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 469 records · Page 26Linked to original sources
PubMed2003

Classifying hepatitis B virus genotypes.

In 1988, hepatitis B virus (HBV) was classified into four genotypes by a sequence divergence in the entire genome exceeding 8%, and designated by capital letters of the alphabet from A to D. There are seven genotypes of HBV (A-G) at present, and an eighth is on the horizon. They have an uneven geographical distribution, and only a few of them are prevalent in a given area of the world. Thus genotype A is frequent in northwest Europe, Sub-Saharan Africa, India and the North, Central and South America, B as well as C are common in Southeast Asia and Oceania, and D is prevalent in the Mediterranean area, Central Asia and South America. Genotype E is restricted to West Africa, and F is localized in Central and South America. The distribution of genotype G added to the alphabet list very recently has yet to be determined. Coinfection with HBV of distinct genotypes is not infrequent and found in about 10% of infected individuals, and is responsible for intertypic recombination of HBV genomes. The mutation for a stop codon in the precore region (G1896A) for aborting the translation of hepatitis B e antigen (HBeAg) is prohibited in HBV genomes of genotype A, as well as some of genotypes C and F, because they possess C at position 1858 that makes a Watson-Crick pair with G at position 1896. Hence, seroconversion to antibody to HBeAg is forbidden or delayed in individuals who carry them. Evidence is accumulating as regards the influence of HBV genotypes on the progression of chronic hepatitis B and response to antiviral therapies. HBV isolates even of the same genotype can differ in virological and clinical characteristics, and therefore, the genotype needs to be classified further into subtypes, especially if they are clinically relevant.

Genotype↗
PubMed1981

Discrimination analysis as a tool for classifying the value of biochemical and immunological tests in lung cancer.

The data obtained with routine biochemical, immunological and haematological tests in healthy blood donors and in patients with lung cancer were subjected to discrimination analysis with the aim to select the minimum possible combination of methods with the highest probability of distinguishing between the two groups. Transformation of lymphocytes (TR.LY) after PHA stimulation, formation of E-rosettes (E-ROS), and blood glycoproteins (GP) were shown to permit the most complete differentiation between the group of patients and the blood donors. On the basis of the above-mentioned tests the discrimination rule could be defined: IF 52,73249 - 0.45913 X E-ROS - 0.40752 X TR.LY + 0.02790 X GP greater than or equal to 0, then the investigated person is classified as a cancer patient while at values less than 0 the person is classified as a healthy individual.

Computers↗
PubMed2001

Symbiogenesis in learning classifier systems.

Symbiosis is the phenomenon in which organisms of different species live together in close association, resulting in a raised level of fitness for one or more of the organisms. Symbiogenesis is the name given to the process by which symbiotic partners combine and unify, that is, become genetically linked, giving rise to new morphologies and physiologies evolutionarily more advanced than their constituents. The importance of this process in the evolution of complexity is now well established. Learning classifier systems are a machine learning technique that uses both evolutionary computing techniques and reinforcement learning to develop a population of cooperative rules to solve a given task. In this article we examine the use of symbiogenesis within the classifier system rule base to improve their performance. Results show that incorporating simple rule linkage does not give any benefits. The concept of (temporal) encapsulation is then added to the symbiotic rules and shown to improve performance in ambiguous/non-Markov environments.

Algorithms↗
PubMed2007

Using the XCS classifier system for multi-objective reinforcement learning problems.

We investigate the performance of a learning classifier system in some simple multi-objective, multi-step maze problems, using both random and biased action-selection policies for exploration. Results show that the choice of action-selection policy can significantly affect the performance of the system in such environments. Further, this effect is directly related to population size, and we relate this finding to recent theoretical studies of learning classifier systems in single-step problems.

Artificial Intelligence↗
PubMed2006

A probabilistic Classifier System and its application in data mining.

The article is about a new Classifier System framework for classification tasks called BYP-CS (for BaYesian Predictive Classifier System). The proposed CS approach abandons the focus on high accuracy and addresses a well-posed Data Mining goal, namely, that of uncovering the low-uncertainty patterns of dependence that manifest often in the data. To attain this goal, BYP-CS uses a fair amount of probabilistic machinery, which brings its representation language closer to other related methods of interest in statistics and machine learning. On the practical side, the new algorithm is seen to yield stable learning of compact populations, and these still maintain a respectable amount of predictive power. Furthermore, the emerging rules self-organize in interesting ways, sometimes providing unexpected solutions to certain benchmark problems.

Algorithms↗
PubMed1992

The DSM-III-R impulse control disorders not elsewhere classified: clinical characteristics and relationship to other psychiatric disorders.

OBJECTIVE: The authors reviewed available studies of DSM-III-R impulse control disorders not elsewhere classified in order to determine the relationship of these disorders to one another and to other psychiatric disorders. METHOD: The review focused on the demographic and clinical characteristics, phenomenology, family history, biology, and response to treatment of individuals with intermittent explosive disorder, kleptomania, pathological gambling, pyromania, and trichotillomania. Analysis was restricted to reports which either indicated use of operational diagnostic criteria or provided descriptions of the impulsive behavior detailed enough that patients could be judged as probably meeting the DSM-III-R criteria. RESULTS: Although different impulse control disorders have different sex ratios, all have similar ages at onset and courses. Studies on phenomenology, family history, and response to treatment suggest that intermittent explosive disorder, kleptomania, pathological gambling, pyromania, and trichotillomania may be related to mood disorders, alcohol and psychoactive substance abuse, and anxiety disorders (especially obsessive-compulsive disorder). Biological studies indicate that intermittent explosive disorder and pyromania may share serotonergic abnormalities similar to those reported in mood disorders. CONCLUSIONS: The impulse control disorders not elsewhere classified appear to be related to one another and to mood, anxiety, and psychoactive substance use disorders. Thus, like major depression, obsessive-compulsive disorder, panic disorder, bulimia nervosa, and attention deficit hyperactivity disorder, they may represent forms of "affective spectrum disorder."

Disruptive, Impulse Control, and Conduct Disorders↗
PubMedDate not supplied

Neurodevelopmental characteristics of children with learning impairments classified according to the double-deficit hypothesis.

The double-deficit model has been examined primarily in relation to reading. We investigated whether children classified according to the double-deficit model would exhibit differences in other neuropsychological domains. Children referred for learning problems (N = 188), ages 7 to 11, were classified by double-deficit subtype. Only three of the four groups predicted by the model could be identified. There were no group differences in IQ or attention problems. The three groups showed different neuropsychological profiles, involving functional domains other than reading and language. Differences also emerged in nonverbal low-level information processing. The double-deficit group was generally most severely affected. The double-deficit groupings identify children with different neuropsychological profiles and variation in the efficiency of basic online information processing, extending beyond the oral and written language domain.

Anomia↗
PubMedDate not supplied

Comparison of students classified as LD who petitioned for or fulfilled the college foreign language requirement.

In this study, we compared the cognitive, academic achievement, and demographic profiles of 46 students from one university who had been classified as learning disabled (LD) and had received permission to substitute courses for the university's foreign language (FL) requirement (petition group) with the profiles of 21 students from the same university who had been classified as LD and had fulfilled the university's FL requirement by passing FL courses (nonpetition group). Results showed no significant differences between the two groups on measures of reading, mathematics, written language, American College Testing score, and graduating grade point average when IQ was used as a covariate. More petition than nonpetition students had at least a 1.0 SD discrepancy between IQ and achievement and had been referred only for FL learning problems. More nonpetition than petition students had taken an FL in college and received accommodations in the FL. The two groups together appeared to constitute a heterogeneous group of learners, with more than half failing to meet a minimum discrepancy criterion for classification as LD. The discussion addresses the classification system for LD, the process for determining the presence of FL learning problems and how to address them, and directions for further research.

Achievement↗
PubMedDate not supplied

Using in vitro prediction models instead of the rabbit eye irritation test to classify and label new chemicals: a post hoc data analysis of the international EC/HO validation study.

The international validation study on alternative methods to replace the Draize rabbit eye irritation test, funded by the European Commission (EC) and the British Home Office (HO), took place during 1992-1994, and the results were published in 1995. The results of this EC/HO study are analysed by employing discriminant analysis, taking into account the classification of the in vivo data into eye irritation classes A (risk of serious damage to eyes), B (irritating to eyes) and NI (non-irritant). A data set for 59 test items was analysed, together with three subsets: surfactants, water-soluble chemicals, and water-insoluble chemicals. The new statistical methods of feature selection and estimation of the discriminant functions classification error were used. Normal distributed random numbers were added to the mean values of each in vitro endpoint, depending on the observed standard deviations. Thereafter, the reclassification error of the random observations was estimated by applying the fixed function of the mean values. Moreover, the leaving-one-out cross-classification method was applied to this random data set. Subsequently, random data were generated r times (for example, r = 1000) for a feature combination. Eighteen features were investigated in nine in vitro test systems to predict the effects of a chemical in the rabbit eye. 72.5% of the chemicals in the undivided sample were correctly classified when applying the in vitro endpoints lgNRU of the neutral red uptake test and lgBCOPo5 of the bovine opacity and permeability test. The accuracy increased to 80.9% when six in vitro features were used, and the sample was subdivided. The subset of surfactants was correctly classified in more than 90% of cases, which is an excellent performance.

Animal Testing Alternatives↗
PubMed2000

A revised protocol for more clearly classifying a nonunion.

Classifying a nonunion solely on the extent of callus formation on a radiograph is insufficient. We have conducted a study with a revised protocol to classify a nonunion more clearly, which uses both radiographic observation and fixation stability. 47 consecutive femoral shaft nonunions were studied using this protocol and 42 nonunions were followed up for at least one year after discovery (range, 1-3 years). 40 nonunions healed after treatment with a union rate of 95.2% (40/42) and a union period of 4.7 plus or minus 0.9 months. The 2 cases of failure were due to implant failure and both healed after reoperation. We recommend this revised protocol as a pretreatment assessment for all nonunions because of its integrity and effectiveness.

Journal Article↗
PubMed1976

Performance of the LARC classifier in clinical laboratories.

As part of the installation procedure of the LARC leukocyte differential classifier in a clinical laboratory, a 100-slide protocol is carried out to establish the performance of the classifier in the laboratory. The detailed make-up of this protocol and its relationship to key performance parameters for the leukocyte differential are described in detail. Data from the first ten of these protocols are presented which establish the (a) normal ranges, (b) reproducibility, (c) accuracy, (d) false-positive/false-negative rates for the detection of left shifts and (e) false-positive/false-negative rates for the detection of bloods with abnormal cells.

Autoanalysis↗
PubMed1979

A practical application of computer pattern recognition research: the Abbott ADC-500 differential classifier.

The ADC-500 is a new blood cell differential classifier manufactured by Abbott Laboratories. It performs 500-cell leukocyte differentials on both normal and abnormal cells, evaluates red cell morphology and estimates platelet sufficiency at a rate of 40 to 50 samples per hour in stand-alone operation. The ADC-500 system consists of a spinner which prepares a uniform blood monolayer on a slide, a stainer which reproducibly stains the slide with Wright's stain, an encoder which attaches an instrument and human readable identification to the slide and an analyzer which accepts a stack of up to 50 slides, evaluates these slides and prints the results and the slide identification on report forms. The system's analysis rate, which represents a 5- to 10-fold increase over other commercially available differential counters, requires a number of specialized techniques for its realization. One key to this performance is the development of a high speed X-Y slide positioning stage which can move to a new cell and settle in 50 msec. Another is the high degree of parallelism used in the system structure and the pipelining of the data processing. A third is the development of uniform and repeatable sample preparation modules. Within the analyzer module, the autofocus, white cell acquisition and high resolution cell analysis systems are independent and operate in parallel. At the same time within the high resolution cell analysis system, one cell is acquired; the digitized image of a second processed; and a third is classified using pattern recognition techniques. All of these tasks, except focus, are under the control of a minicomputer system. Tests of the system reveal good accuracy and an improvement in precision due to the increase in the number of counted cells.

Blood Cells↗
PubMed2005

Recognition of protein/gene names from text using an ensemble of classifiers.

This paper proposes an ensemble of classifiers for biomedical name recognition in which three classifiers, one Support Vector Machine and two discriminative Hidden Markov Models, are combined effectively using a simple majority voting strategy. In addition, we incorporate three post-processing modules, including an abbreviation resolution module, a protein/gene name refinement module and a simple dictionary matching module, into the system to further improve the performance. Evaluation shows that our system achieves the best performance from among 10 systems with a balanced F-measure of 82.58 on the closed evaluation of the BioCreative protein/gene name recognition task (Task 1A).

Biomedical Research↗
PubMed2006

A probabilistic classifier for olfactory receptor pseudogenes.

BACKGROUND: Olfactory receptors (ORs), the largest mammalian gene superfamily (900-1400 genes), has >50% pseudogenes in humans. While most of these inactive genes are identified via coding frame (nonsense) disruptions, seemingly intact genes may also be inactive due to other deleterious (missense) mutations. An ultimate assessment of the actual size of the functional human OR repertoire thus requires an accurate distinction between genes and pseudogenes. RESULTS: To characterize inactive ORs with intact open reading frame, we have developed a probabilistic Classifier for Olfactory Receptor Pseudogenes (CORP). This algorithm is based on deviations from a functionally crucial consensus, constituting sixty highly conserved positions identified by a comparison of two evolutionarily-constrained OR repertoires (mouse and dog) with a small pseudogene fraction. We used a logistic regression analysis to assign appropriate coefficients to the conserved position and thus achieving maximal separation between active and inactive ORs. Consequently, the algorithms identified only 5% of the mouse functional ORs as pseudogenes, setting an upper limit of 0.05 to the false positive detection. Finally we used this algorithm to classify the 384 purportedly intact human OR genes. Of these, 135 were predicted as likely encoding non-functional proteins, and 38 were segregating between active and inactive forms due to missense polymorphisms. CONCLUSION: We demonstrated that the CORP algorithm is capable to distinguish between functional and non-functional OR genes with high precision even when the encoded protein would differ by a single amino acid. Using the CORP algorithm, we predict that approximately 70% of human OR genes are likely non-functional pseudogenes, a much higher number than hitherto suspected. The method we present may be employed for better annotation of inactive members in other gene families as well. CORP algorithm is available at: http://bioportal.weizmann.ac.il/HORDE/CORP/

Algorithms↗
PubMed2006

Building multiclass classifiers for remote homology detection and fold recognition.

BACKGROUND: Protein remote homology detection and fold recognition are central problems in computational biology. Supervised learning algorithms based on support vector machines are currently one of the most effective methods for solving these problems. These methods are primarily used to solve binary classification problems and they have not been extensively used to solve the more general multiclass remote homology prediction and fold recognition problems. RESULTS: We present a comprehensive evaluation of a number of methods for building SVM-based multiclass classification schemes in the context of the SCOP protein classification. These methods include schemes that directly build an SVM-based multiclass model, schemes that employ a second-level learning approach to combine the predictions generated by a set of binary SVM-based classifiers, and schemes that build and combine binary classifiers for various levels of the SCOP hierarchy beyond those defining the target classes. CONCLUSION: Analyzing the performance achieved by the different approaches on four different datasets we show that most of the proposed multiclass SVM-based classification approaches are quite effective in solving the remote homology prediction and fold recognition problems and that the schemes that use predictions from binary models constructed for ancestral categories within the SCOP hierarchy tend to not only lead to lower error rates but also reduce the number of errors in which a superfamily is assigned to an entirely different fold and a fold is predicted as being from a different SCOP class. Our results also show that the limited size of the training data makes it hard to learn complex second-level models, and that models of moderate complexity lead to consistently better results.

Algorithms↗
PubMed2003

Classifying the precancers: a metadata approach.

BACKGROUND: During carcinogenesis, precancers are the morphologically identifiable lesions that precede invasive cancers. In theory, the successful treatment of precancers would result in the eradication of most human cancers. Despite the importance of these lesions, there has been no effort to list and classify all of the precancers. The purpose of this study is to describe the first comprehensive taxonomy and classification of the precancers. As a novel approach to disease classification, terms and classes were annotated with metadata (data that describes the data) so that the classification could be used to link precancer terms to data elements in other biological databases. METHODS: Terms in the UMLS (Unified Medical Language System) related to precancers were extracted. Extracted terms were reviewed and additional terms added. Each precancer was assigned one of six general classes. The entire classification was assembled as an XML (eXtensible Mark-up Language) file. A Perl script converted the XML file into a browser-viewable HTML (HyperText Mark-up Language) file. RESULTS: The classification contained 4700 precancer terms, 568 distinct precancer concepts and six precancer classes: 1) Acquired microscopic precancers; 2) acquired large lesions with microscopic atypia; 3) Precursor lesions occurring with inherited hyperplastic syndromes that progress to cancer; 4) Acquired diffuse hyperplasias and diffuse metaplasias; 5) Currently unclassified entities; and 6) Superclass and modifiers. CONCLUSION: This work represents the first attempt to create a comprehensive listing of the precancers, the first attempt to classify precancers by their biological properties and the first attempt to create a pathologic classification of precancers using standard metadata (XML). The classification is placed in the public domain, and comment is invited by the authors, who are prepared to curate and modify the classification.

Decision Support Systems, Clinical↗
PubMed2005

Prognostic factors and life expectancy in myelodysplastic syndromes classified according to WHO criteria: a basis for clinical decision making.

PURPOSE: The aim of this study was to evaluate the prognostic value of the WHO proposal, to assess the role of the main prognostic factors in myelodysplastic syndromes (MDSs) classified into WHO subgroups, and to estimate mortality (standardized mortality ratio [SMR]) and life expectancy in these groups as a basis for clinical decision making. PATIENTS AND METHODS: Four hundred sixty-seven patients who were diagnosed as having de novo MDS at the Division of Hematology, University of Pavia (Pavia, Italy), between 1992 and 2002, were evaluated retrospectively for clinical and hematologic features at diagnosis, overall survival (OS), and progression to leukemia (leukemia-free survival). RESULTS: Significant differences in survival were noted between patients with refractory anemia (RA), refractory cytopenia with multilineage dysplasia, RA with excess blasts, type 1 (RAEB-1), and RAEB-2. The effect of demographic factors on OS was observed in MDS patients without excess blasts (age, P = .001; sex, P = .006), as in the general population. The mortality of RA patients 70 years or older did not differ significantly from that of the general population (SMR, 1.62; P = .06). Cytogenetics was the only International Prognostic Scoring System variable showing a prognostic value in MDS classified into WHO subgroups. Transfusion-dependent patients had a significantly shorter survival than patients who did not require transfusions (P < .001). Developing a secondary iron overload significantly affected the survival of transfusion-dependent patients (P = .003). CONCLUSION: These data show that the WHO classification of MDSs has a relevant prognostic value. This classification, along with cytogenetics, might be useful in decisions regarding transplantation. MDS with isolated erythroid lineage dysplasia identifies a subset of truly low-risk patients, for whom a conservative approach is advisable.

Adult↗
PubMed1985

Prolactin changes after seizures classified by EEG monitoring.

We examined the postictal hyperprolactinemia after seizures classified by EEG and video telemetry. Prolactin did rise after complex partial seizures that involved motor behaviors and was not further increased by secondary generalization. Nontemporal partial seizures or pseudoseizures did not demonstrate this increase. There was no difference in prolactin elevation after generalized tonic-clonic seizures, whether secondarily generalized or generalized from onset. A discriminant function accurately classified 94% of patients with pseudoseizures and 66% of patients with true cerebral seizures for an overall classification accuracy of 72%.

Adolescent↗