Computational Modeling of Epileptic Activity Using a Hybrid Compartment Technique

描述（由申请人提供）：试图理解影响癫痫的变量的巨大参数空间的一个有用的工具是计算建模。计算机建模允许用户快速更改描述性变量，如离子通道和突触属性、连接和突触消息传递的时间特征。各种细胞几何形状和神经元数量已经被实现，以努力描述连接细胞网络活动的各种特征，所有这些都是为了重现各种实验结果。我们建议在当前的努力中比较和校准来自单室霍奇金-赫胥黎模型和更复杂的多室细胞模型的活动，在这两种情况下，细胞都以一种结构逼真的方式排列和连接，以模拟新皮层。此外，更简单的单室模型将被用作更精确的多室模型的周围边界区域，以努力扩大模型区域，并仍然提供有意义的电生理相关。我们将研究从模型中作为癫痫焦点的区域产生的具体爆发活动。这项比较工作将用于实现以下目标：目标1)使用混合隔室组织测量校准的结构逼真神经网络中的临界传播率，并对这种传播作为第II/III层和第V层锥体细胞系统中连接的函数进行敏感性分析。目的2)在该模型的背景下，通过分级、层流特异性药物阻断来测量致痫发作的潜在活动阈值，并与实验进行比较。目的3)将我们的模拟研究结果与新的硬膜下网格设计的记录进行比较，该设计包含用于局部场电位测量的标准电极元件和微线。这些研究的目的是以定量的方式提供有关癫痫发作的空间扩散和时间演变作为邻近或内在病理的功能的信息。我们计划模拟相对较大的皮质区域（6.4 mm X 6.4 mm），同时检查非常高精度的较小区域（250 <m X 250 <m）的活动。

DESCRIPTION (provided by applicant): One useful tool in trying to understand the vast parameter space of variables influencing epilepsy is computational modeling. Computer modeling allows the user to quickly change descriptive variables such as ion channels and synaptic properties, connections, and the temporal characteristics of synaptic messaging. A variety of cell geometries and numbers of neurons have been implemented in an effort to describe various features of the activities of linked cellular networks all in an attempt to reproduce various experimental findings. We propose in this current effort to compare and calibrate the activity derived from a single compartment Hodgkin-Huxley based model with a more sophisticated multicompartmental cellular model where in both cases the cells are arranged and connected in an architecturally realistic fashion to mimic neocortex. Additionally, the simpler single compartment model will be used as the surrounding boundary region for the more precise multicompartment model in an effort to expand the modeled region and still provide meaningful electrophysiologic correlates. We will examine specifically bursting activity derived from a region of the model which acts as an epileptic focus. This comparative work will be used to pursue the following aims: Aim 1) To measure the rate of ictal spread within a calibrated architecturally realistic neuronal network using a hybrid compartment organization, and to perform a sensitivity analysis of this spread as a function of connectivities in the Layer II/III and Layer V pyramidal cell systems. Aim 2) To measure the underlying activity threshold for ictal onset through graded, laminar specific pharmacologic blockade in the context of this model for comparison with experiment. Aim 3) To compare the results obtained in our simulation studies with recordings obtained from a new subdural grid design incorporating standard electrode elements and microwires for local field potential measurements. The goal of these studies is to provide information in a quantitative manner on seizure spatial spread and temporal evolution as a function of neighboring or intrinsic pathology. We plan to model relatively large regions of cortex (6.4 mm X 6.4 mm), while examining the activity of a very high precision smaller (250 <m X 250 <m) representation. PUBLIC HEALTH RELEVANCE: This grant describes the development of a computer model that accurately portrays seizure evolution in a realistic cortical architecture. This would provide an additional tool in understanding seizure dynamics and the application of new drug or stimulation therapies. Estimates of seizure origin, rate of spread and the volume of tissue involved in pathologic behavior will aid in understanding the disruptive effects of epilepsy.