CASIA OpenIR  > 毕业生  > 博士学位论文
基于功能磁共振和脑电信号融合的情绪图片加工研究
其他题名Affective Picture Processing Research Based on Signal Integration of fMRI and EEG
胡瑾
学位类型工学博士
导师田捷
2007-06-19
学位授予单位中国科学院研究生院
学位授予地点中国科学院自动化研究所
学位专业计算机应用技术
关键词情绪图片加工 功能磁共振成像 脑电图 事件相关电位 多模态融合 Affective Picture Processing Funcional Magnetic Resonance Imaging (Fmri) Electroencephalogram (Eeg) Event-related Potential (Erp) Multimodality Integration
摘要本文的研究重点是对功能磁共振数据和脑电数据进行融合与分析,探讨大脑的情绪图片加工的反应机理。具体描述如下: (1)基于功能磁共振成像与脑电时空激活模式的情绪加工对比研究:我们对同一批被试采用相同实验设计进行fMRI和EEG实验。对比fMRI激活区与EEG时空分析结果,发现二者在很大程度上保持一致,也在一些激活区表现为不一致,但是这并不意味着结果的矛盾。fMRI和EEG可以相互印证,相互补充,从而有助于更好地揭示大脑活动的反应机制。(2)提出一种基于功能磁共振与脑电信号相关分析来研究情绪影响注意的神经机制的方法。fMRI和EEG实验分别进行,采用了同一批被试和相同的实验任务。数据处理采用提取其BOLD响应变化量与ERP的P2和N2波幅呈相关的fMRI激活区的方法。结果从情绪调节注意的两条通路来看,P2阶段主要体现了自下而上的皮层下通路,而N2阶段主要体现了自上而下的皮层通路。(3)提出了一种基于数据时空重构的双重融合方法及框架,包括两步:首先,利用联合ICA方法对fMRI的空间信息和EEG的时间信息进行初步融合,提取成分(ERP时间序列及与之对应的fMRI激活区),并进行数据重构,获得一个动态的fMRI激活模式。然后,在BEM真实头模型中,以第一步求得的fMRI激活区为先验信息,用已经通过仿真实验验证了有效性的FC-DECD方法进行二次融合。最后把本方法应用到情绪加工实验的数据分析中,结果与已有报导一致。
其他摘要The emphasis of this study is the brain mechanism during affective picture processing based on the signal integration of fMRI and EEG. The main work of this study is as follows: (1)To investigate emotion processing based on the comparison between fMRI imaging and EEG spatiotemporal activation patterns. fMRI and EEG experiments were conducted with the same subjects and the same task. The results of these two modalities were compared. The two results have showed great consistency. Although differed in some activation, this didn’t mean the conflict between the two results. fMRI and EEG can support and complement each other to help reveal the brain activation mechanisms comprehensively.(2)A Method based on the correlation analysis of fMRI and EEG signals to investigate the neural mechanisms in which emotion influences attention was proposed. In our study, two modalities were integrated. fMRI and EEG experiments were conducted with the same subjects and the same task. A set of voxels which Strongly correlated BOLD response with the selected ERP peaks (P2 and N2) were extracted. The results showed that according to the two routes with which emotion modulates attention, P2 phase were mostly related to the bottom-up subcortical route, while N2 phase to the top-down cortical route.(3)A dual-integration method based on data spatiotemporal reconstruction was proposed, which included two steps: first, integrated the spatial information of fMRI and the temporal information of EEG by using joint-ICA to find the components (ERP time course and the corresponding fMRI activation), then a fMRI spatiotemporal activation patterns was obtained by spatiotemporal reconstruction. Secondly, with BEM real head model, using the fMRI activation obtained by the first step as priori, implemented the re-integration with FC-DECD method, which has been validated with the simulation experiment. Finally, the present method was applied to the human experiment data with emotion processing, and the results were in accordance with those reported in the previous studies.
馆藏号XWLW1129
其他标识符200418014628056
语种中文
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/6026
专题毕业生_博士学位论文
推荐引用方式
GB/T 7714
胡瑾. 基于功能磁共振和脑电信号融合的情绪图片加工研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2007.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
CASIA_20041801462805(1846KB) 暂不开放CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[胡瑾]的文章
百度学术
百度学术中相似的文章
[胡瑾]的文章
必应学术
必应学术中相似的文章
[胡瑾]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。