Reduced-order modeling of hemodynamics across macroscopic through mesoscopic circulation scales

Olivier Adjoua,Stéphanie Pitre-Champagnat,Didier Lucor

DOI: https://doi.org/10.48550/arXiv.1907.11439

2019-07-26

Abstract:We propose a hemodynamic reduced-order model bridging macroscopic and meso-scopic blood flow circulation scales from arteries to capillaries. In silico tree like vascular geometries, mathematically described by graphs, are synthetically generated by means of stochastic growth algorithms constrained by statistical morphological and topological principles. Scale-specific pruning gradation of the tree is then proposed in order to fit computational budget requirement. Different compliant structural models with respect to pressure loads are used depending on vessel walls thicknesses and structures, which vary considerably from macroscopic to mesoscopic circulation scales. Nonlinear rheological properties of blood are also included and microcirculation network responses are computed for different rheologies. Numerical results are in very good agreement with available experimental measurements. The computational model captures the dynamic transition between large-to small-scale flow pulsatility speeds and magnitudes and wall shear stresses, which have wide-ranging physiological influences.

Medical Physics,Computational Physics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to develop a reduced - order model (ROM) of hemodynamics across macro - to meso - scales to simulate blood flow from arteries to capillaries. Specifically, the researchers proposed a new method to generate the vascular tree structure and improve the accuracy of the model by considering the mechanical properties of the vessel walls at different scales and the nonlinear rheological properties of blood. In addition, this model can also capture the dynamic changes of pulsatile velocity, amplitude, and wall shear stress at large - to - small scales, and these changes have a wide range of impacts on physiological functions. The model aims to overcome the limitations of existing models in multi - scale simulation, especially in terms of pulsatility and nonlinear blood rheological properties in the microcirculation.

The key points of the paper include:
- **Generation of the vascular tree**: An idealized but realistic vascular tree structure is generated using a random growth algorithm, taking into account statistical morphology and topological principles.
- **Vessel wall models at different scales**: Different mechanical models are adopted according to the thickness and structural differences of the vessel walls to adapt to the changes from macro - to meso - scales.
- **Nonlinear rheological properties of blood**: The non - Newtonian fluid characteristics of blood are considered, especially the responses of different rheological models in the microcirculation network.
- **Integration of multi - scale modeling components**: The above multi - scale modeling components are integrated under a unified computational framework, so that the dynamic effects generated in the circulatory network from large arteries to capillaries can be simulated.

Through these improvements, this model can accurately simulate the complex behavior of blood flow at a relatively low computational cost, especially in the microcirculation, which is of great significance for understanding physiological mechanisms and disease development.