Motor imagery combined with brain-computer interface for stroke patients: a meta-analysis

Objective To systematically evaluate the effects of motor imagery combined with brain-computer interface (MI-BCI) on stroke patients. Methods Randomized controlled trials (RCTs) on MI-BCI for stroke patients were retrieved from CNKI, Wanfang, VIP, CBM, PubMed, Cochrane Library, Embase, and Web of Science databases from inception to June 2025. Data were analyzed using RevMan 5.2 software. Results Eight RCTs involving 357 stroke patients were included. The meta-analysis showed that MI-BCI was associated with an improvement in upper limb motor function, although this did not reach conventional statistical significance (SMD = 0.86, 95% CI = −0.04 to 1.75, p = 0.06). In contrast, a statistically significant, moderate-to-large improvement was found in activities of daily living (SMD = 1.47, 95% CI = 0.51 to 2.44, p = 0.003). Subgroup analyses indicated that the efficacy in motor function was primarily evident when MI-BCI was administered as an adjunct to conventional rehabilitation or with an intervention duration of ≥4 weeks. Conclusion The efficacy of MI-BCI is contingent upon its therapeutic context. When used as an adjunct to conventional rehabilitation, MI-BCI can significantly improve both upper limb motor function and activities of daily living in stroke patients. However, current evidence does not support its superiority over motor imagery alone when applied as a standalone therapy. An intervention duration of ≥4 weeks is recommended to achieve significant functional gains.