| BackgroundInfantile spasms(IS)as one kind of epileptic encephalopathy,it mostly occurs during the first year of life and the prognosis is poor.Nearly half of IS children develop other seizure types after treatment.Even the patients with seizure cessation also present with cognitive disorder.Until nowadays the researches mainly focus on drug therapy,gene deficiency,and metabolic disorders.The pathogenesis is still unclear.Recent years,graph theory is gradually being applied to the imaging data and the results confirmed that brain network exists topologies like small-world,modularity and so on.In addition,the occurrence and progression of neuropsychiatric disorders such as schizophrenia and several kinds of epilepsy are not only related with single region but several regions coordinated with each other.IS is rarely studied by researchers so far with graph theory,functional connecitivity(FC),and amplitude of low-frequency fluctuation(ALFF).Therefore,in this article,we try to detect the pathogenesis in IS at the brain network level with functional magnetic resonance imaging(fMRI)data and find out imaging markers which maybe related with ISPurposeThrough the analysis to brain network with the FC,ALFF and topological properties on IS patients,we hope to find out specific indicators to locate abnormal brain activity precisely.Materials and MethodsIn this study,we collected the MRI data from IS children and controls.Then the whole brain functional network was constructed with graph theory.We analyzed the difference between groups through directly network matrix comparison.We analyzed the difference in global metrics through calculating small world characteristics and network efficiency,calculated the FC strength inner-and inter-modules,and analyzed the difference of nodal metrics between the two groups.In addition,we also detected the difference in BOLD signals within DMN between groups with FC and ALFFResults1.Through direct comparison of FC at the whole brain network with the controls,we achieved three connected components which showed increased FC and five decreased connected components.The increased components mainly distributed between posterior cingulate cortex and different lobes.The significantly decreased components mainly distributed in sensorimotor areas,occipital lobe,and DM-N regions.2.Through the comparison in the whole brain topological properties,we found that the normalized characteristic path length,normalized clustering coefficient,the small-worldness,and local efficiency significantly decreased.However,the global efficiency significantly enhanced than the controls.3.Through the analysis of modular interaction,we found that the strength within DMN and fronto-parietal network significantly decreased.The FC within DMN showed significant negative correlation with seizure frequency.The interaction between occipital network and cingulo-opercular network significantly enhanced and was significantly positively correlated with seizure frequency.4.Through the comparison of the nodal metrics,the results showed that the clustering coefficient of inferior temporal gyrus,precuneus,posterior cingulate and occipital lobe significantly decreased compared with controls.5.Through the comparison of FC and ALFF within DMN regions,the results showed that the BOLD signal within the precuneus,parahippocampal gyri,fusiform gyri,and middle temporal gyrus significantly decreased.ConclusionThrough the analysis to the functional network of IS patients,we found that the IS patients exist abnormal connected components which maybe related with IS.The topological properties of functional brain network were disrupted and became randomized.The synchronicity of functional signals of brain regions within DMN and between occipital and cingulo-opercular network maybe related with epileptic seizure.The regions precuneus and posterior cingutate cortex may be disrupted within DMN.The abnormal metrics might be related with IS.Further analysis of multimode imaging technology will be carried out. |
PDF Full Text Request