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 127 records · Page 7Linked to original sources
PubMed1977

Quantitative nuclear image analysis: differentiation between normal, hyperplastic, and malignant appearing uterine glands in a paraffin section. II. Computer assisted recognition by discriminant analysis.

Quantitative image analysis was applied to data from microscope photometry of nuclei in a paraffin section. The data were essentially the same as described in a previous publication (Baak and Diegenbach, 1977). The results from measurements on the nuclei of morphologically normal, atypical hyperplastic and carcinomatous uterine glands were used in discriminant analysis. With this method it is possible to discriminate between the three groups of nuclei. Depending on the (sub)set of the variables used, 60-70% of all nuclei are correctly classified in one of the three groups. Discrimination of one of the groups against the other two results in up to 81% correct classifications. Therefore, discriminant analysis offers a possibility of assisting diagnosis in an objective way.

Cell Nucleus↗
PubMed2006

Discriminant analysis of DNA polymorphisms in herpes simplex virus type 1 strains involved in primary compared to recurrent infections.

The restriction fragment length polymorphism (RFLP) was analyzed using a set of herpes simplex virus type 1 (HSV-1) strains isolated from oro-facial lesions (oro-facial site collection), which was composed of two subsets: one subset consisted of 57 strains from primary oro-facial lesions, and the other of 47 strains from recurrent oro-facial lesions of patients complicated by factors possibly triggering the recurrence (e.g. malignancy, operation, and treatment of dental caries). RFLP analysis was carried out previously on two other sets of HSV-1 strains: one set was from genital lesions (genital site collection), and the other was from non-genital lesions (non-genital site collection). Discriminant analysis was carried out on the three sets of HSV-1 strains: the criterion variable had two values of primary infection or recurrence, and the predictor variables were 20 RFLPs. The degrees of separation between primary infection and recurrence increased in the order oro-facial site collection, genital site collection, and non-genital site collection. The results of discriminant analysis in this study confirmed that reactivation of HSV-1 infection is influenced by triggering factors and the site of infection, thereby suggesting the capabilities of multivariate analysis (including discriminant analysis) with DNA polymorphisms for molecular epidemiological studies.

Animals↗
PubMed1995

Discriminant analysis of MRI measures as a method to determine the presence of dementia of the Alzheimer type.

Multivariate discriminant analysis of brain volumes obtained from semiautomated magnetic resonance image (MRI) quantification was used in an attempt to identify demented patients very early in the course of the disease. Temporal and posterior frontal brain volumes were quantified from MRIs in a cross-sectional study of 31 male and female patients with dementia of the Alzheimer type (DAT) and 29 age- and sex-matched healthy comparison subjects. Mean scores on the Folstein Mini-Mental State Examination (MMS) were in the mild range for the DAT group (20 +/- 6.6), but patients with moderate and severe dementia were also included (MMS range of entire DAT group = 4-28). Significant mean differences in frontal and temporal lobe brain volumes were found between the DAT group and the age- and sex-matched healthy comparison group, but the sensitivity of any single measure was limited to 87% with a specificity of 83%. Initial multivariate discriminant analysis revealed significant gender differences among the healthy subjects, but not the DAT patients. The large group size allowed for subsequent discriminant analyses to be performed by gender. All healthy subjects and DAT patients were correctly classified by gender-specific discriminant functions. The male discriminant variables included brain volume, age, and temporal lobe measures. Inclusion of age in the male discriminant function accounted for age-related brain atrophy, a finding that may have emerged as a consequence of the broad age range of the male DAT population (50-81 years). The male discriminant function was also successfully applied to an independent group of mildly demented subjects that included patients for whom the diagnosis of dementia was uncertain but verified by follow-up clinical evaluations. Measures of temporal lobe brain matter and temporal lobe cerebrospinal fluid volumes were the significant discriminator variables for the women. Quantitative MRI and multivariate discriminant analysis showed promise in distinguishing the dementing process from healthy aging in a group of 60 subjects. Moreover, while not diagnostic of DAT, the approach appeared to offer additional information about the probability of a diagnosis being later confirmed in patients with very mild dementia for whom the clinical identification of DAT is uncertain.

Adult↗
PubMed2003

Canonical discriminant analysis of cadmium content levels in unpolished rice using a portable near-infrared spectrometer.

Near-infrared analysis (NIRA) was applied to discriminate and determine the cadmium content levels of unpolished rice using canonical discriminant analysis (CDA). The overall correct classification rate was 85.2% for 318 NIR spectra, repeated three times for 106 samples. After a further improvement through decreased misclassification rates, NIRA could be useful as a rapid, nondestructive and convenient analytical method for primary screening and detecting of cadmium-polluted rice.

Cadmium↗
PubMed2005

Discriminant analysis as a tool to identify compounds with potential as transdermal enhancers.

Structure-activity relationships were sought for 73 enhancers of hydrocortisone permeation from propylene glycol across hairless mouse skin. Enhancers had chain lengths (CC) from 0 to 16 carbon atoms, 1 to 8 H-bonding atoms (HB), molecular weight 60 to 450, log P (calculated) -1.7 to 9.7 and log S (calculated) -7.8 to 0.7. These predictive properties were chosen because of their ready availability. Enhancement ratio (ER) was defined as hydrocortisone transferred after 24 h relative to control. Values for the ER ranged from 0.2 to 25.3. Multiple regression analysis failed to predict activity; ER values for the 'good' enhancers (ER > 10) were underestimated. Simple guidelines suggested that high ER was associated with CC > 12 and HB 2-5. This was refined by multivariate analysis to identify significant predictors. Discriminant analysis using CC, HB, and molecular weight correctly assigned 11 of the 12 'good' enhancers (92%). The incorrectly assigned compound was a known, idiosyncratic Br compound. Seventeen of the 61 'poor' enhancers (28%) were incorrectly assigned but four could be considered marginal (ER > 8). The success of this simple approach in identifying potent enhancers suggested its potential in predicting novel enhancer activity.

Administration, Cutaneous↗
PubMed1997

Discriminant analysis of normal and malignant breast tissue based upon INAA investigation of elemental concentration.

Discriminant analysis of six trace element concentrations measured by instrumental neutron activation analysis (INAA) in 26 paired-samples of malignant and histologically normal human breast tissues shows the technique to be a potentially valuable clinical tool for making malignant-normal classification. Nonparametric discriminant analysis is performed for the data obtained. Linear and quadratic discriminant analyses are also carried out for comparison. For this data set a formal analysis shows that the elements which may be useful in distinguishing between malignant and normal tissues are Ca, Rb and Br, providing correct classification for 24 out of 26 normal samples and 22 out of 26 malignant samples.

Breast↗
PubMed1992

Discriminant analysis when all variables are ordered.

Determination of the equation that relates an ordered dependent variable to ordered independent variables is sought. One solution, non-parametric discriminant analysis (NPD), involves obtaining the best monotonic step function by means of a computer search procedure. Although one can use alternative selection criteria in obtaining the equation, the illustrative examples use absolute distance. This paper compares the prediction procedures obtained from NPD with those from linear discriminant analysis, linear regression (with and without transformed variables), and logistic regression. We show that NPD is analogous to regression tree analysis with incorporation of ordered variables and monotonicity. We use various prediction functions to predict the example data, the data using the leave-one-out technique, and a verification set. Consistently, non-parametric discriminant analysis performs as good as or better than the tested alternatives.

Craniocerebral Trauma↗
PubMed2003

Segmentation of magnetic resonance brain images through discriminant analysis.

Segmentation (tissue classification) of medical images obtained from a magnetic resonance (MR) system is a primary step in most applications of medical image post-processing. This paper describes nonparametric discriminant analysis methods to segment multispectral MR images of the brain. Starting from routinely available spin-lattice relaxation time, spin-spin relaxation time, and proton density weighted images (T1w, T2w, PDw), the proposed family of statistical methods is based on: (i) a transform of the images into components that are statistically independent from each other; (ii) a nonparametric estimate of probability density functions of each tissue starting from a training set; (iii) a classic Bayes 0-1 classification rule. Experiments based on a computer built brain phantom (brainweb) and on eight real patient data sets are shown. A comparison with parametric discriminant analysis is also reported. The capability of nonparametric discriminant analysis in improving brain tissue classification of parametric methods is demonstrated. Finally, an assessment of the role of multispectrality in classifying brain tissues is discussed.

Brain↗
PubMed2006

Analysis of fiber blends using horizontal attenuated total reflection Fourier transform infrared and discriminant analysis.

We have investigated the utility of a horizontal attenuated total reflection Fourier transform infrared spectrometer (HATR/FT-IR) for the analysis of fiber and textile blends. The identification of a blended textile can be accomplished by subtracting a reference spectrum of the textile's most abundant component, leading to a difference spectrum that infers the identity of the second constituent of the blended textile. Mathematical post-processing of the spectra employing discriminant analysis provided a useful statistical tool to confirm the fiber blend components.

Discriminant Analysis↗
PubMedDate not supplied

The Leicestershire Perinatal Mortality Study: a case study of multi-group discriminant analysis with complex sampling.

Case-control studies are usually analysed by two-group discriminant analysis or by a related method. However, in case-control studies of perinatal mortality the cases (perinatal deaths) are far from homogeneous, and it is likely that some risk factors are relevant only to certain subgroups of cases. This paper proposes a seven-category classification of perinatal deaths and reports an analysis using multi-group discriminant analysis. The problem is further complicated by non-random sampling of controls.

Analysis of Variance↗
PubMed2002

Digital dermoscopy analysis of atypical pigmented skin lesions: a stepwise logistic discriminant analysis approach.

BACKGROUND: Digital microscopy is a non-invasive diagnostic technique enabling determination of characteristics that cannot be appreciated by direct observation. If correctly applied, this technique can be useful for the diagnosis of pigmented skin lesions. PURPOSE: To evaluate the utility of digital microscopy for analysing atypical benign and malignant pigmented skin lesions exploiting digital numerical filtering and automatic measurements. METHODS: Forty-eight parameters were identified as possible discriminating variables, and were grouped in four categories: geometries, colours, textures, and islands of colour. Statistical analysis was used to identify the variables with the highest discriminating power. RESULTS: The high quality of the digital image made it possible to observe diagnostic signs in pigmented skin lesion images, acquired by the present technique, in great detail. Specially designed filtering enhanced certain diagnostic patterns. Stepwise discriminant analysis selected only 10 variables (the means of these variables were higher in melanomas than in nevi). CONCLUSIONS: The combined use of digital dermoscopy and stepwise logistic discriminant analysis made it possible to single out the best objective variables for distinguishing atypical nevi and early melanoma.

Diagnosis, Computer-Assisted↗
PubMedDate not supplied

Neuropsychological and behavioral assessment of ADHD in seven- to twelve-year-old children: a discriminant analysis.

We hypothesized that children with attention-deficit/hyperactivity disorder (ADHD) would underperform on neuropsychological tests that are sensitive to executive function impairments. We further proposed that a linear discriminant analysis using behavioral and neuropsychological variables as dependent variables would pinpoint the capability of behavioral questionnaires and neuropsychological tests to predict the correct classification of ADHD and control children. We designed a transversal study with 62 seven- to twelve-year-old children with ADHD and 62 Full Scale IQ-matched controls using two behavioral scales and 13 neuropsychological tests. Using analysis of variance and covariance with age and school achievement and multifactor analysis of variance, we found that behavioral variables established robust, statistically significant differences between groups (p < .001). Children with ADHD scored worse than controls on 31 out of 61 neuropsychological variables. Children with ADHD, however, obtained better scores that were statistically significant on WISC-R Comprehension and Picture completion. A linear discriminant analysis with nine behavioral variables correctly classified 100% of the participants in both groups. Another linear discriminant analysis using 10 neuropsychological variables correctly classified 85.48% of the participants in both groups. We propose a core battery of selected tests for assessing children with ADHD. The significance of cross-cultural analyses of different developmental disturbances is emphasized.

Child↗
PubMed2001

Discriminant analysis of minor physical anomalies as dysontogenetic markers of schizophrenia.

AIM: The aim of the study was to evaluate the predictive value of minor physical anomalies in the binomial model schizophrenic patient-mentally healthy subject as an index of developmental origin of schizophrenia. MATERIAL AND METHODS: Sixty-eight schizophrenic patients (38 males and 30 females) and 69 mentally healthy subjects (36 males and 33 females) of Bulgarian ancestry were examined for minor physical anomalies using the Waldrop Physical Anomaly Scale. The predictive value of minor physical anomalies for schizophrenia was determined by discriminant analysis. RESULTS: The two-group discriminant analysis distinguishes well the schizophrenic patients from the controls with 6 predictive variables. The variables with significant contribution to prediction of the patient-control status are fine electric hair (p < 0.001), gap between I and II toes (p < 0.001), epicanthus (p < 0.001), high/steepled palate (p < 0.001), tongue with rough/smooth spots (p < 0.001), III toe > or = II toe (p < 0.001). The model overall correct classification is 78.63%, predicting slightly better the schizophrenics than the controls (79.7% vs. 77.4%). CONCLUSIONS: The constellation of variables entered into the discriminant analysis model confirms the hypothesis that at least in one subpopulation of schizophrenic patients the minor physical anomalies reflect early developmental processes determining occurrence of the disease in later periods of life.

Adult↗
PubMedDate not supplied

Discriminant analysis of WAIS results in different types of dementia and depressed patients.

A Multivariate Analysis of Covariance and Discriminant Analysis were carried out on complete WAIS profiles obtained from three groups of demented patients: Multi-Infarct Dementia patients, Senile Dementia of Alzheimer Type patients, and Alcoholic Dementia patients. A group of middle-aged Depressed patients was also included. WAIS did not differentiate among dementias, but Picture Completion and Block Design subtests proved to be effective in differentiating dementia from depression.

Aged↗
PubMed1990

[Spontaneous bacterial peritonitis in the cirrhotic patient. A study by discriminant analysis].

Infection of peritoneal fluid in cirrhotic patients requires an early diagnosis and treatment. Different parameters have been used for the diagnosis of this complication, however, the statistical study has not used multivariable analysis techniques. We have therefore designed this work performing a discriminating analysis in order to find the best variable combination. Twenty-six cirrhotic patients who were consecutively admitted to our hospital have been studied. Total adn polymorphonuclear leukocytes, glucose, LDH, and lactic acid were determined in peritoneal fluid. Eighteen peritoneal fluids were classified as non infected and the rest as infected. The criteria obtained from the discriminating function which combines total leukocytes, LDH, and glucose has been chosen as optimal. The utility of discriminating analysis is discussed and we conclude with the need to verify in a later study the value of the calculated function.

Ascites↗
PubMed1988

Multivariate discriminant analysis of the relationship between the hypo-osmotic swelling test and the in-vitro fertilizing capacity of human sperm.

Multivariate discriminant analysis was used to evaluate the usefulness of routine semen parameters and the hypo-osmotic swelling test (HOST) as predictors of the in-vitro fertilizing capacity of human sperm as assessed by the zona-free hamster egg penetration assay (HEPA). Eighty-eight semen samples from untreated patients attending an infertility clinic were analysed. Semen samples were classified into the following three groups before statistical analysis: group 1--positive sperm penetration (greater than or equal to 10%, n = 39); group 2--borderline penetration rates for HEPA (greater than 0% but less than 10%, n = 39) and group 3--negative sperm penetration (0%, n = 10). The percentage of sperm with normal morphology and sperm count were found to be significant in discriminating between semen samples exhibiting different in-vitro fertilizing capacity. These two discriminating variables in combination gave an overall correct classification rate of 45.5%. The multivariate discriminant analysis was also performed after excluding the data of group 2 semen samples (n = 39), which exhibited borderline sperm penetration rates. As a result, three discriminating variables including semen volume, sperm count and the percentage of sperm with normal morphology were selected. These three variables in combination could accurately predict whether a semen sample would exhibit positive sperm penetration (group 1) or negative sperm penetration (group 3) with an overall accuracy of 75.5%. The percentage of swollen sperm after hypo-osmotic treatment was not related to the HEPA result, as determined by linear correlation and multiple regression analyses, and did not give additional information about the in-vitro fertilizing capacity of sperm as evaluated by multivariate discriminant analysis.

Female↗
PubMed2005

QSAR modeling of carcinogenic risk using discriminant analysis and topological molecular descriptors.

A discriminant analysis model is presented for carcinogenic risk. The data set is obtained from the two-year rodent study FDA/CDER database and was divided into a training set of 1022 organic compounds and an external validation test set of 50 compounds. The model is designed to use as a decision support tool for a defined decision threshold, and is thus a binary discrimination into "high risk" and "low risk" categories. The carcinogenic risk classification is based on the method for estimating human risk from two-year rodent studies developed at the FDA/CDER/ICSAS. The paradigm chosen for this model allows a straightforward risk analysis based on historic information, as well as the computation of coverage, probability and confidence metrics that can further qualify the computed result. The molecular structures were represented as MDL mol files. The molecular structure information was obtained as topological structure descriptors, including atom-type and group-type E-State and hydrogen E-State indices, molecular connectivity chi indices, topological polarity, and counts of molecular features. The MDL QSAR software computed all these descriptors. Furthermore, the discriminant analyses were all performed with the MDL QSAR software. The reported model is based on fifty-three descriptors, using the nonparametric normal kernel method and the Mahalanobis distance to determine proximity. The model performed very well on the fifty compounds of the test set, yielding the following statistics: 76% correctly classified "high risk" (carcinogenic) and 84% correctly classified as "low risk" (non-carcinogenic).

Animals↗