Kernel machines for epilepsy diagnosis via EEG signal classification: A comparative study

Objective: We carry out a systematic assessment on a suite of kernel-based learning machines while coping with the task of epilepsy diagnosis through automatic electroencephalogram (EEG) signal classification. Methods and materials: The kernel machines investigated include the standard support vector machine (SVM), the least squares SVM, the Lagrangian SVM, the smooth SVM, the proximal SVM, and the relevance vector machine. An extensive series of experiments was conducted on publicly available data, whose clinical EEG recordings were obtained from five normal subjects and five epileptic patients. The performance levels delivered by the different kernel machines are contrasted in terms of the criteria of predictive accuracy, sensitivity to the kernel function/parameter value, and sensitivity to the type of features extracted from the signal. For this purpose, 26 values for the kernel parameter (radius) of two well-known kernel functions (namely. Gaussian and exponential radial basis functions) were considered as well as 21 types of features extracted from the EEG signal, including statistical values derived from the discrete wavelet transform, Lyapunov exponents, and combinations thereof. Results: We first quantitatively assess the impact of the choice of the wavelet basis on the quality of the features extracted. Four wavelet basis functions were considered in this study. Then, we provide the average accuracy (i.e., cross-validation error) values delivered by 252 kernel machine configurations; in particular, 40%/35% of the best-calibrated models of the standard and least squares SVMs reached 100% accuracy rate for the two kernel functions considered. Moreover, we show the sensitivity profiles exhibited by a large sample of the configurations whereby one can visually inspect their levels of sensitiveness to the type of feature and to the kernel function/parameter value. Conclusions: Overall, the results evidence that all kernel machines are competitive in terms of accuracy, with the standard and least squares SVMs prevailing more consistently. Moreover, the choice of the kernel function and parameter value as well as the choice of the feature extractor are critical decisions to be taken, albeit the choice of the wavelet family seems not to be so relevant. Also, the statistical values calculated over the Lyapunov exponents were good sources of signal representation, but not as informative as their wavelet counterparts. Finally, a typical sensitivity profile has emerged among all types of machines, involving some regions of stability separated by zones of sharp variation, with some kernel parameter values clearly associated with better accuracy rates (zones of optimality). (C) 2011 Elsevier B.V. All rights reserved.

Noise reduction in biomedical speech signal processing based on time and frequency Kalman filtering combined with spectral subtraction

The goal of this paper is to study and propose a new technique for noise reduction used during the reconstruction of speech signals, particularly for biomedical applications. The proposed method is based on Kalman filtering in the time domain combined with spectral subtraction. Comparison with discrete Kalman filter in the frequency domain shows better performance of the proposed technique. The performance is evaluated by using the segmental signal-to-noise ratio and the Itakura-Saito`s distance. Results have shown that Kalman`s filter in time combined with spectral subtraction is more robust and efficient, improving the Itakura-Saito`s distance by up to four times. (C) 2007 Elsevier Ltd. All rights reserved.

Void fraction measurement and signal analysis from multiple-electrode impedance sensors

Void fraction sensors are important instruments not only for monitoring two-phase flow, but for furnishing an important parameter for obtaining flow map pattern and two-phase flow heat transfer coefficient as well. This work presents the experimental results obtained with the analysis of two axially spaced multiple-electrode impedance sensors tested in an upward air-water two-phase flow in a vertical tube for void fraction measurements. An electronic circuit was developed for signal generation and post-treatment of each sensor signal. By phase shifting the electrodes supplying the signal, it was possible to establish a rotating electric field sweeping across the test section. The fundamental principle of using a multiple-electrode configuration is based on reducing signal sensitivity to the non-uniform cross-section void fraction distribution problem. Static calibration curves were obtained for both sensors, and dynamic signal analyses for bubbly, slug, and turbulent churn flows were carried out. Flow parameters such as Taylor bubble velocity and length were obtained by using cross-correlation techniques. As an application of the void fraction tested, vertical flow pattern identification could be established by using the probability density function technique for void fractions ranging from 0% to nearly 70%.

Ultrasonic Viscosity Measurement Using the Shear-Wave Reflection Coefficient with a Novel Signal Processing Technique

Real-time viscosity measurement remains a necessity for highly automated industry. To resolve this problem, many studies have been carried out using an ultrasonic shear wave reflectance method. This method is based on the determination of the complex reflection coefficient`s magnitude and phase at the solid-liquid interface. Although magnitude is a stable quantity and its measurement is relatively simple and precise, phase measurement is a difficult task because of strong temperature dependence. A simplified method that uses only the magnitude of the reflection coefficient and that is valid under the Newtonian regimen has been proposed by some authors, but the obtained viscosity values do not match conventional viscometry measurements. In this work, a mode conversion measurement cell was used to measure glycerin viscosity as a function of temperature (15 to 25 degrees C) and corn syrup-water mixtures as a function of concentration (70 to 100 wt% of corn syrup). Tests were carried out at 1 MHz. A novel signal processing technique that calculates the reflection coefficient magnitude in a frequency band, instead of a single frequency, was studied. The effects of the bandwidth on magnitude and viscosity were analyzed and the results were compared with the values predicted by the Newtonian liquid model. The frequency band technique improved the magnitude results. The obtained viscosity values came close to those measured by the rotational viscometer with percentage errors up to 14%, whereas errors up to 96% were found for the single frequency method.

An In-Depth Study of the Barkhausen Emission Signal Properties of the Plastically Deformed Fe-2%Si Alloy

This paper presents the results of the in-depth study of the Barkhausen effect signal properties for the plastically deformed Fe-2%Si samples. The investigated samples have been deformed by cold rolling up to plastic strain epsilon(p) = 8%. The first approach consisted of time-domain-resolved pulse and frequency analysis of the Barkhausen noise signals whereas the complementary study consisted of the time-resolved pulse count analysis as well as a total pulse count. The latter included determination of time distribution of pulses for different threshold voltage levels as well as the total pulse count as a function of both the amplitude and the duration time of the pulses. The obtained results suggest that the observed increase in the Barkhausen noise signal intensity as a function of deformation level is mainly due to the increase in the number of bigger pulses.

In silico analysis of candidate genes involved in light sensing and signal transduction pathways in soybean

Several aspects of photoperception and light signal transduction have been elucidated by studies with model plants. However, the information available for economically important crops, such as Fabaceae species, is scarce. In order to incorporate the existing genomic tools into a strategy to advance soybean research, we have investigated publicly available expressed sequence tag ( EST) sequence databases in order to identify Glycine max sequences related to genes involved in light-regulated developmental control in model plants. Approximately 38,000 sequences from open-access databases were investigated, and all bona fide and putative photoreceptor gene families were found in soybean sequence databases. We have identified G. max orthologs for several families of transcriptional regulators and cytoplasmic proteins mediating photoreceptor-induced responses, although some important Arabidopsis phytochrome-signaling components are absent. Moreover, soybean and Arabidopsis gene-family homologs appear to have undergone a distinct expansion process in some cases. We propose a working model of light perception, signal transduction and response-eliciting in G. max, based on the identified key components from Arabidopsis. These results demonstrate the power of comparative genomics between model systems and crop species to elucidate several aspects of plant physiology and metabolism.

SacRALF1, a peptide signal from the grass sugarcane (Saccharum spp.), is potentially involved in the regulation of tissue expansion

Rapid alkalinization factor (RALF) is part of a growing family of small peptides with hormone characteristics in plants. Initially isolated from leaves of tobacco plants, RALF peptides can be found throughout the plant kingdom and they are expressed ubiquitously in plants. We took advantage of the small gene family size of RALF genes in sugarcane and the ordered cellular growth of the grass sugarcane leaves to gain information about the function of RALF peptides in plants. Here we report the isolation of two RALF peptides from leaves of sugarcane plants using the alkalinization assay. SacRALF1 was the most abundant and, when added to culture media, inhibited growth of microcalli derived from cell suspension cultures at concentrations as low as 0.1 mu M. Microcalli exposed to exogenous SacRALF1 for 5 days showed a reduced number of elongated cells. Only four copies of SacRALF genes were found in sugarcane plants. All four SacRALF genes are highly expressed in young and expanding leaves and show a low or undetectable level of expression in expanded leaves. In half-emerged leaf blades, SacRALF transcripts were found at high levels at the basal portion of the leaf and at low levels at the apical portion. Gene expression analyzes localize SacRALF genes in elongation zones of roots and leaves. Mature leaves, which are devoid of expanding cells, do not show considerable expression of SacRALF genes. Our findings are consistent with SacRALF genes playing a role in plant development potentially regulating tissue expansion.

A matching pursuit-based signal complexity measure for the analysis of newborn EEG

This paper presents a new relative measure of signal complexity, referred to here as relative structural complexity, which is based on the matching pursuit (MP) decomposition. By relative, we refer to the fact that this new measure is highly dependent on the decomposition dictionary used by MP. The structural part of the definition points to the fact that this new measure is related to the structure, or composition, of the signal under analysis. After a formal definition, the proposed relative structural complexity measure is used in the analysis of newborn EEG. To do this, firstly, a time-frequency (TF) decomposition dictionary is specifically designed to compactly represent the newborn EEG seizure state using MP. We then show, through the analysis of synthetic and real newborn EEG data, that the relative structural complexity measure can indicate changes in EEG structure as it transitions between the two EEG states; namely seizure and background (non-seizure).

Comparison of BR and QR Eigenvalue Algorithms for Power System Small Signal Stability Analysis

The BR algorithm is a novel and efficient method to find all eigenvalues of upper Hessenberg matrices and has never been applied to eigenanalysis for power system small signal stability. This paper analyzes differences between the BR and the QR algorithms with performance comparison in terms of CPU time based on stopping criteria and storage requirement. The BR algorithm utilizes accelerating strategies to improve its performance when computing eigenvalues of narrowly banded, nearly tridiagonal upper Hessenberg matrices. These strategies significantly reduce the computation time at a reasonable level of precision. Compared with the QR algorithm, the BR algorithm requires fewer iteration steps and less storage space without depriving of appropriate precision in solving eigenvalue problems of large-scale power systems. Numerical examples demonstrate the efficiency of the BR algorithm in pursuing eigenanalysis tasks of 39-, 68-, 115-, 300-, and 600-bus systems. Experiment results suggest that the BR algorithm is a more efficient algorithm for large-scale power system small signal stability eigenanalysis.

Analysis of small signal stability margins using genetic optimization

Power system small signal stability analysis aims to explore different small signal stability conditions and controls, namely: (1) exploring the power system security domains and boundaries in the space of power system parameters of interest, including load flow feasibility, saddle node and Hopf bifurcation ones; (2) finding the maximum and minimum damping conditions; and (3) determining control actions to provide and increase small signal stability. These problems are presented in this paper as different modifications of a general optimization to a minimum/maximum, depending on the initial guesses of variables and numerical methods used. In the considered problems, all the extreme points are of interest. Additionally, there are difficulties with finding the derivatives of the objective functions with respect to parameters. Numerical computations of derivatives in traditional optimization procedures are time consuming. In this paper, we propose a new black-box genetic optimization technique for comprehensive small signal stability analysis, which can effectively cope with highly nonlinear objective functions with multiple minima and maxima, and derivatives that can not be expressed analytically. The optimization result can then be used to provide such important information such as system optimal control decision making, assessment of the maximum network's transmission capacity, etc. (C) 1998 Elsevier Science S.A. All rights reserved.

Autoinhibition by an internal nuclear localization signal revealed by the crystal structure of mammalian importin alpha

Importin alpha is the nuclear import receptor that recognizes classical monopartite and bipartite nuclear localization signals (NLSs). The structure of mouse importin alpha has been determined at 2.5 Angstrom resolution. The structure shows a large C-terminal domain containing armadillo repeats, and a less structured N-terminal importin beta-binding domain containing an internal NLS bound to the NLS-binding site. The structure explains the regulatory switch between the cytoplasmic, high-affinity form, and the nuclear, low-affinity form for NLS binding of the nuclear import receptor predicted by the current models of nuclear import. Importin beta conceivably converts the low- to high-affinity form by binding to a site overlapping the autoinhibitory sequence. The structure also has implications for understanding NLS recognition, and the structures of armadillo and HEAT repeats.

Signal Transduction, Plasma Membrane Calcium Movements, and Pigment Translocation in Freshwater Shrimp Chromatophores

Crustacean color change results from the differential translocation of chromatophore pigments, regulated by neurosecretory peptides like red pigment concentrating hormone (RPCH) that, in the red ovarian chromatophores of the freshwater shrimp Macrobrachium olfersi, triggers pigment aggregation via increased cytosolic cGMP and Ca(2+) of both smooth endoplasmatic reticulum (SER) and extracellular origin. However, Ca(2+) movements during RPCH signaling and the mechanisms that regulate intracellular [Ca(2+)] are enigmatic. We investigate Ca(2+) transporters in the chromatophore plasma membrane and Ca(2+) movements that occur during RPCH signal transduction. Inhibition of the plasma membrane Ca(2+)-ATPase by La(3+) and indirect inhibition of the Na(+)/Ca(2+) exchanger by ouabain induce pigment aggregation, revealing a role for both in Ca(2+) extrusion. Ca(2+) channel blockade by La(3+) or Cd(2+) strongly inhibits slow-phase RPCH-triggered aggregation during which pigments disperse spontaneously. L-type Ca(2+) channel blockade by gabapentin markedly reduces rapid-phase translocation velocity; N- or P/Q-type blockade by omega-conotoxin MVIIC strongly inhibits RPCH-triggered aggregation and reduces velocity, effects revealing RPCH-signaled influx of extracellular Ca(2+). Plasma membrane depolarization, induced by increasing external K(+) from 5 to 50 mM, produces Ca(2+)-dependent pigment aggregation, whereas removal of K(+) from the perfusate causes pigment hyperdispersion, disclosing a clear correlation between membrane depolarization and pigment aggregation; K(+) channel blockade by Ba(2+) also partially inhibits RPCH action. We suggest that, during RPCH signal transduction, Ca(2+) released from the SER, together with K(+) channel closure, causes chromatophore membrane depolarization, leading to the opening of predominantly N- and/or P/Q-type voltage-gated Ca(2+) channels, and a Ca(2+)/cGMP cascade, resulting in pigment aggregation. J. Exp. Zool. 313A:605-617, 2010. (C) 2010 Wiley-Liss, Inc.

Quantum analysis of the nondegenerate optical parametric oscillator with injected signal

In this paper we study the nondegenerate optical parametric oscillator with injected signal, both analytically and numerically. We develop a perturbation approach which allows us to find approximate analytical solutions, starting from the full equations of motion in the positive-P representation. We demonstrate the regimes of validity of our approximations via comparison with the full stochastic results. We find that, with reasonably low levels of injected signal, the system allows for demonstrations of quantum entanglement and the Einstein-Podolsky-Rosen paradox. In contrast to the normal optical parametric oscillator operating below threshold, these features are demonstrated with relatively intense fields.