Large-scale simulation of the human arterial tree.

Area of Science:

• Computational fluid dynamics
• Multiscale modeling
• Biomedical engineering

Background:

• Full-scale simulations of the virtual physiological human (VPH) necessitate advancements in modeling, multiscale mathematics, scientific computing, and medical imaging.
• Current computational demands for 3D blood flow simulations in the human arterial tree are prohibitive, even for petaflop supercomputers.

Purpose of the Study:

• To review key challenges and propose solutions for enabling feasible 3D simulations of blood flow in the human arterial tree.
• To introduce a hierarchical modeling approach for efficient simulation of blood flow across different vascular scales.

Main Methods:

• A three-level hierarchical approach based on vessel size: macrovascular network (MaN), mesovascular network (MeN), and microvascular network (MiN).
• Solving 3D Navier-Stokes equations for MaN simulations on arterial networks with tens of arteries and bifurcations, incorporating boundary conditions for neglected dynamics.

Main Results:

• Demonstration of recent MaN simulations on arterial networks.
• The proposed hierarchical approach is expected to enable multiscale simulations (MaN-MeN-MiN) within hours per cardiac cycle on petaflop computers in the near future.

Conclusions:

• Advancements in modeling, multiscale mathematics, scientific computing, and medical imaging are crucial for VPH simulations.
• The hierarchical multiscale simulation approach offers a feasible pathway for complex cardiovascular modeling and disease studies.