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 703 records · Page 39Linked to original sources

PubMed2003

Statistical method for analysis of the disease curve in animals with experimental autoimmune encephalomyelitis.

Experimental autoimmune encephalomyelitis (EAE) is a procedure used in the laboratory to examine drugs that may have utility in treating multiple sclerosis (MS). The problem of modeling the disease curve in animals with EAE is studied. The classification of animals after each experiment is considered and the chi-square test is proposed to test a homogeneity between treatment groups. A mixture type of nonlinear mixed-effects model with repeated measurements is considered, assuming that the onset and/or remission of disease is the fixed effect as well as the random effect. Statistical inference on the parameters of the disease curves is discussed. The proposed model is shown to be efficient for comparing the disease curves with different treatments. Examples concerning the study of the effects of test compounds in EAE are presented to illustrate the proposed model and statistical methodologies.

Animals↗

PubMed2004

Recursive partitioning analysis classifications I and II: applicability evaluated in a randomized trial for resected single brain metastases.

PURPOSE: Radiation Therapy Oncology Group (RTOG) recursive partitioning analysis (RPA) prognostic classes I and II for patients with brain metastases is derived from a database made up primarily of patients with unresected and multiple metastases. An analysis of a previously published randomized trial was performed to determine the applicability of these RPA prognostic classes in the setting of resection of single metastases to the brain. PATIENTS AND METHODS: Ninety-five patients with single metastases to the brain that were treated with complete surgical resection entered this study. Patients were randomly assigned to treatment with postoperative whole brain radiotherapy (WBRT) (n = 49 patients) or no further brain treatment (n = 46 patients). All patients entered on this study had a Karnofsky performance status of > or =70. Therefore, although the RTOG RPA has 3 classes, only patients with RPA classes I (n = 26) or II (n = 69) were eligible for this study analysis. RESULTS: For RPA class I, the median survival was 10.9 months versus 9.8 months for class II patients (P = 0.45). Multivariate analysis showed that only postoperative WBRT, independent of RPA class I or II, lessened the risk of brain tumor recurrence (P < 0.0001). CONCLUSION: This analysis of a randomized trial evaluating postoperative WBRT in the treatment of single metastases to the brain showed no difference in survival between RPA class I or II patients. In addition, the use of postoperative WBRT after complete surgical resection of single brain metastases results in substantially better control of disease in the brain independent of RPA classes I or II.

Adult↗

PubMed2005

Bayesian model averaging of Bayesian network classifiers over multiple node-orders: application to sparse datasets.

Bayesian model averaging (BMA) can resolve the overfitting problem by explicitly incorporating the model uncertainty into the analysis procedure. Hence, it can be used to improve the generalization performance of Bayesian network classifiers. Until now, BMA of Bayesian network classifiers has only been performed in some restricted forms, e.g., the model is averaged given a single node-order, because of its heavy computational burden. However, it can be hard to obtain a good node-order when the available training dataset is sparse. To alleviate this problem, we propose BMA of Bayesian network classifiers over several distinct node-orders obtained using the Markov chain Monte Carlo sampling technique. The proposed method was examined using two synthetic problems and four real-life datasets. First, we show that the proposed method is especially effective when the given dataset is very sparse. The classification accuracy of averaging over multiple node-orders was higher in most cases than that achieved using a single node-order in our experiments. We also present experimental results for test datasets with unobserved variables, where the quality of the averaged node-order is more important. Through these experiments, we show that the difference in classification performance between the cases of multiple node-orders and single node-order is related to the level of noise, confirming the relative benefit of averaging over multiple node-orders for incomplete data. We conclude that BMA of Bayesian network classifiers over multiple node-orders has an apparent advantage when the given dataset is sparse and noisy, despite the method's heavy computational cost.

Algorithms↗

PubMed1978

Are there two logistic regressions for retrospective studies?

A comparison is made between two different approaches to the linear logistic regression analysis of retrospective study data: the prospective model wherein the dependent variable is a dichotomous indicator of case/control status; and the retrospective model wherein the dependent variable is a binary or polychotomous classification of exposure. The two models yield increasingly similar estimates of the relative risk with increasing degrees of covariate adjustment. When the covariate effects are saturated with parameters, the relative risk estimates are identical. The prospective model is generally to be preferred for studies involving multiple quantitative risk factors.

Child↗

PubMed1991

[Computer-assisted data analysis of injuries of the skeletal system in polytrauma patients].

The data of 366 patients with multiple injuries were evaluated by application of data processing over a period of seven years. Age and sex distribution, location and combination of injuries, mortality, mechanism of accident, diagnostics as well as therapy were analysed. 91.3% of our patients had injuries of the skeleton. The male to female ratio was 2:1. In 55.4% the age of the patients ranged from 10 to 39 years. The mortality rate of all patients was 27.2%. In the third decade only 16.9% of the patients died. Whereas the mortality of the patients in the 1. decade as well as between the 7. and 9. decades of life was very high (33% and 50%). Most of the skeletal injuries were located in the shank (24% open fractures). The mortality rate increases with the number of injuries (Table III). Patients with concomitant injuries, such as head injuries and fractures of the extremities were the most common combination of multiple injuries. 20% of the patients in this group died. Patients with combined injuries of the head, chest, abdomen and skeleton had highest mortality (57%). X-ray examination of the skull, chest, extremities and mainly of the spine and pelvis should be subjects of routine diagnostics. In patients who were suspected of having a blunt abdominal trauma, the first diagnostic technique will be ultrasound. On account of the various number of concomitant injuries as well as the patients individuality, the estimation of prognosis by systems for the classification of the severity of injury is fraught with difficulties. The analysis of the evaluated data showed that previously small number of variables may predict the prognosis of the course of disease.

Adolescent↗

PubMed2000

Night-time sleep and daytime sleepiness in narcolepsy.

This report describes night-time sleep and daytime sleepiness in a large (N=530) sample of patients meeting the International Classification of Sleep Disorders criteria for diagnosis of narcolepsy. Sleep data were obtained from polysomnographic recordings on two consecutive nights. Sleepiness was assessed using the Multiple Sleep Latency Test, the Maintenance of Wakefulness Test and the Epworth Sleepiness Scale. Analysis revealed that sleep was mild to moderately disturbed on both recording nights. A first-night effect was suggested by decreased REM latency and increased percentage REM and slow-wave sleep on the second night. Sleepiness and sleep disturbance varied across patient subgroups created based on patient ethnicity and on the presence/absence of cataplexy, sleep apnoea, and periodic limb movements. Covariation of sleep and sleepiness measures across patients was significant but weak. Strong association was found between subgroup means of sleep and sleep disturbance measures. The findings reported here show that sleepiness and sleep disturbance vary across patient subgroups and that sleep disturbance is related to, although unable to account, for the pathological sleepiness of narcolepsy.

Adolescent↗

PubMed2005

An expression index for Affymetrix GeneChips based on the generalized logarithm.

MOTIVATION: Affymetrix GeneChip high-density oligonucleotide arrays interrogate a single transcript using multiple short 25mer probes. Usually, a necessary step in the analysis of experiments using these GeneChips is to summarize each of these probe sets into a single expression index that can then be used for determining differential expression, for classification, for clustering, and for other analyses. In this paper, we propose a new expression index that is competitive with the best existing methods, and superior in many cases. We call this expression index method GLA, for GLog Average, since after normalization at the probe level, we take the mean generalized logarithm of perfect match probes. RESULTS: In this paper, we use Affycomp as the primary tool to assess the weaknesses and strengths of GLA. Comparisons are made between GLA and most widely used summary methods (RMA, MAS5.0 and MBEI) in great detail. The substantial reduction in variability and increased ability to detect differential expression, together with the simplicity of implementation, make GLA a plausible candidate for analysis of Affymetrix GeneChip data.

Algorithms↗

PubMed1995

Tumor angiogenesis as a prognostic assay for invasive ductal breast carcinoma.

BACKGROUND: Tumor angiogenesis, as assayed by microvessel density, has been proposed as an independent prognostic marker for clinical outcome in breast cancer patients. PURPOSE: The present study evaluated the microvessel density assay and assessed its utility, alone and together with the evaluation of other tumor characteristics, in predicting outcome in patients with invasive ductal carcinomas. METHODS: In a blinded design, cases of invasive ductal carcinoma were selected from a registry containing the records of and tumor specimens from 386 breast cancer patients treated at the Massachusetts General Hospital from 1977 through 1982. After the exclusion of ineligible patients and inadequate specimens, 220 patients were included in the study; their median time of follow-up was 11.5 years. Half of these patients (n = 110) were positive for axillary lymph node metastases. Histologic sections of the tumors were stained immunocytochemically for factor VIII, a coagulation protein expressed by blood vessel endothelium, and for p53 protein. Independently, two analysts counted microvessels in three microscope fields selected from separate vascular regions of the tumor. Variability in microvessel scores between analysts and among different fields of the same tumor was summarized by the coefficient of variation. The kappa statistic tested for agreement between the analysts while correcting for chance agreement. The effects of tumor characteristics on metastasis-free survival and overall survival were tested univariately by the Harrington-Fleming rank test procedure. The effect of multiple factors on survival was tested under a Cox multivariate proportional hazards model. RESULTS: Microvessel count showed considerable variability between the two analysts and among regions within each tumor, with an overall concordance for tumor classification of 73%. Univariate analysis revealed no association between microvessel count and any other tumor or patient characteristic. Multivariate analysis indicated, for these patients, that nodal status and p53 staining predicted metastasis-free survival and that nodal status, estrogen receptor (ER) status and tumor grade predicted overall survival. CONCLUSIONS: Microvessel count showed much variation among different regions of each tumor. It did not predict metastasis-free survival or overall survival. Nodal status was the most powerful criterion to stratify these patients with invasive ductal carcinoma of the breast into different survival groups. Only ER status, tumor grade, and p53 staining had additional prognostic utility for these patients after they had been stratified by nodal status.

Analysis of Variance↗

PubMed1981

Immunology of archaebacteria that produce methane gas.

The antigenic map of 17 methanogenic bacteria representing the entire range of available species was determined by multiple assay with antibody probes. Four major clusters of antigenically related strains coincide with the females proposed on the basis of 16S ribosomal RNA analysis. Immunological mapping uncovered relationships not yet shown by other methods and allowed identification and classification of two new bacterial isolates.

Antigens, Bacterial↗

PubMed1985

Prognostic factors in non-Hodgkin's lymphomas.

Prognostic factors were investigated in 67 patients with non-Hodgkin's lymphomas, homogeneously staged and treated (COP or CHOP according to low or high malignant histotype). A large number of parameters were scrutinized in order to recognize those exhibiting a prognostic value regarding length of survival. All the parameters that singly appeared to influence survival were entered into a multiple regression factor analysis. The erythrocyte sedimentation rate (ESR), higher or lower than 35 mm at the 1st h, better discriminated the groups of patients surviving or not at a given time. The histologic type, according to the Kiel classification of malignancy, was the second best prognosticator when a short-term prediction was requested (survival or death after no more than 2.5 years), but showed insufficient statistical weight for predicting longer survivals (greater than 4 years). Stage seemed to be the third best prognosticator for the first years of survival, but only the second best for longer survivals. Other parameters had very low prognostic importance when compared with those above. The results were substantially confirmed by 28 other patients, taken as controls. The importance that such a simple and easy test as ESR may be adequate with regard to prognosis is emphasized.

Adult↗

PubMed1991

Dental fluorosis and fluoride exposure in Western Australia.

Dental fluorosis in children is reported from many locations, and its prevalence may be increasing. This study aimed to measure fluorosis in 12-year-olds in fluoridated and non-fluoridated areas of Western Australia and to relate this to exposure. School dental clinics in Perth (F- 0.8 mg/L) and the Bunbury area (F- less than 0.2 mg/L) were the sampling units. Parents provided data on residence in fluoridated areas and on use of supplements and toothpaste. Exposure to water and supplemental fluoride between birth and four years was calculated. Clinical examinations (upper left central incisor) based on the TF classifications of fluorosis (which requires teeth to be dried) were conducted for 338 children in Perth and 321 in the Bunbury region. Re-examinations (n = 50) gave a weighted kappa of 0.78. Contingency analysis estimated bivariate relationships, and multiple logistic regression estimated odds ratios (OR) for risk factors. Fluorosis prevalence in the Perth region was 0.40, and in the Bunbury region 0.33 (chi 2 = 3.69, df = 1, p = 0.055). Prevalence was 0.44 in children with fluoride exposure equivalent to optimal water fluoridation and 0.20 among those with the lowest exposure (chi 2 = 35.99, df = 1, p = 0.0001). Increasing exposure was associated with higher fluorosis prevalence and higher TF scores, but overall, 27.3% of participants had TF score 1 (barely discernible), and only 9.4% had TF score greater than or equal to 2.(ABSTRACT TRUNCATED AT 250 WORDS)

Analysis of Variance↗

PubMed2005

Multimodality computerized diagnosis of breast lesions using mammography and sonography.

RATIONALE AND OBJECTIVES: The purpose of this study is to investigate the use of computer-extracted features of lesions imaged by means of two modalities, mammography and breast ultrasound, in the computerized classification of breast lesions. MATERIAL AND METHODS: We performed computerized analysis on a database of 97 patients with a total of 100 lesions (40 malignant, 40 benign solid, and 20 cystic lesions). Mammograms and ultrasound images were available for these breast lesions. There was an average of three mammographic images and two ultrasound images per lesion. Based on seed points indicated by a radiologist, the computer automatically segmented lesions from the parenchymal background and automatically extracted a set of characteristic features for each lesion. For each feature, its value averaged over all images pertaining to a given lesion was input to a Bayesian neural network for classification. We also investigated different approaches to combine image-based features into this by-lesion analysis. In that analysis, mean, maximum, and minimum feature values were considered for all images representing a lesion. We considered performance by using a leave-one-lesion-out approach, based on image features from mammography alone (two to five features), ultrasound alone (three to four features), and a combination of features from both modalities (three to five features total). RESULTS: For the classification task of distinguishing cancer from other abnormalities in a lesion-based analysis by using a single modality, areas under the receiver operating characteristic curves (A(z) values) increased significantly when the computer selected the manner (mean, minimum, or maximum) in which image-based features were combined into lesion-based features. The highest performance was found for lesion-based analysis and automated feature selection from mean, maximum, and minimum values of features from both modalities (resulting in a total of four features being used). That A(z) value for the task of distinguishing cancer was 0.92, showing a statistically significant increase over that achieved with features from either mammography or ultrasound alone. CONCLUSION: Computerized classification of cancer significantly improved when lesion features from both modalities were combined. Classification performance depended on specific methods for combining features from multiple images per lesion. These results are encouraging and warrant further exploration of computerized methods for multimodality imaging.

Breast Neoplasms↗

PubMed2003

An analysis of the risk of B-lymphocyte malignancies in industrial cohorts.

Among numerous studies of occupational groups with varied chemical exposures (e.g., farmers, petroleum workers, and rubber workers), some have reported excess risk for non-Hodgkin's lymphoma (NHL), multiple myeloma, and other cancers of the B-lymphocyte cell line. While not conclusive, these studies raise questions about the effects of chemical exposures on the lymphocytic versus myeloid cell lines. Almost 70 occupational cohort studies were identified that addressed B-cell cancer risks in 9 major industrial categories, in order to look for common patterns across industries. This effort was substantially limited by the inconsistent nature of lymphohematopoietic (LH) classification schemes across studies and over time, and the relative paucity of B-cell-specific results in studies for any given industry. Taking these limitations into consideration, a descriptive, graphical analysis suggested a pattern of B-cell cancer elevations in the rubber and "general chemical" industries, but no consistent patterns in petroleum production/distribution or petrochemical production. The limited data sources, which lack detail about differences in hazard and exposure for different types of products/chemicals, did not allow a comprehensive look at possible common exposures associated with B-cell cancer elevations across industries. This study suggests that evaluation of possible associations between specific chemical exposures and B-cell malignancies would require additional studies with clear and common definitions of B-cell outcomes. The article concludes by giving an example of a possible common framework for categorizing NHL, the diseases for which most classification issues arise.

Butadienes↗

PubMed2003

The first report from the patient outcomes registry for transplant effects on life (PORTEL): differences in side-effects and quality of life by organ type, time since transplant and immunosuppressive regimens.

BACKGROUND: Post-transplant patient quality of life (QOL) is affected by a number of different factors. A nationwide patient registry has been established to evaluate QOL and determine the effects of transplant and immunosuppressive regimens on patient outcomes. METHODS: Patients were contacted directly at national meetings, through transplant centers, and patient support groups and invited to participate in the registry. All transplant patients aged 16 and over were eligible to enroll. Patients completed a 100-item self-administered questionnaire consisting of questions about patient demographics, organ functioning, and other post-transplant outcomes. General QOL was measured by the Short form - 12 (SF-12). The Memphis Survey, an instrument developed and psychometrically validated at the University of Tennessee, was administered to patients to evaluate side-effects associated with immunosuppression. Data were analyzed from the first 722 patients who entered the registry. Side-effect profile and QOL outcomes were evaluated by organ type, time since transplant and immunosuppressive regimen. Multiple regression analyses were conducted to determine predictors of post-transplant QOL. RESULTS: When outcomes were analyzed by organ type, there were no differences in SF-12 or total weighted Memphis scores. Analysis by time since transplant demonstrated that side-effects in the mobility domain increased with patient age and time since transplant. Analysis by immunosuppressive regimen focused on cyclosporine and tacrolimus-based regimens congruent with similar classifications reported in previous studies (Pirsch JD et al. Transplantation 1997: 63: 977, Shield CF III et al. Transplantation 1997: 64: 1738). When analyzed by regimen, there were no differences between the groups in terms of patients reporting good to excellent organ function, treatment for rejection, infection, and over-immunosuppression. Statistically significant differences were observed when side-effect profile was analyzed by immunosuppressive regimen. Patients on cyclosporine-based regimens reported greater overall side-effect severity and more problems with mobility and life roles. Cyclosporine patients also reported more problems in the miscellaneous subscale, including high blood pressure, enlarged gums and hair growth, but less trouble with trembling hands. Multiple stepwise regression models identified several side-effect subscales as having profound effects on mental and physical QOL. CONCLUSION: Transplant recipients report good to excellent levels of QOL, however, side-effects associated with immunosuppressive regimens impair post-transplant QOL. Problems in certain domains, such as mobility, are found to increase with time since transplant. Tacrolimus-based regimens are associated with fewer and less severe side-effects than cyclosporine-based regimens in key domains that affect post-transplant QOL.

Activities of Daily Living↗

PubMed1993

Gulf war wounds: application of the Red Cross wound classification.

The Red Cross wound classification was applied to 63 casualties requiring surgery in the recent Gulf war. The majority of wounds affect only soft tissue, caused predominantly by antipersonnel fragments. Bullet wounds were fewer but tended to be more severe, often involving a fracture or vascular damage. We recommend minor modification to the classification to include scoring of significant neurological injury. Further, we feel that by recording the distribution of all wounds as well as scoring the casualty's two worst injuries, the incidence and pattern of multiple wounds are ascertained, which is useful in military surgical research. We believe that the Red Cross wound classification is valuable in assessing a wound as part of a secondary survey, but that this wound score has little part to play in triage. It may help to decide management of individual wounds in clinical practice and is useful for recording the nature of wounds for future analysis.

Evaluation Studies as Topic↗

PubMed2000

Single and multiple dose pharmacokinetics of nelfinavir and CYP2C19 activity in human immunodeficiency virus-infected patients with chronic liver disease.

AIMS: To evaluate the single-dose and multiple-dose pharmacokinetics of nelfinavir and its active M8 metabolite in eight HIV-seropositive patients with liver disease, and to examine the relationship between CYP2C19 activity (genotype and plasma M8/nelfinavir metabolic ratio) and the severity of liver disease in these patients. METHODS: Nelfinavir was given as a single dose (500 or 750 mg) to patients beginning therapy and twice (500, 750 or 1000 mg) or three times (250 or 750 mg) daily during chronic therapy. Single-dose pharmacokinetic values were used to predict multiple-dose regimens. Peak and total plasma exposures between 2-4 microg ml-1 and 45-75 microg ml-1 h, respectively, and predose levels > 0.7 microg ml-1 were targeted for multidose nelfinavir. Genotype was determined by analysis for CYP2C19*1, CYP2C19*2, and CYP2C19*3. Individuals were grouped according to their genotype, molar M8/nelfinavir AUC ratio (low: < 0.1, intermediate: 0.1-0.3, high > 0.3), and Child-Pugh classification for severity of liver disease. RESULTS: Nelfinavir pharmacokinetics were characterized by wide interindividual variability, low clearance (181-496 ml min-1 70 kg-1, n = 7), and prolonged half-life (5-20 h, n = 7). M8/nelfinavir AUC ratio increased 58% (n = 4) and alpha 1-acid glycoprotein levels decreased up to 39% (n = 5) from single to multiple dosing. CYP2C19 activity was low (metabolic AUC ratio < 0.1) in four patients with moderate to severe liver disease even though they were genetically extensive CYP2C19 metabolizers (*1/*1 or *1/*2). Three patients required lower daily doses than the standard regimen of 750 mg every 8 h to achieve target concentrations and maintain virologic suppression at < 50 RNA copies ml-1 (up to 20 months). CONCLUSIONS: Acquired CYP2C19 deficiency from moderate or severe liver disease resulted in decreased M8 formation. Long-term HIV suppression is possible using low nelfinavir doses in patients with liver disease.

Adult↗

PubMed1992

[Judgement of prognosis in multiple trauma of patients with blunt abdominal injuries based on different useful trauma scores].

The use of trauma indices for characterisation of polytraumatized patients concerning prognoses is being discussed for nearly 30 years. In the seventies a couple of trauma indices were developed. Some of them are in use all over the world. In the Department for Trauma surgery in the General Hospital of Salzburg three of the common trauma scores were analysed and compared with one another: Injury Severity Score (Baker), Polytraumascore of the University Clinic of Hannover, Trauma index (Schreinlechner and Eber). As a rule most studies presently published are based on the Injury Severity Score. In a retrospective analysis of 62 polytraumatized patients we could show that the Trauma index of Schreinlechner and Eber is more practicable for classification of polytrauma because it shows a good correlation of score, fatality and prognoses.

Abdominal Injuries↗

PubMed2006

Mammographic masses characterization based on localized texture and dataset fractal analysis using linear, neural and support vector machine classifiers.

OBJECTIVE: Localized texture analysis of breast tissue on mammograms is an issue of major importance in mass characterization. However, in contrast to other mammographic diagnostic approaches, it has not been investigated in depth, due to its inherent difficulty and fuzziness. This work aims to the establishment of a quantitative approach of mammographic masses texture classification, based on advanced classifier architectures and supported by fractal analysis of the dataset of the extracted textural features. Additionally, a comparison of the information content of the proposed feature set with that of the qualitative characteristics used in clinical practice by expert radiologists is presented. METHODS AND MATERIAL: An extensive set of textural feature functions was applied to a set of 130 digitized mammograms, in multiple configurations and scales, constructing compact datasets of textural "signatures" for benign and malignant cases of tumors. These quantitative textural datasets were subsequently studied against a set of a thorough and compact list of qualitative texture descriptions of breast mass tissue, normally considered under a typical clinical assessment, in order to investigate the discriminating value and the statistical correlation between the two sets. Fractal analysis was employed to compare the information content and dimensionality of the textural features datasets with the qualitative information provided through medical diagnosis. A wide range of linear and non-linear classification architectures was employed, including linear discriminant analysis (LDA), least-squares minimum distance (LSMD), K-nearest-neighbors (K-nn), radial basis function (RBF) and multi-layer perceptron (MLP) artificial neural network (ANN), as well as support vector machine (SVM) classifiers. The classification process was used as the means to evaluate the inherent quality and informational content of each of the datasets, as well as the objective performance of each of the classifiers themselves in real classification of mammographic breast tumors against verified diagnosis. RESULTS: Textural features extracted at larger scales and sampling box sizes proved to be more content-rich than their equivalents at smaller scales and sizes. Fractal analysis on the dimensionality of the textural datasets verified that reduced subsets of optimal feature combinations can describe the original feature space adequately for classification purposes and at least the same detail and quality as the list of qualitative texture descriptions provided by a human expert. Non-linear classifiers, especially SVMs, have been proven superior to any linear equivalent. Breast mass classification of mammograms, based only on textural features, achieved an optimal score of 83.9%, through SVM classifiers.

Artificial Intelligence↗