Search

In this paper, we investigated the relationships between EEG and semantic long-term memory retrieval using spectral analysis and functional connectivity. We employed an EEG data set from a previous study, which was recorded from 34 healthy participants during a memory retrieval process. The data set consists of EEG collected during two tasks, a learning task, in which participants studied animated learning material, and a memory recall task, in which they recalled the learned information. We analyzed both EEG signals and behavioral responses. We conducted spectral analyses for the absolute power and relative power of EEG frequency bands, EEG coherence, absolute phase delay, and EEG amplitude asymmetry features. The delta-band EEG and theta-band absolute power were negatively correlated with task performance at frontal regions, while the alpha-band activity was positively correlated in frontal regions and negatively correlated in parietal and temporal sites. High-beta in the frontal lobe was positively correlated with memory recall score. EEG coherence was negatively correlated in the delta band at the fronto-parietal and fronto-central sites, and positively correlated in the gamma band at widespread regions, including all lobes. EEG phase delay showed the highest degree of connectivity in the delta and alpha bands at frontal sites, and was negatively correlated with memory recall score. The negative correlation in phase delay demonstrated the synchronization and positive correlation reflected the de-synchronization of electrode pairs during memory recollection in the delta and alpha frequencies. We concluded that neuronal networks producing low EEG frequencies at frontal regions and those producing high frequencies in widespread regions are active during the memory recollection process. Thus, semantic memory storage involves widespread neuronal networks across the brain.

International audience; Photoplethysmography and Ballistocardiography are two concepts that are used to measure heart rate from human, by using facial videos. Heart rate estimation is essential to determine the physiological and pathological state of a person. This paper presents a critical review of digital camera based heart rate estimating method on facial skin. This review extends the investigation on to the principles and theory behind photoplethysmography and ballistocardiography. The article contains reviews on the significance of the methods and contributions to overcome challenges such as; poor signal strength, illumination variance, and motion variance. The experiments were conducted to validate the state of the art methods on a challenging database that is available publicly. The implemented methods were validated using the database, on 27 subjects for a range of skin tones from pearl white, fair, olive to black. The results were computed using statistical methods such as: mean error, standard deviation, the root mean square error, Pearson correlation coefficient, and Bland-Altman analysis. The results derived from the experiments showed the reliability of the state of the art methods and provided direction to improve for situations involving illumination variance and motion variance. (C) 2017 Elsevier Ltd. All rights reserved.

In this paper we present a novel health monitoring method by estimating the heart rate and respiratory rate using an RGB camera. The heart rate and the respiratory rate are estimated from the photoplethysmography (PPG) and the respiratory motion. The method mainly operates by using the green spectrum of the RGB camera to generate a multivariate PPG signal to perform multivariate de-noising on the video signal to extract the resultant PPG signal. A periodicity based voting scheme (PVS) was used to measure the heart rate and respiratory rate from the estimated PPG signal. We evaluated our proposed method with a state of the art heart rate measuring method for two scenarios using the MAHNOB-HCI database and a self collected naturalistic environment database. The methods were furthermore evaluated for various scenarios at naturalistic environments such as a motion variance session and a skin tone variance session. Our proposed method operated robustly during the experiments and outperformed the state of the art heart rate measuring methods by compensating the effects of the naturalistic environment.

International audience; Create a complete system of video surveillance using camera mounted on a robot like UAV to maintain optimized vast area coverage and reconstruct an image by using mosaicing techniques. This paper demonstrated the efficiency of using one UAV to cover vast area using optimized positions.

International audience; Subcutaneous veins localization is basic and important step for any intravenous medication administration. Due to different physiological characteristics, mainly darker skin tone, scars or dehydrated condition of patients, medical staff face difficulty in veins localization. Through near infrared imaging technology the veins can be visualized due to high contrast between veins and skin tissue in this modality. Information on the depth of veins is equally important for proper catheterization or venipuncture procedures. Patients have different veins depth due to the different amount of fat present in the subcutaneous layer. The depth of veins from the skin surface cannot be estimated by simple imaging technique. In this paper a mathematical model to estimate the depth of veins based on measured diffused reflectance is presented. A layered model of Monte Carlo simulations for light transport in turbid medium was used to validate the results.

Cloud computing is a methodology and not a technology. Adaptation of cloud computing services for robotic applications is relatively straightforward while adaptation of underlying ideas will require a new design attitude. Cloud computing is a cost-effective and dynamic business model. Currently cloud robotics is understood as a client server methodology which enables robots utilize resources and services placed at centralized servers. These cloud servers treat robots as any other client computer offering them platform, infrastructure, process or algorithm as a service. HTM5 is an OMG MDA based multi-view meta-model for agent oriented development of cloud robotic systems. HTM5 encourages design of peer-to-peer service ecosystems based on an open registry and matchmaking mechanism. In peer-to-peer cloud robotics, a robot can trade its hardware, software and functional resources as a service to other robots in the ecosystem. The peer-to-peer trade in such systems may be driven by contracts and relationships between its member agents. This article discusses trade-view model of HTM5 methodology and its use in developing a cloud robotic ecosystem that implements peer-to-peer, contract based economy. The article also presents a case study with experiments that implement distributed artificial intelligence and peer-to-peer service oriented trade on simulated and real robot colonies.

Provisioning network survivability is especially crucial in wireless sensor and actor network (WSAN) because nodes deployed in hostile environments are prone to frequent failures. Failure of an actor significantly impact actor connected coverage which is essential for effective network operation. Existing mobility-based recovery schemes are either geared towards restoring inter-actor connectivity or area coverage. None of them consider sustaining actor coverage (i.e., having sensors reachable to actors) while restoring inter-actor connectivity. This paper presents RACE, a novel mechanism to Restore Actor Connected Coverage with reduced recovery overhead. RACE distinguishes critical/non-critical actors based on 2-hop information to better assess the scope of the failure and optimize the recovery procedure. Neighbors of a failed actor employ a cooperative failure detection scheme and only perform a limited-scale network reconfiguration to adopt any bereaved sensors left unreachable (uncovered by an actor) due to failure of a non-critical actor. In case a critical actor fails, RACE substitutes it with a non-critical neighbor that has the least impact on coverage (i.e., number of sensors). If it is necessary to engage critical actors in the recovery, RACE is recursively applied by relocating actors until a non-critical node is picked. Simulation results confirm the performance advantage of RACE compared to the best contemporary schemes. 2015 IEEE. Scopus 2-s2.0-84953727973

International audience; Intravenous (IV) catheterization is the most important phase in medical practices of daily life. It is hard to localize veins in patients who have deep veins, minor age or dark skin; hence multiple attempts become indispensable for proper catheterization in such cases. Near Infrared (NIR) Imaging allow to visualize the veins underneath the skin of persons having non-visibility of veins problem. This paper reports the pre-selection of illuminants that ensure best veins/tissues contrast for patients having different skin tone. The sample subjects have been divided in four different classes based on the Luminance value of their skin tone in order to extract the best illuminant wavelengths range for each class. A multispectral approach has been used which provides the flexibility of wavelength range from visible to NIR (380 to 1040nm). The veins/tissue reflectance contrast obtained helps in determining the best wavelengths range where the contrast is maximum for each of the four classes. Using these results, we are planning to build a prototype system which can automatically select the illuminants based on different physiological characteristics of a subject.

Driving distraction shift the attention away from safe driving towards a competing task. Dual task paradigm defines distraction in terms of excessive workload and limited attentional resources. To fill the gap in driver’s safety research, the thesis aims to uncover the mechanics of detecting visual cognitive distraction by using physiological signals. There are 2 sets of experiments conducted with 18 participants. The first experiment with 10 participants aims at evaluating the effects of distraction and synchronising EEG signals. For statistical analysis, Shapiro Wilk test and two-way ANOVA have been used. The results state that the two visual cognitive distraction tasks are performed sequentially and reaction time is affected by the switch. For the second experiment with 8 participants, an integrated system has been created in C++, which involves a 4-wall CAVE system, a driving simulator using SCANeR simulation software, a simple EEG headset with 20 channels (Enobio 20), SMI Eye tracker, BeGaze software. As qualitative methods, SSQ and NASA TLX have been used to evaluate user feedback. As quantitative methods, maths equations, eye movements, EEG signals, and driving performance have been used. For data analysis, Matlab and SPSS and for statistical analysis, Shapiro Wilk test and one-way ANOVA have been used. The results indicate that experiment demands mental capacity, rather than physical capacity. Delta band frequencies in frontal lobe of EEG signals do not have a significant impact on discriminating between driving and distraction, but delta and theta band at central and parietal do. The results also revealed that theta and alpha band frequencies are indeed an important features in both experiments. In a task where subject can control when they are engaging towards the secondary task, beta band showed its statistical importance being the only one able to discriminate up to different level of cognitive tasks. The most striking similarity between both experiments were, some of the statistically significant features are overlapped between brain regions. A comparative analysis of machine learning techniques (SVMs vs Random Forests) in selecting relevant features of EEG signals, referring to the coherence values, has also been conducted. Results indicate that although in general random forests perform better than SVM classifier, F1-score for SVM demonstrates the highest scores (0.90).; Le principe de la distraction au volant est de détourner l'attention du conducteur en phase de conduite en toute sécurité au profit d'une tâche concurrentielle. Le paradigme de la double tâche repose sur l’hypothèse que la distraction se traduit par une charge cognitive excessive et des ressources restreintes en termes d'attention. Pour combler les lacunes des travaux de recherche portant sur l’étude de la sécurité des conducteurs, cette thèse vise à découvrir les mécanismes de détection de la distraction cognitive visuelle à l'aide de signaux physiologiques. A cette fin, deux séries d'expérimentations ont été menées avec 18 participants. La première expérimentation avec 10 participants vise à évaluer les effets de la distraction et de la synchronisation des signaux de l’EEG. Pour l'analyse statistique, le test de Shapiro-Wilk et l'analyse de variance à deux facteurs ANOVA ont été utilisés. Les résultats indiquent que les deux tâches de distraction cognitive visuelle sont exécutées de façon séquentielle et que le temps de réaction est affecté par le passage d'une tâche à l'autre. Pour la deuxième expérimentation avec 8 participants, un système intégré a été créé en C++, qui implique un système CAVE à 4 parois, un simulateur de conduite utilisant le logiciel de simulation SCANeR, un simple casque EEG à 20 canaux (Enobio 20), un SMI de suivi oculaire et un logiciel BeGaze. Comme paramètres qualitatifs, les réponses aux questionnaires SSQ et NASA-TLX ont été utilisés pour évaluer les commentaires des utilisateurs. Comme paramètres quantitatifs, des solutions d’équations mathématiques, des mesures de mouvements oculaires, de signaux EEG et de performances de conduite ont été utilisées. Pour l'analyse des données, les logiciels Matlab et SPSS ont été utilisés et pour l'analyse statistique, le test Shapiro-Wilk et ANOVA à un facteur. Les résultats indiquent que l'expérience exige une capacité mentale plutôt qu'une capacité physique. Les fréquences de la bande delta dans le lobe frontal des signaux EEG n'ont pas d'impact significatif sur la différentiation entre conduite et distraction, mais les bandes delta et thêta au centre et au pariétal en ont. Les résultats ont également révélé que les fréquences des bandes thêta et alpha sont effectivement des caractéristiques importantes dans les deux expérimentations. Dans une tâche où le sujet peut contrôler quand il s'engage dans la tâche secondaire, la bande bêta a montré son importance d’un point de vue statistique, étant la seule capable de faire la distinction entre différents niveaux des tâches cognitives. La similitude la plus frappante entre les deux expérimentations est que certaines des caractéristiques statistiquement significatives se chevauchent entre les régions du cerveau. Par conséquent, une analyse comparative des techniques d'apprentissage machine (SVM-RFE c. les forêts aléatoires) dans la sélection des caractéristiques pertinentes des signaux EEG, en combinaison avec les valeurs de cohérence, a également été réalisée. Les résultats indiquent que, bien qu'en général, les forêts aléatoires donnent de meilleurs résultats que le séparateur SVM, le score F1 du SVM avec SVM-RFE montre les scores les plus élevés (0.90).