Research On Brain Network Characteristics Of ADHD Children Based On FMRI Of Visual Attention Capture Task

Attention-hyperactivity deficit disorder(ADHD),as a common neurodevelopmental disorder in children,has become a serious public health problem in my country and requires timely intervention and timely treatment.Functional magnetic resonance imaging(fMRI)provides a technical means for the study of neurological disorders in ADHD.Based on the fMRI data of ADHD children with visual attention capture task and normal control children,the changes in brain function network characteristics of ADHD children under specific stimuli are studied,as follows:First,collect the fMRI data of ADHD children and normal children under the task of visual attention capture,and explore the difference of brain function modal characteristics between ADHD children and normal children.First construct the visual area brain function network,and then obtain the characteristic indicators of the visual area brain function network,including:degree distribution,average shortest path,betweenness,etc.,and compare and analyze with the traditional whole brain network;at the same time,use support vector machine(SVM)And other classifiers classify feature indicators to distinguish ADHD children from normal children.Secondly,after preliminary analysis of the differences in the specific brain function sub-networks of the two groups of children,further study the characteristics of the relationship between the brain function areas of the subjects during the task,and in-depth exploration of the brain area functions of ADHD children from the perspective of network overlapping modules Sexual synergy.The non-negative matrix factorization(NMF)method based on Bayesian prior is used to preset different values of the number of overlapping communities,and the overlapping community detection is performed on the brain function network constructed by the two sets of data.By analyzing the detection results of overlapping communities,explore the consistency and difference of the overlapping communities of AHHD children and normal children’s brain function network under task mode.On the other hand,the brain function network of ADHD children is real-time and dynamic,and it is necessary to adopt methods to improve the time resolution of the recognition brain network in order to deeply explore the dynamic brain activity state mode of ADHD children.On the basis of the above-mentioned research on the differences in brain modal features of ADHD children and the features of overlapping modules in various brain regions,the analysis is carried out in combination with an adaptive brain state extraction algorithm.A new type of spatial standard deviation(SSD)algorithm is used to dynamically divide the entire time period,refine the dynamic time window brain network with independent brain activities,to improve the time resolution of the brain network in the recognition task;at the same time,through aggregation Class algorithm(K-means++),realizes the separation of the brain activity pattern state during the task process,so as to obtain the core brain node characteristics of each modal brain function network.The above research contents were carried out from the perspectives of exploring the changes in brain function network characteristics of ADHD children,the overlapping module characteristics of brain function areas and improving the time resolution of the brain network in the recognition task state.The results show that the visual area brain function network analysis method is adopted,and the visual area brain is selected.The network index is used as a feature to classify ADHD children and normal children,with a classification accuracy of up to 96.0%.Compared with the traditional method of constructing a whole brain network,the accuracy is improved by about10%,which can better distinguish ADHD children from normal children.In the study of brain function network overlap community detection,it is found that the brain function overlap ratio index of ADHD children is 10.7%,which is slightly lower than that of normal children,indicating that ADHD children have a lower brain function synergy efficiency in the task;at the same time,compared to normal children,The frontal lobe-amygdala-occipital lobe network of the brain function network of ADHD children has abnormal connection.On the other hand,the new method of adaptive brain state extraction theory adopted in the research can effectively obtain the dynamic brain function connection characteristics of ADHD children under the task,and divide the brain activity in the task into 4 different states,with differences between states Obvious;compared with the traditional fixed time window method,the brain network constructed by the new method used in the study,the network modularity Q value is increased by nearly 20% compared with the traditional method.In short,the above-mentioned studies can dig deeply into the node association and connection strength characteristics of the ADHD children’s brain function network under the task state from different angles,and the experimental results are good.The subject research has certain help to distinguish the brain network of ADHD children from normal children,provides a certain theoretical basis for the clinical adjuvant treatment of ADHD children,and is conducive to the auxiliary diagnosis and follow-up treatment of ADHD children.