中文摘要：

目的利用磁共振脑功能成像的功能连接分析技术，研究焦虑症的神经病理机制。方法共20名受试者参加实验。焦虑症患者10例，正常对照组10名。首先根据词语任务刺激的fMRI实验，确定感兴趣脑区。实验1情绪中性词与无词的空白对照交替，实验2威胁词与情绪中性词交替。实验模式为Block Design，刺激材料通过听觉呈现。脑激活结果显示患者组两侧颞上回BOLD激活强度明显强于正常对照组，提示颞上回的功能异常可能与其神经病理机制有关。因此选定两侧颞上回作为感兴趣区，利用我们自行研发的频域一致性功能连接分析方法，计算两组受试者两侧颞上回的功能连接度，进行组间比较与分析。结果词语刺激的脑激活模式：实验1患者组与对照组激活脑区大部分重叠。包括两侧颞上回、颞中回；左侧额下回；双侧前运动区；辅助运动区及小脑半球。实验2患者组激活两侧颞上回、颞中回；两侧额下回；左前额叶背外侧；两侧顶下小叶；两侧前运动区；扣带回前部；辅助运动区。两次实验中患者组两侧颞上回激活强度强于对照组。频域一致性功能连接分析结果显示实验1焦虑症患者与正常对照组两侧颞上回之间的功能连接度差异无统计学意义（正常对照η=0．5636，患者η=0．5500，P〉0．05）。实验2两组受试者（正常对照η=0．5616，患者η=0．4926，P〈0．05）功能连接度η的差异有统计学意义。结论焦虑症患者两侧颞上回激活强度强于对照组，但两者之间的功能连接度降低，尤其在处理威胁词时更显著。这提示脑功能连接度降低可能与焦虑症患者的神经病理机制有关。

英文摘要：

Objective Neuronal circuits involved in the pathophysiology of anxiety are not yet fully understood. We used functional connectivity MRI to explore the characteristic of functional connectivity in anxiety disorders patient and the neural mechanism of this disease. This work was selected as an oral presentation in 2006 ISMRM. Methods Twenty right-handed subjects were included in this study, and were divided into two groups. The anxiety （P） group （n = 10; 7 male, mean age 42 years） consisted of patients meeting DSM-Ⅳ criteria for a principal diagnosis of anxiety disorder. The control （ C ） group consisted of volunteers free of psychiatric symptoms, and was matched on age and gender （n = 10; 7 male ） with the panic patients. The subjects underwent noninvasive functional magnetic resonance imaging while listening actively to （ 1 ） ： emotionally neutral word alternating with no word as the control condition （ CN, PN ） , and （2）： threat-related words alternating with emotionally neutral word as the experimental condition （CT, PT）. Each word was presented in pseudorandom order in each 16 s block of 12 words of the same type. Eight alternating blocks of neutral words were presented for about 256 s. The subject was only asked to passively listen to each word. All MRI data were obtained on a 1.5-Tesla scanner Data analysis was performed with SPM99 to find significant activations in two tasks for two groups. Based on group t-test, we chose two anatomically defined regions： left superior temporal gyms （GTs） and right GTs. Then, based on individual t-map, the voxel with the largest t-value within two regions was taken as the subject-specific peak voxel. We define clusters based on faces and edges, but not corners, so each voxel has 18 neighbors. Subject-specific averaged time series were extracted by averaging the time series of 19 voxels. Since healthy control subjects showed no significant activation （ corrected, P 〈 0. 05 ） during processing of anxiety word to neutral word,