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 757 records · Page 42Linked to original sources

PubMed1986

[Clinico-morphological analysis of spontaneous testicular tumors in dogs].

The study of 83 cases of testicular tumours in dogs based on the International Histological Classification of tumours in domestic animals (WHO) has shown that these tumours can be divided into three groups: sex cord-stromal tumours (43.3%), germ cell tumours (36.3%) and multiple primary tumours (20.4%). The latter consisted of germ cell and sex cord-stromal components. Choriocarcinomas and teratomas were not found. By morphological criteria and histogenesis testicular tumours in dogs have shown to be very similar to analogous tumours in humans.

Animals↗

PubMed2005

Analysis of recursive gene selection approaches from microarray data.

MOTIVATION: Finding a small subset of most predictive genes from microarray for disease prediction is a challenging problem. Support vector machines (SVMs) have been found to be successful with a recursive procedure in selecting important genes for cancer prediction. However, it is not well understood how much of the success depends on the choice of the specific classifier and how much on the recursive procedure. We answer this question by examining multiple classifers [SVM, ridge regression (RR) and Rocchio] with feature selection in recursive and non-recursive settings on three DNA microarray datasets (ALL-AML Leukemia data, Breast Cancer data and GCM data). RESULTS: We found recursive RR most effective. On the AML-ALL dataset, it achieved zero error rate on the test set using only three genes (selected from over 7000), which is more encouraging than the best published result (zero error rate using 8 genes by recursive SVM). On the Breast Cancer dataset and the two largest categories of the GCM dataset, the results achieved by recursive RR are also very encouraging. A further analysis of the experimental results shows that different classifiers penalize redundant features to different extent and this property plays an important role in the recursive feature selection process. RR classifier tends to penalize redundant features to a much larger extent than the SVM does. This may be the reason why recursive RR has a better performance in selecting genes.

Algorithms↗

PubMed2004

Laminin, Her2/neu and Ki-67 as prognostic factors in non-small cell lung cancer.

PURPOSE: The aim of this study was to determine the values of Ki-67 antigen, oncoprotein Her2/neu and laminin, as prognostic and predictive factors in NSCLC, and on the basis of these markers to create a prognostic model which would make it possible to identify patients with a high risk of disease recurrence. MATERIAL AND METHODS: The material for the study came from 64 patients with NSCLC, who underwent surgery in Dolnoślaskie Centrum Gruźlicy i Chorób Płuc i 1996-2000, and subsequently were given radiation therapy in Dolnoślaskie Centrum Onkologii. RESULTS: Among the markers researched, a high level of (intracellular) laminin in carcinoma cells was found to be an unfavourable prognostic factor. Also, another group of patients with an overexpression of oncoprotein Her2/neu were found to have a poorer prognosis, although the influence of proliferative index Ki-67 on patient survival could not be explained. The prognostic model LAMHER, which was defined on the basis of the intracellular laminin level and expression HER2/neu, enables the identification of a group of patients with a high risk of disease recurrence. In a multidimensional analysis of the "classification tree", it was found that patients with the highest risk of disease recurrence were those with LAMHER = 1 (overexpression Her2/neu and/or a high level of intracellular laminin), and patients with LAMHER = 0 but a Fractionation Dilution Factor higher than 1.57. CONCLUSIONS: The conclusion of this study is that multiple molecular marker testing is necessary to detect an independent prognostic impact on survival and is therefore superior to single marker testing. Based on LAMHER testing, two groups of patients could be defined: a low-risk group (LAMHER = 0) and a high-risk group (LAMHER = 1) for failure of standardized treatment.

Biomarkers, Tumor↗

PubMed1984

Analysis of 90 step-sectioned cystectomized specimens of bladder cancer.

For systematic understanding of papillary superficial carcinoma and nonpapillary invasive carcinoma of the bladder 90 cystectomized specimens of transitional cell carcinoma of the bladder were analyzed by step-section. From gross and microscopic observations bladder cancer was classified into 3 types: type 1--multiple papillary superficial carcinomas with or without small areas of dysplasia or carcinoma in situ, type 2--multiple coexistent papillary and nonpapillary carcinomas with widespread dysplasia and carcinoma in situ, and type 3--solitary nonpapillary invasive carcinoma without dysplasia and carcinoma in situ. Actuarial 5-year survival rates for types 1 to 3 were 97, 50 and 17 per cent, respectively. Although this classification is somewhat overlapping and there is a bias that basic data are obtained from the cystectomized specimens, we believe that such an analysis is necessary for considering the histogenesis and progression of human bladder cancer.

Adult↗

PubMed2005

A heuristic fuzzy logic approach to EMG pattern recognition for multifunctional prosthesis control.

This paper presents a heuristic fuzzy logic approach to multiple electromyogram (EMG) pattern recognition for multifunctional prosthesis control. Basic signal statistics (mean and standard deviation) are used for membership function construction, and fuzzy c-means (FCMs) data clustering is used to automate the construction of a simple amplitude-driven inference rule base. The result is a system that is transparent to, and easily "tweaked" by, the prosthetist/clinician. Other algorithms in current literature assume a longer period of unperceivable delay, while the system we present has an update rate of 45.7 ms with little postprocessing time, making it suitable for real-time application. Five subjects were investigated (three with intact limbs, one with a unilateral transradial amputation, and one with a unilateral transradial limb-deficiency from birth). Four subjects were used for system offline analysis, and the remaining intact-limbed subject was used for system real-time analysis. We discriminated between four EMG patterns for subjects with intact limbs, and between three patterns for limb-deficient subjects. Overall classification rates ranged from 94% to 99%. The fuzzy algorithm also demonstrated success in real-time classification, both during steady state motions and motion state transitioning. This functionality allows for seamless control of multiple degrees-of-freedom in a multifunctional prosthesis.

Adult↗

PubMed1995

Dynamic model comparing the bionomics of two isolated Culex tarsalis (Diptera: Culicidae) populations: sensitivity analysis.

A simulation model described (Eisenberg et al. 1994a) was used to compare the population dynamics of Culex tarsalis Coquillett in the Coachella and southern portion of the San Joaquin Valleys of California. Model outputs were classified as a pass if they met criteria that defined typical seasonal abundance patterns established by CO2 and New Jersey light trap data. The sensitivity of this classification to the model parameters was assessed by running multiple simulations for each valley site. Parameter sets associated with a pass were first analyzed separately for each valley and then compared. The two study sites were distinguished by the distributional characteristics of two parameters associated with temperature dependency. One of these parameters described the temperature dependence of larval mortality and the other the temperature dependence of adult egg development. We hypothesize that these isolated Cx. tarsalis populations evolved separately to maximize survival in their respective temperature regimes by adapting to different optimal larval survival temperatures and egg-development rates.

Analysis of Variance↗

PubMed1997

Prognostic classification for esophageal atresia and tracheoesophageal fistula: Waterston versus Montreal.

Since 1962, the Waterston classification has been used to stratify neonates who have esophageal atresia (EA) and/or tracheoesophageal fistula (TEF) into prognostic categories based on birth weight, the presence of pneumonia, and the identification of other congenital anomalies. In response to advances in neonatal care, the surgeons from the Montreal Children's Hospital proposed a new categorization system in 1993 in an attempt to define the current risk factors for patients who have EA/TEF. In the Montreal experience only two characteristics independently affected survival: preoperative ventilator dependence and associated major anomalies. The goal of this study was to determine which system had the greatest validity for the evaluation of prognosis in our patients with EA/TEF. The charts of 94 patients who had EA/TEF treated between 1972 and 1991 were reviewed. Patients were classified using both the Waterston and Montreal systems. Groups were compared with Fisher's Exact test using a 95% confidence level for statistical significance. Eleven infants were ventilator dependent preoperatively; 62 children had major associated anomalies, 8 of which were considered life threatening. Sixteen children died within 4 years, eight during their initial hospital stay. Five of the eight early postoperative deaths occurred in the highest-risk patients (Waterston C or Montreal II). Analysis was performed for multiple risk factors and mortality. As in the Montreal study, the presence of life-threatening and major congenital anomalies represented significant risk factors for death. Pulmonary disease as delineated by ventilator dependence appeared to be more accurate than pneumonia. This study confirms the accuracy of the Montreal classification in defining prognosis for EA/TEF. The Montreal system more accurately identifies children at highest risk than the Waterston classification.

Abnormalities, Multiple↗

PubMed2003

Tumid lupus erythematosus: criteria for classification with immunohistochemical analysis.

OBJECTIVE: To define comprehensive criteria for the classification and differential diagnosis of tumid lupus erythematosus (LE). METHODS: A prospective study of patients fulfilling the classical description of tumid LE was performed. Clinical evaluation, histopathologic and direct immunofluorescence analyses of skin specimens, and serologic evaluation were conducted. The inflammatory cell infiltrate was quantitatively investigated by immunohistochemical analysis of fresh frozen skin specimens using multiple lymphocytic markers. RESULTS: Fifteen patients were followed for a mean of 7 years. Smooth, indurated, nonscarring, pink to violaceous papules, plaques, or nodules, devoid of surface changes were distributed on sun exposed sites. The mean lesion duration was 2 years, female:male ratio was 8:7, and racial distribution was 11 white, 2 Hispanic, and 2 African American patients. Histopathologic findings included a superficial and deep, perivascular, and frequently periadnexal infiltrate of lymphocytes, mucin deposition throughout the dermis, and absent to focal dermal-epidermal junctional involvement. Direct immunofluorescence immunoreactants and low titer antinuclear antibodies were variably present. Immunohistochemical findings included a predominance of pan-T cell marker CD3-expressing (78.0% +/- 6.3%) T lymphocytes. Most were CD4 expressing (82.7% +/- 8.0%) helper T cells; a minority were CD8 expressing (31.3% +/- 14.0%) cytotoxic T cells. The CD4:CD8 ratio was 3.1 (+/-1.3):1. One patient developed systemic LE and one a discoid LE lesion. CONCLUSION: Comprehensive clinical, histopathologic, and immunohistochemical criteria for the classification of tumid LE are proposed that differentiate tumid LE from other cutaneous disorders that may be clinically and histologically indistinguishable. The chronic, benign course indicates that tumid LE be classified as a form of chronic cutaneous LE, although it may be a cutaneous feature of systemic LE.

Adult↗

PubMed1991

Clinicopathologic description of myocarditis.

Histologic evidence of myocarditis was demonstrated in 35 of 348 patients submitted to endomyocardial biopsy over 5 years. Analysis of the histologic findings and clinical course of these patients resulted in a new clinicopathologic classification of myocarditis in which four distinct subgroups are identified. Patients with fulminant myocarditis become acutely ill after a distinct viral prodrome, have severe cardiovascular compromise, multiple foci of active myocarditis by histologic study and ventricular dysfunction that either resolves spontaneously or results in death. Patients with acute, chronic active and chronic persistent myocarditis have a less distinct onset of illness. Patients with acute myocarditis present with established ventricular dysfunction and may respond to immunosuppressive therapy or their condition may progress to dilated cardiomyopathy. Those with chronic active myocarditis initially respond to immunosuppressive therapy, but they have clinical and histologic relapses and develop ventricular dysfunction associated with chronic inflammatory changes including giant cells on histologic study. Chronic persistent myocarditis is characterized by a persistent histologic infiltrate, often with foci of myocyte necrosis but without ventricular dysfunction despite other cardiovascular symptoms such as chest pain or palpitation.

Acute Disease↗

PubMed2005

Refractive error, ocular biometry, and lens opalescence in an adult population: the Los Angeles Latino Eye Study.

PURPOSE: To characterize age- and gender-related differences in refractive error, ocular biometry, and lens opalescence (NOP) in a population-based sample of adult Latinos. Also assessed were the determinants of age-related refractive differences. METHODS: Participants in the Los Angeles Latino Eye Study (LALES), a population-based study of Latinos aged 40 years and more, underwent an ophthalmic examination, including ultrasonic measurements of axial length (AL), vitreous chamber depth (VCD), anterior chamber depth (ACD), lens thickness (LT), and noncycloplegic automated and subjective refraction. Corneal curvature/power (CP) was measured using an autorefractor. NOP was graded at the slit lamp by an ophthalmologist using the Lens Opacity Classification System II. Age- and gender-related differences were calculated. Multiple regression models were used to identify the determinants of age-related refractive differences. RESULTS: Of the 6357 LALES participants, 5588 phakic individuals with biometric data were included in this analysis. Older individuals had shallower ACDs, thicker lenses, more NOP, and more hyperopia compared to younger individuals (P < 0.001). There was no age-related difference in AL (P > or = 0.05). Women had significantly shorter AL, shallower ACD and VCD, than did men (P < or = 0.01). The strongest determinants of refractive error were AL (primarily VCD) and CP. NOP was a small but significant determinant of refractive error in older individuals. CONCLUSIONS: Age- and gender-related differences in ocular biometric, refractive error, and NOP measurements are present in adult Latinos. While the relative contribution of NOP in determining refractive error is small, it is greater in older persons compared to younger individuals.

Adult↗

PubMed2006

Interpreting cancer trends.

The interpretation of cancer incidence trends is complicated by short-term random variation, artifactual fluctuations introduced by screening, changes in diagnosis or disease classification, completeness of reporting, and by the multiplicity of factors that may affect risk for specific cancer sites. We analyzed trends in 56 different cancer sites and subsites in the U.S. SEER registries in the period 1975-2002 using join-point analysis. The increase in cancer incidence for all sites combined that became evident with the inception of the SEER registries in the mid-1970s has abated since the early 1990s. Among the 15 most common cancer sites in men, sites with increasing incidence rates during the most recent time period include melanoma of the skin and cancers of the prostate, kidney and renal pelvis (kidney), and esophagus. Among women, incidence rates are increasing for leukemia, non-Hodgkin's lymphoma, melanoma, and cancers of the breast, thyroid, urinary bladder, and kidney. Incidence rates for all childhood cancers combined increased 0.6% per year from 1975 to 2002. Cancer mortality rates have decreased in the United States since 1991 in both men and in women; site-specific death rates have decreased in the most recent time period for 12 of the top 15 cancer sites in men and 9 of the top 15 cancer sites in women. Similar trends in cancer incidence and mortality have been reported in other industrialized countries. Possible reasons for these trends are discussed.

Child, Preschool↗

PubMed1995

Loss of periodontal attachment in HIV-seropositive military personnel.

The cross-sectional relationship between severe loss of periodontal attachment (LPA) and worsening immune status due to HIV infection was evaluated in 474 HIV-infected subjects (416 men, 58 women) aged 18 to 49 years who had been classified at stages 1 through 6 of the Walter Reed Army Institute of Research (WR) Staging Classification System. LPA was measured at four sites per tooth using a manual probe; severe LPA was defined as > or = 1 site/subject exhibiting > or = 5 mm LPA. Severe LPA was found in 94 (20%) of the subjects. Modeling with multiple logistic regression analysis revealed that WR stage and peripheral CD4+ lymphocyte cell counts were not significant independent predictors of severe LPA. Severe LPA was more common in subjects at WR stage 5 or 6 who exhibited oral candidiasis (OC), a marker of immune system damage, than in persons at those WR stages without OC (odds ratio = 7.85; 95% confidence interval (CI) = 1.94-31.81). After the analysis controlled for WR stage, younger subjects receiving AZT had greater odds of severe LPA than same-age subjects not taking the drug (e.g., odds ratio for subjects aged 30 years = 2.59; 95% CI = 1.22, 5.49). Other significant predictors in the model included male sex; retired military status; cigarette smoking; and presence of cratered, ulcerated, or necrotic interdental papillae. HIV-associated immune deficiency may be associated with localized severe LPA, but this may be an indirect association due to medication use, opportunistic infection, or other factors not captured by the WR staging system or peripheral CD4+ cell counts.(ABSTRACT TRUNCATED AT 250 WORDS)

AIDS-Related Opportunistic Infections↗

PubMed2002

Gene expression data analysis with a dynamically extended self-organized map that exploits class information.

MOTIVATION: Currently the most popular approach to analyze genome-wide expression data is clustering. One of the major drawbacks of most of the existing clustering methods is that the number of clusters has to be specified a priori. Furthermore, by using pure unsupervised algorithms prior biological knowledge is totally ignored Moreover, most current tools lack an effective framework for tight integration of unsupervised and supervised learning for the analysis of high-dimensional expression data and only very few multi-class supervised approaches are designed with the provision for effectively utilizing multiple functional class labeling. RESULTS: The paper adapts a novel Self-Organizing map called supervised Network Self-Organized Map (sNet-SOM) to the peculiarities of multi-labeled gene expression data. The sNet-SOM determines adaptively the number of clusters with a dynamic extension process. This process is driven by an inhomogeneous measure that tries to balance unsupervised, supervised and model complexity criteria. Nodes within a rectangular grid are grown at the boundary nodes, weights rippled from the internal nodes towards the outer nodes of the grid, and whole columns inserted within the map The appropriate level of expansion is determined automatically. Multiple sNet-SOM models are constructed dynamically each for a different unsupervised/supervised balance and model selection criteria are used to select the one optimum one. The results indicate that sNet-SOM yields competitive performance to other recently proposed approaches for supervised classification at a significantly reduced computational cost and it provides extensive exploratory analysis potentiality within the analysis framework. Furthermore, it explores simple design decisions that are easier to comprehend and computationally efficient.

Algorithms↗

PubMed2001

Immunohistochemical panels for differentiating epithelial malignant mesothelioma from lung adenocarcinoma: a study with logistic regression analysis.

Immunohistochemistry provides an important indicator for differential diagnosis between pleural malignant mesothelioma and lung adenocarcinoma, which have complex therapeutic and medicolegal implications. To pinpoint a reliable, restricted panel of markers, the authors evaluated the efficacy of select commercial antibodies in a series of patients with confirmed clinicopathologic diagnosis of mesothelioma or lung adenocarcinoma with the aid of multiple logistic classification tables. Specimens of 46 mesotheliomas and 20 lung adenocarcinomas were examined with calretinin, thrombomodulin, cytokeratins (CKs) 5/6, and high-molecular weight CKs (indicators of mesothelioma), alongside MOC 31, Ber-EP4, and carcinoembryonic antigen (CEA; indicators of lung adenocarcinoma). Of the mesotheliomas, 40 of 46 (87%) were positive with calretinin, 29 of 46 (63%) with thrombomodulin, 40 of 46 (87%) with CKs 5/6, and 41 of 46 (89%) with high-weight CKs; five of 46 mesotheliomas (11%) were focally reactive with MOC 31, four of 46 (9%) with Ber-EP4, and two of 46 (4%) with CEA. Of the lung adenocarcinomas, 18 of 20 (90%) were positive with MOC 31, 20 of 20 (100%) with Ber-EP4, and 17 of 20 (85%) with CEA; and two of 20 (10%) were focally reactive with calretinin, one of 20 (5%) with thrombomodulin, none of 20 (0%) with CKs 5/6, and five of 20 (25%) with high-weight CKs. Multiple logistic modeling indicated two batteries of three antibodies permitting more than 98% overall accuracy: Ber-EP4 plus CKs 5/6 plus calretinin, and Ber-EP4 plus CKs 5/6 plus CEA.

Adenocarcinoma↗

PubMed2003

PCA disjoint models for multiclass cancer analysis using gene expression data.

MOTIVATION: Microarray expression profiling appears particularly promising for a deeper understanding of cancer biology and to identify molecular signatures supporting the histological classification schemes of neoplastic specimens. However, molecular diagnostics based on microarray data presents major challenges due to the overwhelming number of variables and the complex, multiclass nature of tumor samples. Thus, the development of marker selection methods, that allow the identification of those genes that are most likely to confer high classification accuracy of multiple tumor types, and of multiclass classification schemes is of paramount importance. RESULTS: A computational procedure for marker identification and for classification of multiclass gene expression data through the application of disjoint principal component models is described. The identified features represent a rational and dimensionally reduced base for understanding the basic biology of diseases, defining targets for therapeutic intervention, and developing diagnostic tools for the identification and classification of multiple pathological states. The method has been tested on different microarray data sets obtained from various human tumor samples. The results demonstrate that this procedure allows the identification of specific phenotype markers and can classify previously unseen instances in the presence of multiple classes.

Acute Disease↗

PubMed2001

Contributions of strength and body composition to the gender difference in anaerobic power.

BACKGROUND: The purpose of this study was to determine the significant factors contributing to the difference between men and women in anaerobic power (AP) generation. METHODS. PARTICIPANTS: College men (n = 113) and women (n = 175) were randomly selected for evaluation from a fitness class prior to training. MEASURES: AP was determined from the Margaria-Kalamen stair run. Dynamic strength was evaluated from the 1-RM single-leg extension. Body composition was estimated from gender-specific skinfold prediction equations. RESULTS: Discriminant analysis identified %fat, fat-free mass (FFM), and leg extension strength as significant components differentiating between men and women and allowed proper gender classification in 99% of the cases. When both strength and body composition variables were held constant by covariance, there was no significant difference between men and women in AP (F = 1.25). A multiple regression equation using %fat, FFM, and leg extension strength predicted AP (R = 0.91) with an error of 150 W. When the multiple regression equations was applied to a cross-validation sample, AP could be accurately predicted (r = 0.91, t = 0.17) with an average error of 3% +/- 21%. CONCLUSIONS: Body size and dynamic strength are the major factors explaining the explosive leg power difference between men and women without a specific or unique gender quality.

Adolescent↗

PubMed2025

Circulating microRNA panels for multi-cancer detection and gastric cancer screening: leveraging a network biology approach.

BACKGROUND: Screening tests, particularly liquid biopsy with circulating miRNAs, hold significant potential for non-invasive cancer detection before symptoms manifest. METHODS: This study aimed to identify biomarkers with high sensitivity and specificity for multiple and specific cancer screening. 972 Serum miRNA profiles were compared across thirteen cancer types and healthy individuals using weighted miRNA co-expression network analysis. To prioritize miRNAs, module membership measure and miRNA trait significance were employed. Subsequently, for specific cancer screening, gastric cancer was focused on, using a similar strategy and a further step of preservation analysis. Machine learning techniques were then applied to evaluate two distinct miRNA panels: one for multi-cancer screening and another for gastric cancer classification. RESULTS: The first panel (hsa-miR-8073, hsa-miR-614, hsa-miR-548ah-5p, hsa-miR-1258) achieved 96.1% accuracy, 96% specificity, and 98.6% sensitivity in multi-cancer screening. The second panel (hsa-miR-1228-5p, hsa-miR-1343-3p, hsa-miR-6765-5p, hsa-miR-6787-5p) showed promise in detecting gastric cancer with 87% accuracy, 90% specificity, and 89% sensitivity. CONCLUSIONS: Both panels exhibit potential for patient classification in diagnostic and prognostic applications, highlighting the significance of liquid biopsy in advancing cancer screening methodologies.

Neoplasms↗

PubMed1983

The analysis of dependence in cross-classifications having ordered categories, using log-linear models for frequencies and log-linear models for odds.

To analyse the dependence of a qualitative (dichotomous or polytomous) response variable upon one or more qualitative explanatory variables, log-linear models for frequencies are compared with log-linear models for odds, when the categories of the response variable are ordered and the categories of each explanatory variable may be either ordered or unordered. The log-linear models for odds express the odds (or log odds) pertaining to adjacent response categories in terms of appropriate multiplicative (or additive) factors. These models include the 'null log-odds model', the 'uniform log-odds model', the 'parallel log-odds model', and other log-linear models for the odds. With these models, the dependence of the response variable (with ordered categories) can be analyzed in a manner analogous to the usual multiple regression analysis and related analysis of variance and analysis of covariance. Application of log-linear models for the odds sheds light on earlier applications of log-linear models for the frequencies in contingency tables with ordered categories.

Biometry↗