Computer Engineering and Applications ›› 2020, Vol. 56 ›› Issue (21): 272-278.DOI: 10.3778/j.issn.1002-8331.1907-0408

Previous Articles    

Analysis and Application Research of Brain Activation Task Differentiation

WANG Gongshu, REN Zunxiao, LI Dandan, XIANG Jie, WANG Bin   

  1. 1.School of Information and Computer Science, Taiyuan University of Technology, Taiyuan 030000, China
    2.College of Big Data, Taiyuan University of Technology, Taiyuan 030000, China
  • Online:2020-11-01 Published:2020-11-03

脑激活任务区分度的分析及应用研究

王工书,任尊晓,李丹丹,相洁,王彬   

  1. 1.太原理工大学 信息与计算机学院,太原 030000
    2.太原理工大学 大数据学院,太原 030000

Abstract:

The patterns of brain activation vary greatly when brain performing different types of tasks. Based on this, a new method of task differentiation is proposed. The functional Magnetic Resonance Imaging(fMRI) of task state is analyzed based on similarity calculation to measure the degree of differentiation of activation patterns of each brain region when different conditions are performed, and reveal the ability of each brain region to characterize tasks. fMRI data of memory extraction tasks of normal people and bipolar disorder patients are analyzed. Three commonly used similarity measurement methods, namely Pearson correlation analysis, Euclidean distance calculation and cosine similarity, are used to calculate task discrimination of each brain region. The results show that the brain regions with high discrimination are involved in memory, attention and visual information, which indicates the accuracy and scientificity of this method. Bipolar disorder patients have a lower level of task differentiation in areas of the brain responsible for memory and attention, indicating impaired brain function. In addition, it is found that the discrimination calculation based on Pearson correlation analysis performs well. In conclusion, the method of brain activation task discrimination based on similarity analysis can be applied to fMRI analysis of task-state and the corresponding analysis of brain function.

Key words: functional Magnetic Resonance Imaging(fMRI), task activation, similarity analysis

摘要:

大脑在执行不同类型任务时激活模式各不相同,变化很大,各个脑区的变化程度也不同。据此,提出任务区分度计算这一全新的方法。用相似性度量对任务态功能磁共振成像(functional Magnetic Resonance Imaging,fMRI)分析,衡量大脑在执行不同条件时各个脑区激活模式的区分程度,揭示大脑各个区域对任务的表征能力。实验对正常人和狂躁症患者记忆提取任务的fMRI数据进行分析,使用皮尔逊相关分析、余弦相似度分析和欧几里德距离计算3种常用的相似性度量方法,并计算各个脑区的任务区分度。结果表明区分度较高的脑区参与记忆、注意和视觉信息等功能,表明了该方法的准确性和科学性。狂躁症患者在负责记忆和注意等脑区的任务区分度较正常人低,表明患者脑功能受损。此外,研究还发现基于皮尔逊相关分析的区分度计算表现较好。通过与SVM方法的对比证明了该方法在区分不同任务的激活模式时的优越性。综上,基于相似性度量的脑激活任务区分度的方法能够适用于任务态fMRI分析及其相应的脑功能分析。

关键词: 功能磁共振成像(fMRI), 任务激活, 相似性度量