Information Flow-Based Brain Network Analysis of Healthy and Epileptic Syndromes Children

Analyzing the trends in brain information flow of children with epilepsy and normal children can provide a theoretical basis for the pathogenesis of childhood epilepsy and brain growth and development. The paper studied the electroencephalogram (EEG) recorded during sleep in children aged 0 ~ 14y, including 29 healthy children and 32 children with epilepsy syndrome. The directed transfer function (DTF) was used to calculate the correlation characteristics between EEG channels, which were then used to construct the connectivity matrix. To reduce individual differences, generalized sequential forward selection (GSFS) was used for feature screening. A group-level connectivity matrix was constructed, representing the connectivity and differential brain networks across brain regions. Finally, directed graph theory features were used to assess the speed and reliability of information flow. Through comparative analysis of developmental trends and information flow-related features, the main findings include: (1) The speed and reliability of the flow of information between the two groups show similar growth and development trends, albeit to different degrees. (2) Abnormal developmental trends were observed in the age group of 5 ~ 8y, which may be attributed to the prevalence of absence seizures in epileptic children in this age group, often without noticeable spasms. (3) Brain regions show a bidirectional flow of information between central and parietal regions, and between frontal and temporal regions, across all age groups.