最小生成树脑网络分析及自闭症分类研究

Brain Network Analysis Based on Minimum Spanning Tree and Classification in Autism

不同年龄的自闭症患者所表现出来的临床表征差异很大,但这些差异在影像学指标上却难以发现。为了解决这一问题,在静息态功能脑网络基础上,引入最小生成树分析方法,利用度、介数、离心率三个节点指标,对不同年龄分组(儿童-青少年,青少年-成人)之间进行差异分析。进而,根据统计显著性差异提取分类特征,结合SVM分类算法,构建一个准确率较高的模型。结果表明,在两组(儿童-青少年,青少年-成人)对比分析中均得到显著性差异区域,分类准确率分别为80.38%和81.88%.该方法为自闭症不同年龄患者影像学分析及辅助诊断提供了新的方法和思路。

There＇is a huge difference in the clinical characterization of autism in different ages,which is difficult to be detected on the basis of imaging indicators.In order to solve this problem,this study employed the minimum spanning tree analysis method based on the static state functional brain network.Node attributes,such as degree,betweenness centrality and eccentricity,were used to analyze the differences between different age groups（children-adolescents,adolescents-adults）.At the same time,with the significant regions as the feature,a the multi-parameter optimization framework was proposed and a model with high diagnostic accuracy of autism was constructed.The results show that significant differences existed between the two groups.The classification accuracy was 80.38% and 81.88% respectively.This research provides an important method and a new idea for imaging analysis and auxiliary diagnosis of austism patients with different age.