Impact of motor imagery-based brain-computer interface combined with virtual reality on enhancing attention, executive function, and lower-limb function in stroke: A pilot study.

Background: Brain-computer interface combined with virtual reality (BCI-VR) can reduce the difficulty of motor imagery execution and improve training performance. Few studies have focused on the effects of BCI-VR on attention, executive function, and lower-limb function in stroke.

Objective: To evaluate feasibility and preliminary efficacy of BCI-VR pedaling training on the attention, executive function, and lower-extremity function in people after stroke. It will also provide data support for future research, especially sample size calculations.

Design: A single group before-after trial design was used. All participants had a stable level of function over a 2-week period to ensure that their functional recovery was all attributable to BCI-VR training.

Setting: The study was conducted in a specialized rehabilitation hospital.

Participants: Twelve participants with stroke, a certain level of motor imagery ability, capable of walking 10 meters continuously.

Interventions: All participants received a 4-week BCI-VR pedaling training program, 5 days per week, 30 minutes each session.

Outcome measures: Primary outcomes are feasibility and safety. Secondary outcomes were lower-extremity mobility, attention, and executive functions.

Results: Twelve patients were recruited from inpatient rehabilitation and nine completed the study (six males/three females; 56.6 ± 11.6 years). Recruitment and retention rates were 34% and 75%, respectively. Excellent adherence rate (97.7%) was obtained. No adverse events or equipment issues were reported. Following the intervention, significant improvements were found in the lower-extremity strength, balance, walking stability, attention, and general cognitive function (p < .05). A significant correlation was found between improved Berg balance scale change values and symbol digit modalities test change values (p < .05, r = 0.677).

Conclusions: BCI-VR pedaling training provides a depth of feasibility and safety data, methodological detail, and preliminary results. This could provide a useful basis for future studies of BCI-VR pedaling training for stroke rehabilitation.

Clinicaltrials: gov registration number: ChiCTR2300071522 (http://www.chictr.org.cn).