Browsing by Subject "FSI"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Computational construction and simulation of novel heart valve and vein valve designs
    (2023-04) Li, Jirong
    As a heart valve substitute with growth potential and improved durability, tissue-engineered heart valves can prevent reinterventions that are currently often needed in children with congenital heart defects. Our group successfully made a pediatric tri-tube/tri-leaflet valved conduit which has shown growth capability in growing lambs. However, the optimal design for valve performance is unknown. To obtain optimal values of valve geometry parameters which can provide efficient guidance for animal tests to save costs and time, we utilized computer-aided simulations to evaluate multiple valve designs. We performed both the valve construction process and this optimization in silico. This is a complex optimization problem due to multiple components of the objective function for valve performance. We thus applied a multi-objective genetic algorithm, which is an elitist strategy, a parameter-free, simple, yet efficient, constraint-handling method used in many applications. A robust finite element method (FEM)-based algorithm for in silico construction of the valve was developed to facilitate optimization for the case of valve closure, identifying the optimal leaflet height and tube diameter that minimizes peak diastolic stress and maximizes coaptation. A strain energy constitutive equation for our novel material, an aligned tissue grown in the lab, was developed based on biaxial testing and implemented in the FEM model. Fluid-structure interaction (FSI) simulation of the optimal design under steady flow was also conducted as a prelude to a next-stage optimization that includes valve dynamic performance metrics. The same methods for the heart valve were also applied to the computational construction and simulation of a stented bileaflet venous valve that includes a sinus design. Steady FSI simulations were used to investigate the effects of the sinus geometry and non-Newtonian blood rheology on the hemodynamics of the novel transcatheter venous valve.
  • Thumbnail Image
    Item
    The Effect of Endografts on the Propagation of Aortic Dissection: A Fluid-Structure Interaction (FSI) Finite-Element Model of Flow through the Descending Thoracic Aorta
    (2017-07) Datta, Ankurita
    Tears in the lining of the aorta are termed aortic dissections. Aortic dissections affect 12,000 new patients per year in the US and are associated with significant morbidity and mortality. The primary management method is an invasive and difficult surgery to replace the dissected portion of the aorta with a synthetic graft. However, recent randomized trials on the use of minimally invasive stent-grafts to treat aortic dissection have proved promising. Though the precise etiology of aortic dissections is unclear, the underlying pathophysiology may be related to hypertension induced injury in an aorta that has lost compliance due to aging. Though stent-grafts can effectively protect the initial location of the dissection, stent-grafts are essentially stiff tubes and after implantation, residual portions of the native aorta proximal and distal to the stent-graft are needed to dissipate systolic pressure. These regions of the native aorta proximal to the stent-graft required to limit pressure differences are exposed to larger stresses thus increasing the potential of aortic tearing. Stent-graft-induced new entry, SINE, has been recorded in 30\% of patients treated with a stent-graft for aortic dissection. It is hypothesized that the stiffness of the stent-graft itself may exacerbate pressure in the proximal and distal native aorta and may be responsible for stent-graft induced new tears. A mathematical lumped parameter model and computational finite-element model were developed to evaluate this. The lumped parameter model uses electrical analogs for blood flow and compliances of the native aorta and stent-graft respectively. The mathematical model predicts an increased peak pressure for a stiff stent-graft as compared to a compliant stent-graft. Additionally, the mathematical model predicts an increase in peak pressure with an increase in stent-graft length. Motivated by the need to incorporate more anatomically accurate material properties and capture complexities of the flow field, a three-dimensional computational model of the flow through the descending thoracic aorta and stent-graft was developed. For an aortic dissection patient treated with a stiff stent-graft, the computational model predicts a peak cardiac cycle pressure of 190 mmHg, as compared to the predicted pressure of 176 mmHg for an aortic dissection patient treated with a compliant stent-graft. The model suggests that the increase in stent-graft length increases peak pressure during the length of the cardiac cycle, but a change in stent-graft modulus and stent-graft position does not affect the peak pressure during the length of the cardiac cycle. The modeling developed in this thesis confirms that the stiffness of the stent-grafts used to treat aortic dissection affects pressures in the proximal native aorta and could contribute to the formation of stent-graft induced new entry. This information is critical to the development of future stent-grafts to treat aortic dissection.