A 1D CNN for high accuracy classification and transfer learning in motor imagery EEG-based brain-computer interface.

Objective.Brain-computer interface (BCI) aims to establish communication paths between the brain processes and external devices. Different methods have been used to extract human intentions from electroencephalography (EEG) recordings. Those based on motor imagery (MI) seem to have a great potential for future applications. These approaches rely on the extraction of EEG distinctive patterns during imagined movements. Techniques able to extract patterns from raw signals represent an important target for BCI as they do not need labor-intensive data pre-processing.Approach.We propose a new approach based on a 10-layer one-dimensional convolution neural network (1D-CNN) to classify five brain states (four MI classes plus a 'baseline' class) using a data augmentation algorithm and a limited number of EEG channels. In addition, we present a transfer learning method used to extract critical features from the EEG group dataset and then to customize the model to the single individual by training its late layers with only 12-min individual-related data.Main results.The model tested with the 'EEG Motor Movement/Imagery Dataset' outperforms the current state-of-the-art models by achieving a99.38%accuracy at the group level. In addition, the transfer learning approach we present achieves an average accuracy of99.46%.Significance.The proposed methods could foster the development of future BCI applications relying on few-channel portable recording devices and individual-based training.

Emotional processing in RRMS patients: Dissociation between behavioural and neurophysiological response.

BACKGROUNG: Multiple sclerosis (MS) results in a broad range of symptoms, including motor, visual, cognitive, and neuropsychiatric deficits. Some studies, considering affective facial expressions to study emotion processing, demonstrated emotion recognition difficulties in MS patients. OBJECTIVE: We investigated the impact of MS on the emotional-behaviour rating and neurophysiological response (Event Related Potentials-ERP) through a battery of affective visual stimuli selected from the International Affective Picture System (IAPS). METHODS: Twenty patients with diagnosis of Relapsing Remitting MS (RRMS) and 20 Healthy Controls (HC) matched by age, gender and education were enrolled. Each of them, after a neuropsychological assessment, were asked to evaluate arousal and valence of affective visual stimuli. RESULTS: Our results showed higher P300 amplitudes in RRMS patients than HC group for pleasant and unpleasant images. Moreover, RRMS patients showed lower Reaction Time (RT) respect HC in valence rating. No other effect did emerge between groups. CONCLUSION: Our study shows early compensatory cerebral mechanisms in RRMS patients throughout emotional information processing, particularly for unpleasant and pleasant stimuli. We hypothesize that this compensatory cerebral mechanism reduces the behavioural dissimilarity between patients and HC.

Global signal modulation of single-trial fMRI response variability: Effect on positive vs negative BOLD response relationship.

In functional magnetic resonance imaging (fMRI), the relationship between positive BOLD responses (PBRs) and negative BOLD responses (NBRs) to stimulation is potentially informative about the balance of excitatory and inhibitory brain responses in sensory cortex. In this study, we performed three separate experiments delivering visual, motor or somatosensory stimulation unilaterally, to one side of the sensory field, to induce PBR and NBR in opposite brain hemispheres. We then assessed the relationship between the evoked amplitudes of contralateral PBR and ipsilateral NBR at the level of both single-trial and average responses. We measure single-trial PBR and NBR peak amplitudes from individual time-courses, and show that they were positively correlated in all experiments. In contrast, in the average response across trials the absolute magnitudes of both PBR and NBR increased with increasing stimulus intensity, resulting in a negative correlation between mean response amplitudes. Subsequent analysis showed that the amplitude of single-trial PBR was positively correlated with the BOLD response across all grey-matter voxels and was not specifically related to the ipsilateral sensory cortical response. We demonstrate that the global component of this single-trial response modulation could be fully explained by voxel-wise vascular reactivity, the BOLD signal standard deviation measured in a separate resting-state scan (resting state fluctuation amplitude, RSFA). However, bilateral positive correlation between PBR and NBR regions remained. We further report that modulations in the global brain fMRI signal cannot fully account for this positive PBR-NBR coupling and conclude that the local sensory network response reflects a combination of superimposed vascular and neuronal signals. More detailed quantification of physiological and noise contributions to the BOLD signal is required to fully understand the trial-by-trial PBR and NBR relationship compared with that of average responses.

Functional and structural balances of homologous sensorimotor regions in multiple sclerosis fatigue.

Fatigue in multiple sclerosis (MS) is a highly disabling symptom. Among the central mechanisms behind it, an involvement of sensorimotor networks is clearly evident from structural and functional studies. We aimed at assessing whether functional/structural balances of homologous sensorimotor regions-known to be crucial for sensorimotor networks effectiveness-decrease with MS fatigue increase. Functional connectivity measures at rest and during a simple motor task (weak handgrip of either the right or left hand) were derived from primary sensorimotor areas electroencephalographic recordings in 27 mildly disabled MS patients. Structural MRI-derived inter-hemispheric asymmetries included the cortical thickness of Rolandic regions and the volume of thalami. Fatigue symptoms increased together with the functional inter-hemispheric imbalance of sensorimotor homologous areas activities at rest and during movement, in absence of any appreciable parenchymal asymmetries. This finding supports the development of compensative interventions that may revert these neuronal activity imbalances to relieve fatigue in MS.

Movement-induced uncoupling of primary sensory and motor areas in focal task-specific hand dystonia.

INTRODUCTION: Due to growing evidence of sensorimotor integration impairment in focal task-specific hand dystonia, we aimed at describing primary sensory (S1) and primary motor (M1) cortex source activities and their functional cross-talk during a non-dystonia-inducing sensorimotor task free of biases generated by the interfering with the occurrence of dystonic movements. METHOD: Magnetoencephalographic brain signals and opponens pollicis (OP) electromyographic activities were acquired at rest and during a simple isometric contraction performed either alone or in combination with median nerve stimulation. The task was performed separately with the right and left hand by eight patients suffering from focal task-specific hand dystonia and by eight healthy volunteers. Through an ad hoc procedure Functional Source Separation (FSS), distinct sources were identified in S1 (FSS1) and M1 (FSM1) devoted to hand control. Spectral properties and functional coupling (coherence) between the two sources were assessed in alpha [8,13]Hz, beta [14,32]Hz and gamma [33,45]Hz frequency bands. RESULTS: No differences were found between spectral properties of patients and controls for either FSM1 or FSS1 cerebral sources. Functional coupling between FSM1 and FSS1 (gamma band coherence), while comparable between dystonic patients and healthy controls at rest, was selectively reduced in patients during movement. All findings were present in both hemispheres. DISCUSSION: Because previous literature has shown that gamma-band sensory-motor synchronization reflects an efficiency index of sensory-motor integration, our data demonstrate that, in dystonic patients, uncoupling replaces the functional coupling required for efficient sensory-motor control during motor exertion. The presence of bi-hemispheric abnormalities in unilateral hand dystonia supports the presence of an endophenotypic trait.

Cortico-muscular coherence as an index of fatigue in multiple sclerosis.

BACKGROUND: Highly common in multiple sclerosis (MS), fatigue severely impacts patients' daily lives. Previous findings of altered connectivity patterns led to the hypothesis that the distortion of functional connections within the brain-muscle circuit plays a crucial pathogenic role. OBJECTIVE: The objective of this paper is to identify markers sensitive to fatigue in multiple sclerosis. METHODS: Structural (magnetic resonance imaging with assessment of thalamic volume and cortical thickness of the primary sensorimotor areas) and functional (cortico-muscular coherence (CMC) from simultaneous electroencephalo- and surface electromyographic recordings during a weak handgrip task) measures were used on 20 mildly disabled MS patients (relapsing-remitting course, Expanded Disability Status Scale score ≤ 2) who were recruited in two fatigue-dependent groups according to the Modified Fatigue Index Scale (MFIS) score. RESULTS: The two groups were similar in terms of demographic, clinical and imaging features, as well as task execution accuracy and weariness. In the absence of any fatigue-dependent brain and muscular oscillatory activity alterations, CMC worked at higher frequencies as fatigue increased, explaining 67% of MFIS variance (p=.002). CONCLUSION: Brain-muscle functional connectivity emerged as a sensitive marker of phenomena related to the origin of MS fatigue, impacting central-peripheral communication well before the appearance of any impairment in the communicating nodes.

A neurally-interfaced hand prosthesis tuned inter-hemispheric communication.

PURPOSE: This work investigates how a direct bidirectional connection between brain and hand prosthesis modifies the bi-hemispheric sensorimotor system devoted to the movement control of the lost limb. Hand prostheses are often unable to satisfy users' expectations, mostly due to the poor performance of their interfacing system. Neural Interfaces implanted inside nerves of the stump offer the advantage of using the bidirectional neural pathways 'naturally' dispatching signals to control proper hand actions and feed-back sensations. Learning to control a neurally-interfaced hand prosthesis and decode sensory information was previously observed to reduce the inter-hemispheric asymmetry of cortical motor maps and the clinical symptoms of phantom limb syndrome. METHODS: Electroencephalographic (EEG) data was analysed using Functional Source Separation (FSS), a semi-blind method that incorporates prior knowledge about the signal of interest into data decomposition to give access to cortical patch activities. RESULTS: Bi-hemispheric cortices showed normalization of their activity (topographical and spectral patterns) and of functional connectivity between homologous hand controlling areas, during the delivery of the motor command to the cybernetic prosthesis. CONCLUSIONS: The re-establishment of central-peripheral communication with the lost limb induced by a neurally-interfaced hand prosthesis produces beneficial plastic reorganization, not only restructuring contralateral directly-connected control areas, but also their functional balance within the bi-hemispheric system necessary for motor control.

Sensorimotor integration in focal task-specific hand dystonia: a magnetoencephalographic assessment.

To obtain a direct sensorimotor integration assessment in primary hand cortical areas (M1) of patients suffering from focal task-specific hand dystonia, magnetoencephalographic (MEG) and opponens pollicis electromyographic (EMG) activities were acquired during a motor task expressly chosen not to induce dystonic movements in our patients, to disentangle abnormalities indicating a possible substrate on which dystonia develops. A simple isometric contraction was performed either alone or in combination with median nerve stimulation, i.e. when a non-physiological sensory inflow was overlapping with the physiological feedback. As control condition, median nerve stimulation was also performed at rest. The task was performed bilaterally both in eight patients and in 16 healthy volunteers. In comparison with results in controls we found that in dystonic patients: i) MEG-EMG coherence was higher; ii) it reduced much less during galvanic stimulation in the hemisphere contralateral to the dystonic arm, simultaneously with iii) stronger inhibition of the sensory areas responsiveness due to movement; iv) the cortical component including contributions from sensory inhibitory and motor structures was reduced and v) much more inhibited during movement. It is documented that a simultaneous cortico-muscular coherence increase occurs in presence of a reduced M1 responsiveness to the inflow from the sensory regions. This could indicate an unbalance of the fronto-parietal functional impact on M1, with a weakening of the parietal components. Concurrently, signs of a less differentiated sensory hand representation--possibly due to impaired inhibitory mechanisms efficiency--and signs of a reduced repertoire of voluntary motor control strategies were found.

Sensory-motor interaction in primary hand cortical areas: a magnetoencephalography assessment.

Movement control requires continuous and reciprocal exchange of information between activities of motor areas involved in the task program execution and those elaborating proprioceptive sensory information. Our aim was to investigate the sensorimotor interactions in the region dedicated to hand control in healthy humans, focusing onto primary sensory and motor cortices, by selecting the time window at very early latencies. Through magnetoencephalographic recordings, we obtained a simultaneous assessment of sensory cortex activity modulation due to movement and of motor cortex activity modulation due to sensory stimulation, by eliciting a galvanic stimulation to the nerve (the median nerve) innervating a muscle (the opponens pollicis), at rest or during voluntary contraction. The primary sensory and motor cortices activities were investigated respectively through excitability in response to sensory stimulation and the cortico-muscular coherence. The task was performed bilaterally. A clear reduction of the cortico-muscular coherence was found in the short time window following stimuli (between around 150-450 ms). In the same time period, the motor control of isometric contraction was preserved. This could suggest that cortical component of voluntary movement control was transiently mediated by neuronal firing rate tuning more than by cortico-muscular synchronization. In addition to the known primary sensory cortex inhibition due to movement, a more evident reduction was found for the component known to include a contribution from primary motor areas. Gating effects were lower in the dominant left hemisphere, suggesting that sensorimotor areas dominant for hand control benefit of narrowing down gating effects.