基于神经振荡模型分析的母体免疫激活诱导子代孤独症风险预测研究

孤独症谱系障碍（ASD）是一种常见的神经发育障碍疾病，通常发病于3岁之前的儿童，患病率高达1/45，严重损伤了患者的健康和生活质量，给患者家庭、社会都造成了极大负担。母体传染病和自身免疫性疾病患病率增加等因素诱导的母体免疫激活是导致目前孤独症发病率增加的重要原因。目前没有预防和根治孤独症的有效方法，但早期干预和治疗可以有效改善患者的预后。因此，寻找孤独症早期生物标志物成为早期筛查孤独症的关键。本项目基于国内外研究现状分析和课题组前期基础，认为神经振荡模式异常很可能是孤独症早期诊断的重要脑生物标志物，可作为预警子代孤独症异常行为的重要依据。本项目将在母体免疫激活的高风险模型中，分析不同时期子代神经振荡计算结果，并将其与子代的社交行为障碍、偏执行为等认知异常和突触可塑性损伤结果进行相关性分析。构建新的分析算法，利用神经振荡模型分析结果对母体免疫激活模型下子代孤独症异常行为进行早期风险预测。

Autism spectrum disorder (ASD) is a common neurodevelopmental disorder, which usually occurs in children before the age of three. The prevalence rate is as high as 1 / 45, indicating that ASD seriously damages the health and quality of life of the patients, and causes great burden to the patient families and the society. Maternal immune activation, induced by maternal infectious diseases and maternal autoimmune diseases, is a major cause of the increase in the autism incidence rate in recent years. At present, there is almost no way to prevent and treat autism in clinical practice, but intervention in the early-stage of ASD can significantly improve the prognosis of autistic patients. Looking for biomarkers in the early stage of autism is a key to screen autism in the early stage. Based on reviewing research status at home and abroad, and our previous study, one can speculate that the abnormal neural oscillation pattern is likely to be a significant and essential brain biomarker for the early-stage diagnosis of autism. It can be used as an important basis for early warning of the abnormal behavior of children with autism. In the high-risk model of maternal immune activation, we are going to perform the correlation analysis between the data, obtained from the neural oscillatory pattern measurements at different developmental stages, and the data from ASD behavior tests including social behavior disorder, paranoid behavior, other cognitive impairments and synaptic plasticity deficits. A new analysis algorithm will be developed to predict the early risk of the littermate of autism abnormal behavior in the maternal immune activation model by using the results of neural oscillation pattern analysis.