This thesis uses one of Brain Computer Interface (BCI) products, NeuroSky headset, to design a prototype model to control window blind by using headset’s single channel electrode. Seven volunteers performed eight different exercises while the signal from the headset was recorded. The dataset was analyzed, and exercises with strongest power spectral density (PSD) were chosen to continue to work with. Matlabs spectrogram function was used to divide the signal in time segments, which were 0.25 seconds. One segment from each of these eight exercises was taken to form different combinations which were later classified.The classification result, while using two of proposed exercises (tasks) was successful with 97.0% accuracy computed by Nearest Neighbor classifier. Still, we continued to investigate if we could use three or four thoughts to create three or four commands. The result presented lower classification accuracy when using either 3 or 4 command thoughts with performance accuracy of 92% and 76% respectively.Thus, two or three exercises can be used for constructing two or three different commands.

Brain-computer interfacing (BCI) is a new method of human-machine interaction. It involves the extraction of information from the electroencephalogram (EEG) through signal processing and pattern recognition. The technology has far reaching implications for those with severe physical disabilities and has the potential to enhance machine interaction for the rest of the population. In this work we investigate time-frequency analysis in motor-imagery BCI. We consider two methods for signal analysis: adaptive autoregressive models (AAR) and wavelet transform (WAV). There are three major contributions of this research to single-trial analysis in motor-imagery BCI. First, we improve classification of AAR features over a conventional method by applying a temporal evidence accumulation (TEA) framework. Second, we compare the performance of AAR and WAV under the TEA framework for three subjects and find that WAV outperforms AAR for two subjects. The subject for whom AAR outperforms WAV has the lowest overall signal-to-noise ratio in their BCI output, an indication that the AAR model is more robust than WAV for noisier signals. Lastly, we find empirical evidence of complimentary information between AAR and WAV and propose a fusion scheme that increases the mutual information between the BCI output and classes.

Brain-computer interfacing (BCI) is a new method of human-machine interaction. It involves the extraction of information from the electroencephalogram (EEG) through signal processing and pattern recognition. The technology has far reaching implications for those with severe physical disabilities and has the potential to enhance machine interaction for the rest of the population. In this work we investigate time-frequency analysis in motor-imagery BCI. We consider two methods for signal analysis: adaptive autoregressive models (AAR) and wavelet transform (WAV). There are three major contributions of this research to single-trial analysis in motor-imagery BCI. First, we improve classification of AAR features over a conventional method by applying a temporal evidence accumulation (TEA) framework. Second, we compare the performance of AAR and WAV under the TEA framework for three subjects and find that WAV outperforms AAR for two subjects. The subject for whom AAR outperforms WAV has the lowest overall signal-to-noise ratio in their BCI output, an indication that the AAR model is more robust than WAV for noisier signals. Lastly, we find empirical evidence of complimentary information between AAR and WAV and propose a fusion scheme that increases the mutual information between the BCI output and classes.

Les effets de la stimulation électrique sur la dynamique des systèmes neuronaux épileptiques sont encore méconnus. L'objectif principal de cette thèse est de progresser dans la compréhension des effets attendus en fonction des paramètres de stimulation. Dans la première partie du manuscrit, un modèle mésoscopique (population neuronale) de la boucle thalamocorticale est proposé pour étudier en détails les effets de stimulation indirecte (thalamique), avec une attention particulière sur la fréquence. Des signaux EEG intracérébraux acquis chez un patient souffrant d'épilepsie pharmaco-résistante ont d'abord été analysés selon une approche temps-fréquence (algorithme de type Matching Pursuit). Les caractéristiques extraites ont ensuite été utilisées pour identifier les paramètres du modèle proposé en utilisant une approche exhaustive (minimisation de la distance entre signaux simulés et réels). Enfin, le comportement dynamique du modèle a été étudié en fonction de la fréquence du signal de stimulation. Les résultats montrent que le modèle reproduit fidèlement les signaux observés ainsi que la relation non linéaire entre la fréquence de stimulation et ses effets sur l'activité épileptique. Ainsi, dans le modèle, la stimulation à basse fréquence (SBF ; fs <20 Hz) , et la stimulation à haute fréquence (SHF ; fs > 60 Hz) permettent d'abolir les dynamiques épileptiques, alors que la stimulation à fréquence intermédiaire (SFI; 20 < fs < 60 Hz) n'ont pas d'effet , comme observé cliniquement. De plus, le modèle a permis d'identifier des mécanismes cellulaires et de réseau impliqués dans les effets modulateurs de la stimulation. La deuxième partie du manuscrit porte sur les effets polarisants de la stimulation directe en courant continu (CC) de la zone épileptogène dans le contexte de l'épilepsie mésiale du lobe temporal (EMLT). Un modèle biomathématique bien connu de la région hippocampique CA1 a été adapté pour cette étude. Deux modifications sont été intégrées au modèle, 1) une représentation physiologique de l'occurrence des décharges paroxystiques hippocampiques (DPH) basée sur une identification de leurs statistiques d'occurrence basée sur des données expérimentales (modèle in vivo d'EMLT)et 2) une représentation électrophysiologiquement plausible de la stimulation prenant en compte l'interface électrode-électrolyte. L'analyse de la sortie du modèle en fonction de la polarité de stimulation, a montré qu'une réduction (resp. augmentation) significative des DPH (en durée et en fréquence) sous stimulation anodale (resp. cathodole). Un protocole expérimental a ensuite été proposé et utilisé afin de valider les prédictions du modèle. / The effects of electrical stimulation on the dynamics of epileptic neural systems are still unknown. The main objective of this thesis is to progress the understanding of the expected effects as a function of stimulation parameters. In the first part of the manuscript, a mesoscopic model (neural population) of the thalamocortical loop is proposed to study in details the effects of indirect stimulation (thalamic), with a particular attention to stimulation frequency. Intracerebral EEG signals acquired from a patient with drug-resistant epilepsy were first analyzed using a time-frequency approach (Matching Pursuit algorithm). The extracted features were then used to optimize the parameters of the proposed model using a Brute-Force approach (minimizing the distance between simulated and real signals). Finally, the dynamical behavior of the model was studied as a function of the frequency of the stimulation input. The results showed that the model reproduces the real signals as well as the nonlinear relationship between the frequency of stimulation and its effects on epileptic dynamics. Thus, in the model, low-frequency stimulation (LFS; fs <20 Hz) and high-frequency stimulation (HFS; fs > 60 Hz) suppress epileptic dynamics, whereas intermediate-frequency stimulation (IFS; 20 < fs <60 Hz) has no effect, as observed clinically. In addition, the model was used to identify the cellular and network mechanisms involved in the modulatory effects of stimulation. The second part of the manuscript addresses the polarizing effects of direct current (DC) stimulation of the epileptogenic zone in the context of the mesial temporal lobe epilepsy (MTLE). A well-known computational model of the hippocampal CA1 region was adapted for this study. Two modifications were added to the model: 1) a physiological representation of the occurrence of hippocampal paroxysmal discharges (HPD) based on the statistical identification of their occurrence in experimental data (in vivo model of MTLE) and 2) an electrophysiologically plausible representation of the stimulation inputs taking into account the electrode-electrolyte interface. The analysis of the model output as a function of the polarity of stimulation, showed a significant reduction (resp. increase) of HPDs (duration and frequency) in anodal stimulation (resp. cathodol). An experimental protocol was then proposed and used to validate the model predictions.

Studying newborns in the first days of life prior to experiencing the world provides remarkable insights into the neurocognitive predispositions that humans are endowed with. First, it helps us to improve our current knowledge of the development of a typical brain. Secondly, it potentially opens new pathways for earlier diagnosis of several developmental neurocognitive disorders such as Autism Spectrum Disorder (ASD). While most studies investigating early cognition in the literature are purely behavioural, recently there has been an increasing number of neuroimaging studies in newborns and infants. Electroencephalography (EEG) is one of the most optimal neuroimaging technique to investigate neurocognitive functions in human newborns because it is non-invasive and quick and easy to mount on the head. Since EEG offers a versatile design with custom number of channels/electrodes, an ergonomic wearable solution could help study newborns outside clinical settings such as their homes. Compared to adult EEG, newborn EEG data are different in two main aspects: 1) In experimental designs investigating stimulus-related neural responses, collected data is extremely short in length due to the reduced attentional span of newborns; 2) Data is heavily contaminated with noise due to their uncontrollable movement artifacts. Since EEG processing methods for adults are not adapted to very short data length and usually deal with well-defined, stereotyped artifacts, they are unsuitable for newborn EEG. As a result, researchers manually clean the data, which is a subjective and time-consuming task. This thesis work is specifically dedicated to developing (semi-) automated novel signal processing methods for noise removal and for extracting reliable neural responses specific to this population. The solutions are proposed for both high-density EEG for traditional lab-based research and wearable EEG for clinical applications. To this end, this thesis, first, presents novel signal processing methods applied to newborn EEG: 1) Local Outlier Factor (LOF) for detecting and removing bad/noisy channels; 2) Artifacts Subspace Reconstruction (ASR) for detecting and removing or correcting bad/noisy segments. Then, based on these algorithms and other preprocessing functionalities, a robust preprocessing pipeline, Newborn EEG Artifact Removal (NEAR), is proposed. Notably, this is the first time LOF is explored for EEG bad channel detection, despite being a popular outlier detection technique in other kinds of data such as Electrocardiogram (ECG). Even if ASR is already an established artifact real algorithm originally developed for mobile adult EEG, this thesis explores the possibility of adapting ASR for short newborn EEG data, which is the first of its kind. NEAR is validated on simulated, real newborn, and infant EEG datasets. We used the SEREEGA toolbox to simulate neurologically plausible synthetic data and contaminated a certain number of channels and segments with artifacts commonly manifested in developmental EEG. We used newborn EEG data (n = 10, age range: 1 and 4 days) recorded in our lab based on a frequency-tagging paradigm. The chosen paradigm consists of visual stimuli to investigate the cortical bases of facelike pattern processing, and the results were published in 2019. To test NEAR performance on an older population with an event-related design (ERP) and with data recorded in another lab, we also evaluated NEAR on infant EEG data recorded on 9-months-old infants (n = 14) with an ERP paradigm. The experimental paradigm for these datasets consists of auditory stimulus to investigate the electrophysiological evidence for understanding maternal speech, and the results were published in 2012. Since authors of these independent studies employed manual artifact removal, the obtained neural responses serve as ground truth for validating NEAR’s artifact removal performance. For comparative evaluation, we considered the performance of two state-of-the-art pipelines designed for older infants. Results show that NEAR is successful in recovering the neural responses (specific to the EEG paradigm and the stimuli) compared to the other pipelines. In sum, this thesis presents a set of methods for artifact removal and extraction of stimulus-related neural responses specifically adapted to newborn and infant EEG data that will hopefully contribute to strengthening the reliability and reproducibility of developmental cognitive neuroscience studies, both in research laboratories and in clinical applications.