Gui Zhuolan, Liu Yuan, Qiu Shiyin, Zhang Yujian, Dong Kailun, Ming Dong
Annu Int Conf IEEE Eng Med Biol Soc. 2024 Jul;2024:1-4. doi: 10.1109/EMBC53108.2024.10782372.
Lower limb motor dysfunction is a prevalent complication of stroke that significantly impacts patients' quality of life. Current research indicates that motor imagery-based brain-computer interface (BCI-MI) training can assist stroke patients in enhancing motor function and reconstructing neural pathways. Nevertheless, 40% of stroke patients struggle with effective motor imagery (MI), leading to challenges in applying lower limb MI in clinical settings. Electrical stimulation (ES) has demonstrated the ability to induce muscle contractions, generating a kinesthetic illusion that effectively guides subjects in performing MI. However, the existing study lacks clarity regarding the effectiveness of the ES-MI paradigm in improving lower limb MI in stroke patients. To address this gap, we recruited seven stroke patients to participate in an experiment involving the ES-MI enhancement paradigm, aiming to validate its performance in stroke patients. The results revealed that the ES-MI paradigm augmented the activation of the motor cortex in the lower limb and reactivated dormant areas, suggesting that MI training based on the ES-MI paradigm holds promise for enhancing neural remodeling effects in stroke patients. Additionally, the paradigm enhanced the classification accuracy of SVM(+1.17%), KNN(+0.93%), RF(+7.13%), LDA(+5.29%), and EEGNet(+0.96%), indicating potential improvements in the efficiency and quality of human-robot interaction in brain-controlled lower limb rehabilitation robots.
下肢运动功能障碍是中风常见的并发症,严重影响患者的生活质量。目前的研究表明,基于运动想象的脑机接口(BCI-MI)训练可以帮助中风患者增强运动功能并重建神经通路。然而,40%的中风患者难以进行有效的运动想象(MI),这给在临床环境中应用下肢MI带来了挑战。电刺激(ES)已被证明能够诱发肌肉收缩,产生动觉错觉,有效地引导受试者进行MI。然而,现有研究对于ES-MI范式在改善中风患者下肢MI方面的有效性尚不清楚。为了填补这一空白,我们招募了7名中风患者参与一项涉及ES-MI增强范式的实验,旨在验证其在中风患者中的表现。结果显示,ES-MI范式增强了下肢运动皮层的激活,并重新激活了休眠区域,这表明基于ES-MI范式的MI训练有望增强中风患者的神经重塑效果。此外,该范式提高了支持向量机(SVM)(提高1.17%)、K近邻算法(KNN)(提高0.93%)、随机森林(RF)(提高7.13%)、线性判别分析(LDA)(提高5.29%)和EEGNet(提高0.96%)的分类准确率,表明在脑控下肢康复机器人中,人机交互的效率和质量可能得到改善。