Research On State Observation And Similarity Measurement Method Of Dynamic Brain Functional Network

Functional Magnetic Resonance Imaging(f MRI)is one of the most widely used non-invasive brain imaging techniques in recent years.f MRI uses the important relationship between the collected signals and the underlying neuron activity as an expression.An important way of cortical function for imaging functionally active brain regions in health and disease.The most common f MRI technique is Blood Oygenation Level Dependent(BOLD)imaging,which has dominated the field since its discovery.BOLD-f MRI uses hemoglobin as an endogenous contrast agent and relies on the magnetization difference between oxygenated and deoxyhemoglobin to generate f MRI signals.BOLD-f MRI signals change with the continuous activity and changes of the brain,so the functional connection network between brain regions also changes with time.During the research process,f MRI technology provided a good foundation for the reconstruction of the human brain functional network with time attributes and further state analysis,but because the time-varying features of the brain network state were not considered,the inter-sample Individual differences and state transition trends make it difficult to investigate the state transition laws of brain functional networks over time distribution.This paper proposes a dynamic brain function network state observation and similarity measurement method.This method is based on the statistics of brain function network state transition time points,and builds a brain function network state transition observation model to observe the brain network state transition trend.According to the trend of brain network state transition,this paper considers that the state of the brain network has undergone state transition over time,and is composed of a stable state and a transition state.Through the change of the state of the brain network at the time point,then the brain state network is constructed,and the similarity measurement analysis is performed on the brain state network.In order to verify the effectiveness of dynamic brain function network state observation and similarity measurement methods,this paper carried out experiments and analysis from the following three aspects: constructing a brain network state transition observation model;reconstructing the brain network state network;usingtopological overlap coefficients The similarity measurement analysis was performed on the brain state network between healthy children and autistic children.The experimental results show that by observing the brain network state transition observation model to observe the brain network state transition trend,the brain network state transition occurs in a time interval,not a transient.By observing the changes in the state of the brain network,this paper finds that the state of the brain network is composed of a stable state and a transition state,and then the topological overlap coefficient is used to measure the network similarity of the brain state network after reconstruction.The experimental results show that in a healthy child sample experiment,The network similarity between the steady state and the transition state is significantly different,and the method in this paper can effectively distinguish between healthy child samples and autistic child samples.