关键词: 核心节点, 人脑结构网络, 点介中心性, K-shell分解法, 度中心性, 邻近中心性

Abstract: Identification of core nodes is of great significance for the research of whole human brain network which is reconstructed based on nuclear magnetic resonance imaging technology. A novel core node evaluation method with K-shell decomposition and betweenness centrality is presented in this paper. Firstly the important nodes in human brain network are evaluated by degree centrality, betweenness centrality and closeness centrality respectively and the relativity between the results is discussed, then the K-shell decomposition method is used to find those core nodes in the shell, and finally these two methods are integrated together to identify the most important core nodes in the human brain network taking into account the overall characteristics and local characteristics of the brain network node. Experimental results show this method provides a more comprehensive and accurate assessment for the core nodes in human brain network.

Key words: core node, human brain structure network, betweenness centrality, [K]-shell decomposition method, degree centrality, closeness centrality