February 26, 2021 @ 4:00 pm – 5:00 pm

Join on-line via Zoom:
Friday, February 12, 2021
4:10 p.m. – 4:35 p.m. (EST)

“A Versatile Reduced-Order Model for Patient-Specific Simulations of Blood Flow across the Human Aortic Valve”

(Adviser: Prof. Rajat Mittal)

Simulating blood flow in the human aorta poses several modeling challenges: pulsatile inflow, transitional to turbulent Reynolds numbers, complex dynamic boundary motion, multiscale anatomical features, secondary flow arising from vessel curvature, non-Newtonian rheology, and vascular compliance to name a few. Due to the difficulty associated with addressing all these challenges, most computational models choose to focus on either transvalvular hemodynamics (fluid mechanics) or tissue biomechanics (structural mechanics). Models which focus on hemodynamics typically resolve inflow conditions, important flow features and blood rheology, while simplifying morphological features, valve motion and vessel compliance. However, accurate characterization of the aortic jet shape, tilt and arch curvature are critical to determining hemodynamic loading on the aorta wall, in terms of surface pressure and shear distributions. Therefore, accurate representations of valve kinematics, morphology and ascending aorta features are desirable. In this talk, I will show how a simple reduced-order valve model, developed for an idealized aorta, can be adapted to CT scan-enabled patient-specific anatomy. The model is parametrized using virtual mass and tissue elasticity. Leaflet dynamics are driven by a simplified governing equation accounting for driving pressure forces and restoring tissue elasticity. Specific kinematic features like annulus cross-section, leaflet angular span, flutter and opening shapes can be incorporated using simple geometric transformations. The model is validated against echocardiographic measurements of aortic jet velocity and transvalvular pressure gradient. The most significant advantage of such a valve model is that it permits neglecting leaflet thickness thereby relaxing constraints on time-step size and facilitates generating a large database of patient-specific simulations. Such a database can be used for establishing hemodynamic trends across a patient cohort, surgical planning, and medical device design.

Back to top