| |
By Mike Slack, Fluent Europe Ltd.
View the pdf of this article
There is growing interest in the healthcare industry to use CFD to develop
models of living systems. These models are being used to better understand
biological processes, drug delivery, and interactions between the body
and implanted medical devices. One area of interest is the human respiratory
system. The geometrical, physical, and biological descriptions of this
system are complex, and are the subject of ongoing research. The nose,
in particular, is an effective filter that has undergone thousands of
years of evolutionary change. It is well known that particles larger than
10 microns are normally captured within the nasal cavity while smaller
particles or droplets can make it beyond this region and travel deeper
into the respiratory system. The convoluted flow passageways of the nasal
cavity promote inertial deposition at high flow rates, while turbulent
diffusion is believed to cause deposition at lower flow rates.
Static pressure contours on the nasal cavity and trachea during inhalation
Through a joint collaboration involving Materialise and Fluent Europe
Ltd, a CFD model of the human airways has been developed to study the
inhalation of particulate matter. Two geometries, taken from MRI scans,
were made available for the study: a nasal cavity that extends from the
nostrils to the top of the larynx, and a trachea from the same patient
that extends from the larynx to the first stages of the bronchi. The movement
of the volunteer's heart and lungs during the MRI procedure reduced the
resolution of the smaller bronchial branches, so these regions were not
included in the study. Materialise's Magics software was used to blend
the two geometries together, and GAMBIT and TGrid were used to create
a fine, unstructured tetrahedral mesh. A time-varying inhalation and exhalation
profile representing a nasal breather at rest was applied to the bronchi
branch ends as a boundary condition. Turbulence, and turbulence-particle
interaction were included in the simulations. To correctly capture turbulent
deposition, it was necessary to model the turbulence all the way to the
surfaces of the model. The mucus layer lining the nasal cavity was modeled
as a thin porous boundary layer capable of capturing particulates. Deformation
of the passageways during the breathing cycle was not included in the
model.
Velocity contours on several slices through the nasal cavity 
Large (12 microns, red) and small (1 micron, blue) particles at three
times during the breathing cycle; the large particles are trapped in the
nasal cavity; the small ones are inhaled (left) and then expelled during
exhalation (center and right) 0.30 seconds 2.20 seconds 3.80 seconds
The initial CFD work has been a sensitivity study to investigate the
impact of different boundary conditions and turbulence-particle interactions
on the deposition of large and small particles. Particles 1 and 12 microns
in diameter were followed during a complete breathing cycle. To account
for the particles passing beyond the extents of the model and into the
deep lung, those that passed beyond the branch ends during inhalation
were counted. During exhalation, a percentage of those particles were
re-introduced at the branch ends. The findings showed that particle deposition
is very sensitive to the turbulence-particle interactions, and to the
way that the mucus layer is represented. By refining the assumptions,
it has been possible to develop a model of the upper respiratory system
that captures the appropriate proportion of material reaching the different
surfaces of the respiratory system.
The project represents the first stage in the development of a computational
lung modeling capability. It is hoped that this type of model will eventually
help both clinicians and medical device manufacturers to better understand
and design drug delivery and life support systems in the future.
|
|
|
