparing two-dimensional electrophoretic gels, with particular reference to the study of mutation.

An algorithm dedicated to the detection of presumed mutational events involving the polypeptides displayed with two-dimensional polyacrylamide gel electrophoresis has been described. Because of the large number of gels necessary in most studies of mutation, the algorithm has been designed to minimize operator intervention in its execution. The basic principle involves a comparison of the graph structures of the gels of a father, mother, and one or more children, searching for protein spots in the child not found in either parent. These so-called "orphan" spots are considered a probable manifestation of mutation only after other possible causes of such an isolated event have been excluded as rigorously as possible. At present, the analysis of gels prepared from a platelet or erythrocyte lysate yields about 2% "false-positive" findings, i.e., results in the incorrect designation of a unique spot in a child. These errors can be disposed of by technician intervention. In an experiment designed to simulate the occurrence of mutational events, the algorithm operated with 70% accuracy. Most of the "errors" ("false negatives") occurred when the position of the simulated mutant polypeptide coincided in whole or part with that of a preexisting polypeptide, resulting in a class of mutation not detectable by the eye either. With correction for this fact, the accuracy was 84%. Possible improvements in the algorithm which would substantially increase accuracy have been discussed at some length, as have some ideas as to how to manage the large body of data resulting from the operation of the algorithm. A murine experiment designed to validate the approach has been outlined.

Blood Proteins↗

PubMed1984

A synaptic modification algorithm in consideration of the generation of rhythmic oscillation in a ring neural network.

In consideration of the generation of bursts of nerve impulses (that is, rhythmic oscillation in impulse density) in the ring neural network, a synaptic modification algorithm is newly proposed. Rhythmic oscillation generally occurs in the regular ring network with feedback inhibition and in fact such signals can be observed in the real nervous system. Since, however, various additional connections can cause a disturbance which easily extinguishes the rhythmic oscillation in the network, some function for maintaining the rhythmic oscillation is to be expected to exist in the synapses if such signals play an important part in the nervous system. Our preliminary investigation into the rhythmic oscillation in the regular ring network has led to the selection of the parameters, that is, the average membrane potential (AMP) and the average impulse density (AID) in the synaptic modification algorithm, where the decrease of synaptic strength is supposed to be essential. This synaptic modification algorithm using AMP and AID enables both the rhythmic oscillation and the nonoscillatory state to be dealt with in the algorithm without distinction. Simulation demonstrates cases in which the algorithm catches and holds the rhythmic oscillation in the disturbed ring network where the rhythmic oscillation was previously extinguished.

Animals↗

PubMed1984

An automated algorithm for radionuclide angiocardiographic quantitation of circulatory shunting.

Circulatory shunting may be quantitated by analysis of time-activity curves obtained from radionuclide angiocardiography. A new automated algorithm for performing this analysis is proposed. The algorithm uses mathematical deconvolution techniques to increase the temporal separation of the components of this curve and thereby improves the accuracy of the analysis. The stability of the algorithm to random data errors was assessed by experiments on simulated time-activity curves degraded with pseudorandom noise. Excellent performance was obtained on a set of test problems previously used in the literature. The algorithm was used to quantitate left-to-right shunting in patients undergoing radionuclide angiocardiography during cardiac catheterization. A strong correlation (r = 0.96) was found between pulmonary to systemic flow ratios (Qp:Qs) obtained using the algorithm on radionuclide angiocardiographic data and Qp:Qs values obtained by oximetry at cardiac catheterization.

Adolescent↗

PubMed1983

Algorithms for processing spatial information.

Pairs of stimuli taken from a psychometric measure of spatial aptitude were shown to 9-year-olds, 13-year-olds, and adults. The stimuli in pairs were (a) either identical or mirror images, and (b) presented in orientations that differed by 0-150 degrees. Individuals judged, as rapidly as possible, if the stimuli in a pair would be identical or mirror images if presented at the same orientation. In Experiment 1, in which the stimuli were letter-like characters, at all ages most persons solved the problems using an algorithm in which an individual encodes the stimuli in working memory, mentally rotates one stimulus to the orientation of the other, compares them to determine if they are identical, and responds. In Experiment 2, the stimuli were multielement flags; here, the modal algorithm for both 9- and 13-year-olds differed from the previously described algorithm in that if the comparison process revealed that the stimuli were dissimilar, individuals did not respond immediately, but continued processing until a self-imposed deadline was reached. Among adults, the modal algorithm was the same one used in Experiment 1. Results are discussed in terms of the roles of encoding in contributing to the use of a particular algorithm.

Adolescent↗