Study On The Characteristics Of Dynamic Brain Network In Major Depressive Disorder

Major depressive disorder（MDD）is a common mental illness,and the typically symptoms are persistent depressed mood,anhedonia and so on.Electroconvulsive therapy（ECT）is one of the most effective methods for the treatment of drug-resistant/refractory MDD.The current studies have found that there are abnormalities in brain functional network and time-varying dynamic network of brain function in MDD.However,the transmission characteristics of brain dynamic network in MDD and the changes of dynamic characteristics of brain network after ECT are still unclear.This thesis explores the dynamic transmission characteristics of brain dynamic network in MDD and the plasticity changes of brain dynamic network before and after ECT through using nonhomogeneous Markov reachability probability model and rest state functional connectivity.This thesis includes the following two parts:1.The transmission characteristics of brain dynamic network in patients with depression.Time-varying network of brain function uses the time sliding window method to detect the evolution of brain functional network with time.It is still unclear how the time-varying neural circuit is layer and flexibly organized in the functional system.Nonhomogeneous Markov reachable probability approach can describe the stability of brain network and the hierarchical structure of brain functional system at the dynamic level.This thesis first confirmed that the functional brain network system of MDD and healthy control converged to the default mode network and occipital lobe.And these transfer probability changes were related to the behavior scores of patients.Next,this thesis explored the path of the maximum connection flow in the affect network composed of prefrontal-limbic system.and abnormalities through the anterior cingulate and insula circuits was identified.These results illustrated that the dynamic brain network has high temporal stability in the default mode network and visual cortex,and the abnormal connectivity pattern in the affect network of MDD patients.2.Changes in plasticity of the connectivity pattern from habenular to whole-brain voxel before and after ECT in MDD.The habenular is one of the hottest nodes in the affect network of MDD.This thesis used rest state functional connectivity and granger causality analysis to explore the changes of functional connectivity pattern and causality change in patients with MDD after ECT.The support vector machine classification model is established by using the characteristics of functional connectivity and causality analysis to realize the classification and recognition of MDD and healthy controls.The results show that the left angular gyrus-bilateral habenular circuit plays an important role in the improvement of depression.Changes in functional connectivity and causality are associated with changes in depressive symptoms and impaired delayed memory and recall.In addition,this change can be used as a biomarker to distinguish MDD from healthy controls.These results provided new evidences for the importance of habenular in MDD.In conclusion,based on brain connectivity analysis,this thesis detected the transmission characteristics of brain dynamic network and the plasticity changes of connectivity pattern from habenular to whole-brain voxels before and after ECT in MDD,which provided evidences of the relationship between brain image and clinical for the early diagnosis and ECT treatment of MDD.