Johnston Peter R
School of Biomolecular and Physical Sciences, Griffith University, Nathan, QLD, Australia.
Comput Methods Biomech Biomed Engin. 2010;13(2):157-70. doi: 10.1080/10255840903067072.
This paper presents an implementation of the finite volume method with the aim of studying subendocardial ischaemia during the ST segment. In this implementation, based on hexahedral finite volumes, each quadrilateral sub-face is split into two triangles to improve the accuracy of the numerical integration in complex geometries and when fibre rotation is included. The numerical method is validated against previously published solutions obtained from slab and cylindrical models of the left ventricle with subendocardial ischaemia and no fibre rotation. Epicardial potential distributions are then obtained for a half-ellipsoid model of the left ventricle. In this case it is shown that for isotropic cardiac tissue the degree of subendocardial ischaemia does not affect the epicardial potential distribution, which is consistent with previous findings from analytical studies in simpler geometries. The paper also considers the behaviour of various preconditioners for solving numerically the resulting system of algebraic equations resulting from the implementation of the finite volume method. It is observed that each geometry considered has its own optimal preconditioner.
本文介绍了有限体积法的一种实现方式,旨在研究ST段期间的心内膜下缺血情况。在该实现方式中,基于六面体有限体积,每个四边形子面被分割为两个三角形,以提高复杂几何形状以及包含纤维旋转时数值积分的精度。该数值方法针对先前发表的、从存在心内膜下缺血且无纤维旋转的左心室平板和圆柱模型获得的解进行了验证。然后获得了左心室半椭球体模型的心外膜电位分布。在这种情况下表明,对于各向同性心肌组织,心内膜下缺血程度不影响心外膜电位分布,这与先前在更简单几何形状的分析研究结果一致。本文还考虑了各种预条件器在数值求解有限体积法实现过程中产生的代数方程组时的性能。观察到所考虑的每种几何形状都有其自身的最优预条件器。