Challenges and prospects of BCI-based robot systems for stroke rehabilitation

Stroke remains a leading cause of adult disability worldwide, with millions of survivors experiencing motor, sensory, and cognitive impairments that drastically affect their quality of life. Traditional rehabilitation approaches, which primarily depend on physical training and manual assistance, have several limitations, including prolonged treatment duration, low patient engagement, and limited enhancement of neural plasticity. To address these challenges, recent research has increasingly focused on integrating brain-computer interface (BCI) technology with robotic rehabilitation systems (Wolpaw et al., 2020) [1]. Among various BCI technologies, noninvasive electroencephalogram (EEG)-based BCI has emerged as a promising approach for motor function rehabilitation owing to its non-invasiveness, safety, and high accessibility. This technique records brain signals in real time using electrodes placed on the scalp (Wang, 2021) [2], analyzes motor intentions using artificial intelligence algorithms, and integrates exoskeletons, functional electrical stimulation or virtual reality devices to establish a “perception-feedback” closed-loop training system, accelerating the reconstruction and rehabilitation of the motor pathway (Stoyell et al., 2021) [3].