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 973 records · Page 54Linked to original sources
PubMedDate not supplied

Factors affecting the discriminant function analysis of blood chemistry profiles.

The discriminant function analysis (DFA) of a battery of blood chemistry tests is generally more sensitive and specific than single biochemical test in detecting alcoholism or hazardous drinking. This study examined factors affecting the DFA of blood chemistry profiles (BCP) from alcoholics (ALC) in treatment, primary care outpatients (PC), and the general population (GP). Our data indicate that of the 32 variables in the BCP, 14 are not correlated with alcohol intake or consequences and can be deleted from the set considered for use in the DFA. Confounding medical conditions in the data set used in the DFA did not adversely affect the correct classification of ALC or heavy drinkers. It was not necessary to use only extreme groups (e.g., alcoholics vs. light drinkers/abstainers) to develop the discriminant function. Transformation of some blood variables greatly improved the selection of key variables in the DFA. Another finding was that there was little benefit from fitting the DFA to the demographics of subjects. When the BCP was combined with the results of a brief questionnaire, the TWEAK, it only improved slightly the classification power of the DFA based only on the TWEAK questions.

Adolescent↗
PubMed2005

Latent structure of dermatoglyphs in the population of Selska Valley.

The historical records of Selska Valley reveal that the eastern part of this area was first settled by Slovene agrarian colonists, the western part by German colonists and the central part by Friulians. These were later followed by Slovene and Slovenized settlers, who penetrated the valley from north to south. Because of its reproductive isolation, the population of Selska Valley is highly suitable for the study of population structures. The quantitative traits of the digital and palmar dermatoglyphs are polygenetically determined characteristics, which, due to their selective inertness to changes, may provide an insight into microevolutionary processes. The purpose of our study was to identify the possible differences between the populations of villages in the valley and the mountain villages attributable to various migration flows through history. Altogether 340 finger and palm prints of 163 males and 177 females were collected in two groups of villages: (1) the lowland villages (Praprotno, Bukovica, Sevlje, Dolenja vas, Selca, Zelezniki and Zali log), and (2) the mountain villages (Podlonk, Prtovc, Spodnje Danje, Zgornja Sorica and Spodnja Sorica). The 18 dermatoglyphic variables were analyzed. A statistical analysis using standard methods was performed and the latent structure evaluated using factor analysis. The discriminant analysis and latent structure of the quantitative properties of dermatoglyphs suggest the presence of certain differences in gene pools of two studied populations (the group of villages in the valley and the group of mountain villages). It is highly probable that these differences can be attributed to low migration in the Selska Valley and to the 'selective inertness' of quantitative dermatoglyphic traits. In a previous study, no significant biological differences between the studied populations were found in qualitative dermatoglyphic traits. This indicates that Selska Valley and its village populations represent a specific isolate, and therefore expanded studies of this type could significantly contribute to a better understanding of the population concerned and isolates in general.

Analysis of Variance↗
PubMed1999

Host-dependent and seasonal variation in opisthaptoral hard parts of Gyrodactylus cf. arcuatus from three Pomatoschistus spp. and G. arcuatus from Gasterosteus aculeatus: a multivariate approach.

Gyrodactylus arcuatus-like monogeneans collected from the gills of 3 goby species, Pomatoschistus minutus, P. lozanoi and P. pictus, from the Voordelta Area (SW Netherlands) were compared on a morphometric basis with G. arcuatus from its type host, the three-spined stickleback, Gasterosteus aculeatus. Univariate and multivariate analysis was carried out on measurements of 17 characters of the opisthaptoral hard parts of 268 specimens. The contribution of host-dependent and seasonal factors to the observed morphological variation are discussed. Univariate statistics (ANOVA) were used to detect features that were useful in separating the gyrodactylids from the different host groups. Subsequent factor analysis and discriminant analysis, combining all variables, led to the separation of 3 distinct forms (possibly species) dependent on the host species harbouring them. G. arcuatus (sensu stricto) was only present on G. aculeatus, while P. minutus and P. lozanoi carried a common species, Gyrodactylus sp. 1, and P. pictus hosted Gyrodactylus sp. 2. Seasonal variation in size of the sclerities was demonstrated on the Gyrodactylus species from P. minutus and P. lozanoi, and should be taken into consideration when specimens are compared. The findings are discussed in view of the possibility of phylogenetic host specificity of the gyrodactylids occurring on goby species that live sympatrically in the Voordelta area.

Analysis of Variance↗
PubMed1982

Genetic relationships among 5 inbred strains established from common ancestor, dd mouse, as assessed by mandible analysis.

By multivariate analysis using mandible measurements genetic relationships were investigated among 5 inbred strains, DDD, DDK, DSD, DD/Tbr and DDY, derived from the common ancestor, dd mouse. (1) In canonical discriminant analysis (discriminant analysis with reduction of dimensionality), 3 DD strains, DDD, DSD, and DDY, were closer to one another than to any control strain on planes Z1-Z2, Z1-Z3, and Z2-Z3. Only the DDK strain was quite far from those 3 strains. (2) In the discriminant analysis, however, the 5 inbred strains could be classified correctly at the rate of 100%. They were genetically different from one another. (3) The causes for this genetical difference seemed to be the genetic heterogeneity of the ancestor dd mouse and the method of inbreeding which had been independent since the separation.

Analysis of Variance↗
PubMed1991

[Analysis of maximal expiratory flow-volume curves--visual evaluation and quantitative evaluation].

Maximal expiratory flow-volume (MEFV) curves have been utilized in the diagnosis of chronic obstructive lung diseases, in the survey of air pollution, and in studies of respiratory functions in school children. Indices for the lower lung volumes of MEFV curves have been used in statistical analysis or individual evaluation because of their effort independence, while indices for higher lung volumes of MEFV curves, which were considered to be effort-dependent, have not been used. A few reports using indices for higher and lower lung volumes have been described because the indices for higher lung volumes are as reproducible to the maximal expiration as those for lower lung volumes. Recently, MEFV-type analysis has been studied and this analysis has been useful in assessing the effects of smoking on MEFV curves, or in assessing subjects with nasal allergies or those with prior nasal allergies. From the aspects of hygiene, this review mainly describes MEFV-type analysis and the quantitative analysis of discriminant analysis.

Adult↗
PubMed1996

The contribution of frictional torque to loosening at the cement-bone interface in Tharies hip replacements.

We evaluated the results in a series of 170 osteoarthrotic hips (156 patients) at a median of 10.0 years (range, 3.0 to 16.4 years) after a Tharies resurfacing arthroplasty performed with cement. Our purpose was to assess the role of frictional torque in loosening at the cement-bone interface. The hips were divided into three groups on the basis of the diameter of the bearing surface: small (thirty-six, thirty-nine, or forty-three millimeters), medium (forty-seven millimeters), and large (fifty-one or fifty-four millimeters). Comparisons were made with Kaplan-Meier survivorship analysis, with stepwise discriminant analysis of risk factors for aseptic loosening. Revision for aseptic loosening was the end point for survivorship analysis. Despite higher frictional torques due to the increased diameter of the bearing surface and the increased average load, the larger prostheses survived significantly longer than the smaller ones (p=0.002). Stepwise covariate discriminant analysis indicated that the size of the bearing surface was the only factor identified that significantly affected survival. In six patients who had had a bilateral surface replacement with a component of a different size in each hip, the smaller prosthesis always failed first (p=0.001), at an average of seventy-seven months; the larger prosthesis failed at an average of 113 months. Nine patients had had a bilateral replacement with components of the same size, and there was no significant difference in the durations of survival of the two prostheses. Analysis of radiographs and retrieved specimens indicated that, regardless of the size of the component, the mechanism of loosening on both the acetabular and the femoral side of this so-called double-cup replacement was progressive resorption of bone induced by polyethylene wear particles that compromised the fixation of the components. More time was required for the process to disrupt++ the larger fixation area of the larger components. These data indicate that frictional torque was not the primary factor in the loosening of these prostheses with a large bearing surface and that higher friction and frictional torques can be tolerated if the generation of wear debris is sufficiently limited. These findings may be important as alternatives to polyethylene bearing surfaces are investigated.

Acetabulum↗
PubMed1998

Development of medical decision support system for leukemia management.

A leukemia management system was developed in this study which is comprised of four modules: registry, diagnosis, prognosis of treatment, and CAI. The emphasis was on patient management as a whole. Three knowledge models were developed to predict accurate diagnosis and prognosis of treatment: case-based reasoning, neural network, and discriminant analysis, Of these, discriminant analysis model produced the most accurate diagnosis, whereas neural network produced the most accurate prognosis of treatment.

Decision Making, Computer-Assisted↗
PubMed2002

Development and testing of the Labor Support Scale.

Nurses use interventions and supportive strategies to help mothers throughout labor, yet little research examines intrapartum nurses' labor support techniques. The purpose of the study was to develop and test a self-report instrument, the Labor Support Scale (LSS), to describe the frequency with which nurses perform interventions and nurses' perceptions of the helpfulness of interventions. Steps for instrument development were item generation, content validity testing, piloting, refining, and administering the questionnaire in two studies (n = 307, n = 472). Internal consistency reliability was .90 and .92 for the frequency and helpfulness portions of the instrument (respectively). Exploratory factor analysis, known groups technique, content analysis, and discriminant analysis evaluated validity. In both phases, instrument psychometrics provided evidence of content, construct, and discriminant validity.

Discriminant Analysis↗
PubMed1993

Diagnostic ECG classification based on neural networks.

This study illustrates the use of the neural network approach in the problem of diagnostic classification of resting 12-lead electrocardiograms. A large electrocardiographic library (the CORDA database established at the University of Leuven, Belgium) has been utilized in this study, whose classification is validated by electrocardiographic-independent clinical data. In particular, a subset of 3,253 electrocardiographic signals with single diseases has been selected. Seven diagnostic classes have been considered: normal, left, right, and biventricular hypertrophy, and anterior, inferior, and combined myocardial infarction. The basic architecture used is a feed-forward neural network and the backpropagation algorithm for the training phase. Sensitivity, specificity, total accuracy, and partial accuracy are the indices used for testing and comparing the results with classical methodologies. In order to validate this approach, the accuracy of two statistical models (linear discriminant analysis and logistic discriminant analysis) tuned on the same dataset have been taken as the reference point. Several nets have been trained, either adjusting some components of the architecture of the networks, considering subsets and clusters of the original learning set, or combining different neural networks. The results have confirmed the potentiality and good performance of the connectionist approach when compared with classical methodologies.

Cardiomegaly↗
PubMed2006

Evaluation of four modelling techniques to predict the potential distribution of ticks using indigenous cattle infestations as calibration data.

Efficient tick and tick-borne disease control is a major goal in the efforts to improve the livestock industry in developing countries. To gain a better understanding of the distribution and abundance of livestock ticks under changing environmental conditions, a country-wide field survey of tick infestations on indigenous cattle was recently carried out in Tanzania. This paper evaluates four models to generate tick predictive maps including areas between the localities that were surveyed. Four techniques were compared: (1) linear discriminant analysis, (2) quadratic discriminant analysis, (3) generalised regression analysis, and (4) the weights-of-evidence method. Inter-model comparison was accomplished with a data-set of adult Rhipicephalus appendiculatus ticks and a set of predictor variables covering monthly mean temperature, relative humidity, rainfall, and the normalised difference vegetation index (NDVI). The data-set of tick records was divided into two equal subsets one of which was utilised for model fitting and the other for evaluation, and vice versa, in two independent experiments. For each locality the probability of tick occurrence was predicted and compared with the proportion of infested animals observed in the field; overall predictive success was measured with mean squared difference (MSD). All models exhibited a relatively good performance in configurations with optimised sets of predictors. The linear discriminant model had the least predictive success (MSD>or=0.210), whereas the accuracy increased in the quadratic discriminant (MSD>or=0.197) and generalised regression models (MSD>or=0.173). The best predictions were gained with the weights-of-evidence model (MSD>or=0.141). Theoretical as well as practical aspects of all models were taken into account. In summary, the weights-of-evidence model was considered to be the best option for the purpose of predictive mapping of the risk of infestation of Tanzanian indigenous cattle. A detailed description of the implementation of this model is provided in an annex to this paper.

Animals↗
PubMed1998

Comparison of classification approaches applied to NIR-spectra of clinical study lots.

NIR-spectroscopy combined with pattern recognition approaches is applied to classify samples of clinical study lots in the pharmaceutical industry. The performance of linear discriminant analysis (LDA), quadratic discriminant analysis (QDA) and K-nearest neighbour (KNN) method is evaluated on a tablet data set and a capsule data set. To establish a classification model a strategy is followed, which is described in this work. Frequently, in the pharmaceutical industry, several batches of the same clinical study lot are produced. We tested whether it is possible to merge several batches in one class for modelling or, instead, whether it is necessary to model each batch individually.

Capsules↗
PubMedDate not supplied

Characterization of spermatozoa by planar morphometry.

OBJECTIVE: To define the morphometric characteristics of normal sperm heads, and compare them to sperm head measurements used to define normal morphology using strict criteria. DESIGN: Computerized image analysis of selected normal and abnormal seminal fluid specimens collected for routine male fertility studies. SETTING: Research laboratory at the Department of Medical Technology, Bowling Green State University, Bowling Green, OH. PATIENTS: Sixty adult male patients who submitted semen samples for routine analysis. Fifty percent had normal seminal fluid analysis results. The remaining 50% demonstrated abnormal sperm morphology. CRITERION STANDARD: Microscopic evaluation of sperm head morphology. Sperm fitting the criteria of normal as defined by WHO (1987 and 1992) and Kruger (1988) were classified as normal. Sperm with a post nuclear area of less than 40% were classified as acrosomal deficient. MAIN OUTCOME MEASURES: Measurements made from stained seminal fluid smears included sperm head area, perimeter, acrosomal area, percent acrosome, Ferret's diameters, aspect ratio, shape factor, and specific length. Normal sperm heads (NL group) were compared to sperm heads demonstrating an acrosomal deficiency (AD group) for statistically significant differences using multivariate analysis of variance (MANOVA) and analysis of variance (ANOVA). Stepwise discriminant analysis was used to remove duplicating variables. Discriminant analysis was used to classify the sperm heads into NL and AD groups. Receiver-operating characteristic (ROC) curves were applied to the 2 most influential variables in order to identify a cutoff that best distinguishes normal from acrosomal deficient sperm heads. RESULTS: MANOVA and ANOVA showed all 10 variables to be statistically significant (p < .002). Discriminant analysis correctly assigned 98.7% of the normal sperm heads to the NL group and 99.0% of sperm with acrosomal deficiency to the AD group. The percent acrosome and acrosomal area were determined to be the 2 most influential variables. ROC analysis identified a cutoff of 3.6 mu2 for acrosomal area as having the highest sensitivity and specificity (99.7% and 88.0%, respectively). Similarly, a cutoff of 44% for percent acrosome gave a sensitivity of 92.3% and a specificity of 88.7%. The coefficient of variation (CV) for each of the 10 variables determined from 20 day-to-day replicates of a normal semen smear ranged from 2.6% to 8.4%. CONCLUSION: Computerized image analysis is able to define a reference range for sperm head area, percent acrosome, and acrosomal area that may be used to differentiate normal form abnormal sperm heads. Maximum and minimum Ferret's diameters measure sperm head length and width, respectively. Mean maximum Ferret's diameter, minimum Ferret's diameter, and maximum-minimum Ferret's diameter ratio correspond closely to the WHO (1987) midpoint for normal sperm head length, width, and length-width ratio. The average percent acrosome of normal sperm heads determined by morphometry closely correlate to the WHO (1992) and Kruger (1988) midpoints for percent acrosome.

Acrosome↗
PubMed2003

Morphologic reappraisal of serrated colorectal polyps.

The "hyperplastic polyp" is considered a benign lesion with no malignant potential, whereas "serrated adenoma" is a precursor of adenocarcinoma. The morphologic complexity of the serrated adenoma varies from being clearly adenomatous to being difficult to distinguish from hyperplastic polyp, which creates a need for more detailed morphologic analysis of all serrated polyps. We evaluated 24 morphologic variables in 289 serrated polyps from the colon and rectum. Cluster analysis and discriminant analysis were performed. A subset of polyps was immunostained for hMLH1 and hMSH2. Major differences were found between right-sided and left-sided polyps. A distinct group of serrated polyps with abnormal proliferation was identified throughout the colon and rectum. These polyps demonstrated decreased expression of hMHL1 and hMSH2 compared with polyps with normal proliferation. Left-sided serrated polyps with normal proliferation further clustered into three groups: vesicular cell-type, goblet cell-type, and mucin-poor-type. We recommend evaluation of the localization, size, and morphologic features when serrated polyps are included in colorectal carcinogenesis research. Polyps with abnormal proliferation are similar to the polyps in "hyperplastic polyposis" and, because of their decreased expression of hMLH1 and hMSH2, may be the subset of polyps associated with the development of colorectal carcinoma via the microsatellite instability pathway.

Adaptor Proteins, Signal Transducing↗
PubMed1986

Clinical significance of statistical discrimination employing serum TA-4, TPA, and CEA levels in uterine cervical cancer.

The serum levels of TA-4, TPA, and CEA were measured simultaneously in 52 patients with uterine cervical cancer and 99 healthy females. The findings were examined statistically by discriminant analysis method. The diagnostic usefulness of discriminant analysis employing the serum values for these three tumor markers was studied for uterine cervical cancer and compared with the diagnostic efficacy using each tumor marker alone. The following results were obtained: In the uterine cervical cancer patients, the frequencies of cases with the elevated serum levels were 55.8% for TA-4, 46.2% for TPA, and 36.5% for CEA. The functional formula, Z = 1.8395 log TA-4 + 3.5314 log TPA + 1.0130 log CEA-7.7352, was obtained by the stepwise discriminant analysis method. With the values obtained from this formula, 71.2% of patients with uterine cervical cancer were correctly classified in the uterine cervical cancer group. In the uterine cervical cancer patients, the values obtained from this formula increased according to the progression of the clinical stage and reflected the disappearance or persistence of tumor lesions after several therapies. From the above results, it was concluded that the discriminant analysis employing the serum values for three tumor markers was extremely useful in determining complete remission of uterine cervical cancers after several therapies, as well as for early diagnosis of the recurrences in uterine cervical cancer which were not detected by the cytologic smear examinations, compared with the measurement for each tumor marker alone. This discriminant analysis, however, was less effective for early diagnosis of uterine cervical cancer than the cervical cytologic smear test.

Analysis of Variance↗
PubMed1985

Average, binge and maximum alcohol intake in healthy young men: discriminant function analysis.

Blood samples and data on alcohol consumption, diet, exercise, smoking and drug use were collected from 89 undergraduate men affiliated with Greek houses. This population has previously been identified as one that consumed large amounts of alcohol among university undergraduates. Energy and nutrient intake from foods, exercise and drug use (except caffeine) were not different between those reporting consuming more than or less than 1 oz of alcohol per day. Multiple discriminant analysis using a panel of 35 blood tests correctly classified all subjects according to their self-reported intake of more than or less than 2 oz of alcohol per day; 96% were classified correctly at a cut point of 1.5 oz per day. Correct classification was 96-99% using a measure of binge consumption (quantity-frequency of heavy drinking). When maximum number of drinks per occasion was used in the discriminant function, 90-92% were correctly placed. Discriminant analysis based on a panel of common blood tests appears to be a promising technique to identify young drinkers who consume high amounts of alcohol.

Adolescent↗
PubMed1976

[Attempts at a nosological classification with two standardized psychiatric rating scales (author's transl)].

Three different procedures for classification of psychiatric patients, a modified form of Q factor analysis, the classical discriminant analysis and a simple procedure based on clinical concepts, were compared empirically. On the basis of the psychopathological state rated by means of the Inpatient Multidimensional Psychiatric Scale (IMPS) and psychopathological and somatic scales of the AMP documentation system (AMP scale), the three procedures were applied in order to reproduce the psychiatric diagnosis. The classification was made by paired comparison of the following four diagnoses: schizophrenia, paranoid form (n = 45), schizophrenia, unspecified form (n = 47), depressive psychosis (n = 44), and depressive neurosis (n = 53). The procedures were evaluated according to their crossvalidation results. The mean percentages of correct classification were the following: 83% for the modified form of Q factor analysis, 77% for the classical discriminant analysis, 73% for the simple procedure based on clinical concepts using the IMPS, and 79% for the same procedure, but using the AMP scale, all three procedures reaching about the same level of validity.

Bipolar Disorder↗
PubMed1990

A discriminant function analysis of deciduous teeth to determine sex.

Studies of deciduous teeth have concluded that crown size differences in these teeth between males and females are not reliable sex discriminators, in contrast to such differences in permanent teeth. This study measured the mesiodistal and faciolingual crown diameters of all deciduous teeth, as well as those of the permanent first molars, of 162 children from the Burlington Orthodontic Growth Study, conducted earlier in Burlington, Ontario, Canada. All 40 deciduous tooth diameters (20 mesiolingual and 20 faciolingual) were significantly different between the sexes, as were the permanent tooth diameters. Using three to five measurements of deciduous teeth, discriminant analyses of several samplings of these children produced discriminant functions in which 76 to 90% of the holdout samples were correctly classified by sex. Combinations of deciduous and permanent measurements were used to classify 83 to 85% of the holdout samples correctly. When compared with published data on other sample populations, the Burlington group is the most dimorphic for deciduous teeth and is within the range of permanent tooth dimorphisms of other populations. The level of classification accuracy, when using discriminant analysis of the deciduous teeth, can approach the accuracy levels of analysis using the permanent teeth.

Analysis of Variance↗
PubMed2002

Chemical profiling to differentiate geographic growing origins of coffee.

The objective of this research was to demonstrate the feasibility of this method to differentiate the geographical growing regions of coffee beans. Elemental analysis (K, Mg, Ca, Na, Al, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, S, Cd, Pb, and P) of coffee bean samples was performed using ICPAES. There were 160 coffee samples analyzed from the three major coffee-growing regions: Indonesia, East Africa, and Central/South America. A computational evaluation of the data sets was carried out using statistical pattern recognition methods including principal component analysis, discriminant function analysis, and neural network modeling. This paper reports the development of a method combining elemental analysis and classification techniques that may be widely applied to the determination of the geographical origin of foods.

Africa, Eastern↗