Improving motor imagery classification during induced motor perturbations.

Objective.Motor imagery is the mental simulation of movements. It is a common paradigm to design brain-computer interfaces (BCIs) that elicits the modulation of brain oscillatory activity similar to real, passive and induced movements. In this study, we used peripheral stimulation to provoke movements of one limb during the performance of motor imagery tasks. Unlike other works, in which induced movements are used to support the BCI operation, our goal was to test and improve the robustness of motor imagery based BCI systems to perturbations caused by artificially generated movements.Approach.We performed a BCI session with ten participants who carried out motor imagery of three limbs. In some of the trials, one of the arms was moved by neuromuscular stimulation. We analysed 2-class motor imagery classifications with and without movement perturbations. We investigated the performance decrease produced by these disturbances and designed different computational strategies to attenuate the observed classification accuracy drop.Main results.When the movement was induced in a limb not coincident with the motor imagery classes, extracting oscillatory sources of the movement imagination tasks resulted in BCI performance being similar to the control (undisturbed) condition; when the movement was induced in a limb also involved in the motor imagery tasks, the performance drop was significantly alleviated by spatially filtering out the neural noise caused by the stimulation. We also show that the loss of BCI accuracy was accompanied by weaker power of the sensorimotor rhythm. Importantly, this residual power could be used to predict whether a BCI user will perform with sufficient accuracy under the movement disturbances.Significance.We provide methods to ameliorate and even eliminate motor related afferent disturbances during the performance of motor imagery tasks. This can help improving the reliability of current motor imagery based BCI systems.

Enhancing sensorimotor BCI performance with assistive afferent activity: An online evaluation.

An important goal in Brain-Computer Interfacing (BCI) is to find and enhance procedural strategies for users for whom BCI control is not sufficiently accurate. To address this challenge, we conducted offline analyses and online experiments to test whether the classification of different types of motor imagery could be improved when the training of the classifier was performed on the data obtained with the assistive muscular stimulation below the motor threshold. 10 healthy participants underwent three different types of experimental conditions: a) Motor imagery (MI) of hands and feet b) sensory threshold neuromuscular electrical stimulation (STM) of hands and feet while resting and c) sensory threshold neuromuscular electrical stimulation during performance of motor imagery (BOTH). Also, another group of 10 participants underwent conditions a) and c). Then, online experiments with 15 users were performed. These subjects received neurofeedback during MI using classifiers calibrated either on MI or BOTH data recorded in the same experiment. Offline analyses showed that decoding MI alone using a classifier based on BOTH resulted in a better BCI accuracy compared to using a classifier based on MI alone. Online experiments confirmed accuracy improvement of MI alone being decoded with the classifier trained on BOTH data. In addition, we observed that the performance in MI condition could be predicted on the basis of a more pronounced connectivity within sensorimotor areas in the frequency bands providing the best performance in BOTH. These finding might offer a new avenue for training SMR-based BCI systems particularly for users having difficulties to achieve efficient BCI control. It might also be an alternative strategy for users who cannot perform real movements but still have remaining afferent pathways (e.g., ALS and stroke patients).

Brain-machine interfaces for rehabilitation in stroke: A review.

BACKGROUND: Motor paralysis after stroke has devastating consequences for the patients, families and caregivers. Although therapies have improved in the recent years, traditional rehabilitation still fails in patients with severe paralysis. Brain-machine interfaces (BMI) have emerged as a promising tool to guide motor rehabilitation interventions as they can be applied to patients with no residual movement. OBJECTIVE: This paper reviews the efficiency of BMI technologies to facilitate neuroplasticity and motor recovery after stroke. METHODS: We provide an overview of the existing rehabilitation therapies for stroke, the rationale behind the use of BMIs for motor rehabilitation, the current state of the art and the results achieved so far with BMI-based interventions, as well as the future perspectives of neural-machine interfaces. RESULTS: Since the first pilot study by Buch and colleagues in 2008, several controlled clinical studies have been conducted, demonstrating the efficacy of BMIs to facilitate functional recovery in completely paralyzed stroke patients with noninvasive technologies such as the electroencephalogram (EEG). CONCLUSIONS: Despite encouraging results, motor rehabilitation based on BMIs is still in a preliminary stage, and further improvements are required to boost its efficacy. Invasive and hybrid approaches are promising and might set the stage for the next generation of stroke rehabilitation therapies.

Brain-computer interfaces in the completely locked-in state and chronic stroke.

Brain-computer interfaces (BCIs) use brain activity to control external devices, facilitating paralyzed patients to interact with the environment. In this chapter, we discuss the historical perspective of development of BCIs and the current advances of noninvasive BCIs for communication in patients with amyotrophic lateral sclerosis and for restoration of motor impairment after severe stroke. Distinct techniques have been explored to control a BCI in patient population especially electroencephalography (EEG) and more recently near-infrared spectroscopy (NIRS) because of their noninvasive nature and low cost. Previous studies demonstrated successful communication of patients with locked-in state (LIS) using EEG- and invasive electrocorticography-BCI and intracortical recordings when patients still showed residual eye control, but not with patients with complete LIS (ie, complete paralysis). Recently, a NIRS-BCI and classical conditioning procedure was introduced, allowing communication in patients in the complete locked-in state (CLIS). In severe chronic stroke without residual hand function first results indicate a possible superior motor rehabilitation to available treatment using BCI training. Here we present an overview of the available studies and recent results, which open new doors for communication, in the completely paralyzed and rehabilitation in severely affected stroke patients. We also reflect on and describe possible neuronal and learning mechanisms responsible for BCI control and perspective for future BMI research for communication in CLIS and stroke motor recovery.

Decoding of motor intentions from epidural ECoG recordings in severely paralyzed chronic stroke patients.

OBJECTIVE: Recently, there have been several approaches to utilize a brain-computer interface (BCI) for rehabilitation with stroke patients or as an assistive device for the paralyzed. In this study we investigated whether up to seven different hand movement intentions can be decoded from epidural electrocorticography (ECoG) in chronic stroke patients. APPROACH: In a screening session we recorded epidural ECoG data over the ipsilesional motor cortex from four chronic stroke patients who had no residual hand movement. Data was analyzed offline using a support vector machine (SVM) to decode different movement intentions. MAIN RESULTS: We showed that up to seven hand movement intentions can be decoded with an average accuracy of 61% (chance level 15.6%). When reducing the number of classes, average accuracies up to 88% can be achieved for decoding three different movement intentions. SIGNIFICANCE: The findings suggest that ipsilesional epidural ECoG can be used as a viable control signal for BCI-driven neuroprosthesis. Although patients showed no sign of residual hand movement, brain activity at the ipsilesional motor cortex still shows enough intention-related activity to decode different movement intentions with sufficient accuracy.

Movement related cortical potentials in severe chronic stroke.

Movement related cortical potentials (MRCPs) have been studied for many years and controlled using brain computer interfaces (BCIs). Furthermore, MRCPs have been proposed as reliable and immediate indicators of cortical reorganizations in motor learning and after stroke. In this study MRCPs preceding and during hand movements in severe chronic stroke were investigated. Eight severely impaired (no residual finger extension) chronic stoke patients underwent EEG and EMG recordings during a cue triggered hand movement paradigm. Four patients presented subcortical lesions only while the other four presented mixed (cortical and subcortical) lesions. MRCPs were measured before (slow cortical potentials SCPs) and at movement onset (motor potentials MPs). SCPs were observed during paretic hand movements only. Latencies were longer and reached their negativity peak earlier during paretic hand movement. When dividing the patients in subcortical only and mixed lesion patients, we observed significantly bigger MP peak amplitudes over the lesioned hemisphere during paretic and healthy hand movements in subcortical stroke patients. Furthermore, we observed a significant difference in MP peak latency between subcortical and mixed stroke patients during paretic hand movements. We demonstrated for the first time significant differences between subcortical only and mixed (cortical and subcortical) stroke patients' MRCPs during motor preparation and execution. Furthermore, we demonstrated how stroke produces a longer MRCP and that lesion location affects MP peak amplitude and latency. Finally, we propose the use MRCP based BCIs to reduce their duration (towards normal) and induce motor function recovery.

Neuromuscular electrical stimulation induced brain patterns to decode motor imagery.

OBJECTIVE: Regardless of the paradigm used to implement a brain-computer interface (BCI), all systems suffer from BCI-inefficiency. In the case of patients the inefficiency can be high. Some solutions have been proposed to overcome this problem, however they have not been completely successful yet. METHODS: EEG from 10 healthy users was recorded during neuromuscular electrical stimulation (NMES) of hands and feet and during motor imagery (MI) of the same limbs. Features and classifiers were computed using part of these data to decode MI. RESULTS: Offline analyses showed that it was possible to decode MI using a classifier based on afferent patterns induced by NMES and even infer a better model than with MI data. CONCLUSION: Afferent NMES motor patterns can support the calibration of BCI systems and be used to decode MI. SIGNIFICANCE: This finding might be a new way to train sensorimotor rhythm (SMR) based BCI systems for healthy users having difficulties to attain BCI control. It might also be an alternative to train MI-based BCIs for users who cannot perform real movements but have remaining afferents (ALS, stroke patients).