PubMed Health⌕ Search

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 865 records · Page 48Linked to original sources
PubMed2005

Shelf life of Crescenza cheese as measured by electronic nose.

The shelf life of Crescenza, a traditional Italian soft cheese, was measured using classical analysis and a commercial electronic nose. Two lots of samples directly supplied by a manufacturer at the beginning of their commercial life were stored at 2 constant temperatures (8 and 15 degrees C) and analyzed until their respective expiration dates. Among the physicochemical parameters, pH, acidity, hue, and apparent yield rheological index appeared to be the best predictors of the quality decay. Changes in these indices were described with a sigmoidal transition function allowing definition of a loose and a severe shelf-life protocol, based on the trend of first and second time derivatives. A time range of 1 to 3 d at 15 degrees C and 4 to 8 d at 8 degrees C was accordingly assessed to maintain the freshness of Crescenza cheese. The quality decay of cheese aroma was evaluated by inspecting the headspace fingerprint of the same set of samples using the electronic nose. Sample classification through the aroma fingerprint confirmed the predicted shelf-life time ranges. A clear discrimination between "fresh," "aged," and "very aged" samples was obtained using principal components analysis, cluster analysis, and linear discriminant analysis statistical techniques. The predictive ability of the linear discriminant analysis classification model was confirmed by considering a new set of cheese samples purchased at the beginning of their commercial life from a local market and analyzed until their expiration date.

Cheese↗
PubMedDate not supplied

Developmental history of the Glover-Nilsson smoking behavioral questionnaire.

OBJECTIVE: To develop a simple, easily administered pencil-and-paper questionnaire to determine the degree to which behavioral patterns play a role in smoking dependence. METHODS: A modified Delphi technique was used to identify initial questions and to eliminate obvious duplications. Phase 2 utilized multiple statistical methods (principal components analysis, cluster analysis, stepwise multiple linear regression, cross tables, Mantel-Haenzel c2-test, and a Gamma test) to evaluate and reduce the number of questions from 18. RESULTS: These analyses yielded an 11-item questionnaire that can potentially assess behavioral dependence. CONCLUSION: It is hoped that the GN-SBQ will assist physicians, health care providers, and tobacco interventionists in identifying aspects of smoking addiction that are behavioral in nature. The need for future research is discussed.

Attitude to Health↗
PubMed2003

Size and shape modifications in Sardinian adolescent girls.

In females, menarche is the defining moment of puberty, the period of life when the greatest body changes occur. In the present study, the metric and morphological variations associated with sexual maturation are defined in 155 Sardinian girls (10-17 years) and the role of some potentially influential variables is discussed: age, age at menarche and time since menarche. We studied thirty-eight anthropometric variables, the fat-free mass and the fat mass estimated by Bioelectrical Impedance Analysis. Statistical analyses were performed to evaluate the difference between pre- and post-menarcheal girls of the same age (Student's t-test) and to evaluate the different role played by the variables (principal components analysis, cluster analysis, multiple regression). The results demonstrate that the body dimensions of the adolescent girls mainly increase in concomitance with sexual maturation. The age at menarche influences the fat mass but not the distribution of visceral and subcutaneous fat. The time since menarche has also no effect on the distribution of subcutaneous fat.

Adolescent↗
PubMed2004

[Multiple nonlinear statistical method of population genetic structure based on the allelic polymorphism data].

The distribution and structure of the allelic polymorphism data are analyzed and it is pointed out that the distribution of allelic polymorphism data reveals the characteristic of closed data (also named as compositional data or data of constant sum). It is interpreted that the correlation structure of the allelic polymorphism data contains null correlations introduced by "closure" and the statistical distribution of the data is not normal because of its constant row sum, which resulted in great difficulties in analyzing the data with traditional multiple linear statistical methods such as principal component analysis, factor analysis, cluster analysis and canonical correlation analysis. Based on the theory of compositional data analysis proposed by Aitchison in 1982, a multiple nonlinear statistical method originating from the "logratios" approach to the statistical analysis of compositional data is put forward in this paper. As an example, the "logratios" method was used to analyze the genetic structure of TH01 polymorphic loci in Chinese population and the results were compared with those of multiple linear methods such as component principal. It is concluded that the "logratios" multiple nonlinear principle component analysis is a better method with the virtue of sensitivity and specificity for analyzing the genetic structure of population from the data of allelic polymorphism.

Alleles↗
PubMed1987

Multivariate statistical analysis of L-dopa esters as potential anti-parkinsonian prodrugs.

This paper reports a QSAR study of thirteen L-DOPA esters previously synthesised and examined for their physicochemical, biochemical and behavioural properties. Multivariate statistical analysis (principal component analysis, cluster analysis, simple and multiple linear regression) reveals favourable and unfavourable structural features of L-DOPA prodrugs. Some of the many biochemical and behavioural activities investigated are shown to be redundant. These indications may contribute to the design of novel L-DOPA prodrugs.

Analysis of Variance↗
PubMed1986

[Correlation study of cancer mortality and food consumption].

The geographical relationships between cancer mortality and food consumption were analyzed by means of extensive uses of correlation coefficiencies, factor analysis, multiple regression analysis, cluster analysis and path analysis. The results for the countries in the world varied broadly due to selection of the countries analyzed. The results for prefectures in Japan showed differences between males and females especially with regard to stomach cancer. The present study shows the importance of careful selection of geographical and other variables when making correlation studies.

Alcohol Drinking↗
PubMed1987

[Recovery in aphasia (Part 3)--Study by multivariate analysis].

In order to elucidate the factors which have influence on the prognosis of aphasia, correlation between recovery rate of aphasia and the five factors, i.e. initial evaluation of speech test, age, educational level, time between onset and institution of therapy and size of abnormal findings in CT (computed tomography), was studied in 76 right-handed aphasic patients, using multivariate analysis. Cluster analysis and factor analysis were used for analysis of five factors and multiple regression analysis was used for estimation of recovery rate of aphasia. The method of calculation of recovery rate of aphasia is the same to our previous report. The results obtained were as follows: As to analysis of factors, first similarity group among initial evaluation of speech test, recovery rate of aphasia and educational level and second similarity group among time between onset and institution of therapy, size of abnormal findings in CT and age were observed. As to the estimation of recovery rate of aphasia, multiple correlation coefficient of regression analysis using five factors were 0.758, 0.444, 0.627, 0.620, 0.810 and 0.375 respectively in total score, hearing, speaking, reading, writing and calculation on standard language test of aphasia. As to the weight or partial correlation coefficient of regression analysis to the five factors, initial evaluation of speech test, time between onset and institution of therapy and educational level were 129.46, -49.93 and 45.65 respectively and the one of the size of abnormal findings in CT was not high. Stepwise multiple regression analysis of five factors were made.(ABSTRACT TRUNCATED AT 250 WORDS)

Analysis of Variance↗
PubMed1980

[Aortic valve replacement with Starr-Edwards and Björk-Shiley prosthesis. Long term results and statistical analysis of some risk factors (author's transl)].

285 patients who underwent aortic valve replacement between 1972 and 1977 have been studied. 197 patients had a Starr-Edwards prosthesis and 88 patients had a Björk-Shiley prosthesis, both groups presented similar risk factors. The follow-up varied between a period of 6 months minimum and 72 months maximum, the mean period is 3.6 years. The statistical results have been realized based on the multi-factor analysis (cluster) which were the basis to define the homogeneous groups of patients with the Starr prosthesis and the Björk prosthesis. One may observe the best survival percentage after 72 months from surgery date in patients with the Björk prosthesis comparing them with those patients who underwent aortic valve replacement with the Starr prosthesis (resp. 93 and 71%). The difference is important from the statistical point of view. The embolic incidence after such a period is nearly equal (16% for Björk and 14% for Starr). The cluster analysis of the considered parameters (age, duration of symptoms before surgery, NYHA class, C/T) did not evince real and proper risk factors such to condition, at long term, the results even if differences in mortality and in embolic incidence, with regards to the relation with NYHA class and symptoms; duration, have been observed. One may conclude that one of most important risks derives form the type of used prosthesis.

Adolescent↗
PubMed1981

Interpretation of research data: other multivariate methods.

Four techniques of multivariate data analysis are described: discriminant analysis, factor analysis, cluster analysis, and multidimensional scaling. Objectives and applications, data requirements, mechanism of operation, methods of evaluation, and advantages and disadvantages of each method are covered. Multivariate methods are not substitutes for the researcher's use of proper study design. Investigators should try to avoid using a favorite or familiar technique across all research situations.

Research Design↗
PubMed2001

Feature-space clustering for fMRI meta-analysis.

Clustering functional magnetic resonance imaging (fMRI) time series has emerged in recent years as a possible alternative to parametric modeling approaches. Most of the work so far has been concerned with clustering raw time series. In this contribution we investigate the applicability of a clustering method applied to features extracted from the data. This approach is extremely versatile and encompasses previously published results [Goutte et al., 1999] as special cases. A typical application is in data reduction: as the increase in temporal resolution of fMRI experiments routinely yields fMRI sequences containing several hundreds of images, it is sometimes necessary to invoke feature extraction to reduce the dimensionality of the data space. A second interesting application is in the meta-analysis of fMRI experiment, where features are obtained from a possibly large number of single-voxel analyses. In particular this allows the checking of the differences and agreements between different methods of analysis. Both approaches are illustrated on a fMRI data set involving visual stimulation, and we show that the feature space clustering approach yields nontrivial results and, in particular, shows interesting differences between individual voxel analysis performed with traditional methods.

Algorithms↗
PubMed1998

Between- and within-cluster covariate effects in the analysis of clustered data.

Standard methods for the regression analysis of clustered data postulate models relating covariates to the response without regard to between- and within-cluster covariate effects. Implicit in these analyses is the assumption that these effects are identical. Example data show that this is frequently not the case and that analyses that ignore differential between- and within-cluster covariate effects can be misleading. Consideration of between- and within-cluster effects also helps to explain observed and theoretical differences between mixture model analyses and those based on conditional likelihood methods. In particular, we show that conditional likelihood methods estimate purely within-cluster covariate effects, whereas mixture model approaches estimate a weighted average of between- and within-cluster covariate effects.

Adolescent↗
PubMed2004

Minimum entropy clustering and applications to gene expression analysis.

Clustering is a common methodology for analyzing the gene expression data. In this paper, we present a new clustering algorithm from an information-theoretic point of view. First, we propose the minimum entropy (measured on a posteriori probabilities) criterion, which is the conditional entropy of clusters given the observations. Fano's inequality indicates that it could be a good criterion for clustering. We generalize the criterion by replacing Shannon's entropy with Havrda-Charvat's structural alpha-entropy. Interestingly, the minimum entropy criterion based on structural alpha-entropy is equal to the probability error of the nearest neighbor method when alpha = 2. This is another evidence that the proposed criterion is good for clustering. With a non-parametric approach for estimating a posteriori probabilities, an efficient iterative algorithm is then established to minimize the entropy. The experimental results show that the clustering algorithm performs significantly better than k-means/medians, hierarchical clustering, SOM, and EM in terms of adjusted Rand index. Particularly, our algorithm performs very well even when the correct number of clusters is unknown. In addition, most clustering algorithms produce poor partitions in presence of outliers while our method can correctly reveal the structure of data and effectively identify outliers simultaneously.

Algorithms↗
PubMed2002

Discriminant analysis to evaluate clustering of gene expression data.

In this work we present a procedure that combines classical statistical methods to assess the confidence of gene clusters identified by hierarchical clustering of expression data. This approach was applied to a publicly released Drosophila metamorphosis data set [White et al., Science 286 (1999) 2179-2184]. We have been able to produce reliable classifications of gene groups and genes within the groups by applying unsupervised (cluster analysis), dimension reduction (principal component analysis) and supervised methods (linear discriminant analysis) in a sequential form. This procedure provides a means to select relevant information from microarray data, reducing the number of genes and clusters that require further biological analysis.

Animals↗
PubMed2001

Food intake patterns and body mass index in observational studies.

OBJECTIVE: To review studies of patterns of food intake, as assessed by diet index, factor analysis or cluster analysis, and their associations with body mass index or obesity (BMI/Ob). DESIGN: Systematic literature review MEDLINE search with crosscheck of references. STUDIES: Thirty observational studies relating food intake patterns to anthropometric information were identified and reviewed. Food intake patterns were defined using a diet index, factor or cluster analysis in 12, nine and nine studies, respectively. Measures of body mass were made concurrently with the diet assessment in all studies, and only in a few cases were the primary outcomes related to BMI/Ob. RESULTS: The food intake patterns identified could, in most factor or cluster analysis studies, be categorised as: (a) meat, fatty, sweet or energy dense foods; (b) vegetables, fruit, whole grain and low-fat foods; or (c) by high alcohol consumption. The diet indexes were designed to capture a high diversity and/or food combinations matching the recommendations. The relationships with BMI/Ob were inconsistent-ten studies found that intake patterns, which we categorised as fatty, sweet or energy dense were positively associated with BMI/Ob, while similar patterns in four other studies were negatively associated with BMI. The significant associations between diet index score and BMI/Ob were consistently negative, while the associations between factor scores or cluster membership and BMI/Ob were less clear in terms of food intake pattern. Men and women had similar food intake patterns, but food intake patterns were less often positively associated with BMI/Ob in women. In 11 studies, there were no significant associations between food intake pattern and BMI/Ob. CONCLUSION: This review showed that no consistent associations could be identified between BMI or Ob and food intake patterns, derived from diet index, factor analysis or cluster analysis. However, the heterogeneity of food intake patterns identified by such analyses and the lack of gold standards for the application of these techniques hampers consistent analysis of a relation between food intake patterns and health.

Adolescent↗
PubMed2003

Multiple goals, motivation and academic learning.

BACKGROUND: The type of academic goals pursued by students is one of the most important variables in motivational research in educational contexts. Although motivational theory and research have emphasised the somewhat exclusive nature of two types of goal orientation (learning goals versus performance goals), some studies (Meece, 1994; Seifert, 1995, 1996) have shown that the two kinds of goals are relatively complementary and that it is possible for students to have multiple goals simultaneously, which guarantees some flexibility to adapt more efficaciously to various contexts and learning situations. AIM: The principal aim of this study is to determine the academic goals pursued by university students and to analyse the differences in several very significant variables related to motivation and academic learning. SAMPLE: Participants were 609 university students (74% women and 26% men) who filled in several questionnaires about the variables under study. METHOD: We used cluster analysis ('quick cluster analysis' method) to establish the different groups or clusters of individuals as a function of the three types of goals (learning goals, performance goals, and social reinforcement goals). By means of MANOVA, we determined whether the groups or clusters identified were significantly different in the variables that are relevant to motivation and academic learning. Lastly, we performed ANOVA on the variables that revealed significant effects in the previous analysis. RESULTS: Using cluster analysis, three groups of students with different motivational orientations were identified: a group with predominance of performance goals (Group PG: n = 230), a group with predominance of multiple goals (Group MG: n = 238), and a group with predominance of learning goals (Group LG: n = 141). CONCLUSIONS: Groups MG and LG attributed their success more to ability, they had higher perceived ability, they took task characteristics into account when planning which strategies to use in the learning process, they showed higher persistence, and used more deep learning strategies than did the students with predominance of performance goals (Group PG). On the other hand, Groups MG and PG took the evaluation criteria more into account when deciding which strategies to use in order to learn, and they attributed their failures more to luck than did Group LG. Students from Group MG attributed their success more to effort than did the other two groups and they attained higher achievement than Group PG. Group LG tended to attribute their failures more to lack of effort than did the other two groups.

Achievement↗
PubMed1981

Statistical analysis of potential cancer clusters.

Cluster analysis of disease occurrence has a long history in acute models' it has only recently begun to be used for chronic models. This extrapolation of methods has resulted in severe limitations and several modifications. Two cluster reports from Florida are given as examples for a review of the prevalent statistical techniques available for cancer cluster analysis. The methods are described and discussed. The significance of the two clusters is left unsolved.

Colonic Neoplasms↗
PubMed1999

[A study on soil suitability for growth of rhizome of Curcuma longa L].

OBJECTIVE: To seek for the type of soil that influences favorably the yield and quality of Curcuma longa. METHOD: Field contrast control experiment, correlation analysis, path analysis and cluster analysis were used. RESULT: The rhizome yield and quality of Curcuma longa. varied greatly with the types of soil texture. There are clear correlations among the soil factors, as well as between the soil factors and the rhizome yield and quality of the plant. A regression equation of the rhizome yield and soil factors has been built up. CONCLUSION: The soil suitable for growing Curcuma longa. can be classified into three types according to the results of correlation analysis and cluster analysis.

Curcuma↗