基于个性化导联选择策略的脑机交互康复训练临床初步研究

Individualized channel-selection strategy-based brain-machine interaction rehabilitation training:A pilot study of clinical experiments

脑机交互(BMI)技术可识别人体运动意图,并控制机器人辅助卒中患者患肢动作,有效刺激肢体运动及感觉反馈神经环路。卒中患者的损伤类型、损伤程度和损伤部位极具个性化,大脑皮层中运动意图的表达可能会出现不同程度的迁移现象。该文提出一种基于个性化导联选择策略的脑机交互康复训练方法,将患者运动意图表达脑区作为BMI系统信号采集点。该研究共纳入3位卒中患者,分别完成10 d的康复训练,采用基于个性化导联选择策略的BMI康复训练方法,所有患者分别在训练前、后进行临床量表评价和脑电信号(EEG)评估。结果显示:受试者在训练后Fugl-Meyer量表评分提升,其中1位出现运动诱发电位,表征神经通路激活。该基于个性化导联选择策略的脑机交互康复训练方法应在大量患者样本中进一步验证。

Brain-machine interactions(BMI)can recognize motor intentions and transfer the intentions to control a rehabilitation robot assisting paralyzed limb movement,which can effectively stimulate the neural pathways for motor control and sensory feedback.Lesion types,lesion degrees and lesion sites vary from patient to patient.Motor intention expression may then shift to different brain regions depending on the lesion characteristics.This study presents an individualized channel-selection strategy-based BMI rehabilitation training method which uses the individualized motor expression encephalic area as the electroencephalogram(EEG)collection site.Three stroke patients were recruited for this study and were trained for 10 days using individualized channel-selection strategy-based BMI rehabilitation training.All the patients were assessed using a clinical evaluation scale and EEG tests before and after the 10 day training.The results show that after the training the subjects have improved Fugl-Meyer assessments and one of the subjects regains motor evoked potential which indicates the activation of the neural pathway.This individualized channel-selection strategy-based BMI rehabilitation training method should be further validated with a large number of stroke patients in the future.