SEARCH · PubMed Health

Results for “correlation matrix”

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 505 records · Page 28Linked to original sources

PubMed2001

Algorithm for data clustering in pattern recognition problems based on quantum mechanics.

We propose a novel clustering method that is based on physical intuition derived from quantum mechanics. Starting with given data points, we construct a scale-space probability function. Viewing the latter as the lowest eigenstate of a Schrödinger equation, we use simple analytic operations to derive a potential function whose minima determine cluster centers. The method has one parameter, determining the scale over which cluster structures are searched. We demonstrate it on data analyzed in two dimensions (chosen from the eigenvectors of the correlation matrix). The method is applicable in higher dimensions by limiting the evaluation of the Schrödinger potential to the locations of data points.

Journal Article↗

PubMed2005

Entanglement detection by local orthogonal observables.

We propose a family of entanglement witnesses and corresponding positive maps that are not completely positive based on local orthogonal observables. As applications the entanglement witness of a 3x3 bound entangled state [P. Horodecki, Phys. Lett. A 232, 333 (1997)] is explicitly constructed and a family of dxd bound entangled states is introduced, whose entanglement can be detected by permuting local orthogonal observables. The proposed criterion of separability can be physically realized by measuring a Hermitian correlation matrix of local orthogonal observables.

Journal Article↗

PubMed1998

Multivariate autoregressive models for classification of spontaneous electroencephalographic signals during mental tasks.

This article explores the use of scalar and multivariate autoregressive (AR) models to extract features from the human electroencephalogram (EEG) with which mental tasks can be discriminated. This is part of a larger project to investigate the feasibility of using EEG to allow paralyzed persons to control a device such as a wheelchair. EEG signals from four subjects were recorded while they performed two mental tasks. Quarter-second windows of six-channel EEG were transformed into four different representations: scalar AR model coefficients, multivariate AR coefficients, eigenvalues of a correlation matrix, and the Karhunen-Loève transform of the multivariate AR coefficients. Feature vectors defined by these representations were classified with a standard, feedforward neural network trained via the error backpropagation algorithm. The four representations produced similar results, with the multivariate AR coefficients performing slightly better and more consistently with an average classification accuracy of 91.4% on novel, untrained, EEG signals.

Electroencephalography↗

PubMed2005

Encoding strategy for maximum noise tolerance bidirectional associative memory.

In this paper, the basic bidirectional associative memory (BAM) is extended by choosing weights in the correlation matrix, for a given set of training pairs, which result in a maximum noise tolerance set for BAM. We prove that for a given set of training pairs, the maximum noise tolerance set is the largest, in the sense that this optimized BAM will recall the correct training pair if any input pattern is within the maximum noise tolerance set and at least one pattern outside the maximum noise tolerance set by one Hamming distance will not converge to the correct training pair. This maximum tolerance set is the union of the maximum basins of attraction. A standard genetic algorithm (GA) is used to calculate the weights to maximize the objective function which generates a maximum tolerance set for BAM. Computer simulations are presented to illustrate the error correction and fault tolerance properties of the optimized BAM.

Electricity↗

PubMed2005

Convergence analysis of a deterministic discrete time system of Oja's PCA learning algorithm.

The convergence of Oja's principal component analysis (PCA) learning algorithms is a difficult topic for direct study and analysis. Traditionally, the convergence of these algorithms is indirectly analyzed via certain deterministic continuous time (DCT) systems. Such a method will require the learning rate to converge to zero, which is not a reasonable requirement to impose in many practical applications. Recently, deterministic discrete time (DDT) systems have been proposed instead to indirectly interpret the dynamics of the learning algorithms. Unlike DCT systems, DDT systems allow learning rates to be constant (which can be a nonzero). This paper will provide some important results relating to the convergence of a DDT system of Oja's PCA learning algorithm. It has the following contributions: 1) A number of invariant sets are obtained, based on which we can show that any trajectory starting from a point in the invariant set will remain in the set forever. Thus, the nondivergence of the trajectories is guaranteed. 2) The convergence of the DDT system is analyzed rigorously. It is proven, in the paper, that almost all trajectories of the system starting from points in an invariant set will converge exponentially to the unit eigenvector associated with the largest eigenvalue of the correlation matrix. In addition, exponential convergence rate are obtained, providing useful guidelines for the selection of fast convergence learning rate. 3) Since the trajectories may diverge, the careful choice of initial vectors is an important issue. This paper suggests to use the domain of unit hyper sphere as initial vectors to guarantee convergence. 4) Simulation results will be furnished to illustrate the theoretical results achieved.

Algorithms↗

PubMed2000

GEE with Gaussian estimation of the correlations when data are incomplete.

This paper considers a modification of generalized estimating equations (GEE) for handling missing binary response data. The proposed method uses Gaussian estimation of the correlation parameters, i.e., the estimating function that yields an estimate of the correlation parameters is obtained from the multivariate normal likelihood. The proposed method yields consistent estimates of the regression parameters when data are missing completely at random (MCAR). However, when data are missing at random (MAR), consistency may not hold. In a simulation study with repeated binary outcomes that are missing at random, the magnitude of the potential bias that can arise is examined. The results of the simulation study indicate that, when the working correlation matrix is correctly specified, the bias is almost negligible for the modified GEE. In the simulation study, the proposed modification of GEE is also compared to the standard GEE, multiple imputation, and weighted estimating equations approaches. Finally, the proposed method is illustrated using data from a longitudinal clinical trial comparing two therapeutic treatments, zidovudine (AZT) and didanosine (ddI), in patients with HIV.

Anti-HIV Agents↗

PubMed1994

Hydration, transepidermal water loss, pH and skin surface parameters: correlations and variations between dominant and non-dominant forearms.

The aim of this study was to examine differences between the skin of both forearms with regard to several parameters: hydration (skin capacitance), transepidermal water loss (TEWL), pH and skin surface parameters. Possible correlations between measurements were investigated. Thirty-three healthy women (32.6 +/- 8.4 years) participated in the study. The different parameters were measured on each volar forearm (dominant and non-dominant) at precisely delineated sites. TEWL and developed length (DL) were significantly increased on the dominant forearm (P < 0.001 and P < 0.01, respectively). In addition, a correlation matrix was calculated with the differences between dominant and non-dominant forearms for all the parameters measured. Only the skin surface parameters were found to be interrelated with this statistical approach. These data confirm that randomization between forearms is essential in the design of a reliable experimental plan, and indicate the lack of relationship between these measurements as assessed in a group of healthy young women.

Adult↗

PubMed1976

Sex-differences in the agonistic behavior of spiny-mice (Acomys cahirinus).

Observations of 8 groups each containing three adult spiny mice (Acomys cahirinus) revealed that instances of chasing and physical displacement were quite common, while fighting and biting were rarely observed. The relationships between the most frequent behavioral categories were presented in a correlation matrix. In a second study, females tested in their home cages were dominant over males. In the males' home cages, however, no differences in the frequency of female vs. male aggression were observed. More instances of aggression were observed in the cages of the females than in the cages of the males.

Aggression↗

PubMed1980

The assessment of protein-calorie malnutrition in patients with gastrointestinal cancer.

Daily protein and calorie intake, three plasma protein estimations, two upper limb anthropometric measurements, and estimated weight loss were all determined in 31 patients attending a surgical oncology outpatient department. The patients, who had had prior resection of a gastrointestinal cancer, could be divided into three groups depending on (i) the absence of clinically detectable tumour; (ii) the presence of clinically detectable tumour with survival over the ensuing four-month period; and (iii) clinically detectable tumour without survival over the ensuing four-month period. The variables least able to discriminate between these groups, and to interrelate to the other variable in a cross-correlation matrix, were the dietary intake data and the estimated weight loss. It is concluded that protein-calorie malnutrition can be adequately assessed in patients with advanced cancer from studies of the plasma albumin, prealbumin, transferrin, arm fat area, and arm muscle area.

Analysis of Variance↗

PubMed1994

Standardized in vitro mapping with multiple core biopsies of total prostatectomy specimens: localization and prediction of tumour volume and grade.

OBJECTIVE: To analyse the ability of systematic core biopsy mapping to provide prognostic information in patients with prostatic cancer. MATERIALS AND METHODS: The prostates of 60 men with prostatic cancer, stages T0d-T2, who had undergone total retropubic prostatectomy were studied. The average age of the patients was 63 years (range 49-72). Ten core biopsies (1.2 x 35 mm) were taken from the fresh specimens according to a standardized procedure. The total prostatectomy specimens were serially step-sectioned at 5 mm intervals and were assessed regarding tumour volume, grade and pT-stage. The World Health Organization (WHO) grade obtained in the mapping biopsies was compared with that in the operative specimens. Undergrading (WHO) decreased substantially by mapping biopsies, but was still present with the Gleason system. RESULTS: The volume of extracapsular tumours with extensively positive margins was significantly larger than that of intracapsular tumours (P < 0.01). In addition, the fraction of cancer obtained in the biopsies from tumours with grossly positive margins was significantly smaller than that observed in biopsies from pT2 tumours (P < 0.01). The cancer volume calculated from the result of the mapping correlated positively with the tumour volume determined by planimetry (R = 0.83). A weaker correlation was found when only the six dorsal mapping biopsies were taken into consideration (R = 0.68), but the correlation increased to R = 0.75 when the six most significant biopsies were selected with the help of a correlation matrix. Biopsies from the ventral part of the prostate were also important, to obtain an accurate assessment of the tumour fraction within the whole gland. CONCLUSION: Mapping of the prostate gland with multiple (six or more) core biopsies is necessary for preoperative assessment of tumour volume, grade and pT stage; these are all of importance when assigning patients with clinically localized prostatic cancer to prognostic classes.

Aged↗

PubMedDate not supplied

Developmental regulation of skull morphology II: ontogenetic dynamics of covariance.

Canalization may play a critical role in molding patterns of integration when variability is regulated by the balance between processes that generate and remove variation. Under these conditions, the interaction among those processes may produce a dynamic structure of integration even when the level of variability is constant. To determine whether the constancy of variance in skull shape throughout most of postnatal growth results from a balance between processes generating and removing variation, we compare covariance structures from age to age in two rodent species, cotton rats (Sigmodon fulviventer) and house mice (Mus musculus domesticus). We assess the overall similarity of covariance matrices by the matrix correlation, and compare the structures of covariance matrices using common subspace analysis, a method related to common principal components (PCs) analysis but suited to cases in which variation is so nearly spherical that PCs are ambiguous. We find significant differences from age to age in covariance structure and the more effectively canalized ones tend to be least stable in covariance structure. We find no evidence that canalization gradually and preferentially removes deviations arising early in development as we might expect if canalization results from compensatory differential growth. Our results suggest that (co)variation patterns are continually restructured by processes that equilibrate variance, and thus that canalization plays a critical role in molding patterns of integration.

Anatomy, Comparative↗

PubMed1994

Cognitive screening of nursing home residents: factor structures of the Mini-Mental State Examination.

OBJECTIVE: To examine factor structures of the Mini-Mental State Examination, attempting first to replicate any of previously proposed 2-factor solutions; and to explore, secondly, the presence of clinically more differentiated and statistically stable factor structures representing common neurocognitive dimensions. DESIGN: Factor analytic investigation of descriptive dataset collected on nursing home residents. Two factor analyses were performed, one in which the number of factors was fixed at 2 in an effort to replicate previous studies, and one in which the number of factors to retain was determined by the scree test. Both factor analyses used established methods for judging the adequacy of the correlation matrix and the significance of factor loadings, and both applied principal components analysis for initial factor extraction and the equamax criterion for orthogonal rotation. SETTING: Seven nursing homes with a total of 894 beds. PARTICIPANTS: 922 assessments on nursing home residents were performed, of which 892 were complete and entered into the factor analyses. The observation-to-variable ratio exceeded 81:1, assuring the statistical stability of factor solutions derived. MEASUREMENT: The Mini-Mental State Examination, with standardization of words to be recalled and the inverted spelling of "world" as the mental reversal task. MAIN RESULTS: Two factor structures were derived. A 2-factor solution, explaining 36.5% of the variance and statistically and conceptually different from those obtained in previous studies, distinguished between Perceptual-Organizational and Psychomotor skills. A 4-factor solution, which explained 56.1% of the variance, included a factor named Executing Psychomotor Commands, while also further differentiating the perceptual-organizational processes into the factors of Memory, Concentration, and Language. CONCLUSION: The 2-factor solution shows that, notwithstanding previous claims to the contrary, the MMSE can make stable and independent distinctions between psychomotor and perceptual-organizational processes. However, this solution is statistically and conceptually limited and, therefore, of limited clinical and scientific relevance. The 4-factor solution of the MMSE maps well onto commonly recognized dimensions of neurocognitive ability. It offers a stable, intuitively sound, and statistically supported framework for clinical differentiation of cognitive screening data into independent clinical dimensions of neurocognitive functioning. Thus, it offers clinicians and researchers a 4-dimensional framework for interpreting data obtained by means of the MMSE. Studies with other populations of cognitively impaired and intact elderly are recommended to validate and extend the present findings.

Aged↗

PubMed2005

A randomized crossover trial comparing three plateletpheresis machines.

BACKGROUND: The aim of the study was to compare three different apheresis machines with the same donors regarding the processing time required to obtain a 3.5 x 10(11) platelet (PLT) dose and acceptance by donors. STUDY DESIGN AND METHODS: A randomized crossover trial was performed to evaluate the differences between the Amicus Crescendo (Baxter Biotech Corp.), the MCS Plus (Haemonetics Corp.), and the Trima Accel (Gambro BCT). Donations from 51 donors were compared for time adjusted to obtain a standard 3.5 x 10(11) PLT dose (TSD3.5), efficiency, adverse reactions, yield, leukodepletion, machine accuracy, and donor preferences. Processing times were measured by chronometer. The same vein access was used during all three processes in each donor. In the statistical analysis, to take into account the nonindependence of several measurements from the same donor, generalized estimating equations were used with an autoregressive correlation matrix. RESULTS: The Accel produced a TSD3.5 (mean +/- SEM) of 47.9 +/- 1.0 min; the Amicus, 60.3 +/- 1.0 min; and the MCS Plus, 66.7 +/- 1.0 (p < 0.0001). The Amicus presented the greatest efficiency (87.5%; p < 0.0028). The MCS Plus demonstrated the highest capacity for leukodepletion (p < 0.0002) despite one process presenting more than 1 x 10(6) white blood cells per unit. The MCS Plus also measured the processing time with the greatest accuracy. No severe adverse effects were observed. The donors preferred the Accel (61%) followed by the Amicus (35%) and the MCS Plus (4%; p < 0.0001) and the processing speed was the most highly valued measure (55%). CONCLUSIONS: The Accel is the fastest and, because of this advantage, the machine preferred by donors. The Amicus was the most efficient and the MCS Plus was the only one not to underestimate the processing time.

Cross-Over Studies↗

PubMed2006

Hematomas in multicomponent apheresis: searching for related factors.

BACKGROUND: The objective of this study was to identify donation variables that could be related to the development of hematoma during multicomponent apheresis collections (MACs). STUDY DESIGN AND METHODS: This is an observational retrospective study where 1375 donors donated 5177 MACs during a 2-year period with two different machines (Amicus Crescendo [AC], Baxter Healthcare Corp.; and Trima Accel [TA], Gambro BCT). Variable data were recorded prospectively. In the multiple logistic regression analysis, generalized estimating equations were used with an exchangeable correlation matrix to take into account the nonindependence of several measurements from the same donor. RESULTS: During the study period, 170 procedures failed due to hematoma (3.3%). Several variables were related to hematoma development in the adjusted model: operator experience (less than 500 procedures supervised vs. more; odds ratio [OR], 1.66; 95% confidence interval [CI], 1.19-2.31), previous apheresis donations (first time vs. more than 16 donations; OR, 2.87; 95% CI, 1.52-5.45), vein canalized (basilic vs. intermediate antebrachial or cephalic; OR, 1.42; 95% CI, 1.04-1.94), diastolic blood pressure (units divided by 10 mmHg; OR, 0.79; 95% CI, 0.66-0.94), and type of machine used (TA high return limit configuration [RLC] setting configuration vs. AC; OR, 1.94; 95% CI, 1.27-2.96; TA low RLC setting configuration vs. AC; OR, 1.24; 95% CI, 0.83-1.84). CONCLUSIONS: Our study suggests that hematoma in MAC is not a random event. Appropriate machine configuration in the TA could reduce the hematoma rate to a level comparable with that of the AC. Operator training and donor blood pressure are also interesting variables for study because these could be modified to reduce the hematoma rate.

Adolescent↗

PubMed2000

"Factor analyses of the family assessment device," by Ridenour, Daley, & Reich.

Ridenour, Daley, and Reich conducted a series of factor analyses using the correlational matrix of the subscale scores of the Family Assessment Device (FAD), published in Family Process, December, 1999. They conclude that "the FAD subscales be reorganized from their current seven-subscale format" (p. 507). We propose that this suggestion for reorganization is premature and based on the inappropriate application of an "internal consistency" model of scale construction to the FAD. We further suggest that the most important criteria regarding an assessment instrument are those of reliability, validity, and clinical utility. In the absence of this kind of data regarding alternative organizations of the FAD, we believe that the original subscales remain the best choice.

Factor Analysis, Statistical↗

PubMed2004

Quantitative validation of a general competency composite assessment evaluation.

OBJECTIVES: The authors sought to modify and validate a composite assessment evaluation process that assesses resident acquisition of the Accreditation Council for Graduate Medical Education (ACGME) general competencies (GCs). METHODS: This study critically analyzed the evaluation process used in a multicenter study (150 emergency medicine resident evaluations) to determine whether the procedure was psychometrically valid. For each GC, principal component analysis (PCA) was used to determine whether certain evaluation items could be eliminated, as well as to determine the magnitude of variability explained by up to three linear combinations or "principal components." The factor proportions (factor loadings) of various eigenvectors were measured to determine the degree of variability (determined by the square of the factor proportion) within a data or item set. The factor proportions essentially measure the length of the eigenvector as determined from a correlation matrix. RESULTS: The first three principal components are reported as factor proportion sum (% of total variability) as follows: patient care 0.91 (83%), medical knowledge 0.87 (76%), practice-based learning and improvement 0.90 (81%), interpersonal and communication skills 0.84 (71%), professionalism 0.74 (55%), and systems-based practice 0.80 (64%). PCA showed that evaluating certain traditional categories such as medical knowledge seemed to capture a single element, whereas professionalism appeared to measure a more complex, multidimensional phenomenon. CONCLUSIONS: By using a structured development process, the authors were able to create valid evaluation items for determining resident acquisition of the ACGME GCs.

Clinical Competence↗

PubMed1980

Stepwise regression analysis of an intensive 1-year study of delirium tremens.

An intensive 1-year study was carried out on 41 male patients, mean age 49, mean hospitalization time 49 days, admitted to a special ward of the Beckomberga Hospital with the diagnosis of delirium tremens and 50 concomitant somatic and psychiatric diagnoses (1--9 per capita), and given a standardized treatment. The mean duration of delirium tremens after admission was 2 days; 76% recovered within 48 h. The duration after admission was positively correlated to age, number of previous delirium tremens, negatively correlated to B-haemoglobin and B-haematocrit for laboratory data obtained within the first 24 h and was positively correlated to blood sugar and S-creatinine on data taken within 40 h (Pearson correlation matrix). Stepwise multiple regression (SWR) based on 46 quantitative and dummy variables (the latter used to represent the presence of various concomitant diseases) was employed to identify the factors predicting the duration of delirium tremens. On final SWR analysis, which limited the number of observations to cases with complete observation vectors, the following regression equation was obtained: Duration after admission = 3.57--0.93 (S-magnesium)--0.29 (B-eosinophils) + 0.62 (liver disease), P greater than 0.05, n = 14. Although the regression coefficients were not statistically significant, S-magnesium, negatively associated with the duration after admission, offered 20% out of the total 38% of explanation given, whereas B-eosinophils, negatively associated, offered 12%, and liver disease, positively associated, 6%. The choice by the SWR program of S-magnesium as the most important factor in predicting the duration of delirium tremens is consistent with clinical evidence that alcohol ingestion causes magnesium diuresis and that magnesium deficiency is present in chronic alcoholism. In view of this knowledge, it is reasonable to assume that the lack of statistical significance is due to the small sample size rather than to the alternative that no explanation is offered by S-magnesium. Furthermore, B-haemoglobin, S-potassium, S-ASAT, and S-ALAT, known to be characteristically altered in delirium tremens, were found on forcing (a variant of SWR) to be of secondary importance to S-magnesium as explaining factors, whereas blood sugar and S-creatinine derived part of their explaining power from S-magnesium. In conclusion, extensive use of SWR analysis based on 46 potential explaining variables points to serum magnesium concentration as the most important factor in predicting the duration of delirium tremens.

Adult↗

PubMed1980

Monitoring and evaluation of results of community fluoride programs in Norway during the 1960's and 1970's.

The purpose of this paper was to report on the introduction, functioning and results of community fluoride programs in Norway during the 1960's and 1970's. Data on the treatment carried out in the School and Public Dental Services were available in official publications. Information on systematic community topical fluoride programs for priority groups was collected by postal questionnaire from all School and Public Dental Officers. Nine variables were described and used to construct a correlation matrix. During the 1970's, there has been an increase in rinsing and professional topical application at the expense of brushing programs. Though the participation-rate increased steadily among 7--15-year-olds and had reached 60% by 1970 and 90% by 1977. the mean number of surfaces filled did not show a consistent downward trend till 1972/73. The reduction in mean number of surfaces filled form 1972 to 1976 tended to be biggest in countries with a high mean number of surfaces filled in 1972 (r = 0.79). Regional variations in mean number of surfaces filled have decreased markedly during the 1970's. Possible explanations for this and problems concerning collection of data on preventive activities are discussed.

Adolescent↗