BCI-assisted MI Intervention in Subacute Stroke

Stroke is a major public health and social care concern worldwide, being the leading cause of long-term disability in adults. The upper limb motor impairment commonly persists after stroke affecting patients' everyday life functional independence. Despite the intensive rehabilitation, the variability in the nature and the extent of upper limb recovery remains a crucial factor effecting rehabilitation outcomes. Electroencephalography (EEG) -based Brain Computer Interface (BCI) technology is a potential tool to promote functional motor recovery of upper limbs after stroke as shown in several randomized controlled trials. The investigators' multidisciplinary team was successful in designing, implementing and clinically validating a sensorimotor rhythm-based BCI combined with realistic visual feedback of upper limb to support hand motor imagery (MI) practice in stroke participants. However, important questions remain to be addressed to translate BCI in clinical practice such as defining whether the expected BCI-induced early improvements in functional motor outcomes can be sustained in a long-term after stroke. This requires advancements in the knowledge on brain functional re-organization after stroke and how this re-organization would correlate with the functional motor outcome (evidence-base medicine). Last but not least, the definition of the determinants of the patient response to-treatment is paramount to optimize the process of personalized medicine in rehabilitation. The fundamental of this project stems from the investigators' previous findings on the efficacy of BCI-assisted MI training in subacute stroke participants. These promising findings corroborated the idea that a relatively low-cost technique (i.e. EEG-based BCI) can be exploited to deliver a rehabilitative intervention (in this case MI) and prompted the research team to undertake a further translational effort by implementing an all-in-one BCI-supported MI training station- the Promotoer. In this project, the investigators will provide evidence for a persistency (up to 6 months) of the significant early improvement of hand motor function induced by the BCI-assisted MI training operated via the Promotoer. Task-specific training was recently reported to induce long-term improvements (6 months follow-up) in arm motor function after stroke. Thus, the hypothesis is that the BCI-based rewarding of hand MI tasks would promote long-lasting retention of early induced positive effect on motor performance with respect to MI task practiced in an open loop condition (i.e., without BCI). Further hypothesis is that such clinical improvement would be sustained by a long-lasting neuroplasticity changes that would be harnessed by the BCI -based intervention. This hypothesis rises from current evidence for an early enhancement of post-stroke plastic changes enabled by BCI- based trainings. To test this hypothesis, a longitudinal assessment of the brain network organization derived from advanced EEG signal processing will be performed. The heterogeneity of stroke makes prediction of treatment responder a great challenge. The investigators hypothesize that the longitudinal functional and neurophysiology assessment over 6 months from the intervention will allow for insights in biomarkers and potential predictors of stroke participants' response to the Promotoer training. Some of the well-recognized factors contributing to functional motor recovery after stroke such as the relation between lesion characteristics and patterns of post-stroke motor cortical re-organization (e.g., ipsilesional/contralesional primary and non-primary motor areas; cortico-spinal tract integrity, severity of motor deficits at baseline) will be taken into account.

Inclusion criteria

• first ever unilateral stroke - confirmed by MRI;
• hemiplegia/hemiparesis from 1 to 6 months since stroke;
• age between 18 and 80 years;

Exclusion criteria

• severe neglect and aphasia;
• dementia;
• severe spasticity - Modified Ashworth Scale >4 at shoulder/elbow/wrist;
• Upper Extremity Fugl-Meyer Assessment (UE-FMA) >47/60 score (60 is without considering 6 score point for tendon reflexes);
• Token test >29 score;
• concomitant neurological disorders