How to extract high discriminative features that help classification from complex resting-state fMRI (rs-fMRI) data is the key to improving the accuracy of brain disease recognition such as schizophrenia. In this work, we use a weighted sparse model for brain network construction, and utilize the Kendall correlation coefficient (KCC) to extract the discriminative connectivity features for schizophrenia classification, which is conducted with the linear support vector machine. Experimental results based on the rs-fMRI of 57 schizophrenia patients and 64 healthy controls show that our proposed method is more effective ( i.e., achieving a significantly higher classification accuracy, 81.82%) than other competing methods. Specifically, compared with the traditional network construction methods (Pearson's correlation and sparse representation) and the commonly used feature selection methods (two-sample t-test and Least absolute shrinkage and selection operator (Lasso)), the algorithm proposed in this paper can more effectively extract the discriminative connectivity features between the schizophrenia patients and the healthy controls, and further improve the classification accuracy. At the same time, the discriminative connectivity features extracted in the work could be used as the potential clinical biomarkers to assist the identification of schizophrenia.
如何从复杂的静息态功能核磁共振成像（rs-fMRI）数据中提取高鉴别性特征，是提升精神分裂症识别精度的关键。本文使用一种加权稀疏脑网络构建方法，采用肯德尔相关系数（KCC）从脑网络中提取连接特征，并基于线性支持向量机对 57 例精神分裂症患者与 64 例健康受试者进行分类研究，最终得到了较高的分类精度（81.82%）。本文研究结果表明，相较于传统的皮尔逊相关和基于稀疏表示的脑网络构建方法，以及常用的双样本 t检验（ t-test）和最小绝对收缩与选择算子（Lasso）特征选择方法，本文提出的算法可以更有效地提取出能够区分精神分裂症患者与健康人群的脑功能网络连接特征，进而提升分类精度；同时本研究中所提取的鉴别性连接特征或可作为潜在的临床生物学标志物，用以辅助精神分裂症的诊断。.
Keywords: brain functional network; classification; feature selection; functional magnetic resonance imaging; schizophrenia.