脑源成像技术及其应用研究进展  

Progress in brain source imaging and its applications

作　　者：魏玉烜 王旭 江海腾 孟建军[1] WEI YuXuan;WANG Xu;Jiang HaiTeng;MENG JianJun(Institute of Robotics,School of Mechanical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;Department of Neurobiology,Affiliated Mental Health Center&Hangzhou Seventh People’s Hospital,Zhejiang University School of Medicine,Hangzhou 310013,China)

机构地区：[1]上海交通大学机械与动力工程学院机器人研究所,上海200240 [2]浙江大学医学院附属精神卫生中心(杭州市第七人民医院),杭州310013

出　　处：《中国科学：技术科学》2024年第2期196-224,共29页Scientia Sinica(Technologica)

基　　金：“科技创新2030-脑科学与类脑研究”重大项目(编号:2022ZD0212400);国家自然科学基金项目(批准号:52175023)资助。

摘　　要：脑电图(EEG)和脑磁图(MEG)为观察和分析人脑活动提供了重要的无创测量工具. EEG/MEG可采集大规模神经集群放电活动引起的体表电势/磁场,采集的信号具备高时间分辨率,因而在神经科学、临床医学与神经工程等领域得到广泛的应用.然而, EEG/MEG测得的是颅内神经集群活动经过脑组织、脑脊液、颅骨和头皮等人体容积导体组织弥散传播后引起的颅外电势/磁场,其空间分辨率低,无法反映神经活动的真实空间分布.源成像是一种结合人体头部组织的数学模型,用EEG/MEG来推导颅内神经活动源的成像技术,是目前公认的提高无创测量空间分辨率的有效技术.典型的源成像问题包括正问题建模和逆问题求解两个方面,其中逆问题求解是源成像的核心,而正问题建模是逆问题求解的基础,两者相辅相成.本文通过文献检索梳理和归纳目前存在的正问题建模和逆问题求解的典型方法和常用软件,介绍影响源成像精度的因素和评价源成像性能的指标和方法.另外,本文还着重介绍了源成像技术在神经科学、临床医学以及神经工程-脑机接口方面的应用和研究进展.最后,讨论了当前源成像技术面临的挑战,以及针对这些挑战该领域内的研究前沿和趋势,展望了源成像在推动神经科学、临床医学等科学研究和技术进步方面的潜力.Electroencephalogram (EEG) and magnetoencephalogram (MEG) provide important noninvasive tools for inspecting and analyzinghuman brain activities. EEG/MEG captures the scalp potential/magnetic field produced by large-scale neural activity, and the acquiredsignals have a high temporal resolution. Therefore, they have been widely used in neuroscience, clinical medicine, andneuroengineering. However, EEG/MEG measures the extracranial potential/magnetic field produced by the diffuse transmission ofintracranial neural activity through volume conductors such as brain tissue, cerebrospinal fluid, skull, and scalp. As a result, EEG/MEG measurements suffer from low spatial resolution and cannot reflect the actual spatial distribution of neural activity. Sourceimaging is an imaging technology that combines EEG/MEG with mathematical models of the human head to estimate the underlyingbrain electrical source. It is currently recognized as an effective way to improve the spatial resolution of noninvasive measurements. Atypical source imaging problem is composed of a forward problem and an inverse problem. The inverse problem is the core of sourceimaging, and the forward problem is the basis of the inverse problem. These two components complement each other. This papersummarized typical methods and commonly used software for forward and inverse problems through a literature review. Then, wedescribed factors affecting source imaging accuracy, as well as metrics and methods for evaluating source imaging performance. Inaddition, applications and research progress of source imaging technology in neuroscience, clinical medicine, and neuroengineering(especially brain-computer interface) were highlighted. Finally, we discussed current challenges in source imaging technology, as wellas frontiers and trends in this field in response to these challenges, and looked into the future of source imaging to promote scientificresearch and technology in neuroscience, clinical medicine and other areas.

关 键 词：源成像 源定位 源重建 脑电图 脑磁图 逆问题 

分 类 号：R741.044[医药卫生—神经病学与精神病学]