CFC2023

Keynote

Computational analysis of turbulent flow structures in the left ventricle of the heart using patient-specific data

  • Kronborg, Joel (KTH Royal Institute of Technology)
  • Hoffman, Johan (KTH Royal Institute of Technology)

Please login to view abstract download link

The blood flow in the left ventricle is characterized by the change in ventricle volume over the heart cycle, the structure of the inner ventricular wall, blood characteristics, and the associated inflow and outflow regulated by the mitral and aortic valves, respectively. In systole the ventricle contracts which leads to an increase in blood pressure, which in a healthy heart closes the mitral valve to the left atrium and opens the aortic valve for the blood to flow into the aorta. Analogously, in diastole, the heart relaxation phase, the ventricle expands, blood pressure drops, the aortic valve is closed and the mitral valve opens to let blood flow into the ventricle from the left atrium. The inflow in diastole has two phases, the E-wave followed by the A-wave, each representing jet flow which breaks down into turbulent flow in the ventricle. Heart valve disease may lead to restricted valve openings, or leaking valves, that changes the blood flow in the ventricle. Similarly, clinical interventions may lead to changes in the ventricular blood flow. Blood cells and endocardial cells (which form the innermost layer of the ventricle wall) respond to mechanical stresses in the blood flow, and changes in the mechanical stresses may be a risk factor e.g. for thrombosis. In this work we use patient-specific data to build a computational model of the blood flow in the left ventricle of the heart, in which we analyse the mechanical stresses in the blood flow, with a focus on the turbulent flow structures in diastole. The computational model is based on solving the Navier-Stokes equations with a finite element method on a deforming mesh, and the mechanical stresses are analyzed using the triple decomposition of the velocity gradient of the flow. In previous work we have established a clinical pathway for patient-specific simulations through which we analyzed turbulent flow structures in a finite element model where the mitral valve was modelled as a time varying inflow boundary condition. We here extend this framework to a more realistic simulation using a fluid-structure interaction model of the mitral valve based a unified continuum approach.