Use of Computational Fluid Dynamics and Fluid-Structure Interaction to Simulate the Cardiovascular System

摘要

The aim of this session is to discuss how computational methods can be used to investigate the fluid dynamics of the cardiovascular system. The session will focus on the heart and its natural valves. This is important as heart valve failure leads to fluid dynamics which are detrimental to the cardiovascular system that require surgery for correction. In the United Kingdom, the British Heart Foundation has estimated that cardiovascular disease is the cause of one out of every three deaths. Computational methods provide a non-invasive method to investigate the cardiovascular system. In particular computational fluid dynamics and fluid-structure interactions are useful as they can be used to predict fluid dynamics. The former enables shear stresses to be predicted, while the latter enables the induced stress and deformation to be predicted. Such predictions are useful when either understanding pathology or investigating surgical repair or replacement. However, use of computational fluid dynamics on its own ignores moving structures (e.g. heart valves) which can lead to inaccurate predictions of stress and flow. Fluid-structure interaction simulations can solve some of these limitations, for example, through the use of an Arbitrary-Lagrange Euler moving mesh. Its use, though, introduces further technical challenges such as contact modelling or including relatively large deformations in models (e.g. 10-20% strain). Further challenges include selection of representative geometry, boundary conditions, material properties for blood (e.g. viscosity) and soft tissues (e.g. heart valve Young's modulus). In the case of geometry or boundary conditions a major problem is accurate measurement. Defining suitable material properties presents the problem of large variability associated with natural tissues (e.g. heart valves). Currently in our laboratory we have been using both computational fluid dynamics and fluid-structure interaction to aid understanding of the cardiovascular system focusing on heart valves but also aid diagnosis and investigate failure of heart valves. The session includes: 1. background to the cardiovascular system, in particular the heart and its natural valves; 2. description of the basic physiology involved in the heart and heart valve fluid dynamics; 3. discussion of the benefits and limitations in using computational fluid dynamics and fluid-structure interaction to investigate the cardiovascular system; 4. discussion of the technical challenges that simulating the cardiovascular system presents; 5. current findings from our studies on computational methods in investigating the cardiovascular system and its fluid dynamics.