SEARCH · PubMed Health

Results for “Feature selection”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 415 records · Page 23Linked to original sources

PubMed1979

The spandrels of San Marco and the Panglossian paradigm: a critique of the adaptationist programme.

An adaptationist programme has dominated evolutionary thought in England and the United States during the past 40 years. It is based on faith in the power of natural selection as an optimizing agent. It proceeds by breaking an oragnism into unitary 'traits' and proposing an adaptive story for each considered separately. Trade-offs among competing selective demands exert the only brake upon perfection; non-optimality is thereby rendered as a result of adaptation as well. We criticize this approach and attempt to reassert a competing notion (long popular in continental Europe) that organisms must be analysed as integrated wholes, with Baupläne so constrained by phyletic heritage, pathways of development and general architecture that the constraints themselves become more interesting and more important in delimiting pathways of change than the selective force that may mediate change when it occurs. We fault the adaptationist programme for its failure to distinguish current utility from reasons for origin (male tyrannosaurs may have used their diminutive front legs to titillate female partners, but this will not explain why they got so small); for its unwillingness to consider alternatives to adaptive stories; for its reliance upon plausibility alone as a criterion for accepting speculative tales; and for its failure to consider adequately such competing themes as random fixation of alleles, production of non-adaptive structures by developmental correlation with selected features (allometry, pleiotropy, material compensation, mechanically forced correlation), the separability of adaptation and selection, multiple adaptive peaks, and current utility as an epiphenomenon of non-adaptive structures. We support Darwin's own pluralistic approach to identifying the agents of evolutionary change.

Adaptation, Biological↗

PubMed2001

Improvement of mammographic mass characterization using spiculation meausures and morphological features.

We are developing new computer vision techniques for characterization of breast masses on mammograms. We had previously developed a characterization method based on texture features. The goal of the present work was to improve our characterization method by making use of morphological features. Toward this goal, we have developed a fully automated, three-stage segmentation method that includes clustering, active contour, and spiculation detection stages. After segmentation, morphological features describing the shape of the mass were extracted. Texture features were also extracted from a band of pixels surrounding the mass. Stepwise feature selection and linear discriminant analysis were employed in the morphological, texture, and combined feature spaces for classifier design. The classification accuracy was evaluated using the area Az under the receiver operating characteristic curve. A data set containing 249 films from 102 patients was used. When the leave-one-case-out method was applied to partition the data set into trainers and testers, the average test Az for the task of classifying the mass on a single mammographic view was 0.83 +/- 0.02, 0.84 +/- 0.02, and 0.87 +/- 0.02 in the morphological, texture, and combined feature spaces, respectively. The improvement obtained by supplementing texture features with morphological features in classification was statistically significant (p = 0.04). For classifying a mass as malignant or benign, we combined the leave-one-case-out discriminant scores from different views of a mass to obtain a summary score. In this task, the test Az value using the combined feature space was 0.91 +/- 0.02. Our results indicate that combining texture features with morphological features extracted from automatically segmented mass boundaries will be an effective approach for computer-aided characterization of mammographic masses.

Automation↗

PubMed2004

Heidelberg retina tomograph measurements of the optic disc and parapapillary retina for detecting glaucoma analyzed by machine learning classifiers.

PURPOSE: To determine whether topographical measurements of the parapapillary region analyzed by machine learning classifiers can detect early to moderate glaucoma better than similarly processed measurements obtained within the disc margin and to improve methods for optimization of machine learning classifier feature selection. METHODS: One eye of each of 95 patients with early to moderate glaucomatous visual field damage and of each of 135 normal subjects older than 40 years participating in the longitudinal Diagnostic Innovations in Glaucoma Study (DIGS) were included. Heidelberg Retina Tomograph (HRT; Heidelberg Engineering, Dossenheim, Germany) mean height contour was measured in 36 equal sectors, both along the disc margin and in the parapapillary region (at a mean contour line radius of 1.7 mm). Each sector was evaluated individually and in combination with other sectors. Gaussian support vector machine (SVM) learning classifiers were used to interpret HRT sector measurements along the disc margin and in the parapapillary region, to differentiate between eyes with normal and glaucomatous visual fields and to compare the results with global and regional HRT parameter measurements. The area under the receiver operating characteristic (ROC) curve was used to measure diagnostic performance of the HRT parameters and to evaluate the cross-validation strategies and forward selection and backward elimination optimization techniques that were used to generate the reduced feature sets. RESULTS: The area under the ROC curve for mean height contour of the 36 sectors along the disc margin was larger than that for the mean height contour in the parapapillary region (0.97 and 0.85, respectively). Of the 36 individual sectors along the disc margin, those in the inferior region between 240 degrees and 300 degrees, had the largest area under the ROC curve (0.85-0.91). With SVM Gaussian techniques, the regional parameters showed the best ability to discriminate between normal eyes and eyes with glaucomatous visual field damage, followed by the global parameters, mean height contour measures along the disc margin, and mean height contour measures in the parapapillary region. The area under the ROC curve was 0.98, 0.94, 0.93, and 0.85, respectively. Cross-validation and optimization techniques demonstrated that good discrimination (99% of peak area under the ROC curve) can be obtained with a reduced number of HRT parameters. CONCLUSIONS: Mean height contour measurements along the disc margin discriminated between normal and glaucomatous eyes better than measurements obtained in the parapapillary region.

Area Under Curve↗

PubMed2005

Efficient visual search without top-down or bottom-up guidance.

Two types of mechanisms have dominated theoretical accounts of efficient visual search. The first are bottom-up processes related to the characteristics of retinotopic feature maps. The second are top-down mechanisms related to feature selection. To expose the potential involvement of other mechanisms, we introduce a new search paradigm whereby a target is defined only in a context-dependent manner by multiple conjunctions of feature dimensions. Because targets in a multiconjunction task cannot be distinguished from distractors either by bottom-up guidance or top-down guidance, current theories of visual search predict inefficient search. While inefficient search does occur for the multiple conjunctions of orientation with color or luminance, we find efficient search for multiple conjunctions of luminance/size, luminance/shape, and luminance/topology. We also show that repeated presentations of either targets or a set of distractors result in much faster performance and that bottom-up feature extraction and top-down selection cannot account for efficient search on their own. In light of this, we discuss the possible role of perceptual organization in visual search. Furthermore, multiconjunction search could provide a new method for investigating perceptual grouping in visual search.

Form Perception↗

PubMed2006

Reliability of the ICRP'S dose coefficients for members of the public: IV. basis of the human alimentary tract model and uncertainties in model predictions.

The biokinetic and dosimetric model of the gastrointestinal (GI) tract applied in current documents of the International Commission on Radiological Protection (ICRP) was developed in the mid-1960s. The model was based on features of a reference adult male and was first used by the ICRP in Publication 30, Limits for Intakes of Radionuclides by Workers (Part 1, 1979). In the late 1990s an ICRP task group was appointed to develop a biokinetic and dosimetric model of the alimentary tract that reflects updated information and addresses current needs in radiation protection. The new age-specific and gender-specific model, called the Human Alimentary Tract Model (HATM), has been completed and will replace the GI model of Publication 30 in upcoming ICRP documents. This paper discusses the basis for the structure and parameter values of the HATM, summarises the uncertainties associated with selected features and types of predictions of the HATM and examines the sensitivity of dose estimates to these uncertainties for selected radionuclides. Emphasis is on generic biokinetic features of the HATM, particularly transit times through the lumen of the alimentary tract, but key dosimetric features of the model are outlined, and the sensitivity of tissue dose estimates to uncertainties in dosimetric as well as biokinetic features of the HATM are examined for selected radionuclides.

Dose-Response Relationship, Radiation↗

PubMed1995

Spatial features in body-surface potential maps can identify patients with a history of sustained ventricular tachycardia.

BACKGROUND: Regional disparities of ventricular primary-repolarization properties contribute to an electrophysiological substrate for arrhythmias. Such disparities can be assessed from body-surface distributions of ECG QRST areas. Our objective was to isolate and test those features of QRST-area distributions that would be suitable for identifying patients at risk for life-threatening ventricular arrhythmias. METHODS AND RESULTS: We recorded ECGs simultaneously from 120 leads during sinus rhythm for 204 patients taking no antiarrhythmic drugs: half had had sustained ventricular tachycardia (VT); the other half, a myocardial infarction but no history of VT. For each patient, we calculated the QRST area in each lead and, using Karhunen-Loeve (K-L) expansion, reduced these data to 16 coefficients (each relating to one spatial feature, an eigenvector, derived from the total set of 204 QRST-area maps). Using stepwise discriminant analysis, we selected feature subsets that best discriminated between the two groups, and we estimated by a bootstrap procedure using 1000 trials how these subsets would perform on a prospective patient population. The mean diagnostic performance of the classifier for 1000 randomly selected training sets (n = 102 in each, with both groups equally represented) increased monotonically with the number of features used for classification. The initial trend for the corresponding test sets (n = 102 in each) was the same but reversed when the number of features exceeded eight. For an optimal set of eight spatial features, the sensitivity and specificity of the classifier for detecting patients with VT in 1000 test sets were (mean +/- SD) 90.3 +/- 4.3% and 78.0 +/- 6.1%, and its positive and negative predictive accuracies were 80.7 +/- 4.2% and 89.2 +/- 4.2%, respectively. Use of QRS duration as a supplementary feature to eight K-L coefficients can, in the test sets, increase specificity to 80.9 +/- 5.4% and positive predictive accuracy to 82.8 +/- 3.9% compared with the results for the optimal number of eight K-L features alone. CONCLUSIONS: Multiple body-surface ECGs contain valuable spatial features that can identify the presence of an arrhythmogenic substrate in the myocardium of patients at risk for ventricular arrhythmias. Our results compare very favorably with those achieved by any other known test, invasive or noninvasive, for arrhythmogenicity.

Body Surface Potential Mapping↗

PubMed2005

Disseminated intravascular coagulation in catastrophic antiphospholipid syndrome: clinical and haematological characteristics of 23 patients.

BACKGROUND: Disseminated intravascular coagulation (DIC) is an acquired syndrome characterised by formation of microthrombi and fibrin deposition in the microvasculature. The catastrophic antiphospholipid syndrome (APS) is characterised by multiorgan thrombosis, mainly involving small vessels. A broad spectrum of disorders may develop DIC features; however, the catastrophic APS has not previously been recognised as a cause of DIC. OBJECTIVE: To analyse the clinical and laboratory characteristics of catastrophic APS patients with DIC features. METHODS: The web site based international registry of patients with catastrophic APS (CAPS registry) (http://www.med.ub.es/MIMMUN/FORUM/CAPS.HTM) was analysed and the cases with DIC features selected. RESULTS: In 173 patients with catastrophic APS, 23 (13%) were found with DIC features. The clinical and immunological characteristics were similar in catastrophic APS patients with and without DIC features; a significant difference was found only in the prevalence of thrombocytopenia (100% in patients with DIC features v 59% in those without DIC features). CONCLUSIONS: DIC features are not rare in catastrophic APS, supporting the need for systematic screening of antiphospholipid antibodies in all patients with DIC features without precipitating factors. The presence of DIC features in the context of an APS makes it imperative to rule out the catastrophic variant of this syndrome.

Adolescent↗

PubMed2006

Accurate prediction of BRCA1 and BRCA2 heterozygous genotype using expression profiling after induced DNA damage.

PURPOSE: In this study, the differential gene expression changes following radiation-induced DNA damage in healthy cells from BRCA1/BRCA1 mutation carriers have been compared with controls using high-density microarray technology. We aimed to establish if BRCA1/BRCA2 mutation carriers could be distinguished from noncarriers based on expression profiling of normal cells. EXPERIMENTAL DESIGN: Short-term primary fibroblast cultures were established from skin biopsies from 10 BRCA1 and 10 BRCA2 mutation carriers and 10 controls, all of whom had previously had breast cancer. The cells were subjected to 15 Gy ionizing irradiation to induce DNA damage. RNA was extracted from all cell cultures, preirradiation and at 1 hour postirradiation. For expression profiling, 15 K spotted cDNA microarrays manufactured by the Cancer Research UK DNA Microarray Facility were used. Statistical feature selection was used with a support vector machine (SVM) classifier to determine the best feature set for predicting BRCA1 or BRCA2 heterozygous genotype. To investigate prediction accuracy, a nonprobabilistic classifier (SVM) and a probabilistic Gaussian process classifier were used. RESULTS: In the task of distinguishing BRCA1 and BRCA2 mutation carriers from noncarriers and from each other following radiation-induced DNA damage, the SVM achieved 90%, and the Gaussian process classifier achieved 100% accuracy. This effect could not be achieved without irradiation. In addition, the SVM identified a set of BRCA genotype predictor genes. CONCLUSIONS: We conclude that after irradiation-induced DNA damage, BRCA1 and BRCA2 mutation carrier cells have a distinctive expression phenotype, and this may have a future role in predicting genotypes, with application to clinical detection and classification of mutations.

Adult↗

PubMed1999

Computer-supported diagnosis of melanoma in profilometry.

Laser profilometry offers new possibilities to improve non-invasive tumor diagnostics in dermatology. In this paper, a new approach to computer-supported analysis and interpretation of high-resolution skin-surface profiles of melanomas and nevocellular nevi is presented. Image analysis methods are used to describe the profile's structures by texture parameters based on co-occurrence matrices, features extracted from the Fourier power spectrum, and fractal features. Different feature selection strategies, including genetic algorithms, are applied to determine the best possible subsets of features for the classification task. Several architectures of multilayer perceptrons with error back-propagation as learning paradigm are trained for the automatic recognition of melanomas and nevi. Furthermore, network-pruning algorithms are applied to optimize the network topology. In the study, the best neural classifier showed an error rate of 4.5% and was obtained after network pruning. The smallest error rate in all, of 2.3%, was achieved with nearest neighbor classification.

Diagnosis, Differential↗

PubMed1996

Is there feature-based attentional selection in visual search?

A new paradigm combines attentional cuing and rapid serial visual presentation to disentangle the effects of perceptual filtering and location selection. Observers search successive, superimposed arrays, in which feature values are alternated for a target numeral among letters. Two dimensions, size (small, large) and color (red, green) are tested. Selective attention to feature values is jointly manipulated by instructions, presentation probabilities, and payoffs. In Experiment 1, the attended feature provides temporal, not spatial, information; observers show no attentional costs or benefits in response accuracy. In Experiment 2, the attended feature indicates a unique location; observers show consistent attentional costs and benefits. Selective attention to a particular size or color does not cause perceptual exclusion or admission of items containing that feature; it acts by guiding search processes to spatial locations that contain the to-be-attended feature.

Attention↗

PubMed1995

Wavelet analysis of olfactory nerve response to stimulus.

Multiunit electrophysiological activity recorded by gross electrodes from the olfactory nerve was analyzed by wavelet decomposition, a relatively new method of signal processing. The analysis was run on data from the unstimulated olfactory system as well as on data evoked in response to six different odorant stimuli. Like Fourier analysis, wavelet analysis provides a spectral decomposition of the signal. Unlike Fourier, wavelet analysis also locates the dominant spectral features in time. The output of a wavelet analysis can be further processed to enhance selected features. The increased amplitude of the nerve response evoked by stimulation was the most obvious feature, but efforts to learn from it were unproductive. The temporal pattern of receptor cell activity was much more yielding. The analysis resolved the nerve activity into three classes of events based on duration. On wavelet maps these classes of events separate out into three shifting and overlapping but distinct bands, one of which was interpreted as being associated with individual receptor cell firings and the other two as short and somewhat longer duration bursts of activity that was attributed to the synchronized firing of a group of receptor cells. This interpretation is supported by experiments in which waveforms simulating action potentials and bursts of action potentials are added to recorded data. Stimulation of the olfactory system with odorant molecules evokes a significant increase in the number of short duration bursts, and an amplitude increase that can be related to the number of receptor cells responding. Changes in the patterns of wavelet events can be associated with synchrony of cell firing, reset times for bursts of firing, and possibly other physiological dynamics. A number of differences in activity patterns with different odorants were observed, but without sufficient repeatability to allow reliable discrimination among them. While this study is clearly preliminary in that regard, it shows the potential of the wavelet method for contributing to the understanding of olfaction.

Animals↗

PubMed2001

Stepwise logistic regression analysis of tumor contour features for breast ultrasound diagnosis.

To assist the ultrasound (US) differential diagnosis of solid breast tumors by using stepwise logistic regression (SLR) analysis of tumor contour features, we retrospectively reviewed 111 medical records of digitized US images of breast pathologies. They were pathologically proved benign breast tumors from 40 patients (i.e., 40 fibroadenomas) and malignant breast tumors from 71 patients (i.e., 71 infiltrative ductal carcinomas). Radiologists, before analysis by the computer-aided diagnosis (CAD) system, segmented the tumors manually. The contour features were calculated by measuring the radial length of tumor boundaries. The features selection process was accomplished using a stepwise analysis procedure. Then, an SLR model with contour features was used to classify tumors as benign or malignant. In this experiment, cases were sampled with "leave-one-out" test methods to evaluate the SLR performance using a receiver operating characteristic (ROC) curve. The accuracy of our SLR model with contour features for classifying malignancies was 91.0% (101 of 111 tumors), the sensitivity was 97.2% (69 of 71), the specificity was 80.0% (32 of 40), the positive predictive value was 89.6% (69 of 77), and the negative predictive value was 94.1% (32 of 34). The CAD system using SLR can differentiate solid breast nodules with relatively high accuracy and its high negative predictive value could potentially help inexperienced operators to avoid misdiagnoses. Because the SLR model is trainable, it could be optimized if a larger set of tumor images were supplied.

Adolescent↗

PubMed2005

Proteomic cancer classification with mass spectrometry data.

The ultimate goal of cancer proteomics is to adapt proteomic technologies for routine use in clinical laboratories for the purpose of diagnostic and prognostic classification of disease states, as well as in evaluating drug toxicity and efficacy. Analysis of tumor-specific proteomic profiles may also allow better understanding of tumor development and the identification of novel targets for cancer therapy. The biological variability among patient samples as well as the huge dynamic range of biomarker concentrations are currently the main challenges facing efforts to deduce diagnostic patterns that are unique to specific disease states. While several strategies exist to address this problem, we focus here on cancer classification using mass spectrometry (MS) for proteomic profiling and biomarker identification. Recent advances in MS technology are starting to enable high-throughput profiling of the protein content of complex samples. For cancer classification, the protein samples from cancer patients and noncancer patients or from different cancer stages are analyzed through MS instruments and the MS patterns are used to build a diagnostic classifier. To illustrate the importance of feature selection in cancer classification, we present a method based on support vector machine-recursive feature elimination (SVM-RFE), demonstrated on two cancer datasets from ovarian and lung cancer.

Biomarkers, Tumor↗

PubMed2005

Computer-aided diagnosis of the solitary pulmonary nodule.

RATIONALE AND OBJECTIVES: We sought to investigate the utility of a computer-aided diagnosis in the task of differentiating malignant nodules from benign nodules based on single thin-section computed tomography image data. MATERIALS AND METHODS: Eighty-one thin-section computed tomography data sets of solitary pulmonary nodules with proven diagnoses (48 malignant and 33 benign) were contoured manually on a single representative slice by a thoracic radiologist (>10 years of experience). Two separate contours were created for each nodule, one including only the solid portion of the nodule and one including any ground-glass components. For each contour 75 features were calculated that measured the attenuation, shape, and texture of the nodule. These features were than input into a feature selection step and four different classifiers to determine if the diagnosis could be predicted from the feature vector. Training and testing was conducted in a resubstitution and leave-one-out fashion and performance was evaluated using ROC techniques. RESULTS: In a leave-one-out testing methodology the classifiers resulted with areas under the ROC curve (A(Z)) that ranged from 0.68 to 0.92. When evaluating with resubstitution the A(Z) ranged from 0.93 to 1.00. CONCLUSION: Computer-aided diagnosis has the potential to assist radiologists in the task of differentiating solitary pulmonary nodules and in the management of these patients.

Diagnosis, Differential↗

PubMed2005

Wavelet neural network modeling in QSPR for prediction of solubility of 25 anthraquinone dyes at different temperatures and pressures in supercritical carbon dioxide.

A wavelet neural network (WNN) model in quantitative structure property relationship (QSPR) was developed for predicting solubility of 25 anthraquinone dyes in supercritical carbon dioxide over a wide range of pressures (70-770 bar) and temperatures (291-423 K). A large number of descriptors were calculated with Dragon software and a subset of calculated descriptors was selected from 18 classes of Dragon descriptors with a stepwise multiple linear regression (MLR) as a feature selection technique. Six calculated and two experimental descriptors, pressure and temperature, were selected as the most feasible descriptors. The selected descriptors were used as input nodes in a wavelet neural network (WNN) model. The wavelet neural network architecture and its parameters were optimized simultaneously. The data was randomly divided to the training, prediction and validation sets. The predictive ability of the model was evaluated using validation data set. The root mean squares error (RMSE) and mean absolute errors were 0.339 and 0.221, respectively, for the validation data set. The performance of the WNN model was also compared with artificial neural network (ANN) model and the results showed the superiority of the WNN over ANN model.

Anthraquinones↗

PubMed1986

Identification of best electrocardiographic leads for diagnosing anterior and inferior myocardial infarction by statistical analysis of body surface potential maps.

In view of the increasing interest in quantifying and modifying the size of myocardial infarction (MI), it is important to look for clinically practical subsets of electrocardiographic leads that allow the earliest and most accurate diagnosis of the presence and electrocardiographic type of MI. A practical approach is described, taking advantage of the increased information content of body surface potential maps over standard electrocardiographic techniques for facilitating clinical use of body surface potential maps for such a purpose. Multivariate analysis was performed on 120-lead electrocardiographic data, simultaneously recorded in 236 normal subjects, 114 patients with anterior MI and 144 patients with inferior MI, using as features instantaneous voltages on time-normalized QRS and ST-T waveforms. Leads and features for optimal separation of normal subjects from, respectively, anterior MI and inferior MI patients were selected. Features measured on leads originating from the upper left precordial area, lower midthoracic region and the back correctly identified 97% of anterior MI patients, with a specificity of 95%; in patients with inferior MI, features obtained from leads located in the lower left back, left leg, right subclavicular area, upper dorsal region and lower right chest correctly classified 94% of the group, with specificity kept at 95%. Most features were measured in early and mid-QRS, although very potent discriminators were found in the late portion of the T wave.(ABSTRACT TRUNCATED AT 250 WORDS)

Action Potentials↗

PubMed2001

Machine classification of dental images with visual search.

RATIONALE AND OBJECTIVES: The authors performed this study to assess the performance of a computer-based classification system that uses gaze locations of observers to define the subspace for machine learning. MATERIALS AND METHODS: Thirty-two dental radiographs were classified by an expert viewer into four categories of disease of the periapical region: no disease (normal tooth), mild disease (widened periodontal ligament space), moderate disease (destruction of the lamina dura), and severe disease (resorption of bone in the periapical area). There were eight images in each category. Six observers independently viewed the images while their eye gaze position was recorded. They then classified the images into one of the four categories. A sample of image space was used as input to a machine learning routine to develop a machine classifier. Sample space was determined with three techniques: visual gaze, random selection, and constrained random selection. K analyses were used to compare classification accuracies with the three sampling techniques. RESULTS: With use of the expert classification as a standard of reference, observers classified images with 57% accuracy, and the machine classified images with 84% accuracy by using the same gaze-selected features and image space. Results of kappa analyses revealed mean values of 0.78 for gaze-selected sampling, 0.69 for random sampling, 0.68 for constrained random selection, and 0.44 for observers. The use of sample space selected with the visual gaze technique was superior to that selected with both random-selection techniques and by the observers. CONCLUSION: Machine classification of dental images improves the accuracy of individual observers using gaze-selected image space.

Artificial Intelligence↗

PubMed1975

Development of multiple dimension use in form classification.

The use of multiple form dimensions in pattern classification was studied with adults and children in grades 2 and 5. Each subject sorted 30 8-sided random polygons first into 2, then into 3, and finally into 4 groups and repeated the procedure 1 week later. A series of discriminant analyses, using 9 physical form characteristics as predictors, was used to answer several developmental questions. Reliability of classification, number and saliency of features selected, and accuracy with which they were used all implied continuous development of perceptual skills. Multiple feature use in classification was evidenced at all age levels.

Achievement↗