Conception d'un outil informatique basé sur un réseau de neurones artificiels pour la classification automatique des stades du sommeil

L’objectif de notre travail est de développer un outil d’analyse automatique des stades du sommeil basé sur les réseaux de neurones artificiels (RNA). Dans ce papier nous présentons notre démarche pour la conception de cet outil. La première difficulté consiste dans le choix de la représentation des signaux physiologiques et en particulier de l’électroencéphalogramme (EEG). Une fois la représentation adoptée, l’étape suivante est la conception du réseau de neurones optimal déterminé par un processus d’apprentissage et de validation sur les données issues d’un ensemble d'enregistrements de nuits de sommeil. Le résultat obtenu avec un taux de 63% de bonne classification pour six stades, nous incite à approfondir l’étude de cette problématique aux niveaux représentation et conception pour améliorer les performances de notre outil.

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.

Theta EEG oscillatory activity and auditory change detection

The mismatch negativity is an electrophysiological marker of auditory change detection in the event-related brain potential and has been proposed to reflect an automatic comparison process between an incoming stimulus and the representation of prior items in a sequence. There is evidence for two main functional subcomponents comprising the MMN, generated by temporal and frontal brain areas, respectively. Using data obtained in an MMN paradigm, we performed time-frequency analysis to reveal the changes in oscillatory neural activity in the theta band. The results suggest that the frontal component of the MMN is brought about by an increase in theta power for the deviant trials and, possibly, by an additional contribution of theta phase alignment. By contrast, the temporal component of the MMN, best seen in recordings from mastoid electrodes, is generated by phase resetting of theta rhythm with no concomitant power modulation. Thus, frontal and temporal MMN components do not only differ with regard to their functional significance but also appear to be generated by distinct neurophysiological mechanisms.

A novel approach to the detection of synchronisation in EEG based on empirical mode decomposition

Transient neural assemblies mediated by synchrony in particular frequency ranges are thought to underlie cognition. We propose a new approach to their detection, using empirical mode decomposition (EMD), a data-driven approach removing the need for arbitrary bandpass filter cut-offs. Phase locking is sought between modes. We explore the features of EMD, including making a quantitative assessment of its ability to preserve phase content of signals, and proceed to develop a statistical framework with which to assess synchrony episodes. Furthermore, we propose a new approach to ensure signal decomposition using EMD. We adapt the Hilbert spectrum to a time-frequency representation of phase locking and are able to locate synchrony successfully in time and frequency between synthetic signals reminiscent of EEG. We compare our approach, which we call EMD phase locking analysis (EMDPL) with existing methods and show it to offer improved time-frequency localisation of synchrony.

A graph-theoretical approach in brain functional networks. Possible implications in EEG studies

Abstract Background Recently, it was realized that the functional connectivity networks estimated from actual brain-imaging technologies (MEG, fMRI and EEG) can be analyzed by means of the graph theory, that is a mathematical representation of a network, which is essentially reduced to nodes and connections between them. Methods We used high-resolution EEG technology to enhance the poor spatial information of the EEG activity on the scalp and it gives a measure of the electrical activity on the cortical surface. Afterwards, we used the Directed Transfer Function (DTF) that is a multivariate spectral measure for the estimation of the directional influences between any given pair of channels in a multivariate dataset. Finally, a graph theoretical approach was used to model the brain networks as graphs. These methods were used to analyze the structure of cortical connectivity during the attempt to move a paralyzed limb in a group (N=5) of spinal cord injured patients and during the movement execution in a group (N=5) of healthy subjects. Results Analysis performed on the cortical networks estimated from the group of normal and SCI patients revealed that both groups present few nodes with a high out-degree value (i.e. outgoing links). This property is valid in the networks estimated for all the frequency bands investigated. In particular, cingulate motor areas (CMAs) ROIs act as ‘‘hubs’’ for the outflow of information in both groups, SCI and healthy. Results also suggest that spinal cord injuries affect the functional architecture of the cortical network sub-serving the volition of motor acts mainly in its local feature property. In particular, a higher local efficiency El can be observed in the SCI patients for three frequency bands, theta (3-6 Hz), alpha (7-12 Hz) and beta (13-29 Hz). By taking into account all the possible pathways between different ROI couples, we were able to separate clearly the network properties of the SCI group from the CTRL group. In particular, we report a sort of compensatory mechanism in the SCI patients for the Theta (3-6 Hz) frequency band, indicating a higher level of “activation” Ω within the cortical network during the motor task. The activation index is directly related to diffusion, a type of dynamics that underlies several biological systems including possible spreading of neuronal activation across several cortical regions. Conclusions The present study aims at demonstrating the possible applications of graph theoretical approaches in the analyses of brain functional connectivity from EEG signals. In particular, the methodological aspects of the i) cortical activity from scalp EEG signals, ii) functional connectivity estimations iii) graph theoretical indexes are emphasized in the present paper to show their impact in a real application.

Topographic time-frequency decomposition of the EEG

Frequency-transformed EEG resting data has been widely used to describe normal and abnormal brain functional states as function of the spectral power in different frequency bands. This has yielded a series of clinically relevant findings. However, by transforming the EEG into the frequency domain, the initially excellent time resolution of time-domain EEG is lost. The topographic time-frequency decomposition is a novel computerized EEG analysis method that combines previously available techniques from time-domain spatial EEG analysis and time-frequency decomposition of single-channel time series. It yields a new, physiologically and statistically plausible topographic time-frequency representation of human multichannel EEG. The original EEG is accounted by the coefficients of a large set of user defined EEG like time-series, which are optimized for maximal spatial smoothness and minimal norm. These coefficients are then reduced to a small number of model scalp field configurations, which vary in intensity as a function of time and frequency. The result is thus a small number of EEG field configurations, each with a corresponding time-frequency (Wigner) plot. The method has several advantages: It does not assume that the data is composed of orthogonal elements, it does not assume stationarity, it produces topographical maps and it allows to include user-defined, specific EEG elements, such as spike and wave patterns. After a formal introduction of the method, several examples are given, which include artificial data and multichannel EEG during different physiological and pathological conditions.

Brain sources of EEG gamma frequency during volitionally meditation-induced, altered states of consciousness, and experience of the self.

Multichannel EEG of an advanced meditator was recorded during four different, repeated meditations. Locations of intracerebral source gravity centers as well as Low Resolution Electromagnetic Tomography (LORETA) functional images of the EEG 'gamma' (35-44 Hz) frequency band activity differed significantly between meditations. Thus, during volitionally self-initiated, altered states of consciousness that were associated with different subjective meditation states, different brain neuronal populations were active. The brain areas predominantly involved during the self-induced meditation states aiming at visualization (right posterior) and verbalization (left central) agreed with known brain functional neuroanatomy. The brain areas involved in the self-induced, meditational dissolution and reconstitution of the experience of the self (right fronto-temporal) are discussed in the context of neural substrates implicated in normal self-representation and reality testing, as well as in depersonalization disorders and detachment from self after brain lesions.

Chow–Liu trees are sufficient predictive models for reproducing key features of functional networks of periictal EEG time-series

Seizure freedom in patients suffering from pharmacoresistant epilepsies is still not achieved in 20–30% of all cases. Hence, current therapies need to be improved, based on a more complete understanding of ictogenesis. In this respect, the analysis of functional networks derived from intracranial electroencephalographic (iEEG) data has recently become a standard tool. Functional networks however are purely descriptive models and thus are conceptually unable to predict fundamental features of iEEG time-series, e.g., in the context of therapeutical brain stimulation. In this paper we present some first steps towards overcoming the limitations of functional network analysis, by showing that its results are implied by a simple predictive model of time-sliced iEEG time-series. More specifically, we learn distinct graphical models (so called Chow–Liu (CL) trees) as models for the spatial dependencies between iEEG signals. Bayesian inference is then applied to the CL trees, allowing for an analytic derivation/prediction of functional networks, based on thresholding of the absolute value Pearson correlation coefficient (CC) matrix. Using various measures, the thus obtained networks are then compared to those which were derived in the classical way from the empirical CC-matrix. In the high threshold limit we find (a) an excellent agreement between the two networks and (b) key features of periictal networks as they have previously been reported in the literature. Apart from functional networks, both matrices are also compared element-wise, showing that the CL approach leads to a sparse representation, by setting small correlations to values close to zero while preserving the larger ones. Overall, this paper shows the validity of CL-trees as simple, spatially predictive models for periictal iEEG data. Moreover, we suggest straightforward generalizations of the CL-approach for modeling also the temporal features of iEEG signals.

The Brain as a Distributed Intelligent Processing System: An EEG Study

Background: Various neuroimaging studies, both structural and functional, have provided support for the proposal that a distributed brain network is likely to be the neural basis of intelligence. The theory of Distributed Intelligent Processing Systems (DIPS), first developed in the field of Artificial Intelligence, was proposed to adequately model distributed neural intelligent processing. In addition, the neural efficiency hypothesis suggests that individuals with higher intelligence display more focused cortical activation during cognitive performance, resulting in lower total brain activation when compared with individuals who have lower intelligence. This may be understood as a property of the DIPS. Methodology and Principal Findings: In our study, a new EEG brain mapping technique, based on the neural efficiency hypothesis and the notion of the brain as a Distributed Intelligence Processing System, was used to investigate the correlations between IQ evaluated with WAIS (Whechsler Adult Intelligence Scale) and WISC (Wechsler Intelligence Scale for Children), and the brain activity associated with visual and verbal processing, in order to test the validity of a distributed neural basis for intelligence. Conclusion: The present results support these claims and the neural efficiency hypothesis.

Methods of imagetic information representation

The aim of this paper is to highlight some of the methods of imagetic information representation, reviewing the literature of the area and proposing a model of methodology adapted to Brazilian museums. An elaboration of a methodology of imagetic information representation is developed based on Brazilian characteristics of information treatment in order to adapt it to museums. Finally, spreadsheets that show this methodology are presented.

The game inside the school: a representation of 5(th) graders physical education classes

Since the beginning of Physical Education entrance in the brazilin public schools, the game has been frequently used as content, and in the course of time that practice seems to be intensified. In spite of many approaches of different purposes to justify its pedagogic usefulness, the game has been used as an indiscriminate way due to the fascination that it provides to the students. The present study searches for a description and analysis of children`s (10-12 years old) attitudes behaviors in games, on Physical Education classes, inside a public school. The study was accomplished with the researcher also attending as a teacher (action research). For the accomplishment of the study 55 children were filmed in four different games, of different kinds (exposed, transformed, and spontaneous). The classes` description and analysis were focused in the attitude axis and it was defined four topics for the discussion: Conflicts, Respect of rules, Expressiveness, and Competitiveness. The relationship between the individual with the game and its culture were pointed as the main characteristics in the configuration of the ludicrous activity atmosphere. It was also possible to observe specific situations of this relationship, once the games were limited to the social games (Piaget category), in a school atmosphere where children have students roles. Due to the obtained results, the study proposes a reflexive practice in which the students notice their own attitudes and try to adapt the game to their needs and not he other way around. In this perspective, the teacher has an important mediator roll, once he will be responsible to point out the students` difficulties and promote discussions in favor to provide teamwork.

Antinociceptive effect of stimulating the occipital or retrosplenial cortex in rats

A role for the occipital or retrosplenial cortex in nociceptive processing has not been demonstrated yet, but connections from these cortices to brain structures involved in descending pain-inhibitory mechanisms were already demonstrated. This study demonstrated that the electrical stimulation of the occipital or retrosplenial cortex produces antinociception in the rat tail-flick and formalin tests. Bilateral lesions of the dorsolateral funiculus abolished the effect of cortical stimulation in the tail-flick test. Injection of glutamate into the same targets was also antinociceptive in the tail-flick test. No rats stimulated in the occipital or retrosplenial cortex showed any change in motor performance on the Rota-rod test, or had epileptiform changes in the EEG recording during or up to 3 hours after stimulation. The antinociception induced by occipital cortex stimulation persisted after neural block of the retrosplenial cortex. The effect of retrosplenial cortex stimulation also persisted after neural block of the occipital cortex. We conclude that stimulation of the occipital or retrosplenial cortex in rats leads to antinociception activating distinct descending pain-inhibitory mechanisms, and this is unlikely to result from a reduced motor performance or a postictal phenomenon. Perspective: This study presents evidence that stimulation of the retrosplenial or occipital cortex produces antinociception in rat models of acute pain. These findings enhance our understanding of the role of the cerebral cortex in control of pain. (C) 2010 by the American Pain Society

Experimental results on the nonlinear H(infinity) control via quasi-LPV representation and game theory for wheeled mobile robots

In this paper, nonlinear dynamic equations of a wheeled mobile robot are described in the state-space form where the parameters are part of the state (angular velocities of the wheels). This representation, known as quasi-linear parameter varying, is useful for control designs based on nonlinear H(infinity) approaches. Two nonlinear H(infinity) controllers that guarantee induced L(2)-norm, between input (disturbances) and output signals, bounded by an attenuation level gamma, are used to control a wheeled mobile robot. These controllers are solved via linear matrix inequalities and algebraic Riccati equation. Experimental results are presented, with a comparative study among these robust control strategies and the standard computed torque, plus proportional-derivative, controller.

A novel continuous time representation for the scheduling of pipeline systems with pumping yield rate constraints

Pipeline systems play a key role in the petroleum business. These operational systems provide connection between ports and/or oil fields and refineries (upstream), as well as between these and consumer markets (downstream). The purpose of this work is to propose a novel MINLP formulation based on a continuous time representation for the scheduling of multiproduct pipeline systems that must supply multiple consumer markets. Moreover, it also considers that the pipeline operates intermittently and that the pumping costs depend on the booster stations yield rates, which in turn may generate different flow rates. The proposed continuous time representation is compared with a previously developed discrete time representation [Rejowski, R., Jr., & Pinto, J. M. (2004). Efficient MILP formulations and valid cuts for multiproduct pipeline scheduling. Computers and Chemical Engineering, 28, 1511] in terms of solution quality and computational performance. The influence of the number of time intervals that represents the transfer operation is studied and several configurations for the booster stations are tested. Finally, the proposed formulation is applied to a larger case, in which several booster configurations with different numbers of stages are tested. (C) 2007 Elsevier Ltd. All rights reserved.

MODELING THE UNVOICED COMPONENT IN THE CANONICAL REPRESENTATION OF SPEECH

The canonical representation of speech constitutes a perfect reconstruction (PR) analysis-synthesis system. Its parameters are the autoregressive (AR) model coefficients, the pitch period and the voiced and unvoiced components of the excitation represented as transform coefficients. Each set of parameters may be operated on independently. A time-frequency unvoiced excitation (TFUNEX) model is proposed that has high time resolution and selective frequency resolution. Improved time-frequency fit is obtained by using for antialiasing cancellation the clustering of pitch-synchronous transform tracks defined in the modulation transform domain. The TFUNEX model delivers high-quality speech while compressing the unvoiced excitation representation about 13 times over its raw transform coefficient representation for wideband speech.