Efficacy and safety of brain–computer interface for stroke rehabilitation: an overview of systematic review
Jiajun Liu, Yiwei Li, Dongjie Zhao, Lirong Zhong, Yan Wang, Man Hao, Jianxiong Ma
Abstract
Background: Stroke is a major global health challenge that significantly influences public health. In stroke rehabilitation, brain-computer interfaces (BCI) offer distinct advantages over traditional training programs, including improved motor recovery and greater neuroplasticity. Here, we provide a first re-evaluation of systematic reviews and meta-analyses to further explore the safety and clinical efficacy of BCI in stroke rehabilitation. Methods: A standardized search was conducted in major databases up to October 2024. We assessed the quality of the literature based on the following aspects: AMSTAR-2, PRISMA, publication year, study design, homogeneity, and publication bias. The data were subsequently visualized as radar plots, enabling a comprehensive and rigorous evaluation of the literature. Results: We initially identified 908 articles and, after removing duplicates, we screened titles and abstracts of 407 articles. A total of 18 studies satisfied inclusion criteria were included. The re-evaluation showed that the quality of systematic reviews and meta-analyses concerning stroke BCI training is moderate, which can provide relatively good evidence. Conclusion: It has been proven that BCI-combined treatment can improve upper limb motor function and the quality of daily life for stroke patients, especially those in the subacute phase, demonstrating good safety. However, its effects on improving speech function, lower limb motor function, and long-term outcomes require further evidence. Multicenter, long-term follow-up studies are needed to increase the reliability of the results. Clinical Trial Registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD42024562114, CRD42023407720.