Study On Brain Network Of DOC Patients In The Consciousness Recovery Process

The diagnosis and prognosis of disorder of consciousness(DOC)in the recovery process is one of the most urgent problems in clinical treatment.At present,the multi-behavior assessment scale is mainly used to assess the status of consciousness.The revised CRS-R score is the most widely used behavioral assessment scale in clinical practice,but the misdiagnosis rate of this method is high.Therefore,clinical and academic circles have been exploring more objective and effective diagnostic methods of consciousness,such as functional magnetic resonance imaging,positron emission tomography and electroencephalography.Among them,EEG has the advantages of noninvasiveness and painlessness,low cost and high time resolution,which can objectively and accurately reflect the brain dynamic information of patients.It is widely used as an aid in clinical examination of patients with severe disturbance of consciousness.The brain network analysis method in EEG can reflect the information integration and information transfer ability between different brain regions.It has certain advantages in the diagnosis of clinically significant brain diseases,which was widely used in the diagnosis of diseases such as Alzheimer's disease,Parkinson's disease and epilepsy.However,the application of brain network method in the diagnosis of consciousness status in patients with severe mental disorder is still in the preliminary stage of exploration.28 DOC patients in the resting state of EEG in the process of recovering consciousness were collected.After the signal was pretreated,the brain function network and brain effect network were constructed to analysis the change of the characteristic parameters in the process of recovering consciousness.Finally,the correlation between the brain network characteristic parameters and the CRS-R score was analyzed.It provided a theoretical basis for the use of brain network parameters as clinical diagnostic evaluation.The main results of this paper were as follows:(1)The brain function network was constructed after denoising,and the small world attribute of brain function network based on synchronization likelihood.The results showed that as the CRSR score increased during the recovery process,the small world attributes of the subjects were enhanced.Moreover,the larger the increase of CRS-R score,the more the network small world attributes increased.(2)The brain effect network was constructed after denoising,and the local efficiency,casual flow and casual density were calculated based on Granger causality model.The results showed that as the CRS-R score increased during the recovery process,there was no significant change in the brain effect network parameters in the whole brain and non-damaged areas.Only the local efficiency of the damaged area node,the causality of the damage area and the causality density of the damage area were all enhanced.(3)Based on the results of analysis above,correlation between small world properties and CRSR score,correlation analysis between local efficiency and CRS-R score,correlation analysis between casual flow of damaged area and CRS-R score,correlation analysis between casual density of damaged area and CRS-R score were analyzed.The result showed that the correlation coefficients were 0.360±0.129,0.477±0.109,0.808±0.098,0.359±0.128 respectively.The correlation coefficient between causality and CRS-R score in the damage area was the highest,and the correlation coefficient of each subject was the least fluctuant,which was the most suitable for the evaluation of patients' consciousness.