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 361 records · Page 20Linked to original sources
PubMed2006

Automatic transcription factor classifier based on functional domain composition.

To understand the transcriptional regulatory mechanism, it is indispensable to identify transcription factors (TF) from the whole genome and to classify transcription factors into different classes. New computational approaches have been developed to identify TFs/non-TFs, and furthermore to classify TFs into four different classes, based on the protein functional domain composition [K.C. Chou, Y.D. Cai, Using functional domain composition and support vector machines for prediction of protein subcellular location, J. Biol. Chem. 277 (2002) 45765-45769]. We trained and tested our method on a non-redundancy dataset consisting of 74 transcription factors collected from TRANSFAC v7.0 [V. Matys, O.V. Kel-Margoulis, E. Fricke, I. Liebich, S. Land, A. Barre-Dirrie, I. Reuter, D. Chekmenev, M. Krull, K. Hornischer, N. Voss, P. Stegmaier, B. Lewicki-Potapov, H. Saxel, A.E. Kel, E. Wingender, TRANSFAC(R) and its module TRANSCompel(R): transcriptional gene regulation in eukaryotes, Nucleic Acids Res. 34 (2006) D108-D110] and 1558 non-transcription factors from UniProtKB/Swiss-Prot Release 49.3 of 21-Mar-2006. The overall success rates of jackknife cross-validation tests reached 98.4% for TF/non-TF identification and 97.2% for classifications of TF classes: basic domains, zinc-coordinating DNA-binding domains, helix-turn-helix, and beta-scaffold factors.

Artificial Intelligence↗
PubMed2025

A functional assay to classify RB1 variants of uncertain significance.

PURPOSE: The RB1 gene encodes the retinoblastoma protein (pRB) playing a major role in cell cycle control, particularly by its interaction with E2F transcription factors. Familial forms of retinoblastoma are caused by germline pathogenic variants in the RB1 gene predisposing to retinoblastoma and other tumors. By analyzing the RB1 gene in patients with retinoblastoma, we found that missense variants often remain variants of uncertain significance (VUS). METHODS: To classify RB1 VUS, we developed a functional assay evaluating their impact on the ability of pRB to inhibit the activity of the E2F1 promoter, with a luciferase reporter gene. A set of 14 pathogenic/likely pathogenic and benign/likely benign RB1 variants was used for validation. RESULTS: We tested 16 VUS detected in patients with retinoblastoma and found that 9 VUS reduced the ability of pRB to inhibit E2F1 promoter. Among them, the (RB1) c.2263T>G p.(Phe755Val) variant showed a reduced level of pRB on Western blot, suggesting a defect in pRB stability. By applying the criterion PS3_moderate of the American College of Medical Genetics and Genomics/Association for Molecular Pathology classification to this functional assay, 5 of the 9 VUS with functional impact could be classified as likely pathogenic. CONCLUSION: This functional assay can improve the molecular diagnosis of retinoblastoma predisposition by a better determination of pathogenic/likely pathogenic RB1 variants.

Humans↗
PubMed2005

Assisting medical annotation in Swiss-Prot using statistical classifiers.

Bio-medical knowledge bases are valuable resources for the research community. Original scientific publications are the main source used to annotate them. Medical annotation in Swiss-Prot is specifically targeted at finding and extracting data about human genetic diseases and polymorphisms. Curators have to scan through hundreds of publications to select the relevant ones. This workload can be greatly reduced by using bio-text mining techniques. Using a combination of natural language processing (NLP) techniques and statistical classifiers, we achieve recall points of up to 84% on the potentially interesting documents and a precision of more than 96% in detecting irrelevant documents. Careful analysis of the document pre-processing chain allows us to measure the impact of some steps on the overall result, as well as test different classifier configurations. The best combination was used to create a prototype of a search and classification tool that is currently tested by the database curators.

Databases, Protein↗
PubMed2007

Proctitis after external-beam radiotherapy for prostate cancer classified by Vienna Rectoscopy Score and correlated with EORTC/RTOG score for late rectal toxicity: results of a prospective multicenter study of 166 patients.

PURPOSE: To evaluate the Vienna Rectoscopy Score (VRS) as a feasible and effective tool for detecting and classifying pathologic changes in the rectal mucosa after radiotherapy (RT) for prostate cancer, and, also, to correlate its findings with the European Organization for Research and Treatment of Cancer (EORTC)/Radiation Therapy Oncology Group (RTOG) score for late rectal toxicity. METHODS AND MATERIALS: A total of 486 patients with localized prostate cancer underwent external-beam RT up to 70 or 74 Gy within an Austrian-German prospective multicenter trial. In 166 patients, voluntary rectal sigmoidoscopy was performed before and at 12 and/or 24 months after RT. Pathologic findings such as telangiectasia, congested mucosa, and ulcers were graded (Grades 0-3) and summarized according to the VRS. Late rectal side effects (EORTC/RTOG) were documented and correlated with the corresponding VRS. RESULTS: Before RT, 99% had a VRS score of 0. The median follow-up was 40 months. Overall, a late rectal side effects grade or score 1-3 was detected in 43% by EORTC/RTOG compared with 68% by VRS (p < 0.05). Grades 0, 1, 2, and 3 late rectal side effects were found using EORTC/RTOG in 57%, 11%, 28%, and 3%, respectively; the corresponding percentages were 32%, 22%, 32%, and 14% for a VRS of 0, 1, 2, and 3, respectively. A significant coherence between the VRS and EORTC/RTOG was found (p < 0.01). CONCLUSIONS: The VRS is a feasible and effective tool for describing and classifying pathologic findings in the rectal mucosa after RT within a multicenter trial. The VRS and EORTC/RTOG showed a high coherence. However the VRS was significantly more sensitive.

Aged↗
PubMed2005

Responsiveness to autologous sweat and serum in cholinergic urticaria classifies its clinical subtypes.

BACKGROUND: It has been reported that patients with cholinergic urticaria have a type 1 allergy to autologous sweat; however, the pathogenesis of that disorder has not been fully elucidated. OBJECTIVE: We investigated the responsiveness to autologous sweat and serum in patients with cholinergic urticaria in relation to their clinical characteristics. We further classified the clinical subtypes that are clearly characterized by responsiveness to in vivo and in vitro tests as well as their clinical features. METHODS: Intradermal tests with autologous sweat and serum were performed in 18 patients with cholinergic urticaria. Histamine release from peripheral blood basophils induced by autologous sweat was measured. RESULTS: Eleven of 17 patients with cholinergic urticaria showed positive reactions in skin tests with their own diluted sweat. Substantial amounts of sweat-induced histamine release from autologous basophils were observed in 10 of 17 patients. Eight of 15 patients with cholinergic urticaria showed positive reactions in the autologous serum skin tests. All 6 patients who developed satellite wheals after the acetylcholine test showed hypersensitivity to sweat. Further, patients whose eruptions were coincident with hair follicles showed positive responses to the skin test with autologous serum, whereas patients whose eruptions were not coincident with hair follicles did not. CONCLUSION: On the basis of these findings, we propose that cholinergic urticaria should be classified into 2 distinct subtypes. The first (nonfollicular) subtype shows strong positive reactions to autologous sweat and negative reactions to autologous serum. The second (follicular) subtype shows weak reactions to autologous sweat and positive reactions to autologous serum.

Acetylcholine↗
PubMed2005

Number versus distribution in classifying regional lymph node metastases from colon cancer.

BACKGROUND: Metastasis to regional lymph nodes from colon cancer is an important prognostic factor. In the TNM classification, node metastases are classified into three grades based on the number of metastatic nodes. In the Japanese General Rules for Clinical and Pathologic Studies on Cancer of the Colon, Rectum, and Anus (JGR), node metastases are classified into four grades based on the distribution of metastatic nodes. STUDY DESIGN: Based on the findings of node metastases in 164 patients with colon cancer obtained by the clearing method, node classifications by the JGR and TNM classifications were compared. RESULTS: The case distribution by the JGR grading was 41.5% in n (-), 29.3% in n1 (+), 18.3% in n2 (+), and 11.0% in n3 (+) disease. In the TNM classification, the distribution was 23.8% in pN1 and 34.8% in pN2 disease. The 5-year survival rate by the JGR was 98.4% in n (-), 74.3% in n1 (+), 51.2% in n2 (+), and 30.0% in n3 (+) disease; in TNM classification, this rate was 76.0% in pN1 and 45.0% in pN2 disease. CONCLUSIONS: In the classification of regional node metastases from colon cancer, the JGR classification showed a wider range in distribution and 5-year survival rate compared with the TNM system.

Cecal Neoplasms↗
PubMed2005

Prospective recruitment of patients with congestive heart failure using an ad-hoc binary classifier.

This paper addresses a very specific problem of identifying patients diagnosed with a specific condition for potential recruitment in a clinical trial or an epidemiological study. We present a simple machine learning method for identifying patients diagnosed with congestive heart failure and other related conditions by automatically classifying clinical notes dictated at Mayo Clinic. This method relies on an automatic classifier trained on comparable amounts of positive and negative samples of clinical notes previously categorized by human experts. The documents are represented as feature vectors, where features are a mix of demographic information as well as single words and concept mappings to MeSH and HICDA classification systems. We compare two simple and efficient classification algorithms (Naïve Bayes and Perceptron) and a baseline term spotting method with respect to their accuracy and recall on positive samples. Depending on the test set, we find that Naïve Bayes yields better recall on positive samples (95 vs. 86%) but worse accuracy than Perceptron (57 vs. 65%). Both algorithms perform better than the baseline with recall on positive samples of 71% and accuracy of 54%.

Artificial Intelligence↗
PubMed2006

LogitBoost classifier for discriminating thermophilic and mesophilic proteins.

A novel classifier, the so-called LogitBoost classifier, was introduced to discriminate the thermophilic and mesophilic proteins according to their primary structures. When the 20-amino acid composition was chosen as the feature vector, the overall accuracy of the self-consistency check and a five-fold cross-validation procedure was 97.0% and 86.6%, respectively. To test if the method was also applicable to a wide range of biological targets, an independent testing dataset was also used. The method based on LogitBoost algorithm has achieved an overall classification accuracy of 88.9%. According to the three different validation check approaches, it was demonstrated that LogitBoost outperformed AdaBoost and performed comparably with RBF neural network and support vector machine. The influence of protein size on discrimination was addressed.

Algorithms↗
PubMed2006

Adjustment to chronic pain in back pain patients classified according to the motivational stages of chronic pain management.

UNLABELLED: According to Prochaska's transtheoretical model, the Freiburg Questionnaire stages of chronic pain management (FQ-STAPM) were used to classify chronic back patients into 4 distinct motivational stages. The FQ-STAMP was completed by 163 chronic back pain patients. Pain chronicity was measured by the Mainz Pain Staging System; pain intensity was measured by the numeric rating scale. Healthcare system expenses were considered as number of consulted physicians, number of stays in hospital, and number of rehabilitation programs. As psychometric tests, the lower pain disability index (PDI), the Hospital Anxiety and Depression Scale (HADS), and a quality of life score (SF36) were used. Patients were in the following motivational stages: precontemplation in 30%, preparation in 19%, action in 30%, maintenance in 21%. The intensity of pain in the precontemplation stage patients was significantly higher compared to patients in the maintenance stage. A lower pain chronicity was related to a significantly higher motivation. Moreover, there was a significant increase in healthcare system expenses by the lesser motivated patients. Patients in the maintenance stage used significantly less opioids than patients in the precontemplation stage. The higher motivated patients had a significantly lower PDI, a significantly lower HADS, and a significantly higher quality of life compared to less motivated patients. PERSPECTIVE: The study indicates that the FQ-STAPM might be a useful tool to classify chronic back pain patients and to work out a strategy together with the patient relevant to the outcome of pain management among chronic back pain patients.

Adaptation, Psychological↗
PubMedDate not supplied

Handling missing values in support vector machine classifiers.

This paper discusses the task of learning a classifier from observed data containing missing values amongst the inputs which are missing completely at random. A non-parametric perspective is adopted by defining a modified risk taking into account the uncertainty of the predicted outputs when missing values are involved. It is shown that this approach generalizes the approach of mean imputation in the linear case and the resulting kernel machine reduces to the standard Support Vector Machine (SVM) when no input values are missing. Furthermore, the method is extended to the multivariate case of fitting additive models using componentwise kernel machines, and an efficient implementation is based on the Least Squares Support Vector Machine (LS-SVM) classifier formulation.

Algorithms↗
PubMedDate not supplied

Classifying mental retardation and specific strength and deficit areas in severe and profoundly mentally retarded persons with the MESSIER.

The MESSIER, a measure of social and communication skills and the Vineland Social Maturity Scale were administered to 618 severe and profoundly mentally retarded adults. The goal of the study was to establish the potential utility of the MESSIER for classifying level of intellectual disability in this group. Comparing MESSIER scores to previously established DSM-IV-TR diagnosis, 86% of the severe and 80% of the profoundly mentally retarded persons were correctly classified. The implications of these data in using the MESSIER for classification and treatment planning are discussed.

Adolescent↗
PubMed2006

Characteristics of COPD phenotypes classified according to the findings of HRCT.

The present study was performed to clarify the clinical characteristics of chronic obstructive pulmonary disease (COPD) patients classified into phenotypes according to the dominancy of emphysema and the presence of bronchial wall thickening evaluated by chest high-resolution computed tomography (HRCT). Eighty-five patients with stable COPD (FEV1 < or = 80%) were examined by chest HRCT. Emphysematous changes and bronchial wall thickening were evaluated visually, and COPD patients were classified into three phenotypes: absence of emphysema, with little emphysema with or without bronchial wall thickening (A phenotype), emphysema without bronchial wall thickening (E phenotype), and emphysema with bronchial wall thickening phenotype (M phenotype). Clinical characteristics were compared among the three phenotypes. The A phenotype group showed a higher prevalence of subjects who had never smoked and patients with wheezing, higher values of body mass index (BMI) and DLco, milder lung hyperinflation, and greater reversibility of airflow limitation responsive to inhaled beta2-agonist as compared with the other phenotypes. The degree of emphysema was significantly associated with Brinkman index, lower BMI, decrease in DLco, lower FEV1/FVC. The presence of bronchial wall thickening in A- and M- phenotype was significantly associated with reversibility responsive to treatment with inhaled corticosteroid and sputum eosinophilia. These findings suggest that the morphological phenotypes of COPD show several clinical characteristics and different responsiveness to treatment with bronchodilators and inhaled corticosteroids.

Administration, Oral↗
PubMed2005

UISESS scale for staging and classifying clinical-epidemiological risk in type 2 diabetes mellitus and for establishing multidisciplinary preventive actions.

PURPOSE: The goal of this study is the creation and use of a five-stage scale for the description of the health status of a type 2 diabetes mellitus patient population, as individuals and as a set. The scale, named UISESS, ranges from the pre-pathological to the sequelae stage, and can help orient multidisciplinary prevention actions. SETTING AND SAMPLING: A sample of nondiabetics (n = 404), together with a census of all type 2 diabetic patients (n = 2,702) over 30 years of age, were obtained from a family medicine clinic of the Mexican Federal Social Security System (IMSS) in Guadalajara, Mexico. DESIGN AND METHOD: A descriptive study in which the Natural History of the Disease and an epidemiological focus on risk were used to create the UISESS scale. Case histories of a total of 3,106 persons from a family practice clinic were used. The data was classified in five stages and three risk control levels were used for analysis. RESULTS: In the type 2 diabetic group, the average age was 57.6 years; 58% were women. In the nondiabetic group, the average age was 55.38 years; 63% were women. The UISESS scale was applied. A large majority (86%) of the studied type 2 diabetic patients were classified within unfavorable stages for controlling their illness, and 65% of the 3,106 people were at high or very high risk of increasing deterioration of their health status. CONCLUSIONS: The UISESS scale appears to be a useful tool to detect the distribution, evolution and control of type 2 diabetes in a population attended at the outpatient clinic of a family medicine unit, according to the Natural History of the Disease.

Adult↗
PubMed1997

A system for classifying mechanical injuries of the eye (globe). The Ocular Trauma Classification Group.

PURPOSE: To develop a classification system for mechanical injuries of the eye. METHODS: The Ocular Trauma Classification Group, a committee of 13 ophthalmologists from seven separate institutions, was organized to discuss the standardization of ocular trauma classification. To develop the classification system, the group reviewed trauma classification systems in ophthalmology and general medicine and, in detail, reports on the characteristics and outcomes of eye trauma, then established a classification system based on standard terminology and features of eye injuries at initial examination that have demonstrated prognostic significance. RESULTS: This system classifies both open-globe and closed-globe injuries according to four separate variables: type of injury, based on the mechanism of injury; grade of injury, defined by visual acuity in the injured eye at initial examination; pupil, defined as the presence or absence of a relative afferent pupillary defect in the injured eye; and zone of injury, based on the anteroposterior extent of the injury. This system is designed to be used by ophthalmologists and nonophthalmologists who care for patients or conduct research on ocular injuries. An ocular injury is classified during the initial examination or at the time of the primary surgical intervention and does not require extraordinary testing. CONCLUSIONS: This classification system will categorize ocular injuries at the time of initial examination. It is designed to promote the use of standard terminology and assessment, with applications to clinical management and research stud ies regarding eye injuries.

Adult↗
PubMedDate not supplied

Comparison of the Milner and Bisiach Landmark Tasks: can neglect patients be classified consistently?

The aim of the presented studies was to investigate whether classifications of neglect patients into perceptual (i.e. identifying a patient as suffering from mainly attentional/space representation deficits) and premotor (judging the main impairment to be related towards actions into contralesional space) categories are consistent across similar Landmark techniques that have, in the past, been designed to tease these potentially overlapping aspects of hemispatial neglect apart. Thirteen patients with hemispatial neglect were tested both with the Landmark Test, adapted from Milner et al. (1992; 1993) in which they had to manually point to the half of a centrally pre-bisected line that, to them, appeared shorter and the motor version of the Bisiach Landmark Test (Bisiach et al., 1998) in which, rather than just judging a centrally prebisected line, they had to judge asymmetrically bisected lines as well. The specific question was whether these two tasks, which are very similar, would categorise the same set of patients in the same way. Most patients could be classified into either the premotor or perceptual category in each task, but no consistent categorisation emerged across the two tests. Just three out of the thirteen patients were consistently classified across both tests. Despite the apparent similarity of the two tests the Milner Landmark Test proved to be much more sensitive to identifying even a slight perceptual bias and seems therefore the test of choice if identification of perceptual bias is the major interest.

Aged↗
PubMed2003

A method for classifying metabolites in topological pathway analyses based on minimization of pathway number.

Metabolic pathway analysis based on the concept of elementary flux mode is a valuable tool for reconstruction of bacterial metabolisms and in predicting optimal conversion yields in biotechnology. However, pathway analysis of large and highly entangled metabolic networks meets the problem of combinatorial explosion of possible routes across the networks. Here we propose a method for coping with this problem by suitably classifying metabolites as external or internal. External metabolites are considered to have buffered concentrations while internal metabolites have to fulfil a balance condition at steady state. For many substances such as nutrients and excreted products, there are biochemical reasons to classify them as external. In addition, other substances (especially at central branching points) can operationally be considered external in order to avoid combinatorial explosion. We suggest to find such a classification of metabolites that minimizes the number of elementary flux modes (pathways). This is motivated by the objectives of finding such a description of the system that reduces as much as possible the amount of necessary data and of removing the ambiguity and arbitrariness in the classification of metabolites in an automated, systematic way. For networks of moderate size, the solution to this combinatorial minimization problem can be found by exhaustive search. To tackle also larger systems, a stochastic optimization program based on the Metropolis algorithm was developed. Both methods are applied, for illustration, to several reaction schemes including a larger network representing glutathione metabolism.

Combinatorial Chemistry Techniques↗
PubMed2003

Classifying lower limb dynamics in Parkinson's disease.

To classify lower limb dynamics in patients with Parkinson's disease (PD), we conducted a clinical study by using pedaling exercise.Twenty-seven patients with idiopathic PD were included in this study. We measured rotational velocities of pedals during pedaling movements with a newly developed ergometer. The velocity waveforms exhibited different characteristics among patients, which could be categorized into four different clusters. In cluster 1, the amplitude on each side was constant and the relative phase was locked at 180 degrees. The pattern was the same as seen in normal subjects. In cluster 2, the amplitude on each side was constant, but the relative phase was locked at 90 degrees. In cluster 3, the amplitude on each side was modulated, and the relative phase drifted monotonously from 0 to 360 degrees during pedaling cycles. In cluster 4, the amplitude on each side was synchronously and irregularly modulated, and the relative phase fluctuated with intermittent spike-like decrement. In order to evaluate, the correlation between pattern and severity of PD, we divided 13 patients, who underwent measurement of pedaling patterns more than three times, into three groups, and found that the abnormal coordination pattern correlated with the presence of freezing phenomenon in patients with PD. Our clinical analysis may contribute in analyzing and classifying the dynamics of PD.

Adult↗
PubMed2000

Implications of physical symmetries in adaptive image classifiers.

It is demonstrated that rotational invariance and reflection symmetry of image classifiers lead to a reduction in the number of free parameters in the classifier. When used in adaptive detectors, e.g. neural networks, this may be used to decrease the number of training samples necessary to learn a given classification task, or to improve generalization of the neural network. Notably, the symmetrization of the detector does not compromise the ability to distinguish objects that break the symmetry.

Artificial Intelligence↗