In noisy and complex environments, human listeners must segregate the mixture of sound sources arriving at their ears and selectively attend a single source, thereby solving a computationally difficult problem called the cocktail party problem. However, the neural mechanisms underlying these computations are still largely a mystery. Oscillatory synchronization of neuronal activity between cortical areas is thought to provide a crucial role in facilitating information transmission between spatially separated populations of neurons, enabling the formation of functional networks. In this thesis, we seek to analyze and model the functional neuronal networks underlying attention to speech stimuli and find that the Frontal Eye Fields play a central 'hub' role in the auditory spatial attention network in a cocktail party experiment. We use magnetoencephalography (MEG) to measure neural signals with high temporal precision, while sampling from the whole cortex. However, several methodological issues arise when undertaking functional connectivity analysis with MEG data. Specifically, volume conduction of electrical and magnetic fields in the brain complicates interpretation of results. We compare several approaches through simulations, and analyze the trade-offs among various measures of neural phase-locking in the presence of volume conduction. We use these insights to study functional networks in a cocktail party experiment. We then construct a linear dynamical system model of neural responses to ongoing speech. Using this model, we are able to correctly predict which of two speakers is being attended by a listener. We then apply this model to data from a task where people were attending to stories with synchronous and scrambled videos of the speakers' faces to explore how the presence of visual information modifies the underlying neuronal mechanisms of speech perception. This model allows us to probe neural processes as subjects listen to long stimuli, without the need for a trial-based experimental design. We model the neural activity with latent states, and model the neural noise spectrum and functional connectivity with multivariate autoregressive dynamics, along with impulse responses for external stimulus processing. We also develop a new regularized Expectation-Maximization (EM) algorithm to fit this model to electroencephalography (EEG) data.

Active movements are a key feature of human behavior. Even when we do not move our limbs we almost never stop guiding our eyes. As a minimal but omnipresent form of behavior, fast eye movements, called saccades, sample the visual world and determine to a large extent what we perceive. Despite being an integral part of visual perception, prevalent research practice treats the human subject as a passive observer who fixates a spot on the screen and is not allowed to move. Yet, learning sensorimotor interactions by active exploration in order to predict future changes and guide actions seems to be a fundamental principle of neural organization. This results in neural patterns of active behavior that can be fundamentally different from the neural processes revealed in movement-restricted laboratory settings questioning the transferability of results from experimental paradigms demanding fixation to real-world free viewing behavior. In this thesis, we aim at studying the neural mechanisms underlying free viewing behavior. In order to assess the fast, flexible and possibly distributed neural dynamics of active vision, we established a procedure for studying eye movements in magnetoencephalography (MEG) and investigated oscillatory signatures associated with sensorimotor processes of eye movements and saccade target selection, two fundamental processes of active vision. Electroencephalography (EEG) and MEG can non-invasively measure fast neural dynamics and hence seem ideally suited for studying active vision in humans. However, artifacts related to eye movements confound both EEG and MEG signals, and a thorough handling of these artifacts is crucial for investigating neural activities during active movements. Mostly, cleaning of ocular artifacts has been performed for occasional eye movements and blinks in fixation paradigms in EEG. Less is known about the impact of ocular artifacts and especially the saccadic spike on MEG. As a first step to enable active vision studies in MEG, we investigated ocular artifacts and possible ways of their separation from neural signals in MEG. We show that the saccadic spike seriously distorts the spatial and spectral characteristics of the MEG signal (Study 2). We further tested if electrooculogram (EOG) based regression is feasible for corneo-retinal artifact removal (Study 1). Due to an often-raised concern, we addressed if EOG regression eliminates neural activity when applied for MEG. Our results do not indicate such susceptibility and we conclude that EOG regression for removing the corneo-retinal artifact in MEG is suitable. Based on insights from both studies, we established an artifact handling procedure including EOG regression and independent component analysis (ICA) to assess the neural dynamics of active vision. In Study 3, we investigated spectral signatures of neuronal activity across cortex underlying saccade preparation, execution and re-fixation in a delayed saccade task. During preparation and execution, we found a dichotomic signature of gamma power increases and beta power decreases in widespread cortical areas related to saccadic control, including fronto-parietal structures. Saccade direction specific signatures resided in hemisphere lateralized changes in low gamma and alpha power in posterior parietal cortex during preparation extending to extrastriate areas during re-fixation. Real-world behavior implies the constant need to flexibly select actions between competing behavioral alternatives depending on both sensory input and internal states. In order to assess internally motivated viewing behavior, we compared neuronal activity of externally cued saccades with saccades to freely chosen, equally valuable targets. We found gamma band specific power increases in fronto-parietal areas that are likely to reflect a fast transient process of action guidance for sensory-guided saccades and a sustained process for internally selecting between competing behavioral alternatives. The sustained signature of internal action selection suggests that a decision between spatially oriented movements is mediated within sensorimotor structures by neural competition between assemblies encoding parallel evolving movement plans. Since our observations support the assumption that a decision emerges through the distributed consensus of neural activities within effector specific areas rather than in a distinct decision module, they argue for the importance of studying mental processes within their ecologically valid and active context. This thesis shows the feasibility of studying neural mechanisms of active vision in MEG and provides important steps for studying neurophysiological correlates of free viewing in the future. The observed spectrally specific, distributed signatures highlight the importance of assessing fast oscillatory dynamics across the cortex for understanding neural mechanisms mediating real-world active behavior.

No description available.

Stroke is the main cause of adult motor disability. Nevertheless, recent meta-analyses show that the theoretical models conceived to explain the post-stroke brain reorganization are inaccurate and therefore misleading in laying the theoretical foundation for rehabilitation protocols. Mixed results are reported especially in works investigating the excitability properties of the stroke injured brain. Shedding light on the reasons that brought to such mixed results is the central topic of this doctoral thesis. In particular, this confounding evidence is here discussed and tackled in the light of recent works employing brain-state dependent stimulation protocols. These works have been of paramount importance, as they showed that the effects of non-invasive stimulation (TMS and/or rTMS) on the hand knob of the motor cortex depend on the instantaneous sensorimotor state. This local state is largely determined by the phase of the mu-alpha oscillations, with the negative peak representing a high excitability condition. Brain-state dependent results show that controlling for the local state at the moment of stimulation is crucial in order to reduce variability in studies investigating cortical excitability: an approach that has never been employed in stroke literature, so far. In this doctoral thesis, new evidence is provided on affected and unaffected hemispheres’ excitability properties depending on the local state at the moment of stimulation. This previously uncontrolled state-dependent variability is here proposed as one of the factors at the basis of the mixed results in stroke literature. Furthermore, the current models aimed at explaining post-stroke brain reorganization do not take into account factors that recent works suggest might contribute to stroke recovery. In fact, it is here suggested that: interhemispheric inhibition should not be interpreted as competition, structural reserve should be assessed also at the level of the corpus callosum, diaschisis processes should be taken into account and structural and functional connectivity patterns should be included in patients’ assessment. Finally, the excitability properties of the stroke brain have been often inferred comparing stroke patients’ with young healthy controls’. In this regard, it is here proposed that only healthy peers should be included in the control groups, as brain structural changes due to healthy aging have an impact on corticospinal excitability. The aforementioned functional and structural issues are addressed in the following chapters by means of different techniques (i.e. TMS-EEG, MRI, MEG). In particular, in Chapter 1, a new framework of post-stroke brain reorganization is proposed, in which previously over-looked factors are suggested to be essential in the understanding of the potential plastic changes following stroke. Specifically, a new account where interhemispheric inhibition is interpreted in terms of integration and not competition, is supported. Moreover, the proposed framework includes recent pieces of evidence suggesting that structural reserve should be evaluated in the individual patient not only at the level of the cortex, but also in the different sections of the callosum. Finally, it is proposed that structural damage is not static, but rather dynamic as it continues also after the stroke episode through dischiasis processes. In Chapter 2, new knowledge is provided on the different excitability properties of the two hemispheres of stroke patients. In this chapter, TMS-EEG data of stimulation on both the affected and unaffected motor cortex in severe chronic strokes are analysed with a brain-state dependent approach. For the first time, it is shown that the excitability properties of the affected and unaffected hemispheres differ as the local state at the moment of stimulation influences the two hemispheres’ response differently. In particular, the strong and simplified TMS-evoked response in the affected hemisphere, previously reported in severe patients, is shown to depend on a disruption of the differentiation between the high and low excitability states of the motor cortex, determined by the instantaneous phase of alpha oscillations. This low differentiation between excitability states in the affected hemisphere should be systematically investigated, as it could be a potential feature of patients who experience poor recovery. Furthermore, in Chapter 3, connectivity at the individual alpha peak is investigated in a big cohort of healthy participants, in a resting state MEG dataset. This work was implemented because alpha connectivity networks have been shown to predict stroke recovery. For this reason, there is a necessity to reliably assess connectivity at alpha before and after rehabilitation, as this could be informative on the efficacy of rehabilitation. Specifically, it is shown that using complementary phase-coherence metrics is more effective to estimate connectivity patterns at source level. This compound approach is proposed as a tool to better control the modulatory effects of rehabilitation stimulation protocols, in order to identify which are the changes in activity patterns that are potentially responsible for recovery. Finally, in Chapter 4 brain structural changes associated with healthy aging are investigated in a big cohort of participants aged between 18 and 90 years old, both in terms of cortical thinning and cortical myelin concentration loss. In particular, given recent evidence on the relationship between cortical myelin content and corticospinal excitability, it is shown that age-dependent myelin loss occurs mostly at the level of the premotor, motor and sensory cortices. These structural changes need to be taken into account when stroke patients are compared with controls. In fact, since stroke patients are often in their elderly, these age-related structural changes need to be controlled by including only age-matched healthy participants in control groups, as this is not often a fulfilled criterion in stroke studies. To conclude, this doctoral thesis proposes that the current models’ inaccuracy depends on 1) patients’ individual structural and functional factors that have not been taken into account in previous models of brain reorganization post-stroke (Chapter 1), 2) brain-state dependent variability in stimulation effects that have not been controlled for in stroke literature (Chapter 2), 3) a lack of a systematic method to assess the effects of stimulation rehabilitation protocols (Chapter 3) and 4) structural brain changes due to healthy aging, that affect also the stroke brain, and that are not taken into account when patients are compared with young controls in corticospinal excitability studies (Chapter 4). To the author’s knowledge, this is the first work aimed at explaining mixed results in stroke literature from different perspectives and using different neuroimaging techniques for functional and structural anomalies, exploiting recent brain-state dependent approaches for the analysis of stroke patients’ data.

No description available.

Dans le cadre de l’évaluation pré-chirurgicale de patients épileptiques, il est impératif de déterminer la spécialisation hémisphérique du langage, ainsi que de localiser les aires du langage au sein de cet hémisphère. De nouvelles méthodes d’évaluation non- invasives doivent être mises au point afin de diminuer les risques associés aux méthodes plus invasives telles que le test à l’amobarbital intracarotidien (TAI). L’objectif principal de cette thèse est donc de développer un protocole d’évaluation pré-chirurgicale alternatif et non-invasif à l’aide de la magnétoencéphalographie (MEG) pour la latéralisation et la localisation du langage, incluant la mémoire verbale qui serait éventuellement accessible à une population pédiatrique francophone épileptique. L’article 1 présente une recension de la littérature résumant les différentes études en MEG ayant pour objectif l’évaluation pré-chirurgicale du langage. Trente-sept articles en MEG ont été analysés pour déterminer quelles tâches permettaient d’obtenir les meilleurs résultats de latéralisation intrahémisphérique et de localisation du langage pour l’évaluation du langage réceptif et expressif chez des sujets neurologiquement sains et épileptiques. Parmi les tests retenus, l’épreuve de reconnaissance de mots permet d’évaluer le langage réceptif et la mémoire verbale, tandis que des épreuves de fluence verbale telles que la génération de verbes permettent d’évaluer le langage expressif de façon à obtenir de très bons résultats. L’article 2 a permis de valider un protocole auprès de sujets neurologiquement sains à l’aide des épreuves identifiées dans l’article 1. Le protocole utilisé comprend une tâche de langage réceptif et de mémoire verbale (une épreuve de reconnaissance de mots) et une tâche de langage expressif (une épreuve de fluence verbale). Suite à la validation du protocole à l’aide d’analyses par composantes principales, les épreuves ont été administrées à un groupe de patients épileptiques. Les index de latéralité et les analyses de sources i révèlent que la MEG permet de localiser et de latéraliser les fonctions langagières et pourrait donc être utilisée comme méthode d'évaluation du langage lors de l'évaluation pré- chirurgicale auprès de patients épileptiques. Toutefois, alors que l’épreuve de mémoire verbale a permis d’obtenir les meilleurs résultats auprès de l’ensemble des participants, l’épreuve de fluence verbale n’a fourni des informations supplémentaires que chez un seul patient et chez aucun participant neurologiquement sain. En résumé, les deux articles de cette thèse démontrent le potentiel clinique de la MEG pour l’évaluation pré-chirurgicale de patients souffrant d’une épilepsie réfractaire. / It is imperative to determine hemispheric dominance of language and to localize language areas in this hemisphere as part of the presurgical evaluation of epileptic patients. New non-invasive evaluation methods should be developed to reduce the risks associated with more invasive methods such as the intracarotid sodium amobarbital procedure (ISAP). The main objective of this dissertation is to develop an alternative non-invasive presurgical evaluation protocol using magnetoencephalography (MEG) to determine the lateralisation and localization of language and verbal memory that would eventually be accessible to a French-speaking paediatric population suffering from epilepsy. Article 1 is a literature review summarizing MEG studies on the presurgical evaluation of language. Thirty-seven articles were analyzed to determine which tasks allow to obtain the best results with regards to language lateralization and localization with MEG in the assessment of receptive and expressive language in neurologically healthy individuals as well as epileptic patients. We conclude that the word recognition task is well suited to assess receptive language and verbal memory, while a verbal fluency task, such as the verb generation task allows the assessment of expressive language with very good results. Article 2 allowed the validation of a language protocol with neurological healthy participants based on the tasks identified in article 1. The protocol included a task evaluating receptive language functions and verbal memory (word recognition task) and a task evaluation expressive language (verbal fluency). After having validated the tasks using principal component analysis, they were administered to a group of patients with epilepsy. Laterality findings and source analysis showed that MEG can be used to lateralize and localize language and could therefore be used in the presurgical assessment of patients with epilepsy. However, the verbal fluency task, when compared to the verbal memory task, provided additional information for only one patient and in none of the neurologically healthy participants. In summary, the two articles of this dissertation demonstrate the clinical potential of MEG for the presurgical evaluation of patients with refractory epilepsy.

Il existe deux façons d'analyser l'activité cérébrale acquise en magnétoencéphalographie (MEG) : soit en moyennant les réponses suscitées par la répétition d'un stimulus afin d'observer le « champ évoqué »; soit en décomposant le signal en bandes oscillatoires (tel que l'alpha, le bêta ou le gamma), chacune étant associée à différents rôles fonctionnels. Ces méthodes ne prennent cependant pas compte de la complexité de l'activité cérébrale dont l'essentiel est arythmique, notamment au repos. Pour pallier à cela, une autre approche consiste à analyser le spectre de puissance en 1/f observable dans les très basses fréquences, une caractéristique des systèmes dont la dynamique est invariante d'échelle. Pour savoir si cette propriété joue un quelconque rôle dans le fonctionnement cérébral et si elle a des conséquences sur le comportement, nous avons établit un paradigme d'apprentissage visuel permettant d'observer de la plasticité fonctionnelle au cours d'une session MEG. Pour avoir un entraînement optimal, nous avons développé de nouveaux stimuli audiovisuels (AV) (une texture acoustique associée à un nuage de points colorés en mouvement) permettant une intégration multisensorielle et de ce fait un meilleur apprentissage que celui apporté par un entraînement visuel seul (V) ou accompagné d'un bruit acoustique (AVn). Nous avons ensuite étudié les corrélats neuronaux de ces trois types d'apprentissage par l'analyse classique des champs évoqués. Une fois l'activité reconstruite sur la surface corticale de chaque individu à l'aide de MNE-dSPM, nous avons identifié le réseau impliqué dans la tâche au sein de chaque groupe. En particulier, la plasticité sélective observée dans l'aire hMT+ associée au traitement du mouvement visuel corrélait avec les progressions comportementales des individus et était soutenue en AV par un plus vaste réseau comprenant notamment des aires multisensorielles. Parallèlement, nous avons exploré les liens reliant le comportement et les propriétés d'invariance d'échelle de ces mêmes signaux MEG reconstruits sur le cortex. Tandis que la plupart des études se limitent à analyser l'auto-similarité (une caractéristique globale synonyme de longue mémoire), nous avons aussi considéré les fluctuations locales (c-à-d la multifractalité) au moyen de l'analyse WLBMF. Nous avons trouvé des modulations couplées de l'auto-similarité et de la multifractalité dans des régions similaires à celles révélées par l'analyse des champs évoqués. Plus surprenant, Le degré de multifractalité relevé dans chaque individu convergeait durant l'entraînement vers un même attracteur reflétant la performance comportementale asymptotique. / The analysis of Human brain activity in magnetoencephalography (MEG) can be generally conducted in two ways: either by focusing on the average response evoked by a stimulus repeated over time, more commonly known as an ``event-related field'' (ERF), or by decomposing the signal into functionally relevant oscillatory or frequency bands (such as alpha, beta or gamma). However, the major part of brain activity is arrhythmic and these approaches fail in describing its complexity, particularly in resting-state. As an alternative, the analysis of the 1/f-type power spectrum observed in the very low frequencies, a hallmark of scale-free dynamics, can overcome these issues. Yet it remains unclear whether this scale-free property is functionally relevant and whether its fluctuations matter for behavior. To address this question, our first concern was to establish a visual learning paradigm that would entail functional plasticity during an MEG session. In order to optimize the training effects, we developed new audiovisual (AV) stimuli (an acoustic texture paired with a colored visual motion) that induced multisensory integration and indeed improved learning compared to visual training solely (V) or accompanied with acoustic noise (AVn). This led us to investigate the neural correlates of these three types of training using first a classical method such as the ERF analysis. After source reconstruction on each individual cortical surface using MNE-dSPM, the network involved in the task was identified at the group-level. The selective plasticity observed in the human motion area (hMT+) correlated across all individuals with the behavioral improvement and was supported by a larger network in AV comprising multisensory areas. On the basis of these findings, we further explored the links between the behavior and scale-free properties of these same source-reconstructed MEG signals. Although most studies restricted their analysis to the global measure of self-similarity (i.e. long-range fluctuations), we also considered local fluctuations (i.e. multifractality) by using the Wavelet Leader Based Multifractal Formalism (WLBMF). We found intertwined modulations of self-similarity and multifractality in the same cortical regions as those revealed by the ERF analysis. Most astonishing, the degree of multifractality observed in each individual converged during the training towards a single attractor that reflected the asymptotic behavioral performance in hMT+. Finally, these findings and their associated methodological issues are compared with the ones that came out from the ERF analysis.

Este trabalho teve como objetivo principal propor procedimentos de avaliação de metodologia de coleta e análise de solventes no ar em ambiente de trabalho. Utilizou-se o metanol em mistura MEG (Metanol 33%, Etanol 60% e Gasolina 7%) como exemplo. Na época, a mistura MEG estava sendo indicada para substituir o etanol combustível, devido a sua falta no mercado. Foi montada uma atmosfera controlada dinâmica por injeção, para gerar concentrações conhecidas de vapor metanol em mistura MEG. Essa atmosfera permaneceu estável ao redor de 7,5 horas. O limite de detecção método analítico do CG-DIC, utilizado para o acompanhamento das concentrações geradas, foi de 18mg/m3 para um volume amostrado de 2mL coletado por uma válvula de coleta. Estando a atmosfera controlada apta para os testes de validação, foram realizados estudos de volume de \"breakthrough\" para coleta de metanol em tubos de sílica gel, com a finalidade de se conhecer o volume a ser amostrado em campo. Esses estudos foram realizados levando em conta as possíveis interferências de etanol, gasolina e elevada umidade relativatotal . O tempo de coleta de amostra para a mistura com umidade relativa recomendado para uma vazão de 200mL/min foi de 15 minutos. O método analítico utilizado, para a determinação de metanol coletado em tubos de sílica gel e analisado por CG-DIC. foi uma adaptação de métodos encontrados na literatura do NIOSH, para as condições da mistura. Após encontrar a melhor seletividade para o menor tempo de retenção do metanol em mistura MEG, foi determinado o limite de detecção de 9,7x 10-3µg. A precisão do método analítico encontrada foi de 2,1% e a exatidão esteve entre 89% e 97%. Estabelecida a metodologia analítica, as melhores condições de atmosfera controlada e o volume de coleta de amostra, foram realizados estudos que integram a coleta e análise ou seja, precisão e exatidão de coleta e análise e estabilidade de armazenamento. Para estudos de precisão e exatidão de coleta e análise, foram geradas concentrações de metanol em mistura MEG, com umidade ao redor de 0,5, 1 e 2 vezes o limite de tolerância brasileiro(156ppm); dessorção em água e análise por CGDIC. O coeficiente de variação total calculado para o método de coleta e análise de metanol em mistura MEG foi de 7,3% (precisão) e wazzu exatidão esteve entre 90% e 101%. As concentrações de metanol foram estáveis, em tubos adsorventes armazenados em geladeira, durante o período de estudo de 14 dias. Porém, foi constatada uma migração de metanol através da camada analítica para a camada controle de forma gradativa com o passar do tempo. As soluções, a partir da dessorção dos tubos armazenadas em geladeira, foram estáveis durante o período de estudo de 45 dias. Além desses experimentos foram também realizados estudos de capacidade de adsorção da sílica gel para metanol levando em conta os interferentes, e verificado se eram capazes de dessorver totalmente o metanol, após um certo período de coleta. Constatou-se que após 30 minutos de coleta, os interferentes (etanol, umidade e gasolina) entravam em equilíbrio com o adsorvente e o metanol. / The main objective of this work is to propose procedures for the evaluation of the methodology of the sampling and analysis of solvents in the air in the workplace. Methanol in a MEG mixture (Methanol 33%, Ethanol 60%, and Gasoline 7%) was used as an example. At that time the MEG mixture was being proposed to substitute ethanol fuel due to its shortages on the Brazilian market. An atmosphere, dynamically controlled by injection, was produced to generate the known concentrations of methanol vapour in MEG mixtures. This atmosphere remained stable for about 450 minutes. The detection Iimit, using the GC-FID analytical method to monitor the concentrations produced, was 18mg/m3 for a sample volume of 2mL sampled from a sampling valve. The controlled atmosphere being suitable for the validation tests, studies of the breakthrough volume for the sampling of methanol in silica gel tubes were carried out, with the aim of discovering the total volume to be sampled in the field. These studies were carried out taking into account the possible interference of ethanol, gasoline and the high relative humidity. The sampling time for the mixture with the recommended relative humidity with a flow rate of 200mL/min was fifteen minutes. The analytical method used for the determination of methanol sampled in silica gel tubes and analyzed by GC-FID, was based on the methods found in the NIOSH and OSHA literature, adapted to the mixture conditions. After finding the best selectivity for the lowest retention time for the methanol in the MEG mixture, the detection Iimit of 9.7 x 10-3µg was determined. The precision of the analytical method was 2.1% and the accuracy was between 89% and 97%. Having established the analytical methodology. the best conditions for the controlled atmosphere and the sampling volume, studies were carried out which integrated the sampling and the analysis, that is, the precision and the accuracy of the sampling and the analysis and the storage stability. To study the precision and the accuracy of the sampling and analysis, concentrations 1 or 2 times that of the Brazilian tolerance Iimit of methanol (156 ppm) in MEG mixture, with humidity around 0.5, were generated, using dessorption in water and analysis by GC-FID. The total coefficient of variation calculated for the method of sampling and analysis of methanol in MEG mixture was 7.3% (precision) and the accuracy was between 90% and 101% . The concentrations of methanol were stable, in adsorption tubes stored in the refrigerator, throughout the study period of fifteen days. However, a gradual migration of methanol through the analytical layer to the control layer with the passage of time was noted. The solutions from the dessorption of the tubes, stored in the refrigerator, were stable throughout the study period of 45 days. Besides these experiments, studies of the adsorption capacity of silica gel for methanol, taking into account the interferents, were also carried out, and it was verified whether these were capable of the total dessorption of the methanol, after a certain sampling period. It was noted that after thirty minutes of sampling, the interferents came into equilibrium with the adsorbent and the methanol.

Este trabalho teve como objetivo principal propor procedimentos de avaliação de metodologia de coleta e análise de solventes no ar em ambiente de trabalho. Utilizou-se o metanol em mistura MEG (Metanol 33%, Etanol 60% e Gasolina 7%) como exemplo. Na época, a mistura MEG estava sendo indicada para substituir o etanol combustível, devido a sua falta no mercado. Foi montada uma atmosfera controlada dinâmica por injeção, para gerar concentrações conhecidas de vapor metanol em mistura MEG. Essa atmosfera permaneceu estável ao redor de 7,5 horas. O limite de detecção método analítico do CG-DIC, utilizado para o acompanhamento das concentrações geradas, foi de 18mg/m3 para um volume amostrado de 2mL coletado por uma válvula de coleta. Estando a atmosfera controlada apta para os testes de validação, foram realizados estudos de volume de \"breakthrough\" para coleta de metanol em tubos de sílica gel, com a finalidade de se conhecer o volume a ser amostrado em campo. Esses estudos foram realizados levando em conta as possíveis interferências de etanol, gasolina e elevada umidade relativatotal . O tempo de coleta de amostra para a mistura com umidade relativa recomendado para uma vazão de 200mL/min foi de 15 minutos. O método analítico utilizado, para a determinação de metanol coletado em tubos de sílica gel e analisado por CG-DIC. foi uma adaptação de métodos encontrados na literatura do NIOSH, para as condições da mistura. Após encontrar a melhor seletividade para o menor tempo de retenção do metanol em mistura MEG, foi determinado o limite de detecção de 9,7x 10-3µg. A precisão do método analítico encontrada foi de 2,1% e a exatidão esteve entre 89% e 97%. Estabelecida a metodologia analítica, as melhores condições de atmosfera controlada e o volume de coleta de amostra, foram realizados estudos que integram a coleta e análise ou seja, precisão e exatidão de coleta e análise e estabilidade de armazenamento. Para estudos de precisão e exatidão de coleta e análise, foram geradas concentrações de metanol em mistura MEG, com umidade ao redor de 0,5, 1 e 2 vezes o limite de tolerância brasileiro(156ppm); dessorção em água e análise por CGDIC. O coeficiente de variação total calculado para o método de coleta e análise de metanol em mistura MEG foi de 7,3% (precisão) e wazzu exatidão esteve entre 90% e 101%. As concentrações de metanol foram estáveis, em tubos adsorventes armazenados em geladeira, durante o período de estudo de 14 dias. Porém, foi constatada uma migração de metanol através da camada analítica para a camada controle de forma gradativa com o passar do tempo. As soluções, a partir da dessorção dos tubos armazenadas em geladeira, foram estáveis durante o período de estudo de 45 dias. Além desses experimentos foram também realizados estudos de capacidade de adsorção da sílica gel para metanol levando em conta os interferentes, e verificado se eram capazes de dessorver totalmente o metanol, após um certo período de coleta. Constatou-se que após 30 minutos de coleta, os interferentes (etanol, umidade e gasolina) entravam em equilíbrio com o adsorvente e o metanol. / The main objective of this work is to propose procedures for the evaluation of the methodology of the sampling and analysis of solvents in the air in the workplace. Methanol in a MEG mixture (Methanol 33%, Ethanol 60%, and Gasoline 7%) was used as an example. At that time the MEG mixture was being proposed to substitute ethanol fuel due to its shortages on the Brazilian market. An atmosphere, dynamically controlled by injection, was produced to generate the known concentrations of methanol vapour in MEG mixtures. This atmosphere remained stable for about 450 minutes. The detection Iimit, using the GC-FID analytical method to monitor the concentrations produced, was 18mg/m3 for a sample volume of 2mL sampled from a sampling valve. The controlled atmosphere being suitable for the validation tests, studies of the breakthrough volume for the sampling of methanol in silica gel tubes were carried out, with the aim of discovering the total volume to be sampled in the field. These studies were carried out taking into account the possible interference of ethanol, gasoline and the high relative humidity. The sampling time for the mixture with the recommended relative humidity with a flow rate of 200mL/min was fifteen minutes. The analytical method used for the determination of methanol sampled in silica gel tubes and analyzed by GC-FID, was based on the methods found in the NIOSH and OSHA literature, adapted to the mixture conditions. After finding the best selectivity for the lowest retention time for the methanol in the MEG mixture, the detection Iimit of 9.7 x 10-3µg was determined. The precision of the analytical method was 2.1% and the accuracy was between 89% and 97%. Having established the analytical methodology. the best conditions for the controlled atmosphere and the sampling volume, studies were carried out which integrated the sampling and the analysis, that is, the precision and the accuracy of the sampling and the analysis and the storage stability. To study the precision and the accuracy of the sampling and analysis, concentrations 1 or 2 times that of the Brazilian tolerance Iimit of methanol (156 ppm) in MEG mixture, with humidity around 0.5, were generated, using dessorption in water and analysis by GC-FID. The total coefficient of variation calculated for the method of sampling and analysis of methanol in MEG mixture was 7.3% (precision) and the accuracy was between 90% and 101% . The concentrations of methanol were stable, in adsorption tubes stored in the refrigerator, throughout the study period of fifteen days. However, a gradual migration of methanol through the analytical layer to the control layer with the passage of time was noted. The solutions from the dessorption of the tubes, stored in the refrigerator, were stable throughout the study period of 45 days. Besides these experiments, studies of the adsorption capacity of silica gel for methanol, taking into account the interferents, were also carried out, and it was verified whether these were capable of the total dessorption of the methanol, after a certain sampling period. It was noted that after thirty minutes of sampling, the interferents came into equilibrium with the adsorbent and the methanol.

La magnéto-encéphalographie (MEG) mesure l'activité cérébrale avec un excellent décours temporel mais sa localisation sur la surface corticale souffre d'une mauvaise résolution spatiale. Le problème inverse MEG est dit mal-posé et doit de ce fait être régularisé. La parcellisation du cortex en régions de spécificité fonctionnelle proche constitue une régularisation spatiale pertinente du problème inverse MEG. Nous proposons une méthode de parcellisation du cortex entier à partir de la connectivité anatomique cartographiée par imagerie de diffusion. Au sein de chaque aire d'une préparcellisation, la matrice de corrélation entre les profils de connectivité des sources est partitionnée. La parcellisation obtenue est alors mise à jour en testant la similarité des données de diffusion de part et d'autre des frontières de la préparcellisation. C'est à partir de ce résultat que nous contraignons spatialement le problème inverse MEG. Dans ce contexte, deux méthodes sont développées. La première consiste à partitionner l'espace des sources au regard de la parcellisation. L'activité corticale est alors obtenue sur un ensemble de parcelles. Afin de ne pas forcer les sources à avoir exactement la même intensité au sein d'une parcelle, nous développons une méthode alternative introduisant un nouveau terme de régularisation qui, lorsqu'il est minimisé, tend à ce que les sources d'une même parcelle aient des valeurs de reconstruction proches. Nos méthodes de reconstruction sont testées et validées sur des données simulées et réelles. Une application clinique dans le cadre du traitement de données de sujets épileptiques est également réalisée.