SEARCH · Search PubMed
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.
Within a multidisciplinary team, it is important to establish a common language for the statistician and the electroencephalographer. Using an example, we propose a brief introduction to correspondence analysis. Within a set of 8 variables, we show how to isolate a subset of variables which are characteristic either for partial epilepsy or generalized epilepsy.
Principal components analysis (PCA) followed by linear discriminant analysis (LDA) of the nuclear magnetic resonance (NMR) spectra from 98 instant spray-dried coffees, obtained from 3 different producers, correctly attributed 99% of the samples to their manufacturer. Blind testing of the PCA model with a further 36 samples of instant coffee resulted in a 100% success rate in identifying the samples from the 3 manufacturers. Coffees from one manufacturer were also assigned into 2 groups using these techniques, and the compound 5-(hydroxymethyl)-2-furaldehyde was identified as the primary marker of differentiation.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Multiple regression is widely employed to study the contribution of components to the toxicologic effect of a mixture. Here, use is made of the fact that data obtained from standard curves of substances and from their mixtures are separable in regression analysis. Thus, under an assumption of additivity of responses, regression coefficients obtained for components in mixtures alone should be the same as for the individual substances. A t-test is developed such that nonsignificant t values support additivity, negative significant values support antagonism, and positive significant values support synergism. The results are applied to data on the mutagenicity of binary mixtures of azaserine, 4-nitroquinoline N-oxide, and 9-aminoacridine in TA 100 in the Ames assay.
Analytical pyrolysis-gas chromatography (Py-GC) has been a standard method for the forensic analysis of automotive paint for a number of decades. Automotive paints are often identified by visual comparison of pyrograms for peak presence and intensities; however, such analyses can be subjective and time consuming. A preliminary investigation based on Py-GC-mass spectrometric analysis of 100 automobile paint samples of five different colors is presented. Designed experiments are employed to select pyrolysis conditions for adequate discrimination. Pattern recognition techniques including principal component analysis and canonical variates analysis are used to visualize clustering of pyrograms to validate comparisons between different automotive paint pyrograms. These methods have the potential to ease the interpretation task for data sets involving a large number of comparisons.
Explore the source record for details and available documents.
The nature of depressive phenomena in primary health care was explored with data obtained from three primary health care clinics situated in the periphery of the city of Calcutta in India. The Self-Reporting Questionnaire (SRQ) and the Screening for Depression Questionnaire (SDQ-9) were used as the first stage and the Clinical Interview Schedule (CIS) and the Hamilton Rating Scale for Depression (Hamilton) as the second stage instruments respectively. Health workers with limited training administered the first stage instruments to consecutive adult clinic attenders. Principal components analysis followed by multiple linear regression analysis and discriminant function analysis were applied to the data. It was concluded that depressive phenomena in primary health care settings were largely undifferentiated in nature.
Prognostic factors were studied in a series of 80 patients with aplastic anaemia. The most important and easily obtainable individual variables, isolated in a previous univariate statistical analysis, were placed in a linear logistic regression model. A prognostic formula was derived containing the following 4 variables, in decreasing order of significance: (1) reticulocytes, (2) interval from onset of symptoms to first visit, (3) mean red cell volume, and (4) platelets. The prognostic formula permits an estimate of the probability of death within 3 months from the first visit. Its usefulness in several clinical situations, especially those related to bone marrow transplantation, is illustrated.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Forty-eight continuous culture enrichments were performed on summer and winter water samples from two contrasting rivers. The cell output from each chemostat was dependent on the temperature and nitrogen concentration of each enrichment. The diversity of the populations from the continuous cultures, as assessed by species diversity analysis, was always greater than populations obtained on agar plates. However, the species isolated exclusively by continuous culture in these experiments were not unique to the chemostat. All of these species had been isolated at some time on plates directly. High nitrogen concentrations were found to decrease diversity. Populations sampled during the winter were influenced primarily by the concentration of nitrogen. Summer isolates were affected mainly by the temperature of the culture medium. The nutritional versatility of the population was not affected by enrichment of the continuous culture.
All-night sleep of 5 healthy male subjects was scored on the basis of EEG (paper records), EMG and EOG into the stages of wakefulness, REM sleep and slow sleep stages 1-4 according to Rechtschaffen and Kales. Spectral analysis for 8 EEG channels was then performed and spectral parameters (total power and coherence for certain frequency bands) extracted. Stepwise linear discriminant analysis was applied to these spectral parameters to see how well the polygraphically defined sleep stages were recognized. While the results of each intrasubject analysis were satisfactory (error rates between 10 and 15%), a more detailed analysis revealed that most errors occur between nearby slow sleep stages or between S1 and REM or wakefulness. This is related to the fact that Rechtschaffen and Kales classify into discrete stages an essentially continuous process and this is, to some extent, arbitrary. This interpretation is supported by examples. Cross-classification of each subject on the basis of the remaining 4 increases the error rates drastically. A simple way of avoiding this effect, by standardizing the individual data, is shown to work nicely in this case.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.