SEARCH · PubMed Health

Results for “data 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 433 records · Page 24Linked to original sources

PubMed2002

Functional data analysis with application to periodically stimulated foetal heart rate data. I: functional regression.

Functional regression is used to model longitudinal data where the number of measurements on the functional covariate is much greater than the number of subjects in the study. Thus, functional regression can be thought of as singular longitudinal analysis. We have modified existing functional regression techniques to the case of a functional covariate with a repeated stimulus. We applied this modified functional regression to periodically stimulated foetal heart rates. The heart rate tracings were used as a predictor of the child's psychomotor development at approximately 18 months of age. In the past, this type of data has been analysed using the partially subjective concept of habituation. By using the entire heart rate tracings through functional regression, we have the advantage that habituation does not need to be defined and all available information is used to predict later child development.

Adult↗

PubMed1999

A case study report on integrating statistics, problem-based learning, and computerized data analysis.

This paper addresses the pedagogical advantages of teaching statistics not as a stand-alone subject in itself, but rather as a topic integrated into teaching hands-on, problem-based computer-assisted data analysis. For over 10 years, such a two-term course has been taught at Drexel University in lieu of the usual statistics courses formerly taken by undergraduate majors in psychology and sociology. One virtue of the courses as currently implemented is that students seem to learn not just how to perform statistical procedures but how to apply them on their own.

Electronic Data Processing↗

PubMed2006

Global characterization of coronary plaque rupture phenotype using three-vessel intravascular ultrasound radiofrequency data analysis.

AIMS: To compare the global characteristics of patients with and without evidence of plaque rupture (PR) in their coronary tree and to evaluate the phenotype of ruptured plaques using intravascular ultrasound (IVUS) radiofrequency data analysis (IVUS-VH). METHODS AND RESULTS: Forty patients underwent three-vessel IVUS-VH interrogation. Twenty-eight PRs were diagnosed in 26 vessels (25.7% of the vessels studied) of 20 patients (50% of the population). Ruptures located in the left anterior descending were clustered in the proximal part of the vessel, whereas ruptures located in the right coronary artery were more distally located (P=0.02). Patients with at least one PR presented larger body mass index (BMI) (28.4+/-3.7 vs. 25.8+/-2.6 kg/m(2), P=0.01) and plaque burden (40.7+/-7.6 vs. 33.7+/-8.4%, P=0.01) than patients without rupture, despite showing similar lumen cross-sectional area (9.6+/-3.3 vs. 9.2+/-2.3 mm(2), P=0.60). Among current smokers, 66.7% presented a PR in their coronary tree. Finally, PR sites showed a higher content of necrotic core compared with minimum lumen area sites (17.48+/-10.8 vs. 13.10+/-6.5%, P=0.03) and a trend towards higher calcified component. CONCLUSION: Patients with at least one PR in their coronary tree presented larger BMI and worse IVUS-derived characteristics compared with patients without PR.

Body Mass Index↗

PubMed2003

Randomizing patients by family practice: sample size estimation, intracluster correlation and data analysis.

BACKGROUND: Cluster randomized controlled trials increasingly are used to evaluate health interventions where patients are nested within larger clusters such as practices, hospitals or communities. Patients within a cluster may be similar to each other relative to patients in other clusters on key variables; therefore, sample size calculations and analyses of results require special statistical methods. OBJECTIVE: The purpose of this study was to illustrate the calculations used for sample size estimation and data analysis and to provide estimates of the intraclass correlation coefficients (ICCs) for several variables using data from the Seniors Medication Assessment Research Trial (SMART), a community-based trial of pharmacists consulting to family physicians to optimize the drug therapy of older patients. METHODS: The study was a paired cluster randomized trial, where the family physician's practice was the cluster. The sample size calculation was based on a hypothesized reduction of 15% in mean daily units of medication in the intervention group compared with the control group, using an alpha of 0.05 (one-tailed) with 80% power, and an ICC from pilot data of 0.08. ICCs were estimated from the data for several variables. The analyses comparing the two groups used a random effects model for a meta-analysis over pairs. RESULTS: The design effect due to clustering was 2.12, resulting in an inflation in sample size from 340 patients required using individual randomization, to 720 patients using randomization of practices, with 15 patients from each of 48 practices. ICCs for medication use, health care utilization and general health were <0.1; however, the ICC for mean systolic blood pressure over the trial period was 0.199. CONCLUSIONS: Compared with individual randomization, cluster randomization may substantially increase the sample size required to maintain adequate statistical power. The differences in ICCs among potential outcome variables reinforce the need for valid estimates to ensure proper study design.

Aged↗

PubMed1996

A review of crash data analysis in a defect and recall investigation of the general motors C/K pickup trucks.

In the process of assessing the safety of the fuel-containment system of the 1973-1987 General Motors C/K truck, the National Highway Traffic Safety Administration and General Motors Corporation have written and submitted numerous documents to a public file between October 1992 and April 1993. Five substantial reports have been issued by the National Highway Traffic Safety Administration and General Motors Corporation describing data analysis of crashes recorded on state and federal databases. Both the National Highway Traffic Safety Administration and General Motors Corporation have used crash data, in some cases the same data, to examine the claim that a defect in the fuel system design of General Motors Corporation C/K trucks poses an unreasonable risk of death or injury. The comparative analysis presented in this paper demonstrates how crash databases and their summary statistics can be used to support opposed positions in a safety dispute. Understanding differences in the analysis is fundamental to obtaining an insight into the role of field crash data, including its relevance and shortcomings, in defect and recall investigations.

Accidents, Traffic↗

PubMed1984

Permutation methods for the structured exploratory data analysis (SEDA) of familial trait values.

A collection of functions that contrast familial trait values between and across generations is proposed for studying transmission effects and other collateral influences in nuclear families. Two classes of structured exploratory data analysis (SEDA) statistics are derived from ratios of these functions. SEDA-functionals are the empirical cumulative distributions of the ratio of the two contrasts computed within each family. SEDA-indices are formed by first averaging the numerator and denominator contrasts separately over the population and then forming their ratio. The significance of SEDA results are determined by a spectrum of permutation techniques that selectively shuffle the trait values across families. The process systematically alters certain family structure relationships while keeping other familial relationships intact. The methodology is applied to five data examples of plasma total cholesterol concentrations, reported height values, dermatoglyphic pattern intensity index scores, measurements of dopamine-beta-hydroxylase activity, and psychometric cognitive test results.

Adolescent↗

PubMed2006

Restless legs syndrome in hemodialysis patients: health-related quality of life and laboratory data analysis.

AIMS: To compare clinical data, sleep quality and health-related quality of life (HRQOL) with and without RLS in HD patients. MATERIALS AND METHODS: The international RLS study group diagnosis questionnaire was completed by 228 HD patients. The Pittsburg Sleep Quality Index (PSQI) for the evaluation of sleep quality and the Kidney Disease Quality of Life (KDQOL-SF) for the analysis of HRQOL were also used. RESULTS: 53 (23%) patients were diagnosed as RLS. Age and age at the initiation of HD were significantly younger in the RLS group. Serum calcium concentration (Ca) was significantly higher in the RLS group. Sleep quality evaluated by PSQI was significantly lower in the RLS group. In SF-36 domains of KDQOL-SF, bodily pain, general health perceptions, vitality, role functioning emotional, mental health and mental component score were significantly lower in the RLS group. In kidney targeted scales of KDQOL-SF, symptoms/problems, burden of kidney disease, cognitive function, quality of social interaction, sleep and patient satisfaction were significantly lower in the RLS group. CONCLUSION: High Ca was possibly connected to the pathophysiology of RLS which impaired sleep quality as well as HRQOL including mental health and many kidney disease related scales.

Adult↗

PubMed2006

An alternative competing risk model to the Weibull distribution for modelling aging in lifetime data analysis.

A simple competing risk distribution as a possible alternative to the Weibull distribution in lifetime analysis is proposed. This distribution corresponds to the minimum between exponential and Weibull distributions. Our motivation is to take account of both accidental and aging failures in lifetime data analysis. First, the main characteristics of this distribution are presented. Then, the estimation of its parameters are considered through maximum likelihood and Bayesian inference. In particular, the existence of a unique consistent root of the likelihood equations is proved. Decision tests to choose between an exponential, Weibull and this competing risk distribution are presented. And this alternative model is compared to the Weibull model from numerical experiments on both real and simulated data sets, especially in an industrial context.

Aging↗

PubMed2005

Patient characteristics and clinical management of patients with shoulder pain in U.S. primary care settings: secondary data analysis of the National Ambulatory Medical Care Survey.

BACKGROUND: Although shoulder pain is a commonly encountered problem in primary care, there are few studies examining its presenting characteristics and clinical management in this setting. METHODS: We performed secondary data analysis of 692 office visits for shoulder pain collected through the National Ambulatory Medical Care Survey (Survey years 1993-2000). Information on demographic characteristics, history and place of injury, and clinical management (physician order of imaging, physiotherapy, and steroid intraarticular injection) were examined. RESULTS: Shoulder pain was associated with an injury in one third (33.2% (230/692)) of office visits in this population of US primary care physicians. Males, and younger adults (age < or = 52) more often associated their shoulder pain with previous injury, but there were no racial differences in injury status. Injury-related shoulder pain was related to work in over one-fifth (21.3% (43/202)) of visits. An x-ray was performed in 29.0% (164/566) of office visits, a finding that did not differ by gender, race, or by age status. Other imaging (CT scan, MRI, or ultrasound) was infrequently performed (6.5%, 37/566). Physiotherapy was ordered in 23.9% (135/566) of visits for shoulder pain. Younger adults and patients with a history of injury more often had physiotherapy ordered, but there was no significant difference in the ordering of physiotherapy by gender or race. Examination of the use of intraarticular injection was not possible with this data set. CONCLUSION: These data from the largest sample of patients with shoulder pain presenting to primary care settings offer insights into the presenting characteristics and clinical management of shoulder pain at the primary care level. The National Ambulatory Medical Care Survey is a useful resource for examining the clinical management of specific symptoms in U.S. primary care offices.

Adult↗

PubMed2002

Testing model fit in longitudinal data analysis against alternatives with omitted covariates.

Several types of common model misspecifications can be re-formulated as problems of omitted covariates. These include situations with unmeasured confounders, measurement errors in observed covariates and informative censoring. Longitudinal data present special opportunities for detecting omitted covariates that are related to the observed ones differently across time than across individuals. This situation arises with period and cohort effects, as well as with usual formulations of classical measurement error in observed covariates. In this article we focus on testing for the existence of omitted covariates in longitudinal data analysis when models are fit by generalized estimation equations. When omitted covariates are present, specification of the correct link function conditionally on only observed covariates under the alternative usually involves complicated numerical integration. We propose a quasi-score test statistic that avoids the need to fit such alternative models. The statistic is asymptotically chi-square distributed under the null hypothesis of no omitted covariates with degrees of freedom determined by the assumed alternative structure. We study the significance level and the power of the quasi-score test in linear and logistic regression models. The test is then applied to an analysis of excessive daytime sleepiness.

Cohort Studies↗

PubMed1994

Computer assisted clinical data analysis.

The Diagnosis Related Groups patient classification scheme coupled with desk top PC technology permits sophisticated analysis of patient medical data. Individuals with no programming knowledge can produce sophisticated analysis. The functionality and structure of the 3M Analytical Workstation are described and example analysis reports are presented.

Computer Systems↗

PubMedDate not supplied

A JAVA environment for medical image data analysis: initial application for brain PET quantitation.

Analysis software for medical image data tends to be expensive and usable only in a restricted environment. Therefore the aim of the current project was to implement a flexible framework for medical image processing and visualization which is portable among platforms and open to different data formats including DICOM 3.0. The software was designed as a set of tools which encapsulate specialized functionality. The tools are full stand alone applications, but they are also able to present a co-operating environment within which images and other information are communicated in real time. Currently, the emphasis is on quantitative analysis of PET data by kinetic modelling. However, general viewing capabilities are included, and the design is flexible enough that other types of processing can easily be integrated by simply plugging in Java classes. The software is successfully applied to PET data quantitation in clinical research studies and even patient studies. Portability was aimed at by encoding the programs in Java. Experience shows that the implementation of such a complex and computationally demanding Java application is feasible. Although there are major portability issues to consider, configurations can be found on which the software runs stably and at a speed comparable to C code if just-in-time compilation is available.

Brain↗

PubMed2002

Visualization methods for data analysis and planning in medical applications.

Time plays an important role in medicine, both the past and the future. The medical history of a patient represents the past, which needs to be understood by the physician to make the right decisions. The past contains two different kinds of information: measured data (such as blood pressure) and incidents (such as seizures). Planning therapies, on the other hand, requires looking into the future to a certain extent. Visual representations exist for both the past and the future, and they are very useful for getting a better understanding of data or a plan. This paper surveys visualization techniques for both data analysis and planning, and compares them based on a number of criteria.

Computer Graphics↗

PubMed2002

Correlation of aqueous solubility of salts of benzylamine with experimentally and theoretically derived parameters. A multivariate data analysis approach.

Twenty two salts of benzylamine and p-substituted benzoic acids were prepared and characterized. The p-substituent was varied with regard to electronic, hydrophobic, and steric effects as well as hydrogen bonding potential. A multivariate data analysis was used to describe the relationship between the aqueous solubility of the salts and experimentally determined physicochemical parameters and theoretically derived molecular descriptors. The model, based on all descriptors, gave R(2)=0.86 and Q(2)=0.72. The most significant descriptors exhibiting VIP (variance of importance) values above 1.0 were intrinsic dissolution rate, intrinsic solubility of the unionized acids (S(0)), Hansch's hydrophobic parameter, Charton's steric parameter and molecular weight (MW). Statistically good models for predicting solubility of a selected test set were obtained by using simple models consisting of a few descriptors only: (i) Charton, Hansch and MW (R(2)=0.73; Q(2)=0.70), and (ii) Charton and S(0) (R(2)=0.74; Q(2)=0.72).

Benzylamines↗

PubMed1996

[Transformations of parameters in the generalized Poisson distribution for test data analysis].

The generalized Poisson distribution is a distribution which approximates various forms of mixtures of Poisson distributions. The mean and variance of the generalized Poisson distribution, which are simple functions of the two parameters of the distribution, are more useful than the original parameters in test data analysis. Therefore, we adopted two types of transformations of parameters. The first model has new parameters of mean and standard deviation. The second model contains new parameters of mean and variance/mean. An example indicates that the transformed parameters are convenient to understand the properties of data.

Adult↗

PubMed1998

Absolute metabolite quantification by in vivo NMR spectroscopy: V. Multicentre quantitative data analysis trial on the overlapping background problem.

The goal of this study was to establish the best approach for quantifying nuclear magnetic resonance (NMR) lines, that in the frequency domain are overlapping with broad, unwanted background features. To perform the quantitative data analysis in a controlled way, test signals were designed and utilised, derived from two different real-world in vivo nuclear magnetic resonance signals. One of the main conclusions of the study was that the quantification methods currently available to the biomedical research groups can deliver the correct values of the quantitative parameters, but that great care should be taken in using optimal input parameters for the computer programs concerned.

Bayes Theorem↗

PubMed2003

Bioinformatics meets proteomics--bridging the gap between mass spectrometry data analysis and cell biology.

Proteomics research programs typically comprise the identification of protein content of any given cell, their isoforms, splice variants, post-translational modifications, interacting partners and higher-order complexes under different conditions. These studies present significant analytical challenges owing to the high proteome complexity and the low abundance of the corresponding proteins, which often requires highly sensitive and resolving techniques. Mass spectrometry plays an important role in proteomics and has become an indispensable tool for molecular and cellular biology. However, the analysis of mass spectrometry data can be a daunting task in view of the complexity of the information to decipher, the accuracy and dynamic range of quantitative analysis, the availability of appropriate bioinformatics software and the overwhelming size of data files. The past ten years have witnessed significant technological advances in mass spectrometry-based proteomics and synergy with bioinformatics is vital to fulfill the expectations of biological discovery programs. We present here the technological capabilities of mass spectrometry and bioinformatics for mining the cellular proteome in the context of discovery programs aimed at trace-level protein identification and expression from microgram amounts of protein extracts from human tissues.

Computational Biology↗

PubMed2003

CGH-Plotter: MATLAB toolbox for CGH-data analysis.

CGH-Plotter is a MATLAB toolbox with a graphical user interface for the analysis of comparative genomic hybridization (CGH) microarray data. CGH-Plotter provides a tool for rapid visualization of CGH-data according to the locations of the genes along the genome. In addition, the CGH-Plotter identifies regions of amplifications and deletions, using k-means clustering and dynamic programming. The application offers a convenient way to analyze CGH-data and can also be applied for the analysis of cDNA microarray expression data. CGH-Plotter toolbox is platform independent and requires MATLAB 6.1 or higher to operate.

Cluster Analysis↗