Machine learning for hand pose classification from phasic and tonic EMG signals during bimanual activities in virtual reality.

Myoelectric prostheses have recently shown significant promise for restoring hand function in individuals with upper limb loss or deficiencies, driven by advances in machine learning and increasingly accessible bioelectrical signal acquisition devices. Here, we first introduce and validate a novel experimental paradigm using a virtual reality headset equipped with hand-tracking capabilities to facilitate the recordings of synchronized EMG signals and hand pose estimation. Using both the phasic and tonic EMG components of data acquired through the proposed paradigm, we compare hand gesture classification pipelines based on standard signal processing features, convolutional neural networks, and covariance matrices with Riemannian geometry computed from raw or xDAWN-filtered EMG signals. We demonstrate the performance of the latter for gesture classification using EMG signals. We further hypothesize that introducing physiological knowledge in machine learning models will enhance their performances, leading to better myoelectric prosthesis control. We demonstrate the potential of this approach by using the neurophysiological integration of the "move command" to better separate the phasic and tonic components of the EMG signals, significantly improving the performance of sustained posture recognition. These results pave the way for the development of new cutting-edge machine learning techniques, likely refined by neurophysiology, that will further improve the decoding of real-time natural gestures and, ultimately, the control of myoelectric prostheses.

Sleep-push movement performance in elite field hockey champions with and without training specialization.

Objective: To investigate kinematic and muscle activity differences during the sleep-push movement in elite field hockey players. We hypothesized that players with specialized sleep-push movement training (specialists) would possess a lower center of mass (CoM) and enhanced reproducibility of muscle activations during the movement, compared to players without explicit movement training (non-specialists). Methods: Ten field hockey players of the Belgian national field hockey team performed the sleep-push movement (5 specialists and 5 non-specialists). Muscle activity and kinematic data were recorded using EMG to evaluate the reproducibility of muscle activations by cross-correlation analysis and power spectral features across the movement, while a motion capture system was used to assess kinematics. Results: Compared to non-specialists, specialists had significantly (p < 0.05) increased stick velocity (9.17 ± 1.28 m/s versus 6.98 ± 0.97 m/s) and lower CoM height (141 ± 52 mm versus 296 ± 64 mm), during the second part of the shot. Specialists also showed a significant (p < 0.05) lower power spectrum in the activity of the upper limb muscles before the shot. Superimposition of the auto crosscorrelation results demonstrated a high degree of reproducibility in specialists' muscle activations. Conclusion: Sleep-push movements realized by elite players who are specialists in the sleep-push movement presented significant kinematics and muscular activation differences when compared to the sleep-push movements realized by elite players who were not specialists in such movement. Characterization of the specific movement and the related high-level performer's muscular strategies offers the possibility of translating sport science findings into functional training with concrete applications for coaches, players, and other key stakeholders for the continued development of the field.

Effects of Pulsed-Wave Chromotherapy and Guided Relaxation on the Theta-Alpha Oscillation During Arrest Reaction.

The search for the best wellness practice has promoted the development of devices integrating different technologies and guided meditation. However, the final effects on the electrical activity of the brain remain relatively sparse. Here, we have analyzed of the alpha and theta electroencephalographic oscillations during the realization of the arrest reaction (AR; eyes close/eyes open transition) when a chromotherapy session performed in a dedicated room [Rebalance (RB) device], with an ergonomic bed integrating pulsed-wave light (PWL) stimulation, guided breathing, and body scan exercises. We demonstrated that the PWL induced an evoked-related potential characterized by the N2-P3 components maximally recorded on the fronto-central areas and accompanied by an event-related synchronization (ERS) of the delta-theta-alpha oscillations. The power of the alpha and theta oscillations was analyzed during repeated ARs testing realized along with the whole RB session. We showed that the power of the alpha and theta oscillations was significantly increased during the session in comparison to their values recorded before. Of the 14 participants, 11 and 6 showed a significant power increase of the alpha and theta oscillations, respectively. These increased powers were not observed in two different control groups (n = 28) who stayed passively outside or inside the RB room but without any type of stimulation. These preliminary results suggest that PWL chromotherapy and guided relaxation induce measurable electrical brain changes that could be beneficial under neuropsychiatric perspectives.

Riemannian classification of single-trial surface EEG and sources during checkerboard and navigational images in humans.

OBJECTIVE: Different visual stimuli are classically used for triggering visual evoked potentials comprising well-defined components linked to the content of the displayed image. These evoked components result from the average of ongoing EEG signals in which additive and oscillatory mechanisms contribute to the component morphology. The evoked related potentials often resulted from a mixed situation (power variation and phase-locking) making basic and clinical interpretations difficult. Besides, the grand average methodology produced artificial constructs that do not reflect individual peculiarities. This motivated new approaches based on single-trial analysis as recently used in the brain-computer interface field. APPROACH: We hypothesize that EEG signals may include specific information about the visual features of the displayed image and that such distinctive traits can be identified by state-of-the-art classification algorithms based on Riemannian geometry. The same classification algorithms are also applied to the dipole sources estimated by sLORETA. MAIN RESULTS AND SIGNIFICANCE: We show that our classification pipeline can effectively discriminate between the display of different visual items (Checkerboard versus 3D navigational image) in single EEG trials throughout multiple subjects. The present methodology reaches a single-trial classification accuracy of about 84% and 93% for inter-subject and intra-subject classification respectively using surface EEG. Interestingly, we note that the classification algorithms trained on sLORETA sources estimation fail to generalize among multiple subjects (63%), which may be due to either the average head model used by sLORETA or the subsequent spatial filtering failing to extract discriminative information, but reach an intra-subject classification accuracy of 82%.

Increased brain reactivity to gambling unavailability as a marker of problem gambling.

The unprecedented development and ubiquity of sports betting constitute an emerging public health concern. It is crucial to provide markers that could help to better identify people experiencing sports betting-related harms. The current study investigated whether problem gambling status, sports betting passion, and trait-self-control modulate brain reactivity to sports betting cues. Sixty-five frequent sports bettors (35 "nonproblem bettors" and 30 "problem bettors") were exposed to cues representing real upcoming sport events (with varying levels of winning confidence) that were made available or blocked for betting, during functional magnetic resonance imaging (fMRI) recording. Sports betting passion and trait-self-control were assessed using self-report scales. Sport events nonavailable for betting elicited higher insular and striatal activation in problem bettors, as compared with nonproblem bettors. Within a large cluster encompassing the ventral striatum, hippocampus, and amygdala, lower trait-self-control was associated with increased brain reactivity to sport events with high levels of winning confidence that were nonavailable for betting. No significant effect of sports betting passion was observed. These findings suggest that sports bettors' brain reactivity to gambling unavailability might be a relevant marker of sports betting-related harms, as well as of blunted trait-self-control.

Hyperscanning EEG and Classification Based on Riemannian Geometry for Festive and Violent Mental State Discrimination.

Interactions between two brains constitute the essence of social communication. Daily movements are commonly executed during social interactions and are determined by different mental states that may express different positive or negative behavioral intent. In this context, the effective recognition of festive or violent intent before the action execution remains crucial for survival. Here, we hypothesize that the EEG signals contain the distinctive features characterizing movement intent already expressed before movement execution and that such distinctive information can be identified by state-of-the-art classification algorithms based on Riemannian geometry. We demonstrated for the first time that a classifier based on covariance matrices and Riemannian geometry can effectively discriminate between neutral, festive, and violent mental states only on the basis of non-invasive EEG signals in both the actor and observer participants. These results pave the way for new electrophysiological discrimination of mental states based on non-invasive EEG recordings and cutting-edge machine learning techniques.

Local sleep-like events during wakefulness and their relationship to decreased alertness in astronauts on ISS.

Adequate sleep quantity and quality is required to maintain vigilance, cognitive and learning processes. A decrease of sleep quantity preflight and on the International Space Station (ISS) has been reported. Recent counter-measures have been implemented to better regulate sleep opportunities on ISS. In our study, astronauts were allocated enough time for sleep the night before the recordings. However, for proper sleep recovery, the quality of sleep is also critical. Unfortunately, data on sleep quality have yet to be acquired from the ISS. Here, we investigate sleep pressure markers during wakefulness in five astronauts throughout their 6-month space mission by the mean of electroencephalographic recordings. We show a global increase of theta oscillations (5-7 Hz) on the ISS compared to on Earth before the mission. We also show that local sleep-like events, another marker of sleep pressure, are more global in space (p < 0.001). By analysing the performances of the astronauts during a docking simulation, we found that local sleep-like events are more global when reaction times are slower (R 2 = 0.03, p = 0.006) and there is an increase of reaction times above 244 ms after 2 months in space (p = 0.012). Our analyses provide first evidence for increased sleep pressure in space and raise awareness on possible impacts on visuomotor performances in space.

Facing temptation: The neural correlates of gambling availability during sports picture exposure.

Nowadays, sports betting has become increasingly available and easy to engage in. Here we examined the neural responses to stimuli that represent sporting events available for betting as compared to sporting events without a gambling opportunity. We used a cue exposure task in which football (soccer) fans (N = 42) viewed cues depicting scheduled football games that would occur shortly after the scanning session. In the "betting" condition, participants were instructed to choose, at the end of each block, the game (and the team) they wanted to bet on. In the "watching" condition, participants chose the game they would prefer to watch. After the scanning session, participants completed posttask rating questionnaires assessing, for each cue, their level of confidence about the team they believed would win and how much they would enjoy watching the game. We found that stimuli representing sport events available for betting elicited higher fronto-striatal activation, as well as higher insular cortex activity and functional connectivity, than sport events without a gambling opportunity. Moreover, games rated with more confidence towards the winning team resulted in greater brain activations within regions involved in affective decision-making (ventromedial prefrontal cortex), cognitive inhibitory control (medial and superior frontal gyri) and reward processing (ventral and dorsal striatum). Altogether, these novel findings offer a sensible simulation of how the high availability of sports betting in today's environment impacts on the reward and cognitive control systems. Future studies are needed to extend the present findings to a sample of football fans that includes a samilar proportion of female and male participants.

Proactive and Reactive Motor Inhibition in Top Athletes Versus Nonathletes.

We examined proactive (early restraint in preparation for stopping) and reactive (late correction to stop ongoing action) motor response inhibition in two groups of participants: professional athletes ( n = 28) and nonathletes ( n = 25). We recruited the elite athletes from Belgian national taekwondo and fencing teams. We estimated proactive and reactive inhibition with a modified version of the stop-signal task (SST) in which participants inhibited categorizing left/right arrows. The probability of the stop signal was manipulated across blocks of trials by providing probability cues from the background computer screen color (green = 0%, yellow =17%, orange = 25%, red = 33%). Participants performed two sessions of the SST, where proactive inhibition was operationalized with increased go-signal reaction time as a function of increased stop-signal probability and reactive inhibition was indicated by stop-signal reaction time latency. Athletes exhibited higher reactive inhibition performance than nonathletes. In addition, athletes exhibited higher proactive inhibition than nonathletes in Session 1 (but not Session 2) of the SST. As top-level athletes exhibited heightened reactive inhibition and were faster to reach and maintain consistent proactive motor response inhibition, these results confirm an evaluative process that can discriminate elite athleticism through a fine-grained analysis of inhibitory control.

Neural correlates of correct and failed response inhibition in heavy versus light social drinkers: an fMRI study during a go/no-go task by healthy participants.

The ability to suppress responses that are inappropriate, as well as the mechanisms monitoring the accuracy of actions in order to compensate for errors, is central to human behavior. Neural alterations that prevent stopping an inaccurate response, combined with a decreased ability of error monitoring, are considered to be prominent features of alcohol abuse. Moreover, (i) alterations of these processes have been reported in heavy social drinkers (i.e. young healthy individuals who do not yet exhibit a state of alcohol dependence); and (ii) through longitudinal studies, these alterations have been shown to underlie subsequent disinhibition that may lead to future alcohol use disorders. In the present functional magnetic resonance imaging study, using a contextual Go/No-Go task, we investigated whether different neural networks subtended correct inhibitions and monitoring mechanisms of failed inhibitory trials in light versus heavy social drinkers. We show that, although successful inhibition did not lead to significant changes, neural networks involved in error monitoring are different in light versus heavy drinkers. Thus, while light drinkers exhibited activations in their right inferior frontal, right middle cingulate and left superior temporal areas; heavy drinkers exhibited activations in their right cerebellum, left caudate nucleus, left superior occipital region, and left amygdala. These data are functionally interpreted as reflecting a "visually-driven emotional strategy" vs. an "executive-based" neural response to errors in heavy and light drinkers, respectively. Such a difference is interpreted as a key-factor that may subtend the transition from a controlled social heavy consumption to a state of clinical alcohol dependence.

Brain Oscillations in Sport: Toward EEG Biomarkers of Performance.

Brain dynamics is at the basis of top performance accomplishment in sports. The search for neural biomarkers of performance remains a challenge in movement science and sport psychology. The non-invasive nature of high-density electroencephalography (EEG) recording has made it a most promising avenue for providing quantitative feedback to practitioners and coaches. Here, we review the current relevance of the main types of EEG oscillations in order to trace a perspective for future practical applications of EEG and event-related potentials (ERP) in sport. In this context, the hypotheses of unified brain rhythms and continuity between wake and sleep states should provide a functional template for EEG biomarkers in sport. The oscillations in the thalamo-cortical and hippocampal circuitry including the physiology of the place cells and the grid cells provide a frame of reference for the analysis of delta, theta, beta, alpha (incl.mu), and gamma oscillations recorded in the space field of human performance. Based on recent neuronal models facilitating the distinction between the different dynamic regimes (selective gating and binding) in these different oscillations we suggest an integrated approach articulating together the classical biomechanical factors (3D movements and EMG) and the high-density EEG and ERP signals to allow finer mathematical analysis to optimize sport performance, such as microstates, coherency/directionality analysis and neural generators.

Long-Lasting Cortical Reorganization as the Result of Motor Imagery of Throwing a Ball in a Virtual Tennis Court.

In order to characterize the neural signature of a motor imagery (MI) task, the present study investigates for the first time the oscillation characteristics including both of the time-frequency measurements, event related spectral perturbation and intertrial coherence (ITC) underlying the variations in the temporal measurements (event related potentials, ERP) directly related to a MI task. We hypothesize that significant variations in both of the time-frequency measurements underlie the specific changes in the ERP directly related to MI. For the MI task, we chose a simple everyday task (throwing a tennis ball), that does not require any particular motor expertise, set within the controlled virtual reality scenario of a tennis court. When compared to the rest condition a consistent, long-lasting negative fronto-central ERP wave was accompanied by significant changes in both time frequency measurements suggesting long-lasting cortical activity reorganization. The ERP wave was characterized by two peaks at about 300 ms (N300) and 1000 ms (N1000). The N300 component was centrally localized on the scalp and was accompanied by significant phase consistency in the delta brain rhythms in the contralateral central scalp areas. The N1000 component spread wider centrally and was accompanied by a significant power decrease (or event related desynchronization) in low beta brain rhythms localized in fronto-precentral and parieto-occipital scalp areas and also by a significant power increase (or event related synchronization) in theta brain rhythms spreading fronto-centrally. During the transition from N300 to N1000, a contralateral alpha (mu) as well as post-central and parieto-theta rhythms occurred. The visual representation of movement formed in the minds of participants might underlie a top-down process from the fronto-central areas which is reflected by the amplitude changes observed in the fronto-central ERPs and by the significant phase synchrony in contralateral fronto-central delta and contralateral central mu to parietal theta presented here.

Neural rhythmic symphony of human walking observation: Upside-down and Uncoordinated condition on cortical theta, alpha, beta and gamma oscillations.

Biological motion observation has been recognized to produce dynamic change in sensorimotor activation according to the observed kinematics. Physical plausibility of the spatial-kinematic relationship of human movement may play a major role in the top-down processing of human motion recognition. Here, we investigated the time course of scalp activation during observation of human gait in order to extract and use it on future integrated brain-computer interface using virtual reality (VR). We analyzed event related potentials (ERP), the event related spectral perturbation (ERSP) and the inter-trial coherence (ITC) from high-density EEG recording during video display onset (-200-600 ms) and the steady state visual evoked potentials (SSVEP) inside the video of human walking 3D-animation in three conditions: Normal; Upside-down (inverted images); and Uncoordinated (pseudo-randomly mixed images). We found that early visual evoked response P120 was decreased in Upside-down condition. The N170 and P300b amplitudes were decreased in Uncoordinated condition. In Upside-down and Uncoordinated conditions, we found decreased alpha power and theta phase-locking. As regards gamma oscillation, power was increased during the Upside-down animation and decreased during the Uncoordinated animation. An SSVEP-like response oscillating at about 10 Hz was also described showing that the oscillating pattern is enhanced 300 ms after the heel strike event only in the Normal but not in the Upside-down condition. Our results are consistent with most of previous point-light display studies, further supporting possible use of virtual reality for neurofeedback applications.

Performance of the Emotiv Epoc headset for P300-based applications.

BACKGROUND: For two decades, EEG-based Brain-Computer Interface (BCI) systems have been widely studied in research labs. Now, researchers want to consider out-of-the-lab applications and make this technology available to everybody. However, medical-grade EEG recording devices are still much too expensive for end-users, especially disabled people. Therefore, several low-cost alternatives have appeared on the market. The Emotiv Epoc headset is one of them. Although some previous work showed this device could suit the customer's needs in terms of performance, no quantitative classification-based assessments compared to a medical system are available. METHODS: This paper aims at statistically comparing a medical-grade system, the ANT device, and the Emotiv Epoc headset by determining their respective performances in a P300 BCI using the same electrodes. On top of that, a review of previous Emotiv studies and a discussion on practical considerations regarding both systems are proposed. Nine healthy subjects participated in this experiment during which the ANT and the Emotiv systems are used in two different conditions: sitting on a chair and walking on a treadmill at constant speed. RESULTS: The Emotiv headset performs significantly worse than the medical device; observed effect sizes vary from medium to large. The Emotiv headset has higher relative operational and maintenance costs than its medical-grade competitor. CONCLUSIONS: Although this low-cost headset is able to record EEG data in a satisfying manner, it should only be chosen for non critical applications such as games, communication systems, etc. For rehabilitation or prosthesis control, this lack of reliability may lead to serious consequences. For research purposes, the medical system should be chosen except if a lot of trials are available or when the Signal-to-Noise Ratio is high. This also suggests that the design of a specific low-cost EEG recording system for critical applications and research is still required.

Biological oscillations for learning walking coordination: dynamic recurrent neural network functionally models physiological central pattern generator.

The existence of dedicated neuronal modules such as those organized in the cerebral cortex, thalamus, basal ganglia, cerebellum, or spinal cord raises the question of how these functional modules are coordinated for appropriate motor behavior. Study of human locomotion offers an interesting field for addressing this central question. The coordination of the elevation of the 3 leg segments under a planar covariation rule (Borghese et al., 1996) was recently modeled (Barliya et al., 2009) by phase-adjusted simple oscillators shedding new light on the understanding of the central pattern generator (CPG) processing relevant oscillation signals. We describe the use of a dynamic recurrent neural network (DRNN) mimicking the natural oscillatory behavior of human locomotion for reproducing the planar covariation rule in both legs at different walking speeds. Neural network learning was based on sinusoid signals integrating frequency and amplitude features of the first three harmonics of the sagittal elevation angles of the thigh, shank, and foot of each lower limb. We verified the biological plausibility of the neural networks. Best results were obtained with oscillations extracted from the first three harmonics in comparison to oscillations outside the harmonic frequency peaks. Physiological replication steadily increased with the number of neuronal units from 1 to 80, where similarity index reached 0.99. Analysis of synaptic weighting showed that the proportion of inhibitory connections consistently increased with the number of neuronal units in the DRNN. This emerging property in the artificial neural networks resonates with recent advances in neurophysiology of inhibitory neurons that are involved in central nervous system oscillatory activities. The main message of this study is that this type of DRNN may offer a useful model of physiological central pattern generator for gaining insights in basic research and developing clinical applications.

A subjective assessment of a P300 BCI system for lower-limb rehabilitation purposes.

Recent research has shown that a P300 system can be used while walking without requiring any specific gait-related artifact removal techniques. Also, standard EEG-based Brain-Computer Interfaces (BCI) have not been really assessed for lower limb rehabilitation/prosthesis. Therefore, this paper gives a first baseline estimation (for future BCI comparisons) of the subjective and objective performances of a four-state P300 BCI plus a non-control state for lower-limb rehabilitation purposes. To assess usability and workload, the System Usability Scale and the NASA Task Load Index questionnaires were administered to five healthy subjects after performing a real-time treadmill speed control. Results show that the P300 BCI approach could suit fitness and rehabilitation applications, whereas prosthesis control, which suffers from a low reactivity, appears too sensitive for risky and crowded areas.