Coupled mixed-dimensional multiphase porous media approach for modeling airflow, blood flow, and gas exchange in the human lungs

Abstract

Mechanical ventilation is a life-saving therapeutic intervention for patients with impaired pulmonary function, yet it carries the risk of ventilator-induced lung injury (VILI). At bedside, physicians face the challenge of keeping lung tissue in a healthy state while ensuring sufficient gas exchange. Gas exchange occurs between the air in the alveoli and the dense network of pulmonary blood vessels in their walls, and it strongly depends on the balance between ventilation and perfusion. Mismatches between them are a major cause of impaired gas exchange in pulmonary diseases. However, the precise effects of ventilation, including tissue straining on the pulmonary circulation and the connected gas exchange, are largely unknown. Here, we therefore present an approach to computationally model the respiratory and circulatory systems of the human lungs, including gas exchange. Motivated by the lung’s hierarchical structure, our model represents larger airways and blood vessels as spatially resolved discrete networks of zero-dimensional (0D) models that are embedded into a multiphase porous medium (3D). The porous medium models the smaller respiratory and vascular structures, including lung tissue mechanics, in a homogenized way. Additionally, the respiratory gases—oxygen and carbon dioxide—are incorporated as chemical subcomponents of air and blood, with an exchange model in the porous domain. To connect the homogenized (porous domain) and the discrete (networks) representations of airways and blood vessels, we use a 0D-3D coupling method that allows a non-matching spatial discretization of both domains. This comprehensive coupled approach is physics-based, i.e., based on the underlying physical mechanisms, allowing us to investigate the (often unknown and unmeasurable) interplay between ventilation, tissue deformation, perfusion, and its effects on gas exchange dynamics. We anticipate our approach to be an important milestone towards better addressing clinically relevant questions in respiratory care in silico, which will contribute to developing improved ventilation strategies and better patient outcomes.