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 217 records · Page 12Linked to original sources
PubMed1999

Discriminant analysis and equivalent source localization of the EEG related to cognitive functions.

Discriminant analysis and EEG source localization methods were employed to compare groups of normal subjects during different cognitive conditions using 43-channel EEG recordings in the alpha (8-13 Hz) frequency band. Recordings were obtained from 69 dextral females during 2 passive conditions, Eyes-Open and Eyes-Closed, and 2 active conditions, Word-Finding and Dot-Localization. The cross-spectral matrix between all of the electrode sites was used to characterize the EEGs obtained during each condition. The subjects were partitioned into training and test sets and quadratic discriminant functions were constructed from the training sets to classify the EEGs. The discriminant functions successfully classified both the training and test sets at rates approaching 80%. The classification was repeated using only the diagonal (power spectral) elements of the cross-spectral matrices in the discriminant functions and this approach was successful in discriminating between the EEGs from the passive cognitive conditions but failed to discriminate between the EEGs from the active conditions. Source localization using a modified MUSIC algorithm indicated that the centers of brain electrical activity that distinguished the Eyes-Closed condition from the Eyes-Open condition were located in the medial occipital and right frontal regions. Centers of electrical activity that distinguished the Word-Finding condition from the Dot-Localization condition were located in the right medial posterior and left temporal regions. Validation of the locations of the centers of activity was accomplished by repeating the classification procedures using the spatial patterns generated on the scalp by dipole current sources placed at these locations.

Adult↗
PubMed1991

[Discrimination analysis in the differential analysis of pleural effusions].

The authors applied the mathematical method of discrimination analysis in the differential diagnosis of tumourous and other pleural exudates, as well as for the differentiation of malignant and paramalignant exudates. The basis for the investigation were data obtained during a four-year prospective study of pleural exudates at the Second Clinic of Tuberculosis and Respiratory Diseases in Prague. The analysis was made using a programme elaborated by RNDr. Tomásek, CSc. on a Hewlett-Packard 2100S computer in IHE Prague. After processing of 38 parameters--from 138 patients (without data on the cytology of the exudate, biopsy of the parietal pleura, bronchoscopy and thoracoscopy) the computer selected nine parameters with the greatest discrimination value which made it possible to differentiate tumorous exudates from others in 84%. This concerned the following data: skiagram of chest on admission, CEA in the exudate, temperature, exudate known already before admission to Clinic, LDH in serum on logarithmic scale, general impression of disease during physical examination, red cell sedimentation rate on admission, volume of exudate, alkaline phosphatase in exudate. When it was possible to add to the analyzed parameters also examination of the tissue polypeptide antigen (TPA), the analysis could be performed in 61 subjects. In that case the discriminating capacity rose to 95%. Subsequently we entered in the programme 38 parameters from 88 patients with a tumourous exudate and sought the most suitable ones for differentiation of malignant and paramalignant exudates.(ABSTRACT TRUNCATED AT 250 WORDS)

Adolescent↗
PubMed1991

Discriminant analysis of a sexual-functioning inventory with intrafamilial and extrafamilial child molesters.

Several suggestions have been offered by clinicians who work with sex offenders concerning etiological distinctions between incest offenders and extrafamilial child molesters. In this study we hypothesized that nonincestuous child molesters would have more sexual problems in general than would incestuous offenders. Ninety-five incest offenders and 127 nonincestuous child molesters were administered the Derogatis Sexual Functioning Inventory (DSFI), a self-report instrument. On the global Sexual Functioning Index the total sample of child molesters scored below the first percentile compared to the normative sample. In a direct discriminant analysis of the 10 subscales, satisfaction, fantasy, and experience were the principle variables discriminating between incestuous and nonincestuous offenders. Incest offenders were higher on experience and satisfaction and lower on fantasy. However, the discriminant analysis failed to yield clinically relevant distinctions between the sexual problems of extrafamilial and intrafamilial child molesters. DSFI scores for both groups indicated that their overall level of sexual functioning is poor relative to the general population.

Child↗
PubMed1999

Brain metabolic impairment in non-cerebral and cerebral forms of X-linked adrenoleukodystrophy by proton MRS: identification of metabolic patterns by discriminant analysis.

Cerebral metabolism in six children with X-linked adrenoleukodystrophy (X-ALD) was studied using 1H magnetic resonance spectroscopy (MRS), and the status of the patients was monitored for evaluating disease progression. Spectra were abnormal even in patients with no cerebral impairment. Four different metabolic patterns were identified, and a metabolic classification of the disease was proposed, from grade 0 to grade III. The evolution of the disease toward grade II appears to be systematic, but many patients did not evolve from this grade to grade III, which is the metabolic mark of severe progressive forms. Metabolic data of X-ALD were processed using discriminant analysis, which provides a classification accuracy of 95.2%. Proton cerebral MRS together with discriminant analysis may be useful during the follow-up in X-ALD for monitoring the evolution of the disease and the effects of therapy.

Adolescent↗
PubMed1998

Specific language impairment and grammatical morphology: a discriminant function analysis.

Discriminant function analysis was employed to determine if grammatical morpheme production could be used to classify preschool-age children with specific language impairment (SLI) and children with typically developing language skills. Three variables were included in the discriminant analysis: a finite verb morpheme composite, a noun morpheme composite, and mean length of utterance in morphemes. The children with SLI and age-matched controls were discriminated with high levels of accuracy, though the three variables did not yield identical classifications. Across two samples of typically developing children and children with SLI, the verb morpheme composite showed sensitivity exceeding 85% and specificity of 100%. In light of these results and growing evidence that problems with finite verb morphology continue into the school years in children with SLI, the verb morpheme composite was considered to hold promise as a clinical marker for SLI.

Child↗
PubMed1996

A multiple discriminant analysis of vessel accidents.

A large sample of 936 vessel accident cases occurring between 1979 and 1987 on the lower Mississippi River were cluster analyzed to generate four groups relatively unique in their respective attribute values. The attributes used to cluster the accidents included participation in the U.S. Coast Guard's New Orleans Vessel Traffic Service (NOLA-VTS), type of accident, river stage traffic level, system utilization, accident location, weather conditions, and time of accident. The four-group cluster solution resulted in logical groupings, given the realities of navigating the lower Mississippi River. The four groups resulting from the cluster analysis were characterized as Group 1: Danger Zone (224 cases), 100% NOLA-VTS participants whose accidents occurred primarily on the most dangerous part of the river; Group 2: Bad Conditions for Good Navigators (230 cases), characterized by a high rate of participation and unserious accidents occurring in treacherous navigating conditions; Group 3: Probably Preventable (134 cases), characterized by a low participation rates and serious accidents occurring in not the worst navigating conditions; and Group 4: Accidents That Should Not Have Happened (345 cases), characterized by zero participation and serious accidents occurring in reasonable navigating conditions. Significant marginal participation rates for the marine tracking technology across the four accident clusters (100% for Group 1, 67% for Group 2, 37% for Group 3, and 0% for Group 4) effectively distinguishes between casualty groups. In the subsequent discriminant analysis, three discriminant functions correctly classified 96% of the total accidents, including 100% of Group 1 and Group 4, 90% of Group 2, and 88% of Group 3. The variables contributing most to overall group differentiation were participation in the system, overall system utilization, river stage, traffic level, time and location of accidents. The three discriminant functions were statistically significant, with each individual function accounting for a large relative percentage of the variance between the groups. In order of decreasing discriminating power, the functions could be characterized as (1) System Participation and Utilization, (2) Navigating Conditions, and (3) Time and Place.

Accidents↗
PubMed2003

[Discriminant analysis applied to formants of vowels in Castellano dialect during the phonation with prosthesis and esophageal voice after total laryngectomy].

The objective of the present work consists, in analyzing if there is statistically significant differences between certain quantitative variables of acoustic parameters of the voice, in three samples of individuals, using a discriminative analysis. The study groups are formed by: 20 patients with total laringectomy and esophageal voice, 20 patients with total laringectomy and tracheoesophageal voice with phonatory prosthesis, and 20 normal individuals. We analyzed the energy, bandwidth and frequency of the four first formants of the five Castilian vowels in the context of an established phrase. Based on our results the method of discriminative analysis of the used variables is valid to establish differences between the three studied groups; that is they display their own characteristics from the acoustic point of view. This method allows us to affirm that the tracheoesophageal voice does not come closer to the normal voice that the esophageal voice does.

Discriminant Analysis↗
PubMed1979

[Audiologic diagnosis of sensorineural deafness by discriminant analysis (author's transl)].

The audiometric test-battery surely is a help to the diagnosis of a lesion of the VIIIth nerve. The discriminant analysis is the most suitable statistic procedure for reduction of the test-battery and for clearer decision in cases of ambigous constellation of findings. The persons are divided into three groups: normals, patients with presumable inner-ear disturbance, and those with lesions of the acoustic nerve. By means of the classification-coefficients (a result of the discriminant analysis) in new cases a diagnosis per conclusionem can be found on-line.

Audiometry↗
PubMed2006

A mixture model-based discriminate analysis for identifying ordered transcription factor binding site pairs in gene promoters directly regulated by estrogen receptor-alpha.

MOTIVATION: To detect and select patterns of transcription factor binding sites (TFBSs) which distinguish genes directly regulated by estrogen receptor-alpha (ERalpha), we developed an innovative mixture model-based discriminate analysis for identifying ordered TFBS pairs. RESULTS: Biologically, our proposed new algorithm clearly suggests that TFBSs are not randomly distributed within ERalpha target promoters (P-value < 0.001). The up-regulated targets significantly (P-value < 0.01) possess TFBS pairs, (DBP, MYC), (DBP, MYC/MAX heterodimer), (DBP, USF2) and (DBP, MYOGENIN); and down-regulated ERalpha target genes significantly (P-value < 0.01) possess TFBS pairs, such as (DBP, c-ETS1-68), (DBP, USF2) and (DBP, MYOGENIN). Statistically, our proposed mixture model-based discriminate analysis can simultaneously perform TFBS pattern recognition, TFBS pattern selection, and target class prediction; such integrative power cannot be achieved by current methods. AVAILABILITY: The software is available on request from the authors. CONTACT: lali@iupui.edu SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Algorithms↗
PubMed1997

Evaluation of response to chemotherapy in patients affected with non-small cell lung cancer by means of three tumour markers elaborated by discriminant analysis.

Chemotherapy is the most effective treatment for inoperable patients (70%) affected with non-small cell lung cancer (NSCLC). The early detection of tumour progression is mandative in order to promptly shift these patients towards salvage or supportive therapy. The present authors investigated the clinical value of a panel of tumour markers, elaborated by means of discriminant analysis, as a follow-up indicator for the detection of tumour progression. The serum levels of tissue polypeptide antigen (TPA), CYFRA-21.1, neuron-specific enolase (NSE) and carcino-embryonic antigen (CEA) were determined before chemotherapy and after three cycles of treatment. Discriminant analysis generated a formula (canonic variable) which correctly classified the 87.8% of the 74 subjects (86.1% of the 36 progressive diseases and 89.5% of 38 non-progressive diseases). This approach produces an algorithm able to calculate a progression score in NSCLC patients which can be helpful for following-up care and therapy control of these patients.

Adult↗
PubMed1993

Comparison of discriminant analysis and probabilistic expert system in VCG data classification.

Our previous studies (Valová et al. 1992) have dealt with the possibilities of expert system utilization for electrocardiologic data interpretation. The results obtained in these studies provided evidence that the selected probabilistic expert system is suitable for the solution of VCG data interpretation problems. The aim of this paper was to compare the results obtained by stepwise discriminant analysis with that obtained by a probabilistic expert system. These classification methods were applied to VCG data measured by Frank's lead system. Five groups of patients were investigated: 76 healthy subjects, 36 patients with angina pectoris, 112 patients with old posterior myocardial infarction, 107 patients with old anterior myocardial infarction and 35 patients with old anteroseptal myocardial infarction. The classification was carried out by the leaving-one-out technique. Results of the classification obtained in five groups by a probabilistic expert system are evidently better than those obtained by stepwise discriminant analysis.

Angina Pectoris↗
PubMed2001

Predicting class II MHC/peptide multi-level binding with an iterative stepwise discriminant analysis meta-algorithm.

MOTIVATION: Predicting peptides that bind to both Major Histocompatibility Complex (MHC) molecules and T cell receptors provides crucial information for vaccine development. An agretope is that portion of a peptide that interacts with an MHC molecule. The identification and prediction of agretopes is the first step towards vaccine design. RESULTS: An iterative stepwise discriminant analysis meta-algorithm is utilized to derive a quantitative motif for classifying potential agretopes as high-, moderate- or non-binders for HLA-DR1, a class II MHC molecule. A large molecular online database provides the input for this data-driven algorithm. The model correctly classifies over 85% of the peptides in the database. AVAILABILITY: Stepwise discriminant analysis software is available commercially in SPSS and BMDP statistical software packages. Peptides known to bind MHC molecules can be downloaded from http://wehih.wehi.edu.au/mhcpep/. Peptides known not to bind HLA-DR1 are available from the author upon request. CONTACT: ronna@ucsfresno.edu.

Algorithms↗
PubMed1988

Multivariate analysis of dermatoglyphics of severe mental retardates: an application of the constellation graphical method for discriminant analysis.

We studied the dermatoglyphics of 353 severe mental retardates (excluding those with chromosomal abnormalities and major limb malformations), using multivariate analysis, to determine how early intrauterine factors are related to the etiology of mental retardation. First, dermatoglyphics were compared between 140 individuals with undefined prenatal factors and 700 normal controls. After 6 and 9 dermatoglyphic traits were chosen as discriminative variables for males and females, respectively, the data were subjected separately for each sex to the constellation graphical method for discriminant analysis. The same formula as obtained in the idiopathic group was subsequently applied to data from cases in other etiological categories. When the misclassification rate was 0.03, the rates of correct classification of the male patients into the etiological categories of undefined prenatal, defined prenatal, perinatal, postnatal and unknown (no anamnestic data available) categories were 19.7% (13/66), 20.0% (3/15), 8.8% (5/57), 5.0% (1/20) and 7.7% (2/26), while the correct classification rates of females were 24.3% (18/74), 42.1% (8/19), 18.9% (7/37), 5.1% (1/16) and 13.0% (3/23), respectively. The results suggest that early intrauterine factors such as those producing dermatoglyphic deviations may contribute to the pathogenesis of severe mental retardation not only in patients with undefined prenatal etiological factors but also in those with perinatal factors, especially those of the female sex.

Adolescent↗
PubMed1991

Classification of peripheral and central (pontine infarction) vestibular deficits. Selection of a neuro-otological test battery using discriminant analysis.

The results obtained from a complete neuro-otological test battery were examined statistically in order to select measurement variables which would optimally indicate significant differences between four groups: normal patients, patients with partially compensated unilateral peripheral vestibular deficit, patients with an acoustic neurinoma and patients with central (brainstem) vestibular deficit. A stepwise-discriminant analysis was performed on measurements of slow-phase velocity obtained from each test. The primary measurements selected to assign a subject optimally to one population were the canal paresis (CP) of the caloric test, the eye-tracking gain contralateral to the deficit for a 15 deg/s stimulus, the gain asymmetry for optokinetic nystagmus with a 30 deg/s stimulus, and the level of spontaneous nystagmus. The resulting classifications were 100% correct for normal and central deficit patients. However, the division between peripheral deficit and acoustic neurinoma patients overlapped causing about 30% false classifications of neurinoma patients: some 20% of the peripheral deficit patients were classified as normal. If the CP was not available the discriminant analysis substituted the rotating chair response for 5 deg/s2, in place of CP. This substitution caused a 10 to 20% decrease in classification accuracy.

Diagnosis, Differential↗
PubMed1985

Discriminant analysis in renal histological diagnosis of primary glomerular diseases.

Differences in clinical and laboratory findings between different renal histological lesions were examined in 138 patients with primary glomerular diseases, and discriminant analysis was carried out in 72 patients to determine whether each histological type could be distinguished by the linear combination of these findings. The histological types were classified into 7 groups: minimal change nephrotic syndrome (MCNS); focal glomerular sclerosis (FGS); membranous nephropathy (MN); membranoproliferative GN (MPGN); proliferative GN (PGN); PGN with focal crescents (P X fc); and minor glomerular lesions (MGL). Ten variantes were selected from the clinical and laboratory findings in the early stage of the disease: sex, age of onset, acute onset, oliguria, urine protein, RBC in urinary sediment, serum albumin, serum total cholesterol, serum creatinine, and systolic blood pressure. In the discriminant analysis made regarding all these items collectively as continuous variantes, there was a significant difference (p less than 0.001) in the combination patterns of the variantes among histological types. Therefore, further analysis was performed using canonical axes and a multi-stage discriminant method. The canonical score and data obtained by a multi-stage discriminant method demonstrated that MCNS, MN, MPGN, and the group of PGN, P X fc and MGL could be distinguished from each other well, but that the degree of proliferation or the presence of focal lesions could not be predicted. As a result of these studies, we obtained a discriminant formula with which we could predict, with fairly high accuracy, some histological types on the basis of data on the 10 items mentioned.

Adolescent↗
PubMed1988

Evaluation of some immunological parameters in interstitial lung disease by discriminant analysis.

In 37 subjects affected by interstitial lung diseases (19 patients with pulmonary sarcoidosis, 11 with hypersensitivity pneumonitis, 7 with idiopathic pulmonary fibrosis), we have compared by discriminant analysis (Statistical Package for the Social Sciences, version 8.3) 17 biological parameters derived from bronchoalveolar lavage analysis, gallium-67 scanning and lung biopsy. The aim of the study was to analyze the parameters of these three groups by forming one or more linear combinations of the discriminant variables. In particular, we tried to define the ability of such parameters to define these interstitial lung diseases and the relative importance of the data examined. The functions obtained are highly discriminant, so that the three groups are well distinguished among themselves; it means that the variables employed discriminate among the diseases studied. Among the variables considered, differential cell count, immune complex determination, gallium-67 lung scanning have the most important discriminant capacity. Discriminant analysis emphasizes that the three diseases are mediated by different immune mechanisms and underlines the role of each mechanism in determining the disease.

Adult↗
PubMedDate not supplied

Discriminant analysis in predicting prognosis of Bell's palsy.

A prospective study was carried out to objectively quantificate the most important predictive factors in idiopathic facial paralysis (Bell's palsy). Stepwise discriminant analysis was applied to data prospectively obtained from 570 patients with Bell's palsy treated at the ENT Department of La Paz Hospital between January 1983 and January 1986. Due to statistical requirements the final number of patients included for discriminant analysis was 140. In all, 37 variables were studied in each patient. With data obtained from these variables a linear discriminant function was obtained, with an overall accuracy of 95%. The rate of correct prediction is 95.8% for patients with an expected complete facial recovery, and 90.9% for the group of patients with an expected recovery of less than 100%. The 8 predictor variables selected are: ENoG amplitude, stapedius reflex, familial incidence, Hilger test, maximum degree of the palsy, recurrent facial paralysis, facial pain, and hyperacusis. The discriminant function obtained is an easily adaptable method for routine practice in order to objectively assess the prognosis of patients with Bell's palsy.

Adult↗
PubMed2001

Discriminant analysis of prognostic factors for malignant fibrous histiocytoma in soft tissue.

We prospectively followed 32 patients with soft-tissue malignant fibrous histiocytoma (MFH). Parameters were age; sex; tumor size, location, and depth; operative method; chemotherapy; radiotherapy; and histology. Patients with recurrence or metastases due to MFH within 6 months after the initial operation were separated from those without these characteristics by discriminant analysis with statistically significant difference. The order of influential functions was histology, depth of tumor, operative method, and sex. Male patients with deep-seated storiform-pleomorphic type MFH, receiving less comprehensive surgery, had the greatest risk of local recurrence or early metastases. We have to pay particular attention to patients with these factors and perform adequate surgery, because local recurrence and metastases were found to be closely related, and to have a great influence on the prognosis of this disease. Discriminant analysis to separate patients with MFH recurrence or metastases within 6 months after the initial operation from those without these characteristics is worthwhile for prognostic assessment.

Adult↗