Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Feature selection”

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 1,009 records · Page 56Linked to original sources
PubMed1999

Synthesis of a reaction intermediate analogue of biotin-dependent carboxylases via a selective derivatization of biotin.

[formula: see text] An efficient and practical synthesis of 1, a unique reaction intermediate analogue of biotin-dependent carboxylases, is described. The synthesis features a selective acylation of the 1'-N of biotin. Target 1 inhibits the activity of the biotin carboxylase component of acetyl CoA carboxylase. It is the first known biotin-derived inhibitor of biotin carboxylase and should promote new kinetic and structural studies of the biotin-dependent carboxylases.

Avidin↗
PubMed2002

Risk-factor fusion for predicting multifactorial diseases.

A generalized classification methodology is developed to predict the presence or absence of a multifactorial disease from a set of risk factors thought to be correlated with the disease. The methodology includes fusion to combine risk factors into a single feature vector, normalization to overcome the problems associated with fusing features which have different formats and ranges, discrete Karhunen-Loeve transform (DKLT)-based transformation to facilitate parametric classifier development, the selection of features with high interclass separations, and the design of parametric classifiers. The validity of the method is demonstrated by applying it to predict the occurrence of gout from 14 risk factors. Cross-validation evaluations on 96 patients, 48 clinically diagnosed to have gout and 48 diagnosed to not have gout, showed that an average classification accuracy of 75.7% can be obtained. Even more promising is that higher classification accuracies can be achieved through the careful selection of the DKLT transformation matrix which in turn involves selecting design sets that are good representatives of the gout and nongout classes. It is concluded that the generalized methodology developed in this paper is quite effective in predicting multifactorial diseases and can, therefore, assist/support a physician in diagnosing a multifactorial disease.

Diagnosis, Differential↗
PubMed1991

Colposcopes: a critical review.

A multidimensional professional team consisting of a practicing colposcopist, a mechanical engineer, and a medical photographer independently critically evaluated the features of selected colposcopes. Qualitative observations and quantitative measurements were performed on 13 colposcopes representing eight manufacturers and distributors. Optical characteristics, ocular system operation, mobility and stability, relative ease of use, component characteristics, space for instrumentation, and visual illumination, uniformity, and color were comparatively assessed. The specific colposcope information provided may benefit practicing physicians or novice colposcopists who are contemplating purchasing a colposcope. The presented framework of critical feature evaluation may also facilitate the informed selection of other colposcopes not included in this selective, critical, comprehensive review.

Colposcopes↗
PubMed1996

Digital image analysis and identification of eggs from bovine parasitic nematodes.

Computer-assisted microscopy and multivariate statistics were used to establish and evaluate a procedure for identification of bovine strongylid eggs. Ostertagia ostertagi, Cooperia oncophora, Haemonchus placei, Trichostrongylus axei, and Oesophagostomum radiatum eggs were obtained from faeces voided by monospecifically infected calves. Images of single eggs (400 x magnification) were recorded by a CCD camera fitted onto a microscope and digitized on a PC. After separation of eggs from the image background, the pixel (picture element) positions of the egg outline were analysed by algorithms to describe size and shape. A stepwise discriminant analysis was subsequently used to select and rank descriptive features of 4207 eggs according to discriminatory power. Classification criteria were developed by linear discrimination analysis on the basis of selected features, and the criteria evaluated by cross-validation. A maximum average percentage of correct classification of 85.8% resulted when nineteen features were employed in a linear classification criterion. The percentages correct classification for each species were: O. ostertagi 76.3%, C. oncophora 90.8%, O. radiatum 87.8%, H. placei 90.1%, and T. axei 83.8%. Classification based on the five most important features gave an overall correct classification of 81.5%. Images of "unknown' eggs could be identified automatically by the classification criteria after procedural steps performed by PC were linked in a batch program.

Animals↗
PubMedDate not supplied

Evaluating the diagnostic significance of nuclear grooves in thyroid fine needle aspirates with a semiquantitative approach.

OBJECTIVE: To clarify the diagnostic significance of nuclear grooves in thyroid fine needle aspirates, to search for other associated nuclear features and to identify possible diagnostic pitfalls. STUDY DESIGN: In total, 67 fine needle aspiration cases were retrospectively reviewed for selected nuclear features, including grooves, intranuclear cytoplasmic invaginations, enlargement, pleomorphism and status of nucleoli. A semiquantitative method was used to estimate the percentage of nuclear grooves in 5 high-power fields with the most frequent nuclear grooves. RESULTS: Unequivocal intranuclear cytoplasmic invaginations were a specific diagnostic feature of papillary carcinoma. In the absence of intranuclear cytoplasmic invaginations, the presence of > or = 20% nuclear grooves was virtually diagnostic of thyroid neoplasms, predominantly papillary carcinoma. Less than 10% of nuclear grooves virtually eliminated the possibility of papillary carcinoma. However, 10-19% of nuclear grooves may be seen in papillary carcinoma and benign lesions; therefore, such cases fell into the diagnostic category of "atypical cytology" or "indeterminate" cytology. Nuclear enlargement with elongation was a readily identifiable feature usually associated with nuclear grooves. Significant pleomorphism was more commonly seen in papillary carcinoma than in benign lesions. CONCLUSION: A semiquantitative method is helpful when nuclear grooves are encountered in thyroid cytology in the absence of other diagnostic features. It can significantly improve the diagnostic specificity of nuclear grooves for papillary carcinoma.

Biopsy, Needle↗
PubMed2005

Feature extraction from Doppler ultrasound signals for automated diagnostic systems.

This paper presented the assessment of feature extraction methods used in automated diagnosis of arterial diseases. Since classification is more accurate when the pattern is simplified through representation by important features, feature extraction and selection play an important role in classifying systems such as neural networks. Different feature extraction methods were used to obtain feature vectors from ophthalmic and internal carotid arterial Doppler signals. In addition to this, the problem of selecting relevant features among the features available for the purpose of classification of Doppler signals was dealt with. Multilayer perceptron neural networks (MLPNNs) with different inputs (feature vectors) were used for diagnosis of ophthalmic and internal carotid arterial diseases. The assessment of feature extraction methods was performed by taking into consideration of performances of the MLPNNs. The performances of the MLPNNs were evaluated by the convergence rates (number of training epochs) and the total classification accuracies. Finally, some conclusions were drawn concerning the efficiency of discrete wavelet transform as a feature extraction method used for the diagnosis of ophthalmic and internal carotid arterial diseases.

Automation↗
PubMed1975

Features of prognostic significance in solid ovarian teratoma.

The morphology of 20 ovarian tumors reported as solid teratomas is reviewed and correlated with their clinical behavior, in an attempt to select histologic features of prognostic significance. Utilizing a grading system quantitating immature tissue of embryonic origin a tendency was shown for fully mature tumors to follow a benign course, and for those composed predominantly of immature tissues to prove fatal. This method, however, was prognostically unreliable in 16 tumors. An additional but far more important prognostic guide in this series was provided by extra-embryonic tissue of yolk sac origin, which was present in variable amounts and diverse histologic patterns; all teratomas including this component were fatal, frequently with evidence of extra-abdominal metastases. In certain circumstances, immature embryonic teratomas were associated with widespread peritoneal implants, causing death from intestinal obstruction and recurrent ascites. It is important when considering prognosis and therapy to realize the metastatic potential of extra-embryonic elements in teratomas, as distinct from the tendency of immature embryonic components to implant locally and mature into noninvasive deposits, which may be amenable to surgery.

Adolescent↗
PubMed1997

Lack of carboxyl-terminal tyrosine distinguishes the B*2706-bound peptide repertoire from those of B*2704 and other HLA-B27 subtypes associated with ankylosing spondylitis.

B*2704 and B*2706 are two closely related HLA-B27 subtypes, which differ from the common B*2705 by the Asp > Ser77, Val > Glu152, and Ala > Gly211 amino acid changes. In addition, B*2706 differs from B*2704 by the His > Asp114 and Asp > Tyr116 changes. In spite of their similarity B*2704, but not B*2706, was associated to ankylosing spondylitis in a same population. We have carried out pool sequence analyses of the peptides naturally bound to each of these subtypes, and of several individual peptide ligands. B*2704 and B*2706 shared with B*2705, among other features, their selectivity for Arg2 and their allowance for some aliphatic and aromatic C-terminal residues in their bound peptides. The main features that distinguished both subtypes from B*2705 were: 1) their failure to present peptides with C-terminal basic residues, and 2) their allowance for both polar and nonpolar residues at peptide position 3. A major difference between B*2704 and B*2706 was that C-terminal Tyr was prominent among the peptides bound to B*2704, but was not detected among those from B*2706. The use of Tyr as a C-terminal anchor motif is the only functional feature shared by the disease-associated B*2705, B*2702, and B*2704 subtypes that is absent in B*2706. This suggests that the ability of HLA-B27 to present peptides with C-terminal Tyr might be critical for its association to spondyloarthropathy.

Animals↗
PubMed2006

Computer aided detection of clusters of microcalcifications on full field digital mammograms.

We are developing a computer-aided detection (CAD) system to identify microcalcification clusters (MCCs) automatically on full field digital mammograms (FFDMs). The CAD system includes six stages: preprocessing; image enhancement; segmentation of microcalcification candidates; false positive (FP) reduction for individual microcalcifications; regional clustering; and FP reduction for clustered microcalcifications. At the stage of FP reduction for individual microcalcifications, a truncated sum-of-squares error function was used to improve the efficiency and robustness of the training of an artificial neural network in our CAD system for FFDMs. At the stage of FP reduction for clustered microcalcifications, morphological features and features derived from the artificial neural network outputs were extracted from each cluster. Stepwise linear discriminant analysis (LDA) was used to select the features. An LDA classifier was then used to differentiate clustered microcalcifications from FPs. A data set of 96 cases with 192 images was collected at the University of Michigan. This data set contained 96 MCCs, of which 28 clusters were proven by biopsy to be malignant and 68 were proven to be benign. The data set was separated into two independent data sets for training and testing of the CAD system in a cross-validation scheme. When one data set was used to train and validate the convolution neural network (CNN) in our CAD system, the other data set was used to evaluate the detection performance. With the use of a truncated error metric, the training of CNN could be accelerated and the classification performance was improved. The CNN in combination with an LDA classifier could substantially reduce FPs with a small tradeoff in sensitivity. By using the free-response receiver operating characteristic methodology, it was found that our CAD system can achieve a cluster-based sensitivity of 70, 80, and 90 % at 0.21, 0.61, and 1.49 FPs/image, respectively. For case-based performance evaluation, a sensitivity of 70, 80, and 90 % can be achieved at 0.07, 0.17, and 0.65 FPs/image, respectively. We also used a data set of 216 mammograms negative for clustered microcalcifications to further estimate the FP rate of our CAD system. The corresponding FP rates were 0.15, 0.31, and 0.86 FPs/image for cluster-based detection when negative mammograms were used for estimation of FP rates.

Algorithms↗
PubMed1995

Upper gastrointestinal endoscopy in an urban hospital in northern Nigeria: association of presenting features with endoscopic findings.

In 326 fibreoptic upper gastrointestinal (GI) endoscopies performed in Evangel Hospital (Jos, Nigeria), pathology was found in 210 patients, and of a major nature such as peptic ulcer disease or cancer in 129 of these. The three most useful features to predict the presence of major pathology were epigastric tenderness (the single most useful feature), loss of weight and epigastric pain of a burning nature. These features were selected by stepwise discriminant analysis, which also led to the conclusion that the presence of at least two of these three features is an even more powerful predictor of major pathology.

Adult↗
PubMed2004

Attention to features precedes attention to locations in visual search: evidence from electromagnetic brain responses in humans.

Single-unit recordings in macaque extrastriate cortex have shown that attentional selection of nonspatial features can operate in a location-independent manner. Here, we investigated analogous neural correlates at the neural population level in human observers by using simultaneous event-related potential (ERP) and event-related magnetic field (ERMF) recordings. The goals were to determine (1) whether task-relevant features are selected before attention is allocated to the location of the target, and (2) whether this selection reflects the locations of the relevant features. A visual search task was used in which the spatial distribution of nontarget items with attended feature values was varied independently of the location of the target. The presence of task-relevant features in a given location led to a change in ERP/ERMF activity beginning approximately 140 msec after stimulus onset, with a neural origin in the ventral occipito-temporal cortex. This effect was independent of the location of the actual target. This effect was followed by lateralized activity reflecting the allocation of attention to the location of the target (the well known N2pc component), which began at approximately 170 msec poststimulus. Current source localization indicated that the allocation of attention to the location of the target originated in more anterior regions of occipito-temporal cortex anterior than the feature-related effects. These findings suggest that target detection in visual search begins with the detection of task-relevant features, which then allows spatial attention to be allocated to the location of a likely target, which in turn allows the target to be positively identified.

Adult↗
PubMed2006

Learning-based deformable registration of MR brain images.

This paper presents a learning-based method for deformable registration of magnetic resonance (MR) brain images. There are two novelties in the proposed registration method. First, a set of best-scale geometric features are selected for each point in the brain, in order to facilitate correspondence detection during the registration procedure. This is achieved by optimizing an energy function that requires each point to have its best-scale geometric features consistent over the corresponding points in the training samples, and at the same time distinctive from those of nearby points in the neighborhood. Second, the active points used to drive the brain registration are hierarchically selected during the registration procedure, based on their saliency and consistency measures. That is, the image points with salient and consistent features (across different individuals) are considered for the initial registration of two images, while other less salient and consistent points join the registration procedure later. By incorporating these two novel strategies into the framework of the HAMMER registration algorithm, the registration accuracy has been improved according to the results on simulated brain data, and also visible improvement is observed particularly in the cortical regions of real brain data.

Algorithms↗
PubMed2006

Children of currently depressed mothers: a STAR*D ancillary study.

OBJECTIVE: To assess the current and lifetime prevalence of psychiatric disorders among children of currently depressed mothers and to assess the association of clinical features of maternal depression (i.e., severity, chronicity, and clinical features) with child psychopathology. Mothers were participants in the STAR*D (Sequenced Treatment Alternatives to Relieve Depression) multisite trial, designed to compare effectiveness and acceptability of different treatment options for outpatients with non-psychotic major depressive disorder (MDD). METHOD: Treatment-seeking mothers with a current DSM-IV diagnosis of MDD and with at least 1 child 7 to 17 years old were assessed during a major depressive episode (MDE). For each mother, 1 child was assessed (if a mother had more than 1 child, 1 was randomly selected). Maternal features assessed for this study were history of MDEs, severity of current MDE, comorbid conditions, depressive symptom features, and social functioning. Children were assessed for selected psychiatric diagnoses (Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version [K-SADS-PL]), psychopathologic symptoms and social functioning (Child Behavior Checklist), and global functioning (Children's Global Assessment Scale). Data were gathered from December 2001 to April 2004. RESULTS: A large proportion (72%) of mothers were severely depressed (17-item Hamilton Rating Scale for Depression score >/= 22). About a third (34%) of children had a current psychiatric disorder, including disruptive behavior (22%), anxiety (16%), and depressive (10%) disorders. Nearly half (45%) had a lifetime psychiatric disorder, including disruptive behavior (29%), anxiety (20%), and depressive (19%) disorders. Atypical depressive features in the mother were associated with a 3-fold increase in the odds of having a child with depressive (OR = 3.3 [95% CI = 1.2 to 9.5]; p = .02) or anxiety (OR = 2.6 [95% CI = 1.1 to 6.9]; p = .03) disorders. A history of maternal suicide attempts and the presence of comorbid panic disorder with agoraphobia were associated with a 3-fold increase and an 8-fold increase in the odds of depressive disorders in the offspring, respectively. The final model showed significant associations (p </= .05) between the following characteristics of maternal depression and offspring disorders: maternal comorbid panic disorder with agoraphobia and offspring depressive and anxiety disorders, maternal irritable depression and offspring disruptive behavior disorders and any disorder, and maternal substance use disorders and any disorder. CONCLUSIONS: Children of mothers in the midst of a current MDE are at high risk for disruptive behavior and anxiety disorders. The elevated risk of psychopathology among children of depressed mothers may recommend assessment of these children when clinically suggested. Children of depressed mothers with comorbid panic disorder with agoraphobia are at high risk for depressive and anxiety disorders and deserve special attention from clinicians.

Adult↗
PubMed2006

Fructose-mediated stress signaling in the liver: implications for hepatic insulin resistance.

Organisms reprogram metabolic pathways to adapt to changes in nutrient availability. This requires that nutrient-based stimuli are sensed, signals are transmitted, and highly specific responses are engaged. We propose that in the liver, the mitogen-activated protein kinase, c-jun N-terminal kinase (JNK), links excessive nutrient metabolism with impaired insulin regulation of glucose production. The liver, by virtue of its anatomic position and selective regulatory features, buffers and is highly responsive to changes in nutrient delivery. In particular, sugars such as sucrose and fructose uniquely regulate and are selectively metabolized by the liver. We propose that when hepatic fructose uptake exceeds requirements for glycogen and energy (hepatic sugar excess), the JNK-signaling pathway is engaged as part of the adaptive response.

Animals↗
PubMed1998

Glycerol diversifies phage repertoire selections and lowers non-specific phage absorption.

Using a semi-synthetic phage displayed antibody repertoire, isoform-specific and cross-reactive phage-antibodies to eukaryotic elongation factor 1A (eEF1A) have been selected. Enrichment of specific antibodies was found to depend on the presence of glycerol. Further selections against lactate dehydrogenase (LDH) revealed that the dominance of a phage-antibody clone to LDH was inhibited by glycerol, a notable feature for selection strategies where a broad variety of binding clones is desired. The impact of glycerol in distinct steps of the selection protocol was examined and glycerol found to affect certain antibody-antigen interactions. Furthermore, the nonspecific phage binding was lowered by three orders of magnitude at a 20% (v/v) glycerol concentration.

Antibodies↗
PubMed1989

Event-related potentials during memory search and selective attention to letter size and conjunctions of letter size and color.

The analysis of event-related potentials was used in two experiments to investigate the structure of information processing in a task in which subjects selectively attended to letter size (Experiment 1) or a conjunction of letter size and color (Experiment 2) and searched for target letters within the attended stimulus category. The event-related potentials showed that selective attention to letter size resulted in the enhancement of a central N2b component (onset about 200 ms), which was assumed to reflect feature nonspecific orienting of attention. When attention was directed to conjunctions of letter size and color an earlier effect was found (onset about 150 ms) consisting of positivity at the anterior electrodes and negativity at OZ. This earlier effect was assumed to reflect feature-specific selective processing. Although the early effect showed a hierarchical pattern of results, in which the effect of attending to size was contingent on the relevance of the color attribute, the N2b showed a more independent pattern of results, in which the relevance of either the color or the size attribute resulted in an enhancement of this component, independent of the relevance of the other attribute. An increase in the duration of the memory search process resulted in a prolonged negativity with an onset of about 200 ms which was maximal at CZ. In both experiments the initial phase of this negativity was also found in the event-related potentials to the unattended stimulus categories, suggesting that the search process was initiated nonselectively and terminated after the selection cues were identified. Detection of attended target letters resulted in a parietal P3b component. In both experiments there was an earlier effect discriminating targets and nontargets in the range 200-300 ms.

Adult↗
PubMed2006

Discovery of 3-arylpropionic acids as potent agonists of sphingosine-1-phosphate receptor-1 (S1P1) with high selectivity against all other known S1P receptor subtypes.

A series of 3-arylpropionic acids were synthesized as S1P1 receptor agonists. Structure-activity relationship studies on the pendant phenyl ring revealed several structural features offering selectivity of S1P1 binding against S1P2-5. These highly selective S1P1 agonists induced peripheral blood lymphocyte lowering in mice and one of them was found to be efficacious in a rat skin transplantation model, supporting that S1P1 agonism is primarily responsible for the immunosuppressive efficacy observed in preclinical animal models.

Animals↗
PubMed2005

Development of the "Cell Chip": a new in vitro alternative technique for immunotoxicity testing.

Predictive testing of immunotoxicity associated with chemical compounds is complicated and cannot be accomplished with a single test. As most of the existing tests for immunotoxicity employ experimental animals, there is an increasing need for alternative tests in vitro. We have developed a new system for in vitro immunotoxicity testing, which employs changes in cytokine expression observed in vitro as an endpoint indicating potential for perturbation of the immune system in vivo. This system named "fluorescent cell chip" (FCC) is based on a number of genetically modified cell lines that regulate the expression of a transgene coding for fluorescent protein enhanced green fluorescent protein (EGFP) in a similar way as they regulate expression of IL-1beta, IL-2, IL-4, IFN-gamma, IL-10, TNF-alpha, and beta-actin. Morphological and functional features of selected cell lines expressing EGFP under the control of cytokine promotors were compared with maternal cell lines and this comparison showed that critical functional features of the maternal cell lines were preserved in EGFP expressing cells. Two chemicals with known immunotoxic activities, cyclosporine A and potassium tetrachloro-platinate(II), mediated compound-specific pattern of inhibition and activation of reporter gene expression. Thus, the "fluorescent cell chip" has demonstrated potential for application as a predictive screening test for immunomodulatory activities of chemicals. The major advantage of this approach is the possibility to apply this test in high throughput screening of high number of compounds for their well defined biological activity.

Animals↗