Estimating microscopic biophysical information from macroscopic neuroimaging data via the inversion of neural-glial mass models

近年来，神经成像领域的进展使得创建基于血流动力学和基于电生理学的功能连接图以及基于弥散核磁共振的结构连接图成为可能。然而，我们面临的主要挑战不仅是再现真实的大脑连接和动态，而且还包括识别这些动态和网络结构的重要特征和机制，这些特征和机制对大脑功能至关重要。该研究计划的主要目标是发展大规模的理论大脑模型，以弥合大脑结构和功能之间的差距，并确定经验观察到的动态背后的关键网络特性。更具体地说，这一建议的重点是建立一个由多个相互关联的神经胶质团块模型（NGMMs）构建的大脑活动的宏观生成模型，这将在大脑的微观和宏观空间尺度上建立桥梁。因此，通过从宏观数据估计这个模型，我们可以推断出微观数据，如兴奋性和抑制性突触活动，这对于理解大脑信息处理的机制是必不可少的

Progress in the neuroimaging field in recent years has made it possible to create maps of hemodynamic-based and electrophysiology-based functional connectivity, as well as diffusion MRI-based structural connectivity maps. However, a major challenge we face is not only the reproduction of realistic brain connectivity and dynamics, but also the identification of important features and mechanism of these dynamics and of the network structure that are critical to brain function. The broad goal of this research program is the development of large-scale theoretical brain models to bridge the gap between structure and function in the brain, and identify key network properties underlying the empirically observed dynamics. More specifically, this proposal is focused on the development of a macroscopic generative model of brain activity built from multiple interconnected neural-glial mass models (NGMMs), which will bridge the microscopic and macroscopic spatial scales in the brain. Thus, by estimating this model from macroscopic data we can infer microscopic data such as excitatory and inhibitory synaptic activity, which is indispensable for understanding the mechanism underlying brain information processing.**