Abstract：
In order to comprehensively understand the changes of brain networks in patients with chronic tinnitus, this study combined static and dynamic analysis methods to explore the abnormalities of brain networks. Thirty-two patients with chronic tinnitus and 30 age-, sex- and education-matched healthy controls (HC) were recruited. Independent component analysis was used to identify resting-state networks (RSNs). Static and dynamic functional network connectivity (FNC) were performed. The temporal properties of brain network including mean dwell time (MDT), fraction time (FT) and numbers of transitions (NT) were calculated. Two-sample t test and Spearman\'s correlation were used for group compares and correlation analysis. Four RSNs showed abnormal FNC including auditory network (AUN), default mode network (DMN), attention network (AN) and sensorimotor network (SMN). For static analysis, tinnitus patients showed significantly decreased FNC in AUN-DMN, AUN-AN, DMN-AN, and DMN-SMN than HC [p < 0.05, false discovery rate (FDR) corrected]. For dynamic analysis, tinnitus patients showed significantly decreased FNC in DMN-AN in state 3 (p < 0.05, FDR corrected). MDT in state 3 was significantly decreased in tinnitus patients (t = 2.039, P = 0.046). In the tinnitus group, the score of tinnitus functional index (TFI) was negatively correlated with MDT and FT in state 4, and the duration of tinnitus was positively correlated with FT in state 1 and NT. Chronic tinnitus causes abnormal brain network connectivity. These abnormal brain networks help to clarify the mechanism of tinnitus generation and chronicity, and provide a potential basis for the treatment of tinnitus.
摘要：
为了全面了解慢性耳鸣患者脑网络的变化,本研究采用静态和动态分析相结合的方法探讨脑网络异常。32例慢性耳鸣患者和30岁，招募性别和教育匹配的健康对照（HC）。使用独立成分分析来识别静息状态网络（RSNs）。执行静态和动态功能网络连接(FNC)。脑网络的时间特性，包括平均停留时间(MDT)，计算分数时间(FT)和转变次数(NT)。采用双样本t检验和Spearman相关性进行分组比较和相关性分析。四个RSN显示异常FNC，包括听觉网络(AUN)，默认模式网络(DMN)，注意网络(AN)和感觉运动网络(SMN)。对于静态分析，耳鸣患者显示AUN-DMN中FNC显著降低，AUN-AN,DMN-AN,和DMN-SMN比HC[p<0.05,错误发现率(FDR)校正]。对于动态分析，在状态3中,耳鸣患者显示DMN-AN中的FNC显著降低(p<0.05,FDR校正)。耳鸣患者状态3的MDT显著降低(t=2.039,P=0.046)。在耳鸣组中，状态4的耳鸣功能指数（TFI）评分与MDT和FT呈负相关，状态1和NT的耳鸣持续时间与FT呈正相关。慢性耳鸣导致大脑网络连接异常。这些异常的大脑网络有助于阐明耳鸣产生和慢性的机制，为耳鸣的治疗提供潜在依据。
