You are on page 1of 20

Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity.

Computers in
Biology Medicine 102: p40-50

Examining mesh independence for flow


dynamics in the human nasal cavity
Inthavong, Kiao∗1 , Chetty, Annicka1 , Shang, Yidan 1 and Tu, Jiyuan 1
1 School of Engineering, RMIT University

Abstract

Increased computational resources provide new opportunities to explore so-


phisticated respiratory modelling. A survey of recent publications showed a
steady increase in the number of mesh elements used in computational mod-
els over time. Complex geometries such as the nasal cavity exhibit sharp
gradients and irregular curvatures, leading to abnormal flow development
across their surfaces. As such, a robust method for examining the near-wall
mesh resolution is required. The non-dimensional wall unit y + (often used
in turbulent flows) was used as a parameter to evaluate the near-wall mesh
in laminar flows.
Mesh independence analysis from line profiles showed that the line location
had a significant influence on the result. Furthermore, using a single line
profile as a measure for mesh convergence was unsuitable for representing
the entire flow field. To improve this, a two-dimensional (2D) cross-sectional
plane subtraction method where scalar values (such as the velocity magni-
tude) on a cross-sectional plane were interpolated onto a regularly spaced
grid was proposed. The new interpolated grid values from any two meshed
models could then be compared for changes caused by the different meshed
models. The application of this method to three-dimensional (3D) volume
subtraction was also demonstrated.
The results showed that if the near-wall mesh was sufficiently refined, then
narrow passages were less reliant on the overall mesh size. However, in
wider passages, velocity magnitudes were sensitive to mesh size, requiring
a more refined mesh.

∗ kiao.inthavong@rmit.edu.au

1
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Introduction
Objective tests of disordered nasal airflow, such as anterior rhinomanometry, acous-
tic rhinometry and nasal peak inspiratory flow, suffer from limited sensitivity,
specificity and reliability, providing restricted insight into the specific cause or
site of obstruction. Assessments of nasal morphology for abnormalities and dis-
eases can be conducted through 3D computational models (Garcia et al. 2015; Na
et al. 2012; Vinchurkar et al. 2012; Wang and Elghobashi 2014; Zhu et al. 2011),
providing medical practitioners with the insights necessary to make informed de-
cisions regarding surgical interventions.
To understand the precise causes of nasal obstruction and the effects of corrective
surgery, more qualitative and quantitative information is required, including an
understanding of the airflow pattern (laminar or turbulent), localised velocity and
pressure (at different flow rates and in different parts of the nasal cavity) and wall
shear stress. Computational fluid dynamics (CFD) applied to the modelling of fluid
flow in the nose has been used to investigate airflow patterns (Keyhani et al. 1995)
for common deformities such as nasal bone fractures (Chen et al. 2011), septal
deviation (Chen et al. 2009) and inferior turbinate hypertrophy (Chen et al. 2010b).
Increased computational power has facilitated enhanced studies involving more
airway models (De Backer et al. 2008; Sullivan et al. 2014; Xiong et al. 2008;
Zhang et al. 2008).
With resources readily available, it is timely to explore mesh independence analy-
sis used in computational modelling of human nasal cavity airflows. The accuracy
requirements for computational modelling in clinical and engineering applications
(Frederick et al. 2001; Kimbell and Subramaniam 2001; Shang et al. 2015) are far
less stringent than in research applications. Thus arises the question of the extent
to which a mesh should refined be with respect to the level of flow detail.
Early reported computational studies of airflow through a human nasal cavity in-
clude a study by Elad et al. (1993), which used an idealised nose-like shape (9,888
mesh elements used) and a study by Keyhani et al. (1995), which reconstructed
the nasal cavity from computed tomography (CT) scans of one chamber, abruptly
ending at the choanae (76,950 mesh elements used). A literature review was per-
formed of computational studies of the nasal cavity from the years 1993−2017 and
the number of mesh elements reported in the models was plotted (see Figure 1).
Since 1993, the number of studies published in journals has steadily increased, as

2
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

has computational model fidelity, multiphysics and complexity.

Figure 1: Mesh cell elements reported in the literature between 1993 to 2017. Examples
of some high meshed nasal models are highlighted (Inthavong et al. 2012; Schroeter et al.
2011). The studies with very large number of cells ((Calmet et al. 2016; Li et al. 2017) are
linked to high fidelity turbulent flow using Large Eddy Simulations (LES) or Direct
Numerical Simulations (DNS).

For clinical and engineering applications, a computational model may not require
high numbers of mesh elements when only general flow features are of interest.
These features include: the change of direction of flow streamlines from vertical
(as air enters the nostril inlet) to horizontal (in the main nasal passage); flow ac-
celeration through the nasal valve; recirculating flow in the olfactory region; bulk
flow through the middle nasal cavity and nasal cavity floor; and a 90◦ turn as
air enters the nasopharynx (Wen et al. 2008). Many computational models apply
simplifications to physiological behaviour (e.g., rigid walls, dry, smooth walls and
steady flows); therefore, higher-fidelity models may not yield substantial increases
in accuracy with additional simulation time.
Figure 1 shows a large disparity in the reported mesh elements between 10,000
cells and 44 million cells, despite many publications reporting mesh-independent

3
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

computational models. This is due to a combination of the available computational


power and the grid independence test method used. To quantify errors of mesh
independence/grid convergence, an estimation can be performed using Richardson
extrapolation (Richardson 1910; 1927) and the grid convergence index (GCI), which
was introduced by Roache (1994) as a method for uniform reporting of CFD results
but has since evolved into an uncertainty estimator.
Mesh independence analysis in CFD studies of airflow in the nasal cavity has
mainly applied convergence of a local parameter (such as velocity) along a selected
line profile (Chen et al. 2010a; Inthavong et al. 2008) or convergence of global pa-
rameters such as pressure drop or wall shear stress (King Se et al. 2010; Li et
al. 2012; Wang et al. 2009). Early work by Inthavong et al. (2006) relied on local
velocity profiles for mesh independence, resulting in a mesh count of 586,000 tetra-
hedral elements. Subsequent articles recorded a count of 950,000 cells (Inthavong
et al. 2007; Inthavong et al. 2008; Inthavong et al. 2009). More recent work has
shown approximately 4 to 5 million elements are required for grid independence
(Frank-Ito et al. 2015; Schroeter et al. 2011; Tong et al. 2016; Xi and Longest 2008).
Accurately predicting airflow patterns in the nose to better understand nasal air-
way disorders relies on the use of a high-quality mesh. The near-wall mesh re-
finement in turbulent flow regimes (typically > 20 L/min steady inhalation rate)
is characterised by the y + parameter, which describes the first wall-adjacent cell
height relative to the different sub-layers of the turbulent boundary layer. For
laminar flow regimes (typically <15 L/min steady inhalation rate), no parameter
is used as the laminar boundary layer consists of a single region and thus, studies
do not report quantified measures of the near-wall mesh.
This paper evaluated the use of the y + parameter to quantitatively describe the
near-wall boundary layer of a laminar internal pipe flow (Poiseuille flow). Addi-
tionally, the use of velocity line profiles as a measure of grid independence was
evaluated and extended to 2D planar velocity distributions and 3D volume distri-
butions so that higher-order grid independence analysis could be performed.

Method
Computational model: pipe geometry
Four computational pipe models with mesh variations were created for evaluating
a laminar y + parameter. Pipe flows provide solutions that can be applied to com-

4
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

plex internal flows that behave like passages, such as human airways. The pipe
had a radius of 7.5 mm (R = 7.5 mm) and a length of 16R (120 mm). A structured
mesh with varying densities was used (see Table 1). The cross-sectional mesh is
shown in Figure 2a. A laminar streamwise-periodic flow was applied to the pipe to
produce a fully developed flow (Poiseuille flow) with different Reynolds numbers
of 200, 500, 1000, 1300, 1600 and 2000.

Laminar y + for internal pipe flow


The fully developed laminar velocity profile in a pipe is:
( ( ) )
R −y 2
u(y) = 2U 1 − (1)
R

where U is the mean velocity, R is the pipe radius and y is the distance from the
wall. The non-dimensional wall distance and dimensionless velocity are defined
as:
ρu ∗y u
y+ = ; u+ = (2)
µ u ∗y
where ρ is the fluid density, µ is dynamic viscosity and u ∗ is the friction velocity
(taken at the first cell), defined as:

τw du
u∗ = ; τw = µ (3)
ρ dy

Substituting Eqn. 1 and its derivative (i.e., velocity gradient) into y + and u + pro-
duces:
( ( )2)
√ U 1− R
R−y
2ρy U µ
y+ = (R − y); u+ = √ (4)
µ R2ρ Uµ
(R − y)
R2 ρ

Setting y ′ = y/R and collecting the terms that make up the Reynolds number the
non-dimensionless wall units simplify to


y = y 2Re(1 − y ′)
+
(5)

which are the solution profiles for a fully developed laminar Poiseulle flow. The
term y ′ has range of y ′ = 0 (wall boundary) to y ′ = 1 (pipe centre/radius).

5
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

(a) Cross-section mesh of pipe geometry. Pipe M2 shows the dimensions detailed in Table.1

(b) Nasal cavity geometry and surface mesh

Figure 2: Nasal cavity geometry model with dimensions for overall length, height, and
width. Six planes were created (two in each of the anterior, main passage, and posterior
regions) and labelled as planes-a, -b, -c, -d, -e, -f. The unstructured surface mesh for model
nose M3 (2-million cells) is shown and an inferior view of the nostril inlet is given to show
the mesh quality with prism layers.

6
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Computational Model – nasal cavity geometry


A human nasal cavity geometry was meshed using Ansys-Fluent v18.2 which pro-
vided high quality tetrahedral cells with prism layers. The model was meshed
with five levels of refinement, and labelled nose-M1 (coarsest mesh, 0.5-million
cells) through to nose-M5 (finest mesh 10-million cells), for the mesh indepen-
dence evaluation. The level of skewness was used as the main criteria for mesh
quality, where the most refined model, nose-M5 had a maximum skewness of 0.64.
A steady laminar flow of 15L/min (mass flowrate = 3.0625e-4 kg/s) was defined at
the outlet.
The pressure-based coupled solver was used where the system of equations com-
prising the continuity equation

(ui ) = 0 (6)
∂xi

and the momentum equation


( )
∂ui ∂p ∂ ∂ui
ρui = + µ (7)
∂x j ∂xi ∂x j ∂x j

were solved together. Spatial discretization used the second-order accurate up-
wind scheme where, quantities at cell faces were computed using a multidimen-
sional linear reconstruction. The Green-Gauss node-based gradient of a variable
was used to discretize the convection and diffusion terms in the flow conservation
equations. The simulations used double-precision to avoid errors that may arise
from high aspect ratio cells from the prism layers.

Results
y+ for laminar pipe flow
In turbulent flows, the near wall boundary layer is characterized by y + , where
the viscous dominant sublayer (i.e. laminar layer) occurs when y + < 5. In this
sublayer, u + /y + ≈ 1. Computational results from the pipe simulations of normal-
ized velocity and wall units for all Reynolds numbers were plotted in Figure 3a to
verify that u + /y + ≈ 1 for all cases, suggesting it captures the laminar boundary
layer profile correctly. Figure 3b shows y + increases for increased flow rate, e.g.
Re = 200 to Re = 2000 (which are the plotted lines obtained Eqn.5). The markers

7
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

(a) (b)

Figure 3: Normalized velocity u + profile as a function of wall units y + for a laminar pipe
flow in the near wall region of different meshed pipe models (d) CFD results for
determining the y + profile for different flow rates. The markers for the largest first cell
height (y ′ = 0.055) are for pipe-M1 and moving closest to the wall for models pipe-M2
(y ′ = 0.022), pipe-M3 (y ′ = 0.013), pipe-M4 (y ′ = 0.008) are shown.

plotted were individual results obtained from each pipe model, and for the entire
range of Reynolds numbers. These results verify that the y + value can be used as a
quantitative parameter for examining near wall meshing of laminar flows, where
the profile characteristics are consistent with laminar boundary layers, such as the
behavior found in the viscous dominant sublayers of turbulent flows.

Nasal cavity flow and mesh independence


Velocity contours in three planes (plane-b, plane-c, and plane-d) showed recircu-
lating flow patterns were captured in fine-meshed model nose-M5 (Figure 4). In
coarse meshed model nose-M1, the flow was more diffusive. Although only three
planes were given in the figure for brevity, it showed that where the left and right
chambers were separate, the velocity contours between each nasal cavity model
were similar. However, where the two chambers merged into a single passage
(plane-d), there was greater variation in the contours between the coarse and fine
meshed models.
In reported mesh independence tests, or CFD model comparisons, a single line pro-
file of a flow variable is used to demonstrate convergence or model performance.
Multiple line profiles were extracted and compared between nose-M1 (0.5 million
cells) and nose-M3 (2.3 million cells) models shown in Figure 5(a), (c), (e), and be-
tween nose-M3 (2.3 million cells) and nose-M5 (10.0 million cells) models shown in

8
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Figure 4: Velocity contours in three planes for models, nose-M1, nose-M3, and nose-M5
which shows the variation between models of different mesh density. Plane-b, and plane-c
are found where two nasal chambers exist, while plane-d is in the nasopharynx region
where the chambers have merged.

9
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Figure 5: Multiple velocity line profiles on cross-section (a)(b) plane-b; (c)(d) plane-c;
(e)(f) plane-d for direct comparison between models (a)(c)(e) nose-M1 (dots), and nose-M3
(lines); and (b)(d)(f) nose-M3 (dots), and nose-M5 (lines);

10
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Figure 6: Comparison of velocity profiles at plane-d for (a) nose-M3 (dots) to nose-M4
(lines) and (b) nose-M4 (dots) to nose-M5 (lines) where grid convergence occurs at
nose-M4

Figure 5(b), (d), (f). At plane-b, the profile along Line-3 (and to a lesser extent Line-
2 or Line-5) showed reasonable mesh convergence between nose-M1 and nose-M3.
However, there is significant flow variation along Line-4. For plane-c, the largest
flow variation occurred at Line-2, and Line-4. For plane-d, the flow is significantly
different for all lines, and this is due to the larger open area, where the mesh varia-
tion is substantial, compared to the anterior planes that exhibit narrowed regions.
The results show that the use of line profiles, and in particular using a single line
profile, is a poor and limited method to justify mesh convergence over the entire
computational domain.
It was apparent that plane-d required the most mesh refinement due to mixing
of two flow streams merging at the nasopharynx caused by the two chambers
merging. Plane-b, and plane-c were not shown because the velocity profiles had
already reached mesh independence for the model nose-M3. Line velocity profiles
at plane-d for nose-M3 to nose-M4 (Figure 6a) and nose-M4 to nose-M5 (Figure 6b)
show the velocity profiles converge between nose-M4 to nose-M5.

Effect of near wall mesh in laminar flow


The results showed nose-M4 had converged velocity profiles across three planes
across the nasal cavity. The effect of the near wall mesh was evaluated by chang-
ing the prism layer distribution (Figure 7). The redistribution of mesh elements
changed the averaged y + values where the unmodified model (Figure 7b) had

11
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Figure 7: Changing the near wall mesh refinement on the mesh independent model,
nose-M4 (4.8million cells) where (a) inferior view of model showing the nostirls; and the
(b) original nose-M4 (unchanged) model with refined near wall mesh. Subsequent
+
relaxation of the near wall mesh refinement was created where (c) yave = 0.27; (d)
+ + +
yave = 0.53; (e) yave = 0.98; (f) yave = 1.

y+ave =0.13. Four additional models were created with increasing y+ave values.
The redistribution across the entire surface as a distribution function is shown in
Figure 8a, and despite the larger y + variation, the resulting wall shear stress (Fig-
ure 8b) remained consistent among all the models. Line velocity profile compar-
isons taken at the same locations (planes from Figure 5) were referenced against
+
the unmodified model that had yave = 0.13 shown in Figure 9. When y+ave in-
+
creased to 0.27, there was very little change in all profiles. For, yave = 0.53, signif-
icant differences in the profiles were observed in plane-d, but not in plane-b nor
plane-c. As y+ave increased further, the variations between velocity profiles be-
gan to appear in plane-b and plane-c but not significantly relative to the near wall
mesh coarsening.
+
This suggests that the near wall mesh was sufficiently resolved if yave = 0.27, and
that it influenced velocity profiles in large channels of the nasal cavity, and less
significant for narrow channels of the nasal cavity.

12
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Figure 8: Gaussian density estimation function for (a) y + values and (b) wall shear stress
on the boundary wall surface of nasal cavity geometry M4 (approximately 4.8million
cells), affected by near wall mesh refinements

Multi-dimensional grid independence


The line profile comparison is 1D, and analogous to this, a comparison on cross-
sectional planes is a 2D representation. The velocity (or any scalar) from a plane
was extracted and interpolated (Inthavong et al. 2014) onto an 800 x 800 uniformly
spaced grid for any two different meshed models. The interpolation used Python’s
SciPy library with the ‘nearest’ method, and the variation between two meshed
models was defined as the difference between the interpolated grid values, given
as

1 ∑ abs (qi − pi )
N
σ= × 100 (8)
N i =1 (qi + pi ) /2

where p, and q are scalar values (e.g. velocity) from cell-i taken from the coarser,
and finer meshed models, respectively. N is the total number of cells in the grid.
To constrain extremely large σ (residual) due to a small qi the denominator used
the average value between the two models, (qi + pi )/2.
The six planes from Figure 2b were evaluated where pi and qi were the velocity
magnitude values on the planes in any two models. The σ value for each plane
pair between nose-M2 and nose-M3 models are shown as contours in Figure 10a.

13
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Figure 9: Comparisons of line velocity profiles on cross-sectional planes (see Figure 5)


showing the effect of different numbers of near wall mesh elements. The horizontal axis,
is the x-coordinate values (m), while the vertical axis is velocity (m/s). The profiles are
compared against the reference velocity profiles of the unmodified model that has
+
yave = 0.13 (lines).

14
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Figure 10: (a) Contours of velocity subtraction for planes between nose-M2 and nose-M3
(b) Residual values from subtraction of planes between two models nose-M2 and nose-M3

Plane-e exhibited the largest local discrepancy due to the intense mixing from the
left and right chambers merging. Plane-a and plane-b also displayed some local
discrepancies, but less severe. The averaged residual for each plane was plotted in
Figure 10b, which showed σ increase from nose-M1 to nose-M2 but then decreased
monotonically. The averaged σ reduced to less than 5% between nose-M4 and nose-
M5 models. This was sufficiently small to satisfy mesh independence.
If we extend the interpolation from 2D planes to the airway volume, then the
mesh independence subtraction method becomes 3D. Data from each nasal cavity
model was interpolated onto a 100 × 100 × 100 uniformly spaced grid. The σ for
each pair of models are shown in Figure 11 where the σ initially increased between
nose-M1 and nose-M2, but then decreased. This profile was the same as the 2D
plane subtraction method where the σ values were nearly identical and suggests

15
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Figure 11: 3D volume velocity magnitude residual change between two subsquent mesh
models

that using 6-planes is sufficient to obtain an overall σ representation of the entire


volume.

Discussion
Mesh independence based on convergence of flow parameters on a single line is a
limited approach since it ignores the rest of the domain which can lead to regions of
poorly converged mesh. On the other hand, mesh independence based on a single
averaged value of global flow parameters misses locally poor mesh convergence
since it becomes averaged out. In this study we propose that: multiple line profiles
across multiple cross-section planes should be used or; a comparison of a flow
parameter at each mesh element region, over the entire domain using multiple
planar subtractions or volume subtraction, to ensure sufficient mesh independence.
The CFD results also showed that the near wall mesh for laminar flows can adopt
the y + parameter to ensure a sufficient mesh was applied at the wall.
The velocity line profiles and contour planes with narrow passages did not vary
significantly between all meshed models. However, cross-sectioned regions that
exhibited a larger passage (e.g. nasopharynx) showed greater variations in flow
behavior between the different meshed models. While this was partly due to the
increased mixing of two streams of fluids, the nasopharynx (plane-e) had more
space to fill and the mesh cells varied in size between models (see Table 2). Mesh
refinement in the larger passage regions is therefore essential especially where
rapid changes in flow behavior (such as mixing) exists and has overall importance

16
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

to the analysis.
A variety of mesh independence tests were presented which showed that a mesh
with approximately 4-5million cells provided overall good mesh independence.
More recent work has shown approximately 4–5 million elements were needed
for grid independence (Frank-Ito et al. 2015; Schroeter et al. 2011; Tong et al.
2016; Xi and Longest 2008) which also used 4-5million cells. Coarse meshed mod-
els can provide sufficient gross flow features, if the near wall mesh can maintain
a y + < 0.27, and it may be acceptable for rapid solution turnaround times, be-
fore committing to a higher resolution analysis. This is particularly the case for
the anterior nasal cavity and if we are after rapid results and don’t care too much
about accuracy. This has strong relevance for large scale model analysis model
and provides rapid turnaround of results.

Conclusion
This paper presented the use of the y + parameter for evaluating the near wall
mesh resolution in laminar flows. It was also shown that if the near wall mesh
was refined well then, the anterior half of the nasal cavity was not so reliant on
the overall mesh size since the passageways were very narrow and the flow field
was heavily influenced by the bounded walls. Flow variations were most sensitive
in the nasopharynx where the two chambers merged into a single passageway, and
that this region required more refined meshing in the inner flow region (bulk flow
region).
Mesh independence based on convergence of flow parameters on a single line is a
limited approach since it ignores the rest of the domain which can lead to regions
of poorly converged mesh. The extension of the line profiles to 2D plane, and 3D
volume was used to evaluate the velocity magnitude in different meshed models of
the nasal cavity. This new method showed that it could determine local regions of
discrepancies when each mesh was refined. The methods presented are expected
to provide a tool for quick and easy, yet reliable evaluation of mesh independence,
or comparison of different computational model performances.

Acknowledgement
The authors acknowledge the financial support for the research, authorship, and/or
publication of this article from the Australian Research Council (grant no. DP160101953)

17
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

References
Calmet, H., Gambaruto, A. M., Bates, A. J., Vazquez, M., Houzeaux, G., Doorly, D. J. (2016). Large-
scale CFD simulations of the transitional and turbulent regime for the large human airways during
rapid inhalation. Comput Biol Med 69:166-180.
Chen, X. B., Lee, H. P., Chong, V. F., Wang de, Y. (2009). Assessment of septal deviation effects on
nasal air flow: a computational fluid dynamics model. The Laryngoscope 119:1730-1736.
Chen, X. B., Lee, H. P., Chong, V. F., Wang de, Y. (2010a). A computational fluid dynamics model
for drug delivery in a nasal cavity with inferior turbinate hypertrophy. Journal of aerosol medicine
and pulmonary drug delivery 23:329-338.
Chen, X. B., Lee, H. P., Chong, V. F., Wang de, Y. (2010b). Impact of inferior turbinate hypertrophy
on the aerodynamic pattern and physiological functions of the turbulent airflow - a CFD simula-
tion model. Rhinology 48:163-168.
Chen, X. B., Lee, H. P., Chong, V. F., Wang de, Y. (2011). Assessments of nasal bone fracture effects
on nasal airflow: A computational fluid dynamics study. American journal of rhinology & allergy
25:e39-43.
De Backer, J. W., Vos, W. G., Devolder, A., Verhulst, S. L., Germonpré, P., Wuyts, F. L., Parizel, P.
M., De Backer, W. (2008). Computational fluid dynamics can detect changes in airway resistance
in asthmatics after acute bronchodilation. Journal of Biomechanics 41:106-113.
Elad, D., Liebenthal, R., Wenig, B. L., Einav, S. (1993). Analysis of air flow patterns in the human
nose. Medical and Biological Engineering and Computing 31:585-592.
Frank-Ito, D. O., Wofford, M., Schroeter, J. D., Kimbell, J. S. (2015). Influence of Mesh Density on
Airflow and Particle Deposition in Sinonasal Airway Modeling. Journal of aerosol medicine and
pulmonary drug delivery.
Frederick, C. B., Gentry, P. R., Bush, M. L., Lomax, L. G., Black, K. A., Finch, L., Kimbell, J. S., Mor-
gan, K. T., Subramaniam, R. P., Morris, J. B., Ultman, J. S. (2001). A hybrid computational fluid
dynamics and physiologically based pharmacokinetic model for comparison of predicted tissue
concentrations of acrylic acid and other vapors in the rat and human nasal cavities following in-
halation exposure. Inhalation Toxicology 13:359-376.
Garcia, G. J. M., Schroeter, J. D., Kimbell, J. S. (2015). Olfactory deposition of inhaled nanoparticles
in humans. Inhalation Toxicology 27:394-403.
Inthavong, K., Ge, Q. J., Li, X. D., Tu, J. Y. (2012). Detailed predictions of particle aspiration affected
by respiratory inhalation and airflow. Atmospheric Environment 62:107-117.
Inthavong, K., Shang, Y., Tu, J. (2014). Surface mapping for visualization of wall stresses during
inhalation in a human nasal cavity. Respiratory Physiology and Neurobiology 190:54-61.
Inthavong, K., Tian, Z., Tu, J. (2007). CFD simulations on the heating capability in a human nasal
cavity, in 16th Australasian Fluid Mechanics Conference (AFMC), School of Engineering, The Uni-
versity of Queensland, 842-847.
Inthavong, K., Tian, Z. F., Li, H. F., Tu, J. Y., Yang, W., Xue, C. L., Li, C. G. (2006). A numerical study
of spray particle deposition in a human nasal cavity. Aerosol Sci Tech 40:1034-1045.
Inthavong, K., Tian, Z. F., Tu, J. Y., Yang, W., Xue, C. (2008). Optimising nasal spray parameters for
efficient drug delivery using computational fluid dynamics. Computers in Biology and Medicine
38:713-726.
Inthavong, K., Wen, J., Tu, J., Tian, Z. (2009). From CT scans to CFD modelling–fluid and heat trans-

18
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

fer in a realistic human nasal cavity. Engineering Applications of Computational Fluid Mechanics
3:321-335.
Keyhani, K., Scherer, P. W., Mozell, M. M. (1995). Numerical Simulation of Airflow in the Human
Nasal Cavity. Journal of Biomechanical Engineering 117:429-441.
Kimbell, J. S. and Subramaniam, R. P. (2001). Use of computational fluid dynamics models for
dosimetry of inhaled gases in the nasal passages. Inhalation Toxicology 13:325-334.
King Se, C. M., Inthavong, K., Tu, J. (2010). Inhalability of micron particles through the nose and
mouth. Inhalation Toxicology 22 287-300
Li, C., Jiang, J., Dong, H., Zhao, K. (2017). Computational modeling and validation of human nasal
airflow under various breathing conditions. J Biomech 64:59-68.
Li, X., Inthavong, K., Tu, J. (2012). Particle inhalation and deposition in a human nasal cavity from
the external surrounding environment. Building and Environment 47:32-39.
Na, Y., Chung, K. S., Chung, S. K., Kim, S. K. (2012). Effects of single-sided inferior turbinectomy on
nasal function and airflow characteristics. Respiratory Physiology and Neurobiology 180:289-297.
Phillips, T. S. and Roy, C. J. (2014). Richardson Extrapolation-Based Discretization Uncertainty
Estimation for Computational Fluid Dynamics. Journal of Fluids Engineering 136:121401-121401-
121410.
Richardson, L. F. (1910). On the approximate arithmetical solution by finite differences of physical
problems involving differential equations, with an application to the stresses in a masonry dam.
Proceedings of the Royal Society of London. Series A 83:335-336.
Richardson, L. F. (1927). The deferred approach to the limit. Philosophical Transactions of the
Royal Society of London. Series A 226:299-361. Roache, P. J. (1994). Perspective: A Method for
Uniform Reporting of Grid Refinement Studies. Journal of Fluids Engineering 116:405-413.
Schroeter, J. D., Garcia, G. J. M., Kimbell, J. S. (2011). Effects of surface smoothness on inertial
particle deposition in human nasal models. J Aerosol Sci 42:52-63.
Shang, Y., Inthavong, K., Tu, J. (2015). Detailed micro-particle deposition patterns in the human
nasal cavity influenced by the breathing zone. Computers & Fluids 114:141-150.
Sullivan, C. D., Garcia, G. J. M., Frank-Ito, D. O., Kimbell, J. S., Rhee, J. S. (2014). Perception of
better nasal patency correlates with increased mucosal cooling after surgery for nasal obstruction.
Otolaryngology - Head and Neck Surgery (United States) 150:139-147
Tong, X., Dong, J., Shang, Y., Inthavong, K., Tu, J. (2016). Effects of nasal drug delivery device
and its orientation on sprayed particle deposition in a realistic human nasal cavity. Computers in
Biology and Medicine 77:40-48.
Vinchurkar, S., De Backer, L., Vos, W., Van Holsbeke, C., De Backer, J., De Backer, W. (2012). A case
series on lung deposition analysis of inhaled medication using functional imaging based computa-
tional fluid dynamics in asthmatic patients: Effect of upper airway morphology and comparison
with in vivo data. Inhalation Toxicology 24:81-88.
Wang, S. M., Inthavong, K., Wen, J., Tu, J. Y., Xue, C. L. (2009). Comparison of micron- and nanopar-
ticle deposition patterns in a realistic human nasal cavity. Respiratory Physiology and Neurobiol-
ogy 166:142-151.
Wang, Y. and Elghobashi, S. (2014). On locating the obstruction in the upper airway via numerical
simulation. Respiratory Physiology and Neurobiology 193:1-10.
Wen, J., Inthavong, K., Tu, J. Y., Wang, S. (2008). Numerical simulations for detailed airflow dynam-
ics in a human nasal cavity. Respiratory Physiology & Neurobiology 161:125-135.

19
Inthavong et.al. (2018) Examining mesh independence for flow dynamics in the human nasal cavity. Computers in
Biology Medicine 102: p40-50

Xi, J. and Longest, P. W. (2008). Numerical predictions of submicrometer aerosol deposition in the
nasal cavity using a novel drift flux approach. International Journal of Heat and Mass Transfer
51:5562-5577.
Xiong, G., Zhan, J., Zuo, K., Li, J., Rong, L., Xu, G. (2008). Numerical flow simulation in the post-
endoscopic sinus surgery nasal cavity. Medical and Biological Engineering and Computing 46:1161-
1167.
Zhang, J., Liu, Y., Sun, X., Yu, S., Yu, C. (2008). Computational fluid dynamics simulations of res-
piratory airflow in human nasal cavity and its characteristic dimension study. Acta Mechanica
Sinica/Lixue Xuebao 24:223-228.
Zhu, J. H., Lee, H. P., Lim, K. M., Lee, S. J., Wang, D. Y. (2011). Evaluation and comparison of nasal
airway flow patterns among three subjects from Caucasian, Chinese and Indian ethnic groups using
computational fluid dynamics simulation. Respiratory Physiology and Neurobiology 175:62-69.

20