Abstract：
The neuroimaging mechanisms underlying differences in the outcomes of sound therapy for tinnitus patients remain unclear. We hypothesize that abnormal hierarchical architecture is the neuro-biomarker for treatment outcome explanation. We conducted functional connectome gradient analyses on resting-state functional MRI images that acquired before intervention to investigate differences among the patients with effective treatment (ET, n = 27), ineffective treatment (IT, n = 41), and healthy controls (HC, n = 59). General linear models were used to analyze the associations between intergroup differential regions and clinical characteristics. Partial least squares regression was employed to reveal correlations with gene expression. Compared to HC, both ET and IT groups displayed significant differences in the default mode network. Moreover, the ET group exhibited wider gradient range and greater gradient variance. Also, the gradient scores of the differential regions between the ET and HC groups were significantly correlated with Self-rating Anxiety Scale and Self-rating Depression Scale scores, and exhibited positive correlations with the transcriptional profiles of genes related to depression and anxiety. Our results indicated that the abnormalities of ET group, may be more relevant to psychiatric disorders, bringing a higher possible therapeutic potential due to the plasticity of the nervous system. Connectome gradient dysfunction with genetic evidence may serve as an indicator for identifying diverse treatment outcomes of the sound therapy for tinnitus patients before treatment.
摘要：
耳鸣患者声音治疗结果差异的神经影像学机制尚不清楚。我们假设异常分层结构是治疗结果解释的神经生物标志物。我们对干预前获得的静息状态功能MRI图像进行了功能连接体梯度分析，以调查有效治疗患者之间的差异（ET，n=27),无效治疗(IT，n=41),和健康对照(HC，n=59）。一般线性模型用于分析组间差异区域与临床特征之间的关联。采用偏最小二乘回归来揭示与基因表达的相关性。与HC相比，ET和IT组在默认模式网络中显示出显著差异。此外，ET组表现出更宽的梯度范围和更大的梯度方差。此外，ET和HC组间差异区域的梯度得分与焦虑自评量表和抑郁自评量表得分显著相关,并与抑郁和焦虑相关基因的转录谱呈正相关。我们的结果表明，ET异常组，可能与精神疾病更相关，由于神经系统的可塑性，带来了更高的治疗潜力。具有遗传证据的Connectome梯度功能障碍可以作为确定耳鸣患者在治疗前声音疗法的不同治疗结果的指标。
