Research On Infant Brain Development Based On Multi-modal Neuroimaging

The structure,functional connectivity and cognitive level of the infant brain undergo a dynamic and rapid development period after birth.Studying the brain development during this period will help to understand the rules of early brain development and brain work mechanisms in infants,and then lay a solid foundation for the early diagnosis of brain developmental diseases.Advances in magnetic resonance imaging,computer technology and neuroimaging computing have provided great opportunities for studying infant brain developmental patterns.In this thesis,based on multi-modal infant magnetic resonance datasets,we investigated early developmental patterns of infant brains through neuroimaging computing and statistical analysis.The main work and innovations of the thesis are given as following:1.Research on infant brain tissue volume development based on multi-modal neuroimaging.Exploring the development of brain tissue volume and its pattern differences helps to understand the brain tissue maturation in normal infants.Some neurodevelopmental disorders result in abnormal changes in brain tissue volume,so a better understanding of the pattern of normal infant brain tissue volume development will also helps to investigate the cause of the disease and aids in early clinical diagnosis.Based on this,a controlled analysis of the Eastern and Western infant brains using magnetic resonance imaging was conducted.The developmental patterns and differences of infant brain tissue volumes from the two populations were investigated by neuroimaging computing,statistical analysis and linear regression model analysis.The results of the study showed that the volume of gray matter,and its ratio relative to the sum of gray and white matter volumes were significantly different between Eastern infants and Western ones,with the gray matter volume and its ratio of Eastern infants larger than the Western.The results of the linear regression model analysis show that the developmental patterns of white matter volume,and the gray matter volume ratio were significantly different between Eastern and Western infants during early development.Western infants show a faster white matter growth rate,whereas Eastern infants have a faster growth rate in gray matter volume ratio.2.Research on biomarkers of infant brain developmental diseases based on multi-modal neuroimaging.Autism is one of the common disorders of brain development in infants.The etiology is complex and there are no converging findings or reliable biomarkers for early clinical screening and diagnosis.Therefore,it is necessary to explore and study the potential biomarkers systematically.Thus,the thesis conducts a statistical analysis of current studies of infant autism based on different modality neuroimaging data,and explores possible neuroimaging-derived biomarkers.The relationship between the total brain tissue volume and autism in infants was investigated using Meta-analysis for structural MRI data.Meta-analysis showed that the total brain tissue volume was significantly larger in infants with autism.The total brain tissue volume has a significant statistical association with autism,which can be used as a biomarker for the early diagnosis.Based on the functional MRI and DTI data,a statistical analysis method was employed to study the possible biomarkers.The results of the analysis based on f MRI data suggest that infants with autism already show abnormal functional connectivity changes in language,social and other brain regions at an early age.The results of the analysis based on DTI data showed abnormal changes in white matter fiber tracts and its structure,and white matter network efficiency in infants with autism.Therefore,abnormal changes in total brain tissue volume,functional connectivity between different brain regions,white matter structure,fiber bundles and network efficiency are powerful neuroimaging biomarkers for early diagnosis of infant with autism.