論文ID: 2025.01109
Stroke remains a leading cause of mortality and long-term disability worldwide, frequently resulting in impairments in motor control, cognition, and emotional regulation. Conventional rehabilitation approaches, while partially effective, often lack individualization and yield suboptimal outcomes. In recent years, brain-computer interface (BCI) technology has emerged as a promising neurorehabilitation tool by decoding neural signals and providing real-time feedback to enhance neuroplasticity. This review systematically explores the use of BCI systems in post-stroke rehabilitation, focusing on three core domains: motor function, cognitive capacity, and emotional regulation. This review outlines the neurophysiological principles underpinning BCI-based motor rehabilitation, including neurofeedback training, Hebbian plasticity, and multimodal feedback strategies. It then examines recent advances in upper limb and gait recovery using BCI integrated with functional electrical stimulation (FES), robotics, and virtual reality (VR). Moreover, it highlights BCI's potential in cognitive and language rehabilitation through EEG-based neurofeedback and the integration of artificial intelligence (AI) and immersive VR environments. In addition, it discusses the role of BCI in monitoring and regulating post-stroke emotional disorders via closed-loop systems. While promising, BCI technologies face challenges related to signal accuracy, device portability, and clinical validation. Future research should prioritize multimodal integration, AI-driven personalization, and large-scale randomized trials to establish long-term efficacy. This review underscores BCI's transformative potential in delivering intelligent, personalized, and cross-domain rehabilitation solutions for stroke survivors.
