Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “classifier”

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 523 records · Page 29Linked to original sources
PubMed2004

[Cardiogenic oscillation--utility of classifying types of sleep apnea].

Cardiogenic pulse waves that travel along the airway wall are captured as airflow variations synchronized with the electrocardiogram at the airway exit. We have used airflow variations caused by cardiogenic pulse waves (hereinafter referred to as cardiogenic oscillation) to classify types of sleep apnea. Pulse waves do not travel downstream if there is an airway obstruction or choke point (wave speed theory). Thus, cardiogenic oscillation was detectable in central apnea or hypopnea, but not in obstructive apnea. In mixed-type sleep apnea, cardiogenic oscillation disappeared, and thus airway obstruction was judged to have occurred, during central sleep apnea. It was thought that obstructive sleep apnea followed central sleep apnea because there was an airway obstruction, although respiratory effort resumed after the end of central sleep apnea. The pattern of mixed-type sleep apnea was understood from the observation of cardiogenic oscillation. Cardiogenic oscillation is useful for not only classifying types of sleep apnea but also for detecting an airway obstruction.

Aged↗
PubMed2005

The cross-classified multilevel measurement model: an explanation and demonstration.

The link between the hierarchical generalized linear model (HGLM) and the Rasch model's parameterization has already been demonstrated by several researchers. Extensions have been described that include higher clustering levels to model more appropriately the contextual effects that are frequently encountered in educational research. However, pure hierarchies are relatively rare and instead cross-classified data structures are more frequently encountered. Cross-classified random effect modeling (CCREM) is still not commonly used. Use of CCREM in combination with the multilevel measurement model (MMM) has been recently introduced and is described further in the current study. Specifically, the link between the MMM and the CCREM MMM (termed "CCMMM" model) is provided. A dataset was simulated to demonstrate interpretation of the CCMMM model's parameters and to compare results under a CCMMM versus HGLM analysis. An Appendix is provided to demonstrate SAS GLIMMIX code used to estimate HGLM and CCMMM models' parameters.

Data Interpretation, Statistical↗
PubMed1992

Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects.

The value of PET as an objective diagnostic tool for dementia may depend on the degree to which abnormal metabolic patterns can be detected by quantitative classification methods. In these studies, a neural-network classifier based on coarse region of interest analyses was used to classify normal and abnormal FDG-PET scans. The performance of neural networks and of an expert reader were evaluated by cross-validation testing. When the "abnormal" class was represented by subjects with clinical diagnoses of "Probable Alzheimer's," the areas under the relative-operating-characteristic (ROC) curves were 0.85 and 0.89 for the neural network and the expert reader, respectively. When testing with abnormal subjects represented by "Possible AD" cases, ROC areas for both the network and the expert were 0.81. The neural network out-performed discriminant analysis. It is concluded that PET has potential for the detection of abnormal brain function in dementing diseases, and that the combination of neural networks and PET is a useful diagnostic tool. Despite the low-resolution "view" afforded the neural network, its performance was nearly equivalent to that of an expert reader.

Adult↗
PubMed2006

Risk factors profile and clinical outcome of ischemic stroke patients admitted in a Department of Internal Medicine and classified by TOAST classification.

AIM: A classification of ischemic stroke subtypes tailored for individual patients is hard to achieve. In 1993, the Trial of Org 10172 in Acute Stroke Treatment (TOAST) group developed a new system to classify the subtypes of ischemic stroke. In our study we applied the TOAST classification to a group of consecutive patients affected by ischemic stroke, to evaluate outcome and factors associated to each stroke subtype. METHODS: To evaluate the prognosis and the associated factors of ischemic stroke subtypes, we classified according to the TOAST classification a cohort of 159 consecutive patients affected by an acute ischemic stroke. We evaluated neurological deficit at admission by Scandinavian Stroke Scale and scored disability at discharge and 6 months after discharge using the Rankin disability scale. We determined 30 days survival and anamnestically evaluated major vascular risk factors. RESULTS: Patients with cardioembolic stroke and stroke of undetermined etiology had a greater neurological deficit on admission and the worst prognosis either in terms of disability or mortality. Lacunar stroke had the least neurological deficit at admission and the best prognosis. Hypercholesterolemia and smoking were more frequent among patients with large artery atherosclerotic stroke. Hypertension, a history of transient ischemic attack and diabetes were more frequent among patients with lacunar stroke. A weak association with hypertension and smoking was observed for cardioembolic stroke. CONCLUSIONS: The TOAST classification is useful in the clinical setting because it identifies ischemic stroke subtypes with different prognosis and with a different profile of associated factors.

Acute Disease↗
PubMed2006

Using Bagging classifier to predict protein domain structural class.

Classification and prediction of protein domain structural class is one of the important topics in the molecular biology. We introduce the Bagging (Bootstrap aggregating), one of the bootstrap methods, for classifying and predicting protein structural classes. By a bootstrap aggregating procedure, the Bagging can improve a weak classifier, for instance the random tree method, to a significant step towards optimality. In this research, it is demonstrated that the Bagging performed at least as well as LogitBoost and Support vector machines in predicting the structural classes for a given protein domain dataset by 10 cross-validation test, which indicate that the Bagging method is promising and anticipated that it could be potentially further improved on predicting protein structural classes as well as other bio-macromolecular attributes, if the bagging method and other existing methods can be effectively complemented with each other.

Algorithms↗
PubMed1991

[Cervix smears with mild atypia classified as Pap-class IIIA: results of a changed policy].

To evaluate the changed policy regarding cervical smears with mild atypia in the Netherlands (before 1988 classified as Pap II, thereafter as Pap IIIA), an descriptive study was carried out in the laboratory of gynaecological cytology Cyt-U-Universitair (University of Utrecht). All women who in 1988 had a first smear with mild atypia were included. After a follow-up period of 2 years the final diagnosis was noted. The results of the follow-up of cervical smears with mild atypia were compared with the follow-up results of moderate atypia. A selection was made of 477 women. In 55% of the cases no dysplasia was found during the 2 years of follow-up. In 12% of the cases severe dysplasia, carcinoma in situ or carcinoma was found. Compared with the smears with moderate atypia the dysplasia in the follow-up of mild atypia led to fewer cases of proven dysplasia. In addition the dysplasia in these cases was of a lower grade. The group of women referred to the gynaecologist after the smear with mild atypia was remarkably large (30%). As the results show, the policy of careful follow-up of women with mild atypia in the cervical smear has been a good one. It implies classifying mild atypia as Pap IIIA. As many women (55%) do not display cervical dysplasia within a 2 year follow-up period, the policy of making a repeat smear before referring appears justified.

Adult↗
PubMed1991

Comparison of different methods of classifying patients with systemic lupus erythematosus.

The sensitivity and percent agreement of 3 methods of classifying patients with systemic lupus erythematosus (SLE), the 1982 American Rheumatism Association (ARA) revised criteria and a simple and full classification tree, were compared using data from The Johns Hopkins Lupus Cohort, a prospective study of 198 patients with SLE. The 1982 revised criteria were significantly more sensitive than the simple classification tree, correctly identifying 184 (93%) cases compared to 168 (85%) (p = 0.016). Agreement between these 2 classification schema was noted in 174 (87%) cases with 164 classified correctly and 10 failing to satisfy either criteria set. The full classification tree correctly identified 186 (94%) cases. There was no difference in the sensitivity of either the 1982 revised criteria or the full classification tree by racial group; however, the simple classification tree was significantly less sensitive among black than Caucasian patients (80 vs 91.5% p = 0.038). Our data support the continued use of the 1982 revised ARA criteria for the classification of patients with SLE for clinical and epidemiologic research studies.

Adult↗
PubMed1991

GCDFP-70 protein in cyst fluid identified as albumin and used to classify cysts in women with breast gross cystic disease.

We used sodium dodecyl sulfate-polyacrylamide gel electrophoresis to study cyst fluids from women with breast gross cystic disease. The subjects could be classified into two categories according to the concentrations of protein GCDFP-70 in the cyst fluid: those with Type I cysts had a very low content of this protein; those with Type II cysts had very high concentrations. Analysis of the amino acid sequence of GCDFP-70 from both cyst fluid types confirmed that this protein is human plasma albumin. The average concentration of albumin found in Type I cyst fluids was 0.32 g/L and that corresponding to Type II was 10.16 g/L. Thus, albumin quantification from cyst fluids or analysis by either polyacrylamide or agarose gel electrophoresis provides a simple procedure for classifying these fluids, yielding results that correlate well with previous classifications based on other measurements such as sodium, potassium, and chloride concentrations. This albumin-based quantification method may improve the classification of breast cysts and might be useful in further studies on functional changes in the cysts and their relationship to breast cancer.

Adult↗
PubMedDate not supplied

Classifying basic research designs.

Considerable confusion over terminology for classifying basic types of research design in family medicine stems from the rich variety of substantive topics studied by family medicine researchers, differences in research terminology among the disciplines that family medicine research draws from, and lack of uniform research design terminology within these disciplines themselves. Many research design textbooks themselves fail to specify the dimensions on which research designs are classified or the logic underlying the classification systems proposed. This paper describes a typology based on three dimensions that may be used to characterize the basic design qualities of any study. These dimensions are: 1) the nature of the research objective (exploratory, descriptive, or analytic); 2) the time frame under investigation (retrospective, cross-sectional, or prospective); and 3) whether the investigator intervenes in the events under study (observational or interventional). This three-dimensional typology may be helpful for teaching basic research design concepts, for contemplating research design decisions in planning a study, and as a basis for further consideration of a more detailed, uniform research design classification system.

Cross-Sectional Studies↗
PubMed1989

[Symptomatic mania in HIV infection. 3 cases of classified euphoric psychoses as a sequela of the AIDS disease].

3 case histories of patients with acquired immune deficiency syndrome and acute organic psychoses are reported. Their psychoses can be classified as benefaction or blissfulness psychoses. A handfull of previous reports classify similar acute organic psychoses as symptomatic manias. The respective descriptions make obvious that these psychoses are homogenous among each other and that their contents represent a vivid reflection over a life situation which had been thoroughly changed by the AIDS illness.

Acquired Immunodeficiency Syndrome↗
PubMed1988

Number-comparison learning by children classified as mentally retarded.

A training experiment was undertaken to determine whether children classified as mentally retarded could learn a general magnitude-comparison rule ("The number that comes after another in the number sequence is more than the preceding number"). After a pretest, 22 subjects were randomly assigned to an experimental or a control training group. On both immediate and delayed posttests, the experimental subjects significantly outperformed control children on trained number pairs. A modest amount of transfer was also evident. The results suggest that a counting-based approach that utilizes rule rehearsal can help children classified as mentally retarded to use their representation of the number sequence to make mental magnitude comparisons.

Adolescent↗
PubMed1983

Immunohistologic cellular phenotypes of lymphoproliferative disorders. Comprehensive evaluation of 564 cases including 257 non-Hodgkin's lymphomas classified by the International Working Formulation.

The plethora of classifications for non-Hodgkin's lymphomas (NHLs) and controversy regarding the merits of the individual classification schemes has led to the articulation of an International Working Formulation for NHL classification by a working group sponsored by the National Cancer Institute. This classification is based on both architectural and cytologic features and has been shown to have clinical relevance, but it is not an immunologic approach. With the use of frozen sections and both polyclonal and monoclonal antibodies, a comprehensive immunohistologic study was made of 564 biopsy specimens 1) for determination of the utility of the principle of monoclonality in differentiating benign from malignant lymphoproliferative disorders, 2) for definition of the immunohistochemical phenotypes of histologically benign and malignant cellular proliferations, and 3) for evaluation of the immunologic phenotype of 257 non-Hodgkin's lymphomas classified by the International Working Formulation. Two hundred seven "reactive benign" lymphoproliferations demonstrated polyclonal immunostaining. Monoclonal kappa light chain immunostaining was demonstrated in 3 of 4 cases classified as atypical hyperplasia, two of which had coexistent NHL or subsequently developed overt NHL. Frozen tissue sections were found to be essential for demonstration of immunoglobulin and glycoprotein membrane antigens. The results of immunohistochemical studies were readily integrated with the International Formulation. Although diffuse mixed and small lymphocytic lymphomas were immunologically heterogeneous (both T- and B-cell), follicular lymphomas were invariably of B-cell type, and immunoblastic lymphomas originating from homogeneous T- and B-cell populations were identified.

Antibodies, Monoclonal↗
PubMed1983

Comparison of diving experience factors between divers classified as positive and negative for bone cysts.

Official U.S. Navy diving records for enlisted divers who were 35 years of age and older were examined for differences in diving experience between those who were classified as positive or negative for bone cysts. One-third of the divers (n = 31) were classified as positive, and two-thirds (n = 62) as negative, for bone cysts. Diving experience factors included total years of diving experience, frequency of dives between 15 and 31 m, frequency of dives deeper than 31 m, number of saturation dives, number of dives involving decompression, and number of cases of decompression sickness. None of these factors differed significantly between the two groups. While previous findings showed that bone cysts are twice as prevalent among divers as among a comparison group of Navy enlisted men matched to divers for age, rank, and occupational specialty, these results indicate that this high prevalence rate does not appear to be related to specific types of diving exposure. Other environmental factors such as long-bone injury, disease, or exposure to hazardous or toxic substances may account for differences in the prevalence of bone cysts between the two groups.

Adult↗
PubMed1982

Comparison of the effectiveness of four clinical chemical assays in classifying patients with chest pain.

We compared the usefulness of four serum assays for classifying patients originally suspected of having an acute myocardial infarction. One of these is the long-used measurement of total creatine kinase (CK) activity. The other three are relatively new immunoassays: myoglobin by RIA, CK-BB by RIA, and CK-MB by immunoinhibition. When we evaluated test effectiveness with use of conventionally derived reference ranges, the results were misleading. However, by using receiver operating characteristic curves, we were able to effectively compare the four tests at all possible decision levels, rather than at only one. Multiple closely sequential serum specimens were obtained during the first four days after the onset of chest pain. Total CK, CK-MB, and CK-BB all behaved similarly, reaching peak diagnostic effectiveness at 18-20 h, when all three correctly classified 95% of the infarct patients, with a zero false-positive rate. However, total CK was more useful in identifying infarcts later in their courses than were the two CK isoenzyme tests. Myoglobin assay was most effective earlier in the course, at about 7 to 8 h. Our results indicate (a) that the tests for myoglobin and for CK or its isoenzymes are complementary and (b) that of the three CK tests, measurement of total CK activity provides the most information over the broadest segment of a patient's course.

Creatine Kinase↗
PubMedDate not supplied

An automatic sleep classifier for laboratory rodents.

Two successive main steps are usually performed in automatic sleep classification: a) a periodic quantification of the EEG signal by a set of indices featuring each polygraphic epoch by a point in a multidimensional space, b) a stage decision to determine the cluster that includes the representative point for a given epoch (discriminant analysis). As solid state logic comparators simulate linear discriminant functions, this article describes an improved classifier for automatic sleep scoring in laboratory rodents. Four indices were found to be discriminant on both rats and mice: F(the energy in the EEG theta band relative to that in the delta band), Z(number of EEG zero-crossings), D(variability of the EEG amplitude) and M(integrated EMG). They are measured by analog techniques for 30 sec epochs from an EEG and an EMG lead per animal. At the end of each epoch, the logic results of their respective comparators serves as an input to a hardwired modifiable truthtable where a sleep stage decision is elaborated. The outputs of 4 multiplexed classifiers are controlled by microprocessor and may be used for chart or digital recordings on various peripherals. The overall agreement with visual scoring is more than 95% for both rats and mice.

Animals↗
PubMed1980

[A new ICSB classification of staphylococci stocks previously classified according to Bergey's Manual of Determinative Bacteriology].

A collection of 215 Staphylococci, already classified according to Bergey's Manual of Determinative Bacteriology, has been newly classified by these authors in view of the I C S B classification from 1976. Various stocks coned not be included in this classification. The collection between laboratory tests and patogenic effects of Staphylococci has been considered by the authors.

Staphylococcus↗
PubMed1995

Classifying cervical cells using a recurrent neural network by building basins of attraction.

This paper describes the result of classifying cervical cells employing a novel way of using a Hopfield-style neural network for classification. This method could be used as part of an automated cervical screening system. Rather than storing the exemplars (training elements) as stable points, a connection matrix is determined, using perceptron-type learning, such that the exemplars are placed in basins of attraction. The exemplars from different classes of cells are placed in different basins of attraction, with usually more than one basin per training class. A presented element is then classified by the basin of attraction it falls into. Input to the classification consists of one case of feature vectors derived from the cervical cell images; in another case, input consists of the images themselves. Good results were obtained using this technique.

Cervix Uteri↗
PubMed1995

A hybrid neural and statistical classifier system for histopathologic grading of prostatic lesions.

Neural network and statistical classification methods were applied to derive an objective grading for moderately and poorly differentiated lesions of the prostate, based on characteristics of the nuclear placement patterns. A partly trained multilayer neural network was used as a feature extractor. A hybrid classifier system using a quadratic Bayesian classifier applied to these features allowed grade assignment consensus with visual diagnosis in 96% of fields from a training set of 500 fields and in 77% of 130 fields of a test set.

Humans↗