Peng Yu, Zheng Yang, Yuan Ziwen, Guo Jing, Fan Chunyang, Li Chenxi, Deng Jingyuan, Song Siming, Qiao Jin, Wang Jue
The Key Laboratory of Biomedical Information Engineering of Ministry of Education, School of Life Sciences and Technology, Institute of Biomedical Engineering, Xi'an Jiaotong University, Xi'an, China.
Department of Rehabilitation, First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China.
Front Neurosci. 2023 Aug 16;17:1242543. doi: 10.3389/fnins.2023.1242543. eCollection 2023.
Post-stroke depression (PSD) may be associated with the altered brain network property. This study aimed at exploring the brain network characteristics of PSD under the classic cognitive task, i.e., the oddball task, in order to promote our understanding of the pathogenesis and the diagnosis of PSD.
Nineteen stroke survivors with PSD and 18 stroke survivors with no PSD (non-PSD) were recruited. The functional near-infrared spectroscopy (fNIRS) covering the dorsolateral prefrontal cortex was recorded during the oddball task state and the resting state. The brain network characteristics were extracted using the graph theory and compared between the PSD and the non-PSD subjects. In addition, the classification performance between the PSD and non-PSD subjects was evaluated using features in the resting and the task state, respectively.
Compared with the resting state, more brain network characteristics in the task state showed significant differences between the PSD and non-PSD groups, resulting in better classification performance. In the task state, the assortativity, clustering coefficient, characteristic path length, and local efficiency of the PSD subjects was larger compared with the non-PSD subjects while the global efficiency of the PSD subjects was smaller than that of the non-PSD subjects.
The altered brain network properties associated with PSD in the cognitive task state were more distinct compared with the resting state, and the ability of the brain network to resist attack and transmit information was reduced in PSD patients in the task state.
This study demonstrated the feasibility and superiority of investigating brain network properties in the task state for the exploration of the pathogenesis and new diagnosis methods for PSD.
中风后抑郁（PSD）可能与大脑网络属性改变有关。本研究旨在探索在经典认知任务即Oddball任务下PSD的大脑网络特征，以增进我们对PSD发病机制和诊断的理解。
招募了19名患有PSD的中风幸存者和18名无PSD的中风幸存者（非PSD）。在Oddball任务状态和静息状态下记录覆盖背外侧前额叶皮层的功能近红外光谱（fNIRS）。使用图论提取大脑网络特征，并在PSD组和非PSD组之间进行比较。此外，分别使用静息状态和任务状态下的特征评估PSD组和非PSD组之间的分类性能。
与静息状态相比，任务状态下更多的大脑网络特征在PSD组和非PSD组之间表现出显著差异，从而导致更好的分类性能。在任务状态下，与非PSD组相比，PSD组的 assortativity、聚类系数、特征路径长度和局部效率更大，而PSD组的全局效率小于非PSD组。
与静息状态相比，认知任务状态下与PSD相关的大脑网络属性改变更为明显，并且在任务状态下PSD患者大脑网络抵抗攻击和传递信息的能力降低。
本研究证明了在任务状态下研究大脑网络属性对于探索PSD发病机制和新诊断方法的可行性和优越性。
