PubMed Health⌕ Search

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 1,243 records · Page 69Linked to original sources

Identification by quantitative chromatin pattern analysis of patients at risk for recurrence of superficial transitional bladder carcinoma.

PURPOSE: Based on the actual clinical outcomes of 132 fully documented patients with superficial transitional cell carcinoma of the bladder, we characterize the risk of recurrence and/or progression by computer assisted image microscopy applied to Feulgen stained nuclei. MATERIALS AND METHODS: Each tumor was characterized by the conventional grading and staging systems as well as by cytometry generated variables describing nuclear DNA content, nuclear morphometry and chromatin patterns. These data were submitted to discriminant analysis to establish a model distinguishing between 2 groups of patients. Group 1 included cases with remission for more than 60 months and group 2 cases presented with recurrence with or without progression within 12 months of transurethral bladder resection. This latter model was then validated by Kaplan-Meyer analysis of the full data set. RESULTS: As evidenced by Kaplan-Meier analysis, the discriminant factor generated by discriminant analysis of cytometry generated variables provided a cutoff value for distinguishing between low and high risks of recurrence (p <0.00001). In contrast, conventional grading and staging systems were not able to make such efficient distinction. CONCLUSIONS: These 2 groups can be used as references with which new cases can be compared to prognosticate disease behavior independently of histopathological grading and/or clinical staging.

Adult↗

Comparison of triple-risk assessment of fetal trisomy 21 including total human choriogonadotropin (hCG) or its free beta-subunit (free beta hCG).

OBJECTIVE: Second trimester total hCG and free betahCG levels in maternal serum samples of 33 pregnancies affected by fetal trisomy 21 and of 188 matched controls were compared in a retrospective study. To find out differences of discriminating efficacy by using one of these markers a multivariate discriminant analysis was performed. METHOD: Statistical evaluation was performed for hCG/free betahCG frequency distributions. Discriminant analysis was carried out using the status 'affected' or 'unaffected' as the group variable and the serum markers unconjugated estriol (uE3), alpha-fetoprotein (AFP), and alternatively, hCG or free betahCG, as discriminant variables. RESULTS: The median of free betahCG MoM values in affected pregnancies was slightly higher (1.90 MoM) than the median of total hCG MoM values (1.72 MoM) but a lower standard deviation was stated for the logarithmic hCG MoM values (SD = 0.49) compared with free betahCG MoM values (SD = 0.51). A two-tailed Student's t test revealed no significant differences of hCG and free betahCG MoM values in both the affected and unaffected pregnancies. By inclusion of free betahCG the discriminant analysis classified 26 out of 33 affected cases correctly and 45 out of 188 unaffected cases incorrectly. For the inclusion of hCG these ratios were 25/33 and 41/188, respectively. Taking in account the individual maternal age risks at a defined false-positive rate of 5% including free betahCG yielded a higher detection rate than including hCG. However, using 1:380 (age-related at-term risk of a 35-year-old woman) as a cut-off risk including hCG yielded a higher detection rate than including free betahCG. CONCLUSION: For the observed cases none of the markers, hCG or free betahCG, was superior in Down syndrome screening.

Adult↗

Maxillary dental arch form related to voice classification: a pilot study.

This pilot study evaluated maxillary dental arch form dimensions and volume to determine if these parameters could be predictors for or related to voice classification. Nine white female professional singers ranging in age from 26 years to 53 years were studied. A maxillary dental impression and stone dental casts were made using standard dental procedures. Measurements were made from 10 points on each cast to determine the depth of the palate measured from first molar (depth A) and from first bicuspid (depth B), the width measured from cuspid-to-cuspid (width A) and from second molar to second molar (width B), and the length of the palate. An impression of the palatal arch of each cast was made to determine the volume of the palate using fluid displacement methods. Audio recordings were made for each subject, and based on speaking fundamental frequency, spectral analysis, voice profile, and tessitura confirmation, the actual voice classification of each subject in soprano, mezzo, and alto was achieved. Correlation and discriminant analysis tests were performed on the data. The discriminant analysis revealed that no single measurement is a predictor for voice classification. However, the discriminant analysis applied to the predictors depth A, depth B, and volume gives optimal results, ie, each subject was classified in her true group.

Adult↗

Detection of a "faked" strength task effort in volunteers using a computerized exercise testing system.

The objective of this study was to develop an experimental method to separate a "faked" strength effort from a "best" effort in volunteers. Thirty-four pain-free volunteers (18 males, 16 females) performed a shoulder press and pull-down on an isokinetic computerized exercise testing system (CETS), giving a best effort followed by a faked effort. Two months later, a randomly selected subgroup (6 males) repeated the experiment to test the predictive validity of the derived variables. In the statistical analysis, best efforts were first compared with fake efforts by paired ttest for 80 CETS variables for males and females separately. Variables showing a strong difference between the best and faked effort were then selected for further analysis. In the second step of the analysis, the method of multiple correlations (r2 method) was used to reduce the number of redundant CETS variables to five in both the male and female groups. In the third step, a stepwise discriminant analysis was used to select predictor variables for the male and female groups. For the variables selected by the discriminant analysis for both males and females, sensitivities and specificities were calculated. Finally, the developed discriminant formula was used in the predictive validity part of the study to determine the sensitivities and specificities of the developed method. The discriminant analysis selected the following CETS variables for male and female groups, respectively: duty cycle down, work weight/down, peak value up (males); and average power up, 40% repetition down, duty cycle up (females). For males, using their three variables, the discriminant function classified 77.14% of the efforts correctly with 88.9% sensitivity and 64.7% specificity. For females, using their three variables, the discriminant function classified 90.63% of the efforts correctly with 100% sensitivity and 81.3% specificity. In the predictive validity group, the discriminant function classified 75% of the efforts correctly with 83.3% sensitivity and 66.7% specificity. This pilot study indicates that the method developed here may be useful in the experimental study for the discrimination between faked and best efforts on this isokinetic CETS machine. Future studies using this method will need to involve a larger number of volunteers.

Adult↗

Discriminative factor analysis of juvenile delinquency in South Korea.

PURPOSE: The present study was intended to compare difference in research variables between delinquent adolescents and student adolescents, and to analyze discriminative factors of delinquent behaviors among Korean adolescents. METHODS: The research design of this study was a questionnaire survey. Questionnaires were administered to 2,167 adolescents (1,196 students and 971 delinquents), sampled from 8 middle and high school and 6 juvenile corrective institutions, using the proportional stratified random sampling method. Statistical methods employed were Chi-square, t-test, and logistic regression analysis. RESULTS: The discriminative factors of delinquent behaviors were smoking, alcohol use, other drug use, being sexually abused, viewing time of media violence and pornography. Among these discriminative factors, the factor most strongly associated with delinquency was smoking (odds ratio: 32.32). That is, smoking adolescent has a 32-fold higher possibility of becoming a delinquent adolescent than a non-smoking adolescent. CONCLUSIONS: Our findings, that smoking was the strongest discriminative factor of delinquent behavior, suggest that educational strategies to prevent adolescent smoking may reduce the rate of juvenile delinquency. Antismoking educational efforts are therefore urgently needed in South Korea.

Adolescent↗

Representation of orientation and spatial frequency in perception and memory: a choice reaction-time analysis.

Discrimination and short-term memory for the orientation of sinusoidal gratings that differed in spatial frequency, and for the spatial frequency of gratings that differed in orientation, were measured in a same-different task with 0-10 s interstimulus intervals (ISI) between test and reference stimuli. Introducing a difference between test and reference stimuli on a second dimension, or increasing ISI, did not impair spatial discrimination in terms of accuracy, but choice reaction times for correct decisions were prolonged by both manipulations. Results suggest that perceptual discrimination is based on representations in which orientation and spatial frequency are conjointly coded and that decisions are reached by a serial process scanning multiple-tuned, labeled channels; short-term memory may involve reactivation of these channels.

Analysis of Variance↗

Solid-phase microextraction, gas chromatography, and mass spectrometry coupled with discriminant factor analysis and multilayer perceptron neural network for detection of Escherichia coli.

This study was performed to investigate the ability of using discriminant factor analysis (DFA) and an artificial neural network (ANN) to identify and quantify the number of Escherichia coli (ATCC 25922) in nutrient media from data generated by analysis of E. coli volatile metabolic compounds using solid-phase microextraction (SPME) coupled with gas chromatography (GC) and mass spectrometry (MS). E. coli was grown in super broth and incubated at 37 degrees C for 2 to 12 h. Numbers of E. coli were followed using a colony counting method. An SPME device was used to collect the volatiles from the headspace above the samples, and the volatiles were identified using GC-MS. DFA was used to classify the samples from different incubation times. From DFA, it was possible to differentiate super broth from media containing E. coli when cell numbers were 10(5) CFU or more. The potential to predict the number of E. coli from the SPME-GC-MS data was investigated using a multilayer perceptron (MLP) neural network with back propagation training. The MLP comprised an input layer, one hidden layer, and an output layer, with a hyperbolic tangent sigmoidal transfer function in the hidden layer and a linear transfer function in the output layer. Good prediction was found as measured by a regression coefficient (R2 = 0.996) between actual and predicted data.

Colony Count, Microbial↗

[Automatic recognition of age - specific development of the EEG in infancy and early childhood - a five year follow-up study (author's transl)].

This study was initiated with a view to early detection of dysfunctions of the central nervous system (CNS) and to observe the physiological development of the CNS by means of EEG obtained in the age-groups: new-born, half, 1, 2, 3, 4 and 5 years. This was supported by standardized clinical examinations for neurological and somatic findings and by standardized tests of psychomotoric and intellectual development. In this investigation the data as used for automatic processing consisted of 110 EEGs in the age-groups: half, 1, 2, 3 and 4 years. Bipolar EEGs were recorded using "ten-twenty" method with the following leads: F4--C4, P4--O2, F3--C3 and P3--O1. For feature extraction three methods of data reduction were applied, i.e. interval-amplitude, spectral analysis and the autoregressive model. From the features thus obtained, age-specific frequency parameters were selected by statistical methods; further evaluation was performed by cluster and discriminant analysis. The former showed an unequivocal case grouping for each age-group. Using unmatched samples of only clinically healthy children from the three age-groups 1/2, 1 and 2 years, linear discriminant analysis were applied to the parameters of the three methods of data reduction. This procedure yields in mean recognition rates of 95%, 98% resp. 98% in hold-one-out classification. Similar results are obtained with samples from four or five age groups (1/2 to 4 years). The results of this type of automatic EEG analysis show that discriminant analysis can be used to allocate EEGs to specific age-groups; hence, it may readily be ascertained whether differences between chronological age and development age as evidenced by the EEG exist.

Age Factors↗

Combined evaluation of rest-redistribution thallium-201 tomography and low-dose dobutamine echocardiography enhances the identification of viable myocardium in patients with chronic coronary artery disease.

The purpose of this study was to evaluate whether combined evaluation by discriminant analysis of rest-redistribution thallium-201 tomography and low-dose dobutamine echocardiography enhances the accuracy in identifying viable myocardium in patients with chronic coronary artery disease. Rest-redistribution 201Tl has high sensitivity but low specificity in identifying viable myocardium, while the opposite is true for low-dose dobutamine echocardiography. Forty-six patients underwent low-dose dobutamine echocardiography and rest-redistribution 201Tl tomography on the same day. Rest echocardiography was repeated at least 30 days (mean 40+/-20) after myocardial revascularization. Discriminant analysis was applied to the results of 201Tl tomography and dobutamine echocardiography to classify a/dyskinetic segments as viable or non-viable. In 92 a/dyskinetic segments that were revascularized, rest-redistribution 201Tl tomography yielded an accuracy of 75%, while the accuracy of dobutamine echocardiography was 70% (P<0.05). When discriminant analysis was used, the combined evaluation gave an accuracy of 83% (P<0.05 vs both tests). These findings demonstrate that low-dose dobutamine echocardiography and 201Tl imaging are useful and complementary techniques for identifying viable myocardium in patients with chronic coronary artery disease. Combined evaluation by discriminant analysis significantly improves accuracy, although the cost-effectiveness of such an approach remains to be determined.

Adrenergic beta-Agonists↗

Value of counting colonic mucosal Ig-containing cells in the differential diagnosis of chronic inflammatory bowel disease.

AIMS: To investigate whether counting cells containing immunoglobulin (Ig) subclass in colonic biopsy specimens of patients with chronic inflammatory bowel disease, in addition to conventional histological evaluation, can improve the differentiation of patients with Crohn's disease from those with ulcerative colitis. METHODS: The colonic and rectal biopsy specimens of 40 patients with chronic inflammatory bowel disease, comprising 20 patients with Crohn's disease and 20 with ulcerative colitis, were used and sections were stained specifically for IgA, IgM, and IgG heavy chains using an indirect immune peroxidase method. The immunoglobulin subclass containing cells were counted using an ocular grid counting method in a light microscope. A linear stepwise discriminant analysis was performed on Ig subclass containing cell counts in combination with 16 reproducible histological features. The results of this discriminant analysis were compared with the results of the discriminant analyses in which only the histological features were used. RESULTS: Applying stepwise discriminant analysis, two histological features (an excess of histiocytes in the lamina propria and the villous or irregular aspect of the mucosal surface) in combination with IgMax were selected as the most discriminatory parameters that distinguish Crohn's disease from ulcerative colitis. IgMmax was defined as the maximum value of the mean percentage of IgM containing cells over all the biopsy locations. The use of this combination resulted in a better classification in 20% of the patients with Crohn's disease and in 9% of the patients with ulcerative colitis compared with the use of histological features alone. CONCLUSIONS: Morphometric enumeration of Ig subclass containing cells in colonic mucosal biopsy specimens has diagnostic value as a means of differentiating individual patients with Crohn's disease from those with ulcerative colitis.

Adolescent↗

On the utility of the West Haven-Yale Multidimensional Pain Inventory.

STUDY DESIGN: Two studies evaluated the West Haven-Yale Multidimensional Pain Inventory. This 52-item inventory contains 12 scales divided into 3 parts: 1) interference, support, pain severity, self-control, and negative mood; 2) punishing responses, solicitous responses, and distracting responses; and 3) household chores, outdoor work, activities away from home, and social activities. OBJECTIVES: The objective of study 1 was to investigate the internal structure of the Multidimensional Pain Inventory and its demographic correlates. The objective of study 2 was to explore its external correlates (construct validity). SUMMARY OF BACKGROUND DATA: The Multidimensional Pain Inventory appears promising because of its brevity, scope, ease of administration, and initial results. Despite this, it has attracted very little attention in the orthopedic literature since its introduction nearly a decade ago. METHOD: The internal structure was examined using oblique multiple group confirmatory factor analysis. Demographic correlates were examined using discriminant analysis. The external correlates were examined using multiple regression and discriminant analysis. RESULTS: The items' proposed structure fit well except that two pairs of scales are poorly separated: 1) activities away from home and social activities, and 2) solicitous responses and distracting responses. Correlations with external criteria are meaningful. CONCLUSIONS: The Multidimensional Pain Inventory meets standards of reliability and convergent validity, and it may be an improvement over current psychometric devices used to this same end.

Black or African American↗

Cranial and mandibular morphometry in Leontopithecus Lesson, 1840 (Callitrichidae, primates).

In this paper, we report on a craniometric analysis comparing the species of lion tamarins, Leontopithecus Lesson, 1840. Seventeen cranial and mandibular measures were taken on skulls of 59 adult crania: 20 L. rosalia (14 females and 6 males); 13 L. chrysomelas (6 females and 7 males); 23 L. chrysopygus (8 females and 15 males), and 3 L. caissara (1 female and 2 males). All specimens were from the Rio de Janeiro Primate Center (CPRJ-FEEMA, Brazil), except the specimens of L. caissara. Statistical treatment involved a one-way analysis of variance (the Bonferroni test) and discriminant analysis, comparing cranium and mandibles separately to determine variables which best distinguished groups and to group the specimens, using size corrected methods. The Mahalanobis distance was computed from the centroids of each group. Seven measures distinguished females of L. chrysopygus with L. rosalia, six to L. rosalia with L. chrysomelas, and L. chrysopygus with L. chrysomelas. In males, the numbers of measures statistically different were 5, 4, and 3 of the pairwise comparisons above mentioned. Cranial base length and orbital breadth were the only measures that were significantly different in all three dyads, considering both sexes. For the cranium, function 1 of the Discriminant Analysis accounted for 52.4% of the variance and function 2 accounted for 40.3%. Both functions exhibited a significant value for Wilks' lambda (P<0.0001) and 96.6% of specimens were correctly classified. For the mandible, the first two functions provided a significant discrimination 51.1% and 44.9%, respectively, and 69.5% of the correct classification. Orbital breadth and cranial base length contributed most in the cranial analysis, while mandibular length and mandibular body height to mandibular ones. The analyses performed in this study (univariate and multivariate) demonstrated that cranial and mandibular morphology is significantly different among species of Leontopithecus. Despite of sample size, L. caissara shows morphological distances to L. chrysopygus in cranial analysis. However, other investigations are necessary to confirm this.

Analysis of Variance↗

Bayesian-logit model for risk assessment in coronary artery bypass grafting.

Predictive models for the assessment of operative risk using patient risk factors have gained popularity in the medical community as an important tool for the adjustment of surgical outcome. The Bayes' theorem model is among the various models used to predict mortality among patients undergoing coronary artery bypass grafting procedures. Comparative studies of the various classic statistical techniques, such as logistic regression, cluster of variables followed by a logistic regression, a subjectively created sickness score, classification trees model, and the Bayes' theorem model, have shown that the Bayes' model is among those with the highest predictive power. In this study, the Bayes' theorem model is reformulated as a logistic equation and extended to include qualitative and quantitative risk factors. We show that the resulting model, the Bayesian-logit model, is a mixture of logistic regression and linear discriminant analysis. This new model can be created easily without complex computer programs. Using 12,712 patients undergoing coronary artery bypass grafting procedures at the Department of Veterans Affairs Continuous Improvement in Cardiac Surgery Study between April 1987 and March 1990, the predictive power of the Bayesian-logit model is compared with the Bayes' theorem model, logistic regression, and discriminant analysis. The ability of the Bayesian-logit model to discriminate between operative deaths and operative survivors is comparable with that of logistic regression and discriminant analysis.

Age Factors↗

OASYS-3: improved analysis of serial peak expiratory flow in suspected occupational asthma.

Serial peak expiratory flow (PEF) records are useful for the screening and diagnosis of patients with occupational asthma. We have presented work on a method based on linear discriminant analysis (OASYS-2) and have now tried to improve on this technique and investigate the repeatabilities of expert interpretation. 268 serial PEF records made by workers with possible occupational asthma were divided into four sets. The first two were development sets; development set 1 was used to develop a discriminant analysis to match the human expert, who scored each complex (a sequence of work days, then rest days, then work days, or its counterpart: rest-work-rest) for its likelihood of having a work-related effect. This was modelled using 729 new measurements from each complex, together with the 50 measurements originally used in the development of OASYS-2. Linear discriminant analysis is a statistical technique which can refine from a large number of indices a narrower range that best predict a known outcome. Combinations of these indices are weighted and used to assign outcomes for novel cases. Models were produced containing a combination of measurements that generated scores best matching the expert. Development set 2 was used to test the model. Sets 3 and 4 were "gold standard" sets where the diagnosis had been made independently of the PEF record. Set 3 was used to set the cut-off for an occupational effect, the sensitivity and specificity for the combined model was determined from the fourth (gold standard) set. The fourth set was also used to determine the sensitivity and specificity of the human expert. The repeatability of the human expert re-scoring the same complexes had a weighted kappa score of 0.71. OASYS-3 was 100% repeatable. Comparing the scores awarded to whole records by the new (OASYS-3) and OASYS-2 analysis methods with the scores awarded by the human expert revealed mean (95% CI) differences of -0.28 (0.30, -0.26) and -0.34 (-0.37, -0.31) respectively. Hence both OASYS-3 and OASYS-2 tended to score records less positively for work-related changes in PEF than the expert. OASYS-3 scored complexes marginally better than OASYS-2. The sensitivity of OASYS-3 was better than OASYA-2 (82% and 75% respectively) for an equivalent specificity (94%). The sensitivity of the human expert was 100% with a specificity of 93%. OASYS-3 provides an objective method of interpreting serial peak flow records with a sensitivity and specificity approaching that of a human expert and is a modest improvement on OASYS-2.

Adult↗

The prognostic value of somatosensory evoked potential monitoring and tumor data in supratentorial tumor removal.

The aim of this study was to find appropriate somatosensory evoked potential (SSEP) and tumor data that would predict immediate postoperative outcome. Seventeen patients were evaluated, all with supratentorial mass lesions. Intraoperative SSEP monitoring was carried out, and central conduction time, N20 amplitude, and N20 to N13 amplitude ratio were recorded throughout the operation. The differences between the end and the start of the procedure and between the end of the operation and the most substantial changes during tumor removal were calculated for statistical evaluation. Tumor location and extent were carefully determined and calculated by computed tomographic reconstructions in the axial and sagittal planes. Several tumor parameters were measured: the distances from the midpoint of the tumor to the central sulcus, midline, and base of the skull and the longitudinal and cross-sectional diameters. The electrophysiologic and tumor data sets were correlated with the immediate postoperative neurologic deterioration. For this purpose, patients were divided into two groups: group 1 = no neurologic deterioration after operation (13 patients); group 2 = neurologic deterioration after operation (4 patients). The difference in central conduction times between the end and the start of the procedure was the only variable that exhibited a significant influence on the immediate postoperative condition (P = 0.007), as determined by analysis of variance. The other electrophysiologic data as well as the tumor data failed the significance threshold of P = 0.05. Discriminant analysis was applied to test the classification capacity of the measured variables. Involving all measured variables (electrophysiologic and tumor data), discriminant analysis allowed a correct classification of all 17 patients to their proper neurologic deterioration group. Discriminant analysis for SSEP data alone led to 15 correct classifications. Tumor data, used alone for discriminating procedures, revealed 14 correct classifications. When each variable was analyzed separately, only the difference in central conduction times between the end and the start of the procedure gave significant predictions, namely, 15 correct classifications. This was the same number as achieved by all electrophysiologic variables together. No other variable could on its own yield any valid prognosis for assessment of immediate postoperative neurologic deterioration. The data confirm the importance of central conduction time recovery before the end of the operation on the patient's immediate postoperative neurologic condition.

Adult↗

Artificial neural networks for medical classification decisions.

Neural networks are growing in popularity. However, empirically comparing them with traditional methods of solving medical classification decisions is not trivial. Following a brief description and comparison with the statistical technique, discriminant analysis, the methodology for comparing neural networks with discriminant analysis is described. The advantages and disadvantages of neural networks are summarized to support a conclusion that neural networks are beneficial and destined to proliferate. Areas of fruitful future research are identified.

Decision Making, Computer-Assisted↗

Multivariate QSAR analysis of a skin sensitization database.

There is a regulatory requirement for the potential of a new chemical to cause skin sensitization to be assessed. This requirement is presently fulfilled by the use of animal tests. In this study a data base of heterogeneous organic compounds from the guinea pig maximization test has been subjected to multivariate QSAR analysis. The compounds were described both by whole molecule parameters and structural features associated with likely sites of reactivity. Principal component analysis was applied to the data set and although it functions reasonably well to reduce the dimensionality of a large data matrix, it is only moderately useful as a predictive tool when descriptors were chosen rationally. Stepwise discriminant analysis produces a fourteen parameter model, of which twelve were structural features associated with reactivity. This however predicts only 82.6% of compounds correctly after cross validation. There is trend for the linear discriminant analysis model to predict compounds as non sensitizers, suggesting that the parameters incorporated were not wholly suitable for discriminating between the two classes. Another criticism of linear discriminant analysis is that it may be unable to cope with the likely embedded data structure. With this in mind, the structural alerts may be better employed in an expert system, to identify potential hazard, where they will not suffer the limitations of a statistical model.

Animals↗