Research On Functional Brain Network Pattern Recognition Based On Depression

Depression is a common mental illness.Patients have feelings of sadness,guilty,and hopelessness,and are accompanied by a high degree of suicidal tendencies.Unipolar depression and bipolar disorder are the two subtypes of depressive disorder,the clinical symptoms of unipolar depression and bipolar disorder patients in depression stage are very similar,it is easy to cause misdiagnosis,so that patients cannot get the correct treatment.Therefore,studying the neural mechanism of diseases from the perspective of brain imaging helps us to better understand and distinguish these two diseases.Previous related neuroimaging studies focus on observing the anomaly of a particular brain area or the connection mode of a specific network,ignoring the connection mode of the large scale functional connection network of the whole brain.Here,we constructed a large-scale functional network to investigate abnormal functional connectivity patterns in patients with unipolar depression and bipolar disorder.In this paper,the main work is as follows:1.Using a large scale static functional connectivity analysis method,the static functional brain network of in patients with unipolar depression and bipolar disorder was divided into 10 networks,and functional connection analysis between intra-network and inter-network.The results showed that the functional connectivity specifically decreased were mainly between the auditory network and the sensor-motor network,and the connection between the sensor-motor and the visual networks,the sensor-motor and the auditory networks,the visual and the ventral attention networks,and the auditory and the visual networks in patients with unipolar depression.Compared with healthy controls,functional connectivity decreased between the auditory and the default networks in unipolar depression patients,while the in the cingulo-opercular network in patients with bipolar disorder.In addition,the connection patterns between the visual and dorsal attention networks in patients with unipolar depression and bipolar disorder are inconsistent.Finally,a categorized analysis was conducted to assess the sensitivity of the selected differential connections to differentiate between unipolar depression and bipolar disorder patients.The results showed that the effect of classification based on significant difference network was higher than that based on data driven.This confirms to a certain extent that the significant difference network found in the study can be used as the symbolic features potential possibility to distinguish two kinds of diseases.2.According to the 10 networks which were divided well,the dynamic functional network connection variability between intra-network and inter-network was analyzed in patients with unipolar depression and bipolar disorder.The results showed that the connection variability in patients with unipolar depression in the sensor-motor network increased.Meanwhile,our results showed that there was not any significant effective variability in inter-networks.In order to verify whether the sensor-motor network could become a specific network of cerebral function network in unipolar depression patients,pattern recognition and classification method were used.We found that the sensor-motor network could not be an effective feature to distinguish the unipolar and the bipolar depression patients.This may be related to the small functional connectivity variability of brain in patients with depression.The findings of this study provided new insights to understand the variability of brain functional network in patients with depression.This paper examines the abnormal connectivity of whole brain functional networks in patients with unipolar depression and bipolar disorder from the perspective of static functional connectivity and dynamic functional connectivity.The study found that patients with unipolar depression showed more specific connections between intra-network and inter-network than those with bipolar disorder.The results provide help for further understanding of the pathophysiological mechanisms underlying depression.