SEARCH · PubMed Health

Results for “Discriminant Analysis”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources

PubMed1995

Enzymatic markers of gallstone-induced pancreatitis identified by ROC curve analysis, discriminant analysis, logistic regression, likelihood ratios, and information theory.

We investigated the diagnostic utility of frequent serial determinations of aspartate aminotransferase, alanine aminotransferase (ALT), lipase, amylase, and the lipase/amylase (L/A) ratio for distinguishing patients with acute pancreatitis due to biliary obstruction from those with acute pancreatitis due to other pathogenesis. Analyzed were enzyme activities obtained at admission and peak enzyme activities identified retrospectively from serial measurements in 53 patients with acute pancreatitis due to various causes. We evaluated the data with multiple statistical tools. Discriminant analysis and logistic regression revealed the diagnostic significance of ALT at initial and peak values, and the maximum information provided by peak ALT was confirmed by both logistic regression and stratum-specific likelihood ratios. Stratum-specific likelihood ratios showed peak ALT > 150 U/L was highly diagnostic of biliary pancreatitis. The L/A ratio, either at admission or at peak, was the only other significant variable for identifying patients with acute pancreatitis due to biliary obstruction. A multivariate logistic discriminant function including ALT and the L/A ratio significantly discriminated biliary acute pancreatitis from pancreatitis due to other causes. Evaluation of initial and peak enzyme data by information theory revealed that the optimal test depended on disease prevalence. Initial ALT activities were the test of choice for identifying biliary pancreatitis, up to a disease prevalence of approximately 0.75. At disease prevalence > 0.75, the initial L/A ratio provided the greatest amount of diagnostic information.

Acute Disease↗

PubMed1995

A PC program for unbiased and predictive linear and quadratic discriminant analysis.

Discriminant analysis plays an important role in biological and medical research. The most popular methods of discrimination in practical applications are parametric methods like linear and quadratic discriminant analysis. However, there exist modifications of these approaches, namely unbiased and predictive discriminant analysis, which lead to reduced error rates in certain situations. In this paper a menu-driven, user-friendly PC program written in Borland Pascal 7.0 is introduced which performs unbiased and predictive linear and quadratic discriminant analysis.

Bias↗

PubMed1990

Differentiation between extrahepatic and intrahepatic cholestasis by discriminant analysis.

Discriminant analysis was used to differentiate between extrahepatic biliary atresia and intrahepatic cholestasis. Among the ten laboratory variables tested, three (gamma-glutamyl transpeptidase, alkaline phosphatase and total serum bilirubin) were useful in the differential diagnosis. gamma-Glutamyl transpeptidase contributed most to the discrimination (85%). From a population study of 28 babies with extrahepatic biliary atresia and 24 infants with intrahepatic cholestasis, the procedure achieved a diagnostic accuracy and specificity of 92.9% and an efficiency of 92.3%. The jackknife procedure has also confirmed that the mathematical model was robust for discriminant analysis and therefore it may be valid for screening infants with cholestasis for early surgical intervention. Discriminant analysis is a useful adjunct for differentiation between intrahepatic cholestasis and extrahepatic biliary atresia.

Alkaline Phosphatase↗

PubMed1980

Structure-activity study of antiulcerous and antiinflammatory drugs by discriminant analysis.

Discriminant analysis was used in the structure-activity study of antiulcerous benzoguanamines, antiinflammatory phenylacetic acids, and aminouracils. The usual discriminant analysis requires the equality of covariance matrix for the multivariate normal distribution between observation groups. When this condition is not fulfilled for some pairs of groups, a modified procedure, the "admissible" discriminant analysis after Anderson and Bahadur, was applied. In this procedure, the model of equal covariance is not the prerequisite for the analysis. As the primary criterion for selecting the best combination of variables in the discriminant functions, we used the number of misclassified compounds which is minimum. The discriminant variables were selected from the physiochemical parameters used to analyze the variation in hydrophobicity due to structural modifications. The potency scores divided into three groups for each of the three series of compounds were predicted with more than 80% accuracy, when the two-group analysis was performed for the most potent and least potent groups omitting the intermediary group.

Animals↗

PubMed1985

Prognosis and diagnosis of Reye syndrome by discriminant analysis.

Discriminant analysis was used to discriminate between Reye syndrome (RS) patients and non-RS cases based either on conventional blood chemistry data obtained upon admission, or on the activities of hepatic mitochondrial enzymes in biopsy or necropsy tissue. The control group for blood chemistry measurements contained children with upper respiratory tract infections, varicella, etc. who did not develop RS, as well as healthy children. Subjects with no liver disorder (e.g., accidental death, sudden infant death, etc.) or with non-RS liver disorders were used as controls for hepatic enzyme studies. Hepatic damage indicators (aspartate aminotransferase, AST; alanine aminotransferase, ALT; and bilirubin) correctly classified 86-96% of non-RS cases and 61-71% of RS. By contrast, AST and ALT had little prognostic value (63% overall correct). Ammonia effectively classified favorable outcome cases (95% correct) but not unfavorable (14% correct). However, when ammonia was included with stage of coma information 88% of the favorable and 85% of the unfavorable outcome cases were correctly classified. Discriminant analysis of hepatic enzymes (glutamate dehydrogenase and monoamine oxidase activity) for a RS and a non-RS group correctly classified 80% of non-RS and 95% of RS specimens. The function was suitable for the direct evaluation of RS-like mitochondrial enzyme changes in rat liver.

Ammonia↗

PubMed1990

Morphometry of macrophage development in malignant lymphoma. Predictive value of discriminant analysis.

Discriminant analysis of morphometric data on the ultrastructure of developing macrophages has been used to classify 62 individual subjects into one of the 3 groups of origin, namely normal, Hodgkin's disease and non-Hodgkin's lymphoma, each finding being compared with the known diagnosis. The data had been obtained from blood monocytes grown in suspension culture over a period of 6 days, and related to whole cell, nucleus, nucleoli and mitochondria. Over 80% of subjects were correctly classified as between the 3 groups and over 90% as to their normality or otherwise. Although the non-specific nature of changes in defence cells makes it unlikely that morphometric studies of macrophages will find a place in the diagnosis of specific malignancies, the present work indicates it could be useful in assessing host response and hence prognosis and response to treatment. Discriminant analysis of quantitative differences in cell structure could have wide clinico-pathological application.

Adolescent↗

PubMed1992

Hemophilia A carrier detection by restriction fragment length polymorphism analysis and discriminant analysis based on ELISA of factor VIII and vWf.

We performed carrier determination on female subjects from 32 hemophilia A kindreds with a combination of restriction fragment length polymorphism analysis and discriminant analysis of factor VIII antigen and von Willebrand factor antigen analyzed by enzyme-linked immunosorbent assay. Subjects included 25 obligate carriers, 30 at-risk female subjects from 19 kindreds each with two or more male subjects, with hemophilia and 28 at-risk female subjects from 13 kindreds each with a single sporadic case. Deoxyribonucleic acid (DNA) analysis with factor VIII intragenic probes clarified the carrier status of 15 female subjects, and extragenic probes classified an additional 14. Discriminant analysis, which identified 24 of 25 (96%) obligate carriers with carrier probabilities greater than or equal to 0.71 and 37 of 39 (95.0%) normal female subjects with probabilities less than or equal to 0.30, was useful for clarifying the carrier status of the female subjects who were not helped by DNA analysis and those that were classified by extragenic probes alone. Twenty-nine female subjects could not be categorized by restriction fragment length polymorphism analysis because DNA was unavailable from the subject or from key family members, key female subjects were noninformative, or carriership could not be excluded by restriction fragment length polymorphism in kindreds with sporadic cases. We studied 28 of these female subjects by discriminant analysis; 15 had carrier probabilities of greater than or equal to 0.71, 12 less than or equal to 0.30, and 1 = 0.35. All carriers by extragenic probes had carrier probability values greater than or equal to 0.71, whereas all noncarriers had values less than or equal to 0.30. In some families, particularly those in which gonadal mosaicism was a possibility, extensive family studies together with DNA and discriminant analyses were required for clarifying the source of the mutation, and hence the carrier status of the at-risk female subjects. Thus DNA and discriminant analyses complement each other, and when combined with careful pedigree analysis, the power of carrier determination is increased in some families.

DNA↗

PubMed1984

The improvement of the antenatal detection of fetal neural tube defects using discriminant analysis.

Discriminant analysis was applied to the antenatal diagnosis of fetal neural tube defects. Alpha-fetoprotein, gestational age, and seven easily assayed biochemical variables were used to generate the discriminant function, which was derived from a retrospective study on data from 130 pregnancies, including 25 which were associated with fetal anencephaly and 13 with fetal spina bifida. This function was used to generate a diagnosis in a blind study of 40 previously unassayed samples: 20 problem samples from another centre interspersed with 20 normal samples. The same function was used over a period of 2 years to predict prospectively the outcome of 669 pregnancies with respect to neural tube defects. Of these 55 were found to be associated with fetal neural tube defects. Classification is shown to be more reliable than using alpha-fetoprotein and gestational age alone. The results demonstrate a successful application of discriminant analysis to medical diagnosis.

Amniotic Fluid↗

PubMed1978

Discriminant analysis.

Discriminant analysis (DA) is a pattern recognition technique that has been widely applied in medical studies. It allows multivariate observations ("patterns" or points in multidimensional space) to be allocated to previously defined groups (diagnostic categories). The relationships between DA and other multivariate statistical techniques of interest in medical studies will be briefly discussed. The main emphasis is on linear discriminant functions (LDF). The theoretic assumptions underlying DA using LDFs will be presented, and the effect of violations to these assumptions will be reviewed in detail. Alternative methods will be presented when violations cause serious problems. It has been shown that the familiar LDF is fairly robust to departures from the assumptions. The application of the LDF in less than ideal situations therefore often does not cause much harm (if the violations are not too grotesque). Another set of problems reviewed is how to estimate the misallocation probabilities when using discriminant functions. The selection of the "best" subset of variables out of the complete set will be discussed. Practical guide lines are given based on the theoretic studies reviewed. When possible, available computer programs for various problems of DA will be indicated. The review does not aim at covering all medical studies where DA has been applied, since emphasis is on the practical conclusions of the theory of DA.

Discriminant Analysis↗

PubMed1993

The efficacy of predicting dystocia in yearling beef heifers: II. Using discriminant analysis.

Discriminant analysis was used to develop equations for predicting dystocia in 2-yr-old, first-calf beef heifers. Data collected on 893 yearling heifers from 34 farms were used to derive four sets of prediction equations (Data Set 1). There was a set of equations for British x British (BRBR) heifers at prebreeding, BRBR heifers at pregnancy check, Continental x British-Continental x Continental (CNBR-CNCN) heifers at prebreeding, and CNBR-CNCN heifers at pregnancy check. A second set of data collected on 3,728 heifers at prebreeding and 875 heifers at pregnancy check were used to test these equations (Data Set 2). Measurements taken prebreeding and at pregnancy check included heifer age, weight, body condition score, pelvic height, pelvic width, pelvic area, hip height, and shoulder height. A precalving estimate of average calf birth weight (AVGBWT) and actual calf birth weight (BWT) were also recorded. When actual calf BWT was used, the prediction equations developed for BRBR heifers at prebreeding and at pregnancy check correctly classified 84.6 and 88.5% of the heifers in Data Set 1, respectively. The prediction equations developed for CNBR-CNCN heifers correctly classified 88.1% of the heifers at prebreeding and 87.9% at pregnancy check. When AVGBWT was used, the accuracy of these four sets of prediction equations ranged from 77.1 to 83.1% for heifers in Data Set 1. In Data Set 2, the prebreeding equations correctly classified 78.5% of BRBR and 79.0% of CNBR-CNCN heifers when AVGBWT was used. Culling by these prebreeding equations would have reduced difficult birth rate in BRBR and CNBR- CNCN heifers by 8.5 and 15.7%, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Age Factors↗

PubMed1984

Discriminant analysis.

Discriminant analysis (DA) is a family of statistical methods in which the primary purpose is to predict into which of several discrete groups (eg, diagnostic categories) an individual subject should be classed. This might be called the classification or forecasting function. A second use of DA examines how well the derived discrimination equation fits the data from which it is derived. This is called the evaluation function of the analysis. Examples of familiar clinical situations show how DA might be relevant to health care decisions, especially classification decisions.

Decision Making↗

PubMed1986

Evaluation of the secretin-cholecystokinin test for chronic pancreatitis by discriminant analysis.

Discriminant analysis was used to interpret the results of the secretin-cholecystokinin (CCK) test in the diagnosis of chronic pancreatitis. An allocation rule based on the use of two test variables--mean chymotrypsin concentration and peak bicarbonate output--was constructed to distinguish between 63 patients with chronic pancreatitis and 68 patients without organic disease. These latter patients had signs and symptoms similar to those of the patients with chronic pancreatitis and were used as controls. The allocation rule was applied to a larger set of individuals, including 105 patients with various other diseases. The sensitivity of the test was 83%, and the specificity was 89%. With a prevalence of chronic pancreatitis of 27% in this set of individuals, the positive predictive value was 73%, the negative predictive value was 93%, and the accuracy rate 87%. This diagnostic performance of the secretin-CCK test gives the test a meaningful place in the examination of patients suspected of having chronic pancreatitis.

Adult↗

PubMed1978

Discriminant analysis of bronchial asthma by linear discriminant function with parameters of flow-volumes: discriminant analysis of bronchial asthma in young male non-smokers.

With the parameters of a flow-volume and a volume-time curve, the discriminant analysis of bronchial asthma is described. The subjects were classified into three groups (healthy adults, mild asthmatic patients and moderates ones). The difference of the mean vectors of the parameters of the three groups was made clear by the selection methods of the discriminant analysis between any two of the groups both with 6 parameters (%FVC, FEV1.0%, peak flow rate (PF), flow rate at 50% of FVC (V50), flow rate at 25% of FVC (V25), and V50/V25) and with 8 (6 parameters mentioned above and V75, V10). Forced expiratory volume in 1 second percent (FEV1.0%) or V50 was selected at the first step with 6 parameters, and V75 was selected at the first step with 8 parameters. Probabilities of misclassification with 8 parameters were lower than those with 6 ones and the probability of misclassification at the discriminant analysis between healthy adults and mild asthmatic patients with 8 parameters was 15.75% at the final step.

Adult↗

PubMed1984

Detection of haemophilia carriers: multivariate analysis compared with discriminant analysis using up to five factor VIII variates.

Multivariate analysis is potentially superior to the linear discriminant analysis which is commonly used to identify carriers of haemophilia. Our aim was to compare these two statistical methods, and to find which factor VIII variates most effectively partitioned carrier and normal subjects. In this study we assayed one- and two-stage factor VIII coagulant activity, factor VIII related antigen by electroimmunoassay and by fluoroimmunoassay, and ristocetin co-factor in 50 normal females and 50 carriers of haemophilia. From the results we calculated multivariate ellipses which circumscribed the normal and the carrier populations, and we displayed these on the monitor of a microcomputer. These ellipses separated the two populations better than linear discriminants calculated on the same data. Multivariate analysis correctly identified 94% of the carriers whereas discriminant analysis correctly identified only 84%. Discriminant analysis gave poorer results because the statistical assumption of equal variance was breached, whereas the assumption of multivariate normality was upheld. Of the five factor VIII variates, two-stage factor VIII coagulant activity and factor VIII related antigen by electroimmunoassay correctly identified the most subjects. Ristocetin co-factor did not improve the diagnostic ability, either when in lieu of or when added to factor VIII related antigen.

Antigens↗

PubMed1986

Evaluation of carrier detection rates for Duchenne and Becker muscular dystrophies using serum creatine-kinase (CK) and pyruvate-kinase (PK) through discriminant analysis.

Serum pyruvate-kinase (PK) and creatine-kinase (CK) determinations have been carried out in a sample of 100 obligate carriers for the Duchenne muscular dystrophy (DMD) gene, 23 obligate carriers for the Becker muscular dystrophy (BMD) gene, and 50 normal adult control women. Blood samples were collected from all subjects three times on three independent occasions and the means of these three determinations were considered for both PK and CK activities in the statistical analysis. Discriminant analysis has shown that, in the group of carriers for the DMD gene, the estimated misclassification frequencies (M.F.) using either serum CK, PK, or both enzymes were: 26.5% for CK alone, 19.5% for PK alone, and 19% for both enzymes. In the group of carriers for the BMD gene, the estimated proportions of M.F. were: 31.7% for CK alone, 23.8% for PK alone, and 20.4% for both enzymes. It is concluded that, although a proportion of carries still remains undetected, the use of serum PK determinations enhances the capability of detecting carriers of both DMD and BMD mainly when compared with serum CK alone.

Biometry↗

PubMed1980

The use of concomitant serum pyruvate-kinase (PK) and creatine-phosphokinase (CPK) for carrier detection in Duchenne's muscular dystrophy through discriminant analysis.

Serum pyruvate-kinase (PK) and creatine-phosphokinase (CPK) determinations have been carried out in a sample of 30 obligate carriers for the DMD gene and 50 normal adult control females. In all the subjects under study blood samples have been collected 3 times on 3 independent occasions and the means of these 3 determinations were considered for both CPK and PK activities in the statistical analysis. Discriminant analysis has shown that in the group of 30 obligate carriers the estimated proportions of misclassification using either serum CPK, PK or both enzymes were fourn to be: 10/30 (33.3%) for CPK alone, 6/30 (20%) for PK alone and 5/30 (16.6%) for both enzymes. It is concluded that although a small proportion of carriers still remain undetected the concomitant use of PK and CPK determinations enhances the capability of detecting carriers for the Duchenne gene mainly when compared with CPK alone.

Clinical Enzyme Tests↗

PubMed2000

Chromosome identification using hidden Markov models: comparison with neural networks, singular value decomposition, principal components analysis, and Fisher discriminant analysis.

The analysis of G-banded chromosomes remains the most important tool available to the clinical cytogeneticist. The analysis is laborious when performed manually, and the utility of automated chromosome identification algorithms has been limited by the fact that classification accuracy of these methods seldom exceeds about 80% in routine practice. In this study, we use four new approaches to automated chromosome identification--singular value decomposition (SVD), principal components analysis (PCA), Fisher discriminant analysis (FDA), and hidden Markov models (HMM)--to classify three well-known chromosome data sets (Philadelphia, Edinburgh, and Copenhagen), comparing these approaches with the use of neural networks (NN). We show that the HMM is a particularly robust approach to identification that attains classification accuracies of up to 97% for normal chromosomes and retains classification accuracies of up to 95% when chromosome telomeres are truncated or small portions of the chromosome are inverted. This represents a substantial improvement of the classification accuracy for normal chromosomes, and a doubling in classification accuracy for truncated chromosomes and those with inversions, as compared with NN-based methods. HMMs thus appear to be a promising approach for the automated identification of both normal and abnormal G-banded chromosomes.

Chromosome Mapping↗

PubMed2004

A modified algorithm for generalized discriminant analysis.

Generalized discriminant analysis (GDA) is an extension of the classical linear discriminant analysis (LDA) from linear domain to a nonlinear domain via the kernel trick. However, in the previous algorithm of GDA, the solutions may suffer from the degenerate eigenvalue problem (i.e., several eigenvectors with the same eigenvalue), which makes them not optimal in terms of the discriminant ability. In this letter, we propose a modified algorithm for GDA (MGDA) to solve this problem. The MGDA method aims to remove the degeneracy of GDA and find the optimal discriminant solutions, which maximize the between-class scatter in the subspace spanned by the degenerate eigenvectors of GDA. Theoretical analysis and experimental results on the ORL face database show that the MGDA method achieves better performance than the GDA method.

Algorithms↗