重度抑郁症患者脑功能网络的分类研究

Research on classification of brain functional network in major depressive patient

为了构建辅助诊断模型,为抑郁症的诊断提供一个新的方法,以提高抑郁症诊断的准确率。在连续的阈值空间(8%～32%)内构建所有被试的功能脑网络并使用复杂网络理论对抑郁症患者的脑网络进行分析,力求提取出可以从各个维度来表征抑郁症患者的脑网络的特征值,采用不同的属性组合并使用SVM分类算法对所有被试进行分类研究,结果发现将全局属性与局部属性组合作为分类特征得到的分类正确率最高,因此该方法可以用于抑郁症的辅助诊断中。

In order to construct a computer-aided diagnosis model and propose a new method for depressive diagnosis to im- prove the accuracy of depression diagnosis, this article constructed the whole-brain functional networks of all subjects and ana- lyzed the brain networks of depressive patients from different perspectives to extract characteristic values which could represent brain networks of depressive patients from different dimensions. And it classified all subjects by SVM classification algorithm research. Using different property group, results show that the rate of correct is highest when combination of global and local properties used as classification characteristics, so this method can be used in the auxiliary diagnosis of depression.