A fast approach to estimating Windkessel model parameters for patient-specific multi-scale CFD simulations of aortic flow

• Published in: Computers & fluids Vol. 259; p. 105894
• Main Authors: Li, Zongze, Mao, Wenbin
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 15.06.2023
• Subjects: Aortic flow; Computational Fluid Dynamics; Flow resistance; Lattice Boltzmann Method; Windkessel model; Aortic flow; Flow resistance; Windkessel model; Lattice Boltzmann Method; Computational Fluid Dynamics
• ISSN: 0045-7930; 1879-0747; 1879-0747
• DOI: 10.1016/j.compfluid.2023.105894

•Aortic flow simulation with patient-specific WK3 boundary conditions.•Windkessel model parameters were obtained by a non-iterative procedure.•Patient-specific geometric flow resistance involved in the optimization.•Global optimization toolbox in MATLAB was utilized to tune the parameters.•Up to 0.22% error in flowrate and 4.13 mmHg deviation from the preset values. Computational fluid dynamics (CFD) study of hemodynamics in the aorta can provide a comprehensive analysis of relevant cardiovascular diseases. One trending approach is to couple the three-element Windkessel model with patient-specific CFD simulations to form a multi-scale model that captures more realistic flow fields. However, case-specific parameters (e.g., Rc, Rp, and C) for the Windkessel model must be tuned to reflect patient-specific flow conditions. In this study, we propose a fast approach to estimate these parameters under both physiological and pathological conditions. The approach consists of the following steps: (1) finding geometric resistances for each branch using steady CFD simulation; (2) using the pattern search algorithm from Matlab toolbox to search the parameter spaces by solving the flow circuit system with the consideration of geometric resistances; (3) performing the multi-scale modeling of aortic flow with the optimized Windkessel model parameters. The method was validated through a series of numerical experiments to show flexibility and robustness, including physiological and pathological flow distributions at each downstream branch from healthy or stenosed aortic geometries. This study demonstrates a flexible and computationally efficient way to capture patient-specific hemodynamics in the aorta, facilitating personalized biomechanical analysis of aortic flow.