Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “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 487 records · Page 27Linked to original sources
PubMed1999

ROC analysis of statistical methods used in functional MRI: individual subjects.

The complicated structure of fMRI signals and associated noise sources make it difficult to assess the validity of various steps involved in the statistical analysis of brain activation. Most methods used for fMRI analysis assume that observations are independent and that the noise can be treated as white gaussian noise. These assumptions are usually not true but it is difficult to assess how severely these assumptions are violated and what are their practical consequences. In this study a direct comparison is made between the power of various analytical methods used to detect activations, without reference to estimates of statistical significance. The statistics used in fMRI are treated as metrics designed to detect activations and are not interpreted probabilistically. The receiver operator characteristic (ROC) method is used to compare the efficacy of various steps in calculating an activation map in the study of a single subject based on optimizing the ratio of the number of detected activations to the number of false-positive findings. The main findings are as follows: Preprocessing. The removal of intensity drifts and high-pass filtering applied on the voxel time-course level is beneficial to the efficacy of analysis. Temporal normalization of the global image intensity, smoothing in the temporal domain, and low-pass filtering do not improve power of analysis. Choices of statistics. the cross-correlation coefficient and t-statistic, as well as nonparametric Mann-Whitney statistics, prove to be the most effective and are similar in performance, by our criterion. Task design. the proper design of task protocols is shown to be crucial. In an alternating block design the optimal block length is be approximately 18 s. Spatial clustering. an initial spatial smoothing of images is more efficient than cluster filtering of the statistical parametric activation maps.

Brain↗
PubMed2001

Statistical comparison of dissolution curves.

The purpose of this investigation was to develop statistical procedures to determine if two sets of dissolution curves could have come from the same population of curves. The f(2)statistic developed by the Food and Drug Administration, FDA, has been shown to have many limitations and is too liberal in concluding similarity between dissolution profiles. The procedure currently used by the FDA involves computing the mean amount dissolved at each time and then comparing the two mean curves. This approach ignores all of the variability within sets of profiles, which, from a statistical viewpoint, is a serious limitation. This investigation presents three different statistics for comparison of dissolution curves with associated decision rules and power functions. These three statistics are extensions of existing procedures: (1) an extension of the Mann--Whitney test which compares the variability within each set of profiles and between the two sets; (2) an extension of the Kolmogorov--Smirnov D statistic which compares three empirical cumulative distribution functions; and (3) an adaptation of the well known chi-squared test. A computer program, which includes the algorithm for each of the three statistics and varying sample sizes, is also available.

Algorithms↗
PubMed1977

[Statistical methods in medicine: in instruction, consultancy and research (author's transl)].

This survey covers the use of statistical methods in three areas of medicine: (1) the obligatory instruction of students in statistics, (2) the statistical consulting of clinicians and their candidates for a doctor's degree and (3) the modification and creation of statistical methods by the statisticians in the departments of medical statistics and documentation. Moreover some factors are discussed that reduce the effectiveness of statistical consulting, may it be service, advisory work or collaboration. Other sections of this paper cover the calamity of utilization and evaluation of old clinical records, the avoidance of fallacies and of errors committed by the inappropriate and incorrect application of statistical methods in medicine. Supplementary questions are given as a "first help" in tackling problems in medicine.

Education, Medical, Graduate↗
PubMed2006

Stereotactic statistical imaging analysis of the brain using the easy Z-score imaging system for sharing a normal database.

Statistical brain imaging analysis has good objectivity and reproducibility. In Japan, statistical parametric mapping (SPM) and three-dimensional stereotactic surface projections (3D-SSP) are used nationwide as statistical imaging analysis with standard brain coordinates. They often help to interpret brain single photon emission computed tomography (SPECT) images by avoiding possible pitfalls (e.g., effects of aging, atrophy) with which clinicians are unfamiliar. However, this type of analysis presents a problem: statistical processing requires many normal subject images. The easy Z-score imaging system (eZIS) is one of the statistical analysis methods that uses SPM processing in normalization and smoothing, and it has the function of image conversion leading to statistical analysis without a control database. Therefore, statistical analysis can be used in clinical practice by sharing a prepared normal database. By unifying the image quality by processing a shared database, this program has great potential for sharing patient imaging data in many hospitals. It is expected that the eZIS will help perform detailed analysis in many functional diseases in collaborative studies. This article describes the interpretation of brain SPECT images and suggests the usefulness and potential of eZIS.

Aging↗
PubMed1996

Applications of statistical quality control to cardiac surgery.

BACKGROUND: Although originally developed for use in manufacturing statistical quality control techniques may be applicable to other frequently performed, standardized processes. METHODS: We employed statistical quality control charts (X- s, p, and u) to analyze perioperative morbidity and mortality and length of stay in 1,131 nonemergent, isolated, primary coronary bypass operations conducted within a 17-quarter time period. RESULTS: The incidence of the most common adverse outcomes, including death, myocardial infarction, stroke, and atrial fibrillation, appeared to follow the laws of statistical fluctuation and were in statistical control. Postoperative bleeding, leg-wound infection, and the summation of total and major complications were out of statistical control in the early quarters of the study period but showed progressive improvement, as did postoperative length of stay. CONCLUSIONS: The incidence of morbidity and mortality after primary, isolated, nonemergent coronary bypass operations may be described by standard models of statistical fluctuation. Statistical quality control may be a valuable method to analyze the variability of these adverse postoperative events over time, with the ultimate goal of reducing that variability and producing better outcomes.

Coronary Artery Bypass↗
PubMed1984

An evaluation of Salmonella (Ames) test data in the published literature: application of statistical procedures and analysis of mutagenic potency.

We searched the published literature for Salmonella test data on some 450 chemicals. Only 137 of more than 400 articles containing original data satisfied minimum criteria for a quantitative analysis [1751 experiments, comprising data on 152 chemicals (Table 1)]. Many of these papers did not report basic information about the test protocol (Table 2). We used previously described statistical procedures (Bernstein et al., 1982) to estimate the initial slopes of the dose-response curves and corresponding standard errors. We also applied tests for significance and linear goodness-of-fit. We then used the results of these analyses to examine several issues: (1) Linearity of the low dose region of the dose-response curve. We found that the overwhelming majority of curves were linear, though ability to detect non-linearity of dose-response curves in the standard plate test is only limited. 7% of all experiments to which the goodness-of-fit test was applied were curves of increasing slope, and with a few possible exceptions, these were not obviously associated with any particular mutagens, even those generally considered to produce non-linear effects such as MNNG and EMS (Table 3). (2) Performance of the statistical test for significance. Results of the statistical test for significance of the dose-response were compared with author's opinions as to positivity. In almost all cases (94%) results of the statistical test and authors opinions were the same. In the examples of conflicting opinions, the reasons were: (a) the statistical test places more weight than do most authors on the presence of a linear dose-response; (b) most authors tend to require at least a 2-fold increase over the spontaneous background for 'significance', and (c) when the number of spontaneous revertants is small (e.g., TA1537), authors tend to require a larger increase in induced revertants than when the spontaneous background is large, whereas the statistical procedure makes no such distinction. These factors result in the statistical test tending to identify more experiments as positive than do authors, provided there is a linear dose-response, and authors tending to judge more experiments as positive when the dose-response is not linear. (3) Reproducibility. Among the 1751 experiments there were 122 data-sets (a total of 333 experiments) in which the same chemical was tested by two or more different laboratories under the same protocol. 21 of the 122 data-sets had some disagreement between experiments as to whether results were positive or negative (Table 4).(ABSTRACT TRUNCATED AT 400 WORDS)

Dose-Response Relationship, Drug↗
PubMed1990

Use of statistical decision rules for evaluating laboratory animal carcinogenicity studies.

In the evaluation of long-term rodent carcinogenicity studies, many different tumor sites and types are evaluated, which may increase the likelihood of a statistical false positive. To deal with this issue, a number of statistical decision rules have been proposed that take into account multiple comparisons. This paper discusses the various types of decision rules and evaluates the factors that may lead to different interpretations of experimental results. These concepts are illustrated by examining the statistical decision procedures used by three analysts to evaluate the results of 25 long-term rodent carcinogenicity studies carried out by the National Cancer Institute. Agreement among these decision rules is shown to be greater than originally reported. It is also concluded that while the application of statistical decision rules may be of value in some instances to guard against statistical false positives, the final interpretation of the data should be based on biological as well as statistical considerations. Thus, statistical decision rules should not be employed as a substitute for sound scientific judgment in the overall evaluation of these experiments.

Animals↗
PubMed2003

Misleading statistical calculations in far-advanced glaucomatous visual field loss.

OBJECTIVE: In this study, the capability of statistical analysis indices to characterize static automated visual fields (VFs) accurately in cases of far-advanced glaucoma was assessed. DESIGN: Retrospective observational case series. PARTICIPANTS: Sixteen eyes of 15 patients with end-stage glaucoma and evidence of collapse of VF statistical analysis indices were included in the study. METHODS: End-stage glaucoma was defined as vertical cup-to-disc ratio of 0.9 or more, mean deviation less than -24 dB and with only a central or temporal island remaining in the VF gray scale. Collapse of statistical indices was defined as any of the following: pattern deviation probability plot without a single VF location showing P < 0.5%; corrected pattern standard deviation (CPSD) and pattern standard deviation (PSD) probability less than 5% or within normal limits (WNL); short-term fluctuation (SF) probability WNL; glaucoma hemifield test (GHT) not outside normal limits (ONL); or presence of a low patient reliability comment triggered by 40% or more false-negative (FN) responses. MAIN OUTCOME MEASURES: Visual field statistical indices. RESULTS: Of the 16 VFs showing misleading statistical calculations, 9 of 16 eyes had a normal pattern deviation probability plot. The PSD, SF, and CPSD parameters were normal or barely outside the normal range in 4 of 16, 10 of 16, and 5 of 16 eyes, respectively. The GHT was ONL in 7 of 13 eyes, borderline with generalized reduction of sensitivity (GRS) in three eyes, and only GRS in two additional eyes. Low patient reliability was triggered because of an FN score of 40% or more in 10 of 16 eyes. CONCLUSIONS: Statistical indices are crucial for the interpretation of automated static VFs. However, in end-stage glaucomatous VF loss, both summary statistical indices and reliability indices may not detect abnormality, thus misleading the casual observer.

Data Interpretation, Statistical↗
PubMedDate not supplied

[The use of statistical methods in 2 medical journals with high impact factor].

This paper intends to contribute to the discussion about the use of statistical techniques within the current biomedical and health investigation. We hypothesized that statistical methods necessary to go ahead with most high level research were rather small. Two high impact factor journals about epidemiology and health sciences were selected: New England Journal of Medicine and American Journal of Epidemiology. Both represent what we have called "successful research". The 2386 original and special articles which were published along the 1986-1990 period were carefully examined. Each statistical technique was classified into one of four complexity levels. Each paper was eventually classified on the same grounds. The frequency of each method was used within each level was then accounted. Statistical cited literature was also registered for each article. Finally, a descriptive examination of this information was developed. The results show that techniques widely dominant are those classified as from lower levels. Only the 12% for New England Journal of Medicine, and 17% for American Journal of Epidemiology belong to the most complex type. It was shown that current biomedical research dismisses, to a large extent both, complex statistical techniques and specialized statistical literature. In general, it limits itself to resources no more ahead than elementary multivariate procedures. It was concluded that simpler statistical methods are useful enough not only to understand what is currently published in journals like the ones studied but to produce high level results as well.

Bibliometrics↗
PubMed2003

Statistical design in phase II clinical trials and its application in breast cancer.

Several statistical designs for phase II studies have been proposed, but they are frequently misunderstood or not applied at all. In this review we describe the major characteristics of the available designs. To investigate the extent to which statistical designs were used in some recent phase II studies, and which designs were the most common, we did a survey of 145 trials involving treatment of breast cancer. Studies selected for the survey were published between 1995 and 1999 in one of seven specific oncology journals (all with impact factor consistently higher than 2). 94 of the studies (64.8%) did not have an identifiable statistical design. However, among the 51 studies with statistical design there was a notable heterogeneity in the type of design applied. We put together a list of factors associated with use of statistical design at univariate analysis. These factors included: referral to a previous phase I study, recent trial start date, private sponsorship, single-agent treatment, and multicentre organisation. Single-agent treatment (OR 2.35; 95% CI 1.01-5.51) and multicentre organisation (OR 3.24; 95% CI 1.47-7.15) were independently predictive of the presence of statistical design. Publication in journals with high impact factors and short intervals between the start of the study and publication were also correlated with statistical design.

Analysis of Variance↗
PubMed2002

Statistical phylogeography.

While studies of phylogeography and speciation in the past have largely focused on the documentation or detection of significant patterns of population genetic structure, the emerging field of statistical phylogeography aims to infer the history and processes underlying that structure, and to provide objective, rather than ad hoc explanations. Methods for parameter estimation are now commonly used to make inferences about demographic past. Although these approaches are well developed statistically, they typically pay little attention to geographical history. In contrast, methods that seek to reconstruct phylogeographic history are able to consider many alternative geographical scenarios, but are primarily nonstatistical, making inferences about particular biological processes without explicit reference to stochastically derived expectations. We advocate the merging of these two traditions so that statistical phylogeographic methods can provide an accurate representation of the past, consider a diverse array of processes, and yet yield a statistical estimate of that history. We discuss various conceptual issues associated with statistical phylogeographic inferences, considering especially the stochasticity of population genetic processes and assessing the confidence of phylogeographic conclusions. To this end, we present some empirical examples that utilize a statistical phylogeographic approach, and then by contrasting results from a coalescent-based approach to those from Templeton's nested cladistic analysis (NCA), we illustrate the importance of assessing error. Because NCA does not assess error in its inferences about historical processes or contemporary gene flow, we performed a small-scale study using simulated data to examine how our conclusions might be affected by such unconsidered errors. NCA did not identify the processes used to simulate the data, confusing among deterministic processes and the stochastic sorting of gene lineages. There is as yet insufficient justification of NCA's ability to accurately infer or distinguish among alternative processes. We close with a discussion of some unresolved problems of current statistical phylogeographic methods to propose areas in need of future development.

Animals↗
PubMed2002

Correction of analytical results for recovery: a comparison of the method performance characteristics from recent collaborative trials studies for aflatoxin quantification using conventional and robust statistics.

Results from recently conducted collaborative trials on the determination of aflatoxin B(1) in various matrices have been evaluated to establish whether the use of recovery data would result in a distinct change of the relative between-laboratory standard deviation (RSD(R)) of the corrected data compared with the uncorrected data. In addition, we applied conventional and robust statistics to evaluate whether the impact of the use of recovery data on the estimation of RSD(R) depended on the statistical method applied for data analysis. This investigation was based on means before and after correction for recovery. The method performance characteristics were calculated using results from naturally contaminated test materials, while the results from test materials fortified with the target analytes were used to estimate the recovery. The study revealed that applying conventional and robust statistics in general led to comparable estimates for RSD(R). The comparison about the use of recovery data showed that in most cases, the RSD(R) obtained from the analysis of aflatoxin B(1) decreased after correction of the results for recovery. This tendency was similar when the comparison was done using robust or conventional statistics. However, in three cases, conventional statistics yielded a higher RSD(R) for the corrected data, whereas robust statistics showed the opposite. Looking carefully at the data, the treatment of conventional statistics indicated that the way outliers are detected and removed could result in an under- or overestimation of RSD(R). Applying the law of error propagation revealed that most likely the correlation between the uncorrected data and the recovery rate led to a reduced variability of the data corrected for recovery.

Aflatoxin B1↗
PubMed2000

Removing the effect of statistical uncertainty on dose-volume histograms from Monte Carlo dose calculations.

Dose-volume histograms (DVHs) of the dose distributions calculated by the Monte Carlo method contain statistical uncertainties. The Monte Carlo DVH can be considered as blurred from the noiseless DVH by the statistical uncertainty. The focus of the present work is on the removal of the statistical uncertainty effect on the Monte Carlo DVHs and the reconstruction of the noiseless DVHs. We first study the effect of statistical uncertainty. It is found that the steeper the DVH, the more significant the effect. For typical critical structure DVHs the effect is usually negligible. For the target DVHs the effect could be clinically significant, depending on the value of uncertainty and the slope of the DVH. We then propose an iterative reconstruction algorithm. Using the DVHs and statistical uncertainties from the Monte Carlo simulations, we are able to reconstruct the noiseless DVHs. A hypothetical example and a number of clinical cases have been used to test the proposed algorithm. For each clinical case, two Monte Carlo simulations (denoted A and B) were performed. Simulation A has very large statistical uncertainties (about 10% of dose in the target volume) while simulation B has very small uncertainties (about 1%). DVHs from simulation B were used to approximate the noiseless DVHs. Using the proposed algorithm, the effect of statistical uncertainty can be removed from the DVHs of simulation A. The reconstructed DVHs were in good agreement with the DVHs from simulation B. The proposed approach is expected to be useful in removing the blurring effect on a quickly calculated Monte Carlo DVH when performing the iterative forward treatment planning.

Algorithms↗
PubMed2003

Statistical inversion for medical x-ray tomography with few radiographs: II. Application to dental radiology.

Diagnostic and operational tasks in dental radiology often require three-dimensional information that is difficult or impossible to see in a projection image. A CT-scan provides the dentist with comprehensive three-dimensional data. However, often CT-scan is impractical and, instead, only a few projection radiographs with sparsely distributed projection directions are available. Statistical (Bayesian) inversion is well-suited approach for reconstruction from such incomplete data. In statistical inversion, a priori information is used to compensate for the incomplete information of the data. The inverse problem is recast in the form of statistical inference from the posterior probability distribution that is based on statistical models of the projection data and the a priori information of the tissue. In this paper, a statistical model for three-dimensional imaging of dentomaxillofacial structures is proposed. Optimization and MCMC algorithms are implemented for the computation of posterior statistics. Results are given with in vitro projection data that were taken with a commercial intraoral x-ray sensor. Examples include limited-angle tomography and full-angle tomography with sparse projection data. Reconstructions with traditional tomographic reconstruction methods are given as reference for the assessment of the estimates that are based on the statistical model.

Algorithms↗
PubMed1984

On the statistical significance of nucleic acid similarities.

When evaluating sequence similarities among nucleic acids by the usual methods, statistical significance is often found when the biological significance of the similarity is dubious. We demonstrate that the known statistical properties of nucleic acid sequences strongly affect the statistical distribution of similarity values when calculated by standard procedures. We propose a series of models which account for some of these known statistical properties. The utility of the method is demonstrated in evaluating high relative similarity scores in four specific cases in which there is little biological context by which to judge the similarities. In two of the cases we identify the statistical properties which are responsible for the apparent similarity. In the other two cases the statistical significance of the similarity persists even when the known statistical properties of sequences are modelled. For one of these cases biological significance is likely while the other case remains an enigma.

Base Sequence↗
PubMed1987

Statistical analysis and study design in plastic and reconstructive surgical research.

Appropriate study design and proper statistical analysis are necessary ingredients for improving the quality and reliability of the information in journal articles. General surgery and plastic surgery articles were compared for principal author's academic degree, a Ph.D.'s presence as a coauthor, the study type, the presence of statistical analysis, the analysis' appropriateness, and the types of errors in study design or statistical analysis. Ph. D. authorship was associated with increased percentage of articles using statistical analysis. When compared with general surgery articles, plastic surgery articles performed four times fewer statistical analyses. However, when statistical analyses were performed, there were few differences between these two specialties. Although there were no differences in the types of statistical analysis errors, there were differences in the types of study design errors. The causes of these discrepancies may lie in the nature of plastic surgery; they may be reduced by adherence to Feinstein's principles of study design and result interpretation.

Authorship↗
PubMed2005

Clinical research and statistical methods in the urology literature.

PURPOSE: We provide a systematic assessment of the quality and accuracy of statistical reporting in the urology literature. MATERIALS AND METHODS: All original research publications with adult human subjects in a single issue (August 2004) of 4 leading urology journals were identified for formal review. A standardized evaluation form was developed in consultation with an experienced biostatistician and subsequently tested. Two independent reviewers with at least 1 year of formal training in research design and biostatistics who were blinded to authors and institutions reviewed each article. Discrepancies were settled by consensus and/or adjudication by the biostatistician. RESULTS: Of the 169 articles screened 97 met eligibility criteria for review. Cohort (43 of 97 or 44%) or cross-sectional (28 of 97 or 29%) designs comprised the majority of these studies. Only 10 randomized clinical trials (12.4%) were identified. Statistical tests were identified in 83 studies (93%). Overall 69 of 83 studies (71%) providing statistical comparisons had at least 1 statistical error, including using the wrong test for the data type in 28%, inappropriate use of a parametric test in 22% and failure to account for multiple comparisons in 65%. In studies applying multivariate analysis (29%) over fitting the model with too many variables was the most common statistical flaw (39%). CONCLUSIONS: This formal review suggests that statistical methods are often used inappropriately in the urology literature, thereby, potentially undermining the validity of study results and conclusions. An effort to raise the awareness of appropriate statistical techniques through postgraduate education appears indicated.

Humans↗
PubMed1997

Multiresolution statistical analysis of high-resolution digital mammograms.

A multiresolution statistical method for identifying clinically normal tissue in digitized mammograms is used to construct an algorithm for separating normal regions from potentially abnormal regions; that is, small regions that may contain isolated calcifications. This is the initial phase of the development of a general method for the automatic recognition of normal mammograms. The first step is to decompose the image with a wavelet expansion that yields a sum of independent images, each containing different levels of image detail. When calcifications are present, there is strong empirical evidence that only some of the image components are necessary for the purpose of detecting a deviation from normal. The underlying statistic for each of the selected expansion components can be modeled with a simple parametric probability distribution function. This function serves as an instrument for the development of a statistical test that allows for the recognition of normal tissue regions. The distribution function depends on only one parameter, and this parameter itself has an underlying statistical distribution. The values of this parameter define a summary statistic that can be used to set detection error rates. Once the summary statistic is determined, spatial filters that are matched to resolution are applied independently to each selected expansion image. Regions of the image that correlate with the normal statistical model are discarded and regions in disagreement (suspicious areas) are flagged. These results are combined to produce a detection output image consisting only of suspicious areas. This type of detection output is amenable to further processing that may ultimately lead to a fully automated algorithm for the identification of normal mammograms.

Algorithms↗