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 1,225 records · Page 68Linked to original sources
PubMed2004

Classifying EEG-based motor imagery tasks by means of time-frequency synthesized spatial patterns.

OBJECTIVE: To develop a single trial motor imagery (MI) classification strategy for the brain-computer interface (BCI) applications by using time-frequency synthesis approach to accommodate the individual difference, and using the spatial patterns derived from electroencephalogram (EEG) rhythmic components as the feature description. METHODS: The EEGs are decomposed into a series of frequency bands, and the instantaneous power is represented by the envelop of oscillatory activity, which forms the spatial patterns for a given electrode montage at a time-frequency grid. Time-frequency weights determined by training process are used to synthesize the contributions from the time-frequency domains. RESULTS: The present method was tested in nine human subjects performing left or right hand movement imagery tasks. The overall classification accuracies for nine human subjects were about 80% in the 10-fold cross-validation, without rejecting any trials from the dataset. The loci of MI activity were shown in the spatial topography of differential-mode patterns over the sensorimotor area. CONCLUSIONS: The present method does not contain a priori subject-dependent parameters, and is computationally efficient. The testing results are promising considering the fact that no trials are excluded due to noise or artifact. SIGNIFICANCE: The present method promises to provide a useful alternative as a general purpose classification procedure for MI classification.

Brain Mapping↗
PubMed2004

Classifying G-protein coupled receptors with bagging classification tree.

G-protein coupled receptors (GPCRs) play a key role in different biological processes, such as regulation of growth, death and metabolism of cells. They are major therapeutic targets of numerous prescribed drugs. However, the ligand specificity of many receptors is unknown and there is little structural information available. Bioinformatics may offer one approach to bridge the gap between sequence data and functional knowledge of a receptor. In this paper, we use a bagging classification tree algorithm to predict the type of the receptor based on its amino acid composition. The prediction is performed for GPCR at the sub-family and sub-sub-family level. In a cross-validation test, we achieved an overall predictive accuracy of 91.1% for GPCR sub-family classification, and 82.4% for sub-sub-family classification. These results demonstrate the applicability of this relative simple method and its potential for improving prediction accuracy.

Algorithms↗
PubMed2005

Advances in classification of EEG signals via evolving fuzzy classifiers and dependant multiple HMMs.

Two novel approaches to the problem of brain signals (electroencephalogram (EEG)) classification are introduced in the paper. The first method is based on a modular probabilistic network architecture that employs multiple dependant hidden Markov models (DM-HMM-D) on the input features (channels). The second method, eClass, is based on an on-line evolvable fuzzy rule base of EEG signal prototypes that represent each class and take into consideration the spatial proximity between input signals. Both approaches use supervised learning but differ in their mode of operation. eClass is designed recursively, on-line, and has an evolvable structure, while DM-HMM-D is trained off-line, in a block-based mode, and has a fixed architecture. Both methods have been extensively tested on real EEG data that is recorded during several experimental sessions involving a single female subject who is exposed to mild pain induced by a laser beam. Experimental results illustrate the viability of the proposed approaches and their potential in solving similar classification problems.

Adult↗
PubMed2006

Comparative multiresolution wavelet analysis of ERP spectral bands using an ensemble of classifiers approach for early diagnosis of Alzheimer's disease.

Early diagnosis of Alzheimer's disease (AD) is becoming an increasingly important healthcare concern. Prior approaches analyzing event-related potentials (ERPs) had varying degrees of success, primarily due to smaller study cohorts, and the inherent difficulty of the problem. A new effort using multiresolution analysis of ERPs is described. Distinctions of this study include analyzing a larger cohort, comparing different wavelets and different frequency bands, using ensemble-based decisions and, most importantly, aiming the earliest possible diagnosis of the disease. Surprising yet promising outcomes indicate that ERPs in response to novel sounds of oddball paradigm may be more reliable as a biomarker than the more commonly used responses to target sounds.

Aged↗
PubMed2004

Classifying, diagnosing, and treating the complications of resurfacing the facial skin.

Resurfacing describes a broad range of treatments meant to address visible signs of aging within the skin. This can be achieved by inducing damage in the epidermis, reticular dermis, or even into the papillary dermis. The body's ability to repair that damage is responsible for the result. Complications for a patient can be broadly categorized as infectious,cicatricial, pigmentary, inflammatory (prolonged erythema, contact dermatitis), and systemic. Each of these complications is examined with attention to diagnosis and treatment.

Aged↗
PubMed2006

Sequence and expression of four coral G protein-coupled receptors distinct from all classifiable members of the rhodopsin family.

A measure of the functional importance of G protein-coupled receptors (GPCRs) as signalling molecules is that over seven hundred have been cloned and identified in the human genome alone. Yet few have been characterized in the lower metazoan phyla, especially in the phylum Cnidaria which is well positioned phylogenetically for tracing the early evolution of GPCRs owing to their possession of the first-evolved nervous systems. We report here the cloning and characterization of four novel rhodopsin-like GPCR cDNAs from the staghorn coral Acropora millepora that share significant similarity with each other but not with the majority of other members of the rhodopsin alpha subfamily. The deduced proteins lack many of the conserved residues and motifs that form the signature of the different groups of alpha rhodopsin receptors. Maximum likelihood phylogenetic analysis likewise implies that the coral receptors do not have a simple or close relationship with any of the major groups within the alpha rhodopsin subfamily. In situ hybridization revealed transcripts in endodermal cells of planula larvae of all ages and in post-settlement polyps. These GPCRs appear to belong to a alpha rhodopsin-like group unique to corals. Comparisons with other cnidarian GPCRs suggest also that GPCRs diverged early in metazoan evolution.

Amino Acid Sequence↗
PubMed2005

County level socioeconomic position, work organization and depression disorder: a repeated measures cross-classified multilevel analysis of low-income nursing home workers.

UNLABELLED: This study simultaneously tests the effect of county, organizational, workplace, and individual level variables on depressive disorders among low-income nursing assistants employed in US nursing homes. A total of 482 observations are used from two waves of survey data collection, with an average two-year interval between initial and follow-up surveys. The overall response rate was 62 percent. The hierarchically structured data was analyzed using multilevel modeling to account for cross-classifications across levels of data. Nursing assistants working in nursing homes covered by a single union in three states were asked about aspects of their working conditions, job stress, physical and mental health status, individual and family health-care needs, household economics and household strain. PARTICIPANTS: The 241 nursing assistants who participated in this study were employed in 34 nursing homes and lived in 49 counties of West Virginia, Ohio and Kentucky. MAIN RESULTS: The study finds that emotional strain, related to providing direct care to elderly and disabled clients, is associated with depressive disorder, as is nursing home ownership type (for-profit versus not-for-profit). However, when controlling for county level socioeconomic variables (Gini index and proportion of African Americans living in the county), neither workplace nor organizational level variables were found to be statistically significant associated with depressive disorder. CONCLUSIONS: This study supports previous findings that emotional demand in health-care environments is an important correlate of mental health. It also adds empirical evidence to support a link between financial strain and depression in US women. While this study does not find that lack of a seniority wage benefits--a factor that can conceivably exacerbate financial strain over time--is associated with depressive disorder among low-income health-care workers, it does find county level measures of poverty to be statistically significant predictors of depressive disorder. Longitudinal county level measures of low-income as predictors of depression may even offer a methodological advantage in that they are presumably more stable indicators of cumulative exposure of low income than are more transient workplace indicators. Incorporating measures of cumulative exposure to low income into empirical studies would be particularly timely given the global changes that are currently restructuring the labor force and influencing work organization and labor processes--most notably the growth in low income jobs and the deskilling of labor. Though this study provides evidence that workplace and organizational level variables are associated with depressive disorder among low-wage nursing assistants in US nursing homes, the fact that these relationships do not hold once county level measures of poverty are controlled for, suggests that more distal upstream determinants of workplace mental health problems, such economic inequality, may be at play in determining the mental health of low wage workers.

Adult↗
PubMed2004

Absence of Epstein-Barr virus in anaplastic large cell lymphoma: a study of 64 cases classified according to World Health Organization criteria.

The frequency of Epstein-Barr virus (EBV) in anaplastic large cell lymphoma (ALCL) has been controversial. The interpretation of previous studies is complicated by the use of nonuniform EBV detection methods and the inclusion of cases of CD30-positive diffuse large B-cell lymphoma and so-called "ALCL, Hodgkin-like," as defined in the Revised European-American Lymphoma classification scheme. In the current World Health Organization (WHO) classification system, both of these tumors are excluded from the ALCL category. Also, recently developed antibodies (eg, the antibody specific for PAX-5/B-cell-specific activator protein [BSAP]) provide new, sensitive tools for identifying neoplasms of B-cell lineage that can morphologically resemble ALCL. In this study we evaluated 64 cases of ALCL of T- or null-cell lineage, defined according to the WHO classification system, for the presence of EBV. All tumors were negative for B-cell antigens, including PAX-5/BSAP and CD20 or CD79a. The study group included 27 (42%) anaplastic lymphoma kinase (ALK)-positive (18 T-cell and 9 null-cell) and 37 (58%) ALK-negative (30 T-cell and 7 null-cell) tumors analyzed by in situ hybridization for EBV-encoded RNA (EBER) or immunohistochemistry for EBV-latent membrane protein type 1. All 64 cases were negative for EBV. We conclude, based on the current definition of ALCL in the WHO classification, there is no role for EBV in ALCL arising in Western patients. We suggest that published reports of EBV in a small proportion of ALCL cases in Western patients can be explained by the inclusion of tumors no longer considered to be in the current classification of ALCL, such as CD30-positive anaplastic tumors of B-cell origin.

Anaplastic Lymphoma Kinase↗
PubMed2004

Feasibility studies of utilizing disk intrinsic dissolution rate to classify drugs.

The purpose of this report was to investigate the feasibility of using disk intrinsic dissolution rate (DIDR) to determine solubility class membership. We employed a VanKel dissolution apparatus fitted with a Wood's intrinsic dissolution die. To test the robustness of the method, variations of DIDR with compression force, dissolution volume, distance of the drug disk from the bottom of the dissolution vessel, and drug disk rotation speed were studied using furosemide and metoprolol in pH 4.5 acetate buffer as a model system. The DIDRs of six low solubility and nine high solubility model drugs were then determined at pH 1.2, 4.5, and 6.8 and compared to their BCS solubility class membership. It was found that the compression force, dissolution medium volume, and die position had no significant effect on DIDR for the system studied. The proposed compression force, dissolution volume, die position, and rotation speed are 2000 psi, 900 ml, 0.5 in., and 100 rpm, respectively. The test results obtained from 15 model BCS drugs show a good relationship between the DIDR and BCS solubility classification with 0.1 mg/min/cm(2) as a class boundary unless the dose is either extremely low or high where discrepancies may exist between the solubility and DIDR methods. Therefore, more scientific research and debates are needed before considered for regulatory purpose.

Compressive Strength↗
PubMed2006

Accuracy of 64-slice computed tomography to classify and quantify plaque volumes in the proximal coronary system: a comparative study using intravascular ultrasound.

OBJECTIVES: We evaluated the accuracy of a new 64-slice computed tomography (CT) scanner, compared with intravascular ultrasound, to visualize atherosclerosis in the proximal coronary system. BACKGROUND: Noninvasive determination of plaque composition and plaque burden may be important to improve risk stratification. METHODS: In 20 patients, a 64-slice CT scan (Sensation 64, Siemens Medical Solutions, Forchheim, Germany) and an intravascular ultrasound investigation of vessels without stenosis >50% was performed. Diagnostic image quality with 64-slice CT was obtained in 36 vessels in 19 patients. RESULTS: In these vessels, which were divided in 3-mm sections, 64-slice CT enabled a correct detection of plaque in 54 of 65 (83%) sections containing noncalcified plaques, 50 of 53 (94%) sections containing mixed plaques, and 41 of 43 (95%) sections containing calcified plaques. In 192 of 204 (94%) sections, atherosclerotic lesions were excluded correctly. In addition, 64-slice CT enabled the visualization of 7 of 10 (70%) sections revealing a lipid pool and could identify a spotty calcification pattern in 27 of 30 (90%) sections. The correlation coefficient to determine plaque volumes per vessel was r2 = 0.69 (p < 0.001) with an underestimation of mixed and noncalcified plaque volumes (p < 0.03) and a trend to overestimate calcified plaque volumes by 64-slice CT. The interobserver variability to determine plaque volumes was 37%. Interobserver agreement to identify atherosclerotic sections was good (Cohen's kappa coefficient = 0.75). CONCLUSIONS: We conclude that 64-slice CT reveals encouraging results to noninvasively detect different types of coronary plaques located in the proximal coronary system. The ability to determine plaque burden currently is hampered by mainly an insufficient reproducibility.

Coronary Angiography↗
PubMed2006

Using support vector machines to optimally classify rotator cuff strength data and quantify post-operative strength in rotator cuff tear patients.

Shoulder strength data are important for post-operative assessment of shoulder function and have been used in diagnosis of rotator cuff pathology. Support vector machines (SVM) employ complex analysis techniques to solve classification and regression problems. A SVM, a machine learning technique, can be used for analysis and classification of shoulder strength data. The goals of this study were to determine the diagnostic competency of SVM based on shoulder strength data and to apply SVM analysis in efforts to derive a single representative shoulder strength score. Data were taken from fourteen isometric shoulder strength measurements of each shoulder (involved and uninvolved) in 45 rotator cuff tear patients. SVM diagnostic proficiency was found to be comparable to reported ultrasound values. Improvement of shoulder function was accurately represented by a single score in pairwise comparison of the pre-operative and the 12 month post-operative group (P < 0.004). Thus, the SVM-based score may be a promising metric for summarizing rotator cuff strength data.

Artificial Intelligence↗
PubMed2005

A multicenter study of the revised Edmonton Staging System for classifying cancer pain in advanced cancer patients.

The comparative analysis of analgesic interventions for cancer pain is greatly compromised by the lack of well-validated and clinically acceptable tools, which allow a composite classification of pain and patient population characteristics. Although the Edmonton Staging System (ESS) for cancer pain was developed for this purpose, clinical and research utility has been limited due to problems associated with the assessment of some items, especially in relation to definitions and terminology. To overcome these limitations, we designed a revised ESS (rESS) and conducted a multicenter study to determine its inter-rater reliability and predictive value. In revising the rESS, we hypothesized that patients with less problematic pain features would require a shorter time to achieve stable pain control, require less complicated analgesic regimens, be more responsive to opioid therapy, and use lower opioid doses. The rESS items include mechanism of pain, presence or absence of incidental pain, presence or absence of psychological distress and addictive behavior, and level of cognitive function. Patients with cancer pain who were consecutively admitted to two different hospice centers, an acute care consultation service in a teaching hospital or a tertiary palliative care unit in a second teaching hospital were evaluated for study entry. Two independent palliative care specialists completed the rESS where possible within 24 hours of each other. Patients' pain ratings and opioid consumption were recorded daily until the study endpoint (i.e. achievement of stable pain control, discharge or death). Seven hundred and forty-six patients were eligible for study entry and of these, 619 (83%) had a pain syndrome. Inter-rater reliability estimates ranged from 0.67 (pain mechanism) to 0.95 (presence of addiction). In the univariate Cox regression analysis, younger patients (<60), as well as patients with neuropathic pain, incidental pain, psychological distress, or co-morbid psychological distress and addiction, required a significantly longer time to achieve stable pain control (P<0.05). In the multivariate Cox regression analysis, only age (<60), neuropathic pain and incidental pain were significantly associated with time to reach stable pain control (P<or=0.05). Patients with neuropathic or incidental pain used significantly more modalities to achieve stable pain control (P<0.01). Patients with neuropathic pain, incidental pain, as well as the presence of psychological distress or addiction, required a higher final mean morphine equivalent daily dose (MEDD) (P<0.001). A comparison of the rESS with the ESS demonstrated the ineffectiveness of the ESS prognostic staging system for predicting achievement of stable pain control. These findings confirm the study hypothesis, suggesting that the rESS appears to have good predictive value and a moderate to high inter-rater reliability. We suggest the rESS should prove to be a useful tool in clinical practice, and in the comparison of cancer pain populations in research studies.

Aged↗