The Research Of Brain Functional Network In Generalized Anxiety Disorder Based On Magnetic Resonance Imaging

Generalized anxiety disorder（GAD）is an extremely common mental disorder in adults,and is characterized by excessive,uncontrollable mental and physical anxiety.Exploring the neural activity of the brain’s functional networks in GAD could reveal the pathophysiological mechanism of GAD to a certain extent,thus providing effective scientific basis for the diagnosis and treatment of GAD..However,up to now,the research results on the neural mechanisms of GAD are not repeatable and the methods are relatively simple.Therefore,in order to further characterize the brain image representation of GAD,to reveal its potential pathological mechanism.Based on the functional magnetic resonance imaging（f MRI）technique,this paper combines brain image with clinical behavior,from resting state to task state,to explore the abnormal neural activity of GAD in multiple dimensions.The work of this paper mainly consists of the following two parts:1.To address the specific changes of brain resting state functional network caused by anxiety symptoms in GAD.The multivariate distance-based matrix regression（MDMR）was used,to capture the regions where whole-brain resting-state functional connectivity was associated with anxiety symptoms in GAD patient,and further examined the specificity of the relationship between connectivity within the discovered network and the severity of psychic and somatic symptoms.Finally,a support vector machine regression（SVR）model was used to explore the predictive power of symptom related abnormity brain functional connectivity for clinical anxiety scores in a test sample.The results showed that whole-brain connectivity in multiple regions of the salience network（SN）and the default mode network（DMN）was associated with anxiety symptoms.In addition,we further found that the somatic symptoms was dominated by the hyper-connectivity within SN and psychic symptoms was linked with the hyperconnectivity within DMN.In contrast,the hypo-connectivity between SN and DMN contributed to both types of anxiety symptoms.Importantly,the functional connectivity pattern of DMN could predict the individual psychic symptom in an independent GAD sample.These findings emphasized the dissociable roles of SN and DMN in the pathophysiology of psychic and somatic anxiety symptoms,which might be crucial in providing a promising neuroimaging biomarker for novel personalized treatment strategies for GAD patients.2.Aiming at the problem of dynamic switching and reorganization of brain functional network in GAD patients under emotional stimulus task state,we used sliding window method to construct dynamic functional connectivity matrix,and uses an iterative and ordered Louvain algorithm under each window to detect the community structure of emotion network nodes changing over time during emotional task community detection algorithm was used to reveal the flexibility and module allegiance of nodes in multi-layer dynamic networks.The results showed that,in the context of positive emotion,the patients had less flexibility in the left medial superior frontal gyrus of the prefrontal network than the healthy controls.And in the context of negative emotions,the left triangular inferior frontal gyrus and inferior orbitofrontal gyrus in the prefrontal network,the left putamen and pallidum in the subcortical nucleus were higher flexible in patients with GAD.Further study showed that the decrease of flexibility in left medial superior frontal gyrus was negatively correlated with anxiety symptoms,and positively related with self-esteem and extroversion.These results reveal the GAD patients abnormal brain networks under emotional stimuli of dynamic switching mechanism,it also indicates that patients with GAD have strong negative emotional bias in the process of emotional processing stimulated,this may be explained GAD bottom-up strong emotion production and top-down emotion regulation dysregulation mechanism provides new evidence.