Abstract

The presence of electrooculographic (EOG) artifacts in the electroencephalographic (EEG) signal is a major problem in the study of brain potentials. A variety of algorithms have been proposed to reject these artifacts including methods based on regression and blind source separation (BSS) techniques. None of them has so far been established as the method of choice. In the present study, the performances of five widely used EOG artifact rejection techniques are compared. The compared methodologies include two fully automated regression methods, one based on Least Mean Square (LMS) for its optimization process, and the other on Recursive Least Square (RLS) algorithm, two BSS techniques which use respectively the Extended — Independent Component Analysis (ext — ICA) and the Second Order Blind Identification (SOBI), and finally a time-varying adaptive algorithm based on H ∞ principles (H ∞ — TV). Each algorithm was applied in real EEG data and then their performance quantified in the time domain. The performance of RLS and H ∞ — TV were poor in removing eye — blink artifacts. For the rest of the methods the results supported the use of LMS technique and suggested the need for further research examining the performance of various artifact rejection techniques in both time and frequency domain.

Abstract

Emotion identification is beginning to be considered as an essential feature in human-computer interaction. However, most of the studies are mainly focused on facial expression classifications and speech recognition and not much attention has been paid until recently to physiological pattern recognition. In this paper, an integrative approach is proposed to emotional interaction by fusing multi-modal signals. Subjects are exposed to pictures selected from the International Affective Picture System (IAPS). A feature extraction procedure is used to discriminate between four affective states by means of a Mahalanobis distance classifier. The average classifications rate (74.11%) was encouraging. Thus, the induced affective state is mirrored through an avatar by changing its facial characteristics and generating a voice message sympathising with the user’s mood. It is argued that multi-physiological patterning in combination with anthropomorphic avatars may contribute to the enhancement of affective multi-modal interfaces and the advancement of machine emotional intelligence.

Abstract

The continuously increasing number of neuroscience studies and the difficulties associated with searching for related information and properly tracking neuroscience findings makes it imperative that one may be lead to isolated theories and findings which may be incompatible to each other or partially occluded. Semantically describing several aspects of studies in this field, such as, research groups attributes, aims of studies, experimental procedures followed, hardware and software tools utilised, acquisition systems used, as well as, the emerging neuro-physiological patterns found, may facilitate an integrative view of neuroscience theories. To this end, the current piece of work aims to provide a global theoretical framework using ontologies and semantic rules to describe neuroscience studies. Implementation details and applicability of the proof of concept are illustrated by means of an example targeting the semantic description of an emotion related study. The importance of the proposed framework in facilitating the envisaged personalised healthcare of the information society is discussed.

Abstract

The plethora of artifact rejection (AR) techniques proposed for removing electrooculographic (EOG) artifacts from electroencephalographic (EEG) signals can be separated into two main categories. The first category is composed of regression - based methods, while the second one consists of blind source separation (BSS) - methods. A major disadvantage of BSS-based methodology is that the artifactual components include also neural activity, thus their rejection leads to the distortion of the underlying cerebral activity. The current study tries to solve the aforementioned problem by proposing a new hybrid algorithm for EOG AR. According to this automatic approach, called REG-ICA, independent component analysis (ICA) is used to decompose EEG signals into spatial independent components (ICs). Then an adaptive filter, based on a stable Version of the recursive least square (sRLS) algorithm, is applied to ICs so as to remove only EOG artifacts and maintain the neural signals intact. Then the cleaned ICs are projected back, reconstructing the artifact - free EEG signals. In order to evaluate the performance of the proposed technique, REG-ICA has been compared with the least mean square (LMS) approach, in simulated EEG data. Two criteria were used for the comparison: how successfully algorithms remove eye blinking artifacts, and how much the EEG signals are distorted. Results support the argument that REG-ICA removes successfully EOG activity, while it minimizes the distortion of the underlying cerebral activity in contrast to LMS. 

Abstract

Integrating emotional feedback to educational systems has become one of the main concerns of the affective learning research community. This paper provides evidence that Embodied Conversational Agents (ECAs) could be effectively used as emotional feedback to improve brainwave activity towards learning. Further research, integrating ECAs into tutoring systems is essential to confirm these results. 

Abstract

Olfactometers are widely used in the study of the chemical senses from a neurophysiological point of view. Although there is a plethora of olfactometer designs, all of them lack of flexibility in modification. In more details they are not able to dynamically increase the number of odors that can be provided simultaneously or they are not capable to use other form of odorous material than the one they’ve been designed for. In addition to all these the concentration of the stimulus is estimated indirectly through the ratio of the odorized and the odorless air that is delivered to the subject. Taking into account all these, it is understandable that there is an urgent need for an effective olfactometer which will be able to overcome the aforementioned drawbacks of the existing olfactometers. In this scope, the current study comes to introduce a new computer – operated olfactometer. Its novelty lies on the fact that it has a modular architecture with microcontroller units in every module, which can undoubtfully simplify the system’s modification. On the other hand it can also estimate directly the Volatile Organic Compounds (VOC) with one sensor for every odor, and one sensor for the overall stimulus.

Abstract

Electrical signals detected along the scalp by an Electroencephalogram (EEG), but that originate from non-cerebral origin are called artifacts. Especially when these artifacts are produced by the human body we talk about biological artifacts. The most common biological artifacts are the electrical signals produced by ocular and heart activity. EEG data is almost always contaminated by such artifacts. The last decade Independent Component Analysis (ICA) has a crucial role in neuroscience and it takes great attention for artifact rejection purposes. According to ICA’s methodology, EEG signals are decomposed to statistical Independent Components (IC) and then an EEG specialist is called to recognize the artifactual ICs. Some of the major limitations of the current approach are that the aforementioned selection is subjective, it demands a high skill EEG operator, it is time consuming and it cannot be applied in online processing. Our study employs machine learning techniques in order to recognize the contaminated ICs with ocular or heart artifacts. More specific 19-channel EEG datasets from 86 normal subjects were decomposed using ICA (19×86=1634 ICs in total). Then three independent observers marked an IC as artifactual if it includes ocular or heart artifacts, otherwise it was marked as normal. Then kurtosis was computed in short segments with 1250 sample points fixed length without overlap for each IC. The mean kurtosis value was computed for each IC and the Naïve Bayes Classifier (NBC) classifier was adopted in order to classify the ICs as artifactual or normal. The results suggest that the NBC has correctly classified 1611/1634 ICs (98.5924 %) so it can be suggested that kurtosis seems to be convenient for the classification of contaminated ICs by ocular or heart artifacts. 

Abstract

The aim of this paper is to show that the brain activity of patients with acute respiratory failure hospitalized in Intensive Care Units (ICUs) can provide useful medical information, which is directly related to neurological rehabilitation. It also aims to show that the entropy and kurtosis, widely used indices of the electroencephalographic (EEG) signals, are able to identify EEG changes associated with cerebral hypoxia. EEG signals were recorded from eight adult patients with acute respiratory failure admitted to the ICU. The measurements were recorded in five stages, with FiO2 at 40%, 100%, 60%, 20% and 0% (T-piece) respectively. Total time of recordings was 50min (10 min. for each stage). The EEG signals were filtered and further cleaned from ocular and muscular artifacts as well as from the artifacts introduced by other external devices, electrodes movements and electrode’s bad tangencies. Afterwards the 10-min EEG signals of each stage were segmented in ten epochs with one minute fixed length. Then Kurtosis and Shannon’s Entropy were calculated in each segment. One-Way ANOVA verified the assumption that there are statistically significant differences between the various stages of our protocol, while the Scheffe Post-Hoc tests revealed the homogeneous subsets compiled by the aforementioned stage. The results suggest that the EEG is directly connected with the mechanical ventilator’s changes, so in the future, clinicians could probably use the EEG as particularly useful and time-critical information, especially during the weaning procedure from the mechanical ventilator.

Abstract

Electroencephalographic (EEG) signals were recorded from 28 participants as they passively viewed emotional stimuli from International Affective Picture System (IAPS), categorized in 4 groups ranging in pleasure and arousal. The aim of the study was to examine if the Functional Connectivity Networks estimated during different emotional stimuli, differ in their characteristics. Functional Connectivity Networks were estimated for the four categories of emotional stimuli using coherence between each pair of electrodes on the frequency band of alpha rhythm. Graph metrics were calculated for each network and they were statistically analyzed. Pleasure was found to modify the local efficiency of the networks with unpleasant stimuli appearing to form networks with clusters easier than pleasant stimuli. Arousal also affected the global efficiency of the functional networks, with high arousing stimuli appearing to form networks with more efficient communication among nodes.

Abstract

The aim of this study was to examine whether Functional Connectivity Networks (FCN) obtained during exposing the humans to emotional stimuli, vary according to the characteristics of the stimuli. Twenty-eight participants passively viewed emotional pictures from International Affective Picture System (IAPS) ranging across both arousal and pleasure dimensions. Electroencephalographic (EEG) signals were simultaneously recorded from 19 scalp sites. Relative Wavelet Entropy (RWE) was estimated for all EEG electrode pairs. Different RWE thresholds were set in order to form the corresponding FCNs. Graph theory metrics, such as Density, Cluster Coefficient and Global Efficiency were calculated for all FCNs. It was found that arousal significantly affected small-world properties of FCNs formed from 42 to 170 msec; low arousing pictures elicited FCN with higher Density, higher Cluster Coefficient and higher Global Efficiency as compared to high arousing ones

Abstract

Synchronization analysis of EEG data has been so far performed by means of coherence functions or non-linear similarity quantifications. However, linear methods fail to provide information about the entire frequency spectrum or the direction of the interaction, while non-linear estimates require time-consuming computations, difficult parameter tuning and huge amounts of data. This paper, aims to overcome the above limitations by investigating the feasibility of using the time-evolving Relative Wavelet Entropy (RWE) for the quantification of the similarity degree between homologous electrodes on either hemisphere. Emotional stimuli selected from the International Affective Picture Stimuli (IAPS) collection are employed in order to induce neurophysiological responses. The methodological framework involves the analysis of the EEG data in time intervals of 128 ms duration. The results showing increased similarity during early, mid and late emotional processing indicate the method's robustness providing hope for the dynamic characterization of the cooperative brain activity during cognitive functioning.

Abstract

Many studies investigated the brain responses as a reaction in auditory or visual stimuli separately. However a few studies have been published so far investigating the interactions of the two aforementioned stimuli. The current study comes to examine the impact of the audio-visual stimulation with binaural beats and flickering light in four different colors on low and upper alpha oscillations. For this purpose electroencephalogram (EEG) was adopted and Event Related Desynchronization/Event Related Synchronization (ERD/ERS) has been used as an index in order to investigate the alpha brain responses. Statistically significant results suggest that the combination of audio-visual stimuli with binaural beats and flickering light color at 8 and 10 Hz respectively can evoke significant Following Frequency Response (FFR) of the low and upper alpha oscillations.

Abstract

Gender differences in mathematical thinking is a common concern of scientists from different research fields. Both parents and teachers report that males seem to perform better in complex mathematics compared to females. This study comes to shed light in the different organization of the underlying functional networks, in order to investigate the aforementioned observation, without supporting or rejecting this statement. In this sense, it is generally accepted that females use their both hemispheres to accomplish a certain task, while males use mostly the hemisphere which is properly suited. For the purposes of the current analysis, electroencephalographic recordings were collected from 11 males and 11 females, during a difficult mathematical task. Then a previously proposed model was used in order to pass from the sensor level to the cortical one, in order to examine the networks formed among the cortical dipoles. Mutual information was employed to form the graphs represeting the functional connectivity among the different dipoles, while the density, the global and the local efficiencies were further examined. The results suggest that females use their both hemisphere to solve the complex mathematical task while males use mostly their left hemisphere which is the responsible one for the mathematical thinking.

Abstract

Event-Related Potentials (ERPs) or Event-Related Oscillations (EROs) have been widely used to study emotional processing, mainly on the theta and gamma frequency bands. However, the role of the slow (delta) waves has been largely ignored. The aim of this study is to provide a framework that combines EROs with Event-Related Desynchronization (ERD)/Event-Related Synchronization (ERS), and peak amplitude analysis of delta activity, evoked by the passive viewing of emotionally evocative pictures. Results showed that this kind of approach is sensitive to the effects of gender, valence, and arousal, as well as, the study of interhemispherical disparity, as the two-brain hemispheres interplay roles in the detailed discrimination of gender. Valence effects are recovered in both the central electrodes as well as in the hemisphere interactions. These findings suggest that the temporal patterns of delta activity and the alterations of delta energy may contribute to the study of emotional processing. Finally the results depict the improved sensitivity of the proposed framework in comparison to the traditional ERP techniques, thereby delineating the need for further development of new methodologies to study slow brain frequencies. 

Abstract

Men and women seem to process emotions and react to them differently. Yet, few neurophysiological studies have systematically investigated gender differences in emotional processing. Here, we studied gender differences using Event Related Potentials (ERPs) and Skin Conductance Responses (SCR) recorded from participants who passively viewed emotional pictures selected from the International Affective Picture System (IAPS). The arousal and valence dimension of the stimuli were manipulated orthogonally. The peak amplitude and peak latency of ERP components and SCR were analyzed separately, and the scalp topographies of significant ERP differences were documented. Females responded with enhanced negative components (N100 and N200), in comparison to males, especially to the unpleasant visual stimuli, whereas both genders responded faster to high arousing or unpleasant stimuli. Scalp topographies revealed more pronounced gender differences on central and left hemisphere areas. Our results suggest a difference in the way emotional stimuli are processed by genders: unpleasant and high arousing stimuli evoke greater ERP amplitudes in women relatively to men. It also seems that unpleasant or high arousing stimuli are temporally prioritized during visual processing by both genders. 

Abstract

Recent neuroscience findings demonstrate the fundamental role of emotion in the maintenance of physical and mental health. In the present study, a novel architecture is proposed for the robust discrimination of emotional physiological signals evoked upon viewing pictures selected from the International Affective Picture System (IAPS). Biosignals are multichannel recordings from both the central and the autonomic nervous systems. Following the bidirectional emotion theory model, IAPS pictures are rated along two dimensions, namely, their valence and arousal. Following this model, biosignals in this paper are initially differentiated according to their valence dimension by means of a data mining approach, which is the C4.5 decision tree algorithm. Then, the valence and the gender information serve as an input to a Mahalanobis distance classifier, which dissects the data into high and low arousing. Results are described in Extensible Markup Language (XML) format, thereby accounting for platform independency, easy interconnectivity, and information exchange. The average recognition (success) rate was 77.68% for the discrimination of four emotional states, differing both in their arousal and valence dimension. It is, therefore, envisaged that the proposed approach holds promise for the efficient discrimination of negative and positive emotions, and it is hereby discussed how future developments may be steered to serve for affective healthcare applications, such as the monitoring of the elderly or chronically ill people. 

Abstract

There are so far two main methodological approaches for rejecting ocular artifacts from electroencephalographic (EEG) and magnetoencephalographic (MEG) signals: regression- and Blind Source Separation (BSS)-based techniques, both having merits, as well as, some serious limitations. In this piece of work, a hybrid methodology that combines the main advantages of these two methods is proposed. We hypothesize that the artifactual independent components (ICs) extracted by a BSS method include more ocular and less cerebral activity than the contaminated EEG signals. We thus propose to apply a regression algorithm to the ICs rather than directly to the recorded signals. The analysis was carried out with synthetic mixtures of real EEG and electroocculographic (EOG) recordings. A BSS method, the extended INFOMAX version of Independent Component Analysis (ICA), was initially used to decompose the artificially contaminated EEG signals into spatiotemporal ICs. Then, a regression scheme, based on a stable version of the Recursive Least Squares algorithm, sRLS, was applied to the artifactual components in order to remove only the ocular artifacts, maintaining the underlying neural signals intact. The processed ICs were then projected back, reconstructing the artifact-free EEG signals. The performance of the proposed technique was compared with two automatic techniques; a regression technique based on Least Mean Square (LMS) algorithm and a BSS-based artifact rejection technique called wavelet-ICA (W-ICA) on the artificially contaminated data. For comparison, two metrics were used to assess the different methods’ performance: the first quantified how successful each technique was in removing the ocular artifacts from the EEG recordings, and the second one quantified how much each technique distorted the ongoing brain activity in both time and frequency domains. Confirming our main hypothesis, results have shown that the artifactual ICs contained more ocular and less cerebral activity (p < 0.04) (artifact to signal ratio (ASR) = 1.83 ± 3.65) in contrast to the contaminated electrode signals (ASR = 0.69 ± 3.40). Our results reveal that the proposed methodology, namely REG-ICA, removes the ocular artifacts more successfully than W-ICA (p < 0.01) or LMS (p < 0.01). It also distorts less the brain activity in the time domain when compared to W-ICA and LMS. In the frequency domain, it distorts the brain activity less than the W-ICA in all frequency bands, and less than the LMS for the delta, beta, and gamma bands. Our results suggest that the proposed methodology is evidently an attractive alternative to other already proposed artifact rejection methodologies.

. 

Abstract

Brain functional connectivity has gained increasing interest over the last few years. The application of Graph Theory on functional connectivity networks (FCNs) has shed light into different topics related to physiology as well as pathology. To this end, different connectivity metrics may be used; however, some concerns are often raised related with inconsistency of results and their associated neurophysiological interpretations depending on the metric used. This paper examines how the use of different connectivity metrics affects the small-world-ness of the FCNs and eventually the neuroscientific evidences and their interpretation; to achieve this, electroencephalography (EEG) data recorded from healthy subjects during an emotional paradigm are utilized. Participants passively viewed emotional stimuli from the international affective picture system (IAPS), categorized in four groups ranging in pleasure (valence) and arousal. Four different pair-wise metrics were used to estimate the connectivity between each pair of EEG channels: the magnitude square coherence (MSC), cross-correlation (CCOR), normalized mutual information (NMI) and normalized joint entropy (NJE). The small-world-ness is found to be varying among the connectivity metrics, while it was also affected by the choice of the threshold level. The use of different connectivity metrics affected the significance of the neurophysiological results. However, the results from different metrics were to the same direction: pleasant images exhibited shorter characteristic path length than unpleasant ones, while high arousing images were related to lower local efficiency as compared to the low arousing ones. Our findings suggest that the choice of different metrics modulates the small-world-ness of the FCNs as well as the neurophysiological results and should be taken into account when studying brain functional connectivity using graph theory.

. 

Abstract

Acute alcohol intake is known to enhance inhibition through facilitation of GABAA receptors, which are present in 40% of the synapses all over the brain. Evidence suggests that enhanced GABAergic transmission leads to increased large-scale brain connectivity. Our hypothesis is that acute alcohol intake would increase the functional connectivity of the human brain resting-state network (RSN). To test our hypothesis, electroencephalographic (EEG) measurements were recorded from healthy social drinkers at rest, during eyes-open and eyes-closed sessions, after administering to them an alcoholic beverage or placebo respectively. Salivary alcohol and cortisol served to measure the inebriation and stress levels. By calculating Magnitude Square Coherence (MSC) on standardized Low Resolution Electromagnetic Tomography (sLORETA) solutions, we formed cortical networks over several frequency bands, which were then analyzed in the context of functional connectivity and graph theory. MSC was increased (p<0.05, corrected with False Discovery Rate, FDR corrected) in alpha, beta (eyes-open) and theta bands (eyes-closed) following acute alcohol intake. Graph parameters were accordingly altered in these bands quantifying the effect of alcohol on the structure of brain networks; global efficiency and density were higher and path length was lower during alcohol (vs. placebo, p<0.05). Salivary alcohol concentration was positively correlated with the density of the network in beta band. The degree of specific nodes was elevated following alcohol (vs. placebo). Our findings support the hypothesis that short-term inebriation considerably increases large-scale connectivity in the RSN. The increased baseline functional connectivity can -at least partially- be attributed to the alcohol-induced disruption of the delicate balance between inhibitory and excitatory neurotransmission in favor of inhibitory influences. Thus, it is suggested that short-term inebriation is associated, as expected, to increased GABA transmission and functional connectivity, while long-term alcohol consumption may be linked to exactly the opposite effect. 

Abstract

Introduction. Sensorimotor cortex is activated similarly during motor execution and motor imagery. The study of functional connectivity networks (FCNs) aims at successfully modeling the dynamics of information flow between cortical areas. Materials and Methods. Seven healthy subjects performed 4 motor tasks (real foot, imaginary foot, real hand, and imaginary hand movements), while electroencephalography was recorded over the sensorimotor cortex. Event-Related Desynchronization/Synchronization (ERD/ERS) of the mu-rhythm was used to evaluate MI performance. Source detection and FCNs were studied with eConnectome. Results and Discussion. Four subjects produced similar ERD/ERS patterns between motor execution and imagery during both hand and foot tasks, 2 subjects only during hand tasks, and 1 subject only during foot tasks. All subjects showed the expected brain activation in well-performed MI tasks, facilitating cortical source estimation. Preliminary functional connectivity analysis shows formation of networks on the sensorimotor cortex during motor imagery and execution. Conclusions. Cortex activation maps depict sensorimotor cortex activation, while similar functional connectivity networks are formed in the sensorimotor cortex both during actual and imaginary movements. eConnectome is demonstrated as an effective tool for the study of cortex activation and FCN. The implementation of FCN in motor imagery could induce promising advancements in Brain Computer Interfaces.