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 883 records · Page 49Linked to original sources
PubMedDate not supplied

[The selectivity, polarity and characteristic features of autistic and syntonic schizophrenia].

In previous papers (3, 4, 5) the author presented a hypothesis concerning important features of clinical picture of paranoid schizophrenia, that is selectivity and polarity (bipolarity). Numerous research works aimed at closer recognition of some differences of clinical picture of paranoid schizophrenia have been conducted under the author's direction. The formation of certain psychopathological proportions discerned by the author has been researched with intentionally elaborated scales. The research results have led to distinguishing autistic and syntonic schizophrenia.

Autistic Disorder↗
PubMed2001

The effects of selection for larval behavior on adult life-history features in Drosophila melanogaster.

Selection for late-life fecundity and longevity in adult Drosophila melanogaster is well known to modify numerous characteristics of life history and physiology. We report experiments here in which selection applied to behavior affects features in an identical fashion. Selection for feeding rate of larval D. melanogaster modifies caloric intake, as measured by the uptake and incorporation of labeled glucose. Selection for slow larval feeding produced lines of D. melanogaster in which larvae synthesized significantly less lipid prior to pupation and eclosed to have low early-life fecundity and a long life as adults. They also had greater lifetime fecundity, but lower viability of egg to hatched adult. Alternatively, fast-feeding larvae incorporated more lipid before pupation and eclosed with high early-fecundity that declined rapidly throughout their short adult life. Slow-feeding populations also had a significantly enhanced expression of the stress-resistance genes CuZn-SOD, CATALASE, and HSP70. Selection on larval feeding behavior reproduced the antagonistic evolutionary trade-off found under selection for adult life span and mimicked the physiological response in life span as seen in many species when dietary restriction is imposed on adults. Thus, nutrient acquisition during development appears to share a common evolutionary and genetic basis with the allocation processes that determine adult life-history traits and the related phenotypic dietary restriction phenomena.

Animals↗
PubMed2006

Novel unsupervised feature filtering of biological data.

MOTIVATION: Many methods have been developed for selecting small informative feature subsets in large noisy data. However, unsupervised methods are scarce. Examples are using the variance of data collected for each feature, or the projection of the feature on the first principal component. We propose a novel unsupervised criterion, based on SVD-entropy, selecting a feature according to its contribution to the entropy (CE) calculated on a leave-one-out basis. This can be implemented in four ways: simple ranking according to CE values (SR); forward selection by accumulating features according to which set produces highest entropy (FS1); forward selection by accumulating features through the choice of the best CE out of the remaining ones (FS2); backward elimination (BE) of features with the lowest CE. RESULTS: We apply our methods to different benchmarks. In each case we evaluate the success of clustering the data in the selected feature spaces, by measuring Jaccard scores with respect to known classifications. We demonstrate that feature filtering according to CE outperforms the variance method and gene-shaving. There are cases where the analysis, based on a small set of selected features, outperforms the best score reported when all information was used. Our method calls for an optimal size of the relevant feature set. This turns out to be just a few percents of the number of genes in the two Leukemia datasets that we have analyzed. Moreover, the most favored selected genes turn out to have significant GO enrichment in relevant cellular processes.

Artificial Intelligence↗
PubMed1993

Inattention and the perception of visual features.

Subjects selectively attended to one of two interleaved, novel figures while ignoring the other figure. In subsequent tests administered to determine the extent to which the ignored figure was perceived, recognition of shape and the location of contour gaps was at the chance level. Moreover, recognition of the presence of contour gaps was significantly below the chance level. These results indicate that preattentive visual processing of unattended objects is too crude to encode global shape and local features such as contour gaps. It is suggested that preattentive processing creates visual representations of unattended objects that contain very limited information about features.

Adult↗
PubMed1998

Color selection and location selection in ERPs: differences, similarities and 'neural specificity'.

It was hypothesized that color selection consists of two stages. The first stage represents a feature specific selection in neural populations specialized in processing color. The second stage constitutes feature non-specific selections, related to executive attentional processes and/or motor processes. This hypothesis was tested by investigating the effects of selectively attending to a specific color, location, or conjunction of location and color on the ERPs elicited by briefly flashed gratings. The gratings differed on three dimensions: color (red or blue), location in the visual field (4.4 degrees to the left or right of fixation) and form (target or non-target). Subjects had to respond to the presentation of target gratings in the attended category. Color selection was reflected in an enhanced parietal positivity in the 150-190 ms interval. Source analyses suggested that this color selection positivity might be generated in the basal occipital cortex, possibly human V4, an area of the brain specialized in color processing. The effect was separated from the P1 spatial attention effect both in topography and sources. Color selection was also reflected in a contralateral occipitotemporal negativity, which resembled the N1 spatial attention effect both in timing and topography. And finally, color selection was reflected in an N2b component. This N2b was similar in timing, topography and sources to the N2b's elicited by location selection and conjunction selection. We suggested that the N2b reflects feature non-specific selection processes, elicited by a range of attended stimuli, and possibly reflects activity in the anterior cingulate cortex. The NP80 was unaffected by attention to color and/or location and localized in striate cortex.

Adult↗
PubMed2006

Online control of a brain-computer interface using phase synchronization.

Currently, almost all brain-computer interfaces (BCIs) ignore the relationship between phases of electroencephalographic signals detected from different recording sites (i.e., electrodes). The vast majority of BCI systems rely on feature vectors derived from e.g., bandpower or univariate adaptive autoregressive (AAR) parameters. However, ample evidence suggests that additional information is obtained by quantifying the relationship between signals of single electrodes, which might provide innovative features for future BCI systems. This paper investigates one method to extract the degree of phase synchronization between two electroencephalogram (EEG) signals by calculating the so-called phase locking value (PLV). In our offline study, several PLV-based features were acquired and the optimal feature set was selected for each subject individually by a feature selection algorithm. The online sessions with three trained subjects revealed that all subjects were able to control three mental states (motor imagery of left hand, right hand, and foot, respectively) with single-trial accuracies between 60% and 66.7% (33% would be expected by chance) throughout the whole session.

Adolescent↗
PubMedDate not supplied

Characterizing bitterness: identification of key structural features and development of a classification model.

This work describes the first approach in the development of a comprehensive classification method for bitterness of small molecules. The data set comprises 649 bitter and 13 530 randomly selected molecules from the MDL Drug Data Repository (MDDR) which are analyzed by circular fingerprints (MOLPRINT 2D) and information-gain feature selection. The feature selection proposes substructural features which are statistically correlated to bitterness. Classification is performed on the selected features via a naïve Bayes classifier. The substructural features upon which the classification is based are able to discriminate between bitter and random compounds, and thus we propose they are also functionally responsible for causing the bitter taste. Such substructures include various sugar moieties as well as highly branched carbon scaffolds. Cynaropicrine contains a number of the substructural features found to be statistically associated with bitterness and thus was correctly predicted to be bitter by our model. Alternatively, both promethazine and saccharin contain fewer of these substructural features, and thus the bitterness in these compounds was not identified. Two different classes of bitter compounds were identified, namely those which are larger and contain mainly oxygen and carbon and often sugar moieties, and those which are rather smaller and contain additional nitrogen and/or sulfur fragments. The classifier is able to predict 72.1% of the bitter compounds. Feature selection reduces the number of false-positives while also increasing the number of false negatives to 69.5% of bitter compounds correctly predicted. Overall, the method presented here presents both one of the largest databases of bitter compounds presently available as well as a relatively reliable classification method.

Algorithms↗
PubMedDate not supplied

Segmental autotransplantation of the distal pancreas to the thigh: CT and ultrasound features.

In select patients with pancreatitis, pancreatectomy may be the only alternative for treatment of abdominal pain. Segmental autotransplantation of the distal pancreas to the thigh has been shown to be successful in preventing or reducing the severity of diabetes following pancreatectomy. We present the postoperative anatomy and potential complications identified on cross-sectional imaging.

Chronic Disease↗
PubMed2002

How spiking neurons give rise to a temporal-feature map: from synaptic plasticity to axonal selection.

A temporal-feature map is a topographic neuronal representation of temporal attributes of phenomena or objects that occur in the outside world. We explain the evolution of such maps by means of a spike-based Hebbian learning rule in conjunction with a presynaptically unspecific contribution in that, if a synapse changes, then all other synapses connected to the same axon change by a small fraction as well. The learning equation is solved for the case of an array of Poisson neurons. We discuss the evolution of a temporal-feature map and the synchronization of the single cells' synaptic structures, in dependence upon the strength of presynaptic unspecific learning. We also give an upper bound for the magnitude of the presynaptic interaction by estimating its impact on the noise level of synaptic growth. Finally, we compare the results with those obtained from a learning equation for nonlinear neurons and show that synaptic structure formation may profit from the nonlinearity.

Journal Article↗
PubMed2003

An in-silico method for prediction of polyadenylation signals in human sequences.

This paper presents a machine learning method to predict polyadenylation signals (PASes) in human DNA and mRNA sequences by analysing features around them. This method consists of three sequential steps of feature manipulation: generation, selection and integration of features. In the first step, new features are generated using k-gram nucleotide acid or amino acid patterns. In the second step, a number of important features are selected by an entropy-based algorithm. In the third step, support vector machines are employed to recognize true PASes from a large number of candidates. Our study shows that true PASes in DNA and mRNA sequences can be characterized by different features, and also shows that both upstream and downstream sequence elements are important for recognizing PASes from DNA sequences. We tested our method on several public data sets as well as our own extracted data sets. In most cases, we achieved better validation results than those reported previously on the same data sets. The important motifs observed are highly consistent with those reported in literature.

Base Sequence↗
PubMed1990

Selection criteria for surgery--clinical features and EEG including invasive techniques.

Preconditions, selection criteria for surgical treatment of epilepsy and some aspects of the presurgical evaluation are discussed. Identification of those patients who will profit from epilepsy surgery, exact localization of the seizure generating structures, and evaluation of the potential risks of the proposed operation are the main goals of the preoperative diagnostic procedures. The localizational value of certain seizure patterns and ictal clinical signs and symptoms is reviewed and illustrated with symptoms provoked by electrical stimulation of the brain.

Brain Mapping↗
PubMedDate not supplied

Responses of directionally selective neurons of the frog tectum mesencephali, their adaptation and extrapolation features.

Tectal directionally selective neurons were investigated in frogs. The highest measure of agreement in the character of response and dynamics of latent periods of neuron pairs was found with interelectrode distances of up to 300 mu, a fact which is evidently connected with the dimensions of the zones of ramification of the afferent fibers. Adaptation and extrapolation properties were discovered for most neurons and they were more marked in the nuclear layers of the tectum than in the fibrous layers. The more stable responses of neurons in the fibrous layers indicate the possible representation there of branches of fibers of the retinal ganglion cells, whereas neurons with high adaptivity (novelty neurons) are located in the nuclear layers.

Adaptation, Physiological↗
PubMed2005

Competition, predation, and the evolutionary significance of the cercopithecine cheek pouch: the case of Cercopithecus and Lophocebus.

Reports of cercopithecine cheek pouch use and functional significance are largely anecdotal, and to date there have been no investigations into its use by species living in closed forest habitats. Here, I report on cheek pouch use in Cercopithecus ascanius and Lophocebus albigena in the Kibale National Park, Uganda, between July-October 1997. Two hypotheses were evaluated: this feature was selected for because of its role in 1) increasing feeding efficiency via a reduction in potential feeding competition, and/or 2) reducing vulnerability to predation. Results indicate that both species were more likely to use their cheek pouches when feeding on contestable foods and, after filling cheek pouches, retreated to more densely vegetated ("safer") positions for processing food. There was no influence of age and sex on L. albigena cheek pouch use. Subadult C. ascanius cheek-pouched less frequently than adults, although there were no differences between adult males and females. There was no relationship between feeding-patch size and number of plant food items cheek-pouched in either species. However, the diameter of breast height (dbh; a measure of patch size) of trees in which C. ascanius used their cheek pouches was significantly larger than the dbh of trees in which they did not. Both species were more likely to use their cheek pouches in the presence of greater numbers of conspecifics. These data provide insight into the relationship(s) among oral anatomy and feeding efficiency, and facilitate understanding into the selection for this important oral feature in stem cercopithecines.

Adaptation, Physiological↗
PubMed2006

Separation of isomeric 2-aminopyridine derivatized N-glycans and N-glycopeptides of human serum immunoglobulin G by using a zwitterionic type of hydrophilic-interaction chromatography.

Isomeric oligosaccharides and isomeric glycopeptides are sometimes difficult to separate on normal-phase (NP) and reversed-phase (RP) columns. A zwitterionic type of hydrophilic-interaction chromatography column with sulfobetaine groups (called ZIC-HILIC column) was first applied to the separation of 2-aminopyridine derivatized (PA) N-glycans and tryptic peptides of human serum immunoglobulin G (IgG). It is shown that the ZIC-HILIC column has high capability for structural recognition of isomeric N-glycans as well as high selectivity for glycopeptides. The former feature (i.e., structural recognition) was proven by sufficient separation of neutral PA N-glycan isomers, which are usually difficult to separate on NP and RP columns. In addition, it is noteworthy that IgG glycopeptides consisting of isomeric N-glycans and the same peptide sequences can be sufficiently separated on a ZIC-HILIC column. The latter feature (i.e., selectivity) was also demonstrated by easily separating two peptide groups with/without N-glycans. Thus, we note that the ZIC-HILIC column is highly promising for a simple analysis of N-glycans and N-glycopeptide samples.

Aminopyridines↗
PubMed2002

An outcome prediction model for patients with clear cell renal cell carcinoma treated with radical nephrectomy based on tumor stage, size, grade and necrosis: the SSIGN score.

PURPOSE: Currently outcome prediction in renal cell carcinoma is largely based on pathological stage and tumor grade. We developed an outcome prediction model for patients treated with radical nephrectomy for clear cell renal cell carcinoma, which was based on all available clinical and pathological features significantly associated with death from renal cell carcinoma. MATERIALS AND METHODS: We identified 1,801 adult patients with unilateral clear cell renal cell carcinoma treated with radical nephrectomy between 1970 and 1998. Clinical features examined included age, sex, smoking history, and signs and symptoms at presentation. Pathological features examined included 1997 TNM stage, tumor size, nuclear grade, histological tumor necrosis, sarcomatoid component, cystic architecture, multifocality and surgical margin status. Cancer specific survival was estimated using the Kaplan-Meier method. Cox proportional hazards regression models were used to test associations between features studied and outcome. The selection of features included in the multivariate model was validated using bootstrap methodology. RESULTS: Mean followup was 9.7 years (range 0.1 to 31). Estimated cancer specific survival rates at 1, 3, 5, 7 and 10 years were 86.6%, 74.0%, 68.7%, 63.8% and 60.0%, respectively. Several features were multivariately associated with death from clear cell renal cell carcinoma, including 1997 TNM stage (p <0.001), tumor size 5 cm. or greater (p <0.001), nuclear grade (p <0.001) and histological tumor necrosis (p <0.001). CONCLUSIONS: In patients with clear cell renal cell carcinoma 1997 TNM stage, tumor size, nuclear grade and histological tumor necrosis were significantly associated with cancer specific survival. We present a scoring system based on these features that can be used to predict outcome.

Adult↗
PubMed2003

DNA cytometric features in biopsies of TaT1 urothelial cell cancer predict recurrence and stage progression more accurately than stage, grade, or treatment modality.

OBJECTIVES: To compare retrospectively the predictive value for recurrence and stage progression of DNA ploidy and S-phase fraction by flow cytometry and highly automated ultrafast image cytometry (ICM) in biopsies of TaT1 urothelial cell carcinomas (UCCs) of the urinary bladder with stage, grade, other pathologic features, and treatment. METHODS: Three experienced pathologists reviewed the stage and grade of 228 UCCs; 193 (85%) consensus cases were analyzed further. We had enough material for single-cell suspensions for both flow cytometry and ICM in 183 cases (94.8%). The 2001 European Society for Analytical Cellular Pathology standards for DNA ICM were followed. The predictive value of DNA features, classic prognosticators (stage, grade, carcinoma in situ, multicentricity), and treatment modality for recurrence and stage progression were analyzed with univariate (Kaplan-Meier) survival and multivariate (Cox model) regression analysis. Ta and T1 cases were analyzed separately. RESULTS: Of the 228 cases, 88 (51.5%) recurred and 13 (7.6%) progressed. On univariate analysis, most of the DNA features studied were statistically significant. Treatment modality and grade were only prognostic for progression (not for recurrence) and only in Ta cases. On multivariate analysis, DNA ICM features performed best; the strongest recurrence predictor for Ta UCC was a DNA index (DI) of 1.0 versus all others, and for T1 UCC, a DI of less than 1.3 versus 1.3 or greater. The best stage progression predictor for Ta UCCs was a DI of 1.0 plus an S-phase fraction of less than 10%, and for T1 UCCs, a DI of less than 1.3 versus 1.3 or greater. With multivariate analysis, sex, age, grade, carcinoma in situ, multicentricity, and treatment modality were excluded once the DNA ICM features were selected. CONCLUSIONS: DNA image cytometric features predict recurrence and stage progression in TaT1 UCC biopsies more accurately than classic prognostic factors, independent of treatment modality.

Aged↗
PubMed2006

A rule-based emotion-dependent feature extraction method for emotion analysis from speech.

This paper presents a rule-based method to determine emotion-dependent features, which are defined from high-level features derived from the statistical measurements of prosodic parameters of speech. Emotion-dependent features are selected from high-level features using extraction rules. The ratio of emotional expression similarity between two speakers is defined by calculating the number and values of the emotion-dependent features that are present for the two speakers being compared. Emotional speech from Interface databases is used for evaluation of the proposed method, which was used to analyze emotional speech from five male and four female speakers in order to find any similarities and differences among individual speakers. The speakers are actors that have interpreted six emotions in four different languages. The results show that all the speakers share some universal signs regarding certain emotion-dependent features of emotional expression. Further analysis revealed that almost all speakers in the analysis used unique sets of emotion-dependent features and each speaker used unique values for the defined emotion-dependent features. The comparison among speakers shows that the expressed emotions can be analyzed according to two criteria. The first criterion is a defined set of emotion-dependent features and the second is an emotion-dependent feature value.

Adult↗
PubMed2000

AgNOR analysis of atypical ductal hyperplasia and intraductal carcinoma of the breast.

OBJECTIVE: To determine the diagnostic and prognostic value of argyrophilic nucleolar organizer regions (AgNORs) in atypical ductal hyperplasia (ADH), ductal carcinoma in situ (DCIS) and microinvasive ductal carcinoma (MDCA) of the breast. STUDY DESIGN: Image analysis of histologic sections from biopsies of 46 breast ADH and DCIS and 18 cases of MDCA. Determination of morphometric features of cell nuclei and nucleolar organizer regions by using AMBA software system. Data were compared with the estrogen receptor/progesterone receptor (ER/PR) content as well as with the growth fraction, determined immunohistochemically. RESULTS: AgNOR number and total AgNOR area increased from ADH to DCIS. The highest values were recorded in cases of DCIS with microinvasion. Differences between ADH and intraductal or microinvasive ductal carcinoma were statistically significant. Within the group of intraductal carcinomas, the lowest values were measured in the solid type and highest values in the comedo type. A correlation was found between AgNOR features and growth fraction but not between these features and ER/PR status. CONCLUSION: Selected AgNOR features are relevant for differentiation between ADH and DCIS as well as between low and high grade DCIS and microinvasive ductal carcinoma. Therefore, objective and reproducible data obtained by AgNOR analysis may allow better evaluation of the prognostic significance of these lesions.

Biomarkers, Tumor↗