A model reduction approach for the variational estimation of vascular compliance by solving an inverse fluid-structure interaction problem
A model reduction approach for the variational estimation of vascular compliance by solving an inverse fluid-structure interaction problem
复制标题
DOI：
10.1088/0266-5611/30/5/055006
复制
发表时间：
2014-05-01
期刊：
影响因子：
2.1
通讯作者：
Veneziani, Alessandro
中科院分区：
文献类型：
--
作者：
Bertagna, Luca;Veneziani, Alessandro
科学计算已逐渐成为心血管疾病研究的重要工具。基于数值模拟的定量分析的作用已经从“概念证明”转移到患者特定的调查，这要归功于成像和计算工具之间的强大集成。然而，除了个别的几何形状，数值模型需要的知识，几乎没有从测量，特别是在体内检索的参数。出于这个原因，最近心血管数学考虑数据同化程序，用于从测量和图像中提取患者特定参数的知识。在本文中，我们特别考虑血管顺应性的量化，即量化动脉壁在血液应力下变形的趋势的参数。继以前的文件，其中提出了变分数据同化程序，基于解决逆流体-结构相互作用问题，在这里，我们考虑模型降阶技术的基础上适当的正交分解方法来完成的反问题的解决方案，在计算上有效的方式。
Scientific computing has progressively become an important tool for research in cardiovascular diseases. The role of quantitative analyses based on numerical simulations has moved from 'proofs of concept' to patient-specific investigations, thanks to a strong integration between imaging and computational tools. However, beyond individual geometries, numerical models require the knowledge of parameters that are barely retrieved from measurements, especially in vivo. For this reason, recently cardiovascular mathematics considered data assimilation procedures for extracting the knowledge of patient-specific parameters from measures and images. In this paper, we consider specifically the quantification of vascular compliance, i.e. the parameter quantifying the tendency of arterial walls to deform under blood stress. Following up a previous paper, where a variational data assimilation procedure was proposed, based on solving an inverse fluid-structure interaction problem, here we consider model reduction techniques based on a proper orthogonal decomposition approach to accomplish the solution of the inverse problem in a computationally efficient way.