摘要
目的提出一种新的基于独立成分分析法进行动态脑功能网络分析的方法,并应用该方法探讨精神分裂症患者在动态全脑功能网络上的变异。方法首先基于滑动时间窗方法计算正常被试和精神分裂症患者的动态全脑功能网络,然后使用组信息指导独立成分分析方法,提取每个被试的动态全脑功能网络的功能连接状态及相应的时间波动,比较正常被试和精神分裂症患者在功能连接状态上的差异。结果两组的最重要功能连接状态的模式有相似性。正常被试在额叶、顶叶相关区域较精神分裂症患者具有更强的正功能连接;在小脑相关区域精神分裂症患者呈现出更多的正功能连接,而正常被试呈现出更多的负功能连接。结论组信息指导独立成分分析方法可有效提取动态脑功能网络的功能连接状态,可揭示精神分裂症患者在动态脑功能网络的变异。
Objective To propose a novel independent component analysis（ICA）based method for analyzing dynamic functional network,and to investigate the aberrance of schizophrenia patients in dynamic whole-brain functional network.Method Firstly,dynamic whole-brain functional network was extracted using a sliding time window method for healthy controls and schizophrenia patients.Secondly,group information guided ICA（GIG-ICA）was used to extract functional connectivity states and the related fluctuations from dynamic whole-brain functional network for each subject.Finally,the difference between healthy controls and schizophrenia patients in functional connectivity states was compared.Results The dominant functional connectivity states from the two groups were similar.However,healthy controls had stronger positive connectivity in frontal and parietal lobes than schizophrenia patients.In terms of cerebellum cortex,schizophrenia patients presented more positive connectivity,while healthy controls presented more negative connectivity.Conclusion GIG-ICA is able to extract the functional connectivity states from the dynamic brain functional network efficiently,which can reveal the aberrance of schizophrenia patients.
出处
《中国医学影像技术》
CSCD
北大核心
2015年第6期926-931,共6页
Chinese Journal of Medical Imaging Technology
基金
国家自然科学基金(81471367)
中国科学院百人计划基金
关键词
磁共振成像
动态脑功能网络
独立成分分析
功能连接状态
精神分裂症
Magnetic resonance imaging
Dynamic brain functional network
Independent component analysis
Functional connectivity state
Schizophrenia
