PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Multivariate 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 1,567 records · Page 87Linked to original sources
PubMed2006

Infrared spectroscopy with multivariate analysis potentially facilitates the segregation of different types of prostate cell.

The prostate gland is conventionally divided into zones or regions. This morphology is of clinical significance as prostate cancer (CaP) occurs mainly in the peripheral zone (PZ). We obtained tissue sets consisting of paraffin-embedded blocks of cancer-free transition zone (TZ) and PZ and adjacent CaP from patients (n = 6) who had undergone radical retropubic prostatectomy; a seventh tissue set of snap-frozen PZ and TZ was obtained from a CaP-free gland removed after radical cystoprostatectomy. Paraffin-embedded tissue slices were sectioned (10-mum thick) and mounted on suitable windows to facilitate infrared (IR) spectra acquisition before being dewaxed and air dried; cryosections were dessicated on BaF(2) windows. Spectra were collected employing synchrotron Fourier-transform infrared (FTIR) microspectroscopy in transmission mode or attenuated total reflection-FTIR (ATR) spectroscopy. Epithelial cell and stromal IR spectra were subjected to principal component analysis to determine whether wavenumber-absorbance relationships expressed as single points in "hyperspace" might on the basis of multivariate distance reveal biophysical differences between cells in situ in different tissue regions. After spectroscopic analysis, plotted clusters and their loadings curves highlighted marked variation in the spectral region containing DNA/RNA bands ( approximately 1490-1000 cm(-1)). By interrogating the intrinsic dimensionality of IR spectra in this small cohort sample, we found that TZ epithelial cells appeared to align more closely with those of CaP while exhibiting marked structural differences compared to PZ epithelium. IR spectra of PZ stroma also suggested that these cells are structurally more different to CaP than those located in the TZ. Because the PZ exhibits a higher occurrence of CaP, other factors (e.g., hormone exposure) may modulate the growth kinetics of initiated epithelial cells in this region. The results of this pilot study surprisingly indicate that TZ epithelial cells are more likely to exhibit what may be a susceptibility-to-adenocarcinoma spectral signature. Thus, IR spectroscopy on its own may not be sufficient to identify premalignant prostate epithelial cells most likely to progress to CaP.

Algorithms↗
PubMed1981

Numerical evaluation of cytologic data. X. Introduction to multivariate analysis of variance.

Quantitative measurement on cytologic material usually involves several features for each cell. The data are thus multivariate and are represented as feature vectors. Analysis of variance on univariate data is well established for the detection of small differences between sets of data. For multivariate data, the calculating schemes are much less well known. They are presented in this paper in direct analogy to univariate procedures. The calculating schedule in a factorial design in bivariate data is demonstrated.

Analysis of Variance↗
PubMed1983

Disenrollment from a prepaid group plan. A multivariate analysis.

Although high annual disenrollments from prepaid group practices (PGPs) present significant problems to the manager, studies on this subject generally have been limited to simple descriptions of the phenomenon. The authors' approach utilizes multivariate techniques--discriminant analysis and logistic regression--to overcome the simplicity of these bivariate studies. It allows an examination of the significant correlates of disenrollment in a stepwise manner that adjusts for the effects of all variables included in the models. Two surveys conducted by the Medical Care Group of St. Louis (MCGSL) provide data for this research. After classifying the sample of 2,402 families as continuous enrollees, voluntary disenrollees, and mandatory disenrollees, the analyses show that these groups represent significantly different member populations. The variables that significantly and independently discriminate among them include sociodemographic characteristics (subscriber age, race, education and occupation, and family size and income), satisfaction level with the plan, and presence of alternatives to the PGP (nonplan family members and coverage by alternative health insurance). These findings suggest that voluntary and mandatory disenrollment behavior may be more complex than previously considered. Moreover, the PGP manager is faced with perplexing problems in attempting to control this troublesome phenomenon.

Analysis of Variance↗
PubMed1999

Multivariate analysis ('forensiometrics')--a new tool in forensic medicine. Findings on the victim of sharp-force homicide can predict the inter-relationship with the perpetrator.

A multivariate 'forensiometric' technique (PLS-DA) was used to create a model to predict the relationship between homicide victim and perpetrator on a five-level scale: strangers, acquaintances, drinking companions, relatives and spouses. The model is based on findings on the victim and at the venue, and uses the following ten variables as 'predictors', listed in falling order of covariation with instances of close relationship between victim and perpetrator: Victim found in home, female victim, a single sharp injury, injuries to the upper extremity, superficial sharp injuries to the chest ('scratches'), ten or more sharp injuries, presence of defence injuries, total number of sharp injuries, male victim, and victim found outdoors. The model was based on 87 sharp-force homicides (the model-set) and was validated on another set of 43 sharp-force homicides (the test-set). In this test-set validation, 17 of 39 cases (44%) were correctly predicted according to the results of earlier combined forensic and police investigations, and a prediction one step away from the correct level was given in another 17 cases. These results were significantly better than would have been obtained by chance.

Criminal Psychology↗
PubMed2003

Assessment of 1H NMR spectroscopy and multivariate analysis as a technique for metabolite fingerprinting of Arabidopsis thaliana.

An approach to metabolite fingerprinting of crude plant extracts that utilizes 1H nuclear magnetic resonance (NMR) spectroscopy and multivariate statistics has been tested. Using ecotypes of Arabidopsis thaliana as experimental material, a method has been developed for the rapid analysis of unfractionated polar plant extracts, enabling the creation of reproducible metabolite fingerprints. These fingerprints could be readily stored and compared by a variety of chemometric methods. Comparison by principal component analysis using SIMCA-P allowed the generation of residual NMR spectra of the compounds that contributed significantly to the differences between samples. From these plots, conclusions were drawn with respect to the identity and relative levels of metabolites differing between samples.

Arabidopsis↗
PubMed2005

Correlation of porous and functional properties of food materials by NMR relaxometry and multivariate analysis.

The porous properties of food materials are known to determine important macroscopic parameters such as water-holding capacity and texture. In conventional approaches, understanding is built from a long process of establishing macrostructure-property relations in a rational manner. Only recently, multivariate approaches were introduced for the same purpose. The model systems used here are oil-in-water emulsions, stabilised by protein, and form complex structures, consisting of fat droplets dispersed in a porous protein phase. NMR time-domain decay curves were recorded for emulsions with varied levels of fat, protein and water. Hardness, dry matter content and water drainage were determined by classical means and analysed for correlation with the NMR data with multivariate techniques. Partial least squares can calibrate and predict these properties directly from the continuous NMR exponential decays and yields regression coefficients higher than 82%. However, the calibration coefficients themselves belong to the continuous exponential domain and do little to explain the connection between NMR data and emulsion properties. Transformation of the NMR decays into a discreet domain with non-negative least squares permits the use of multilinear regression (MLR) on the resulting amplitudes as predictors and hardness or water drainage as responses. The MLR coefficients show that hardness is highly correlated with the components that have T2 distributions of about 20 and 200 ms whereas water drainage is correlated with components that have T2 distributions around 400 and 1800 ms. These T2 distributions very likely correlate with water populations present in pores with different sizes and/or wall mobility. The results for the emulsions studied demonstrate that NMR time-domain decays can be employed to predict properties and to provide insight in the underlying microstructural features.

Emulsions↗
PubMed1985

Multivariate analysis and list mode processing of murine hemopoietic subpopulations for cytokinetic studies.

Multivariate analyses and list mode data processing were used to obtain cytokinetic information on flow cytometrically distinct hemopoietic subpopulations. In one application, viable, unfixed hemopoietic subpopulations were discriminated on the basis of cyanine dye fluorescence and orthogonal light scatter; Hoechst dye fluorescence was measured to determine the proliferative status of the subpopulations. In another application, ethanol-fixed mouse bone marrow cells were triply stained with Hoechst dye, rhodamine-conjugated wheat germ agglutinin (WGA), and a fluorescein-labeled monoclonal antibody against bromodeoxyuridine. In both of these studies, flow cytometric data for all three variables were acquired in list mode fashion, stored on magnetic tape, and processed by list mode software on a computer-based multivariable pulse-height analyzer. In the first application, subpopulations distinguished by cyanine dye intensity and light scatter appeared to be more related to cell lineage and cell size than proliferative status. In the second application, WGA affinity discriminated two subpopulations in mouse bone marrow S-phase cells in each subpopulation that actively incorporated bromodeoxyuridine (BrdUrd). List mode data processing obviates the need for routine electronic sorting of cells and thus facilitates characterization of discriminated subpopulations. In this regard, it is particularly useful for the study of flow cytometrically distinct, low frequency subpopulations.

Animals↗
PubMed1990

Multivariate analysis of quantitative chemical and enzymic characterization data in classification of Actinobacillus, Haemophilus and Pasteurella spp.

Chemotaxonomic data for strains of Actinobacillus, Haemophilus and Pasteurella spp. were analysed using three multivariate statistical strategies: principal components, partial least squares discriminant, and soft independent modelling of class analogy. The species comprised Actinobacillus actinomycetemcomitans. Haemophilus aphrophilus, H. paraphrophilus, H. influenzae, Pasteurella multocida, P. haemolytica and P. ureae. Strains were characterized by cell sugar and fatty acid composition, lysis kinetics during EDTA and EDTA plus lysozyme treatment, and methylene blue reduction. In total 23 quantitative variables were compiled from chemotaxonomic analyses of 25 strains. A. actinomycetemcomitans and H. aphrophilus formed distinct classes which differed from those of H. paraphrophilus, H. influenzae and Pasteurella spp. All characterization variables, except those describing fatty acid content, contributed significantly to inter-species discrimination.

Actinobacillus↗
PubMedDate not supplied

Growth of the body in early fetuses studied by multivariate analysis.

The growth of the body was studied in 30 human fetuses ranged from 10 to 22 weeks of gestation. The fetuses were fixed by immersion in 4 percent formaldehyde and the following dimensions were studied: a) lengths: arm, forearm, hand, thigh, leg, foot and crown-rump (sitting height), b) perimeters: head, thorax and abdomen. A covariance matrix was calculated from natural logarithms of all measurements. The relative growth of these measurement was computed by multivariate allometry using a principal components analysis (PCA). All characters were positively correlated with the first principal component which accounted for 94.65 per cent of the total variance. Considering the different measurements in the sequence of the increasing growth rates no one was considered to increase in isometric relationship. PCA showed that the following measurements grew with negative allometry: head perimeter, C-R length, thoracic perimeter, length of the forearm and abdominal perimeter. On the other hand, the following lengths grew with positive allometry: hand, foot, thigh, arm and leg. In conclusion, during the first two trimesters of prenatal life the growth of the body is allometrical. Limbs increase with greater growth rates than the perimeters of the body cavities.

Body Height↗
PubMed2006

Multivariate analysis for 1H-NMR spectra of two hundred kinds of tea in the world.

NMR measurements coupled with pattern-recognition analysis offer a powerful mixture-analysis tool for latent-feature extraction and sample classification. As fundamental applications of this analysis for mixtures, the 1H spectra of 176 kinds of green, black, oolong and other tea infusions were acquired by a 500 MHz NMR spectrometer. Each spectrum pattern was analyzed by a multivariate statistical pattern-recognition method where Principal Component Analysis (PCA) was used in combination with Soft Independent Modeling of Class Analogy (SIMCA). SIMCA effectively selected variables that contribute to tea categorization. The final PCA resulted in clear classification reflecting the fermentation and processing of each tea, and revealed marker variables that include catechin and theanine peaks.

Algorithms↗
PubMed1993

Multivariate analysis of risk factors for thrombus formation in University of Tokyo ventricular assist device.

Of 77 University of Tokyo ventricular assist devices used in a total of 70 patients at 21 institutions, 13 pumps were found to have macroscopic thrombus formations. Because 19 devices that were used for less than 24 hours showed no thrombus deposition, they were considered not to have been sufficiently exposed to the thrombogenic environment for macroscopic thrombus deposition and were removed from the subsequent multivariate study. A total of eight potential risk factors were assessed in relation to thrombosis. Prevalences of thrombus formation were compared between two groups with or without each of the risk factors. In a univariate analysis, the following categoric variables were demonstrated to be significantly associated with complications, in descending order of significance: use of gabexate mesilate (protease inhibitor) as an anticoagulant (p = 0.005), normal platelet count (p = 0.010), duration of support (p = 0.038), leukocytosis (p = 0.042), and minimum pumping flow (p = 0.042). Use of heparin and the consequent increase in activated clotting time showed no relationship. Multivariate discriminant analysis, which was done to identify risk factors rejecting cross correlation between each variable, demonstrated platelet count (p = 0.006), use of gabexate mesilate (p = 0.007), and minimum flow (p = 0.008) to have significant and independent risks. These results indicate the importance of maintaining pumping flow above a certain minimum level, addition of antiplatelet drugs to the antithrombogenic regimen, and nonuse of gabexate mesitate.

Adolescent↗
PubMed2004

Multivariate analysis of quantitative trait loci influencing variation in anxiety-related behavior in laboratory mice.

The number and mode of action of quantitative trait loci (QTL) that contribute to behavioral variation in rodents is still largely unknown. On theoretical grounds, multivariate techniques are expected to yield new insights into this problem, but there are only a few examples of its application in practice. Here we explore the power of multivariate approaches to uncover the genetic architecture of 23 anxiety-related phenotypes in 1636 F2 laboratory mice. We detected QTL with a genome-wide significance threshold of P < 0.05 on 14 chromosomes, of which 10 correspond to those identified by univariate analysis. Novel QTL were found on Chromosomes 3, 9, 13, and 17. Thus, multivariate analyses increased the yield of QTL exceeding a genome-wide significance threshold by 40%. On the basis of these results and by the application of a QTL estimator, we show that the mean number of QTL influencing anxiety-related behavior in mice is 6, with a 95% upper limit of 14.

Animals↗
PubMed1982

[Evaluation of vital prognosis in gas gangrene by a method of multivariate analysis of the results of the initial clinical examination].

Prognosis value of initial clinical symptoms and findings in gas gangrene was evaluated in 72 patients. Fifty eight parameters were recorded. The relationship between the simultaneous effects of several variables upon survival was accomplished using the multivariate technique, discriminant analysis. On the basis of observed patients deaths or survivals, the individual condition weights were statistically derived in a manner that best distinguishes likely survivors from nonsurvivors. Six variables out of the 58 studied gave right classification of outcomes in 97 p. 100 of the patients.

Adolescent↗
PubMed2004

Suicide risk among physicians: a multivariate analysis.

It has been hypothesized that previous research on physician suicide has neglected a possible suppressor effect on suicide from a low risk of divorce among physicians. The present article addresses this question using data on 143,885 deaths from the 1990 National Mortality Detail Files. The results of a multivariate logistic regression analysis indicated that when controls are introduced for marital status, gender, race, and other covariates of social status, physicians were 2.45 times more likely to die of suicide than the rest of the working age population. The results suggested that previous work, which had not controlled for the protective effect of marital status, might have underestimated the incidence of physician suicide.

Journal Article↗
PubMed1983

Multivariate analysis of prognostic variables in patients with metastatic testicular cancer.

A majority of patients with metastatic testicular cancer achieve a complete remission as a result of current treatment programs. However, patients who fail to achieve a complete remission have a very poor prognosis, and nearly all die of their disease. A multivariate logistic regression analysis of several clinical variables associated with prognosis was performed using data from 171 patients treated for metastatic testicular cancer at Memorial Hospital between September 1975 and February 1981. A mathematical model was identified which correctly predicted 94% of complete remissions and 83% of all outcomes. The variables achieving statistical significance were the logarithm of the serum values of lactate dehydrogenase (p less than 0.001) and human chorionic gonadotropin (p less than 0.001) and the total number of sites of metastasis (p less than 0.001). The model was tested against 49 patients with metastatic testicular cancer treated at the University of Minnesota Hospitals, and it correctly predicted 86% of complete remissions and 84% of all outcomes. In a highly curable disease such as testicular cancer, mathematical modeling may enable the clinical investigator to anticipate those patients who are least likely to do well. Alternate treatment strategies would be appropriate for such patients.

Antineoplastic Agents↗
PubMed2006

18FDG PET in vascular dementia: differentiation from Alzheimer's disease using voxel-based multivariate analysis.

The brain metabolic pattern of vascular dementia (VaD) remains poorly characterized. Univariate voxel-based analysis ignores the functional correlations among structures and may lack sensitivity and specificity. Here, we applied a novel voxel-based multivariate technique to a large ((18)F)2-fluoro-2-deoxy-D-glucose positron emission tomography data set. The sample consisted of 153 subjects, one-third each being probable subcortical VaD, probable Alzheimer disease (AD) (matched for Mini-Mental-State examination (MMSE) and age), and normal controls (NCs). We first applied principal component (PC) analysis and removed PCs significantly correlated to age. The remainders were used as feature vectors in a canonical variate analysis to generate canonical variates (CVs), that is, linear combinations of PC-scores. The first two CVs efficiently separated the groups. CV(1) separated VaD from AD with 100% accuracy, whereas CV(2) separated NC from demented subjects with 72% sensitivity and 96% specificity. Images depicting CV(1) and CV(2) showed that lower metabolism differentiating VaD from AD mainly concerned the deep gray nuclei, cerebellum, primary cortices, middle temporal gyrus, and anterior cingulate gyrus, whereas lower metabolism in AD versus VaD concerned mainly the hippocampal region and orbitofrontal, posterior cingulate, and posterior parietal cortices. The hypometabolic pattern common to VaD and AD mainly concerned the posterior parietal, precuneus, posterior cingulate, prefrontal, and anterior hippocampal regions, and linearly correlated with the MMSE. This study shows the potential of voxel-based multivariate methods to highlight independent functional networks in dementing diseases. By maximizing the separation between groups, this method extracted a metabolic pattern that efficiently differentiated VaD and AD.

Aged↗
PubMed1992

A multivariate analysis of the relations between craniofacial structure and occlusion with DiPaolo's quadrilateral measures.

This study examined the relations between craniofacial structure and occlusion with clinically oriented skeletal measures consistent with Enlow's counterpart principle and a combination of univariate and multivariate statistics. The pretreatment lateral cephalograms and dental casts of 59 white adolescents with natural skeletal and occlusal development were selected to provide a random sampling of various malocclusions. Craniofacial structure was described with DiPaolo's quadrilateral measures. The relations between craniofacial structure and occlusion were analyzed with simple correlation analysis, stepwise multiple regression, and stepwise discriminant analysis. Intratester and intertester reliability were established, and a lack of sexual dimorphism was demonstrated. Three types of relations were identified: anteroposterior relations, vertical relations, and coordinations. A strong association was found between the variation in the anteroposterior measures and the variation in sagittal occlusion. A more modest association was found between the variation in the vertical measures and the variation in vertical occlusion. A combined skeletal measure composed of the variable pair mandibular plane angle and the ratio of anterior lower facial height to anterior upper facial height was found to explain the greatest percent of the variance in vertical occlusion. Several coordinations were found which, when taken together, described an overall pattern of associations.

Adolescent↗