A hybrid EEG-EMG BMI improves the detection of movement intention in cortical stroke patients with complete hand paralysis

Motor rehabilitation based on brain-machine interfaces (BMI) has been shown as a feasible option for stroke patients with complete paralysis. However, the pathologic EEG activity after a stroke makes the detection of movement intentions in these patients challenging, especially in those with damages involving the motor cortex. Residual electromyographic activity in those patients has been shown to be decodable, even in cases when the movement is not possible. Hybrid BMIs combining EEG and EMG activity have been recently proposed, although there is little evidence about how they work for completely paralyzed stroke patients. In this study we propose a neural interface, relying on EEG, EMG or EEG+EMG features, to detect movement attempts. Twenty patients with a chronic stroke affecting their motor cortex were recruited, and asked to open and close their paralyzed hand while their electrophysiological signals were recorded. We show how EEG and EMG activities provide complementary information for detecting the movement intentions, being the accuracy of the hybrid BMI significantly higher than the EEG-based system. The obtained results encourage the integration of hybrid BMI systems for motor rehabilitation of patients with paralysis due to stroke.