| Autism is a highly heterogeneous neurodevelopmental disorder with individual differences in the severity of symptoms.At present,the diagnosis and evaluation of autism still has a certain degree of subjectivity.Studies have shown that structural magnetic resonance morphology can provide an objective basis for the diagnosis and evaluation of autism,and many studies have found that brain development in autism is related to abnormalities in brain structure.Therefore,exploring the brain development and behavioral characteristics of autism based on structural magnetic resonance images is of great significance for understanding its neurobiological mechanism and establishing an objective diagnosis and evaluation model of autism.Currently,studies on the brain development and behavioral characteristics of autism are mainly based on publicly available brain atlases.However,studies using atlas-based approaches may lack detailed information on the internal structure of brain regions.In this article,we extracted vertex morphological features from structural magnetic resonance imaging data and constructed a more detailed high-resolution brain atlas based on the publicly available standard brain atlas.From the two perspectives of standard brain atlas and high-resolution brain atlas,We explored the brain development of autism by predicting brain age based on morphological features.Additionally,we explored the relationship between brain structure and autism behavioral characteristics by predicting behavioral indicators based on morphological features.Specifically,the research is divided into the following three modules:(1)Construction of a high-resolution brain atlas.In order to refine the existing brain atlas and investigate the impact of the internal structure information of brain regions on the brain development and behavioral characteristics of autism,we established a high-resolution brain atlas based on the morphological features of structural magnetic resonance data.In each brain region of the standard brain atlas,a morphological connectivity network was established based on the morphological features of the vertices.The modularity selection algorithm was used to identify the best community detection algorithm to partition the morphological connectivity network.The segmentation results were then mapped to the standard brain space to complete the construction of the entire highresolution brain atlas.Through this study,each brain region of the standard brain atlas is divided into different numbers of subregions,which can reveal the structural information inside the brain region.The high-resolution brain atlas of this study is the basis for subsequent exploration of the mechanism of autism brain development and prediction of autism behavior.(2)Brain development research for autism.The relationship between brain development and structure in individuals with autism has always been a key area of research.In this study,we established a model for predicting brain age based on the morphological features in the two brain atlases,and explored the differences in brain development between the autism group and the normal control group.The results based on the standard brain atlas showed that five brain regions in the right brain,including the olfactory cortex and lingual gyrus,and five brain regions in the left brain,including the olfactory cortex,inferior frontal gyrus,and parahippocampal gyrus,were significantly different from those in the normal control group(P<0.05).Based on the high-resolution brain atlas,nine brain regions in the right brain,including the temporal lobe,lingual gyrus,and parahippocampal gyrus,and nine brain regions in the left brain,including the parahippocampal gyrus,posterior cingulate gyrus,precuneus,olfactory cortex,fusiform gyrus,and inferior frontal gyrus,were significantly different from those in the normal control group(P<0.05).In summary,the brain development of autism is closely related to the abnormality of its brain structure.Based on the perspective of high-resolution brain atlas,more brain regions that affect the development of autistic brain can be found.(3)Behavioral characteristics research for autism.In order to study the relationship between behavioral characteristics of autism and brain structure,this paper established an Autism Diagnostic Observation Schedule(ADOS)scores prediction model based on the morphological features of the two brain atlases.Standard brain atlas experiments showed that morphological features were only predictive of ADOS stereotypic behavior scores(r = 0.28,P < 0.01).The high-resolution brain atlas experiments showed that among the four behavioral indicators of ADOS,some brain regions showed certain predictive ability,and the number of brain regions in the left brain was more than that in the right brain.In summary,the behavioral performance of autism is correlated with the brain structure of autism.As a more refined brain atlas,high-resolution brain atlas plays an important role in the study of the brain structure mechanism of autism behavior.It is beneficial to clarify the brain mechanism related to autism and assist in the establishment of objective diagnosis and evaluation model for autism.In conclusion,there are significant differences between the brain development of autism and the normal control group,which is related to the abnormal brain structure of autism;the behavioral characteristics of autism(social impairment,communication impairment,stereotyped behavior)and brain structure have relevance.This study proposes a new brain atlas division method,which divides the standard brain regions into finer subregions,establishes models through subregional morphological features,and reveals more brain structure that affect brain development and autistic behavior in autism.Research for autism based on high-resolution brain atlas constructed by morphological features helps better understanding its inner neural and brain mechanisms. |
PDF Full Text Request