Near infrared spectroscopy of deep brain structures in the neonate

包括功能性近红外光谱（fNIRS）和功能性磁共振成像（fMRI）在内的对大脑血流动力学变化敏感的功能性神经成像技术发展迅速，为深入了解成人大脑功能连接提供了帮助。然而，使用fNIRS和fMRI来研究新生儿的大脑发育是有限的。本实验旨在利用fMRI作为验证方法，推进当前的fNIRS分析方法，以便使用fNIRS研究新生儿的功能性脑连接。fNIRS是一种新兴的新生儿脑成像技术，用于测量皮质表面的血流动力学变化。功能磁共振成像是测量新生儿脑部血流动力学的首选方法，因为它能够检测整个大脑的T2*信号。虽然fNIRS为评估新生儿种群提供了许多优势，包括其高时间分辨率（~100ms）、低成本和便携性，但由于获取皮质表面下血流动力学变化的技术挑战，其用于皮质下-皮质连接研究的应用受到限制。为了定位新生儿大脑深部结构的血流动力学变化，提出了一种新的机器学习算法，将fNIRS扩展到新生儿新出现的认知和运动系统中皮层下和皮层血流动力学的测量。为了实现利用fNIRS研究新生儿皮层下-皮层连通性的长期目标，我们需要(1)开发一种预测方法来推断皮层下血流动力学；(2)利用fMRI验证基于已建立的皮质-皮质下连接方法的预测模型；(3)将预测方法应用于fNIRS新生儿血流动力学数据。fNIRS系统的发展将为研究涉及特定大脑系统的早期网络连接如何促进特定领域的认知成熟提供新的见解。通过更好地了解大脑成熟的个体差异如何预测大脑连接的改变，这一知识可能为更有效的早期干预教学方法奠定基础，这些方法可以针对特定功能大脑系统发展的关键时期进行校准。

Functional neuroimaging technologies that are sensitive to haemodynamic changes in the brain including functional near infrared spectroscopy (fNIRS) and functional magnetic resonance imaging (fMRI) have advanced rapidly, providing insight into adult brain functional connectivity. However, using fNIRS and fMRI to study newborn brain development has been limited. The proposed set of experiments seeks to advance current fNIRS analytic methods, using fMRI as a validation method, in order to use fNIRS to study functional brain connectivity in newborns. fNIRS is an emerging neonatal brain-imaging technique that measures haemodynamic changes at the cortical surface. fMRI is a preferred method for measuring neonatal brain hemodynamics due to its ability to detect T2* signal throughout the brain. While fNIRS offers a number of advantages for assessing neonatal populations including its high temporal resolution (~100ms), low-cost and portability, its use for the study of subcortical-cortical connectivity has been limited due the technical challenges of obtaining haemodynamic changes below the cortical surface. In order to localize haemodynamic changes in deep brain structures in the neonatal brain, a novel machine-learning algorithm is proposed to expand upon applications of fNIRS to the measurement of subcortical and cortical haemodynamics underlying emerging cognitive and motor systems in the newborn. To achieve the long-term goal of utilizing fNIRS for the study of neonatal subcortical-cortical connectivity, we need to (1) develop a prediction method to extrapolate subcortical haemodynamics; (2) validate the prediction model based on established methods of cortical-subcortical connectivity using fMRI; and (3) apply the prediction method to fNIRS neonatal haemodynamic data. The development of fNIRS systems will offer new insights into how early network connectivity involving specific brain systems contributes to cognitive maturation in particular domains. By providing a better understanding of how individual differences in brain maturation may predict alterations in brain connectivity, this knowledge may lay the foundation for more effective early-intervention teaching methods calibrated to critical periods in the development of specific functional brain systems.