Predicting the developmental trajectories of cognitive and motor dimensions from preterm neonatal vocalizations

摘要 在美国，每10个婴儿中就有一个是早产的，这导致了较高的长期早产率 对健康造成负面影响。预计在不久的将来，早产率将会更高。 因孕产妇冠状病毒病2019年(新冠肺炎)感染。虽然早产不会 必然导致发育性神经精神障碍，它与更高的 自闭症谱系障碍(ASD)1-9，智力残疾10，11，注意力缺陷/多动障碍 (ADHD)2、12、学习障碍13、脑性瘫痪13以及认知、社交和运动方面的延迟 发展2、4、8、10、14-16。与早产相关的发育风险和不确定性 可能会让照顾者对未来感到困惑、压力和焦虑。此外， 早产儿认知、社会和运动发展轨迹的异质性 混淆了准确、及早发现发展问题和及早实施 治疗性干预17-21。由于某些形式的早期干预改善了婴儿的预后 患有(或高风险)发育性神经精神障碍22-28，客观的、定量的 预测(或改进)早产儿发育轨迹的方法 是迫切需要的。一些研究小组试图使用计算模型来 区分29-32岁早产儿和足月儿的哭声。然而，可以预测的计算模型 早产儿的不同认知、社会和运动发展轨迹 不存在。此外，早产如何改变认知、社交和运动的神经基础 人们对发育轨迹知之甚少。根据我们的初步数据显示， 杏仁核的体积受到一种与社会缺陷有关的基因变体的选择性影响 我们假设早产也会改变认知、社交和运动发育 通过相关脑结构的变化从新生儿到成人期早期的轨迹 认知能力、社交能力和运动能力。为了验证这一假设，我们将利用我们的专业知识 计算特征选择，使用新生儿发声的变量和不同的 小鼠大脑中最能解释认知、社交和运动能力差异的区域。一个 积极的结果将提供急需的预测模型，以进一步探索机制基础 对小鼠模型的认知、社交和运动发育的影响。这应该使调查人员能够 进一步评估不同认知、社会和运动的机制和结构基础 环境和遗传风险因素的临床前模型中的轨迹。

Summary One in every 10 babies is born prematurely in the US, and this contributes to high rates of long-term negative health consequences. An even higher rate of preterm births is expected in the near future due to maternal coronavirus disease 2019 (COVID-19) infection. While premature birth does not necessarily result in developmental neuropsychiatric disorders, it is associated with elevated rates of autism spectrum disorder (ASD)1-9, intellectual disability10, 11, attention-deficit/hyperactivity disorder (ADHD)2, 12, learning disabilities13, cerebral palsy13, and delays in cognitive, social, and motor development2, 4, 8, 10, 14-16. The developmental risks and uncertainty associated with premature birth may overwhelm caregivers with confusion, stress, and anxiety about the future. Moreover, the heterogeneity of cognitive, social, and motor developmental trajectories among preterm infants confounds the accurate, early detection of developmental issues and the early implementation of therapeutic interventions17-21. As some forms of early intervention improve the prognoses of infants with (or at high risk of) developmental neuropsychiatric disorders22-28, an objective, quantitative method of predicting (or improving the prediction of) the developmental trajectories of preterm infants is urgently needed. A number of research groups have attempted to use computational models to differentiate the cries of preterm and term infants29-32. However, computational models that can predict the heterogeneous cognitive, social, and motor developmental trajectories among preterm infants do not exist. Moreover, the neural basis of how preterm birth alters the cognitive, social, and motor developmental trajectories is poorly understood. Based on our preliminary data, which showed that the volume of the amygdala is selectively impacted by a gene variant that is linked to social deficits in mice, we hypothesize that preterm birth also alters the cognitive, social, and motor developmental trajectories from the neonatal to early adult periods via variable alterations of brain structures relevant to cognitive, social, and motor capacities. To test this hypothesis, we will leverage our expertise in computational feature selection, using variables of neonatal vocalization and volumes of various mouse brain regions that best account for variances in cognitive, social, and motor capacities. A positive outcome will provide much-needed predictive models to further explore the mechanistic bases of cognitive, social, and motor development in mouse models. This should enable investigators to further evaluate the mechanistic, structural bases for heterogeneous cognitive, social, and motor trajectories in preclinical models of environmental and genetic risk factors.