基于自适应细化网格的初期血栓形成计算模型

A computational model of primary thrombosis based on adaptive mesh refinement

随着计算机应用领域的扩大,使用建立血栓形成机制计算模型或计算机软件模拟血栓形成过程逐渐成为血栓研究的热点。本文提出了一个基于笛卡尔自适应细化网格的混合计算模型用于模拟初期血栓的形成。计算模型中利用水平集方法的运动界面追踪功能对初期血栓的形态和生长过程进行建模,采用网格细化和局部水平集方法实现建模网格的自适应细化,提高水平集方法的计算效率。计算模型被应用于狭窄血管中初期血栓形成的仿真实验,仿真结果显示的血管狭窄对血小板聚集的影响与相关研究的动物实验结果一致;并且,自适应细化网格有效降低了实验过程中因界面形态复杂化增加的网格数量。

With the development of computer applications, generation or simulation of thrombosis mechanism via computational models or computer software have become hotspots of the research on thrombus. In this paper, a hybrid computational model based on Descartes adaptive mesh refinement is proposed to simulate primary thrombus formation.In this model, the shape and growth of thrombus are modeled by using free surface tracking function of the level set method. The computational efficiency of level set method is improved by realizing adaptive mesh refinement of models with the mesh refinement and the local level set methods. The computational model is applied to the simulation experiments of primary thrombus formation in narrow vascular stenosis. The results of the simulation experiments are consistent with those of relevant animal experiments in the effect of stenosis on platelets aggregation. And the adaptive mesh refinement reduces the increasing grid quantity effectively due to complex interface pattern during experiments.