Journal of Computer Applications ›› 2021, Vol. 41 ›› Issue (9): 2767-2772.DOI: 10.11772/j.issn.1001-9081.2020111794

Special Issue: 前沿与综合应用

• Frontier and comprehensive applications • Previous Articles     Next Articles

Degree centrality based method for cognitive feature selection

ZHANG Xiaofei1,2,3, YANG Yang3,4, HUANG Jiajin1,3, ZHONG Ning1,3,5   

  1. 1. Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China;
    2. School of Computer, Jiangsu University of Science and Technology, Zhenjiang Jiangsu 212003, China;
    3. Beijing Key Laboratory of MRI and Brain Informatics, Beijing 100124, China;
    4. School of Humanities and Social Sciences, Beijing Forest University, Beijing 100089, China;
    5. Department of Life Science and Informatics, Maebashi Institute of Technology, Maebashi Gunma 371-0816, Japan
  • Received:2020-11-17 Revised:2021-01-07 Online:2021-09-10 Published:2021-05-08
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61420106005), the Beijing Municipal Natural Science Foundation (4182005), the Science and Technology Plan of Beijing Municipal Education Commission (KM201710005026), the Major Project of Philosophy and Social Sciences in Jiangsu Universities (2020SJZDA065).

基于度中心性的认知特征选择方法

张笑非1,2,3, 杨阳3,4, 黄佳进1,3, 钟宁1,3,5   

  1. 1. 北京工业大学 信息学部, 北京 100124;
    2. 江苏科技大学 计算机学院, 江苏 镇江 212003;
    3. 磁共振成像脑信息学北京市重点实验室, 北京 100124;
    4. 北京林业大学 人文社会科学学院, 北京 100089;
    5. 前桥工科大学 生命科学与信息学系, 群马 前桥 371-0816, 日本
  • 通讯作者: 钟宁
  • 作者简介:张笑非(1980-),男,江苏高邮人,讲师,博士研究生,CCF会员,主要研究方向:脑信息学、人工智能;杨阳(1984-),男,云南曲靖人,讲师,博士,主要研究方向:心理学、脑信息学、认识神经科学;黄佳进(1977-),男,贵州遵义人,讲师,博士,CCF会员,主要研究方向:人工智能;钟宁(1956-),男,北京人,教授,博士生导师,博士,CCF会员,主要研究方向:脑信息学、Web智能、粗糙集。
  • 基金资助:
    国家自然科学基金资助项目(61420106005);北京市自然科学基金资助项目(4182005);北京市教委科技计划项目(KM201710005026);江苏高校哲学社会科学重大项目(2020SJZDA065)。

Abstract: To address the uncertainty of cognitive feature selection in brain atlas, a Degree Centrality based Cognitive Feature Selection Method (DC-CFSM) was proposed. First, the Functional Brain Network (FBN) of the subjects in the cognitive experiment tasks was constructed based on the brain atlas, and the Degree Centrality (DC) of each Region Of Interest (ROI) of the FBN was calculated. Next, the difference significances of the subjects' same cortical ROI under different cognitive states during executing cognitive task were statistically compared and ranked. Finally, the Human Brain Cognitive Architecture-Area Under Curve (HBCA-AUC) values were calculated for the ranked regions of interest, and the performances of several cognitive feature selection methods were evaluated. In the experiments on functional Magnetic Resonance Imaging (fMRI) data of mental arithmetic cognitive tasks, the values of HBCA-AUC obtained by DC-CFSM on the Task Positive System (TPS), Task Negative System (TNS), and Task Support System (TSS) of the human brain cognitive architecture were 0.669 2, 0.304 0 and 0.468 5 respectively. Compared with Extremely randomized Trees (Extra Trees), Adaptive Boosting (AdaBoost), random forest, and eXtreme Gradient Boosting (XGB), the recognition rate for TPS of DC-CFSM was increased by 22.17%, 13.90%, 24.32% and 37.19% respectively, while its misrecognition rate for TNS was reduced by 20.46%, 29.70%, 44.96% and 33.39% respectively. DC-CFSM can better reflect the categories and functions of the human brain cognitive system in the selection of cognitive features of brain atlas.

Key words: human brain cognitive architecture, cognitive feature, brain atlas, functional Magnetic Resonance Imaging (fMRI), Region Of Interest (ROI), degree centrality

摘要: 针对大脑图谱认知特征选择的不确定性提出了基于度中心性的认知特征选择方法(DC-CFSM)。首先,基于大脑图谱构建认知实验任务中被试的脑功能网络(FBN),并计算得到FBN每个兴趣点(ROI)的度中心性(DC);其次,统计对比被试相同皮质兴趣点在执行认知任务时不同认知状态间的差异显著性并对其进行排序;最后,根据排序后的ROI计算人脑认知体系曲线下面积(HBCA-AUC)值,并评估几种认知特征选择方法的性能。在心算认知任务功能核磁共振成像(fMRI)数据上进行的实验中,DC-CFSM在人脑认知体系的任务正相关系统(TPS)、任务负相关系统(TNS)及任务支撑系统(TSS)上得到的HBCA-AUC值分别为0.669 2、0.304 0、0.468 5。与极限树、自适应提升、随机森林、极限梯度提升(XGB)等方法相比,DC-CFSM对TPS的识别率分别提高了22.17%、13.90%、24.32%和37.19%,对TNS的误识率分别减小了20.46%、29.70%、44.96%和33.39%。可见DC-CFSM在大脑图谱认知特征的选择上更能反映人脑认知体系的类别和功能。

关键词: 人脑认知体系, 认知特征, 大脑图谱, 功能核磁共振成像, 兴趣点, 度中心性

CLC Number: