精细亚区尺度难治性抑郁症异常神经环路的多模态MRI解析

Treatment-resistant depression (TRD) is one subtype of depression, which is with the greatest difficulty in treatment, the most difficult of social function recovery, and the heaviest disease burden. Currently, there is lack of objective standard to recognize TRD from non-TRD early. There is also lack of quantitative, objective biological indicators to assess the treatment effects on TRD. Therefore, understanding the abnormal mechanism of neural circuits in TRD and finding effective objective biomarkers are urgently needed for basic research and clinical practice of depression disorder. This project will use multimodal MRI datasets acquired from different clinical diagnosis and treatment phases to perform the following three studies: 1) to map the abnormal brain sub-regions and related neural circuits of TRD at a more fine-grained subregional scale; 2) By fusing multimodality MRI findings and extracting effective features, to identify the potential imaging biomarkers for early diagnosis and treatment assessment of TRD based on machine learning; 3) Based on the above findings, we will observe the structural and functional changes of TRD related neural circuits during repetitive TMS (rTMS) treatment, and try to optimize the precise stimulation localization of rTMS. Through the above studies, the current project will provide a scientific basis for diagnosis, treatment methods, and will provide potential biomarker for the clinical treatment and prognosis of TRD.

难治性抑郁症（TRD）是抑郁症中治疗难度最大、对患者社会功能恢复影响最大、疾病负担最重的亚型。如何准确无创的测量脑结构与功能信息，解析TRD异常神经环路并寻找有效的生物学标记，成为TRD临床诊疗迫切需要解决的问题。多模态脑成像技术的发展成为解决这一问题的关键。本课题将立足于抑郁症异常神经环路假说，开展以下研究：1）通过采集TRD不同临床诊疗阶段的多模态磁共振数据，在脑区精细亚区尺度上系统解析相关神经环路的结构功能异常；2）融合多模态脑影像发现并提取有效特征，利用机器学习分类算法，寻找应用于TRD与非TRD早期识别的影像学标记，使TRD干预治疗的窗口期提前；在此基础上，3) 观察重复经颅磁刺激（rTMS）治疗前后TRD异常神经环路的改变，并尝试优化TRD的刺激靶区定位。通过上述研究，本项目旨在为TRD临床早期诊断、疗效评价提供可靠、定量的影像学标记，并为其治疗方案的优化提供科学依据和新思路。

研究表明情绪神经环路的结构与功能异常可能是难治性抑郁症的重要生物标志，如果能对抑郁症异常情绪环路进行精细解析，就可能获得抑郁症早期诊断的生物标志和治疗的精准靶区。但目前国际上已有的人类情绪环路结构与功能研究是基于临床常用的Brodmann脑图谱，对于情绪环路的结构和功能了解较为模糊，还不能绘制出精细情绪环路图谱。本课题利用多模态脑成像技术，建立了全脑尺度的脑网络组图谱，解析了情绪功能相关的精细亚区尺度的神经环路；2）在此基础上建立了情绪环路精细图谱的临床应用范式，为该图谱的大规模临床应用研究奠定基础。