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 793 records · Page 44Linked to original sources

Ion currents through mutant phospholemman channel molecules.

Phospholemman (PLM), a small membrane protein with a single transmembrane domain, is a major substrate for protein kinases in muscle. In lipid bilayers, PLM forms ion channels with two unusual features--selectivity for the zwitterion taurine, and switching among cation- and anion-selective conformations. To investigate the molecular determinants of channel behavior, we measured ion currents through (1) PLM channels with point mutations in the extracellular and cytoplasmic domains, (2) chimeric PLM channels with the transmembrane domain of IsK, a K channel subunit, and (3) truncated PLM channels lacking the cytoplasmic domain. Truncated channels and those with a mutation at the cytoplasmic face had altered selectivity. Channels with a point mutation near the extracellular face were non-selective. Switching among channel conformations was less frequent in truncated channels and in channels with point mutations in either the extracellular or cytoplasmic domain. Chimeric channels with a different transmembrane domain had only a small reduction in conductance. We conclude that both the extracellular and cytoplasmic domains play roles in channel selectivity and conformational changes. Cation-binding sites appear to lie in the cytoplasmic domain, and an anion-binding site may reside in the extracellular domain. The cytoplasmic domain may function as a "ball and chain" to regulate channel selectivity.

Amino Acid Sequence↗

[Single use insulin injector evaluated by patients (questionnaire study)].

The aim of the study was the analysis of perception of the disposable pen injector HumaJect by diabetic patients. Selected features of the insulin delivery systems were evaluated, and on that basis, the comparison between HumaJect and other insulin injectors was made. Research material was collected in questionnaires filled out by doctors after interviewing patients who were using HumaJect for at least one month. 1802 diabetic individuals aged 9 to 88 participated in the study. HumaJect was ranked "very good" (which is the highest possible rank) by 71% of patients. Among specific features, the highest ranks were assigned to "Ease of dose setting", "Dosing range" and "Disposable form". Most of patients assigned higher ranks to HumaJect than to other insulin delivery systems. 89% of patients indicated desire to continue treatment with this injector.

Adolescent↗

The topography of tactile learning in humans.

The spatial distribution of learned information within a sensory system can shed light on the brain mechanisms of sensory-perceptual learning. It has been argued that tactile memories are stored within a somatotopic framework in monkeys and rats but within a widely distributed network in humans. We have performed experiments to reexamine the spread of tactile learning across the fingertips. In all experiments, subjects were trained to use one fingertip to discriminate between two stimuli. Experiment 1 required identification of vibration frequency, experiment 2 punctate pressure, and experiment 3 surface roughness. After learning to identify the stimuli reliably, subjects were tested with the trained fingertip, its first and second neighbors on the same hand, and the three corresponding fingertips on the opposite hand. As expected, for all stimulus types, subjects showed retention of learning with the trained fingertip. However, the transfer beyond the trained fingertip varied according to the stimulus type. For vibration, learning did not transfer to other fingertips. For both pressure and roughness stimuli, there was limited transfer, dictated by topographic distance; subjects performed well with the first neighbor of the trained finger and with the finger symmetrically opposite the trained one. These results indicate that tactile learning is organized within a somatotopic framework, reconciling the findings in humans with those in other species. The differential distribution of tactile memory according to stimulus type suggests that the information is stored in stimulus-specific somatosensory cortical fields, each characterized by a unique receptive field organization, feature selectivity, and callosal connectivity.

Adult↗

Diagnosis supporting rules of the Hepar system.

The "Hepar" system comprises a clinical database and the shell of procedures that aim at data analysis and the support of diagnosis. The database consists of hepatological patient cases. Each case is described by about 200 medical findings and histopathologically verified diagnosis. The diagnosis supporting rules of "Hepar" are based on visualizing data transformations and on the similarity based techniques. The applied linear visualizing transformations of data sets on the plane aim at separating of the groups of patients associated with different diseases. The resulting diagnostic maps by the visual inspection allow to find such cases in the database that are similar to the previously diagnosed patients. This paper examines combining of data transformations with the nearest neighbors techniques in the support of diagnosis. We report the results on the experimental comparisons of different decision rules including the feature selection procedure.

Decision Support Techniques↗

Boosted decision tree analysis of surface-enhanced laser desorption/ionization mass spectral serum profiles discriminates prostate cancer from noncancer patients.

BACKGROUND: The low specificity of the prostate-specific antigen (PSA) test makes it a poor biomarker for early detection of prostate cancer (PCA). Because single biomarkers most likely will not be found that are expressed by all genetic forms of PCA, we evaluated and developed a proteomic approach for the simultaneous detection and analysis of multiple proteins for the differentiation of PCA from noncancer patients. METHODS: Serum samples from 386 men [197 with PCA, 92 with benign prostatic hyperplasia (BPH), and 96 healthy individuals], randomly divided into training (n = 326) and test (n = 60) sets, were analyzed by surface-enhanced laser desorption/ionization (SELDI) mass spectrometry. The 124 peaks detected by computer analyses were analyzed in the training set by a boosting tree algorithm to develop a classifier for separating PCA from the noncancer groups. The classifier was then challenged with the test set (30 PCA samples, 15 BPH samples, 15 samples from healthy men) to determine the validity and accuracy of the classification system. RESULTS: Two classifiers were developed. The AdaBoost classifier completely separated the PCA from the noncancer samples, achieving 100% sensitivity and specificity. The second classifier, the Boosted Decision Stump Feature Selection classifier, was easier to interpret and used only 21 (compared with 74) peaks and a combination of 21 (vs 500) base classifiers to achieve a sensitivity and specificity of 97% for the test set. CONCLUSIONS: The high sensitivity and specificity achieved in this study provides support of the potential for SELDI, coupled with a bioinformatics learning algorithm, to improve the early detection/diagnosis of PCA.

Algorithms↗

[The application of artificial neural network in QSAR research of PCDD].

The study of QSAR between molecular structure and retention behavior of solutes is an important topic of chromatographic basic theory. PCDD (polychlorinated dioxin) is a kind of important environmental toxin. The basic skeleton of PCDD is [symbol: see text]. PCDDs are all chlorine substituted, so they are only different from each other in the number and position of chlorine substitutents on the skeleton. There are 76 kinds of homologues (1 to 8 chlorine substituent) for 3, 4, 5 and 6 chlorine substituted compounds including more than ten kinds of isomerides. In this paper, the description of molecular structure and feature selection have been investigated. Artificial neural network (ANN) is used to relate the molecular structure and retention time of PCDD. The predicted results of 49 PCDDs retention time by ANN are listed. It was shown that as to the relative error more than 95% data was less than 10% and more than 80% data was less than 5%.

Chromatography↗

Medical recommendations--followed or ignored? Factors influencing compliance in arthritis.

Physiatrists are frequently confronted by the failure of patients to follow medical advice, with resultant adverse effects on rehabilitation outcomes. This paper examines the compliance of arthritis patients with an exercise regimen. A number of factors are examined in this study in assessing elements related to patient compliance. Included are patient attributes and perceptions, professional home visits and form of instruction in the regimen and selected features of the patient's social environment. Age, activity status and marital status, use of medications, patient understanding of arthritis and professional home visits appear to have a bearing upon patient compliance. Careful attention to factors likely to influence patient compliance will prove helpful in efforts designed to enhance rehabilitative progress and, ultimately, improve the lives of patients.

Adult↗

[Texture analysis of ultrasonograms--a new method in the diagnosis of thyroid diseases?].

Ultrasonography is a generally accepted method for diagnosing both the diffuse and focal (nodal) lesions of the thyroid gland. The main limit of this method is the restricted ability of the human eye to analyse all information included. Moreover, the proportion of subjectivity when evaluating the picture is greater than with other imaging methods. Computer texture analysis in combination with automatic classification may prove a potent tool that could enable--immediately after the standard examination--to assign the finding to a particular type of diffuse disorder, with an accuracy of up to 100 percent. From a variety of procedures, which we have tested until now, the best results were obtained with a combination of spatial, co-occurrence and systematically constructed features, selected by a method that is based on the magnitude of classification error. From several tested ways of classification the Bayes's classificator in combination with the criterion of majority was found to be the best approach.

Humans↗

Learning Boolean queries for article quality filtering.

Prior research has shown that Support Vector Machine models have the ability to identify high quality content-specific articles in the domain of internal medicine. These models, though powerful, cannot be used in Boolean search engines nor can the content of the models be verified via human inspection. In this paper, we use decision trees combined with several feature selection methods to generate Boolean query filters for the same domain and task. The resulting trees are generated automatically and exhibit high performance. The trees are understandable, manageable, and able to be validated by humans. The subsequent Boolean queries are sensible and can be readily used as filters by Boolean search engines.

Algorithms↗

Penalized binary regression for gene expression profiling.

OBJECTIVES: A typical bioinformatics task in microarray analysis is the classification of biological samples into two alternative categories. A procedure is needed which, based on the expression levels measured, allows us to compute the probability that a new sample belongs to a certain class. METHODS: For the purpose of classification the statistical approach of binary regression is considered. Highdimensionality and at the same time small sample sizes make it a challenging task. Standard logit or probit regression fails because of condition problems and poor predictive performance. The concepts of frequentist and of Bayesian penalization for binary regression are introduced. A Bayesian interpretation of the penalized log-likelihood is given. Finally the role of cross-validation for regularization and feature selection is discussed. RESULTS: Penalization makes classical binary regression a suitable tool for microarray analysis. We illustrate penalized logit and Bayesian probit regression on a well-known data set and compare the obtained results, also with respect to published results from decision trees. CONCLUSIONS: The frequentist and the Bayesian penalization concept work equally well on the example data, however some method-specific differences can be made out. Moreover the Bayesian approach yields a quantification (posterior probabilities) of the bias due to the constraining assumptions.

Bayes Theorem↗

Quantitative changes in thyroid lymphoid cells as a marker of malignancy.

OBJECTIVE: To investigate the possibility of using morphometric changes in lymphoid cells in thyroid tissue as a marker of malignant tumors. STUDY DESIGN: Morphometry of biopsy material was performed with a semiautomated, computerized image analyzer. RESULTS: Nine features were selected. These features formed the criteria for differentiation between malignant and benign disease. Equations for the calculation of a weighting coefficient for each feature and a diagnostic index were derived. The values for the diagnostic index for the groups with malignant and benign pathology differed significantly. CONCLUSION: One can determine the presence of a malignant tumor in the thyroid gland from the diagnostic index value.

Biomarkers, Tumor↗

Ultrasonographic appearance of the uterus, placenta, fetus, and fetal membranes throughout accurately timed pregnancy in beagles.

Serial ultrasonographic examinations were performed on 8 Beagle bitches from 20 to 60 days pregnant to determine time of first detection, appearance, and sizes of selected features of pregnancy. Gestation was timed from the day of the preovulatory luteinizing hormone surge. Findings related to gestational age were consistent among bitches. Gestational ages at earliest detection of the following features were: chorionic cavity at day 20; placental layers in the uterine wall at day 22 to 24; zonary placenta at day 27 to 30; embryo and heartbeat at day 23 to 25; yolk sac membrane at day 25 to 28; allantoic membrane at day 27 to 31; choroid plexus of the brain at day 31 to 35; fetal movement at day 34 to 36; skeleton at day 33 to 39; bladder and stomach at day 35 to 39; kidney at day 39 to 47; and liver hypoechoic, compared with lung, at day 38 to 42. Extra-fetal structures were measurable from day 20 or 22 through day 48. Chorionic cavity diameter increased from 0.2 +/- 0.0 cm to 3.3 +/- 0.2 cm, outer uterine diameter increased from 0.8 +/- 0.03 cm to 4.8 +/- 0.2 cm, length of chorionic cavity or zonary placenta increased from 0.3 +/- 0.03 cm to 4.9 +/- 0.05 cm, uterine wall thickness increased from 0.3 +/- 0.03 cm to 0.8 +/- 0.01 cm, and placental thickness increased from 0.1 +/- 0.0 cm to 0.5 +/- 0.05 cm. Chorionic cavity diameter, outer uterine diameter, and placental length each increased at a linear rate through day 37, after which time, each had a marked plateau in growth. Of the extra-fetal structures, chorionic cavity diameter was the most accurate for estimation of gestational age. All of the fetal structures studied increased at an increasing (second order) rate. Crown-rump length increased from 0.3 +/- 0.05 cm on day 24 to 9.2 +/- 0.2 cm on day 48. Body diameter increased from 0.2 +/- 0.03 cm on day 24 to 4.6 +/- 0.15 cm on day 60. Head diameter increased from 0.8 +/- 0.05 cm on day 34 to 2.7 +/- 0.04 cm on day 60. Of the fetal structures, head diameter was the most accurate for estimation of gestational age.

Animals↗

[Image segmentation in tongue characterization].

Tongue diagnosis is one of the essential methods of traditional Chinese medical diagnosis. The accuracy of tongue diagnosis can be improved by tongue characterization. Tongue area segmentation and homogeneous regions segmentation in tongue are important contents of preprocess of tongue image. An algorithm based on edge detection and Gradient vector flow (GVF) active contour for tongue area segmentation and another algorithm based on unsupervised segmentation of color-texture for homogeneous regions segmentation in tongue were presented. Totally about 1500 tongue images were collected. Results of tongue area segmentation achieved accuracy rate of 94.3% and results of homogeneous regions segmentation in tongue were approved by traditional Chinese medical experts. The experiments results show robustness of the algorithms. This work establishes solid foundation for feature selecting of Tongue diagnosis.

Algorithms↗

The immunobiology of heat shock proteins.

A cell's reaction to heat shock is one of the most conserved responses among evolutionary diverse species. Although selected features of the response may vary between organisms, the synthesis of heat shock proteins is universal. The induction of hsps is rapid and occurs immediately following exposure to only a few degrees above normal physiological temperature. The array of assigned functions as well as those hypothesized for heat shock proteins imply that hsps are designed to render cells capable of coping with a variety of physiological stresses, including heat and oxidative injury. Interestingly, the gene for the hsp70 is located between the complement and tumor necrosis factor genes within the class III region of the human major histocompatibility complex on the short arm of chromosome 6. Hsp70 proteins have been linked with a protective role during stress and as unfoldase(s) in protein translocation into the endoplasmic reticulum and mitochondria. Moreover, hsp analogs are often presented as antigens in bacterial or protozoal infections. The mapping of hsp70 in the MHC region may have significant implications with regard to immune mediated diseases. Moreover, the relationship of hsps to immunity and cancer represents the beginning of a major link of the role of hsps in human pathology. Finally, with its pleiotropic roles in essential cellular physiology and immune function, we predict that future experiments will define a relationship between hsps and allergy.

Animals↗

Diagnosis of early relapse in ovarian cancer using serum proteomic profiling.

Surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF) mass spectrometry data has been increasingly analyzed for identifying biomarkers to help early detection of the disease. Ovarian cancer commonly recurs at the rate of 75% within a few months or several years later after standard treatment. Since recurrent ovarian cancer is relatively difficult to be diagnosed and small tumors generally respond better to treatment, new methods for the detection of early relapse in ovarian cancer are urgently needed. Here, we propose a new algorithm SVM-MB/RFE (SVM-Markov Blanket/Recursive Feature Elimination) based on SVM-RFE, which identifies biomarkers for predicting the early recurrence of ovarian cancer. In this approach, we first apply t-test for feature pruning and then binning using 5-fold cross validation. Finally, 58 peaks are obtained from 27,000 of the raw data. Such dramatically reduced features relax the computational burden in the next step of our algorithm. We compare the performance of three feature selection algorithms and demonstrate that SVM-MB/RFE outperforms other methods.

Algorithms↗

Alpha- and beta-adrenergic induction of the expression of the uncoupling protein thermogenin in brown adipocytes differentiated in culture.

In order to examine the control of expression of the gene coding for the brown fat specific uncoupling protein thermogenin (UCP), brown fat cells isolated as undifferentiated precursors from the interscapular brown adipose tissue of young mice were grown in culture. In these cells, it was possible by norepinephrine (NE) addition to induce specifically the expression of the UCP gene. The effect of NE was due to activation of transcription. The ability to express the UCP gene was maximal in cells around confluence; cell cultures younger or older than this showed a lower response. The response to NE showed a sharp optimum around 0.1 microM and was linear with time over the 4-h period studied. The presence of insulin or thyroid hormones facilitated the NE response. Pharmacological analysis of the adrenergic response indicated that UCP gene expression could be induced both via beta-receptors (probably beta 3) and via alpha 1-receptors; these effects were synergistic. It was concluded that it is possible to promote these precursor cells to advance to such a state of differentiation that they can demonstrate the selective feature of the brown fat cell, i.e. the ability to express UCP. The expression of the UCP gene is regulated via interacting adrenergic mechanisms.

8-Bromo Cyclic Adenosine Monophosphate↗

Knowledge discovery: Detecting elderly patients with impaired mobility.

Immobility is an important health concern for the elderly patients and healthcare providers who care for the elderly. The purpose of this study was to test a knowledge discovery method to detect elderly patients with impaired mobility in a large clinical dataset. The research method applied an exploratory design and a data mining classification method (cost sensitive Decision Tree J48 from WEKA) to classify patients. Important factors were identified by the Feature Selection method. The Decision Tree algorithm classified patients in the dataset with 65% sensitivity and 72% specificity for a reduced model. The results were evaluated by 10-fold cross validation. Examples of decision rules were also extracted. The study can be applied to classify different health problems in different populations and serves as a foundation for the development of healthcare decision support systems.

Aged↗

[Postnatal cellular pleomorphosis in retinal pigment epithelium].

Selected features of cell morphology of the retinal pigment epithelia (RPE) were compared in newborn (42 eyes) and adult albinotic rats (34 eyes) at the age of 9 months. The RPE of newborn rats was found to consist of nearly homeomorphous, hexagonal, mononucleate, and relatively small cells (monomorphism). Minimal regional differences were found to exist. In adult albinotic rats the density of the RPE cell population was found to be considerably reduced, but the cells and nuclei were generally enlarged. A high percentage of RPE cells was found to contain two nuclei, and as a rule there were always some cells with three, four or even more nuclei. The picture of the cell mosaic was characterized by marked pleomorphism with regionally determined graduation. Pleomorphous changes were more pronounced in the central than in the more peripheral areas. The genesis and dynamics of cellular pleomorphism are also discussed.

Animals↗