PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Multiple Classification 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 505 records · Page 28Linked to original sources
PubMed2005

Cerebral perfusion abnormality in narcolepsy with cataplexy.

To investigate abnormal cerebral perfusion in narcoleptics with cataplexy, 25 narcoleptics with cataplexy and 25 normal controls were enrolled in this study. Cerebral perfusion was measured by brain single photon emission computed tomography (SPECT) using 99mTc-ethylcysteinate dimer. Patients and normal controls had not received any medication prior to the SPECT scan. Differences in cerebral perfusion between narcoleptics and normal controls were subjected to statistical parametric mapping (SPM) analysis. Overnight polysomnography and multiple sleep latency test (MSLT) were performed in all patients. Brain SPECT was carried out on all patients and normal controls during the waking state. Clinical symptoms and MSLT results of all patients are in accord with the International Classification of Sleep Disorders criteria for narcolepsy. MSLT showed a short mean sleep latency (1.69 +/- 1.0 min) and 2-5 sleep onset REM periods in individual patient. SPM analysis of brain SPECT showed hypoperfusion of the bilateral anterior hypothalami, caudate nuclei, and pulvinar nuclei of thalami, parts of the dorsolateral/ventromedial prefrontal cortices, parahippocampal gyri, and cingulate gyri in narcoleptics [P < 0.05 by Student's t test with false discovery rate (FDR) correction]. Significant hypoperfusion in the white matter of frontal and parietal lobes was also noted in narcoleptics. This study shows reduced cerebral perfusion in subcortical structures and cortical areas in narcoleptics. The distribution of abnormal cerebral perfusion is concordant with the pathway of the cerebral hypocretin system and may explain the characteristic features of narcolepsy, i.e., cataplexy, emotional lability, and attention deficit.

Adolescent↗
PubMed2006

The impact of temporal compression and space selection on SVM analysis of single-subject and multi-subject fMRI data.

In the present study, we compared the effects of temporal compression (averaging across multiple scans) and space selection (i.e. selection of "regions of interest" from the whole brain) on single-subject and multi-subject classification of fMRI data using the support vector machine (SVM). Our aim was to investigate various data transformations that could be applied before training the SVM to retain task discriminatory variance while suppressing irrelevant components of variance. The data were acquired during a blocked experiment design: viewing unpleasant (Class 1), neutral (Class 2) and pleasant pictures (Class 3). In the multi-subject level analysis, we used a "leave-one-subject-out" approach, i.e. in each iteration, we trained the SVM using data from all but one subject and tested its performance in predicting the class label of the this last subject's data. In the single-subject level analysis, we used a "leave-one-block-out" approach, i.e. for each subject, we selected randomly one block per condition to be the test block and trained the SVM using data from the remaining blocks. Our results showed that in a single-subject level both temporal compression and space selection improved the SVM accuracy. However, in a multi-subject level, the temporal compression improved the performance of the SVM, but the space selection had no effect on the classification accuracy.

Adult↗
PubMed1984

Clinical relevance of immunologic phenotype in diffuse large cell lymphoma.

The immunologic phenotypes of 78 diffuse large cell lymphomas were determined by an immunoperoxidase technique using a panel of monoclonal antibodies. The phenotypes were correlated with clinical and morphological parameters by univariate and multivariate analysis. Forty-one lymphomas (53%) expressed immunoglobulin (Ig+). Of the 37 cases that did not express immunoglobulin (Ig-), 9 expressed T cell antigens. Although the T cell phenotypes were antigenically heterogeneous, all cases represented mature T cell phenotypes. The majority of the remaining 28 cases expressed the B cell-associated antigen, B1. At 5 yr, actuarial survival for the Ig- patients was 63%, compared with 15% for the Ig+ patients. A significantly greater proportion of patients with Ig+ lymphomas were over the age of 65 at diagnosis. All of the 9 patients with marrow involvement were Ig+. Multiple factors were analyzed by the Cox regression procedure for their impact on survival, including antigenic profile, histologic grade, morphological classification, and numerous clinical parameters previously recognized to be of prognostic significance. In this analysis, stage, age greater than 65 yr, systemic symptoms, and marrow involvement had the greatest influence on survival. The survival difference between Ig- and Ig+ patients is explained by a higher proportion of Ig+ patients with these unfavorable prognostic factors. With our current immunologic methods, retrospective cell phenotyping analysis has not provided independent prognostic significance in diffuse large cell lymphoma. A prospective evaluation of similarly treated patients is needed to characterize the influence of phenotype fully and to determine its potential usefulness for therapy.

Adolescent↗
PubMed1998

Associations between hemorrhoids and other diagnoses.

PURPOSE: The risk factors and mechanisms that contribute to the occurrence of hemorrhoids are not well understood. The study of the comorbid occurrences of hemorrhoids with other diagnoses in identical patients may point to a common underlying pathophysiology. The present study was undertaken to determine which diagnoses are associated with the occurrence of hemorrhoids. METHODS: A case-control study compared the occurrence of comorbid diseases in case subjects with hemorrhoids with that of control subjects without hemorrhoids. The case population comprised all patients with hemorrhoids (International Classification of Diseases codes 455.0-455.9), who were discharged from hospitals of the U.S. Department of Veterans Affairs between 1986 and 1996. In a multiple logistic regression analysis, the occurrence of hemorrhoids served as outcome variable, and age, gender, ethnicity, and the comorbid occurrence of other diagnoses served as predictor variables. RESULTS: A total of 96,314 individual patients with hemorrhoids and the same number of control subjects were identified. In a chart review of a random sample of 100 cases, the diagnosis of hemorrhoids could be confirmed in 97 percent of all instances checked. The variety of diagnoses associated with hemorrhoids could be broken down into five large categories: 1) diseases associated with diarrhea (odds ratio, 1.30; 95 percent confidence interval, 1.27-1.33); 2) spinal cord injuries (odds ratio, 1.17; 95 percent confidence interval, 1.09-1.26); 3) constipation and related diseases (odds ratio, 1.48; 95 percent confidence interval, 1.43-1.54); 4) various types of anorectal diseases (odds ratio, 4.71; 95 percent confidence interval, 4.44-5.0); and 5) conditions that could be considered manifestations or sequelae of the hemorrhoidal disease itself (odds ratio, 3.41; 95 percent confidence interval, 3.30-3.51). CONCLUSIONS: The types and spectrum of comorbid diagnoses associated with hemorrhoids suggest that an increased tone of the anal sphincter constitutes a common pathophysiologic mechanism for the development of hemorrhoids.

Adult↗
PubMed2001

Predicting days of high allergenic risk during Betula pollination using weather types.

The aim of this study was to build up a picture of the influence of meteorological conditions on pollen and pollinosis, taking account of weather types, pollen concentrations in the air and pollinosis symptoms, with the aim of preventing allergic responses. The study took place in Burgundy from 1996 to 1998, during the pollination of the birch (Betula), which is the most important arborean allergen in this region. We used daily pollen data from four Hirst volumetric traps, identified weather types by Bénichou's classification, and obtained data on the occurrence of rhinitis, conjunctivitis, asthma and coughing from a sample of 100 patients. These data were analysed by multiple-component analysis. The results show that pollen dispersal is favoured by windy conditions, low relative humidity, precipitation below 2 mm and temperatures above 6 degrees C. Such weather also favours pollinosis, but other particular meteorological situations, even if they do not assist pollen dispersal, can act directly on the development of symptoms: a decrease of temperature (3 degrees C) led to the development of rhinitis and conjunctivitis, while strong winds were associated with many cases of conjunctivitis and asthma, owing to the irritant effect of cold or wind; asthma was favoured by temperature inversions with fog, probably because such weather corresponds to high levels of pollution, which act on bronchial hyperreactivity. Because the weather types favouring pollination and pollinosis are predicted by the meteorological office, this can constitute a tool for reducing the effect of high-risk allergenic days.

Asthma↗
PubMed2000

Predicting arm recovery following stroke: value of site of lesion.

OBJECTIVES: The aims of this study were to assess whether the site of lesion is predictive of upper limb recovery after stroke and to determine whether this information adds to the predictive ability of the clinical examination. MATERIAL AND METHODS: Forty-five patients were examined at entry to the study and at 2 and 12 months after stroke. The Brunnström-Fugl-Meyer test was used as outcome measurement. Predictor variables included clinical parameters and classifications of lesion site (obtained by CT/MRI). RESULTS: Correlation analysis revealed small to moderate relationships between lesions of subcortical structures and arm outcome at 2 months. In multiple regression analysis, the best model for predicting recovery at 2 months was found to be a combination of the clinical parameters with a purely subcortical lesion. Motor recovery at 12 months was best predicted by the clinical tests alone. The results further indicated that patients with subcortical damage tended to take longer to recover. CONCLUSIONS: Clinical assessment is most useful for determination of the prognosis of upper limb recovery after stroke. Neuroanatomical parameters measured by CT or MRI can only act as an adjunct.

Adult↗
PubMed2001

Management of lymphoma patients in a cancer unit: an audit of 15 years' experience in a district general hospital.

The objectives were to analyse data obtained in a district general hospital medical oncology unit over a 15-year period, and to provide a comparator for standards of care for lymphoma patients in new cancer units. Prospectively collected data were analysed on 208 consecutive patients presenting with lymphoma, on an 'all-comers' basis, between 1981 and 1996. Treatment was with radiotherapy and/or chemotherapy, according to local protocols. The outcome measure was survival from the date of diagnosis. The 5-year actuarial survival was 72.7% for Hodgkin's disease and 55.7% for non-Hodgkin's lymphoma. Age and stage of disease were the only predictors of survival in a multivariate analysis. Histological classification was not a useful predictor of survival in this analysis. Survival figures comparable with those obtained nationally and across Europe are attainable in a cancer unit. Multiple pathways of referral of lymphoma patients operate in our region.

Adolescent↗
PubMedDate not supplied

Evaluation of gemfibrozil therapy: predictive response from lipoprotein subfraction analysis.

Subjects with high-density lipoprotein cholesterol (HDL-C) values of less than 47 mg/dL (mean 35.6 +/- 5.5 mg/dL) were selected for this study to examine relationships between plasma lipids, lipoprotein components, and the outcome of gemfibrozil therapy. Changes in plasma lipoprotein subfractions were determined to better understand the previously observed variability of the responses in both HDL-C and triglycerides to gemfibrozil. Based on the data collected, an attempt was made to identify pretreatment lipid parameters that may be predictive regarding the efficacy of gemfibrozil therapy. Serum samples were analyzed at the outset and after the conclusion of 12 weeks of gemfibrozil therapy. Because the HDL-C response to therapy was highly variable, the data from patients were separated into two groups, responders (>20% increase in HDL-C) and nonresponders (<20% increase in HDL-C). The lipid components of lipoprotein subfractions were evaluated using multiple regression analysis yielding predictive models that show the relationship between specific lipoprotein subfractions and the percentage change in HDL-C and posttreatment triglyceride levels. Group classification was then predicted with 78% accuracy using specific lipoprotein subfractions to estimate an individual's percentage change in HDL-C. The major difference between the responder and nonresponder groups was their respective correlations between triglyceride-lowering and changes in HDL-C. In the responder group, there was a significant correlation between the changes in HDL-C and the lowering of triglycerides (r = 0.61, p = 0.03), whereas the nonresponder group showed no such correlation (r = 0.17, p = 0.52). The predictive model also proved to be highly accurate in forecasting the effectiveness of the triglyceride-lowering action of gemfibrozil in this group of patients.

Adult↗
PubMed2006

Multivariate information bottleneck.

The information bottleneck (IB) method is an unsupervised model independent data organization technique. Given a joint distribution, p(X, Y), this method constructs a new variable, T, that extracts partitions, or clusters, over the values of X that are informative about Y. Algorithms that are motivated by the IB method have already been applied to text classification, gene expression, neural code, and spectral analysis. Here, we introduce a general principled framework for multivariate extensions of the IB method. This allows us to consider multiple systems of data partitions that are interrelated. Our approach utilizes Bayesian networks for specifying the systems of clusters and which information terms should be maintained. We show that this construction provides insights about bottleneck variations and enables us to characterize the solutions of these variations. We also present four different algorithmic approaches that allow us to construct solutions in practice and apply them to several real-world problems.

Algorithms↗
PubMed1989

Somatosensory evoked response and recovery from stroke.

The purpose of this study was to evaluate median nerve short latency somatosensory evoked potentials (SEP) as prognostic indicators of functional outcome after right cerebral infarction. Twenty-six patients with right cerebral infarction were admitted to a rehabilitation unit and were classified into three groups based on the evoked potentials measured on admission. The SEP classification and nine other variables reported to predict recovery after stroke were evaluated for prediction of the final outcome measure, the Barthel Scale, using stepwise multiple regression analysis. Patients with normal SEP achieved a mean Barthel score of 95 +/- 3.9 SD, while patients with an amplitude asymmetry on SEP showed a mean Barthel score of 77.5 +/- 17.5 SD. The group with absence of cortical potentials had a mean Barthel score of 59.5 +/- 21.3 SD. Six predictors (Barthel admission score, SEP, electroencephalogram, Brunnstrom staging of the arm at the time of admission, joint position sense, and hemihypesthesia) were highly correlated with the Barthel outcome score (p less than 0.05 for all). The linear regression equation with three predictors (Barthel admission score, SEP, and homonymous hemianopsia) provided the best prediction of Barthel outcome score (p = 0.005). These data indicate that the Barthel admission score is the best predictor of functional level after stroke rehabilitation. Knowledge of the median SEP and homonymous hemianopsia improve this prediction.

Adult↗
PubMed1996

Spirometry of healthy adult South African men. Part II. Interrelationship between socio-environmental factors and 'race' as determinants of spirometry.

AIM: To examine the interaction of socio-environmental background, 'race' and anthropometry on spirometry. METHODS: A study population of white and black bank workers with broadly comparable current socio-economic circumstances was identified. Detailed questionnaires regarding a number of indicators of socio-environmental status (SES) were administered in addition to anthropometric and spirometric measurements. Multiple linear regression analyses were performed. The Kappa statistic was used to determine the best predictors of SES. RESULTS: 'Race' was found to be the best predictor of SES. Within the black group a sub-classification of high and low SES was possible. In all respects the white SES indicators were higher than those of the black high SES group. Multiple regression analysis showed that sitting height was a better anthropometric predictor of spirometry than standing height. Replacing 'race' with other indicators of SES gave similar measures of prediction (by Mallow's CP statistic) of spirometry in the over-30-year age group. CONCLUSION: The impact of 'race' on spirometry can be reduced substantially by incorporating sitting height and indicators of SES in the regression equations. This observation lends support to the hypothesis that 'race' is not a direct genetic predictor of lung size, but may have an influence because of SES and anthropometric features associated with 'race'.

Adult↗
PubMed1993

Metacarpophalangeal pattern profile analysis in Williams syndrome.

Many patients with Williams syndrome (WS) are not diagnosed until they are old enough to demonstrate the characteristic personality and facial changes. A number of these changes are quite subtle and none of them is present in all affected individuals. The cause of WS remains obscure and consequently, there are no cytogenetic, biochemical, or molecular studies to help in the diagnosis of patients in whom the diagnosis is uncertain. We have generated a mean WS metacarpophalangeal pattern profile (MCPP) on 21 clinically diagnosed individuals with WS. This mean syndrome profile shows that WS hands are smaller than average age-matched control hands and that the distal phalanx of the thumb is disproportionately large with respect to the rest of the hand. A mathematical model, which effectively discriminates WS patients from unaffected control individuals, was developed using discriminant analysis of the MCPP data. Of the 21 WS patients classified by this method, only 2 were misclassified as "normal." Similarly, 2 of the 24 control individuals were misclassified as "WS," yielding an over-all successful classification rate of 91%.

Abnormalities, Multiple↗
PubMed2025

Integrative multi-omics analysis proposes a metabolic classification of gliomas: distinct metabolic states, immune infiltration, and prognosis.

BACKGROUND: The tumor microenvironment (TME) of glioma harbors diverse cell types; however, cell metabolic heterogeneity remains to be explored. This study aims to characterize the metabolic features of different cell types in the TME by integrating multiple datasets, including genomics, bulk and single-cell transcriptomics, and metabolomics. METHODS: Unsupervised machine learning was used to construct an energy metabolic classifier based on the metabolic pathways identified from bulk RNA-seq of gliomas in the TCGA dataset. The classifier was externally validated using multiple datasets, including genomics, bulk RNA-seq, snRNA-seq, and the metabolomics data. Furthermore, metabolic heterogeneity associated with the classifier was further characterized at single-cell resolution. RESULTS: The energy metabolism-based classifier stratified patients into two prognostic clusters: patients in cluster 1 were characterized by high pathway activity of glycolysis, the pentose phosphate pathway (PPP), and fatty acid oxidation (FAO), whereas patients in cluster 2 exhibited higher activity in glutaminolysis. This metabolic classifier revealed both intratumoral and intertumoral metabolic heterogeneity, and the complexity was further validated by the metabolomics profiling and snRNA-seq data from the CPTAC dataset. Notably, OSMR, highly expressed in cluster 1, showed significant co-expression with key glycolytic enzyme genes. The OSM/OSMR/JAK1/STAT3 axis potently drives malignant progression of glioma cells, specially enhancing their invasive and migratory capabilities. Single-cell resolution analyses demonstrated that tumor metabolic heterogeneity is primarily driven by malignant cells rather than non-malignant components, while tumor microenvironment (TME) factors were also found to modulate malignant cell metabolism. Significantly, glycolytic activity in glioma cells increased during the phenotypic transition from PN (proneural) to MES (mesenchymal), with cluster 1 metabolic phenotypes predominating in the tumor core. Compared to cluster 2, cluster 1 patients exhibited higher mRNA expression of immunosuppressive checkpoint genes, which correlated with pronounced immunosuppression in the TME. Furthermore, various immune cells demonstrated distinct metabolic preferences at single-cell resolution. CONCLUSIONS: This study developed an energy metabolic-based classifier for gliomas with prognostic and therapeutic potential. Metabolic reprogramming was linked with the PN-to-MES transition of glioma cells and immunosuppression in the tumor microenvironment. Multi-omics data, especially snRNA-seq, offered insights into metabolism heterogeneity at single-cell resolution, enabling personalized treatment strategies.

Humans↗
PubMed2002

Analysis of multivariate frequency data by graphical models and generalizations of the multidimensional row-column association model.

Models used to analyze cross-classifications of counts from psychological experiments must represent associations between multiple discrete variables and take into account attributes of stimuli, experimental conditions, or characteristics of subjects. The models must also lend themselves to psychological interpretations about underlying structures mediating the relationship between stimuli and responses. To meet these needs, the author extends the graphical latent variable models for nominal and/or ordinal data proposed by C. J. Anderson and J. K. Vermunt (2000) to situations in which dependencies between observed variables are not fully accounted for by the latent variables. The graphical models provide a unified framework for studying multivariate associations that include log-linear models and log-multiplicative association models as special cases.

Humans↗
PubMed1980

Hierarchic classification of multiple types of urothelial cells by computer.

A two-level hierarchic classification of 915 urothelial cell images from the urinary sediment was performed by computer using the TICAS programs. The purpose of the analysis was to determine whether adequate discrimination could be obtained among several classes of cells, such as degenerated (DG), multinucleated (MN) and mononucleated well-preserved (WP) ones. The first level in the classification hierarchy showed that the three classes of cells could be identified by computer and that the identification of the group of WP cells was particularly satisfactory, the analysis of WP cells, using training and object sets, documented once again that the identification of diagnostically significant subgroups could be achieved with a very small misclassification error, which was less than 1% for benign (NEG) and malignant (POS) cells. In view of the prior successful application of the classification of images of WP cells to establish patient profiles, the results of the hierarchic classification support the concept of automated analysis of cells in the urinary sediment by computer.

Computers↗
PubMed2001

Predicting "charge outliers" after spinal cord injury: a multicenter analysis of demographics, injury characteristics, outcomes, and rehabilitation charges.

OBJECTIVE: To describe the distribution of charges, to distinguish between "charge outliers" and nonoutliers, and to identify a model that uses demographics and injury characteristics to predict charge outlier status in individuals with spinal cord injury (SCI). DESIGN: Retrospective data analysis of patients admitted to 24 acute inpatient rehabilitation national Spinal Cord Injury Model Systems centers. Statistical analysis, including proportions, means, and standard deviations (SDs), were compiled for the following variables: demographic and injury information, rehabilitation charges, medical complications, associated injuries, and surgical procedures. SETTING: Tertiary, university medical centers participating in the National Institute on Disability and Rehabilitation Research's SCI Model Systems project. PARTICIPANTS: A total of 13,392 patients who were admitted to 24 acute, intensive, interdisciplinary rehabilitation settings after traumatic SCI between November 1972 and August 1996. MAIN OUTCOME MEASURES: Statistical data analysis was used to determine significance between charge outliers and nonoutliers on the basis of demographic, injury characteristics, and clinical factors. Outliers, under the current diagnostic-related group system, are defined as cases in which lengths of stay exceed the mean by the lesser of 20 days or 1.94 SDs. RESULTS: Statistically significant differences were found between SCI charge outliers and nonoutliers based on ethnicity, education, employment, level of injury, American Spinal Injury Association impairment classification, and sponsor of hospitalization. On average, outliers were 4 years older than nonoutliers, and tended to have more associated injuries, pressure ulcers, surgical procedures, and medical complications. A forward-conditional stepwise multiple logistic regression analysis was used to confirm univariate analysis and to predict the presence or absence of outliers based on the predictor variables. A model for the prediction of SCI charge outlier status was defined. CONCLUSIONS: SCI charge outliers are most likely to be retired, insured, have high cervical level injuries, and be educated beyond high school. Improved treatment efficiency serves as a means of cost reduction and is a reason to identify outlier characteristics.

Adolescent↗
PubMed2004

Cross-sectional analysis of different variables of patients with non-insulin dependent diabetes and their periodontal status.

The periodontal condition of 72 non-insulin dependent diabetes patients was compared with that of 92 nondiabetic individuals. Plaque Index (PII), Calculus Index (CI), Gingival Index (GI), and attachment loss (AL) were measured on four surfaces of six teeth in each subject. All four parameters were significantly higher in the diabetic group. No significant difference in the frequency of toothbrushing was found between the groups. For all age groups, GI and AL were higher in the diabetic group. In each group, GI was not changed with age, while AL increased with age. Classification of the groups based on PII showed that the diabetic group's GI was higher than the nondiabetic group for low, medium, or high PII values. The diabetic group showed higher AL for only the medium and high PII groups. Classification by CI revealed that the diabetic group's GI and AL were significantly higher than those of the nondiabetic group for subjects with low, medium, or high values of CI. Multiple regression analysis revealed that the main factor affecting GI was the presence or absence of diabetes. PII and CI both showed a significant relationship with GI; age was the second most significant factor. The most significant factors influencing AL were CI and the presence or absence of diabetes; age was the second most significant factor. Patients who had had diabetes for more than 10 years had a higher AL than those who had suffered from diabetes for less than 10 years. Patients with average HbA1c values > or = 10% had more serious mean GI values than those with HbA1c values < 10%. In patients with diabetes, age, plaque accumulation, and calculus formation have more detrimental effects on the periodontal apparatus than in healthy individuals.

Age Factors↗