No description available.

User-centered computer based learning is an emerging field of interdisciplinary research. Research in diverse areas such as psychology, computer science, neuroscience and signal processing is making contributions the promise to take this field to the next level. Learning systems built using contributions from these fields could be used in actual training and education instead of just laboratory proof-of-concept. One of the important advances in this research is the detection and assessment of the cognitive and emotional state of the learner using such systems. This capability moves development beyond the use of traditional user performance metrics to include system intelligence measures that are based on current neuroscience theories. These advances are of paramount importance in the success and wide spread use of learning systems that are automated and intelligent. Emotion is considered an important aspect of how learning occurs, and yet estimating it and making adaptive adjustments are not part of most learning systems. In this research we focus on one specific aspect of constructing an adaptive and intelligent learning system, that is, estimation of the emotion of the learner as he/she is using the automated training system. The challenge starts with the definition of the emotion and the utility of it in human life. The next challenge is to measure the co-varying factors of the emotions in a non-invasive way, and find consistent features from these measures that are valid across wide population. In this research we use four physiological sensors that are non-invasive, and establish a methodology of utilizing the data from these sensors using different signal processing tools. A validated set of visual stimuli used worldwide in the research of emotion and attention, called International Affective Picture System (IAPS), is used. A dataset is collected from the sensors in an experiment designed to elicit emotions from these validated visual stimuli. We describe a novel wavelet method to calculate hemispheric asymmetry metric using electroencephalography data. This method is tested against typically used power spectral density method. We show overall improvement in accuracy in classifying specific emotions using the novel method. We also show distinctions between different discrete emotions from the autonomic nervous system activity using electrocardiography, electrodermal activity and pupil diameter changes. Findings from different features from these sensors are used to give guidelines to use each of the individual sensors in the adaptive learning environment.

L'hypovigilance du conducteur (que ce soit provoquée par la distraction ou la somnolence) est une des menaces principales pour la sécurité routière. Cette thèse s'encadre dans le projet Toucango, porté par la start-up Innov+, qui vise à construire un détecteur d'hypovigilance en temps réel basé sur la fusion d'un flux vidéo en proche infra-rouge et d'informations environnementales. L'objectif de cette thèse consiste donc à proposer des techniques d'extraction des indices pertinents ainsi que des algorithmes de fusion multimodale qui puissent être embarqués sur le système pour un fonctionnement en temps réel. Afin de travailler dans des conditions proches du terrain, une base de données en conduite réelle a été créée avec la collaboration de plusieurs sociétés de transports. Dans un premier temps, nous présentons un état de l'art scientifique et une étude des solutions disponibles sur le marché pour la détection de l'hypovigilance. Ensuite, nous proposons diverses méthodes basées sur le traitement d'images (pour la détection des indices pertinents sur la tête, yeux, bouche et visage) et de données (pour les indices environnementaux basés sur la géolocalisation). Nous réalisons une étude sur les facteurs environnementaux liés à l'hypovigilance et développons un système d'estimation du risque contextuel. Enfin, nous proposons des techniques de fusion multimodale de ces indices avec l'objectif de détecter plusieurs comportements d'hypovigilance : distraction visuelle ou cognitive, engagement dans une tâche secondaire, privation de sommeil, micro-sommeil et somnolence. / Driver hypovigilance (whether caused by distraction or drowsiness) is one of the major threats to road safety. This thesis is part of the Toucango project, hold by the start-up Innov+, which aims to build a real-time hypovigilance detector based on the fusion of near infra-red video evidence and environmental information. The objective of this thesis is therefore to propose techniques for extracting relevant indices as well as multimodal fusion algorithms that can be embedded in the system for real-time operation. In order to work near ground truth conditions, a naturalistic driving database has been created with the collaboration of several transport companies. We first present a scientific state of the art and a study of the solutions available on the market for hypovigilance detection. Then, we propose several methods based on image (for the detection of relevant indices on the head, eyes, mouth and face) and data processing (for environmental indices based on geolocation). We carry out a study on the environmental factors related to hypovigilance and develop a contextual risk estimation system. Finally, we propose multimodal fusion techniques of these indices with the objective of detecting several hypovigilance behaviors: visual or cognitive distraction, engagement in a secondary task, sleep deprivation, microsleep and drowsiness.

In the analysis of heart rate variability (HRV) are used temporal series that contains the distances between successive heartbeats in order to assess autonomic regulation of the cardiovascular system. These series are obtained from the electrocardiogram (ECG) signal analysis, which can be affected by different types of artifacts leading to incorrect interpretations in the analysis of the HRV signals. Classic approach to deal with these artifacts implies the use of correction methods, some of them based on interpolation, substitution or statistical techniques. However, there are few studies that shows the accuracy and performance of these correction methods on real HRV signals. This study aims to determine the performance of some linear and non-linear correction methods on HRV signals with induced artefacts by quantification of its linear and nonlinear HRV parameters. As part of the methodology, ECG signals of rats measured using the technique of telemetry were used to generate real heart rate variability signals without any error. In these series were simulated missing points (beats) in different quantities in order to emulate a real experimental situation as accurately as possible. In order to compare recovering efficiency, deletion (DEL), linear interpolation (LI), cubic spline interpolation (CI), moving average window (MAW) and nonlinear predictive interpolation (NPI) were used as correction methods for the series with induced artifacts. The accuracy of each correction method was known through the results obtained after the measurement of the mean value of the series (AVNN), standard deviation (SDNN), root mean square error of the differences between successive heartbeats (RMSSD), Lomb\'s periodogram (LSP), Detrended Fluctuation Analysis (DFA), multiscale entropy (MSE) and symbolic dynamics (SD) on each HRV signal with and without artifacts. The results show that, at low levels of missing points the performance of all correction techniques are very similar with very close values for each HRV parameter. However, at higher levels of losses only the NPI method allows to obtain HRV parameters with low error values and low quantity of significant differences in comparison to the values calculated for the same signals without the presence of missing points. / Na análise da variabilidade da frequência cardíaca (Heart Rate Variability - HRV) são usadas séries temporais que contém as distancias entre batimentos cardíacos sucessivos, com o m de avaliar a regulação autonômica do sistema cardiovascular. Estas séries são obtidas a partir da análise de sinais de eletrocardiograma (ECG), as quais podem ser afetados por distintos tipos de artefatos, levando a interpretações incorretas nas análises feitas sob as séries da HRV. Abordagem clássica para lidar com esses artefatos implica a utilização de métodos de correção, alguns deles com base na interpolação, substituição ou técnicas estatísticas. No entanto, existem poucos estudos que mostram a precisão e desempenho destes métodos de correção em sinais reais da HRV. Assim, o presente estudo tem como objetivo determinar cómo os diferentes níveis de artefatos presentes no sinal afetam as caraterísticas da mesma, utilizando-se diferentes métodos lineares e não lineares de correção e posteriormente quanticação dos parâmetros da HRV. Como parte da metodología utilizada, sinais ECG de ratos obtidas mediante a técnica da telemetria foram usadas para gerar séries de HRV reais sem nenhum tipo de erro. Nestas séries foram simulados batimentos perdidos para diferentes taxas de pontos a m de emular a situação real com a maior precisão possível. Adicionalmente, foram aplicados os métodos de eliminação de segmentos (DEL), interpolação linear (LI) e cúbica (CI), janela de média móvel (MAW) e interpolação preditiva não lineal (NPI) como métodos de correção dos artefatos simulados sob as séries com erros. A precisão de cada método de correção foi conhecida através dos resultados obtidos com a quanticação do valor médio da série (AVNN), desvio padrão (SDNN), erro quadrático médio das diferenças entre batimentos sucessivos (RMSSD), periodograma de Lomb (LSP), análise de flutuações destendenciadas (DFA), entropia multiescala (MSE) e dinâmica simbólica (SD) sob cada sinal de HRV com e sem erros. Os resultados obtidos mostram que para baixos níveis de perdas de batimentos o desempenho das técnicas de correção é similar, com valores muito semelhantes para cada parámetro quanticado da HRV. Não obstante, em níveis de perdas maiores só NPI permite obter valores muito próximos e sem muitas diferenças signicativas para os mesmos parâmetros da HRV, em comparação com os valores calculados para as séries sem perdas.

Os potenciais evocados somato-sensitivos são úteis para detectar e localizar lesões nas vias sensoriais. Sua obtenção exige a média síncrona de mais de mil respostas individuais. A redução do número de estímulos elétricos para obter o potencial evocado resulta na diminuição do tempo do exame e do desconforto do paciente. O objetivo desta dissertação foi o de estudar o potencial de contribuição de duas técnicas de tempo-freqüência (ondaletas e filtros associados a trechos temporais específicos) à estimação de potenciais evocados somato-sensitivos, quando se utilizam apenas 100 respostas individuais. Quanto aos filtros, sugere-se o uso de dois passa-baixas. O primeiro, com freqüência de corte em 900Hz, deve ser utilizado no trecho inicial de 3ms a 35ms e o segundo, com freqüência de corte em 200Hz, no trecho final de 25ms a 60ms. Em relação aos parâmetros da técnica baseada em ondaletas, recomenda-se a utilização da ondaleta-mãe biortogonal 5.5, pois ela fornece erros menores e apresenta curvas visualmente boas. Além disso, ela apresenta a vantagem de ter fase linear, que é mais adequada ao processamento de potenciais evocados. Os 20% maiores coeficientes das escalas D3, D4 e D5 e os 50% maiores coeficientes da escala D6 que se encontram em trechos temporais específicos, além de todos os coeficientes de aproximação da escala 6, reconstroem adequadamente o potencial evocado. A análise estatística dos erros quadráticos normalizados indicou que a estimação por ondaletas é a melhor dentre as técnicas testadas. Também se verificou que ambas as técnicas resultaram na redução do erro quadrático normalizado, quando comparadas à média síncrona de 100 respostas individuais. Conclui-se que tanto as ondaletas quanto os filtros contribuem de forma positiva à obtenção de melhores estimativas do potencial evocado, mesmo quando um número reduzido de respostas individuais é utilizado. / Somatosensory evoked potentials are useful to detect and locate lesions in sensory pathways. In order to obtain somatosensory evoked potentials, more than one thousand single sweeps must be synchronously averaged. The smaller the number of electrical stimuli used for evoked potentials, the lower is the examination length and the patient discomfort. The objective of this thesis is to study the contribution potential of two time-frequency techniques (wavelets and filters associated to specific time intervals) to the estimation of somatosensory evoked potentials, when only one hundred individual responses are used. For the filtering technique, it is suggested that two low-pass filters be used. The first filter has a 900Hz cutoff frequency and must be used in the 3ms-35ms time interval. The second one has a 200Hz cutoff frequency and should be applied to the 25ms-60ms section. Regarding wavelet parameters, it is recommended that a biorthogonal 5.5 mother wavelet be used, because it provides smaller errors and the results are visually good. Besides it, this mother wavelet has linear phase, which is useful to the evoked potential processing. The 20% greatest coefficients in D3, D4, D5 scales, and the 50% greatest D6 coefficients are candidates to the reconstruction. Those that fall in specific time intervals are used together with all the A6 coefficients. They reconstruct evoked potentials in a satisfactory manner. The statistical analysis of the normalized squared errors indicates that the wavelet estimation is the best technique among the tested ones. This work also shows that both techniques resulted in the reduction of the normalized squared errors, when compared to the synchronous averaging of 100 individual responses. As a conclusion, both wavelets and filters contribute in a positive manner to improve evoked potential estimation, even when a reduced number of individual responses is used.

Os potenciais evocados somato-sensitivos são úteis para detectar e localizar lesões nas vias sensoriais. Sua obtenção exige a média síncrona de mais de mil respostas individuais. A redução do número de estímulos elétricos para obter o potencial evocado resulta na diminuição do tempo do exame e do desconforto do paciente. O objetivo desta dissertação foi o de estudar o potencial de contribuição de duas técnicas de tempo-freqüência (ondaletas e filtros associados a trechos temporais específicos) à estimação de potenciais evocados somato-sensitivos, quando se utilizam apenas 100 respostas individuais. Quanto aos filtros, sugere-se o uso de dois passa-baixas. O primeiro, com freqüência de corte em 900Hz, deve ser utilizado no trecho inicial de 3ms a 35ms e o segundo, com freqüência de corte em 200Hz, no trecho final de 25ms a 60ms. Em relação aos parâmetros da técnica baseada em ondaletas, recomenda-se a utilização da ondaleta-mãe biortogonal 5.5, pois ela fornece erros menores e apresenta curvas visualmente boas. Além disso, ela apresenta a vantagem de ter fase linear, que é mais adequada ao processamento de potenciais evocados. Os 20% maiores coeficientes das escalas D3, D4 e D5 e os 50% maiores coeficientes da escala D6 que se encontram em trechos temporais específicos, além de todos os coeficientes de aproximação da escala 6, reconstroem adequadamente o potencial evocado. A análise estatística dos erros quadráticos normalizados indicou que a estimação por ondaletas é a melhor dentre as técnicas testadas. Também se verificou que ambas as técnicas resultaram na redução do erro quadrático normalizado, quando comparadas à média síncrona de 100 respostas individuais. Conclui-se que tanto as ondaletas quanto os filtros contribuem de forma positiva à obtenção de melhores estimativas do potencial evocado, mesmo quando um número reduzido de respostas individuais é utilizado. / Somatosensory evoked potentials are useful to detect and locate lesions in sensory pathways. In order to obtain somatosensory evoked potentials, more than one thousand single sweeps must be synchronously averaged. The smaller the number of electrical stimuli used for evoked potentials, the lower is the examination length and the patient discomfort. The objective of this thesis is to study the contribution potential of two time-frequency techniques (wavelets and filters associated to specific time intervals) to the estimation of somatosensory evoked potentials, when only one hundred individual responses are used. For the filtering technique, it is suggested that two low-pass filters be used. The first filter has a 900Hz cutoff frequency and must be used in the 3ms-35ms time interval. The second one has a 200Hz cutoff frequency and should be applied to the 25ms-60ms section. Regarding wavelet parameters, it is recommended that a biorthogonal 5.5 mother wavelet be used, because it provides smaller errors and the results are visually good. Besides it, this mother wavelet has linear phase, which is useful to the evoked potential processing. The 20% greatest coefficients in D3, D4, D5 scales, and the 50% greatest D6 coefficients are candidates to the reconstruction. Those that fall in specific time intervals are used together with all the A6 coefficients. They reconstruct evoked potentials in a satisfactory manner. The statistical analysis of the normalized squared errors indicates that the wavelet estimation is the best technique among the tested ones. This work also shows that both techniques resulted in the reduction of the normalized squared errors, when compared to the synchronous averaging of 100 individual responses. As a conclusion, both wavelets and filters contribute in a positive manner to improve evoked potential estimation, even when a reduced number of individual responses is used.

Optimal CD4+ T cell activation is key to the generation of a productive immune response. Naïve circulating CD4+ T cells are quiescent under normal conditions and undergo activation only upon encounter of the T cell receptor (TCR) with Major Histocompatibility Complex (MHC)-encoded class II molecules on antigen presenting cells (APCs). Processed antigens (derived from pathogens, tumors or self tissue during autoimmunity) in complex with MHC class II are recognized by specific TCRs on CD4+ T cells. After this encounter, the highly complex and regulated process of CD4+ T cell activation results in the differentiation of naïve T cells into effectors and their clonal expansion. Apart from binding to its cognate peptide-MHC-II complex, several other factors define the extent and magnitude of T cell activation. This context is an important determinant of the nature of the subsequent T cell response. One of the factors involved is the strength of the signal (SOS) which is delivered to the cell upon ligation of the TCR to the MHC-peptide complex. The SOS, which can vary from weak to strong, is determined by the affinity/avidity of the TCR for the MHC-peptide complex, antigen concentrations, the duration of engagement, etc. Extreme weak or strong signals can lead to non-productive T cell responses with the former resulting mostly in anergy and the latter in cell death. Signals of optimal strength are the ones that translate into functional T cell responses. However, mechanisms by which signal strength information is translated into distinct T cell responses are still not very well understood. Binding of the TCR to the MHC-peptide complex triggers several signaling cascades and leads to generation of intracellular signaling intermediates, including Ca2+. Rise in intracellular Ca2+ levels is one of the first events to occur upon initiation of T cell activation. The initial increase is brought about due to release of Ca2+ from intracellular smooth endoplasmic reticulum stores. Once intracellular stores have been emptied, the increase is sustained by a process termed as capacitative Ca2+ entry, involving opening of Ca2+ channels in the plasma membrane known as Ca2+ release activated channels (CRACs). Consequently, Ca2+ flows from the extracellular milieu into the cell. A sustained Ca2+ increase is essential for activation of the transcription factor, NF-AT whose primary job is to initiate transcription of IL-2, a cytokine crucial for CD4+ T cell proliferation. The other intracellular signaling intermediates which are the focus of work presented in this study are reactive oxygen species (ROS). TCR signaling leads to generation of ROS, which may be either mitogenic or detrimental to T cell activation. Low levels of ROS, especially H2O2, inactivate phosphatases leading to activation of kinases and signaling pathways resulting in increased proliferation. However, high levels of ROS cause oxidative stress leading to reduced T cell activation, hyporesponsiveness and death. The experimental system used for this study consists of purified mouse lymph node CD4+ T cells. These cells were activated with varying strengths of the primary signal to better understand the roles of Ca2+ and ROS in modulating T cell activation and function. The signal strength was either varied by addition of different concentrations of ionomycin or thapsigargin, pharmacological agents that increase intracellular Ca2+ concentrations. Alternatively, signal through the surface TCR-CD3 complex was initiated using anti-CD3 in two modes: soluble (weak signal) or plate immobilized (strong signal). Increasing concentrations of ionomycin or thapsigargin or changing the mode of anti-CD3 from soluble to plate bound enhances IL-2 amounts, thereby converting a weak signal to a strong one. The work presented has been divided into three parts, each dealing with a distinct aspect of T cell activation. I. SOS and CTLA4-CD80/CD86 interactions: The binding of the TCR to its cognate MHC-peptide complex delivers the primary signal. However, this alone is not sufficient to drive T cell activation and an additional costimulatory signal emanating from the binding of CD28, a constitutively expressed receptor on T cells, to its ligands CD80 and CD86 is required. Another receptor that binds to CD80 and CD86 is CTLA-4 although it does so with a ~100 fold higher affinity. CTLA-4, unlike CD28, is expressed upon T cell activation and is considered to downregulate T cell activation. Its role as a negative regulator is highlighted by the phenotype of Ctla4 -/-mice which die of massive lymphoproliferation. However, there have also been reports of some plasticity in the effects mediated by CTLA-4. Previous work from our laboratory showed that CTLA-4-CD80/CD86 interactions could either inhibit or stimulate T cell activation depending on the SOS. To identify the molecular mediators of the differential effects of CTLA-4, the role of Ca2+ and ROS was evaluated. During activation with phorbol myristate acetate (PMA) and low amounts of ionomycin, intracellular amounts of Ca2+ were greatly reduced upon blockade of CTLA-4-CD80/CD86 interactions. Further experiments demonstrated that CTLA4-CD80/CD86 interactions reduced cell cycling upon activation with PMA and high amounts of ionomycin or thapsigargin (strong SOS) but the opposite occurred with PMA and low amounts of ionomycin or thapsigargin (weak SOS). These results were confirmed by activating cells with anti-CD3 either in the soluble or plate bound form. Considerably higher amounts of intracellular Ca2+ were present in cells activated with plate bound anti-CD3 compared to those activated with soluble anti-CD3. These amounts, further augmented by CTLA-4-CD80/CD86 interactions, probably became toxic to cells as increased proliferation was observed, using reagents that blocked these interactions. The opposite, however, was seen in cells activated with soluble anti-CD3. Also, CTLA4-CD80/CD86 interactions enhanced the generation of ROS. Studies with catalase revealed that H2O2 is required for IL-2 production and cell cycle progression during activation with a weak SOS. However, the high amounts of ROS produced during activation with a strong SOS reduced cell cycle progression. Together, this study identifies intracellular Ca2+ and ROS to play important roles in the modulation of T cell responses by CTLA4-CD80/CD86 interactions. II. SOS and CD4 downregulation: This study was initiated to identify early T cell functional responses that would help predict the strength of the primary signal. Using the in vitro culture system of varying signal strengths, it was found that CD4 surface expression was controlled by signal strength. CD4 is a surface glycoprotein expressed on the TH subset along with the TCR. It performs two main functions: First, binding to MHC class II and strengthening the TCR-MHC interaction, i.e. functioning as a coreceptor. Second, due to its association with p56lck a src family tyrosine kinase, the presence of CD4 along with the TCR enhances signal transduction. Also, CD4 acts as a receptor for entry for the AIDS virus. It is known that CD4 is downregulated from the surface and degraded upon T cell activation by a protein kinase-C dependent process in human and mouse T cells. Experiments presented in this study showed increased CD4 downregulation with a strong signal. The roles of intracellular mediators were assessed and high intracellular Ca2+ amounts, but not PMA activation, was required for sustained CD4 downregulation. Also, increased H2O2 amounts in cells activated with a strong SOS inhibited CD4 downregulation. Most interestingly, the pattern of CD4 downregulation was different between peripheral T cells and thymocytes, suggesting a correlation with CD4+ T cell development. III. Modulation of CD4+ T cell activation by small molecule plant growth regulators: An important area of investigation in T cell biology is the identification of molecules that modulate T cell activation. Towards this end, the mechanisms by which small molecule plant growth regulators, naphthalene acetic acid (NAA), 2,4 dichlorophenoxyacetic acid (2,4D) and indole acetic acid (IAA), influence CD4+ T cell activation was studied. It is useful to recall that IAA is the natural auxin present in plants, NAA is a synthetic auxin and 2,4D is a herbicide. These compounds, but not structurally similar control molecules, increased the activation and IL-2 production in CD4+ T cells activated with either soluble anti-CD3 or a combination of PMA and ionomycin. An investigation into the mechanisms of action by these compounds revealed increased early generation of intracellular ROS and Ca2+. Interestingly, the nature of their effects was found to rely on the strength of the primary signal: IL-2 and proliferation were increased in cells activated with a weak signal, but lowered proliferation was observed in cells activated with a strong signal. Cells activated with strong signal posses high amounts of ROS and Ca2+ and further increase in their amounts by IAA, NAA and 2,4D resulted in growth suppression. However, augmentation of Ca2+ and ROS amounts in cells activated with a weak signal was mitogenic. The role of these compounds during in vivo T cell responses needs to be addressed. Taken together, results presented in this study emphasize the importance of the role of SOS in determining T cell responses. In addition, the roles of Ca2+ and ROS downstream of the primary signal in modulating CD4+ T cell activation were demonstrated.

In the analysis of heart rate variability (HRV) are used temporal series that contains the distances between successive heartbeats in order to assess autonomic regulation of the cardiovascular system. These series are obtained from the electrocardiogram (ECG) signal analysis, which can be affected by different types of artifacts leading to incorrect interpretations in the analysis of the HRV signals. Classic approach to deal with these artifacts implies the use of correction methods, some of them based on interpolation, substitution or statistical techniques. However, there are few studies that shows the accuracy and performance of these correction methods on real HRV signals. This study aims to determine the performance of some linear and non-linear correction methods on HRV signals with induced artefacts by quantification of its linear and nonlinear HRV parameters. As part of the methodology, ECG signals of rats measured using the technique of telemetry were used to generate real heart rate variability signals without any error. In these series were simulated missing points (beats) in different quantities in order to emulate a real experimental situation as accurately as possible. In order to compare recovering efficiency, deletion (DEL), linear interpolation (LI), cubic spline interpolation (CI), moving average window (MAW) and nonlinear predictive interpolation (NPI) were used as correction methods for the series with induced artifacts. The accuracy of each correction method was known through the results obtained after the measurement of the mean value of the series (AVNN), standard deviation (SDNN), root mean square error of the differences between successive heartbeats (RMSSD), Lomb\'s periodogram (LSP), Detrended Fluctuation Analysis (DFA), multiscale entropy (MSE) and symbolic dynamics (SD) on each HRV signal with and without artifacts. The results show that, at low levels of missing points the performance of all correction techniques are very similar with very close values for each HRV parameter. However, at higher levels of losses only the NPI method allows to obtain HRV parameters with low error values and low quantity of significant differences in comparison to the values calculated for the same signals without the presence of missing points. / Na análise da variabilidade da frequência cardíaca (Heart Rate Variability - HRV) são usadas séries temporais que contém as distancias entre batimentos cardíacos sucessivos, com o m de avaliar a regulação autonômica do sistema cardiovascular. Estas séries são obtidas a partir da análise de sinais de eletrocardiograma (ECG), as quais podem ser afetados por distintos tipos de artefatos, levando a interpretações incorretas nas análises feitas sob as séries da HRV. Abordagem clássica para lidar com esses artefatos implica a utilização de métodos de correção, alguns deles com base na interpolação, substituição ou técnicas estatísticas. No entanto, existem poucos estudos que mostram a precisão e desempenho destes métodos de correção em sinais reais da HRV. Assim, o presente estudo tem como objetivo determinar cómo os diferentes níveis de artefatos presentes no sinal afetam as caraterísticas da mesma, utilizando-se diferentes métodos lineares e não lineares de correção e posteriormente quanticação dos parâmetros da HRV. Como parte da metodología utilizada, sinais ECG de ratos obtidas mediante a técnica da telemetria foram usadas para gerar séries de HRV reais sem nenhum tipo de erro. Nestas séries foram simulados batimentos perdidos para diferentes taxas de pontos a m de emular a situação real com a maior precisão possível. Adicionalmente, foram aplicados os métodos de eliminação de segmentos (DEL), interpolação linear (LI) e cúbica (CI), janela de média móvel (MAW) e interpolação preditiva não lineal (NPI) como métodos de correção dos artefatos simulados sob as séries com erros. A precisão de cada método de correção foi conhecida através dos resultados obtidos com a quanticação do valor médio da série (AVNN), desvio padrão (SDNN), erro quadrático médio das diferenças entre batimentos sucessivos (RMSSD), periodograma de Lomb (LSP), análise de flutuações destendenciadas (DFA), entropia multiescala (MSE) e dinâmica simbólica (SD) sob cada sinal de HRV com e sem erros. Os resultados obtidos mostram que para baixos níveis de perdas de batimentos o desempenho das técnicas de correção é similar, com valores muito semelhantes para cada parámetro quanticado da HRV. Não obstante, em níveis de perdas maiores só NPI permite obter valores muito próximos e sem muitas diferenças signicativas para os mesmos parâmetros da HRV, em comparação com os valores calculados para as séries sem perdas.

This master’s thesis deals with half-wave and full-wave rectifiers and instrumentation amplifier design in CMOS technology, suitable for biomedical signal processing. Properties of optional solutions are analyzed and appropriate circuits are designed. Their functionality is verified with simulation. Designed circuits are then used to form a circuit converting differential input voltage into rectified output current.

Na área de processamento de sinais biomédicos a extração de informação, baseada em um conjunto de medidas adquiridas no tempo, é considerada de suma importância. A qualidade desta informação extraída permite avaliar o funcionamento dos diversos órgãos. Objetivos: (1) propor o método de análise de componentes dependentes para auxiliar a extração de componentes de interesse, a partir de medidas multivariadas; (2) caraterizar as componentes extraídas através de representações em termos de tempo e freqüência, e espectro de potência; e, (3) aplicar o método e avaliar as componentes de interesse extraídas no contexto real MCGf, MGG e fMRI. A proposta para a extração fundamenta-se no método chamado de Análise de Componentes Dependentes ACD. As medidas a serem processadas são multivariadas a partir de sensores distribuídos, espacialmente, no corpo humano dando origem a um conjunto de dados correlacionados no tempo e/ou no espaço. Observa-se que os sinais de interesse raramente são registrados de forma isolada, e sim misturados com outros sinais superpostos, ruído e artefatos fisiológicos ou ambientais, onde a relação sinal-ruído é geralmente baixa. Nesse contexto, a estratégia a ser utilizada baseia-se na ACD, que permitirá extrair um pequeno número de fontes, de potencial interesse, com informações úteis. A estratégia ACD para extração de informação é aplicada em três importantes problemas, na área de processamento de sinais biomédicos: (1) detecção do sinal do feto em magnetocardiografia fetal (MCGf); (2) detecção da atividade de resposta elétrica do estômago em magnetogastrografia (MGG); e, (3) detecção de regiões ativas do cérebro em experimentos de imagens por ressonância magnética funcional (Functional Magnetic Resonance Imaging, fMRI). Os resultados, nos três casos estudados, mostraram que o método utilizado, como estratégia, é efetivo e computacionalmente eficiente para extração de sinais de interesse. Concluímos, baseados nas aplicações, que o método proposto é eficaz, mostrando seu potencial para futuras pesquisas clínicas._________________________________________________________________________________________ ABSTRACT: An important goal in biomedical signal processing is the extraction of information based on a set of physiological measurements made along time. Generally, biomedical signals are electromagnetic measurements. Those measurements (usually made with multichannel equipment) are registered using spatially distributed sensors around some areas of the human body, originating a set of time and/or space correlated data. The signals of interest are rarely registered alone, being usually observed as a mixture of other spurious, noisy signals (sometimes superimposed) and environmental or physiological artifacts. More over, the signal-to-noise ratio is generally low. In many applications, a big number of sensors are available, but just a few sources are of interest and the remainder can be considered noise. For such kind of applications, it is necessary to develop trustful, robust and effective learning algorithms that allow the extraction of only a few sources potentially of interest and that hold useful information. The strategy used here for extraction of sources is applied in three important problems in biomedical signal processing: (1) detection of the fetal magnetocardiogram signal (fMCG); (2) detection of the electrical activity of the stomach in magnetogastrograms (MGG); and (3) detection of active regions of the brain in experiments in functional Magnetic Resonance Imaging (fMRI). The results, within the three cases of study, showed that the DCA method used as strategy is effective and computationally efficient on extraction of desired signals.