In silico modeling of aerosol deposition in lungs

The mathematical representation of particle transport in the airflow within the pulmonary system is of interest in a wide range of fields: from inhalation toxicology to infectious disease proliferation to drug delivery. The complexities of the airway geometry, the highly non-constant nature of the airflow involving inhalation–exhalation cycles, intermittent turbulence and the shedding of turbulent vortices into otherwise non-turbulent flows, complicated moving boundaries in the deep lung with heretofore undetermined mechanics, and the complex interaction with the surfactant layer and its own flows, all of which are coupled with significant variability from patient to patient, all point to a problem that is computationally intractable today. Yet, the physics of each of these complicating factors is well understood, and, in fact, models of most of them have already been implemented into sophisticated CFD codes. Thus, the overall modeling problem appears to really be one of identifying which physical phenomena are involved, including their in silico representations from an existing library. The simulation problem of porting this model to appropriate hardware and mining it to extract meaningful results is the open challenge today. Given Moore's law of the doubling of computational power every 18–24 months, it does appear to be only a matter of time before this happens. The potential for improved understanding of pulmonary particle deposition patterns, on an individual-by-individual basis, with its attendant implications for personalized medicine is enormous.

Section editor:Steven C. George – Departments of Biomedical Engineering, and Chemical Engineering and Materials Science, University of California, Irvine, CA 92697-2575, USA