单相抑郁和双相抑郁是常见的情绪障碍。由于没有公认的生物学标记可以作为研究和应用的金标准，导致目前单双相抑郁的临床误诊率高，且治疗效果参差不齐。已有研究表明，情感环路的异常模式很可能是区分单双相抑郁的重要“窗口”。为此，本项目使用最新的方法进行情感环路核心脑区的精确分割与定位。在此基础上，结合多模态磁共振影像、情感量表和情感功能任务的测查，探索单双相抑郁情感环路中关键节点的相互作用和调节机制，情感环路异常的功能机制、结构机制以及它们之间的相互关系。同时基于机器学习方法和神经网络预测模型，尝试实现对抑郁类型的准确区分和个体情感功能的预测。本项目将进一步揭示单、双相抑郁情感环路相关的特异性神经缺陷，为临床客观诊断和精准治疗提供影像学依据。

The major depression disorders (MDD) and bipolar disorders (BD) are common psychiatric disorders in our society. However, the absence of validated biomarkers has resulted in significant clinical issues, including high misdiagnosis rate and poor treatment outcome. Recent studies suggested that the distinct abnormal patterns in emotional circuit might be a crucial “window” for differentiating MDD and BD patients. Thus, in current project, we segmented and located the core brain areas in emotional circuit by using novel neuroimaging approaches. Then, emotion-related tasks, emotional scales, combined with multi-model MRI image data were used to explore the interactions and regulatory mechanism between the critical nodes in emotional circuit, and the abnormal function mechanism, structure mechanism and the relationship between them. Moreover, we conducted the machine learning and the connectome-base predictive modeling, to accurately distinguish BD from MDD, and to predict emotional function in individual level, respectively. This project might shed light on the specific neural deficit of emotional circuit in MDD and BD, which would benefit current clinical diagnosis and treatment system.