Physics-Informed Neural Networks Simulation and Validation of Airflows in Three-Dimensional Upper Respiratory Tracts

Accurate and efficient simulation of airflows in human airways is critical for advancing the understanding of respiratory physiology, disease diagnostics, and inhalation drug delivery. Traditional computational fluid dynamics (CFD) provides detailed predictions but is often mesh-sensitive and computationally expensive for complex geometries. In this study, we explored the usage of physics-informed neural networks (PINN) to simulate airflows in three geometries with increasing complexity: a duct, a simplified mouth-lung model, and a patient-specific upper airway. Key procedures to implement PINN training and testing were presented, including geometry preparation/scaling, boundary/constraint specification, training diagnostics, nondimensionalization, and inference mapping. Both laminar PINN and SDF-mixing-length PINN were tested. PINN predictions were validated against high-fidelity CFD simulations to assess accuracy, efficiency, and generalization. Results demonstrated that nondimensionalization of the governing equations was essential to ensure training accuracy for respiratory flows at 1 m/s and above. Hessian-matrix-based diagnosis revealed a quick increase in training challenges with flow speed and geometrical complexity. Both the laminar and SDF-mixing-length PINNs achieved comparable accuracy to corresponding CFD predictions in the duct and simplified mouth-lung geometry. However, only the SDF-mixing-length PINN adequately captured flow details unique to respiratory morphology, such as obstruction-induced flow diversion, recirculating flows, and laryngeal jet decay. The results from this study highlight the potential of PINN as a flexible alternative to conventional CFD for modeling respiratory airflows, with adaptability to patient-specific geometries and promising integration with static or real-time imaging (e.g., 4D CT/MRI).