Biomedical Engineering Reference
In-Depth Information
The modelling of the lung airway is restricted to the CT-scans obtained, which often
are end at the trachea. Therefore, the influence of the respiratory system upstream of
the trachea is missing and the modelling requirements, assumptions, and expected
errors are discussed.
8.5.1
Geometry and Computational Setup
Two models were developed from CT scans by a helical 64-slice multidetector row
CT scanner (General Electric) of a 66-year-old, non-smoking, asthmatic Asian male
(height 171 cm and weight 58 kg) on the day after hospital admission with an acute
exacerbation of asthma. At the time, his lung function by spirometry (Spirocard, QRS
Diagnostic, Plymouth, Minnesota, USA) showed severe airflow obstruction with a
forced expiratory volume in 1 s (FEV
1
) of 1.02 L (41 % predicted). Data was acquired
with 1 mm collimation, a 40 cm field of view (FOV), 120 kV peak and 200 mA.
At baseline, 2 cm axial length of lung caudad to the inferior pulmonary ligament
was scanned during a single full inhalation total lung capacity breath-hold (FEV1
was measured at 2.31 L), which yielded 146 contiguous images (slices in transverse
direction, Z) of 1 mm thickness with voxel size 0
.
25 mm
×
×
1 mm.
An identical protocol was used to acquire images following recovery 30 days later,
when his FEV1 was measured at 2.27 L (91 % predicted). The two computationally
reconstructed models were then identified as the AA-model (acute asthma model)
and the REC-model (recovered model).
For this study, a parabolic inlet profile from an extended trachea is used. In a recent
study by Li et al. (2007), it was found that the type of velocity inlet condition and
existence of cartilaginous rings influence the air flow field; however, their impact is
less important in comparison with the variations in the upper airway geometry, e.g.,
branch curvature. A flow rate of 12 L/min was applied at the inlet, which corresponds
to a Reynolds number of 863 at the inlet. The flow regime has been considered
transitional for flow rates of 15-60 L/min (Zhang et al. 2008b) and for flow rates
greater than 30 L/min (Re
0
.
25 mm
2,500) (Jayarajua et al. 2008) based on the existence
of the laryngeal jet producing transitional behaviour. Other researchers however,
have applied a laminar flow regime for flow rates of 28-30 L/min (Re
∼
2,000-
2,500) (Nowak et al. 2003; van Ertbruggen et al. 2005) based on the argument that
RANS turbulence models cannot predict the transitional behaviour reliably, that the
flow within the lung branches rapidly becomes laminar after the initial bifurcations,
and that the flow is mostly laminar because the Reynolds number is globally below
the transitional value. In an investigation into the turbulence structures using Large
Eddy Simulations (LES), (Kleinstreuer and Zhang 2003) showed that at a flow rate of
15 L/min inlet turbulence is damped out and that laminar flow prevails. Interestingly,
deposition patterns were found to be unaffected by turbulent dispersion at 15 L/min,
although particle deposition is enhanced by turbulence for flow rates of 30 and
60 L/min when
St
∼
≤
0
.
06.
Based on these findings and the limitation of this study (exclusion of the larynx)
a steady laminar flow is used for a flow rate of 12 L/min. In addition, using a tur-
bulence model that is not suitable for flow rates that exhibit some light transitional
Search WWH ::
Custom Search