Journal of Biosciences and Medicines, 2013, 1, 28-32 JBM Published Online October 2013 (
Remodeling of livin g human nasal cavity under the
assistance of acoustic rhinometry technique*
Jun Zhang
Advanced Technology of Transportation Vehicle Key Laboratory of Liaoning Province, Dalian Jiaotong University, Dalian, China
Received 2013
Acoustic rhinometry could numerically describe up-
per airway condition of air draft by drawing a graph
plotting the distance from the nostril vs. the cross-
sectional area. Some d ecreases on the graph corres-
pond to the typical anatomic structures of human
nasal cavity. The 3-dimensional, computing fluid dy-
namic model of the same person was developed based
on computed tomography scans. The veracity of the
CFD model was valued by contrasting the relevant
areas of stenosis site between the model and the AR
graph. The aim in this study is to make clear how to
use an AR to help improve and enrich the CFD model
with the information of graph acquired from the
measurement. The combination of AR and CT can be
used to establish a living human nasal cavity model
with higher significant information content.
Keywords: Acoustic Rhinometry; Nasal Cavity;
Computed Tomography; 3-Dimensional Reconstruction;
Computing Fluid Dynamic Model
The human nose is the first line of defense that every-
body has to protect itself from outer invasion all one’s
life. With the currently development of research towards
the pathogenic mechanism and the application of iatrical
apparatus such as endoscopes, it has been gradually
proved that many diseases such as rhinitis or nasosinusi-
tis are related to the nasal anatomy [1]. Abnormal nasal
structure leads to abnormal nasal function and conse-
quently, the abnormal nasal function causes further ab-
normalities of nasal structure. Many rhino-diseases so
occur. By simulating the structure and function of the
nasal cavity with 3-dimensional reconstruction theory
with a computer, we can profoundly explore the outbreak,
treatment and prevention of nasal diseases. Nowadays,
there is a focus on the quantificational study of the rela-
tionship between environmental stimulus and body re-
modeling [2] which is called self-adaptation by estab-
lishing of computing fluid dynamic (CFD) models and
analyzing the results with inversion methods. In this pa-
per, the 3-dimensional model was reconstructed based on
computed tomography (CT) images. More instruments
should be applied to help adjust the model. Acoustic ref-
lectometry is a helpful tool that can measure the coronal
section information of the nasal cavity.
Acoustic reflectometry, which generally include
acoustic rhinometry (AR) and acoustic pharyngometry
(AP) i n the ear-nose-throat (ENT) department clinic, is a
relatively new modality of evaluating the construction
and function of the nasal cavities, and the technique can
quantify upper airway airing status [3]. A computer
draws a graph (Figure 1 show an AR report in detail).
under corresponding AR analysis system plotting the
distance from the nostril relative to the CSA. In this
graph, the x-axis represents the distance into the nasal
cavity, and the y-axis represents the 2-dimensional area
relative to distance. Most subjects demonstrate some
sudden decreases, which look like a series of valleys on
the graph. The valleys appear at ar ound 2 cm, 4 cm, 6 cm
and 8 cm, which are termed the minimal CSA 1 to 4 [4].
Numerical studies using realistic nasal geometries
have not been possible until recent advances in both
computer hardware and software. Up to the present,
many researchers have conducted numerical simulations
of human nasal airflow. Among them, Ravi P. Subrama-
niam et al.’s [5] is the most detailed of these studies. In
their numerical work, they studied airflow through the
whole human nose in a three-dimensional model and
used computational fluid dynamics simulations to re-
search the nasal airflow field for steady-state inspiration
in an anatomically accurate computer reconstruction of
both sides of the human nasal passages. And the project
included the posterior nasal airway and nasopharynx.
The model simulated in this study contains the whole
nasal region from nostril till to nasopharynx (see Figure
2) include maxillary sinus. The volunteers were asked for
acoustic rhinometry examine soon af ter CT scans. Acous-
*Project of Liaoning Province Education Department, LS2010030.
J. Zhang / Journal of Biosciences and Medicines 1 (2013) 28-32
Copyright © 2013 SciRes. OPEN ACCESS
Figure 1. AR report. Label a, b, c and d refer to MCSA1,
MCSA2, MCSA3 and MCSA4 respectively.
Figure 2. Lateral view of the left human nose.
Labels a-g mark locations of the coronal sections
in Figu re 1.
tic rhinometry was widely used in clinic and research
ever since it was invented. Bulent Mamikoglu and Ste-
ven M. Houser [6,7] used it to diagnose nasal septal dev-
iation and mild sleep apnea respectively which are the
main diseases it can diagnose.
The nasal passages have a complex geometry [8]. A lat-
eral outline of the human nasal passages with various
sections labeled is shown in Fig ure 2. We assume a
standing position in this discussion of the nasal anatomy
and in our model simulations and acoustic rhinometry
measurements. The anterior three-quarter of the nose is
divided into two enantiomorphic parts by the nasal sep-
tum. Entrance to these two cavities is through the nostrils,
which are slightly slanted with respect to the hard palate
about 45˚. The nasal vestibule is a funnel-shaped region
extending abou t 1 - 2 cm from the anterior tip of the nose
and includes the nostril section. The vestibule leads to
the constricted nasal valve region, the posterior boundary
of which is roughly presumed to be the anterior limit of
the inferior turbinate [9]. The nasal valve, which has
been conventionally defined as the site of narrowest
cross-sectional area of the nasal passage posterior to the
nostril opening, accounts for more than half of the total
nasal resistance [10]. Here, we define the exact location
and orientation of the nasal valve follow the tradition
opinion [9,11] in referring broadly to the nasal valve
region. The nasal vestibule and valve region form a flow-
limiting structure that is liable to disturb when the pres-
sure difference between the ambient and the respired air
in this portion of the airway exceeds a critical value. In
our research, we assume the boundary of the nasal air-
way to be rigid throughout. In contrast with this rigid
characterization, the nasal airway is, in fact, a dynami-
cally changing structure. Most individuals demonstrate a
coordinated nasal cycle where the nasal passages on one
side of the nose are occluded to various degrees due to
tissue congestion for a certain period of time [12]. This
physiological effect will lead to dynamically changing
differences in the airflow profiles on the two sides of the
nose. Here we ignored that effect.
At the constriction around the nasal valve region, the
airway undergoes a right-angled bend toward the nasal
floor, with a sudden increase in the coronal cross-sec-
tional area of the passages [9]. As a consequence, airflow
in humans has been reported to become turbulent in this
region [12]. The region between the nasal valve region
and the nasopharynx is referred to as the main nasal air-
Thirty volunteers (15 males and 15 females) were ran-
domly selected for our study. These patients were fully
evaluated by nasal anterior rhinoscopy and endoscopy.
Our initial physical examination allowed us to qualita-
tively designate a patient’s septum as having no devia-
tion or being mildly, moderately, severely, or markedly
severely deviated. All CT studies were done for evalua-
tion of each patient’s sinus problems and for the final
3-parameter reconstruction. The layer thickness was 3
mm to make the model more accurate. The CT scans and
AR graphs were concentratively obtained in one day. The
ages of the patients ranged from 25 to 55 years (median
30 years).
AR testing was carried out without a vasoconstrictor
and 15 minutes when the computer been steady running.
The test was performed in an acoustically adequate en-
vironment, with control of the following factors accord-
ing to the Standards of the International Committee on
AR [13] in order to guarantee the accuracy of the test: a)
the patients remained in the air-conditioned room for 30
minutes before the measurement, b) temperature and
humidity were kep t at 21˚C and 50% - 6 0% respectively,
c) stabilization of the patients’ heads: All volunteers were
examined in the sitting position on a straight-back
J. Zhang / Journal of Biosciences and Medicines 1 (2013) 28-32
Copyright © 2013 SciRes. OPEN ACCESS
chair with head support during measuring, d ) position of
the wave tube: wave tube was placed horizontally paral-
lel to the ground. Volunteers were requested to fix their
gaze at a point on the opposite wall straight ahead to the
gaze level. e) Use of Vaseline to prevent leakage, selec-
tion of proper silica gel coupler and f) training the con-
trol of respiration. All AR studies were performed by the
same examiner when measurement was performing. In
order to guarantee the accuracy of the test, at least three
curves were obtained for each examination. After each
measurement, the nasal adaptor was removed from the
nostril and after a few seconds reconnected. A new mea-
surement was then obtained. The results were considered
to be adequate when the coefficient of variation was less
than 10% [14]. A mean curve for each nostril was then
constructed for each volunteer based on the recorded
The 3-dimensional, finite-element models were de-
veloped from CT scans of the noses of those volunteers.
The scans provided 37 - 43 contiguous coronal images of
each nose at intervals of 3 mm. The points, which were
on the edge of the nasal cavity to the nasal wall, were
first picked up from the gray-scaled CT images. Then the
airway perimeters traced from the images appeared. Se-
condly, connecting the points on the perimeters, we con-
structed the 2-dimensional, finite-element lines of each
coronal slice. After every piece of perimeters was deli-
neated, the third step is to link the lines of one piece to
next together to form the 3-dimensional, finite-element
closed area of the nasal airways. At last, we formed the
columns with the closed area. Thus, the 3-dimensional,
finite-element model of the nasal airways was finished.
The reconstruction utilized commercial software AN-
SYS6.1 (ANSYS Inc., USA). The finite-element models
used 190,000 - 300,000 3D FLOTRAN elements, com-
prising 34,000 - 63,000 nodes. The following conditions
were imposed on the spatial boundaries of the flow field:
a) a no-slip condition was imposed at the surface of the
nasal airway walls (vs = 0 m/s); b) a str ess-free condition
was established at the outlet. This condition is satisfied if
the normal velocity gradients are set to zero at the outlet
(v = constant); c) at the nostril surface (the flow inlet),
stationary pressure, which equal to atmosphere pressure
in numerical value (p = 101325 Pa) were specified. We
applied the finite-element method (FEM) using the
commercial fluid dynamics solution package of AN-
SYS6.1 to solve the equations of motion for airflow. The
FEM obtains solutions to th e full Navier-Stokes and con-
tinuity equations within each element of the fin iteele-
ment mesh.
Thirty cases were performed in this study and here we
take one for illustration. The volunteer was a healthy,
26-year-old, nonsmoking northeast China male. The vo-
lunteer’s septum was no deviation and sinus was normal.
The FEM model contained 42 layers of CT images, and
was meshed into 271,168 elements with 59,382 nodes
after reconstruction. The AR offered numerical value of
CSA each 0.24 cm from the anterior nostril between 7.5
and 20 cm inwards. Connecting the discrete points with
smooth curve, we got the AR graph. Here we intercepted
the curve from 4 - 12 cm for analysis.
The FEM model and the AR graph of the same nasal
cavity are paralleling which is shown in Figure 3. The
longitudinal curve-axis across the nasal airway passage
from the nostril to the nasopharynx referred to the spread
direction of the sound waves and it was corresponding to
the X-axis of the AR graph below. We set a coordinate
point each two centimeter (0, 2, 4, 6, 8, and 10 cm) on
the curve. From the AR graph, we can see four valleys
appear at the positions of about 2, 4, 6 and 8cm. They
correspond to the classical anatomic sites of nasal cavity
as declared above and the corresponding sectional pat-
terns at each point of the model are shown in Figure 4.
The patterns meet the description of valleys. Similarly,
we can get the pattern and area of any cross -section of
nasal cavity.
In the present clinical studies, d eviation of the nasal sep-
tum (DNS) is one of the most common diagnoses in
otorhinolaryngology practice. AR has been extensively
used in the otorhinolaryngologic clinic to diagnose DNS
since it was introduced to medical field. Furthermore, th e
diseases relation to the function and structure of upper
airway such as obstructive sleep apnea (OSA), adenoids
hypertrophy (AH) of children and so on can all be diag-
nosed by AR. The acoustic reflection technique is repro-
ducible, noninvasive, and free from potential side effects.
The good correlation between AR graph and nasal area
Figure 3. Meshed model of human nasal cavity
and its AR graph.
J. Zhang / Journal of Biosciences and Medicines 1 (2013) 28-32
Copyright © 2013 SciRes. OPEN ACCESS
Figure 4. Sectional pattern of nasal cavity model at 2, 4, 6, 8
and 10 cm along the direction of sound waves.
adds to the potential of this technique. Integrated with
some other tools, such as rhinomanameter, AR could be
applied to more extensive fields.
This study describes how to use the AR technique to
assess a 3-dimensional model of nasal cavity. The me-
thod of 3-dimensional construction based on CT scans in
this paper is widely used in the medical model construc-
tion. CT detects the nasal cavity wall with transmitted
technique, while AR does with reflected technique. The
AR and the CT technique are complementary to each
other. Their testing results can be compared to assess if
the nasal model is acceptable. Moreover, the results be-
tween CFD simulations of the airflow profile and the AR
graph can also examine the quality of the model. In this
way, we can receive more amount of objective informa-
tion from the model of the nasal cavity. In brief, they all
offer objective documentation, however, are all useful in
the reconstruction of nasal cavity by combined of each
Knowledge of airflow in the human nose is important
for understanding many aspects of the biology and pa-
thology of the respiratory tract. We get to know that ab-
normal nasal structure leads to abnormal nasal function
and as a consequence, abnormal nasal function causes
further abnormalities of nasal structure. Simulating the
relationship between structure and function of nasal cav-
ity with 3-dimensional reconstruction theory and CFD
analysis by a computer, we can deeply research the prin-
ciple of the outbreak, treatment and prevention of nasal
diseases. Nowadays, there is a focus on the quantifica-
tional study of the relationship between environmental
stimulus and body remodeling by inversion methods and
establishing of CFD models. In this paper, the 3-dime n -
sional model was reconstructed based on CT images. AR
scan was applied to help value the model. It shows that
the combination of AR and CT is helpful and offers
another way for the modelin g of living human nasal cav-
AR is a relatively new technique that quantifies upper
airway condition of obstruction. It may provide the re-
searcher with objective coronal CSA data of nasal cavity
beyond that of any other physical examination. A com-
puter draws a graph plotting the distance from the nostril
relative to the cross-sectional area; thus the 3-dime n -
sional nasal cavity is projected into a 2-dimensional
graph. In this graph, some sudden decreases emerge,
which look like a series of valleys. Such segments on the
graph correspond to the typical anatomic structures of
human nasal cavity, where sudden changes of air pres-
sure and flow velocity occur because of the changes of
nasal airway status. The 3-dimensional, CFD models are
developed based on CT scans. Connecting the contours
from CT images by their position to a closed volume, a
3-dimensional nasal passage model is then reconstructed.
The airflow can be simulated under commercial software
or some other programs. CT and AR are complementary
techniques since CT acquires boundary information by
the transmission of X-rays, while AR does it by the ref-
lection of sound waves. We can value the veracity of the
CFD model by contrasting the relevant areas of stenosis
site between the model and AR graph. The aim in this
study was to make clear how to use an AR to help im-
prove and enrich the CFD model with the information of
graph acquired from the measurement. It is proved that
the combination of AR technique and CT is a way for the
establishing of a living human nasal cavity model to be
of more amount of information.
[1] Uliyanov, Y.P. (1997) Surgical reconstruction of nasal
serodynamics. XVI World Congress of Otolaryngology
Head and Neck Surgery, Sydney, 1591-1595.
[2] Wenzhi, Z. (2004) Experimental study on the stress and
remodeling adaptation of vivo bone tissue and its biolog-
ical model. Doctoral Dissertation of Dalian University of
Technology, Dalian.
[3] Corey, J.P., Gungor, A. and Nelson, R. (1998) Normative
standards for nasal cross sectional area by race as meas-
ured by acoustic rhinometry. Otolaryngology Head and
Neck Surgery, 119, 389-393.
[4] Corey, J.P., Kemker, B.J. and Nelson, R. (1997) Evalua-
tion of the nasal cavity by acoustic rhinometry in normal
and allergic subjects. Otolaryngology Head and Neck
Surgery, 117, 8-22.
[5] Subramaniam, R.P., Richardson, R.B. and Morgan, K.T.
(1998) Computational fluid dynamics simulations of in-
spiratory airflow in the human nose and nasopharynx.
Inhalation Toxicology, 10, 91-120.
[6] Mamikoglu, B., Houser, S.M. and Akbar, I. (2000) Acou-
stic rhinometry and computed tomography scans for the
diagnosis of nasal septal deviation, with clinical correla-
tion. Otolaryngology Head and Neck Surgery, 123, 61-68.
[7] Houser, S.M., Mamikoglu, B. and Aquino, B.F. (2000)
Acoustic rhinometry findings in patients with mild sleep
apnea. Otolaryngology He ad and Neck Surgery , 126,
J. Zhang / Journal of Biosciences and Medicines 1 (2013) 28-32
Copyright © 2013 SciRes. OPEN ACCESS
[8] Negus, V. (1958) The comparative anatomy and physi-
ology of the nose and paranasal sinuses. Livingstone,
[9] Proctor, D.F. (1982) The upper airway. Nose, Uppe r Air-
way Physiology and the Atmospheric Environment, 23-
[10] Haight, S.J. and Cole, P. (1983) Site and function of the
nasal valve. Laryngoscope, 49-55.
[11] Lanza, D.C. and Clerico, D.M. (1995)Anatomy of the hu-
man nasal passages. Handbook of Olfaction and Gusta-
tion, 53-73.
[12] Cole, P. (1982) Upper respiratory airflow. Nose, Upper
Airway Physiology and the Atmospheric Environment,
[13] Hilberg, O.F. and Pedersen, (2000) Acoustic rhinometry:
Recommendations for technical specifications and stan-
dard operating procedures. Rhinology, 16, 3-17.
[14] Brown, I.G., Zamel, N. and Hoffstein, V. (1986) Pharyn-
geal crosssectional area in normal men and women. Jour-
nal of Applied Physiology, 61, 890-895.