Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “canonical correlation analysis”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 181 records · Page 10Linked to original sources
PubMed1976

Comparison of left and right ventricular function in acute myocardial infarction.

Pulmonary arterial end-diastolic and mean right atrial pressures were compared in 25 patients with acute myocardial infarction and in one patient with unstable angina. No consistent relationship was observed between these pressures. Simultaneous ventricular function curves relating the stroke work of each ventricle to its respective filling pressure were constructed on 34 occasions, dextran infusion or diuresis being used to alter the filling pressure. The curves from each ventricle were described mathematically by a quadratic (parabolic) function as well as by a straight line function and then compared by canonical correlation analysis. Alterations in the left ventricular function curves occurred with and without depression or right ventricular function curves. These hemodynamic measurements demonstrate that acute myocardial infarction can alter the relationship between left and right ventricular function.

Adult↗
PubMedDate not supplied

Cognitive dysfunction associates with white matter hyperintensities and subcortical atrophy on magnetic resonance imaging of the elderly diabetes mellitus Japanese elderly diabetes intervention trial (J-EDIT).

BACKGROUND: Type 2 diabetes is associated with cognitive dysfunction and increases the risk of dementia in the elderly. The aim of this study was to explore, by means of magnetic resonance (MR) imaging, possible relationships among clinical profiles of diabetes, cognitive function, white matter hyperintensities (WMHs) and subcortical brain atrophy. METHODS: Data were obtained from 95 nondemented type 2 diabetic participants aged 65 years or over, enrolled in an intervention trial for Japanese elderly diabetic patients. Cognitive function was measured with neuropsychiatric tests, including mini-mental state examination (MMSE), verbal memory, digit symbol substitution and Stroop tests. Hyperintensity was classified into periventricular, deep white matter, thalamic and basal ganglia. Four ventricle-to-brain ratios were used to measure subcortical atrophy. To identify clinical features of diabetes, indices of glycemic control, lipid metabolism, blood pressure and complications were examined. Canonical correlation analysis and regression analysis were used to assess correlation. RESULTS: Scores for digit symbol substitution and MMSE negatively correlated with WMHs in the parietal lobe and hyperintensities in the thalamus, respectively. Lower scores for memory and digit symbol substitution showed positive association with enlarged subcortical atrophy adjacent to lateral ventricles. There was no association between clinical pictures of diabetic patients with cognitive dysfunction and of those with morphological changes in the brain. CONCLUSIONS: Impaired cognitive domains of the speed of mental processes and memory were associated with WMHs and subcortical atrophy. Degenerative changes in the cerebral small vessels may constitute predictive factors for the rate of cognitive dysfunction in elderly diabetic patients.

Aged↗
PubMed2003

Novel ROC-type method for testing the efficiency of multivariate statistical methods in fMRI.

The receiver operating characteristic (ROC) method is a useful and popular tool for testing the efficiency of various diagnostic tests applicable to functional MRI (fMRI) data. Typically, the diagnostic tests are applied on simulated and pseudo-human fMRI data, and the area under the ROC curve is used as a measure of the efficiency of the diagnostic test. The effectiveness of such a method depends on how well the simulated data approximate the real data. For multivariate statistical methods, however, this technique is usually inadequate, as the spatial dependence among voxels is ignored for simulated data. In this work a modified ROC method using real fMRI data with a broader scope is proposed. This method can be applied to most fMRI postprocessing techniques, including multivariate analyses such as canonical correlation analysis (CCA). Also, the relationship of the modified ROC method with the conventional ROC method is discussed in detail.

Algorithms↗
PubMed2005

The relationship of maternal attributes, resources, and perceptions of postpartum experiences to depression.

The relationships of maternal attributes, resources, and perceptions of the postpartum experience to postpartum depression (PPD) were examined. One hundred and forty-three mothers completed measures of maternal attributes, resources, and perceptions of the postpartum experience, as well as depressed mood and clinical depression symptoms. Results from canonical correlation analysis revealed two patterns. Pattern 1: Women without history of depression, more emotional support from partners, higher parenting evaluation, lower centrality of the infant in the mother's thoughts and actions, and fewer life changes had lower depressed mood and fewer clinical depression symptoms than their counterparts. Pattern 2: Mothers without history of depression who were married, with higher financial stress, and lower parenting evaluation had higher depressed mood, but not more clinical depression symptoms. Maternal age, parity, time since delivery, income, and help from partner were unrelated to either depression indicator. These patterns suggest multiple paths to PPD, and the need for routine mental health assessment and exploration of women's perceptions of their postpartum experience.

Adolescent↗
PubMed1997

The "Poffenberger" and "Dimond" paradigms: interrelated approaches to the study of interhemispheric dynamics?

Brown and Jeeves (1993) found that an evoked potential estimate of interhemispheric transfer time, from the left to the right hemisphere, correlated negatively with a bilateral field advantage (BFA) in a response-choice letter matching task. We implemented a go no-go dot size matching task to determine whether the crossed-uncrossed difference (CUD) in reaction time, commission errors and omission errors (estimates of the "cost" of interhemispheric transfer) would correlate with BFAs in data from the same experiment and whether the type of decision ("same" versus "different") would modulate the CUDs and/or BFAs. Sixteen normal right-handed subjects were tested. The CUDs were negatively correlated with the BFAs. Canonical correlation analysis of this set of relations was highly significant (r = .95). Estimates of left-to-right relay were far more strongly related to BFA (p = .0001) than were estimates of right-to-left relay (p = .03). "Same" decisions yielded a unilateral field advantage and "different" decisions a bilateral field advantage in omission error data, this crossed interaction reaching significance. More efficient interhemispheric relay favored BFAs, i.e., strongly suggesting in such cases an advantage of interhemispheric over intrahemispheric integration. This effect appeared to be markedly asymmetric. Furthermore, resource sharing within and between the hemispheres was a function of the "same" versus "different" dimensions of the decision to be made, especially in omission errors. The opposite dissociation occurred less markedly in the reaction times, this double dissociation reaching significance, revealing presence of a subtle speed-accuracy trade-off in interhemispheric dynamics.

Adult↗
PubMed1984

The assessment of changes in brain volume using combined linear measurements. A CT-scan study.

All linear measurements employed for evaluation of brain atrophy, were performed on 148 computed tomograms of patients aged 28 to 84 without evidence of any nervous system disorder. These included size of lateral, third and fourth ventricles, width of the Sylvian and frontal interhemispheric fissures and cortical sulci and size of the pre-pontine cistern. Various parameters indicated decrease in brain mass with age. Since the atrophic process is a diffuse phenomenon, integration of several measurements evaluating separate brain regions was made. The bicaudate ratio and the Sylvian fissure ratio (representing both central and cortical atrophy) were combined arithmetically, resulting in a correlation of 0.6390 with age (P less than 0.0005). With a computed canonical correlation analysis; a formula was obtained which combined measurements of the lateral and third ventricles, the Sylvian fissure and the pre-pontine cistern. This formula yielded a correlation of 0.67795 (P less than 0.0005). These linear measurements will allow simple and reliable assessement of reduction in brain volume during the normal aging process and in disorders accompanied by brain atrophy.

Adult↗
PubMed1985

Features of brain atrophy in Parkinson's disease. A CT scan study.

Multiple parameters for brain volume and mass were studied in 85 parkinsonian patients and in 149 normal controls aged 24 to 84 using CT scanning. In controls there was reduction in brain substance with advancing age. Increased brain atrophy in patients with Parkinson's disease (PD) was mainly observed in the younger age group of 24 to 49. This included parameters evaluating the size of the lateral and third ventricles (P less than 0.05 and P less than 0.01 respectively) and the size of the subarachnoid space in the frontal interhemispheric and Sylvian fissures (P less than 0.005 and P less than 0.01). With computed canonical correlation analysis a formula was obtained which expressed the tendency of the atrophic process in PD to involve the areas surrounding the third ventricle and the mesial aspect of the frontal lobes more than during normal aging.

Adult↗
PubMed1998

Measures of hostility as predictors of facial affect during social interaction: evidence for construct validity.

We assessed the construct validity of several self-report measures and an interview-based measure of hostility (Interpersonal Hostility Assessment Technique [IHAT]) by evaluating their associations with a behavioral indicator of hostile emotions (facial expressions during social interaction). Participants in the study were 123 volunteers (44% males and 56% females) who were recruited from local community organizations. Self-report measures (Cook-Medley Hostility Scale, Rotter Interpersonal Trust Scale, Buss-Durkee Hostility Inventory, and Spielberger Anger Expression Scale) were represented by factor scores reflecting Overt Hostility, Covert Hostility, and Hostile Beliefs. A canonical correlation analysis identified significant associations between a set of facial affect scores reflecting animosity and various measures of hostility. Specifically, increases in anger and disgust expressions and decreases in happy facial expressions were associated with high IHAT scores and high scores on self-report measures of Hostile Beliefs and Covert Hostility. Women were more expressive than men, especially concerning positive affect, and women had lower scores on self-report measures of Hostile Beliefs and Overt Hostility. IHAT scores were uncorrelated with any of the self-report factors which suggests the two assessment techniques are tapping different aspects of the hostility construct.

Adult↗
PubMed1996

Multisignal evaluation in ICP-MSDetermination of trace elements in Mo-Zr-alloy.

The existence of isobaric and polyatomic ion interferences has been considered to be a significant shortcoming of inductively coupled plasma - mass spectrometry (ICP-MS). In the special case of Mo-Zr-alloys problems caused by polyatomic ions are substantial. The determination of trace elements by using classical univariate linear regression of single signals or any correction equations presented by the instrumental software is impossible. There are various possibilities to overcome spectral interferences, in principle. The application of high-resolution ICP-MS, separation methods or alternative sample introduction systems is limited by the high cost and the apparative expenditure. In this report methods of multisignal calibration, like partial least squares (PLS) regression and canonical correlation analysis (CCA), are used to overcome isobaric or chemical interferences in ICP-MS. It is shown, that PLS and CCA are able to handle or tolerate, respectively, spectral interferences existing in a Mo-Zr-O-H-F system. The results of quantitative analysis provided by PLS and CCA were comparable.

Journal Article↗
PubMed2003

Exploring two methods for statistical downscaling of Central European phenological time series.

In this study we set out to investigate the possibility of linking phenological phases throughout the vegetation cycle, as a local-scale biological phenomenon, directly with large-scale atmospheric variables via two different empirical downscaling techniques. In recent years a number of methods have been developed to transfer atmospheric information at coarse General Circulation Model's grid resolutions to local scales and individual points. Here multiple linear regression (MLR) and canonical correlation analysis (CCA) have been selected as downscaling methods. Different validation experiments (e.g. temporal cross-validation, split-sample tests) are used to test the performance of both approaches and compare them for time series of 17 phenological phases and air temperatures from Central Europe as microscale variables. A number of atmospheric variables over the North Atlantic and Europe are utilized as macroscale predictors. The period considered is 1951-1998. Temporal cross-validation reveals that the CCA model generally performs better than MLR, which explains 20%-50% of the phenological variances, whereas the CCA model shows a range from 40% to over 60% throughout most of the vegetation cycle. To show the validity of employing phenological observations for downscaling purposes both methods (MLR and CCA) are also applied to gridded local air temperature time series over Central Europe. In this case there is no obvious superiority of the CCA model over the MLR model. Both models show explained variances from 40% to over 70% in the temporal cross-validation experiment. The results of this study indicate that time series of phenological occurrence dates are very compatible with the needs of empirical downscaling originally developed of local-scale atmospheric variables.

Atmosphere↗
PubMed2004

Correlations of life-span variation parameters in 128 successive generations of Drosophila melanogaster with changes in atmospheric pressure and geomagnetic activity.

Correlations between the parameters of life-span (LS) distribution of Drosophila melanogaster, including mean LS (MLS) and the time of 10 and 90% population mortality, and some geophysical parameters that are usually beyond the control of researchers dealing with laboratory cultures, including atmospheric pressure, solar activity indices (Wolf's sunspot numbers and 2,800-MHz radio flux), and geomagnetic activity (planetary index, K(p)), were studied. Geophysical data were obtained from free-access official web sites of the National Oceanic & Atmospheric Administration of the US Department of Commerce and the Institute of Terrestrial Magnetism and Radiowave Propagation of the Russian Academy of Sciences. The geophysical parameters were calculated only for the period corresponding to 10 days of preimaginal development of the flies from egg to imago. Canonical correlation analysis, calculation of the non-parametric Spearman rank-order correlation coefficients, and graphical data analysis were used. Highly significant correlations between parameters of LS distribution in males and females and environmental factors, such as the atmospheric pressure on the 4th and 5th day of development and geomagnetic activity indices (K(p)) on the 6th and 10th day of development were found, with correlation coefficients varying from 0.31 to 0.37 (P<0.02). Assuming a causal relationship between geophysical factors and LS, it may be hypothesized that energetically weak environmental factors determine the formation of LS oscillatory dynamics in laboratory populations. The possible mechanisms underlying the contribution of these environmental factors to the LS variation in successive generations are discussed.

Animals↗
PubMed1989

Beta-endorphin and components of emotionality discriminate between physically active and sedentary men.

Differences between physically active and sedentary men were tested by profile comparison. The study identifies the relative importance of circulating beta-endorphin (BE), atherosclerotic disease risk (ADR) index, and selected components of emotionality in discriminating between physically active and sedentary men. The subjects were psychologically normal and medically healthy middle-aged men. Jogging activity was the subject classification criterion. The data were collected on selected physiological (treadmill), biochemical (blood collected from resting subjects), and psychological (Eysenck and MMPI) variables. The physical fitness score (PFS) was used as an index of fitness. Physically active men with a high PFS (n = 21), when compared to the sedentary men with a low PFS (n = 15), exhibited lower basal plasma BE, lower ADR, lower anxiety index (AI), and lower MMPI depression score (D). Canonical correlation analysis showed that PFS and BE in one set were correlated with D and neuroticism (NS) in another set of variables. Discriminant function analysis showed that the AI was the most powerful discriminator between the physically active and sedentary men, followed by BE and NS. Interestingly, BE and NS exhibited the same magnitude of discrimination power. The ADR exhibited less discrimination power, relative to AI, BE, and NS. In conclusion, the physically active men, compared to the sedentary men in this study, exhibited lower basal plasma BE, which appeared to be associated with less atherosclerotic disease risk, less neuroticism, less anxiety, and less depression.

Adult↗
PubMed1996

Intra-individual patterns of hormonal and affective adaptation to work demands: an n = 2 study of junior doctors.

This paper examines the detailed pattern of hormonal and affective response to natural variations in work demands, in relation to a taxonomy of adaptive response based on Frankenhaeuser's (1986) psychobiological analysis of coping modes and Hockey's (1993) regulatory control model. Two junior doctors were monitored every day over a 7-week period of work on a cardio-thoracic surgery ward. Measurements were made for each morning and afternoon work period of self-rated workload, effort and affective state, and the level of urinary catecholamines and cortisol. The de-trended data were analysed separately for the two individuals, using multivariate methods. Following reduction of work variables by principal components analysis, canonical correlation analyses were carried out for each individual. These revealed differences in the patterns of adaptation to two distinctive work contexts (enabling and demanding work) across the two doctors, as identified through loadings on two significant pairs of canonical variates. As expected, enabling work (high medical demands, with high personal resources) was associated with active coping in both subjects (low fatigue, high effort/adrenaline and low cortisol). However, demanding work (high general demands) was associated with marked differences between them in the pattern of loadings. One subject showed the strain pattern normally associated with effortful engagement in difficult tasks (high anxiety and fatigue, high effort/adrenaline), though without the anticipated reduction in cortisol. The response pattern of the other individual was indicative of passive coping (high anxiety and cortisol, with no effort/adrenaline component). The findings are interpreted in terms of the role of personal coping styles on the adaptive response to work demands. The use of canonical correlation methods with intra-individual data sets, although relatively unusual, appears to provide potentially valuable evidence on the nature of individual differences in the process of psychobiological response to stress underlying work-health relationships.

Adaptation, Psychological↗
PubMed1980

Differential alcohol use patterns and personality traits among three Alcoholics Anonymous attendance level groups: further considerations of the affiliation profile.

The present study was designed to identify personality and drinking behavior correlates associated with membership in Alcoholics Anonymous (A.A.). Affiliation profiles were developed on the basis of frequency of A.A. attendance; subtypes were defined as non, low, and high affiliates, respectively. An initial canonical correlation analysis assessed the amount of common variance between personality and drinking characteristics. Subsequent discriminant function analyses on the second-order scales of the drinking and personality variables sets identified those variables differentiating the A.A. affiliated groups. High levels of affiliation were associated with a greater degree of anxiety, a tendency to be more affected by feelings rather than intellect, and more deterioration in the physical, exhibited significant differences in perceived benefits, style, and consequences of alcohol usage. The potential therapeutic utility of A.A. affiliation profiles was discussed.

Alcohol Drinking↗
PubMed1994

Relationship of body fat distribution with cardiovascular risk factors in healthy Chinese.

The relationships between six body girths (shoulder, midarm, waist, hip, thigh, and calf) and cardiovascular risk factors (systolic and diastolic blood pressures and glucose, triglyceride, lipoprotein cholesterol, and apolipoprotein levels) were examined in 407 healthy Chinese urban workers in Taipei, Taiwan who were between 40 to 59 years old. Canonical correlation analysis revealed significant associations of upper body adiposity (shoulder, midarm, and waist girths) with cardiovascular risk factors in all subgroups assessed: men, premenopausal women, and postmenopausal women. Waist girth and hip girth were consistent and important variables, and weighted in the opposite direction. Waist-hip ratio (WHR) was the best descriptor of centralized adiposity. Centralized fat distribution was positively associated with blood pressure and glucose, triglyceride, and apolipoprotein (apo) B levels, and negatively associated with high-density-lipoprotein (HDL) cholesterol and apo A-I levels in this population. Body fat distribution had an effect independent of body mass index and accounted for some of the differences in triglyceride, HDL cholesterol, apo A-I, and apo B concentrations among men, premenopausal women, and postmenopausal women. Our findings in a Chinese population are similar to data from other studies in Western populations, and are consistent with the hypothesis that centralized adiposity is related to cardiovascular risk factors independent of general obesity.

Adipose Tissue↗
PubMed2007

Relationship between quality of life and child traffic fatalities.

Traffic injuries are the leading cause of morbidity and mortality of children around the world. Economic development and quality of life are main components of public health. Using canonical correlation analysis, this study examined the relationship between quality of life and traffic fatality among children. Data were collected from 19 countries by using World Health Organization and OECD statistics. The results indicated that there is a strong relationship between quality of life and children traffic fatality. Growing quality of life has become protective against child traffic accident mortality. Increasing in gross national income per capita, life expectancy at birth number of years, road motor vehicles per thousand population, and share of persons of working age (15-64 years) in employment also increases children traffic safety.

Accidents, Traffic↗
PubMed2006

Germination polarity of Beauveria bassiana conidia and its possible correlation with virulence.

Three different germination types of conidia; unidirectional, bidirectional and multidirectional, were revealed through microscopic observations for eight Beauveria bassiana isolates germinated on Sabouraud dextrose agar. Canonical correlation analysis indicated that there is a positive correlation between unidirectional germination and virulence against diamondback moth, Plutella xylostella and the Colorado potato beetle, Leptinotarsa decemlineata. Scanning electron microscopy revealed different in vivo behaviors for unipolar- and bipolar-germinated conidia. Unipolar-germinated conidia produced a strong germ tube with mostly appressorium-like structures while bipolar-germinated conidia continued to invasive hyphal growth without any penetration, indicating that germination polarity in one way or another may be an indicator of pathogenic ability.

Animals↗
PubMed2006

Multisubject activation detection in fMRI by testing correlation of data with a signal subspace.

In this article, a generalized likelihood ratio test is proposed to assess the correlation between multisubject functional MRI (fMRI) time series and bases of a signal subspace for detecting the existence of group activation in each voxel of the brain. The signal subspace is generated by a design matrix using the time series of the desired effects. The proposed method leads to testing the product of eigenvalues of a specific matrix. The eigenvector corresponding to the largest eigenvalue is the weighting vector for the linear combination of time series of various subjects that has the maximum correlation with the signal subspace. In another method, namely, canonical correlation analysis, the largest eigenvalue of the above matrix is tested for activation detection. Surrogate data on resting state (no activation) are generated by randomization and used to estimate the statistical distribution of these parameters under the null hypothesis condition. A postprocessing step is applied to prevent false detection of voxels that are not sufficiently active (among subjects) by defining a minimum ratio for the active population. The proposed methods are applied on simulated and experimental fMRI data, and the results are compared with those of the general linear model (GLM; using the SPM and FMRISTAT toolboxes). The proposed methods showed higher detection sensitivity as compared with the GLM for activation detection in simulated data. Similarly, they detected more activated regions than did the GLM from multisubject experimental fMRI data on a visual (sensorimotor) event-related task.

Algorithms↗