Université de Bordeaux, Institut des Neurosciences Cognitives et Intégratives d'Aquitaine, UMR5287 Bordeaux, France ; CNRS, Institut des Neurosciences Cognitives et Intégratives d'Aquitaine, UMR5287 Bordeaux, France.
CEA, LETI Grenoble, France.
Front Comput Neurosci. 2014 Feb 19;8:13. doi: 10.3389/fncom.2014.00013. eCollection 2014.
Nowadays, high-density microelectrode arrays provide unprecedented possibilities to precisely activate spatially well-controlled central nervous system (CNS) areas. However, this requires optimizing stimulating devices, which in turn requires a good understanding of the effects of microstimulation on cells and tissues. In this context, modeling approaches provide flexible ways to predict the outcome of electrical stimulation in terms of CNS activation. In this paper, we present state-of-the-art modeling methods with sufficient details to allow the reader to rapidly build numerical models of neuronal extracellular microstimulation. These include (1) the computation of the electrical potential field created by the stimulation in the tissue, and (2) the response of a target neuron to this field. Two main approaches are described: First we describe the classical hybrid approach that combines the finite element modeling of the potential field with the calculation of the neuron's response in a cable equation framework (compartmentalized neuron models). Then, we present a "whole finite element" approach allowing the simultaneous calculation of the extracellular and intracellular potentials, by representing the neuronal membrane with a thin-film approximation. This approach was previously introduced in the frame of neural recording, but has never been implemented to determine the effect of extracellular stimulation on the neural response at a sub-compartment level. Here, we show on an example that the latter modeling scheme can reveal important sub-compartment behavior of the neural membrane that cannot be resolved using the hybrid approach. The goal of this paper is also to describe in detail the practical implementation of these methods to allow the reader to easily build new models using standard software packages. These modeling paradigms, depending on the situation, should help build more efficient high-density neural prostheses for CNS rehabilitation.
如今，高密度微电极阵列为精确激活空间上受控的中枢神经系统 (CNS) 区域提供了前所未有的可能性。然而，这需要优化刺激设备，而这反过来又需要很好地理解微刺激对细胞和组织的影响。在这种情况下，建模方法为预测电刺激对 CNS 激活的影响提供了灵活的方法。在本文中，我们介绍了最先进的建模方法，提供了足够的细节，使读者能够快速构建神经元细胞外微刺激的数值模型。这些方法包括：(1) 计算组织中刺激产生的电场，以及 (2) 目标神经元对该场的响应。描述了两种主要方法：首先，我们描述了经典的混合方法，该方法将电场的有限元建模与电缆方程框架中的神经元响应计算相结合（分区神经元模型）。然后，我们提出了一种“整体有限元”方法，通过用薄膜近似表示神经元膜，同时计算细胞外和细胞内的电位。这种方法以前是在神经记录的框架中提出的，但从未被用于确定细胞外刺激对亚区水平的神经响应的影响。在这里，我们通过一个例子表明，后一种建模方案可以揭示神经膜的重要亚区行为，而混合方法无法解决这些行为。本文的目的还在于详细描述这些方法的实际实现，以便读者能够使用标准软件包轻松构建新模型。这些建模范例应根据具体情况帮助构建更有效的高密度神经假体，以用于 CNS 康复。
