Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Discriminant Analysis”

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 379 records · Page 21Linked to original sources
PubMed1995

Discriminant analysis of factors distinguishing patients with functional dyspepsia from patients with duodenal ulcer. Significance of somatization.

Patients with duodenal ulcer or functional dyspepsia do not differ on dyspeptic symptoms. The aim of the present study was to test the hypothesis that functional dyspepsia and duodenal ulcer are two different diagnostic entities by examining the discriminating power of several anamnestic, biological, and psychosocial variables. Ninety-four patients with duodenal ulcer and 86 patients with functional dyspepsia were included. Anamnestic data, global assessment, Helicobacter pylori status, blood group, Lewisa+ phenotype, and several measures of psychological distress and somatic complaints were registered. Compared to patients with functional dyspepsia, the duodenal ulcer patients were more often infected by Helicobacter pylori and had their stomach discomfort more often relieved by eating. Compared to patients with duodenal ulcer, patients with functional dyspepsia had higher scores of depression, trait anxiety, general psychopathology and different somatic complaints (called somatization). They were also less satisfied with the health care system, their disorder had a greater negative impact on their quality of life, and their global assessment of own health was poorer. Discriminant analysis including age, smoking, Helicobacter pylori status, global assessment, and somatic complaint classified 86.1% of the patients correctly (77.9% of the patients with functional dyspepsia and 93.6% of the patients with duodenal ulcer). It is concluded that duodenal ulcer and functional dyspepsia are two separate diagnostic entities. Patients with duodenal ulcer are older, smoke more often, and almost all are infected with Helicobacter pylori, while patients with functional dyspepsia are characterized by somatization and a negative assessment of their own health.

Adult↗
PubMed1999

Class II MHC quantitative binding motifs derived from a large molecular database with a versatile iterative stepwise discriminant analysis meta-algorithm.

MOTIVATION: The identification of T-cell epitopes can be crucial for vaccine development. An epitope is a peptide segment that binds to both a T-cell receptor and a major histocompatibility complex (MHC) molecule. Predicting which peptide segments bind MHC molecules is the first step in epitope prediction. RESULTS: An iterative stepwise discriminant analysis meta-algorithm explores a large molecular database to derive quantitative motifs for peptide binding. The applications presented here demonstrate the algorithm's versatility by producing four closely related models for HLA-DR1. Two models use an expert initial estimate and two do not; two models use amino acid residues as the only predictors and two use amino acid groupings as additional predictors. Each model correctly classifies >90% of the peptides in the database. AVAILABILITY: Software is available commercially; data are free over the Internet.

Algorithms↗
PubMed1997

Sex determination by discriminant analysis of calcanei measurements.

Eight measurements taken on the right calcaneus (maximum length, load arm width, minimum width, height of calcaneus, body height, breadth of the facies articularis talaris posterior, breadth and height of the facies articularis cuboidea) of a known contemporary Southern Italian skeletal population (40 males and 40 females) were used to determine sex by multivariate discriminant analysis. Three functions revealed a correct sex-determination of 85%. These functions were obtained by the association of the following parameters: maximum length, load arm width and breadth of the facies art. talaris post. (function no. 1); maximum length and breadth of the facies art. talaris post. (function no. 2); maximum length and height of the facies art. cuboidea and breadth of the facies art. cuboidea (function no. 3). These results may aid the forensic anthropologist when no other remains, useful for skeletal sex determination, are available.

Adult↗
PubMed2005

Application of partial least squares discriminant analysis to two-dimensional difference gel studies in expression proteomics.

Two-dimensional difference gel electrophoresis (DIGE) is a tool for measuring changes in protein expression between samples involving pre-electrophoretic labeling ith cyanine dyes. In multi-gel experiments, univariate statistical tests have been used to identify differential expression between sample types by looking for significant changes in spot volume. Multivariate statistical tests, which look for correlated changes between sample types, provide an alternate approach for identifying spots with differential expression. Partial least squares-discriminant analysis (PLS-DA), a multivariate statistical approach, was combined with an iterative threshold process to identify which protein spots had the greatest contribution to the model, and compared to univariate test for three datasets. This included one dataset where no biological difference was expected. The novel multivariate approach, detailed here, represents a method to complement the univariate approach in identification of differentially expressed protein spots. This new approach has the advantages of reduced risk of false-positives and the identification of spots that are significantly altered in terms of correlated expression rather than absolute expression values.

Analysis of Variance↗
PubMed1998

[Linear discrimination analysis in blood pressure monitoring].

The 24-hours blood pressure monitoring (BPM) is a reliable technique largely used for hypertension diagnosis in clinical practice. Presence of the lack of a nocturnal decrease in blood pressure values is one differentiating criterion between essential and secondary hypertension. New useful method of mathematical and statistical processing of BPM data using the linear discriminant analysis (LDA) was described. This statistical analysis was applied for 123 BPM measurements from patients with essential hypertension without and with different kidney diseases including hemodialyzed patients. The control group was normotensive healthy volunteers. The LDA method successfully separate group of patients with essential hypertension from patients with secondary renal hypertension but the error of return ranking to these groups was 28.5%. The recalculation for groups-controls, hypertensive patients and hypertensive hemodialyzed patients leads to decreasing of the error of return ranking to 5.7%. This retrospective study of nonhomogeneous groups could not clearly differentiate groups of hypertensive patients. The new prospective study on accurately definite groups of patients can eliminate these problems.

Blood Pressure Determination↗
PubMed2004

Classification of individual lung cancer cell lines based on DNA methylation markers: use of linear discriminant analysis and artificial neural networks.

The classification of small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) can pose diagnostic problems due to inter-observer variability and other limitations of histopathology. There is an interest in developing classificatory models of lung neoplasms based on the analysis of multivariate molecular data with statistical methods and/or neural networks. DNA methylation levels at 20 loci were measured in 41 SCLC and 46 NSCLC cell lines with the quantitative real-time PCR method MethyLight. The data were analyzed with artificial neural networks (ANN) and linear discriminant analysis (LDA) to classify the cell lines into SCLC or into NSCLC. Models used either data from all 20 loci, or from five significant DNA methylation loci that were selected by a step-wise back-propagation procedure (PTGS2, CALCA, MTHFR, ESR1, and CDKN2A). The data were sorted randomly by cell line into 10 different data sets, each with training and testing subsets composed of 71 and 16 of the cases, respectively. Ten ANN models were trained using the 10 data sets: five using 20 variables, and five using the five variables selected by step-wise back-propagation. The ANN models with 20 input variables correctly classified 100% of the cell lines, while the models with only five variables correctly classified 87 to 100% of cases. For comparison, 10 different LDA models were trained and tested using the same data sets with either the original data or with logarithmically transformed data. Again, half of the models used all 20 variables while the others used only the five significant variables. LDA models provided correct classifications in 62.5% to 87.5% of cases. The classifications provided by all of the different models were compared with kappa statistics, yielding kappa values ranging from 0.25 to 1.0. We conclude that ANN models based on DNA methylation profiles can objectively classify SCLC and NSCLC cells lines with substantial to perfect concordance, while LDA models based on DNA methylation profiles provide poor to substantial concordance. Our work supports the promise of ANN analysis of DNA methylation data as a powerful approach for the development of automated methods for lung cancer classification.

Carcinoma, Non-Small-Cell Lung↗
PubMed1991

Bioequivalence assessment of diltiazem preparations by means of discriminant analysis of data from solid-phase extraction and liquid chromatography.

A solid-phase extraction technique for sample clean-up coupled with a new LC procedure is reported for the assay of diltiazem in plasma. The use of disposable cartridges provides selective extraction and easy automation. A new LC system based on LiChrospher RP 60 Select B columns is described. For routine analysis, the procedure provides a rapid simultaneous clean-up of several samples prior to chromatography and reproducible recoveries over a concentration range of 10-800 ng. The procedure was used to analyse the plasma samples from a bioequivalence study of three commercial diltiazem preparations. The pharmacokinetic parameters in 12 healthy male volunteers were determined and the assessment of bioequivalence was conducted by discriminant analysis.

Adult↗
PubMed1976

An application of stepwise discriminant analysis to the characterization of military heroin dependents, illicit drug users, and psychiatric patients.

This application of stepwise discriminant analysis explores a way to structure skewed-sample interview data, discusses some implications of the results and the conceptual limitations of assumptions about deviancy, and suggests a focus for future studies of the complex phenomena of drug abuse. The methodology differs from previous reports in that it uses a constant probability level, letting the computational procedure determine the number of significant variables. It also provides a method of determining the "chance" level (the amount of correct classification when there are no differences between the groups) so that the observed correct classification values may be usefully interpreted.

Computers↗
PubMed1993

Personal competencies and community participation in small community residential programs: a multiple discriminant analysis.

We examined whether broad factors assessing dimensions of personal competency and community participation could be used to discriminate between people with mental retardation who were living in small group homes and small foster homes. Using a large national sample and multiple discriminant analysis procedures, we found that personal competencies were least important in distinguishing on the basis of setting between people living in the two types of residential placements. Primary differences emerged in factors assessing extent of community participation, family relationships, and recreation/leisure integration. Implications for policy and future research were presented and discussed.

Activities of Daily Living↗
PubMed1994

Use of discriminant analysis to assess disease activity in pulmonary tuberculosis with a panel of specific and nonspecific serum markers.

Several cell activation markers, acute phase reactants, enzymes, and antituberculous antibody serum levels have been proposed as possible markers to monitor disease activity in patients with tuberculosis. They have all shown limited sensitivity or specificity. The authors therefore attempted to generate a canonical variable using discriminant analysis, including sensitive and specific parameters, to be a reliable marker in classifying patients correctly during the course of pulmonary tuberculosis. The following parameters were selected: two soluble cell activation markers (soluble interleukin-2 receptor and sCD8); the levels of immunoglobulin (Ig) G and IgM antibodies against the A60 antigen complex; and the presence of specific antibodies directed to eight different A60 components, revealed by Western blot analysis. The tests were performed on sera from three groups of patients with pulmonary tuberculosis. The first group comprised 25 patients with onset tuberculosis, clinically active (OTCA), evaluated at the time of admission. The second group included 28 patients with chemotherapy-treated tuberculosis, clinically active (CTCA), 2 months after therapy had begun. The third group included 20 patients with tuberculosis, nonclinically active (TNCA), who had had at least 1 year of effective therapy. The authors obtained an 80.9% rate of correct classification for the three groups and a rate of 100% when OTCA and TNCA were compared. The patients with CTCA were scarcely differentiated and tended to be distributed into the two other groups. To improve the separation between patients with CTCA and those with OTCA, a second canonical variable was generated with a 91.7% rate of correct classification, as compared with 71% obtained using the sCD8 as the best single variable. The mean values of the last canonical variable were statistically different (Mann-Whitney test, P = .049) when stratified for acid fast bacilli-positive or negative CTCA patients (microscopic detection). Three patients, followed during the entire course their disease, were, as expected, correctly positioned with respect to the subsequent disease phases.

Adult↗
PubMed2004

Wavelet-based processing of neuronal spike trains prior to discriminant analysis.

Investigations of neural coding in many brain systems have focused on the role of spike rate and timing as two means of encoding information within a spike train. Recently, statistical pattern recognition methods, such as linear discriminant analysis (LDA), have emerged as a standard approach for examining neural codes. These methods work well when data sets are over-determined (i.e., there are more observations than predictor variables). But this is not always the case in many experimental data sets. One way to reduce the number of predictor variables is to preprocess data prior to classification. Here, a wavelet-based method is described for preprocessing spike trains. The method is based on the discriminant pursuit (DP) algorithm of Buckheit and Donoho [Proc. SPIE 2569 (1995) 540-51]. DP extracts a reduced set of features that are well localized in the time and frequency domains and that can be subsequently analyzed with statistical classifiers. DP is illustrated using neuronal spike trains recorded in the motor cortex of an awake, behaving rat [Laubach et al. Nature 405 (2000) 567-71]. In addition, simulated spike trains that differed only in the timing of spikes are used to show that DP outperforms another method for preprocessing spike trains, principal component analysis (PCA) [Richmond and Optican J. Neurophysiol. 57 (1987) 147-61].

Action Potentials↗
PubMed1992

Detection and evolution of rhythmic components in ictal EEG using short segment spectra and discriminant analysis.

An automated method for analysis of ictal EEG is described which aims to reliably detect one or several rhythmic components in short EEG segments (2 sec) and to display their presence, frequency, amplitude, location, and temporal evolution. Spectra were estimated and compared using fast Fourier transform (FFT) and autoregressive modelling (AR). A subsequent linear discriminant analysis decided whether a spectral peak is likely to be caused by rhythmic activity or by the inherent statistical variability. FFT was found to perform better than AR in the detection of rhythmic components, yielding a false-positive rate of 0.825%, a false-negative rate of 2% (signal to noise ratio -4.6 dB), a frequency resolution of 2 Hz, and a temporal resolution of 0.5 sec. Seizure analysis revealed that the ictal scalp EEG of even short seizures can show a complex evolution of rhythmic patterns which are impossible or difficult to recognize by visual inspection or conventional spectral analysis. The following phenomena are demonstrated: superposition of two rhythmic components suggesting two cerebral regions discharging simultaneously and independently with their own pacemakers, sudden and gradual change of frequencies, and gradual development of harmonic frequencies. It is suggested that a more precise correlation between rhythmic generators and seizure symptomatology might allow more predictable pharmacological responses in antiepileptic therapy.

Adult↗
PubMed2005

Automated design of robust discriminant analysis classifier for foot pressure lesions using kinematic data.

In the recent years, the use of motion tracking systems for acquisition of functional biomechanical gait data, has received increasing interest due to the richness and accuracy of the measured kinematic information. However, costs frequently restrict the number of subjects employed, and this makes the dimensionality of the collected data far higher than the available samples. This paper applies discriminant analysis algorithms to the classification of patients with different types of foot lesions, in order to establish an association between foot motion and lesion formation. With primary attention to small sample size situations, we compare different types of Bayesian classifiers and evaluate their performance with various dimensionality reduction techniques for feature extraction, as well as search methods for selection of raw kinematic variables. Finally, we propose a novel integrated method which fine-tunes the classifier parameters and selects the most relevant kinematic variables simultaneously. Performance comparisons are using robust resampling techniques such as Bootstrap 632+ and k-fold cross-validation. Results from experimentations with lesion subjects suffering from pathological plantar hyperkeratosis, show that the proposed method can lead to approximately 96% correct classification rates with less than 10% of the original features.

Adult↗
PubMed1991

[Use of discriminant analysis in the development of a prognostication system for complications of hypertension].

The knowledge of unfavourable factors affecting the course of essential hypertension (EH) cannot solve the problem of identifying persons for differentiated follow-up and treatment of those with the increased risk for EH complications. In order to elaborate the system of prognosing acute complications of EH among rural population a prospective and retrospective methods of observation were used. To develop a system of prognosing a discriminant analysis was used. The individual risk for acute complications in EH is determined by available signs according to the working table by finding the algebraic sum of prognostic coefficients. The result is compared with the scale of threshold value of summarized influence of factors.

Adult↗
PubMed2004

Combining pattern discovery and discriminant analysis to predict gene co-regulation.

MOTIVATION: Several pattern discovery methods have been proposed to detect over-represented motifs in upstream sequences of co-regulated genes, and are for example used to predict cis-acting elements from clusters of co-expressed genes. The clusters to be analyzed are often noisy, containing a mixture of co-regulated and non-co-regulated genes. We propose a method to discriminate co-regulated from non-co-regulated genes on the basis of counts of pattern occurrences in their non-coding sequences. METHODS: String-based pattern discovery is combined with discriminant analysis to classify genes on the basis of putative regulatory motifs. RESULTS: The approach is evaluated by comparing the significance of patterns detected in annotated regulons (positive control), random gene selections (negative control) and high-throughput regulons (noisy data) from the yeast Saccharomyces cerevisiae. The classification is evaluated on the annotated regulons, and the robustness and rejection power is assessed with mixtures of co-regulated and random genes.

Algorithms↗
PubMed1998

The role of social determinants on men's and women's mobility in Italy. A comparison of discriminant analysis and artificial neural networks.

The paper focuses on the role of the spouse's occupation as a resource for mobile individuals, from the perspective that social positions are held by families, rather than by individuals. Three groups are confronted in terms of the role of the key variables and other relevant factors: men whose spouse does not have a paid job (group 1), men and women whose spouse has a paid job (group 2 and 3). The data set is provided by the national survey on social mobility in Italy, carried out in 1985; social achievements of members of the three groups are considered, including social origins and destinations, social position corresponding to respondent's first job, cultural background (educational achievement of respondent's father and mother found), respondent's education and spouse's social position. The techniques used are discriminant analysis and back propagation Neural Networks. Both techniques traced a clear boundary between group 1 and groups 2 and 3, which were discriminated mainly on the basis of the spouse's occupation; Artificial Neural Networks reached better classification results and allowed a deeper insight into the nonlinear effects of the discriminating variables for the three groups.

Artificial Intelligence↗
PubMed1995

Evaluation of serum peptide banding patterns of patients on hemodialysis therapy by discriminant analysis.

The present study was directed towards determining serum protein abnormalities, with emphasis on protein deficits in patients on short- and long-term hemodialysis. Serum protein profiles of healthy controls and chronic and end-stage renal failure patients on hemodialysis were analysed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis. A total of 104 distinct peptide bands in the molecular weight range of 23 kDa to 300 kDa were encountered for both healthy and patient populations. The SPSS/PC+ discriminant analysis employing group mean statistics revealed significant protein deficits in the patient population. As many as 72 (69.23%) of the 104 individual peptide bands had lower mean scores for the patients as compared to the healthy subjects. In all, 21 (29.2%) of these peptide bands belonged to the high molecular weight range of 185 kDa to 300 kDa while 13 (18.1%) were located in the low molecular weight range of 23 kDa to 36 kDa. To show that patients and healthy controls can be categorized into their respective groups on the basis of their serum protein profiles, a 100% correct classification was established by group membership prediction analysis. Furthermore, using minimization of Wilk's lambda as the criterion for variable selection, it was found that the 285 kDa, 265 kDa, 250 kDa, 155 kDa, 125 kDa, 110 kDa, 32 kDa, 27.5 kDa, 26 kDa and 23.5 kDa serum peptide bands were the most important in classifying an individual as healthy or diseased.

Adolescent↗