COMPUTATIONAL TOOLS FOR MULTISCALE HEART MODELING

这个子项目是许多研究子项目中利用 资源由NIH/NCRR资助的中心拨款提供。子项目和 调查员(PI)可能从NIH的另一个来源获得了主要资金， 并因此可以在其他清晰的条目中表示。列出的机构是 该中心不一定是调查人员的机构。 这个项目是与亨利克斯博士在杜克大学的计算电生理学小组合作完成的。这个项目的目标是使用足够有效的心脏组织模型来模拟大片组织的行为，甚至整个心脏，但仍然保留与细胞和膜行为及其模拟模型的联系。这类研究最常用的模型使用连续的双区域模型，这种方法平均细胞内和细胞外的空间，形成由膜分隔的连续的交错体积。然而，这种类型的模型没有考虑实际膜的形状和位置，也没有处理细胞是离散实体的事实。为了进一步分析比多曼方法固有的平均效果，更具体地说。为了将平均特性与潜在的组织病理学联系起来，该项目旨在使用细胞尺度上的离散几何和计算模型来模拟心脏组织中去极化前沿的传播。 应用的重点已经从基于少量(30-100个)离散细胞模型的心脏组织的被动电特性发展到现在包括兴奋的传播，仍然同样强调在各种条件下决定激活的速度和有效性的参数。主要的医学目标仍然是开发一种方法，允许将来自正常组织和病理组织的组织学细胞水平的特征直接结合到模拟框架中，该模拟框架可以生成驱动Bidomain公式的相关中尺度参数。通过这种方式，我们可以考虑离子浓度、细胞形状、缝隙连接电导率、细胞外体积等一系列病理生理学变化，这些因素既与动作电位传播的良性波动有关，也与危及生命的心律失常有关。 我们的建模方法非常适合于这一领域的参数研究，因为它的物理规模很小，而且对相关参数的控制程度很高。我们能够在通常用于单细胞模型的尺度上研究组织制备，因此可以观察到细胞和细胞膜以及细胞基质和间质空间对兴奋传播的贡献。在这种情况下(可能还有许多其他情况)，孤立细胞的反应将与同一细胞作为合体的一部分(如心肌)的反应截然不同。 我们实现的方法是基于从心肌细胞和组织结构的显微镜研究中获得的图像和参数创建少数细胞的横截面，然后扩展或挤压这些横截面以生成高度逼真的组织三维模型。通过管理边界形状，可以生成与其他类似块互锁的细胞构建块，并且这可以生长到形成所需的大小的组织块，并且仍然计算容易处理。从这些块的网络中，我们创建了非常高分辨率的网格，使用数百万个有限元素来表示连接到组织中的多达100个细胞。然后，有限元方法提供了计算无源电特性和扩展去极化波前的电活性的数值框架。我们将这种方法称为“微域”，因为它明确地包括了表征精细组织行为所必需的微域。

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. This project is a collaboration with Dr. Henriquez' Computational Electrophysiology group at Duke University. This project aims at using models of cardiac tissue that are efficient enough to simulation behavior in large pieces of tissue, even whole hearts, but still retain links to cellular and membrane behavior, and their simulation models. The most commonly used models for this type of study employ continuous bidomain models, an approach which averages out the intra- and extracellular spaces to form continuous interleaved volumes separated by a membrane. This type of model, however, does not take into account the shape and location of the actual membrane nor does it deal with the fact that cells are discrete entities. In order to further analyze the effect of averaging intrinsic to the bidomain approach, and, more specifically. to tie the averaged properties to the underlying tissue pathology, the project aims at using discrete geometric and computational models at a cellular scale to perform simulations of the propagation of the depolarization front in cardiac tissue. The application focus has progressed from studying the passive electrical properties of cardiac tissue based on models of small numbers (30-100) of discrete cells to now include the spread of excitation, still with the same emphasis on the parameters that determine the speed and effectiveness of activation under a variety of conditions. The main medical goals remains to develop an approach that allows for the direct incorporation of cellular level features from histology of both normal and pathological tissues into a simulation framework that can generate the associated meso-scale parameters that drive the bidomain formulation. In this way, we can take into account a host of pathophysiological changes in ionic concentrations, cell shape, gap junction conductivity, extracellular volume, etc., factors that are linked to both benign fluctuations in propagation of the action potential but also to life threatening arrhythmias. Our modeling approach is uniquely suited to parameter studies in this domain simply because of its small physical scale and the level of control over relevant parameters. We are able to study a tissue preparation at the scale typically employed for models of single cells and thus can observe contributions from both the cells and cell membrane as well as the cell matrix and interstitial spaces to the spread of excitation. The response of an isolated cell in this context (and probably many other contexts) will be quite different from the response of that same cell when it is part of a syncytium like the myocardium. The approach we have implemented is based on creating cross sections of a few cells from images and parameters from microscopy studies of cardiac cell and tissue structure and then expanding or extruding these cross sections to generate highly realistic three-dimensional models of the tissue. By managing the boundary shape, it is possible to generate building blocks of cells that interlock with other similar blocks and this can grow to form as large a tissue piece as is required and still computational tractable. From these networks of blocks, we create very high resolution meshes, using millions of finite elements to represent up to 100 cells linked into a tissue. The finite element method then provides the numerical framework to compute passive electrical characteristics and also electrical activity of the spreading depolarization wavefront. We have denoted this approach as "microdomain" as it includes explicitly the microscopic domains that are essential to characterize fine scale tissue behavior.