摘要
小世界是一种以较低的连接和能量成本实现高效的信息分离与整合的网络结构,而人脑网络具有显著的小世界特性。在弥散张量成像(diffusion tensor imaging,DTI)脑网络的研究中,如何有效地量化和评估网络的小世界属性依然是研究中存在的一个关键问题。在研究文中,我们首先概括了已有小世界属性评估指标及其存在的问题,随后提出了一种新的基于网络全局效率和局部效率的小世界属性评估指标。为了验证该指标的有效性,我们基于75个中老年人的DTI脑网络对其进行了应用与评估。与传统指标相比,该指标对研究对象的年龄变化更敏感,并与多项认知评估量表的结果存在显著相关。网络节点随机化和网络失连接这两种攻击测试的结果也表明,新指标在DTI脑网络的研究中具有较高的准确性和稳定性。
Human brain networks exhibit prominent small-world property, which facilitates efficient information segregation and integration at low wiring and energy costs. In the study of DTI（diffusion tensor imaging） brain network, how to effectively quantify and assess the small-world attributes is still a considerable problem. In this article, we firstly summarized the existing attribute indexes of small-world and their corresponding limitations. Then we propose a new index based on global and local network efficiency. To verify the validity of the new indicator, we evaluated its performance in the study of DTI brain networks from 75 elderly cases. The results showed that the new indicator was more sensitive to the age-related brain network changes, and significantly correlated with several cognitive assessment scores. In additional, two attack strategies（network randomization and edge removement） were used to test the sensitivity of the indicator. The results show that the proposed indicator have a higher accuracy and stability in the study of DTI networks compared with the existing small-world network indicators.
作者
付振荣
林岚
靳聪
王婧璇
吴水才
Fu Zhenrong;Lin Lan;Jin Cong;Wang Jingxuan;Wu Shuicai(College of Life Science and Bioengineering,Beijing University of Technology,Beijing,100124,China)
出处
《生命科学仪器》
2018年第3期42-48,41,共8页
Life Science Instruments
基金
北京市教委科技计划一般项目(KM201810005033)
国家科技支撑计划课题(2015BAI02B03)
