=Paper=
{{Paper
|id=None
|storemode=property
|title=Three Dimensional Imaging Based Diagnosis for Obstructive Sleep Apnoea: A Conceptual Framework
|pdfUrl=https://ceur-ws.org/Vol-944/cihealth6.pdf
|volume=Vol-944
}}
==Three Dimensional Imaging Based Diagnosis for Obstructive Sleep Apnoea: A Conceptual Framework==
https://ceur-ws.org/Vol-944/cihealth6.pdf
Three Dimensional Imaging Based Diagnosis for
Obstructive Sleep Apnoea: A Conceptual
Framework
1
Syed M. S. Islam, 1 Mithran S. Goonewardene and 2 Paul Sillifant
1
School of Dentistry, The University of Western Australia, 35 Stirling Highway,
Crawley, WA 6009, Australia
2
Department of Oral and Maxillofacial Surgery, Royal Perth Hospital, 197
Wellington Street, Perth, WA 6000, Australia
1
{syed.islam,mithran.goonewardene}@uwa.edu.au, 2 paulsillifant@hotmail.com
Abstract. Obstructive Sleep Apnoea (OSA) is a disorder in which repet-
itive periodic cessation of breathing for 10 seconds or more occurs during
sleep despite increased effort to breathe. It leads to day-time sleepiness,
poorer health, increased healthcare and higher work-related and road ac-
cidents costing the national economy billions of dollars per year. Early in-
tervention may improve health outcomes for the sufferers. In this article,
a hierarchical diagnostic approach is proposed in which at first a quick
and safe three-dimensional (3D) surface imaging based technique is used
to identify patients susceptible to OSA, thereby allowing a cost-effective
patient screening. The susceptible patients are referred for volume imag-
ing such as Cone Beam Computed Tomography (CBCT) from which the
airway and other hard-tissue anatomical features can be extracted. Age
and gender specific 3D facial norms and different thresholds have been
proposed to compute against which individualized features can be judged
to determine the presence of OSA. Finally, the severity of OSA is mea-
sured by polysomnography sleep study only for those patients who are
confirmed for OSA by both surface and volume image-based analysis.
1 Introduction
Sleep apnoea is a serious health issue with significant public health implications
[6, 13]. There are three types of sleep apnoea: obstructive (OSA), central (CSA)
and mixed (combination of the two). In OSA (84% of cases), mechanical factors
play an integral role in the reduction of airflow despite continued respiratory
effort [1]. In CSA (0.4% of cases) the physiological respiratory control processes
fail to maintain the required respiratory function for optimal health.
OSA is characterised by the presence of apnoeas (i.e. a complete cessation
of breathing despite respiratory effort) or hypopneas, defined as greater than
30% reduction in chest and/or abdominal expansion during breathing or shal-
low breathing lasting at least 10 seconds combined with at least a 4% reduction
in oxygen desaturation. Numerous indices have been developed to express the
55
severity of sleep apnoea diagnosed using polysomnography and include the ap-
noea index (AI) which represents the total number of apnoeas per hour and the
Apnoea-Hypopnea Index (AHI), which represents the total combined apnoeas
and hypopnoeas per hour. The AHI has been divided into severity scales: mild
(5 < AHI < 15), moderate (15 < AHI < 30) and severe (AHI > 30). Additional
indices that have been utilised include sleep arousals (Respiratory Disturbance
Index) and subjective patient perceptions of sleep impact on daytime activities
(Epworth Sleepiness Scale).
During apnoeic episodes, arterial blood oxygen saturation decreases, and
sympathetic activity and blood pressure increases. Each apnoeic episode ends
with an arousal from sleep, resulting in marked fragmentation of sleep in af-
fected individuals. Excessive daytime sleepiness is a major consequence of OSA.
OSA has also been linked to significant conditions such as hypertension [16, 9],
ischaemic heart disease and stroke [18], premature death [17], and impairment of
cognitive functions [8] which may contribute to motor vehicle and workplace re-
lated accidents (comparable to functioning while intoxicated) [7]. A study from
The University of British Columbia demonstrated that a person with OSA is
twice as likely to be involved in a motor vehicle accident [19]. For untreated
individuals, it has been established that there is a 37% higher 5-year morbidity
and mortality rate [14].
It is estimated that 775,000 Australians (4.7% of the adult population) suffer
from OSA [15]. The Busselton (Australia) Health Survey [2] of 294 men aged
40 to 65 years revealed that about 26% of individuals have mild and 10% have
severe levels of sleep apnoea. The total financial and non-financial burden of
OSA in Australia was estimated as 21.2 billion dollars in 2010 including direct
health care cost of $575.42 million and indirect health care cost (due to lost
productivity, deadweight loss, workplace/motor vehicle accidents, social security
payments etc.) of $2.6 billion [18]. In U.S. it was estimated in 2008 that the
average additional annual health care cost of an untreated sleep apnoea patient is
US $1,336 contributing an estimated total of $3.4 billion/year additional medical
costs [1].
In this article, we introduce a novel quantitative diagnostic method for OSA
based on the combination of two approaches related to two different imaging
modalities (surface and volume). The first approach is based on the analysis of
a three-dimensional surface scan of a subject (using e.g. a 3dMD face scanner).
We propose to extract quantitative facial features from the scan to differentiate
between facial morphologies of OSA patients and normal non-apnoeic individ-
uals. The relative position of the upper and lower jaws to the skull base and
in turn to each other can be assessed as represented by the external facial ap-
pearance. These facial features can be evaluated to determine the relationship
between facial morphology and the severity of OSA. 3D surface facial scanning
has the advantage of being a non-invasive imaging tool which does not require
exposure to ionizing radiation. The second approach relates to the application
of state-of-the-art dental imaging in the form of a Cone Beam CT to obtain a
3D (volumetric) representation of the hard and soft tissues. The determination
56
of the morphology (shape and structure) of the airway of OSA patients should
help in revealing any significant deviations from the airway of normal individ-
uals. As Cone Beam CT is a readily available imaging tool in most clinics, the
proposed diagnostic method is easily accessible with many control non-OSA pa-
tients imaged for unrelated dental anomalies. The overall outcome of the article
is the development of improved conservative diagnostic methods which will be
accessible to wider patient groups and will contribute in early intervention.
The rest of the article is organized as follows. Various approaches currently
used for the diagnosis of OSA is described in Section 2. The conceptual frame-
work for our proposed approach is elaborated in Section 3. Proposal for the
evaluation of the new diagnostic method is discussed in Section 4 followed by
the conclusions in Section 5.
2 Existing Diagnostic Approaches for OSA
OSA is seen more frequently in older males and is related to many predisposing
factors such as increased Body Mass Index (BMI), increased neck circumference,
smoking, alcohol consumption and enlarged tonsils and adenoids. Clinicians also
recognise specific dentofacial deformities which predispose individuals to the
development of OSA. The obvious retrusion or underdevelopment of the lower
jaw and and/or the upper jaw alerts the clinician to the possibility of a patient
susceptible to OSA.
Today, overnight polysomnography remains the ‘gold standard’ diagnostic
method for OSA. It is a monitored sleep study to record biophysiological changes
that occur during sleep. Measurements include electroencephalogram, electroocu-
lograms, submental electromyogram, oronasal airflow, chest wall motion, and
arterial oxygen saturation. In addition to the significant inconvenience to the
patient, polysomnography requires sophisticated specialist facilities, technical
and scientific staff and sleep clinicians, which are commonly not available in all
regions.
Imaging techniques have been considered as useful adjunctive tools to diag-
nose and plan the treatment of OSA, with the radiographic head film (cephalo-
metric) analysis being the most convenient and widely used [3]. However, the
cephalometric analysis is inherently limited because of its two dimensional imag-
ing and the lack of information about the airway volume and dimensions [5]. In
addition, measurements are obtained with the patient in the upright position
which may not accurately reflect the distortion of the airway in the supine sleep-
ing position. This may create an underestimation of the degree and pattern of
airway narrowing and/or collapse. Lee et al. [10, 11, 12] analysed facial char-
acteristics to predict OSA with an accuracy of 76.1% using 2D photographic
and cephalometric images. These have limitations compared to 3D surface and
volume data. For example, while they demonstrated a relationship between fa-
cial structural measurements such as alar width and intercanthal distance, they
did not assess 3D positional relationships of the relevant structural components
representing the underlying jaw base, which is the focus of this article.
57
During the last few years, there has been significant interest in developing
conservative, cost-effective, patient-convenient and widely applicable methods to
diagnose and treat OSA. Although the morphology of patients diagnosed with
OSA has been well documented using two dimensional (2D) imaging techniques,
and to a much lesser degree using 3D imaging techniques, no specific strati-
fied evaluation has demonstrated the impact of progressive distortions of the
maxillomandibular structures on airflow and sleep performance.
3 Proposed Methods and Techniques
Considering the cost effectiveness and the simplicity, we propose a hierarchical
framework for diagnosing OSA. We would like to keep the cheaper and widely
accessible measures at the beginning and thus screening out a number of patients
before suggesting for more expensive and exhaustive approaches. The detailed
framework is described in this section.
3.1 Statistical Design
A null hypothesis for developing the new diagnostic approach can be defined
as follows: there will be a statistically significant difference in the proportion
of patients who are correctly diagnosed with OSA using the new method as
compared to the gold standard.
The sample size for the above hypothesis can conservatively be estimated
using an expected sensitivity (probability of correctly identifying a patient as
positive by the proposed approach given they have OSA) of 0.85 and specificity
(probability of correctly identifying a patient as negative by the new approach
given they do not have OSA) of 0.95, and a 95% confidence level. A sample
size of 100 OSA patients and 100 non-OSA participants would provide a 0.07
precision for sensitivity and 0.04 precision for specificity.
3.2 Determination of Norms and Thresholds
This approach requires a prior set up of age and gender specific facial norms
(nn) used as references. For that purpose, we propose to compute the age and
gender specific average faces from a large sample of non-OSA subjects. In ad-
dition to these average-face norms, we also propose to determine some other
thresholds associated with other discriminating features as illustrated in Fig. 1
and explained below.
Threshold t1 can be established as follows from 3D ear to ear facial surface
images (e.g. Fig. 2) of the 100 patients diagnosed with OSA by polysomnography.
The face area can be detected and cropped and various surface features (e.g.
length of the maxilla, mandible and chin and the circumference of the neck) can
be extracted. The relative shape ratios (RSRs) of these different features (e.g.
length of maxilla with respect to the mandible and that of maxilla and mandible
compared to the forehead and neck) can be computed. These features then can
58
Age and gender
specific norms
3D surface (nn) Find the
Determine
image most
age and
1 Detect common
Measure gender t1
and deviations
different specific
extract in OSA
surface deviations
n ear to ear patients
Facial features
face data Compute relative
images of
persons shape ratios
diagnosed (RSR)
with OSA 3D Find RSRs
volumetric that are
image 1 Segment Measure Determine mostly
volumetric correlation related to
airway,
parameters between airway
n mandible t2
of the airway morpholo
and
anatomic morphology gy of
maxilla
components and RSRs most of
the OSA
patients
Find the
Compute
Compute consistent
maxilla-
Soft-tissue factor of
mandible t3
compen- soft-tissue
relative
sation compen-
shape ratios
sation
Compute age and gender specific
t4
average morphology of the airway
Fig. 1. Block diagram of the computation of different thresholds (t1 , t2 , t3 , and t4 ) used
in the proposed diagnostic algorithm.
Fig. 2. 3D textured image of a person’s frontal (left) and right profile.
59
be compared with the age and gender specific norms to outline any deviations
from the norms. The threshold t1 can then be derived from these deviations.
Three more thresholds can be determined from 3D volumetric images which
can be acquired using a Cone Beam CT scanner from the same patients above.
The volumetric data of the airway (Fig. 3) and other anatomical features can be
segmented from these data using commercial software such as Dolphin, 3dMD-
vultus and 3D Slicer. Different volumetric parameters can be measured and
statistically correlated with the facial RSRs computed from the facial surface
images. The RSR (of each age and gender group) with the highest correlation
factor can be used as a threshold (t2 ). The relative shape ratio of maxilla and
mandible computed from volumetric data can be compared with those obtained
from surface data (3dMD) to evaluate the most common soft-tissue compensa-
tion factor (t3 ). The average morphology of the airway (threshold, t4 ) of the
different age and gender subgroups can be computed using the above software
or computer programming using MATLAB.
Hard-tissue
Airway
Soft-tissue
Fig. 3. 3D volumetric image of an OSA patient and his digitally segmented airway
represented in wireframe model.
3.3 Diagnosis Using the New Approach
As illustrated in Fig. 4, in the proposed diagnostic framework, a subject present-
ing for an OSA test will firstly be diagnosed using a surface image. A 3dMD scan
(e.g. Fig. 2) will be taken using the 3dMD Facial Scan System. The captured
image data will be represented as a 3D surface mesh. Then quantitative facial
shape features and ratios will be extracted or derived from the surface data.
An individualized norm will be determined based on the age and gender spe-
cific norms (nn) to localize and quantify any shape deviations (d1 ) of the facial
60
Age and gender
specific average Determine Compute
face (nn) individualized relative shape
norm ratio (RSR)
Localize
3D surface Detect and Measure
and
image extract 2D different (d1+t3-t1)+(RSR-
quantify
and 3D ear to shape t2)>=t5
deviations
ear face data features
A patient (d1)
No
approachi Yes
ng for
OSA test Measure
Localize and
Segment airway
quantify
airway volumetric
Volume deviations (d2)
parameters
image
Average airway norm of the
OSA patients (t4) Yes No
No
d2>=t6
OSA
Perform polysomnography
and other clinical observations
No
AHI>=5
Yes
OSA
Fig. 4. Block diagram of the proposed diagnostic methods.
61
shape of the patient relative to a non-OSA subject of the same gender and age
group. These shape deviations along with a soft-tissue compensation factor (t3 ),
will be compared with the morphological threshold t1 . Furthermore, the RSR of
the patient will also be compared with the threshold t2 . After computation of
these thresholds and deviations obtained from the analysis of the surface image
only, patients will be primary identified as a candidate for OSA if the summation
of the following two differences is greater than or equal to an empirically eval-
uated threshold t5 : (i) the difference of the soft-tissue deviations including any
soft-tissue compensations from the most common deviations in OSA patients
and, (ii) the difference of patients’ relative shape ratio from the similar ratio of
the most of the OSA patients. The condition can be mathematically represented
as in Equation 1.
(d1 + t3 − t1 ) + (RSR − t2 ) >= t5 (1)
The potential subjects will then be exposed to a Cone Beam CT scan for
a segmental airway assessment, and volumetric parameters of the airway will
be measured. Comparing the average airway norm of OSA patients (t4 ), the
deviation (d2 ) in the airway will be calculated. If the subjects’ deviations are
greater than or equal to an empirically determined threshold t6 , they will be
recommended for a polysomnographic sleep study and other clinical observations
in order to finally confirm the presence and severity of OSA expressed in AHI.
4 Evaluation of the Proposed Diagnostic Method
A comparison of the new 3D imaging-based diagnostic method with findings from
polysomnography can be performed through a test for difference in proportions
for the paired-sample design [4]. The diagnosis can be defined as successful if AHI
(found using polysomnographic sleep study) of the positively diagnosed patients
(using the proposed method) is found to be greater than 5 (the threshold measure
of apnoea).
The test for difference in proportions for the paired-sample design can then be
used to reject the hypothesis that there will be a statistically significant difference
in the proportion of patients who are correctly diagnosed with OSA using the
proposed method as compared to the gold standard (polysomnography). If there
is no statistically significant difference in the proportion of patients who benefit
from the proposed 3D image-based approach, then it should be widely adopted.
The test can be specifically described as follows:
1. For the 200 randomized subjects, apply the 3D imaging-based diagnosis ap-
proach (response Y1) and standard polysomnography procedure (matched
control, response Y2) [4].
2. Define a failure by a ‘miss’ and ‘false alarm’ (adopting the terminology of
detection theory), i.e. if the subject is diagnosed with the proposed approach
while they are not diagnosed using polysomnography, then it is a false alarm.
We then determine the proportion of cases when the proposed method re-
sulted in a success (P1) and when polysomnography resulted in success (P2).
62
3. If the proportions P1 and P2 computed above are equal, then reject the
hypothesis that the proportion of successes is the same for our 3D imaging-
based diagnosis method and polysomnography. (test statistics is unit-normally
distributed; the exact formula is given in [4]).
5 Conclusions
The proposed conservative, cost-effective, patient-convenient and widely appli-
cable methods to diagnose OSA will facilitate more accessible diagnosis of a
larger number of patient groups than is possible with polysomnography and will
enhance early intervention.
The purpose of the proposed diagnostic approach is not to replace the sleep
studies but to screen and then to stratify adult OSA patients for various modes
of treatment based on the anatomical features and airflow measurements. Im-
portantly, the proposed approach will also provide guidance to clinicians who
manage significant jaw structure problems in children with occasionally irre-
versible conventional orthodontics with little regard for the consequences of
leaving the child prone to developing sleep apnoea with their underlying jaw
structure remaining atypical. The specific patterns of jaw morphology can be
identified during the diagnosis in individuals who would be considered for surgi-
cal management of their jaw deformity in adulthood rather than attempting to
compensate the teeth for the jaw structure. Clinicians may then modify the way
in which they advise patients with more severe jaw structure problems based on
the impact on predisposition to OSA. Moreover, after screening, morphologically
predisposed patients may be warned about lifestyle habits which may contribute
to the possibility of developing OSA at a later age.
Acknowledgments
This research is sponsored by two Special Donation grants from the Australian
Society of Orthodontists Foundation of Research and Education (PG51311900
and PG51312000) and, by a Research Development Award (PG12104373) and
a Research Collaboration Award (PG12105002) from the University of Western
Australia.
63
Bibliography
[1] Aarab G, Lobbezoo F, Hamburger H, Naeije M (2010) Effects of an oral ap-
pliance with different mandibular protrusion positions at a constant vertical
dimension on obstructive sleep apnea. Clinical Oral Investigations 14(3)
[2] Bearpark H, Elliott L, Grunstein R, et al (1995) Snoring and sleep apnea. a
population study in australian men. Am J Respir Crit Care Med 151:1459–
1465
[3] Conley R (2011) Evidence for dental and dental specialty treatment of ob-
structive sleep apnoea. part 1: the adult osa patient and part 2: the paedi-
atric and adolescent patient. J Oral Rehabil 38(2):136–56
[4] Connor R (1987) Sample size for testing differences in proportions for the
paired-sample design. Biometrics 43(11):207–211
[5] Doff M, Hoekema A, Pruim G, van der Hoeven J, de Bont L, Stegenga B
(2009) Effects of a mandibular advancement device on the upper airway
morphology: a cephalometric analysis. J Oral Rehabil 36(5):330–7
[6] George C (2001) Reduction in motor vehicle collisions following treat-
ment of sleep apnoea with nasal cpap. Thorax 56(7):508–512, DOI
10.1136/thorax.56.7.508
[7] Horne J, Reyner L (1995) Sleep related vehicle accidents. Bmj
310(6979):565–7
[8] Klitzman D, Miller A (1994) Obstructive sleep apnea syndrome: complica-
tions and sequelae. Mt Sinai J Med 61(2):113–21
[9] Lavie P, Herer P, Hoffstein V (2008) Obstructive sleep apnoea syndrome as
a risk factor for hypertension: population study. Bmj 320(7233):479–82
[10] Lee R, Chan A, Grunstein R, Cistulli P (2009) Craniofacial phenotyping in
obstructive sleep apnea - a novel quantitative photographic approach. Sleep
32(1):37–45
[11] Lee R, Petocz P, Prvan T, Chan A, Grunstein R, Cistulli P (2009) Pre-
diction of obstructive sleep apnea with craniofacial photographic analysis.
Sleep 32(1):46–52
[12] Lee R, Sutherland K, Chan A, Grunstein R, Cistulli P (2010) Relationship
between surface facial dimensions and upper airway structures in obstruc-
tive sleep apnea. Sleep 33(9):1249–1254
[13] Lee W, Nagubadi S, Kryger M, Mokhlesi B (2008) Epidemiology of obstruc-
tive sleep apnea: a population-based perspective. Expert Rev Respir Med
2(3):349–364
[14] Marti S, Sampol G, Munoz X, Torres F, Roca A, Lloberes P, et al (2002)
Mortality in severe sleep apnoea/hypopnoea syndrome patients: impact of
treatment. Eur Respir J 20(6):1511–8
[15] Morgenthaler T, Kagramanov V, Hanak V, et al (2006) Complex sleep apnea
syndrome: is it a unique clinical syndrome? Sleep 29:1203–1209
64
[16] Nieto F, Young T, Lind B, Shahar E, Samet J, Redline S, et al (2000)
Obstructive sleep apnoea syndrome as a risk factor for hypertension: pop-
ulation study. Jama 283(14):1829–36
[17] Rapoport D (1994) Treatment of sleep apnea syndromes. Mt Sinai J Med
61(2):123–30
[18] Report DAE (2011) Re-awakening australia - the eco-
nomic cost of sleep disorders in australia. URL
http://www.sleephealthfoundation.org.au/research.html
[19] ScienceDaily (2008) Sleep apnea doubles car crash
risk, study shows. retrieved on oct 30, 2012. URL
http://http://www.sciencedaily.com/releases/2008/02/080218214401.htm
65