SEARCH · PubMed Health

Results for “cluster analysis”

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 55 records · Page 3Linked to original sources

PubMed2006

Comparative strategies for using cluster analysis to assess dietary patterns.

OBJECTIVES: To characterize dietary patterns using two different cluster analysis strategies. DESIGN: In this cross-sectional study, diet information was assessed by five 24-hour recalls collected over 10 months. All foods were classified into 24 food subgroups. Demographic, health, and anthropometric data were collected via home visit. SUBJECTS: One hundred seventy-nine community-dwelling adults, aged 66 to 87 years, in rural Pennsylvania. STATISTICAL ANALYSIS: Cluster analysis was performed. RESULTS: The methods differed in the food subgroups that clustered together. Both methods produced clusters that had significant differences in overall diet quality as assessed by Healthy Eating Index (HEI) scores. The clusters with higher HEI scores contained significantly higher amounts of most micronutrients. Both methods consistently clustered subgroups with high energy contribution (eg, fats and oils and dairy desserts) with a lower HEI score. Clusters resulting from the percent energy method were less likely to differentiate fruit and vegetable subgroups. The higher diet quality dietary pattern derived from the number of servings method resulted in more favorable weight status. CONCLUSIONS: Cluster analysis of food subgroups using two different methods on the same data yielded similarities and dissimilarities in dietary patterns. Dietary patterns characterized by the number of servings method of analysis provided stronger association with weight status and was more sensitive to fruit and vegetable intake with regard to a more healthful dietary pattern within this sample. Public health recommendations should evaluate the methodology used to derive dietary patterns.

Aged↗

PubMed1981

The classification of patients into diagnostic groups using cluster analysis.

The use of cluster analysis for the classification of patients into diagnostic groups remains controversial. Further evidence is provided for assessing the use of cluster analysis in comparison with other multivariate methods. It is suggested that, when appropriately used, cluster analysis is a convenient tool for developing empirically based diagnostic groupings and that the frequently stated limitations of the technique should not invalidate the obtained groupings in such cases.

Diagnosis, Differential↗

PubMedDate not supplied

The use of multivariate methods in the identification of subtypes of Alzheimer's disease: a comparison of principal components and cluster analysis.

Two contrasting multivariate statistical methods, viz., principal components analysis (PCA) and cluster analysis were applied to the study of neuropathological variations between cases of Alzheimer's disease (AD). To compare the two methods, 78 cases of AD were analyzed, each characterised by measurements of 47 neuropathological variables. Both methods of analysis revealed significant variations between AD cases. These variations were related primarily to differences in the distribution and abundance of senile plaques (SP) and neurofibrillary tangles (NFT) in the brain. Cluster analysis classified the majority of AD cases into five groups which could represent subtypes of AD. However, PCA suggested that variation between cases was more continuous with no distinct subtypes. Hence, PCA may be a more appropriate method than cluster analysis in the study of neuropathological variations between AD cases.

Aged↗

PubMedDate not supplied

Determination of liver volume from CT scans using histogram cluster analysis.

The histogram cluster analysis procedure (HICAP), which was developed by NASA for processing satellite images, classifies images into discrete clusters of pixels according to one or more arbitrary imaging variables. We incorporated this nonparametric, multivariate procedure in a semiautomatic computer algorithm for calculating total liver volume from CT scans and compared its performance with that of a human observer. Total liver volumes were calculated from CT scans in adult patients by the algorithm and by an experienced radiologist using the trackball controlled cursor at the CT console. Variability in the computer calculated volumes was determined by repeating calculations three times over the course of 3-12 months. Using HICAP in the univariate mode, we calculated total liver volumes from 28 contrast enhanced CT scans in 27 patients. Liver volumes calculated by the semiautomatic and manual methods had a median absolute difference of 3.6% (Vcomputer = 1.08 * Vmanual - 99.52 cc; r2 = 0.99). Median day-to-day variability of the computer calculated volumes was 1.9% (95% confidence interval: 1.3-2.7%). Using HICAP in a bivariate mode to illustrate its ability to incorporate two image features in one analysis, we studied an additional patient and compared total liver volume calculated from the univariate data set defined by the contrast enhanced CT scan with that calculated from the bivariate data set defined by nonenhanced and contrast enhanced CT scans. The HICAP errors were 4.1% in the univariate analysis and 0.4% in the bivariate analysis. It is concluded that this statistical clustering algorithm provides a clinically accurate, repeatable, and feasible method of in vivo liver volume determination.

Adult↗

PubMedDate not supplied

[The users of centers for AIDS information and prevention in the Comunidad Valenciana, Spain: a study based on cluster analysis].

OBJECTIVE: To measure the usefulness of multiple correspondence analysis (MCA) and cluster analysis applied to the epidemiological research of HIV infection. The specific are to explore the relationships between the different variables that characterize the users of the AIDS Information and Prevention Center (CIPS) and to identify clusters of characteristics which in terms of the attendance to these centers, could be considered similar. METHODS: The clinical history the CIPS in the Valencian region in Spain was used as data source. The target population target were intravenous drug users (IDUSs) attending these centers between 1987 and 1994 (n = 6211). Information about socio-demographic and HIV type I infection-related variables (drug use and sexual behaviour) was collected by means of a semistructured questionnaire. A MCA was carried out to obtain a group of quantitative factors that were used in a cluster analysis. RESULTS: A 44.8% HIV type I prevalence was found. Five factors were detected by MCA that explain 51.14% of the total variability, of which sex, age and the usual sexual partner were the variables best explained. Cluster analysis allowed to describe 5 different subgroups of CIPS users according to their socio-demographics characteristics, risk behaviours and serologic status. It is necessary to highlight the categories 1 and 2, which collect the serologic status and the most relevant characteristics of HIV infection. Category I contains users with a negative serology and characterized by being mainly single adolescent men, with a low educational level; they stated that they have no steady sexual partner, do not share syringes and have been intravenous drug users between 3 and 10 years. They mainly come from the city of Alicante. Category 2 contains mainly people that are HIV positive and older. They also share syringes and have been intravenous drug users for a longer time; they have a higher education level and most of them come from the city of Valencia. CONCLUSIONS: The proposed method of analysis was able to characterise the CIPS users, identifying those socio-demographic variables and risk behaviours that are more related to the serologic status. The applicability of these techniques to epidemiologic studies of HIV type I infection is discussed.

Acquired Immunodeficiency Syndrome↗

PubMed1994

Putative dimeric organization of nuclear receptor hormone-binding domains, deduced from hydrophobic cluster analysis.

2D hydrophobic cluster analysis (HCA) protein sequence processing, efficient at low levels of sequence identity, leads to a coherent scheme for the structural organization of the hormone-binding domains (HBDs) of nuclear receptors. The typical serine protease inhibitor (serpin) fold, previously proposed, is confirmed as a likely framework for the hormone-binding domain and leads to a logical dimerization. Furthermore, homo- or hetero-dimerization creates sites where hormone could likely be bound, itself being an active component of the dimerization. This model fulfils many of the biochemical and biological data.

Amino Acid Sequence↗

PubMed2003

Gene expression profiling of resting and activated vascular smooth muscle cells by serial analysis of gene expression and clustering analysis.

Migration and proliferation of vascular smooth muscle cells (SMCs) are key events in atherosclerosis. However, little is known about alterations in gene expression upon transition of the quiescent, contractile SMC to the proliferative SMC. We performed serial analysis of gene expression (SAGE) of cultured, human SMCs, either grown under resting circumstances or activated with an atherogenic stimulus. Analysis of tags, representing 47,209 and 47,259 mRNAs from a library of resting and activated SMCs, respectively, identified 105 tags induced and 52 tags repressed greater than fivefold. To evaluate the relevance in SMC biology of unmatched, regulated tags, we performed hierarchical clustering analysis, based on their expression profiles in public SAGE databases, and clustered these novel genes in distinct groups. The regulation in SMCs was confirmed by Northern blotting for representative genes of these groups. Plasminogen activator inhibitor-2 has not been associated with atherosclerosis before and was localized to atherosclerotic lesions.

Gene Expression↗

PubMed2005

The use and reporting of cluster analysis in health psychology: a review.

PURPOSE: Cluster analysis is a collection of relatively simple descriptive statistical techniques with potential value in health psychology, addressing both theoretical and practical problems. There are many methods of cluster analysis from which to choose, with no clear guidelines to aid researchers. In the absence of guidelines it is likely that methods already reported by published researchers will be adopted, and so clear reporting of statistical methodology, while always important, is particularly crucial with cluster analysis. The aim of this review is to describe and evaluate the reporting of cluster analysis in health psychology publications. METHODS: Electronic searches of 18 health psychology journals identified 59 articles using cluster analysis published between 1984 and 2002. Articles were submitted to systematic evaluation against published criteria for the reporting of cluster analysis. RESULTS: Just 27% of the papers reviewed met all five criteria, although 61% met at least four. Details of the similarity measure and the computer program used were most frequently omitted. Furthermore, while researchers usually reported the procedures employed to determine the number of clusters and to validate the clusters, these procedures were often lacking in rigour, and were reported in insufficient detail for replication. CONCLUSIONS: The reporting of cluster analysis was found to be generally unsatisfactory, with many studies failing to provide enough information to allow replication or the evaluation of the quality of the research. Clear guidelines for conducting and reporting cluster analyses in health psychology are needed.

Behavioral Medicine↗

PubMed1991

The magnocellular and parvocellular divisions of the monkey subthalamic nucleus as revealed by cluster analysis of neuronal sizes.

Cluster analysis of neuronal somal sizes in the subthalamic nucleus of rhesus monkeys from newborn to adult age allows the segregation of two territories with predominance of small and large cells, respectively. The topographic distribution of the 'parvocellular' and 'magnocellular' segments is similar when samples are obtained from coronal, horizontal and sagittal series of sections. The parvocellular component occupies the rostral pole, the entire rostrocaudal extent of the medial tip and dorsomedial border, and probably also the caudal cap. The magnocellular segment is in the central core extending to the ventrolateral border except for the medial tip. These findings and their correlation with the results of other morphologic and physiologic studies allow the following conclusions. (1) The monkey subthalamic nucleus contains at least two differentially distributed cell subpopulations. (2) The magnocellular division is more related to the pallido-subthalamic-pallidal loop involving the lateral pallidal segment. (3) The parvocellular division appears strategically located to control the pallidal output to diencephalic and mesencephalic targets. (4) Cluster analysis can reveal the existence of more than one neuronal population in a particular brain structure where an overall unimodal distribution of cell sizes may suggest the presence of a single type.

Aging↗

PubMedDate not supplied

Cluster analysis: significance, empty space, clustering tendency, non-uniformity. I--Statistical tests on the significance of clusters.

The agglomerative clustering methods and the tests usually applied to evaluate the significance of clusters are critically evaluated. Many clustering techniques can provide erroneous information about the existence of clusters. The single linkage technique is suggested to identify natural, well separated, clusters. The existing statistical tests on the significance of clusters are not satisfactory. A new statistical test, based on the distribution of the distances between the objects and their first nearest neighbor, is presented. The performances of the test are compared with those of the Sneath test and of the variance-ratio test on some artificial and real data sets.

Chemistry Techniques, Analytical↗

PubMedDate not supplied

Cluster analysis: significance, empty space, clustering tendency, non-uniformity. II--Empty Space index.

The here presented Empty Space index (ES) evaluates the fraction of the information space without experimental points, i.e. the space where the distance from an experimental point is significantly larger than the mean distance between the experimental points themselves. ES can be used to eliminate the ambiguity of the some clustering indexes, that aim to evaluate the separation of the data set in clusters, but these clustering indexes are really a mixed measure of clustering, of empty space (the empty space does not necessarily correspond to the break between clusters) and of the degree of uniformity of the objects. The ES index can be used also to correct the MST index, the clustering index based on the distribution of edge lengths in the minimum spanning tree connecting the objects. The corrected MST index seems to be a reliable measure of the clustering degree.

Chemistry Techniques, Analytical↗

PubMed2001

The search for subtypes of DCD: is cluster analysis the answer?

Studies using cluster analysis as a method to identify distinct subtypes of developmental coordination disorder (DCD) have been inconclusive leading some authors to conclude that the method of cluster analysis should be abandoned while others call for the validation of previously defined subtypes. The objective of the current study was to examine the use of cluster analysis as a method of searching for subtypes of DCD to gain a better understanding of how different samples and different measures influence the interpretation of results. The paper provides a detailed review of three commonly cited cluster analytical studies and then explores the possible reasons for the discrepant results by replicating the approach with a different clinical sample. The results highlight the impact of different measures on cluster structure and the importance of adoption of a common standard to facilitate interpretation across studies.

Child↗

PubMed1989

A cluster analysis for threshold perimetry.

Cluster analysis in perimetry is a technique used in the evaluation of localised visual field loss. It has previously been applied to suprathreshold data and, unlike the indices currently available to indicate localised loss, it is influenced by the relative positions of individual defects. This paper describes a cluster analysis for use with data from Program 31 of the Octopus perimeter. To demonstrate the technique, sensitivity values of normal 60-year-old subject were altered to simulate localised loss. Illustrative examples of clinical cases are given, showing differing degrees of localised loss that do not influence the corrected loss variance (CLV) but influence the computed cluster parameters. It is hoped that the value of this form of analysis will be demonstrated in clinical follow-up of glaucoma patients.

Humans↗

PubMed1983

Identification of subsets of patients with Graves' disease by cluster analysis.

We have applied cluster analysis methods to forty-nine laboratory and clinical characteristics (including 26 HLA-A, B antigens) observed in 196 Graves' disease patients. Three subgroups could be identified: group I (seventy-nine patients) had small goitres, low indices of autoimmunity and a tendency to remission with medical treatment; group IIa (twenty-nine patients) had clinical and laboratory features of 'Hashitoxicosis'; Group IIb (eighty-four patients) had a high incidence of ophthalmopathy, familial aggregation, marked evidence of autoaggression and a tendency to relapsing hyperthyroidism. The prevalence of HLA-B8 was 8.9% in group I, 20.7% in group IIa and 86.9% in group IIb. This study demonstrates that Graves' disease can be subdivided using cluster analysis into clinically relevant subgroups which are further distinguished by their correlation with HLA-B8. Possible immunological bases for these observed patterns are discussed.

Adolescent↗

PubMed2005

Cluster analysis: a useful tool for the analysis of cerebral laser-Doppler scanning data.

Laser-Doppler (LD) fluxmetry (LDF) is a widely used method for the measurement of relative tissue perfusion. Assessing LD-flux at multiple locations using a scanning technique greatly reduces movement artefacts and makes repetitive measurements at the same location possible. However, measurements in brain are often confounded by superficial cortical vessels. Commonly applied strategies to circumvent this problem, such as defining a cut-off point to exclude the high flux data of vessels or calculating the median from multiple locations to estimate regional cerebral blood flow (rCBF) all have specific shortcomings. The aim of this study was to analyse LD-data by mathematically discriminating between parenchymal and vessel data based on the distribution of flux data. Data was obtained by scanning the cortex of 15 male Sprague-Dawley rats using a matrix of 6x10 equidistant (500 microm) points. Standard statistical analysis as well as cluster analysis using the complete linkage algorithm was performed. The LD-data showed a bimodal frequency distribution with low values representing parenchymal and high values representing vessel flux. Parenchyma and vessels were reliably discriminated by cluster analysis. This was shown by mapping the vessel clusters on the scan matrix with the location of the superficial cortical vessels using Chi-square testing (p<0.0001). The parenchymal data followed a Gaussian normal distribution (p<0.851), whereas the vessel data did not (p<0.0001). Thus, cluster analysis is useful to discriminate parenchymal from vessel flux, thereby significantly improving the accuracy of LD-scanning data.

Algorithms↗

PubMed1999

Classification of urinary stones by cluster analysis of ionic composition data.

The cluster analysis technique is considered for classifying kidney stones based on data for nine chemical analysis parameters. A set of 214 stones is used, which has been previously classified using empirical classification rules into three stone types using the percentage concentrations of the urate, oxalate, and phosphate radicals. We investigate whether cluster analysis utilising data on all parameters leads to different classifications and explore the possibility of other effective classifiers. We also compare the performance of various clustering techniques, distance and similarity measures and data standardisation methods. Results indicate that inclusion of the additional six parameters does not improve the classification accuracy. Best matching with the empirical classification (6% error) is achieved using the average linkage (between groups) clustering method and the squared Eculidean distance measure without data standardisation. Excluding these three main radicals causes a 63% matching error. Cluster analysis results suggest that carbon ions alone provide a single classifier for the three stone types, giving a matching error of approximately 10% with the empirical classification.

Cluster Analysis↗

PubMed1989

Profile of in vitro binding affinities of neuroleptics at different rat brain receptors: cluster analysis comparison with pharmacological and clinical profiles.

A series of 21 neuroleptics with different chemical structures (phenothiazines, thioxanthenes, dibenzodiazepines, butyrophenones, benzamides, etc.) was examined for their in vitro interactions with 12 neurotransmitter binding sites in the rat brain (alpha- and beta-noradrenergic, dopaminergic, muscarinic, serotoninergic, histaminic, and opioid receptors, calcium channels, and serotonin uptake binding sites). The biochemical profile obtained from the binding data was compared with reported pharmacological and clinical profiles for this class of compounds by cluster analysis. Cluster analysis on binding data classified the compounds in three main subgroups: benzamides, compounds with an affinity mainly for DA2 and 5-HT2 receptors and inactive at muscarinic receptors, and compounds with a high affinity for alpha 1-adrenergic receptors and muscarinic receptors. The main subgroups resulting from cluster analysis of previously published pharmacological and clinical data for neuroleptics contain compounds common to the present study, with some correlations. The results extend previous observations that a complete binding profile corresponds to the pharmacological and clinical profile of this class of compounds.

Animals↗

PubMed2004

[Determination of metal ions in tobacco and application of clustering analysis].

The contents of potassium, magnesium, sodium, manganese, ferrum and zinc were determined by flame spectrometry and atomic absorption spectrometry respectively. The dataset was clustered by fuzzy c-means (FCM) clustering analysis and hierarchical clustering analysis. It is shown that the results of the FCM clustering analysis are more accurate than those of the hierarchical cluster analysis, and it is with positive meaning to apply FCM clustering analysis to estimating the producing area of tobaccos.

Algorithms↗