IR和SSVEP双模式融合脑-机接口研究  

Research of bimodal fusion brain-computer interface based on IR and SSVEP

作　　者：官金安[1] 廖婉戎 李东阁 周到[1] 高军峰[1] GUAN Jin'an;LIAO Wanrong;LI Dongge;ZHOU Dao;GAO Junfeng(Hubei Provincial Key Laboratory of Medical Information Analysis and Tumor Diagnosis and Treatment&Key Laboratory of Cognitive Science of State Ethnic Affairs Commission,South-Central Minzu University,Wuhan 430074,China;Tongji Hospital of Tongji Medical College,Huazhong University of Science and Technology,Wuhan 430022,China)

机构地区：[1]中南民族大学医学信息分析及肿瘤诊疗湖北省重点实验室&认知科学国家民委重点实验室,武汉430074 [2]华中科技大学同济医学院附属协和医院,武汉430022

出　　处：《中南民族大学学报（自然科学版）》2022年第5期558-564,共7页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：国家自然科学基金资助项目(61773408);中央高校基本科研业务费专项资金资助项目(CZZ19004,CZY20039)。

摘　　要：多模式脑-机接口可以完成更复杂的识别任务,也可以提高系统性能.为了进一步提高模拟阅读(Imitating-reading,IR)脑-机接口的信息传输率,将稳态视觉诱发电位(Steady-state Visual Evoked Potential,SSVEP)融入IR范式,提出了一种新型的双模式脑-机接口字符输入系统.该系统对IR和SSVEP两种范式采用并行控制的方式,通过识别同步诱发的事件相关电位和SSVEP特征分别判断目标字符所在的列和行,进而完成字符输入.对9名受试者进行离线实验,获得的平均正确率为81.5%,最高可达99%;最高信息传输率为95.2 bits/min,比IR范式提高了接近1倍,单个字符输入的时间明显减少.实验结果表明:所提出的IR和SSVEP双模式融合脑-机接口是可行的,且在提高脑-机接口系统性能方面具有较好潜力.The multi-modal brain-computer interface(BCI)can complete more complicated tasks and improve the system performance.To further improve the information transfer rate of imitating-reading(IR)BCI system,a novel bimodal BCI character input system is proposed by integrating the steady-state visual evoked potential(SSVEP)into the IR paradigm.The system uses parallel control for IR and SSVEP paradigms,and determines the column and row of target character by identifying the event related potentials and SSVEP characteristics.The off-line experiment results obtained from nine participants indicated that the average accuracy was 81.5%and the highest was 99%.Importantly,the highest information transfer rate(ITR)was 95.2 bits/min,which was about one time higher than that of IR paradigm and achieved the higher speed of character input using the bimodal BCI.The experimental results indicate that the proposed bimodal fusion BCI based on IR and SSVEP is feasible and it is promising for improving the BCI system performance.

关 键 词：脑-机接口 多模式融合 模拟阅读 稳态视觉诱发电位 字符输入 

分 类 号：R318.6[医药卫生—生物医学工程]