SEARCH · PubMed Health

Results for “Clustering 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 1,387 records · Page 77Linked to original sources

PubMed1996

Pulsatile release and circadian rhythms of thyrotropin and prolactin in children with growth hormone deficiency.

We have measured mean concentrations and have appraised the pulsatile nature of thyrotropin (TSH) and prolactin (PRL) release in children with classical GH deficiency (GHD; n = 4) and neurosecretory GH dysfunction (NSD; n = 4) and have compared the results with those obtained in children with constitutional delay (control; n = 4). Blood samples were obtained at 20-min intervals for 24 h. Pulse analysis of TSH and PRL was undertaken using the Cluster pulse detection algorithm. Circadian rhythmicity of TSH and PRL was assessed using cosinor analysis. The mean 24-h concentration of GH in the control subjects was significantly higher than that obtained in the GHD and NSD groups. With regard to TSH, the mean serum concentration in the GHD and NSD group were higher than that of the control subjects. This augmentation reflects TSH pulses of large amplitude and area, and a higher interpulse valley mean rather than a difference in peak number or peak duration. No differences in mean PRL concentration or characteristics of PRL pulses were found between the control and GHD and NSD subjects. When the 24 h data sets were divided into day (0800-2000 h) and night (2000-0800 h), the mean nighttime TSH concentration was higher than the daytime concentration in the control, GHD, and NSD groups. Although there were no day versus night differences in TSH pulse frequency in either group, peak amplitude, area, and interpulse valley means were increased during the night in the control group, and peak area, duration, and amplitude mean in the NSD group. The nighttime mean PRL concentrations in the control, GHD, and NSD subjects were higher than those found during the day. This increase was accounted for by increases in PRL peak amplitude, area in the control group, and peak area, amplitude, and interpulse valley mean in the GHD and NSD groups. Cosinor analysis of the 24-h TSH and PRL data revealed clear circadian rhythmicity in all groups of subjects. These data suggest that GHD and NSD are associated with an increase in pulsatile TSH secretion due to an increase in pulse amplitude and interpulse valley mean.

Adolescent↗

PubMed1990

Reliability of estimates of pulsatile characteristics of luteinizing hormone and growth hormone release in women.

The test-retest reliability of estimates of pulsatile LH and GH release was evaluated in 23 eumenorrheic women during the early follicular phase of the menstrual cycle. Each subject was studied during two successive or near-successive menstrual cycles by repetitive blood sampling every 10 min for 24 h. Pulsatile parameters for LH and GH release were identified and characterized using the Cluster pulse detection algorithm. For LH, no significant differences existed in any parameter mean between the two 24-h admissions. Correlation coefficients for consecutive 24-h studies ranged from r = 0.22 (P less than 0.32) for number of LH peaks to r = 0.79 (P less than 0.0001) for 24-h integrated LH values (area under the concentration vs. time curve). No significant mean differences in any parameter were observed for consecutive 24-h GH evaluations. Correlation coefficients for 24-h GH ranged from r = 0.25 (P less than 0.34) for nadir to r = 0.71 (P less than 0.002) for incremental peak increase. Cosinor analysis was used to determine significant 24-h variations in LH and GH concentrations. Statistically significant differences existed between admissions for the amplitude of the nyctohemeral LH rhythm and its acrophase (time at which maximal hormone value was attained), but no mean differences were found for mesor (mean concentration). Correlation coefficients for LH were r = 0.10 (P less than 0.65), r = 0.43 (P less than 0.08), and r = 0.78 (P less than 0.0001) for phase, amplitude, and mesor, respectively. No significant mean differences existed for any parameter of nyctohemeral GH rhythms. Correlation coefficients were r = -0.18 (P less than 0.52), r = 0.49 (P less than 0.72), and r = 0.14 (P less than 0.80) for 24-h GH amplitude, phase, and mesor, respectively. We conclude that comparisons of mean and integrated LH and GH concentrations over a 24-h interval in the early follicular phase of the menstrual cycle are reliable; however, certain pulsatile properties responsible for the achievement of the mean daily concentrations of LH and GH may be nonuniform from menstrual cycle to menstrual cycle. In addition, nonuniformities may exist in the nyctohemeral rhythms of serum concentrations of LH and GH in the adult woman between cycles when a single 24-h time series is the basis for the analysis.

Adult↗

PubMed1993

Twenty-four-hour concentration profiles of gonadotropin and estradiol (E2) in prepubertal and early pubertal girls: the diurnal rise of E2 is opposite the nocturnal rise of gonadotropin.

To investigate the detailed pattern of circulating gonadotropin and estradiol (E2) concentrations around the onset of puberty, plasma gonadotropin and E2 were measured at 20-min intervals for 24 h in seven prepubertal and six early pubertal normal short girls. The hormone concentrations obtained were analyzed by Cluster pulse detection algorithm, cosinor analysis, and cross-correlation analysis. All subjects showed spontaneous LH and FSH pulses, and six early pubertal girls showed spontaneous E2 pulses. Cosinor analysis revealed 24-h LH rhythms in all subjects except two early pubertal girls and 24-h FSH rhythms in all subjects except one early pubertal girl. The acrophases (clocktime for maximal value) in the 24-h rhythm of LH and FSH were both found in the late hours of sleep. All subjects except three prepubertal girls showed significant 24-h E2 rhythms. In contrast to the 24-h LH and FSH rhythms, the acrophase of the 24-h E2 rhythm was found in the daytime waking period. Cross-correlation analysis demonstrated significant positive cross-correlations between LH and E2 that were maximum at an E2 lag of 5.7-9.3 h in three of the six early pubertal girls. In conclusion, the E2 concentration profiles in girls around the onset of puberty show marked 24-h rhythm, with acrophase during the daytime waking period. There exists a 5.7- to 9.3-h time lag between LH and E2 time series, and this long time lag might correspond to the time required for aromatization for E2 synthesis.

Adolescent↗

PubMed1996

A longitudinal assessment of hormonal and physical alterations during normal puberty in boys. III. The neuroendocrine growth hormone axis during late prepuberty.

Cross-sectional studies demonstrating that physiological GH secretion varies widely among normally and slowly growing children as well as adults have created uncertainty over the definition of normal GH secretion. Furthermore, recent data indicate that the pattern of GH released under identical physiological conditions may be unique for each individual, and suggest that normality may be an individually defined condition. To investigate and develop further this concept and to avoid the confounding effects of fluctuating gonadal steroid hormone levels, we chose as a model normally growing prepubertal boys and performed a longitudinal assessment of spontaneous GH release characteristics (36 24-h GH secretion studies in 9 boys over 9-19 months). Assessment of serum GH pulse characteristics was accomplished using the Cluster pulse detection algorithm. Characteristics of underlying pituitary GH secretory events were estimated by multiple parameter deconvolution analysis. Approximate entropy was used to quantify the serial regularity or orderliness of GH release over 24 h. Among the group data, mean 24-h GH concentrations spanned a range of more than 4-fold (1.6-7.0 micrograms/L). The intersubject coefficient of variation (CV) was 41%. In contrast, values from individuals exhibited much less variability, not only for mean 24-h GH level (CV = 25 +/- 4%; P = < 0.02), but also for all assessed component GH pulse properties (P < 0.01) vs. intersubject values). Similarly, the estimated daily GH production rate, the calculated GH half-lives, and all parameters of GH secretory events varied much less for intraindividual compared to interindividual values. The sizes of the serum GH pulses gave rise to the greatest differences in overall serum GH level among individuals, as demonstrated by the large within-subject CV (50%). The most constant pulse characteristic among subjects was that of 24-h GH pulse frequency (intersubject CV = 30%). Approximate entropy estimates disclosed high within-subject consistency (mean CV = 15%). Several aspects of GH secretion and serum concentrations varied inversely with the subject's mean age-adjusted body mass index, including the 24-h GH production rate (r = -0.67; P < 0.05), the GH secretory burst amplitude (r = -0.73; P = 0.026), the mean serum GH pulse amplitude (r = -0.79; P = 0.011), and the sum of the GH pulse amplitudes (r = -0.66; P = 0.05). By contrast, no consistent pattern of increase or decrease in serum GH concentrations with advancing age was detectable during the prepubertal period. These data suggest that during prepuberty, 1) individual boys regulate daily GH secretion within relatively confined limits characteristic for that individual, much narrower than the range present in the larger population; 2) differences in mean 24-h GH levels among normally growing prepubertal boys arise primarily from differences in GH pulse size; 3) differences in body composition, as indicated by the body mass index, may influence the GH secretion range characteristic of each individual; and 4) there is no consistent change in mean 24-h GH concentrations or the orderliness (approximate entropy) of the GH release process ever time during late prepuberty as the onset of puberty approaches.

Adult↗

PubMed1999

[Reading screening mammograms with the help of neural networks].

With digital mammography it is possible to assist radiologists in breast cancer screening with computers to improve their reading performance. The need for this has been demonstrated by studies showing a large variability in skill of radiologists reading mammograms. Moreover, retrospective studies show that a significant number of cancers are clearly visible on earlier screening mammograms, even for 'trained' computers. Methods for automated detection of breast cancer in mammograms often use artificial neural networks. These are 'trained' to recognize abnormal mammographic areas using a large database of known cases. For detection of microcalcification clusters very reliable algorithms exist, with such high sensitivity that radiologists can limit their search to areas that have been marked 'suspect' by the computer. The development of methods to recognize malignant masses is much more difficult, but ample progress has been achieved in recent years.

Adult↗

PubMed2005

Random forest similarity for protein-protein interaction prediction from multiple sources.

One of the most important, but often ignored, parts of any clustering and classification algorithm is the computation of the similarity matrix. This is especially important when integrating high throughput biological data sources because of the high noise rates and the many missing values. In this paper we present a new method to compute such similarities for the task of classifying pairs of proteins as interacting or not. Our method uses direct and indirect information about interaction pairs to constructs a random forest (a collection of decision tress) from a training set. The resulting forest is used to determine the similarity between protein pairs and this similarity is used by a classification algorithm (a modified kNN) to classify protein pairs. Testing the algorithm on yeast data indicates that it is able to improve coverage to 20% of interacting pairs with a false positive rate of 50%. These results compare favorably with all previously suggested methods for this task indicating the importance of robust similarity estimates.

Algorithms↗

PubMed1992

New and future developments in screen-film mammography equipment and techniques.

Two improvements in mammography equipment during the last 4 years will greatly affect the practice of mammography during the next decade: dose reduction and improved testing equipment for monitoring the quality of mammography. The increased use of digital radiography has stimulated studies comparing digital mammography with screen-film mammography. Some digital algorithms for detecting clusters of calcifications may have application in screening mammography within the next decade.

Female↗

PubMed1996

Automatic sorting of multiple unit neuronal signals in the presence of anisotropic and non-Gaussian variability.

Neuronal noise sources and systematic variability in the shape of a spike limit the ability to sort multiple unit waveforms recorded from nervous tissue into their single neuron constituents. Here we present a procedure to efficiently sort spikes in the presence of noise that is anisotropic, i.e., dominated by particular frequencies, and whose amplitude distribution may be non-Gaussian, such as occurs when spike waveforms are a function of interspike interval. Our algorithm uses a hierarchical clustering scheme. First, multiple unit records are sorted into an overly large number of clusters by recursive bisection. Second, these clusters are progressively aggregated into a minimal set of putative single units based on both similarities of spike shape as well as the statistics of spike arrival times, such as imposed by the refractory period. We apply the algorithm to waveforms recorded with chronically implanted micro-wire stereotrodes from neocortex of behaving rat. Natural extension of the algorithm may be used to cluster spike waveforms from records with many input channels, such as those obtained with tetrodes and multiple site optical techniques.

Action Potentials↗

PubMedDate not supplied

[Classification approach to prognostication of the severity of bronchial asthma].

To define the severity of bronchial asthma, to predict its subsequent course, and to prescribe adequate glucocorticosteroid therapy, a mathematical model based on the cluster analysis was used. The authors' algorithm of the classification of bronchial asthma was based on the cluster analysis was used. The authors' algorithm of the classification of bronchial asthma was based on the mean distance value between the objects characterizing patients' status and hyperspatial signs. Each patient was described using a set of 15 clinical, functional, and immunological indices which had a meaning of signs describing the objects. Three clusters corresponding to the groups identified a priori up to two patients were formed to make the classification by the algorithm proposed. The findings make it possible to efficiently analyze the baseline indices in a particular patient from his entrance into or highest relationship to a definite cluster. This may define the severity of bronchial obstruction and predict its progression.

Algorithms↗

PubMedDate not supplied

Biclustering algorithms for biological data analysis: a survey.

A large number of clustering approaches have been proposed for the analysis of gene expression data obtained from microarray experiments. However, the results from the application of standard clustering methods to genes are limited. This limitation is imposed by the existence of a number of experimental conditions where the activity of genes is uncorrelated. A similar limitation exists when clustering of conditions is performed. For this reason, a number of algorithms that perform simultaneous clustering on the row and column dimensions of the data matrix has been proposed. The goal is to find submatrices, that is, subgroups of genes and subgroups of conditions, where the genes exhibit highly correlated activities for every condition. In this paper, we refer to this class of algorithms as biclustering. Biclustering is also referred in the literature as coclustering and direct clustering, among others names, and has also been used in fields such as information retrieval and data mining. In this comprehensive survey, we analyze a large number of existing approaches to biclustering, and classify them in accordance with the type of biclusters they can find, the patterns of biclusters that are discovered, the methods used to perform the search, the approaches used to evaluate the solution, and the target applications.

Algorithms↗

PubMed2001

Extraction of correlated gene clusters by multiple graph comparison.

This paper presents a new method to extract a set of correlated genes with respect to multiple biological features. Relationships among genes on a specific feature are encoded as a graph structure whose nodes correspond to genes. For example, the genome is a graph representing positional correlations of genes on the chromosome, the pathway is a graph representing functional correlations of gene products, and the expression profile is a graph representing gene expression similarities. When a set of genes are localized in a single graph, such as a gene cluster on the chromosome, an enzyme cluster in the metabolic pathway, or a set of coexpressed genes in the microarray gene expression profile, this may suggest a functional link among those genes. The functional link would become stronger when the clusters are correlated; namely, when a set of corresponding genes form clusters in multiple graphs. The newly introduced heuristic algorithm extracts such correlated gene clusters as isomorphic subgraphs in multiple graphs by using inter-graph links that are defined based on biological relevance. Using the method, we found E.coli correlated gene clusters in which genes are related with respect to the positions in the genome and the metabolic pathway, as well as the 3D structural similarity. We also analyzed protein-protein interaction data by two-hybrid experiments and gene coexpression data by microarrays in S.cerevisiae, and estimated the possibility of utilizing our method for screening the datasets that are likely to contain many false positive relations.

Algorithms↗

PubMed2005

Mapping the dimensionality, density and topology of data: the growing adaptive neural gas.

Self-organized maps are commonly applied for tasks of cluster analysis, vector quantization or interpolation. The artificial neural network model introduced in this paper is a hybrid model of the growing neural gas model introduced by Fritzke (Fritzke, in Advances in Neural Information Processing Systems 7, MIT Press, Cambridge MA, 1995) and the adaptive resolution clustering modification for self-organized maps proposed by Firenze (Firenze et al., in International Conference on Artificial Neural Networks, Springer-Verlag, London, 1994). The hybrid model is capable of mapping the distribution, dimensionality and topology of the input data. It has a local performance measure that enables the network to terminate growing in areas of the input space that is mapped by units reaching a performance goal. Therefore the network can accurately map clusters of data appearing on different scales of density. The capabilities of the algorithm are tested using simulated datasets with similar spatial spread but different local density distributions, and a simulated multivariate MR dataset of an anatomical human brain phantom with mild multiple sclerosis lesions. These tests demonstrate the advantages of the model compared to the growing neural gas algorithm when adaptive mapping of areas with low sample density is desirable.

Algorithms↗

PubMed2003

Analysis of cytokine functions in graft rejection by gene expression profiles.

BACKGROUND: The function of interferon (IFN)gamma in the regulation of the immune response after allogeneic transplantation is still poorly understood. Previous studies have suggested that IFNgamma can promote rejection and be important in tolerance induction. METHODS: To analyze the various IFNgamma-dependent functions in terms of T helpers 1 and 2 responses during rejection, we investigated mice deficient in the transcription factors (signal transducer of activated T cells [STAT]4 and 6) and IFNgamma in fully major histocompatibility complex-mismatched vascularized cardiac transplants. Serum levels of the cytokines tumor necrosis factor-alpha, IFNgamma, and interleukin (IL)-1beta were evaluated by enzyme-linked immunosorbent assay, and the graft-infiltrating cells were examined by immunohistochemical staining. To analyze a large panel of immune parameters, we determined the expression of chemokines, chemokine receptors, and clusters of differentiation markers by RNAase protection assays. The data were analyzed with algorithms that generated hierarchic clustering dendrograms. Also, the expression profiles of individual genes were determined with self-organizing maps. RESULTS: Our data show that both the STAT4- and STAT6-deficient groups have statistically prolonged graft survival (P<0.04 and P<0.01). Despite the absence of prolongation of graft survival in the IFNgamma-deficient group, our analysis of variance data show that more genes (18) were modulated in the IFNgamma-deficient group compared with the other two STAT4- and STAT6-deficient groups (five each). CONCLUSIONS: Our results indicate that IFNgamma plays a distinct role in the modulation of gene expression that includes STAT4-independent mechanisms. Our study identifies eight genes (IL-1beta, IL-1RA, macrophage inflammatory protein-1beta, monocyte chemoattractant protein-1, CC-chemokine receptor (CCR)-1, CCR2, CCR5, and F4/80) that are highly expressed in all of our experimental groups. Thus, these genes become candidates for essential functions during rejection.

Analysis of Variance↗

PubMed2001

Textquest: document clustering of Medline abstracts for concept discovery in molecular biology.

We present an algorithm for large-scale document clustering of biological text, obtained from Medline abstracts. The algorithm is based on statistical treatment of terms, stemming, the idea of a 'go-list', unsupervised machine learning and graph layout optimization. The method is flexible and robust, controlled by a small number of parameter values. Experiments show that the resulting document clusters are meaningful as assessed by cluster-specific terms. Despite the statistical nature of the approach, with minimal semantic analysis, the terms provide a shallow description of the document corpus and support concept discovery.

Abstracting and Indexing↗

PubMed2000

Backbone cluster identification in proteins by a graph theoretical method.

A graph theoretical algorithm has been developed to identify backbone clusters of residues in proteins. The identified clusters show protein sites with the highest degree of interactions. An adjacency matrix is constructed from the non-bonded connectivity information in proteins. The diagonalization of such a matrix yields eigenvalues and eigenvectors, which contain the information on clusters. In graph theory, distinct clusters can be obtained from the second lowest eigenvector components of the matrix. However, in an interconnected graph, all the points appear as one single cluster. We have developed a method of identifying highly interacting centers (clusters) in proteins by truncating the vector components of high eigenvalues. This paper presents in detail the method adopted for identifying backbone clusters and the application of the algorithm to families of proteins like RNase-A and globin. The objective of this study was to show the efficiency of the algorithm as well as to detect conserved or similar backbone packing regions in a particular protein family. Three clusters in topologically similar regions in the case of the RNase-A family and three clusters around the porphyrin ring in the globin family were observed. The predicted clusters are consistent with the features of the family of proteins such as the topology and packing density. The method can be applied to problems such as identification of domains and recognition of structural similarities in proteins.

Algorithms↗

PubMed2004

Computer-aided diagnosis of mammographic microcalcification clusters.

Computer-aided diagnosis techniques in medical imaging are developed for the automated differentiation between benign and malignant lesions and go beyond computer-aided detection by providing cancer likelihood for a detected lesion given image and/or patient characteristics. The goal of this study was the development and evaluation of a computer-aided detection and diagnosis algorithm for mammographic calcification clusters. The emphasis was on the diagnostic component, although the algorithm included automated detection, segmentation, and classification steps based on wavelet filters and artificial neural networks. Classification features were selected primarily from descriptors of the morphology of the individual calcifications and the distribution of the cluster. Thirteen such descriptors were selected and, combined with patient's age, were given as inputs to the network. The features were ranked and evaluated for the classification of 100 high-resolution, digitized mammograms containing biopsy-proven, benign and malignant calcification clusters. The classification performance of the algorithm reached a 100% sensitivity for a specificity of 85% (receiver operating characteristic area index Az = 0.98 +/- 0.01). Tests of the algorithm under various conditions showed that the selected features were robust morphological and distributional descriptors, relatively insensitive to segmentation and detection errors such as false positive signals. The algorithm could exceed the performance of a similar visual analysis system that was used as basis for development and, combined with a simple image standardization process, could be applied to images from different imaging systems and film digitizers with similar sensitivity and specificity rates.

Algorithms↗

PubMed1983

Numerical taxonomy of Vibrio cholerae and related species isolated from areas that are endemic and nonendemic for cholera.

A total of 165 strains of vibrios isolated from clinical and environmental sources in the United States, India, and Bangladesh, 11 reference cultures, and 4 duplicated cultures were compared in a numerical taxonomic study using 83 unit characters. Similarity between strains was computed by using the simple matching coefficient and the Jaccard coefficient. Strains were clustered by unweighted average linkage and single linkage algorithms. All methods gave similar cluster compositions. The estimated probability of error in the study was obtained from a comparison of the results of duplicated strains and was within acceptable limits. A total of 174 of the 180 organisms studied were divided into eight major clusters. Two clusters were identified as Vibrio cholerae, one as Vibrio mimicus, one as Vibrio parahaemolyticus, three as Vibrio species, and one as Aeromonas hydrophila. The V. mimicus cluster could be further divided into two subclusters, and the major V. cholerae group could be split into seven minor subclusters. Phenotypic traits routinely used to identify clinical isolates of V. cholerae can be used to identify environmental V. cholerae isolates. No distinction was found between strains of V. cholerae isolated from regions endemic for cholera and strains from nonendemic regions.

Aeromonas↗

PubMed2005

Unsupervised fuzzy clustering analysis supports behavioral cutoff criteria in an animal model of posttraumatic stress disorder.

BACKGROUND: Unsupervised fuzzy clustering (UFC) analysis is a mathematical technique that groups together objects in the multidimensional feature space according to a specified similarity measurement, thereby yielding clusters of similar data points that can be represented by a set of prototypes or centroids. METHODS: Since clinical studies of mental disorders distinguish between affected and unaffected individuals, we designed an inclusion/exclusion criteria (cutoff behavioral criteria [CBC]) approach for animal behavioral studies. The effect of classifying the study population into clearly affected versus clearly unaffected individuals according to behaviors on two behavioral paradigms was statistically significant. RESULTS: Here the raw data from previous studies were subjected to UFC algorithms as a means of objectively testing the validity of the concept of the CBC for our experimental model. The first UFC algorithm yielded two clearly discrete clusters, found to consist almost exclusively of the exposed animals in the one and unexposed animals in the other. The second algorithm yielded three clusters corresponding to animals designated as clearly affected, partially affected, and clearly unaffected. The algorithm for physiological data in addition to behavioral data failed to elicit discrete clusters. CONCLUSIONS: The UFC analysis yielded data that support the conceptual contention of the CBC and lends additional validity to our previous behavioral studies.

Acoustic Stimulation↗