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 433 records · Page 24Linked to original sources

PubMed1992

Subtyping of psychiatric patients by cluster analysis of QEEG.

We have previously reported successful classification of patients with a variety of psychiatric disorders, using multiple discriminant functions based upon selected neurometric QEEG variables. In independent replications, these functions accurately separate patients with different DSM-III-R diagnoses from one another and from normals. This capability demonstrates that distinctive and replicable patterns of neurometric abnormalities are correlated with the clinical symptom clusters upon which DSM-III-R diagnostic criteria are based. However, patients with the same clinical diagnoses often respond very differently to the same treatments. Similar symptoms may arise from different pathophysiology. This study explored the 'natural structure' of a population of psychiatric patients in 8 diagnostic categories, using uninformed cluster analysis based upon the same set of neurometric variables found useful in separating each of these categories from normal. This preliminary numerical taxonomic approach reveals that groups of patients in each of these DSM-III-R categories contain subtypes with markedly different pathophysiology; further, patients in different DSM-III-R categories were aggregated together within each cluster, displaying similar pathophysiological profiles. Objective classification based on such physiological measurements may add information useful to improve treatment outcomes.

Brain↗

PubMed1993

Classification of adult day care: a cluster analysis of services and activities.

Using data from a 1986 national census survey of 774 adult day care (ADC) centers, this study (a) determined whether distinct classes of ADC could be identified based on measures of program services and activities, and (b) delineated the distinguishing characteristics of such classes on other available measures of structure, process, and client population. A cluster analysis of 10 "process" measures of services and activities identified 6 classes of ADC centers: Alzheimer's Family Care, Rehabilitation, High Intensity Clinical/Social, Moderate Intensity Clinical/Social, General Purpose, and Low Scoring. Validity was examined by developing a set of expectations for pairs of classes on other available variables. Of 12 expectations, 11 were supported by the statistical tests. Finally, profiles of the 6 classes were developed to describe the classes on 30 other characteristics. The findings clarify the settings to which previous ADC studies are generalizable and indicate a need for effectiveness studies on special classes of ADC.

Adult↗

PubMed1988

Automatic classification of two-dimensional gel electrophoresis pictures by heuristic clustering analysis: a step toward machine learning.

The interpretation of two-dimensional gel electrophoresis (2-DGE) profiles can be facilitated by artificial intelligence and machine learning programs. We have incorporated into our 2-DGE computer analysis system (termed MELANIE-Medical Electrophoresis Analysis Interactive Expert system) a program which automatically classifies 2-DGE patterns using heuristic clustering analysis. This program is a step toward machine learning. In this publication, we describe the classification method and the preliminary results obtained with liver biopsy electrophoretograms. Heuristic clustering is also compared to other classification techniques.

Algorithms↗

PubMed2004

Cluster analysis of genes with significant change in expression in cells conditioned to survive TBOOH.

Immortal murine lens epithelial cells, alphaTN4-1 have been conditioned to survive H2O2, H cells, or TBOOH, T cells, at concentrations that will cause cataract in vitro. Since H cells are killed by TBOOH but T cells survive H2O2, it is of interest to examine the gene expression of these cell lines. We now report the results of cluster analysis of genes whose expression is significantly changed by TBOOH. The analysis has revealed a small group of antioxidative defense genes that contribute to the survival of T and H cells when exposed to oxidative stress.

Animals↗

PubMed2002

Molecular profiles of allograft rejection following inhibition of CD40 ligand costimulation differentiated by cluster analysis.

Recent technological advances in biomedical research, such as genome sequences and DNA microarrays, have dramatically increased the size of relevant databases. A major challenge is the extraction of a limited number of parameters from these databases that can differentiate and diagnose complex biological states. In a model of cardiac transplantation investigating immunosuppression by inhibition of CD40 ligand costimulation, we have applied a combination of cluster algorithms and self-organizing maps to analyze a panel of 60 candidate genes. Dendrograms generated by cluster analysis distinguished different molecular bases of rejection. Using self-organizing maps, we identified nine genes (CD4, CCR3, CCR5, LT beta, MIP-1 alpha, MIP-2, CD8 alpha, IP-10, and RANTES), each with a unique profile of transcriptional expression, that reproduce the differentiation of states of rejection in dendrograms. Using histology and immunohistochemistry, we correlated differential regulation of CD4 and CD8 at the levels of mRNA and protein. Our strategy of data reduction successfully decreased the number of genes to nine, which are sufficient to differentiate distinct states of rejection in our experimental protocol.

Animals↗

PubMed1998

Properties of spinal motoneurons and interneurons in the adult turtle: provisional classification by cluster analysis.

The purpose of the present study was to compare, in motoneurons (MNs) vs. interneurons (INs), selected passive, transitional, and active (firing) properties, as recorded in slices of lumbosacral spinal cord (SC) taken from the adult turtle. The cells were provisionally classified on the basis of (1) the presence (in selected INs) or absence (MNs and other INs) of spontaneous discharge, (2) a cluster analysis of selected properties of the nonspontaneously firing cells, (3) a comparison to previous data on turtle MNs and INs, and (4) a qualitative comparison of the results with those reported for other vertebrate species (lamprey, cat). The provisional nomenclature accommodated properties appropriate for solely MNs (Main MN group) vs. nonspontaneously firing INs (Main IN-N) vs. spontaneously firing INs (IN-S) and for neurons with two degrees of intermediacy between the Main MN and the Main IN-N groups (Overlap MN, Overlap MN/IN). Morphological reconstructions of additional cells, which had been injected with biocytin during the electrophysiological tests, were shown to provide clear-cut support for the provisional classification procedure. The values for the measured parameters in the 96 tested cells covered the spectrum reported previously across adult vertebrate species and were robust in measurements made on different SC slices up to 5 days after their removal from the host animal. The interspecies comparisons permitted the predictions that (1) our Main MN and Overlap MN cells would be analogous to two MN types that innervate fast-twitch and slow-twitch skeletomotor muscle fibers, respectively, in the cat, and (2) the MNs in our Overlap MN/IN group probably innervate slow (nontwitch, tonic) muscle fibers whose presence has recently been established in the turtle hindlimb. In summary, the results bring out the utility of the SC slice preparation of the turtle for study of spinal motor mechanisms in adult tetrapod vertebrates, particularly as an adjunct to the in vivo cat, because of the ease with which robust measurements can be made of the active properties of both MNs and INs.

Action Potentials↗

PubMed2005

Multivariate cluster analysis as a versatile tool for the quality assessment of short chain chloroparaffin quantification in environmental samples.

The compositions of commercially available solutions of short chain chlorinated paraffins (SCCP) and technical mixtures were determined by high resolution gas chromatography (HRGC) electron capture negative ionisation (ECNI) combined with low resolution mass spectrometry (LRMS). Differences and similarities in the relative distribution of congeners and homologues were investigated by cluster analysis. Samples could be separated into two groups/clusters with similar molecular mass, chlorination degree and congener pattern. With the exception of one sample, the use of CH4-ECNI-MS led to an overestimation of the chlorine content. Moreover, the influence of different SCCP mixtures on the quantification was studied. The results showed that small differences in the chlorine content could sometimes cause substantial systematic errors of up to 119% despite similar homologue and congener patterns. Errors of quantification corresponded to the deviations between response factors of sample and standard. As a consequence SCCPs should be quantified with a standard matching the overall response factor of the sample.

Cluster Analysis↗

PubMed2002

Surface antigen expression in chronic lymphocytic leukemia: clustering analysis, interrelationships and effects of chromosomal abnormalities.

Chronic lymphocytic leukemia (CLL) is a phenotypically distinguishable form of B-lymphoid leukemias. The regularity of surface membrane antigen expression patterns, their interrelationships as well as the effects of the three frequent chromosomal aberrations, ie 11q deletion, 13q deletion and trisomy 12, were investigated in 35 classic CLL cases by flow cytometry. The two-way cluster analysis of 31 individual antigens revealed three expression patterns: (1) most cells in most cases positive (CD5, CD19, CD20, CD23, CD27, CD40, CD45, CD45RA); (2) most cells in most cases negative (CD10, CD14, CD34, CD122, CD154, mIgG); and (3) a mixed pattern with a variable number of positive cases and a variable percentage of positive cells in individual cases (CD11c, CD21, CD22, CD25, CD38, CD45RO, CD79b, CD80, CD95, CD124, CD126, CD130, FMC7, mIgD, mIgkappa, mIglambda, mIgM). The expressions of several antigens were strongly interdependent, even when antigens belonged to entirely different gene families. Such antigen pairs were: CD11c/CD21; CD19/CD45; CD19/CD79b; CD22/CD45RA; CD23/Igkappa; CD25/mIgM; CD27/CD45; CD45/CD79b; CD45RA/Igkappa. In contrast, the expression of some antigens was mutually exclusive, the best examples being CD45RA/CD45RO, CD38/CD80 and CD45RA/CD80. Deletion of chromosome arm 11q attenuated expression of splicing variant CD45RA, but enhanced CD45RO expression. In contrast, cases of trisomy 12 were associated with enhanced CD45RA and attenuated CD45RO expression. Similarly, trisomy 12 was associated with enhanced CD27 and mIgkappa expression. The variable levels of signaling surface membrane antigens, their interactions and interference by genetic aberrations are likely to affect the clinical progression and drug response of CLL.

ADP-ribosyl Cyclase↗

PubMed1996

Self-organizing neural networks--an alternative way of cluster analysis in clinical chemistry.

Supervised learning schemes have been employed by several workers for training neural networks designed to solve clinical problems. We demonstrate that unsupervised techniques can also produce interesting and meaningful results. Using a data set on the chemical composition of milk from 22 different mammals, we demonstrate that self-organizing feature maps (Kohonen networks) as well as a modified version of error backpropagation technique yield results mimicking conventional cluster analysis. Both techniques are able to project a potentially multi-dimensional input vector onto a two-dimensional space whereby neighborhood relationships remain conserved. Thus, these techniques can be used for reducing dimensionality of complicated data sets and for enhancing comprehensibility of features hidden in the data matrix.

Animals↗

PubMed2003

Dietary and nutritional patterns in an elderly rural population in Northern and Southern Italy: (I). A cluster analysis of food consumption.

OBJECTIVE: To describe the food consumption patterns of Italian elderly subjects and the factors associated with different dietary habits. DESIGN: Cross-sectional study. SETTING: Population-based study. SUBJECTS: A total of 847 men and 1465 women aged 65 y or older, living in rural areas in the province of Pavia (Northern Italy) and near Cosenza (Southern Italy) in 1992-1993. INTERVENTIONS: The dietary information was collected by means of a quantitative food-frequency questionnaire, with cluster analysis being used to segregate the subjects on the basis of similarities in their food consumption. A generalised logistic regression model including residence, age, living status, education, physical activity and degree of disability was fitted to assess the factors associated with different food consumption habits. RESULTS: Six dietary clusters were selected for men and seven for women. The largest cluster for both genders was "small eaters" (46% of men and 51% of women); "big eaters", "light diet" and "alcohol" clusters were also identified for both genders. The men were also grouped into "balanced diet" and "cheese" clusters, and the women into "sweet", "greens" and "butter" clusters. CONCLUSIONS: In addition to the influence of the area of residence--residents in Northern Italy consume greater amounts of animal fats, sugar and alcoholic beverages, and those living in Southern Italy have a greater intake of fruit, vegetables, fish and olive oil--all of the other studied variables predicted the probabilities of falling into a specific dietary cluster. SPONSORSHIP: National Research Council (Italy)--Invecchiamento Project No. 95.01048. PF40.

Aged↗

PubMedDate not supplied

Profiles of rural nurses' use of personal protective equipment: a cluster analysis.

This study examined the relationship between profiles of rural registered nurses' levels of compliance with the use of personal protective equipment and HIV-related attitudes. Survey data were collected from 395 practicing registered nurses in nine rural counties in New York and Pennsylvania. Cluster analysis grouped respondents according to their use of protective equipment by patient HIV status. Seven profiles were identified from the resulting 13 clusters. Profiles were named according to the characteristics of the protective equipment usage that were most evident in each profile. Usage levels included minimal users, appropriate users, anticipatory users, glove users, discriminate users, maximal users, and optimal users. Of these, three were classified as compliant, one as undercompliant, and three as overcompliant. Selective compliance (changing usage in response to patients' HIV status) was characteristic of five profiles. Knowledge that patients were HIV-positive accounted for overcompliance and selective compliance and was related to nurses' HIV attitudes. Undercompliance was related to care of patients who were HIV-negative or unknown. Rural nurses' use of personal protective equipment is not homogenous but discrete and idiosyncratic. This analysis expands current knowledge and redefines nursing practice of standard precautions.

Adult↗

PubMed2002

A comparison of maximum covariance and K-means cluster analysis in classifying cases into known taxon groups.

Maximum covariance (MAXCOV) is a method for determining whether a group of 3 or more indicators marks 1 continuous or 2 discrete latent distributions of individuals. Although the circumstances under which MAXCOV is effective in detecting latent taxa have been specified, its efficiency in classifying cases into groups has not been assessed, and few studies have compared its performance with that of cluster analysis. In the present Monte Carlo study, the classification efficiencies of MAXCOV and the k-means algorithm were compared across ranges of sample size, effect size, indicator number, taxon base rate, and within-groups covariance. When the impact of these parameters was minimized, k-means classified more data points correctly than MAXCOV. However, when the effects of all parameters were increased concurrently, MAXCOV outperformed k-means.

Cluster Analysis↗

PubMed1994

[Specific characteristics of Crohn disease patients--results of a cluster analysis of personality markers].

The aim of this study was to find out, if patients with Crohn's disease share some personality traits or if it makes more sense to separate them into subgroups. Therefore, 52 patients took part in a psychoanalytic interview. The interview was composed in order to check personality characteristics as they were named in the literature. However, there could not be found any specific personality traits, but a cluster analysis revealed 7 subgroups. Two larger subgroups, together 22 patients, consisted of patients with inhibited aggression, who were closely connected to their families and showed illness inciting stressful life events. Another subgroup, 11 patients, showed a marked independent, aggressive and illness-dissimulating behavior. The hypothesis of a specific personality structure has to be rejected.

Adaptation, Psychological↗

PubMed1991

Knowing when you're drunk: a cluster analysis of cues to intoxication.

The present study was conducted to determine what cues to intoxication untrained subjects report using and how subjects see relationships among these cues. One hundred subjects were asked to list 5 cues that they use to determine their level of intoxication. These responses were tabulated, and the 18 most frequent responses were retained for further analyses. To assess the relationships among cues, all possible pairs of cues were rated for similarity. Mean similarity ratings were collapsed across subjects to form a mean similarities matrix that was subjected to cluster analysis. A taxonomy of cues to intoxication is discussed.

Adult↗

PubMedDate not supplied

[Schizophrenia, manic-depressive psychoses and "middle cases". A clinical and nosologic study with cluster analysis].

Having analyzed the main nosographies concerning the endogenous psychoses problem, the A. aim at obtaining a subdivision of the psychotic patients into classes based on the similarity shown by the same patients about symptoms found out by a transverse examination. For this purpose, 246 psychotic inpatients have been given PSE to evaluate the presence/absence of 18 symptoms considered discriminant to the clinical rating of the patients. Such data have been processed by a Cluster Analysis. We got 7 groups of probands that have stressed the relevant number of cases with a mixture of affective and schizophrenic symptoms. We suggest the use of a bidimensional model to get a proper fitting of psychotics with mixed symptomatology.

Adolescent↗

PubMed1998

Fuzzy cluster analysis--a new method to predict future cardiac events in patients with positive stress tests.

Several studies have shown that combining the change in the ST-segment with another exercise variable improves the predictive value of stress testing. However, no method has been able to combine many stress test variables with the ST-segment change simultaneously and help the clinician better predict future cardiac events. Fuzzy Cluster Analysis (FCA) was used to combine 5 stress test variables with ST-segment deviation to classify each of 232 positive outpatient stress tests as mildly, moderately, or severely abnormal. Cardiac events were recorded in these 3 patient groups up to 96 months (mean 65 months) after the stress tests. Coronary angiography was performed on 159 of these patients within 1 month of their stress tests. FCA better separated the 3 event-free survival curves than classifying the stress tests by three ST-segment (0.5-1.5 mm, 2-2.5 mm, > 3 mm) groups (p < 0.05). At 2 years, 90% of the FCA mild group were compared with 70% for the 0.5-1.5 mm group (p < 0.01). Moderate and severe tests by FCA separated patients with an intermediate from those with a poor prognosis while the 2-2.5 mm and 3 mm or more ST-segment curves did not (p < 0.05). FCA showed overall better correlation with coronary score (r = 0.71) than did the graded ST-segment groups (r = 0.48). FCA predicted both mild and high-grade (triple-vessel and left main) coronary disease better than ST-segment alone. Thus FCA better predicts future cardiac events in patients with positive stress tests than the ST-segment alone. This combined with its usefulness in predicting the extent of coronary disease provides the basis of a clinical strategy for managing patients with positive stress tests.

Adult↗

PubMed2006

Sequential activation of muscle synergies during locomotion in the intact cat as revealed by cluster analysis and direct decomposition.

During goal-directed locomotion, descending signals from supraspinal structures act through spinal interneuron pathways to effect modifications of muscle activity that are appropriate to the task requirements. Recent studies using decomposition methods suggest that this control might be facilitated by activating synergies organized at the level of the spinal cord. However, it is difficult to directly relate these mathematically defined synergies to the patterns of electromyographic activity observed in the original recordings. To address this issue, we have used a novel cluster analysis to make a detailed study of the organization of the synergistic patterns of muscle activity observed in the fore- and hindlimb during treadmill locomotion. The results show that the activity of a large number of forelimb muscles (26 bursts of activity from 18 muscles) can be grouped into 11 clusters on the basis of synchronous co-activation. Nine (9/11) of these clusters defined muscle activity during the swing phase of locomotion; these clusters were distributed in a sequential manner and were related to discrete behavioral events. A comparison with the synergies identified by linear decomposition methods showed some striking similarities between the synergies identified by the different methods. In the hindlimb, a simpler organization was observed, and a sequential activation of muscles similar to that observed in the forelimb during swing was less clear. We suggest that this organization of synergistic muscles provides a means by which descending signals could provide the detailed control of different muscle groups that is necessary for the flexible control of multi-articular movements.

Animals↗

PubMed1993

Self-organizing neural networks as a means of cluster analysis in clinical chemistry.

Connectionist systems (often termed "neural networks") are an alternative way to solve data processing tasks. They differ radically from conventional "von-Neumann" computing devices. Recent work on neural networks in clinical chemistry was done using supervised learning schemes, resulting in models which resemble classical discriminant analysis. The aim of the present study is to make clinical chemists familiar with basic concepts of self-organizing neural networks employing unsupervised learning schemes. Using a benchmark data set on the composition of milk from 22 different mammals, it is demonstrated that self-organizing neural networks are capable of performing tasks similar to classical cluster analysis and principal component analysis. Self-organizing neural networks could be envisaged to provide an alternative way for reducing the dimensionality of complex multivariate data sets, thus producing easily comprehensible low-dimensional "maps" of essential features.

Animals↗