Estimating underlying neuronal activity from EEG using an iterative sparse technique

When mechanical engineering inspired from physiology improves postural-related somatosensory processes

Despite numerous studies uncovering the neural signature of tactile processing, tactile afferent inputs relating to the contact surface has not been studied so far. Foot tactile receptors being the first stimulated by the relative movement of the foot skin and the underneath moving support play an important role in the sensorimotor transformation giving rise to a postural reaction. A biomimetic surface, i.e., complying with the skin dermatoglyphs and tactile receptors characteristics should facilitate the cortical processes. Participants (n = 15) stood either on a biomimetic surface or on two control surfaces, when a sudden acceleration of the supporting surface was triggered (experiment 1). A larger intensity and shorter somatosensory response (i.e., SEP) was evoked by the biomimetic surface motion. This result and the associated decrease of theta activity (5-7 Hz) over the posterior parietal cortex suggest that increasing the amount of sensory input processing could make the balance task less challenging when standing on a biomimetic surface. This key point was confirmed by a second experiment (n = 21) where a cognitive task was added, hence decreasing the attentional resources devoted to the balance motor task. Greater efficiency of the postural reaction was observed while standing on the biomimetic than on the control surfaces.

Cortical facilitation of tactile afferents during the preparation of a body weight transfer when standing on a biomimetic surface

Self-generated movement shapes tactile perception, but few studies have investigated the brain mechanisms involved in the processing of the mechanical signals related to the static and transient skin deformations generated by forces and pressures exerted between the foot skin and the standing surface. We recently found that standing on a biomimetic surface (i.e., inspired by the characteristics of mechanoreceptors and skin dermatoglyphics), that magnified skin-surface interaction, increased the sensory flow to the somatosensory cortex and improved balance control compared to standing on control (e.g., smooth) surfaces. In this study, we tested whether the well-known sensory suppression that occurs during movements is alleviated when the tactile afferent signal becomes relevant with the use of a biomimetic surface. Eyes-closed participants (n = 25) self-stimulated their foot cutaneous receptors by shifting their body weight toward one of their legs while standing on either a biomimetic or a control (smooth) surface. In a control task, similar forces were exerted on the surfaces (i.e., similar skin-surface interaction) by passive translations of the surfaces. Sensory gating was assessed by measuring the amplitude of the somatosensory-evoked potential over the vertex (SEP, recorded by EEG). Significantly larger and shorter SEPs were found when participants stood on the biomimetic surface. This was observed whether the forces exerted on the surface were self-generated or passively generated. Contrary to our prediction, we found that the sensory attenuation related to the self-generated movement did not significantly differ between the biomimetic and control surfaces. However, we observed an increase in gamma activity (30-50 Hz) over centroparietal regions during the preparation phase of the weight shift only when participants stood on the biomimetic surface. This result might suggest that gamma-band oscillations play an important functional role in processing behaviorally relevant stimuli during the early stages of body weight transfer.

Identifying epileptic source location and extent: an iterative sparse electromagnetic source imaging algorithm

In this paper we have introduced a novel electromagnetic source imaging (ESI) technique and demonstrated its validity and excellent performance in imaging the location and extent of underlying epileptic sources in patients suffering from focal epilepsy. The proposed algorithm employs ideas from sparse signal processing literature and convex optimization theories to improve source imaging results obtained from scalp-recorded electroencephalogram (EEG). EEG source imaging results generally use subjective methods to determine the extent of the underlying brain activity. The proposed technique provides significant improvement in dealing with such shortcomings and eliminates the need for thresholding. The results of our computer simulations and clinical validation study demonstrate the excellent performance of the proposed algorithm and suggest it may become a useful tool for objectively determining the location and extent of focal epileptic activity in a noninvasive fashion.