Challenges in Validating CFD-Derived Inhaled Aerosol Deposition Predictions

Abstract

Computational fluid dynamic (CFD) techniques have provided unprecedented opportunity for investigating inhaled particle deposition in realistic human airway geometries. Several recent articles describing local aerosol deposition predictions based upon “validated” CFD models have highlighted the challenges in validating local aerosol deposition predictions. These challenges include: (1) defining what is meant by validation; (2) defining appropriate experimental data for validation; and (3) determining when the agreement is not fortuitous. The term validation has numerous meanings, depending on the field and context in which it is used. For example, in computer programming it means the code executes as intended, to the experimentalist it means predicted results agree with matched experimental measurements, and to the risk assessor it implies that predictions using new parameters can be trusted. Based on the current literature it is not clear that a consensus exists for what constitutes a validated CFD model. It is also not clear what types of experimental data are needed or how closely the CFD input values and experimental conditions should be matched (similar or identical airway geometries, entrance airflow, or aerosol profiles) to validate CFD derived predictions. Due to the complexity of CFD computer codes and the multiplicity of deposition mechanisms, it is possible that total aerosol deposition may be accurately predicted and the resulting local particle deposition patterns are incorrect, or vice versa. Specific examples and suggestions for several challenges to experimentalists and modelers are presented.