Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Multivariate statistics”

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 253 records · Page 14Linked to original sources
PubMed1980

[Place of multivariate statistics in clinical physiological research].

On the basis of their own experience of many years in the use of mathematical methods in medicine the authors provide a critical analysis of the results of using computers in making the diagnosis and the main reasons for dissatisfaction with the results. Regarding clinico-physiological studies as a field for the application of modern mathematical statistics, they suggest a logical scheme which they had tested time and again for statistical analysis of the data by multidimensional methods (particularly with the use of the method of principal components, cluster analysis, latent analysis, lambda-moments, etc.) and also some methodological devices which permit keeping the data at hand throughout the entire analysis, resorting time and again to all possibilities for their complete and exhaustive description. The position of the authors in principle consists in the fact that success can only be achieved by constant joint work of the medical specialist and mathematician, beginning with the first stage of formulating both the medical and statistical tasks on condition that the clinico-physiological essence of the problem is comprehended by the mathematician.

Diagnosis, Computer-Assisted↗
PubMed1996

Chromosome 8 allelic loss and the outcome of patients with squamous cell carcinoma of the supraglottic larynx.

BACKGROUND: Loss of genetic heterogeneity (allelic loss or loss of heterozygosity) on chromosome arm 8p is frequent in squamous cell carcinomas of the head and neck and has been associated with poor prognosis. We have previously demonstrated that there are three minimal regions of allelic loss on this chromosome arm. The location of each region is marked by a microsatellite locus: D8S264 (8p23), D8S552 (8p23-p22), and D8S133 (8p21). These findings imply the existence of at least three putative tumor suppressor genes on this chromosome arm that may become inactivated during the progression of squamous cell carcinoma. PURPOSE: We used allelic loss data from these three loci to determine if inactivation of these putative suppressors is associated with poor prognosis for patients with squamous cell carcinoma of the supraglottic larynx. We also used multivariate statistics to compare the prognostic power of allelic loss at these genetic markers with that of demographic, clinical, and histopathologic parameters. METHODS: We examined the D8S264, D8S552, and D8S133 microsatellites in tumors from a retrospective population of 59 patients. All patients had histologically confirmed squamous cell carcinoma of the supraglottic larynx and had been treated surgically. DNA was extracted from matched sets of normal and microdissected tumor tissue and used for polymerase chain reaction amplification of the microsatellite markers. Reaction products were separated by denaturing gel electrophoresis and visualized by autoradiography. Patient data were obtained from the original pathology report and from the tumor registry of the Department of Otolaryngology-Head and Neck Surgery, Washington University School of Medicine, St. Louis, MO. Histopathologic data were obtained by reviewing the portion of the resection specimen used for DNA extraction. Parameters whose association with reduced disease-free interval and reduced disease-specific survival was statistically significant were identified by use of the Kaplan-Meier method and the logrank statistic. Multivariate Cox proportional hazards models were used to identify independent predictors of poor prognosis. All statistical tests were two-sided. RESULTS: In this patient population, allelic loss at the D8S264 locus was associated with both shorter disease-free interval (logrank P = .028) and reduced disease-specific survival (logrank P = .004). Allelic loss at the next most centromeric locus, D8S552, had a statistically significant association with only reduced disease-specific survival (logrank P = .034), whereas allelic loss at the most centromeric region, D8S133, showed no statistically significant association with reductions in either interval. Multivariate Cox models suggested that D8S264 was the only 8p marker of the three microsatellites with a statistically significant and independent association with shortened disease-free interval (relative risk [RR] = 3.38; P = .0107) and reduced disease-specific survival (RR = 3.41; P = .0105). CONCLUSIONS: Allelic loss in the p23 region of chromosome 8 appears to be a statistically significant, independent predictor of poor prognosis in patients with supraglottic squamous cell carcinoma.

Alleles↗
PubMed1997

Assessing medical students' perceptions of mistreatment in their second and third years.

PURPOSE: To study medical students' perceptions of mistreatment in their second and third years of training. METHOD: A questionnaire was distributed at Wayne State University School of Medicine to the class of 1993 at the end of its third year and to the class of 1994 at the end of its second and third years. The students were asked if they had been subjected to various forms of mistreatment; the third-year students were asked to rate their perceptions of each clinical department's response to them on the basis of gender and race-ethnicity, as well as their overall treatment. The students also completed demographic information about age, gender, and marital status, number of children, and race-ethnicity. Results were analyzed using chi-square statistics, multivariate statistical analyses, analyses of variance, and Duncan's post-hoc comparisons. RESULTS: The response rate for the class of 1993 was 71.5%; response for the class of 1994 were 66.9% in their second year and 75.2% in their third year; 41.7% were women, and the racial-ethnic breakdown was 71.2% white/Caucasian, 11.7% black/African American and 16.8% other. There was a significant difference between the percentages of second-year and third-year students reporting any experience of mistreatment (37.2% vs 75.8%, p < .001). Canonical correlation analysis revealed bias in the third year based on gender (p < .0001) and race-ethnicity (p < .0002); both variates were related to sexual humor. The students' perceptions of mistreatment were lowest for family medicine and highest for obstetrics-gynecology and surgery. Perceptions of mistreatment in departments varied significantly by gender and race-ethnicity. The nonwhite males reported the least favorable treatment in most departments. CONCLUSION: Marked variability in the students' perceptions of mistreatment within departments suggest that a variety of approaches will be required to improve the medical training environment.

Education, Medical, Undergraduate↗
PubMed1989

[A novel method for determining an integrated character in quantitative genetics].

This paper describes a novel method using maximum entropy principle introduced by Shannon to construct a unique integrated character and gives a comparison between this method and multivariate statistics in quantitative genetics. In multivariate genetic analysis multivariate statistics is usually used for a polygenic system of multiple quantitative traits to form several complex character. Genetic analysis can be implemented through treating these correlated elementary traits by using the technique of principal component analysis or canonical correlation. Differing from the multivariate statistics, this paper presents a new method using maximum-entropy principle to obtain a unique integrated character from a set of correlated quantitative traits. This method has the advantages of simple mathematical structure and process, simple and clear results.

Animals↗
PubMed1998

Impact of distal aortic and visceral perfusion on liver function during thoracoabdominal and descending thoracic aortic repair.

PURPOSE: We examined the impact of distal aortic and visceral perfusion on liver function during thoracoabdominal and descending thoracic aortic repair. METHODS: Between January 1991 and July 1996, 367 patients underwent thoracoabdominal and descending thoracic aortic repair. Baseline and postoperative total bilirubin, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, lactate dehydrogenase, fibrinogen, prothrombin time (PT), and partial thromboplastin time (PTT) were measured for 286 patients. We examined the impact of distal aortic and direct visceral perfusion on liver function-related clinical laboratory values. Univariate and multivariate statistical methods for categorical and continuous variables were used. RESULTS: In categorical analysis, type II thoracoabdominal aortic aneurysm, history of hepatitis, and emergency presentation had a statistically significant multivariate association with abnormal laboratory values. In continuous-distributed multivariate data analysis, type II thoracoabdominal aortic aneurysm and visceral perfusion were statistically significant predictors of postoperative alkaline phosphatase, PT, and PTT. Type II aneurysms increased postoperative liver function-related laboratory values significantly above other aneurysm types (alkaline phosphatase, +114 IU, p < 0.0001; PT, +1.99 seconds, p < 0.02; PTT, +6.7 seconds, p < 0.03). Visceral perfusion was associated with a concomitant decrease (alkaline phosphatase, -101.2 IU, p < 0.0001; PT, -1.8 seconds, p < 0.07; PTT, -5.6 seconds, p < 0.02). CONCLUSIONS: Visceral perfusion negates the rise in postoperative liver function-related clinical laboratory values associated with type II thoracoabdominal aortic aneurysm repair.

Alkaline Phosphatase↗
PubMed1998

Analysis of image sequences in fluorescence videomicroscopy of stationary objects.

Fluorescence videomicroscopy allows the temporal behavior of biological specimens to be studied at the cellular level. We describe two types of methods that can be used for extracting quantitative information from image sequences: the modeling approach, which is mainly local, and multivariate statistical analysis, which provides a global approach. The potentials for use of these two methods are illustrated through a simulation example and actual examples dealing with the study of chloride secretion by airway epithelial cells. We define some guidelines for making a choice between these two approaches, bearing in mind that a blend of these two methodologies is also possible.

Cells, Cultured↗
PubMed2001

Source apportionment of gaseous atmospheric pollutants by means of an absolute principal component scores (APCS) receptor model.

A multivariate statistical method has been applied to apportion the atmospheric pollutant concentrations measured by automatic gas analyzers placed on a mobile laboratory for air quality monitoring in Taranto (Italy). In particular, Principal Component Analysis (PCA) followed by Absolute Principal Component Scores (APCS) technique was performed to identify the number of emission sources and their contribution to measured concentrations of CO, NOx, benzene toluene m+p-Xylene (BTX). This procedure singled out two different sources that explain about 85% of collected data variance.

Air Pollutants↗
PubMed2001

Overdispersion in modelling accidents on road sections and in empirical bayes estimation.

In multivariate statistical models of road safety one usually finds that the accident counts are 'overdispersed'. The extent of the overdispersion is itself subject to estimation. It is shown that the assumption one makes about the nature of overdispersion will affect the maximum likelihood estimates of model parameters. If one assumes that the same overdispersion parameter applies to all road sections in the data base, then, the maximum likelihood estimate of parameters will be unduly influenced by very short road sections and insufficiently influenced by long road sections. The same assumption about the overdispersion parameter also leads to an inconsistency when one estimates the safety of a road section by the Empirical Bayes method. A way to avoid both problems is to estimate an overdispersion parameter (phi) that applies to a unit length of road, and to set the overdispersion parameter for a road section of length L to phiL. How this would change the estimates of regression parameters for road section models now in use requires examination. Safety estimation by the Empirical Bayes method is altered substantially.

Accidents, Traffic↗
PubMed2001

The redfish species Sebastes viviparus, Sebastes marinus and Sebastes mentella have different composition of their tissue fatty acids.

The fatty acid profile in the tissue of heart, gill, skull- and otolith-oil of the three redfish species, Sebastes viviparus, S. marinus and S. mentella was determined by a chemometric method, consisting of methanolysis of samples of the tissues and of the oils and gas chromatography of the resulting fatty acid methyl esters. The analytical data were treated by multivariate statistics comprising principal component analysis (PCA) and soft independent modeling of class analogy (SIMCA). Although the differences in fatty acid composition among the two tissues and two oils were the dominating features of the data, the three species had significantly different fatty acid profiles within each tissue. Variation among the individuals was considerable. The fatty acid profiles appeared to be species specific. The mutual relationship between S. marinus and S. mentella is closer than the relationship between either of them and S. viviparus. A comparison with a corresponding investigation on the same three species from another location indicates that the observed differences do not seem to be affected by biotic or abiotic factors.

Animals↗
PubMed1996

An assessment of the risks of syphilis and HIV infection among a sample of not-in-treatment drug users in Houston, Texas.

The research presented in this paper details the results of an assessment of the risk factors associated with having a positive syphilis or HIV serology. The study was conducted using a sample of not-in-treatment drug users volunteering to participate in an HIV risk reduction intervention. The sample was composed of individuals who had injected drugs within 30 days or smoked crack cocaine 48 hours prior to participation in the study. Study participants were approximately 75% male and 66% African-American. All participants provided a blood sample to be tested for HIV and syphilis. Analysis of risk was conducted using univariate and multivariate statistical methods. Multivariate analysis of blood results showed that women, African-Americans, and those having a positive blood test for HIV were at higher odds of having a positive syphilis test. Analysis also showed that being a gay or bisexual male, having a history of drug injection, having less than a high-school education, having a history of trading sex for money, being African-American, and having a positive blood test for syphilis significantly increased the odds of a positive HIV test. Implications for HIV and STD prevention are discussed.

Adult↗
PubMed2003

A multivariate approach applied to microarray data for identification of genes with cell cycle-coupled transcription.

We have analyzed microarray data using a modeling approach based on the multivariate statistical method partial least squares (PLS) regression to identify genes with periodic fluctuations in expression levels coupled to the cell cycle in the budding yeast, Saccharomyces cerevisiae. PLS has major advantages for analyzing microarray data since it can model data sets with large numbers of variables and with few observations. A response model was derived describing the expression profile over time expected for periodically transcribed genes, and was used to identify budding yeast transcripts with similar profiles. PLS was then used to interpret the importance of the variables (genes) for the model, yielding a ranking list of how well the genes fitted the generated model. Application of an appropriate cutoff value, calculated from randomized data, allows the identification of genes whose expression appears to be synchronized with cell cycling. Our approach also provides information about the stage in the cell cycle where their transcription peaks. Three synchronized yeast cell microarray data sets were analyzed, both separately and combined. Cell cycle-coupled periodicity was suggested for 455 of the 6,178 transcripts monitored in the combined data set, at a significance level of 0.5%. Among the candidates, 85% of the known periodic transcripts were included. Analysis of the three data sets separately yielded similar ranking lists, showing that the method is robust.

Algorithms↗
PubMed2003

MHCPred: A server for quantitative prediction of peptide-MHC binding.

Accurate T-cell epitope prediction is a principal objective of computational vaccinology. As a service to the immunology and vaccinology communities at large, we have implemented, as a server on the World Wide Web, a partial least squares-based multivariate statistical approach to the quantitative prediction of peptide binding to major histocom- patibility complexes (MHC), the key checkpoint on the antigen presentation pathway within adaptive cellular immunity. MHCPred implements robust statistical models for both Class I alleles (HLA-A*0101, HLA-A*0201, HLA-A*0202, HLA-A*0203, HLA-A*0206, HLA-A*0301, HLA-A*1101, HLA-A*3301, HLA-A*6801, HLA-A*6802 and HLA-B*3501) and Class II alleles (HLA-DRB*0401, HLA-DRB*0401 and HLA-DRB*0701). MHCPred is available from the URL: http://www.jenner.ac.uk/MHCPred.

Binding Sites↗
PubMed2001

Association of surgical specialty and processes of care with patient outcomes for carotid endarterectomy.

BACKGROUND AND PURPOSE: Because there is considerable variation in practice patterns and outcomes for carotid endarterectomy (CE), there is a need to study the processes of care that are associated with adverse outcomes. The purpose of this study was to examine the impact of processes of care and surgical specialty on adverse outcomes for CE. METHODS: A retrospective cohort study based on a voluntary CE registry containing 3644 patients undergoing CE between April 1, 1997, and March 31, 1999, in New York hospitals was used in the study. A multivariable statistical model was used to identify significant independent patient risk factors and to examine the association of processes of care and surgical specialty with outcomes after adjustment for differences in patient risk factors. RESULTS: The overall adverse outcome (in-hospital death or stroke) rate was 1.84%. After adjustment for differences in 7 patient risk factors that were significantly related to adverse outcomes, the use of >/=1 specific processes of care (eversion endarterectomy, protamine, or shunts) was found to be associated with lower odds of an adverse outcome relative to patients undergoing CE without the processes (OR=0.42, P=0.006). Similarly, patients undergoing surgery performed by vascular surgeons had lower odds of experiencing an adverse outcome (OR=0.36, P=0.009). Processes of care and surgical specialty were highly correlated with one another. CONCLUSIONS: Processes of care and surgical specialty are significant interrelated determinants of adverse outcome for CE.

Aged↗