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 1,045 records · Page 58Linked to original sources
PubMed1990

Improvement of diagnosis in the non-invasive assessment of coronary artery disease: enhanced evaluation of quantitative exercise 201thallium imaging by multivariate analysis.

STUDY OBJECTIVE: The aim was to evaluate discriminant analysis, performed in patients without prior myocardial infarction, in enhancing the diagnostic value of quantitative exercise 201thallium scintigraphy. DESIGN: All clinical, electrocardiographic, and scintigraphic variables were first subjected to a univariate analysis. Afterwards a discriminant analysis was done. PATIENTS: 135 patients (104 male) were studied. Age was 24-70 years, mean 55 years. MEASUREMENTS AND MAIN RESULTS: Two discriminant analyses were done. In the first analysis, the ability to detect the presence of coronary artery disease was tested. Significant variables were: (1) history of angina, (2) sex, (3) quantitative analysis of 201thallium scintigraphy, (4) age, (5) ischaemic ST response, (6) angina during the test, and (7) the pressure-rate product. The sensitivity, specificity, and accuracy of classification using the discriminant function were 91%, 87%, and 90%, respectively. The sensitivity was higher than when using only visual interpretation (sensitivity 70%; p less than 0.0002) or quantitative interpretation (sensitivity = 66%; p less than 0.0001) of thallium scans, without significant loss of specificity (p less than 0.5488; p less than 0.6875). In the second analysis, a discriminant function was calculated to detect multivessel disease. Five input variables were selected: (1) number of vessels with stenosis predicted by quantitative analysis, (2) number of vessels with stenosis predicted by visual analysis, (3) ischaemic ST response, (4) sex, (5) angina during the test. Multivariate analysis showed an increase in sensitivity when compared with visual interpretation (78% v 55%; p less than 0.0043) and quantitative analysis (66%; p less than 0.0156). Using the classification, the discriminant function was more accurate than visual analysis (81% v 69%) or quantitative analysis (77%). CONCLUSIONS: The results show that multivariate analysis of non-invasive test results in quantitative thallium exercise testing allows convenient use for clinical purposes with improved results.

Adult↗
PubMed2001

Pattern recognition techniques for the evaluation of spatial and temporal variations in water quality. A case study: Suquía River Basin (Cordoba-Argentina).

We report a comparative study using three different chemometric techniques to evaluate both spatial and temporal changes in Suquía River water quality, with a special emphasis on the improvement obtained using discriminant analysis for such evaluation. We have monitored 22 parameters at different stations from the upper, middle, and beginning of the lower river basin during at least two years including 232 different samples. We obtained a complex data matrix, which was treated using the pattern recognition techniques of cluster analysis (CA), factor analysis/principal components (FA/PCA). and discriminant analysis (DA). CA renders good results as a first exploratory method to evaluate both spatial and temporal differences, however it fails to show details of these differences. FA/PCA needs 13 parameters to point out 71% of both temporal and spatial changes, consequently data reduction from FA/PCA in this case is not as considerable as expected. However, FA/PCA allows to group the selected parameters according to common features as well as to evaluate the incidence of each group on the overall change in water quality, specially during the analysis of temporal changes. DA technique shows the best results for data reduction and pattern recognition during both temporal and spatial analysis. DA renders an important data reduction using 6 parameters to afford 87% right assignations during temporal analysis. Besides, it uses only 5 parameters to yield 75% right assignations during the spatial analysis of four different basin areas. DA allowed us to greatly reduce the dimensionality of the starting data matrix, pointing out to a few parameters that indicate the biggest changes in water quality as well as variation patterns associated with seasonal variations, urban run-off, and pollution sources, presenting a novel approach for water quality assessments.

Argentina↗
PubMed2002

Classification of disease subgroup and correlation with disease severity using magnetic resonance imaging whole-brain histograms: application to magnetization transfer ratios and multiple sclerosis.

This paper presents a new approach to characterize subtle diffuse changes in multiple sclerosis (MS) using histograms derived from magnetization transfer ratio (MTR) images. Two major parts dominate our histogram analysis; 1) Classification of MTR histograms into control and MS subgroups; 2) Correlation with current disability, as measured by the EDSS scale (a measure of disease severity). Two data reduction schemes are used to reduce the complexity of the analysis: linear discriminant analysis (LDA) and principal component analysis (PCA). LDA is better for the classification of MTR histograms as it takes into account the between-class variation. By using LDA, the space of MTR histograms is transformed to the optimal discriminant space for a nearest mean classifier. In contrast, PCA is useful for correlation with current disability as it takes into account the variation within each subgroup in its process. A multiple regression analysis is used to evaluate the multiple correlation of those principal components with the degree of disability in MS. This is the first application of such classification and correlation techniques to magnetic resonance imaging histogram data. Our MTR histogram analysis approach give improved classification success and improved correlation compared with methods that use traditional histogram features such as peak height and peak location.

Adult↗
PubMed1998

Event-related magnetic fields in the auditory cortex of man during unilateral movements: a discriminant function analysis.

It is often assumed that sensorimotor coordination is a feature of the sensorimotor areas of the neocortex only. The purpose of the present study was to examine how this phenomenon is reflected in the auditory cortex of man. Ten subjects were engaged in a stimulus-reaction paradigm, in which each of two acoustical tones was associated to either of two motor reactions. Magnetic fields recorded with a 122-channel magnetometer were modelled by current dipoles. The spatial coordinates as well as the amplitudes of the dipoles were analyzed from 90 to 110 ms after stimulus onset using discriminant analysis. The results suggest that the dipole trajectory in the auditory cortex of the right hemisphere and amplitudes of the dipoles in the auditory cortex of the left hemisphere already 90-110 ms after the beginning of the stimulus could be affected not only by physical features of the stimulus, but also by the motor task required as a reaction.

Acoustic Stimulation↗
PubMed1990

Discriminant factor analysis of 31P NMR spectroscopic data in myopathies.

Discriminant factor analysis (DFA) enables one to distinguish among diagnostic groups using diagnostic variables. It provides discriminant functions that are linear combinations of the diagnostic variables and that optimally separate diagnostic groups. It was used to enhance the accuracy of 31P NMR data in the diagnosis of myopathies. DFA allowed a good separation of normal subjects, congenital neuromuscular disorders with type I fiber predominance, and McArdle's diseases. It elicited an unexpected homogeneity of idiopathic rhabdomyolyses, the metabolic origin of which is unknown, and suggested that the abnormality could involve the mitochondrial oxidative metabolism in some of them. In mitochondrial myopathies, an expected heterogeneity is shown by DFA. It may allow an automatic diagnosis of some myopathies from 31P NMR data or guide biochemists by comparing biochemical features of a new patient to those of previously established groups.

Adult↗
PubMed1984

Mandible analysis of NOD and NON strains of mice.

10 inbred strains of mice including NOD and NON were identified by discriminant and canonical discriminant analysis of the mandible measurements. The number of erroneous discriminations was 0.5% (1/199) for the females and 0% for males (0/232). In canonical discriminant analysis (discriminant analysis with reduction of dimensionality), NOD and NON inbred strains were separately located on 2-dimensional planes, Z1-Z2, Z1-Z3, and Z2-Z3. These results clearly indicate that the genetic constitution of NOD and NON differs, although they have been established from a common ancestor ICR mouse. Causes of divergence between NOD and NON mice are discussed.

Analysis of Variance↗
PubMed1993

Prediction of behind-the-wheel driving performance in patients with cerebral brain damage: a discriminant function analysis.

Rehabilitation specialists, particularly occupational therapists, are often involved in evaluating clients' fitness to resume driving after cerebral damage due to head injury or stroke. Their available methods include separate or combined administration of predriver, simulator, and behind-the-wheel evaluations. However, use of these methods without a theoretical model on which to base test selection has yielded some criticism and little research about the effectiveness of these methods in predicting driving performance. Using the authors' Cybernetic Model of Driving as the basis for assessing abilities and behaviors relevant to driving outcome, this study sought to determine the effectiveness of the evaluation methods by discriminant analysis and measurements of sensitivity in predicting behind-the-wheel failures. Comprehensive evaluations of 106 patients revealed that residual deficits in cognition per se did not render a person unfit to drive and underscored the importance of considering behaviors in determining fitness. The methods of evaluation were shown to be relatively sensitive in predicting outcome; off-road and on-road evaluation reached sensitivities of 90% and 92% with the inclusion of behavioral measures. Formulae for predicting outcome based on methods of evaluation are provided and reasons for failures in behind-the-wheel evaluations are discussed.

Adolescent↗
PubMed1993

[Sex determination with a discriminant function analysis of deciduous teeth size in plaster models].

Plaster models of the teeth of 3-year-old Japanese children (96 males, 98 females) were used to record the crown length, crown width and crown thickness of 5 maxillary and 5 mandibular deciduous teeth (30 measurement values). These measurements were used to devise a number of sex determination formulae. A sex-determination formula using all 30 values was calculated. Furthermore, a number of practical formulae were derived from only the crown width and crown thickness values because the deciduous teeth wear in 4 years and older children rapidly progress, making the crown length measurement unreliable. These formulae were calculated for the maxillary teeth alone and mandibular teeth alone. The formulae based on only the crown width or thickness were also calculated for both maxillary and the mandibular teeth. A step-wise discriminant analysis was then used to ascertain the most reliable measurements and a practical formula subsequently devised. The results obtained were as follows: 1. The mean value for each measurement was greater in males than in females. 2. Significant differences in the values recorded were seen in 28 out of the 30 measurements taken. The measurement items not exhibiting these significant differences were the crown width of the maxillary lateral incisor and the crown thickness of the mandibular second molar. 3. The accuracy rates for the sex-determination analysis and the step-wise sex determination analysis calculated using all 30 values were 78.6% and 75.7%, respectively. 4. The accuracy ranges for the modified sex-determination formulae and the associated step-wise sex determination analyses were 70.6-78.4% and 67.0-76.8%, respectively.

Child, Preschool↗
PubMed1993

[Application of fuzzy inference to personal identification--sex determination from teeth].

Fuzzy inference has received much attention recently as a new type of computer control technology for application in various fields. In this study I applied fuzzy inference to personal identification in medical jurisprudence. Sex determination program was constructed from teeth using the fuzzy inference technique and was compared with known discriminant analysis. The materials examined were mandibular and maxillary dental plaster models of 100 adult males and females, and 80 infant males and females. For the permanent teeth, measurements were made on crown width and thickness of the mandibular and maxillary central incisors, canines, first premolars, and first molars, and of the width and length of the maxillary dental arch. For the deciduous teeth, measurements were made on crown width and thickness of all teeth. Stepwise discriminant analysis was conducted using these values. The values from 5 selected upper-ranked items were set as input objects of the fuzzy inference program and the probability of maleness obtained from the discriminant analysis was set as the output object. Finally, sex determination program for both permanent and deciduous teeth was constructed using a fuzzy inference software development tool. Each measured value was input into this program and the output results were compared with those of discriminant analysis. The percentages of correct determinations for permanent teeth were 83.0% for males and 86.0% for females using stepwise discriminant analysis, however it increased to 93.0% for males and 89.0% for females using the fuzzy inference program. The percentages of correct determinations for deciduous teeth models were 67.5% for males and 75.0% for females using stepwise discriminant analysis, and increased to 86.3% for males and 81.3% for females using the fuzzy inference program. Among samples with probabilities between 40% and 60%, 3 out of 14 males and 4 out of 13 females were misjudged using stepwise discriminant analysis in cases of permanent teeth. However using fuzzy inference program, it reduced to 0 for males and 3 for females. In case of deciduous teeth, 9 out of 20 males and 9 out of 23 females were misjudged using stepwise discriminant analysis. But it reduced to 5 for males and 4 for females using the fuzzy inference program.

Adult↗
PubMed1990

Dental health, prophylactic antibiotic measures and infective endocarditis: an analysis of the knowledge of susceptible patients.

A questionnaire was administered to 338 cardiac patients susceptible to infective endocarditis seen at the outpatient clinic or a cardiac ward. The questionnaire consisted of a series of questions on the educational, social, familial and professional background, the medical and dental history and the knowledge of the patient on prophylaxis of infective endocarditis. Univariate analysis (chi-square test) and stepwise discriminant analysis were used to evaluate the factors responsible for the presence or absence of an acceptable degree of awareness of infective endocarditis. More than half of the patients had no knowledge on prophylactic measures. The time since the diagnosis of heart disease, an edentulous state, written instructions received in the past on prophylaxis, the location of the follow-up, the level of education and previous prosthetic valve surgery were factors retained by discriminant analysis. A discriminant score was calculated for each patient and permitted classification in subgroups with a high, intermediate and low probability of presence of knowledge. This was simplified by classifying patients according to the number of risk factors. An answer to one of the six variables, selected by discriminant analysis, unfavorable for the presence of knowledge, was considered as a risk factor. Risk factor analysis must be considered as an important tool in the education of patients on the risk and prophylaxis of infective endocarditis.

Adolescent↗
PubMed1977

Diagnostic effectiveness of biochemical liver-function tests, as evaluated by discriminant function analysis.

I evaluated the diagnostic value of routinely ordered liver-function tests in 175 biopsy-proven cases of hepatic disease by use of stepwise discriminant analysis. The tests studied-total and "direct" bilirubin, alkaline phosphatase, lactate dehydrogenase, and aspartate aminotransferase-correctly classified 45-73% of cases, depending on the homogeneity of the diagnostic groups. Aspartate aminotransferase and alkaline phosphatase were the best discriminators. When all tests were used in the most homogeneous groups (tumors, cirrhosis, and hepatitis), there was a stepwise improvement in diagnostic accuracy from 51 to 73%.

Alkaline Phosphatase↗
PubMed1991

Automatic classification of cells in cell cycle phases based on Ki-67 antigen quantification by fluorescence microscopy.

Ki-67 antigen is thought to be a marker of cell proliferation, as it can be detected in cycling cells, i.e. cells in G1, S, G2 and M phases, but not in resting cells. The immunocytochemical staining pattern obtained by the Ki-67 monoclonal antibody varies, depending on the cell cycle phases. Analysis of double staining of Ki-67 antigen and DNA in the MCF-7 cell line by videomicrofluorometry allows the description of both the level and the pattern of Ki-67 staining in the form of a set of parameters defining each cell. These parameters were measured in MCF-7 cell populations characterized according to their position in the cell cycle. They were submitted to a statistical analysis (principal component and discriminant analysis) which allowed the determination of the optimal parameters to characterize a given cellular group and permitted the use of these parameters for an automatic classification of cells in the different cell cycle phases. In G1, S, G2, prophase + metaphase and anaphase + telophase cells, these parameters allowed a classification of cells with a good-classification rate of 94.37%. A comparison of this method with methods based on the DNA histogram and bromodeoxyuridine uptake was performed. The classification coefficients stemming from the discriminant analysis were introduced into a program to obtain, automatically, the Ki-67 labelling index and the percentages of cells in each phase. This method, which allows a quick evaluation of the proliferation and the phase indices, may be more widely applicable.

Breast Neoplasms↗
PubMed1992

Introduction of a neuronal network as a tool for diagnostic analysis and classification based on experimental pathologic data.

A neuronal network, as well as uni- and multivariate statistics and a discriminant analysis were applied to a morphometric database of 58 cases with thyroid neoplasms and normal thyroid tissue. The ability to classify cases correctly according to their diagnosis was compared between the neuronal network and discriminant analysis. For all pairwise comparisons, classification by neuronal network was as least as good as classification by discriminant analysis. For some comparisons, the neuronal network provided more correct diagnoses than discriminant analysis. On the contrary, in a comparison between tumors which are not significantly different according to multivariate statistics, the network reclassifies only half of the cases correctly, whereas discriminant analysis falsely suggests the possibility of classifying cases with either diagnosis. Our results confirm a higher sensitivity of the neuronal network to the diagnostic information contained in the present morphometric database, and we will therefore use this concept for analysis and diagnostic classification in further morphometric studies.

Diagnosis, Computer-Assisted↗
PubMed1997

Classification of collected trot, passage and piaffe based on temporal variables.

The objective was to determine whether collected trot, passage and piaffe could be distinguished as separate gaits on the basis of temporal variables. Sagittal plane, 60 Hz videotapes of 10 finalists in the dressage competitions at the 1992 Olympic Games were analysed to measure the temporal variables in absolute terms and as percentages of stride duration. Classification was based on analysis of variance, a graphical method and discriminant analysis. Stride duration was sufficient to distinguish collected trot from passage and piaffe in all horses. The analysis of variance showed that the mean values of most variables differed significantly between passage and piaffe. When hindlimb stance percentage was plotted against diagonal advanced placement percentage, some overlap was found between all 3 movements indicating that individual horses could not be classified reliably in this manner. Using hindlimb stance percentage and diagonal advanced placement percentage as input in a discriminant analysis, 80% of the cases were classified correctly, but at least one horse was misclassified in each movement. When the absolute, rather than percentage, values of the 2 variables were used as input in the discriminant analysis, 90% of the cases were correctly classified and the only misclassifications were between passage and piaffe. However, the 2 horses in which piaffe was misclassified as passage were the gold and silver medallists. In general, higher placed horses tended toward longer diagonal advanced placements, especially in collected trot and passage, and shorter hindlimb stance percentages in passage and piaffe.

Analysis of Variance↗
PubMedDate not supplied

Selection criteria for endoscopic retrograde cholangiopancreaticography (ERCP) in patients with gallstone disease.

Endoscopic retrograde cholangiopancreatography (ERCP) has been used in patients referred to cholecystectomy when clinical information, biochemical values, or ultrasonography (clinical characterization) have indicated possible presence of common bile duct stones. A retrospective study of 599 patients treated for gallstone disease was used to develop a characterization procedure for predicting common bile duct stones by a discriminant analysis procedure. The variables selected by the analysis as the best combination for CBDS prediction were age (years), the values of bilirubin (micromol/l), ALAT (U/l) and gamma GT (U/l). The characterization was false positive in 22 cases (3.7%) and false negative in 11 cases (1.8%), compared to 198 false positive cases (33.1%) and three false negative cases (0.5%) by the clinical characterization. A leaving-one-out correction did not change the results. In a test set of 157 cholecystectomy patients, clinical characterization was false positive in 44.6% of the patients, compared to 4.5% false positive results when using the discriminant analysis procedure. The discriminant analysis procedure would have missed one patient with common bile duct stones. Selection by the discriminant analysis characterization procedure seems to reduce the frequency of preoperative ERCP significantly without an increase in undetected common bile duct stones.

Adolescent↗
PubMed1995

Surgical strategy for patients with gastric carcinoma with submucosal invasion. A multivariate analysis.

BACKGROUND: Early gastric cancer can be treated by endoscopic excision or simple wedge surgical resection. Standard gastrectomy often is advised if submucosal invasion is found, even though only 15-25% of these patients have lymph node metastases. In this study, the risk of lymph node involvement was examined by multivariate analysis to develop a simple discriminant function for surgical decision making in this setting. METHODS: The authors determined factors significantly correlated with lymph node involvement in a retrospective review of 196 patients with gastric adenocarcinoma invading into, but not beyond, the submucosa. Depth and horizontal spread of cancer in the submucosa were evaluated in addition to ordinary pathologic factors. Discriminant analysis for lymph node involvement was performed using explanatory variables chosen from the results of the univariate analyses. RESULTS: Lymph node involvement correlated significantly with larger tumor size; greater dimension of submucosal invasion; deeper submucosal invasion; gross appearance of Type I, IIc + III or IIa + IIc; and severity of vessel invasion. Of the variables, the amount of lymphatic invasion, macroscopic appearance, and maximum dimension of submucosal infiltration were selected as effective predictors of lymph node involvement according to discriminant analysis. A correct discrimination of 74.8% was obtained with a linear discriminant function using these variables. Lymph node involvement was observed in 50% of the cases with a discriminant score less than -1 and in 25% of those with a score between -1 and 0, whereas no lymph node involvement was observed in those with a score greater than 2. CONCLUSIONS: Discriminant function as used in this study provided a useful criterion for additional surgery for patients with gastric carcinoma invading the submucosa who were treated initially by localized excision. Prophylactic lymph node dissection may not be necessary for a discriminant score greater than 2, whereas extended lymphadenectomy would be recommended for a score less than -1.

Adenocarcinoma↗
PubMed2005

Pattern recognition for the analysis of polymeric materials.

A new method of polymer classification is described involving dynamic mechanical analysis of polymer properties as temperature is changed. The method is based on the chemometric analysis of the damping factor (tan delta) as a function of temperature. In this study four polymer groups, namely, polypropylene, low density polyethylene, polystyrene and acrylonitrile-butadiene-styrene, each characterised by different grades, were studied. The aim is to distinguish polymer groups from each other. The polymers were studied over a temperature range of -50 degrees C until the minimum stiffness was reached, tan delta values were recorded approximately every 1.5 degrees . Principal components analysis was performed to visualise groupings and also for feature reduction prior to classification and clustering. Several clustering and classification methods were compared including k-means clustering, hierarchical cluster analysis, linear discriminant analysis, k-nearest neighbours, and class distances using both Euclidean and Mahalanobis measures. It is demonstrated that thermal analysis together with chemometrics provides excellent discrimination, representing a new approach for characterisation of polymers.

Materials Testing↗
PubMed1983

EEG background activity in childhood epilepsy: valid parameterisation by the partial autocorrelation function.

The relationship between EEG background activity and the clinical course is studied in a group of 42 patients with a primary myoclonic-astatic epilepsy. According to the occurrence of seizures the probands are divided into two groups. One group is comprised of patients with an unfavourable course of the disease and the other group is comprised of probands with a favourable one. It is shown by a combined approach of variance analysis, linear discriminant analysis and factor analysis that the EEG background activity parameterized by the partial autocorrelation function is significantly different in the two clinically defined groups. The error rate for assigning a proband, only from the proband's parameterized EEG, to the correct clinical class is estimated to be 0.24. A factor analysis of EEG parameters yields a discriminant factor interpreted as rhythm-power-factor. The group with an unfavourable course is characterized by rhythms in the theta-band and a greater total power than in the favourable group which shows rhythms in the alpha-band. Hypersynchronous potentials in the EEG of this study are not related to the clinical course of the epilepsy. This study confirms that the EEG background activity contains important information about the convulsibility in primary myoclonic-astatic epilepsy.

Adolescent↗