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 163 records · Page 9Linked to original sources
PubMed2002

Discriminant analysis formulas of optic nerve head parameters measured by confocal scanning laser tomography.

PURPOSE: To evaluate whether discriminant analysis formulas of optic disc variables measured by confocal laser scanning tomography can detect glaucomatous visual field defects, to compare the diagnostic precision of these formulas to detect glaucomatous visual field defects in different types of chronic open-angle glaucoma, and to assess whether gender or refractive error influence the results obtained by the formulas. METHODS: One hundred and sixty-one patients with perimetrically defined glaucomatous optic nerve damage and 194 healthy subjects were recruited. All patients underwent confocal laser scanning tomography of the optic disc. The data were analyzed with three linear discriminant analysis formulas (sectorial, Bathija, and Mikelberg) obtained in sets of data different from those used in the present study. RESULTS: The areas under the receiver-operating characteristic curves of the three formulas and of the cup shape measure as a single parameter ranged from 0.649 to 0.81 in the entire group, and the results did not change when age-matched eyes were considered (0.618-0.812). In each of the glaucoma subgroups with primary open-angle, pseudoexfoliative, and normal-pressure glaucoma, and additionally in the hyperopic, myopic, female, and male subgroups, the sectorial formula had the highest diagnostic precision and the highest correlation coefficients with the visual field indices, followed by the Bathija and Mikelberg formulas, without major differences between the subgroups. All three formulas were more effective than the cup shape measure as a single parameter. CONCLUSION: In the various chronic open-angle glaucomas, the sectorial and Bathija formulas tended to have higher diagnostic precision than the Mikelberg formula and the cup shape measure. Gender and refractive error do not markedly influence the diagnostic precision of the formulas tested. The scores of the formulas are mild indicators of the amount of glaucomatous visual field loss. All three formulas are superior to the single cup shape measure in the detection of glaucomatous optic nerve damage.

Adult↗
PubMed1996

Application of multivariate linear discriminant analysis to lung sounds in some pulmonary diseases.

In the past 15 yrs, a number of investigators have applied spectral analysis to respiratory sounds recorded from the chest wall or the trachea in order to objectively characterize them and to relate them with different pulmonary diseases. In the present study, we have applied multivariate linear discriminant analysis to the spectral features of respiratory sounds. Lung sounds and the airflow velocity were recorded from 15 normal adults and 37 patients falling into three different disease categories: chronic obstructive lung disease, bronchial asthma and bronchiectasis. All patients had prominent adventitious lung sounds (i.e. either wheezes or crackles). Amplitude spectra of five selected inspiratory and expiratory sound segments of each subject were calculated using the Fast Fourier Transform algorithm. Multi-variate linear discriminant analysis was then applied to the normalized and averaged spectral area values calculated for 10 unequal and arbitrarily selected frequency bands for each patient in the frequency range between 80 Hz and 1 kHz. Inspiratory and expiratory sounds were treated separately. Discriminant functions were computed from the spectral area values and plotted on graphs to classify the subjects into one of the disease categories or as normal (training set). While some separation was achieved among the different disease groups, a clearer separation was evident between normals and patients as a whole on the basis both of inspiratory and expiratory sounds. Inspiratory and expiratory sound frequency bands having the largest weights in classification were determined. Admittedly, the specific results of this study are preliminary or even tentative in view of the inadequacies of sound recording and signal conditioning techniques that were available to us at the time of recording. However, we believe that the investigation serves to illustrate the potential of multivariate discriminant analysis in the diagnostic classification of patients on the basis of their lung sound patterns. We suggest that this technique be considered by investigators involved in lung sound research, because it also allows other patient variables to be combined with the selected parameters of lung sounds.

Adult↗
PubMed1992

Classification of turkeys as low or high volume semen producers using discriminant analysis.

Medium White turkeys in Generation 10 of divergent selection for semen ejaculate volume (SEV) were classified into lines that produce either low or high volumes of semen using a multivariate statistical technique, discriminant analysis. The discriminant function was based on records of hen fertility, embryonic mortality in the early and late incubation periods, and the incidence of pipped eggs. A hit ratio, the proportion of birds correctly categorized as low or high, of .60 was obtained. The discriminant function was subsequently used to classify birds in Generation 12. A misclassification rate of .28 and .48 was obtained for birds in the low and high lines, respectively. It was concluded that the relatively low rates of misclassification indicate a potential method for identifying male and female turkeys as belonging to low or high SEV lines, based on fertility and incubation records.

Animals↗
PubMed1996

Discriminant analysis: a technique for adding value to patient satisfaction surveys.

Quality improvement in healthcare organizations requires effective measurement of patient satisfaction. This paper describes a methodology that identifies dimensions of care most closely associated with overall perceptions of quality. A patient satisfaction survey was mailed to 2,055 discharged patients of 13 home health agencies. Patients were asked to evaluate service dimensions of home health relating to scheduling, nursing care, home health aide services, and discharge arrangements. Overall satisfaction with quality of services was used as the dependent variable in two discriminant analysis equations. Eleven dimensions discriminated between "excellent" and "good" quality, and seven dimensions discriminated between "satisfactory" and "unsatisfactory" quality. Using discriminant analysis, items most closely associated with quality indeces can be identified and used in CQI initiatives.

Data Collection↗
PubMed1994

Discrimination between demented patients and normals based on topographic EEG slow wave activity: comparison between z statistics, discriminant analysis and artificial neural network classifiers.

The topographic distributions of absolute delta and theta powers were used to classify demented patients and normals by means of z statistics, discriminant analysis and artificial neural networks (NN). The data were taken from two psychopharmacological studies in mildly to moderately demented patients (111 and 96 patients for studies I and II, respectively) and from 56 normal healthy controls. All patients were diagnosed according to DSM-III criteria and were free of medication for at least 2 weeks. The NN used was a strictly layered feed-forward network with complete connections. The z-transformed absolute power values in the combined delta and theta frequency range at 17 electrodes, recorded in a 3 min vigilance-controlled EEG with eyes closed, were used as input. After having trained the NN successfully by backpropagating of errors, the generalization test with independent data results in a classification performance of 90% determined by "relative operating characteristic" analysis. The NN out-performed z statistics and discriminant analysis. This high percentage of correct classifications may justify the development of further application of NNs based on topographic EEG data.

Aged↗
PubMed1985

[Prediction of pulmonary hypertension from simple, noninvasive parameters using discriminant analysis].

Based on a large number of males with chronic obstructive bronchitis, the usefulness of discriminant analysis for prediction of pulmonary hypertension has been investigated. Two groups were formed: 227 men with overt pulmonary hypertension; 145 patients without pulmonary hypertension Applying the discriminant analysis, 9 variables were selected step by step; 6 out of them were measurably contributing to the discrimination of both groups. As relevant turned out diameter of the right descending pulmonary artery, FVC% predicted, FEV1% FVC, pa2, systolic blood pressure, and RV% TLC. In a procedure of reclassification, based upon the calculated discriminant functions, 82,6% of the cases were correctly classified (sensitivity 80,6%, specificity 86,2%). A pocket calculator programme following the method by Läuter gave identical results. The application of the discriminant function to a new group of 147 unselected men with chronic obstructive bronchitis corroborated the foregoing results. 79,1% of all patients with overt or latent pulmonary hyper-tension were detected. For overt hypertension the sensitivity was 88,4%, for latent hypertension 62,5%. Using this method, decision-making regarding heart catheterism can be improved. It must be stressed, however, that this discriminant function is not valid for women and for other diagnosis groups in males.

Bronchitis↗
PubMed1994

The use of discriminant analysis in predicting the distribution of bluetongue virus in Queensland, Australia.

The climatic variables that were most useful in classifying the infection status of Queensland cattle herds with bluetongue virus were assessed using stepwise linear discriminant analysis. A discriminant function that included average annual rainfall and average daily maximum temperature was found to correctly classify 82.6% of uninfected herds and 72.4% of infected herds. Overall, the infection status of 74.1% of herds was correctly classified. The spatial distribution of infected herds was found to parallel that of the suspected vector, Culicoides brevitarsis. This evidence supports the role of this arthropod species as a vector of bluetongue viruses in Queensland. The effect of potential changes in temperature and rainfall (the so-called 'global warming' scenario) on the distribution of bluetongue virus infection of cattle herds in Queensland was then investigated. With an increase in both rainfall and temperature, the area of endemic bluetongue virus infection was predicted to extend a further 150 km in and in southern Queensland. The implications of this for sheep-raising in Queensland are discussed.

Animals↗
PubMed1993

On the efficacy of the rank transformation in stepwise logistic and discriminant analysis.

We have evaluated the performance of four stepwise variable selection procedures commonly used in medical and epidemiologic research. The four procedures are discriminant and logistic regression and their rank transformed versions, where the independent variables are replaced by their ranks. We generated, by computer, data for two groups from several distributions with a variety of sample sizes and covariance matrices. The two ranking procedures each increased the chance of correctly selecting those variables related to group membership for data generated from log-normal or contaminated distributions. For normally distributed data the ranking procedure had little effect on variable selection. Rank transformed discriminant analysis and rank transformed logistic regression were equally effective in selecting variables when sample sizes exceeded 100. Rank transformed discriminant analysis was superior for smaller data sets. We discuss the implications of the results of this study for clinical and epidemiologic research.

Case-Control Studies↗
PubMed2006

A novel radial basis function neural network for discriminant analysis.

A novel radial basis function neural network for discriminant analysis is presented in this paper. In contrast to many other researches, this work focuses on the exploitation of the weight structure of radial basis function neural networks using the Bayesian method. It is expected that the performance of a radial basis function neural network with a well-explored weight structure can be improved. As the weight structure of a radial basis function neural network is commonly unknown, the Bayesian method is, therefore, used in this paper to study this a priori structure. Two weight structures are investigated in this study, i.e., a single-Gaussian structure and a two-Gaussian structure. An expectation-maximization learning algorithm is used to estimate the weights. The simulation results showed that the proposed radial basis function neural network with a weight structure of two Gaussians outperformed the other algorithms.

Algorithms↗
PubMed1997

Discrimination of iron deficiency anemia from other anemia by multiple discriminant analysis of routine blood count and red cell distribution width.

A computer aided diagnostic program, MDA-3 (Multiple Discriminant Analysis) was designed for clinical use. MDA-3 employs CBC data to analyze using a technique of two-group linear discriminant analysis (a type of multivariate analysis) of anemia and polycythemia. MDA-3 is given discriminant knowledge obtained from a database of 7 CBC items and red cell distribution width (RDW) from 14 groups. We collected 851 CBC data of hematologic abnormalities, designed MDA-3 for IDA screening and evaluated the program's efficacy. The number of cases discriminated as IDA group in the first and the second rank by MDA-3 is up to 80.7%. If false negative cases whose degree of probability is similar to the IDA group are considered as the IDA suspicion group, the diagnostic rate becomes 84.9%. This result demonstrates that MDA-3 is useful for screening of IDA.

Anemia, Iron-Deficiency↗
PubMed1990

[Computer study of pathologic factors influencing survival in colorectal carcinoma--a multivariant stepwise discriminant analysis].

Five quantitative pathologic parameters and three clinicopathologic parameters in 80 colorectal carcinoma patients with different prognosis were studied with a multivariant stepwise discriminant analysis on computer. The results showed that DNA index (DI), nucleocytoplasmic ratio (NCR), histologic grade and Dukes' stage were significant factors influencing the survival. With these factors, a 4-variable Fisher's discriminant function was established, which could correctly identify 81.25% and 83.75% of the predicted cases (Jacknife Procedure and Reclassification) Sensitivity of the two methods was 87.80%. The results indicate that the multivariant stepwise discriminant analysis on quantitative pathologic parameters combined with clinicopathologic parameters is of value in predicting prognosis of colorectal carcinoma patients.

Adenocarcinoma↗
PubMed1985

[Diagnostic usefulness of stepwise discriminant analysis employing the values of CA125, TPA, IAP, CEA and ferritin in sera measured simultaneously for gynecological malignant neoplasms].

The values for CA125, TPA, IAP, CEA, and ferritin in sera were measured simultaneously in 68 healthy nonpregnant females and 133 patients with various gynecological diseases, and examined by stepwise discriminant analysis. The usefulness and the limits for diagnosis of various gynecological diseases were investigated for each tumor marker. Also, the diagnostic usefulness of stepwise discriminant analysis employing the values for five tumor markers in sera was studied for gynecological malignancies compared with that of measuring serum CA125 alone. Because the mean values for CA125 in sera were increased specifically in the ovarian cancer patient group compared with those of other tumor markers in sera, the measurement of serum CA125 was considered to be more useful in diagnosing ovarian cancer than that of the other tumor markers. The mean values for CA125 in sera, however, were also increased more significantly in the groups of patients with endometriosis and normal pregnancies than in the group of healthy nonpregnant females (p less than 0.005). In the stepwise discriminant analysis employing the values for CA125 and four other tumor markers in sera, the diagnostic usefulness of each tumor marker was demonstrated in the early diagnosis, the differential diagnosis, and the determination of complete remission after several therapies for ovarian cancers.

Adenocarcinoma, Mucinous↗
PubMed1990

Discriminant analysis of cellular fatty acids of Candida species, Torulopsis glabrata, and Cryptococcus neoformans determined by gas-liquid chromatography.

We used discriminant analysis of cellular fatty acid compositions determined by gas-liquid chromatography to differentiate yeastlike fungi (a total of 190 strains; including 37 Candida albicans strains, 21 Candida krusei strains, 13 Candida guilliermondii strains, 37 Candida tropicalis strains, 10 Candida pseudotropicalis strains, 24 Candida parapsilosis strains, 32 Torulopsis glabrata strains, and 16 Cryptococcus neoformans strains). Previous results with a standard strain of C. albicans indicated that reproducible fatty acid chromatograms can be obtained with cells grown in a medium of 2% Sabouraud glucose agar at 35 degrees C for between 48 and 72 h. These conditions were also maintained in cultures of the other organisms that we studied. The cellular fatty acid compositions of the organisms were determined quantitatively by gas-liquid chromatography and analyzed by discriminant analysis. The total correct identification expressed as relative peak percent was 95.8% (89.2% for C. albicans to 100% for C. krusei, C. guilliermondii, C. pseudotropicalis, T. glabrata, and C. neoformans). The total correct identification expressed as the common peak (palmitic acid) ratio was 94.7% (87.5% for C. parapsilosis to 100% for C. pseudotropicalis, T. glabrata, and C. neoformans). Both results suggest that cellular fatty acid compositions can be differentiated by this method.

Candida↗
PubMed1999

[Fisher discriminant analysis for carcinogenic potency of aromatic amines].

OBJECTIVE: To study the relationship between carcinogenicity and their structure and physicochemical parameters of aromatic amines. METHODS: Fisher discriminant analysis for carcinogenicity of aromatic amines in two different batches was conducted with estimation of parameters of physicochemical using central atom and functional groups in combination with structural parameters of polycyclic aromatic hydrocarbons (PAHs). RESULTS: The discriminant efficacy approximated to that of the two in a series of equations (I and III) established by Yuta, but inferior to the optimal equation II. CONCLUSION: The selected physicochemical parameters can be applied to QSAR study on carcinogenicity of most aromatic amines, with better results in Fisher discriminant analysis.

Amines↗
PubMed1979

Genetics of classic von Willebrand's disease. II. Optimal assignment of the heterozygous genotype (diagnosis) by discriminant analysis.

In classic von Willebrand's disease (vWd), assignment of the heterozygous genotype for genetic studies and diagnosis for clinical purposes (which are not exactly the same) are formidable problems. We have pointed out in the first report in this series that almost 50% of the members of two large kindred who transmitted this disease, and were therefore heterozygous, were scored as normal by the usual tests of hemostasis. This report describes how this large proportion can be significantly reduced by application of discriminant analysis. Using linear discriminants in three variables--coagulation factor VIII (VIII:C), factor-VII-related antigen (VIIIR:Ag), and the ristocetin cofactor related to factor VIII (VIIIR:WF)--we were able to classify as heterozygous more than 80% of the transmitters in the two large kindred. It was of particular interest that the four parents of two related vWd homozygotes could be scored as heterozygous by discriminant analysis even though all their laboratory tests were within the normal ranges.

Adult↗
PubMed1983

Discriminant analysis of lung function test results in the selection of patients for bronchial carcinoma surgery.

Discriminant analysis was applied to the results of lung function tests carried out on patients prior to thoracotomy for carcinoma of the bronchus. A group of 64 patients who subsequently suffered post-operative complications were compared with one of 78 who were complication free. The largest differences between the mean discriminants of the groups using from 2 to 9 attributes was with combinations which included the residual volume and maximum voluntary ventilation values in every case. The differences increased little from the 2- to the 9-attribute analyses. Using the group mean discriminate as limits, two sets of 240 patients and one of 116 patients were divided into 3 risk categories. The findings were compared with those using a previously published method based upon empirically determined limits for the lung function test results. Consistent findings indicated that it was necessary to add the unweighted maximal mid-expiratory flow rate and whether the operation was a right pneumonectomy to the 2-attribute discriminant procedure to give one that was comparable to the empirical method. It was concluded that little improvement can be brought about using discriminant analysis on the test results although the new procedure may prove more convenient in practice.

Carcinoma, Bronchogenic↗
PubMed1996

Discriminant analysis of antibiotic resistance patterns in fecal streptococci, a method to differentiate human and animal sources of fecal pollution in natural waters.

Discriminant analysis of patterns of antibiotic resistance in fecal streptococci was used to differentiate between human and animal sources of fecal pollution in natural waters. A total of 1,435 isolates from 17 samples of cattle, poultry, human, and wild-animal wastes were obtained, and their ability to grow in the presence of four concentrations of five antibiotics (chlortetracycline, halofuginone, oxytetracycline, salinomycin, and streptomycin) was measured. When the resulting antibiotic resistance patterns were analyzed, an average of 74% of the known isolates were correctly classified into one of six possible sources (beef, chicken, dairy, human, turkey, or wild). Ninety-two percent of human isolates were correctly classified. When the isolates were pooled into four possible categories (cattle, human, poultry, and wild), the average rate of correct classification (ARCC) increased to 84%. Human versus animal isolates were correctly classified at an average rate of 95%. Human versus wild isolates had an ARCC of 98%, and cattle versus poultry isolates had an ARCC of 92%. When fecal streptococci that were isolated from surface waters receiving fecal pollution from unknown origins were analyzed, 72% of the isolates from one stream and 68% of the isolates from another were classified as cattle isolates. Because the correct classification rates of these fecal streptococci are much higher than would be expected by chance alone, the use of discriminant analysis appears to hold promise as a method to determine the sources of fecal pollution in natural waters.

Animals↗
PubMedDate not supplied

Discrimination between migraine patients and normal subjects based on steady state visual evoked potentials: discriminant analysis and artificial neural network classifiers.

Fifty-one migraine patients and 19 control subjects were examined by steady state visual evoked potentials (SSVEPs) procedure. The aim of this study was to develop a discriminant analysis and an artificial neural network (NN) classifier in order to discriminate between migraneurs during attack-free periods and normal subjects. Discriminant analysis correctly classified 72.5% of migraine patients with a false positive rate of 36.8%. The NN method had a sensitivity of 100% with a false positive rate of 15%. The results of this study confirm SSVEP pattern as a marker of migraine and demonstrate that NNs could be a useful method in the statistical analysis of topographic EEG data.

Adult↗