BCI-controlled NMES in Subacute Stroke

Stroke patients with severe upper limb movement deficits have limited treatment options and often remain severely handicapped at the chronic stage. Recent findings have suggested that poor motor recovery can be due to severe damage of the cortico-spinal tract (CST), the neural fibres connecting the movement regions of the brain to the spinal cord. Hence, to improve recovery of upper limb movements it will be crucial to re-establish and strengthen CST projections. Recent studies provided evidence that closed-loop brain computer interface-driven electrical stimulation of the paretic muscles can induce clinically important and lasting recovery of upper limb function, even in patients with chronic, severe motor affection. In this treatment approach, movement intentions of the patients are detected with electroencephalography and real-time analyses. This triggers an electrical stimulation of affected upper limb muscles. In this study, the investigators hypothesize that neuromuscular electrical stimulation (NMES) applied contingent to voluntary activation of primary motor cortex, as detected by a brain-computer interface (BCI), can help restore CST projections. This might improve recovery of patients with severe upper limb movement deficits. Treatment will be started within the first 8 weeks after stroke onset.

Electrical stimulation of paretic upper limb is triggered contigent to voluntary motor cortex activation of the patient, as detected by the brain-computer interface.

Electrical stimulation of paretic upper limb is applied independently of motor cortex activation of the patient by using a prerecorded session of another patient.

From the recorded brain activity (EEG) subject specific patterns will be extracted with machine learning techniques from recordings where the subject executes movements tasks. Whenever a subject-specific pattern can be identified and detected, this is used for triggering neuromuscular electrical stimulation.

Neuromuscular electrical stimulation is triggered independently of the patient's movement intentions.

Change in fractional anisotropy (FA) of the cortico-spinal tract as determined from diffusion tensor imaging

Computed from high-density EEG recordings. Continuous measure. Higher values indicate better outcome.

Inclusion Criteria: Ischemic or hemorrhagic stroke Stroke onset ≤ 8 weeks Severe, unilateral motor upper extremity hemiparesis (≤15 Fugl-Meyer Score) Ability to give informed consent Exclusion Criteria: Second stroke during rehabilitation Skull breach Cardiac pacemaker Metallic implants in the brain Delirium or disturbed vigilance Inability to follow treatments sessions Severe language comprehension deficits Severe dystonia or spasticity Severe co-morbidity (ex, traumatic, rheumatologic, neurodegenerative diseases) Pregnancy