QUALITY CONTROL AND PATIENT DOSIMETRY IN DENTAL

CONE BEAM CT
J. Vassileva* and D. Stoyanov
Department for Radiation Protection at Medical Exposure, National Centre of Radiobiology and Radiation
Protection, 132 Kliment Ohridsky blv., 1756 Soa, Bulgaria
*Corresponding author: j.vassileva@ncrrp.org
This paper presents the initial experience in performing quality control and patient dose measurements in a cone beam
computed tomography (CT) scanner (ILUMA
TM
Ultra, IMTEC Imaging, USA) for oral and maxillofacial radiology. The
X-ray tube and the generator were tested rst, including the kVp accuracy and precision, and the half-value layer (HVL). The
following tests specic for panoramic dental systems were also performed: tube output, beam size and beam alignment to the
detector. The tests specic for CT included measurements of noise and CT numbers in water and in air, as well as the homo-
geneity of CT numbers. The most appropriate dose quantity was found to be the air kerma-area product (KAP) measured
with a KAP-metre installed at the tube exit. KAP values were found to vary from 110 to 185 mGy m
2
for available adult
protocols and to be 54 mGy m
2
for the paediatric protocol. The effective dose calculated with the software PCXMC (STUK,
Finland) was 0.05 mSv for children and 0.090.16 mSv for adults.
INTRODUCTION
Cone beam computed tomography (CBCT) is
increasingly being used in the oral and maxillofacial
radiology practice since 2001 for implant planning,
assessment of bony and dental pathological con-
ditions, orthodontic treatment planning, temporo-
mandibular joint imaging, etc.
(1,2,3)
.
Published reports showed that the effective doses
from CBCT are 580 times higher than doses from
single panoramic radiograph and 123 % of a com-
parable conventional CT
(15)
. The CBCT dose
varies substantially depending on the device, eld of
view (FOV) and selected technique factors.
The basic principles on the use of dental CBCT,
recently established by the European Academy of
DentoMaxilloFacial Radiology and the project
SEDENTEXCT (safety and efcacy of a new and
emerging dental X-ray modality) require a quality
assurance program to be implemented for each
CBCT facility, including acceptance tests and regular
routine tests to ensure optimal radiation protection
for staff and patients
(6,7)
. No standard quality control
(QC) and dosimetry protocols are available in the lit-
erature. Very few publications deal with QC for
CBCT
(8,9)
; a few more present studies of patient
doses
(15)
.
This paper presents the initial experience with
acceptance testing and patient dosimetry for one
type of CBCT.
CBCT SYSTEM
Measurements were carried out on the latest full-
view CBCT for craniofacial imaging (ILUMA
w
Ultra Cone Beam CT Scanner, IMTEC Imaging,
USA). This system has a xed FOV and a pyrami-
dal-shaped X-ray beam, directed towards a at panel
detector on the other side of the patients head. The
C-arm with the X-ray assembly and detector per-
forms a single 3608 rotation around the head,
acquiring multiple 2D projection images. This varies
from a traditional medical CT which uses a fan-
shaped X-ray beam acquiring axial image slices of
the FOV. Reconstruction software generates a 3D
volumetric data set used to provide secondary recon-
structed images in axial, sagittal and coronal planes,
as well as multi-planar (oblique, curved, cross-sec-
tional) reformation and 3D visualisation. The
system characteristics are summarised in Table 1.
The system operates in a continuous mode with a
xed tube voltage of 120 kVp; two steps of the tube
current are used 3.8 mA for adult patients and
1 mA for children. Scanning time is selectable
between 20 and 40 ms.
TESTING PROCEDURE
Because the CBCT system has a rotational geometry
like panoramic dental systems, tomographic recon-
struction like a conventional fan-beam CT and a
large area detector like digital radiography/uoro-
scopy systems, the elaborated QC program included
relevant parts of the tests performed for these three
systems. A similar approach was reported by other
authors
(8,9)
. For most of the parameters were
adopted remedial levels (RLs) and suspension levels
(SLs), similar to the existing national requirements
for conventional radiography, CT and dental
systems
(10)
.
# The Author 2010. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Radiation Protection Dosimetry (2010), Vol. 139, No. 13, pp. 310312 doi:10.1093/rpd/ncq011

a
t

U
n
i
v
e
r
s
i
d
a
d
e

E
s
t
a
d
u
a
l

P
a
u
l
i
s
t
a

J

l
i
o

d
e

M
e
s
q
u
i
t
a

F
i
l
h
o

o
n

J
u
l
y

3
0
,

2
0
1
3
h
t
t
p
:
/
/
r
p
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Tube voltage accuracy and precision
Tube voltage (kVp) was measured with an X-ray
multimetre (Barracuda, RTI Electronics, Sweden).
The multipurpose detector (MPD) was centred at
the image detector surface. Measurements were per-
formed in two possible modesin the scout mode
(without rotation), using the MPD calibration for
radiographic systems, and in CT mode, using the
corresponding calibration of the MPD. Tube voltage
precision was tested with three consecutive
measurements.
Specic tube output
Tube output was measured with a at 30 cm
3
ionis-
ation chamber (type 233612, PTW Freiburg,
Germany) positioned in air at 60 cm from the tube
focus. Measurements were performed in a scout
mode. Consistency of the output was tested with
three repeated measurements.
Half-value layer
The same detector and geometry as for the output
measurements were used also for half-value layer
(HVL) measurements. Aluminium lters of 99.9 %
purity were used.
Size of radiation beam and beam alignment
to the detector
Measurement was performed with a ready pack
X-ray lm with a size larger than the detector area,
positioned directly on the at panel detector surface.
The lm was exposed during the full rotation at
120 kV, 3.8 mA and 40 ms.
CT number uniformity and image noise
CT number uniformity and image noise were
measured with a cylindrical water phantom with a
diameter of 16 cm, positioned at the patients head
support. Using the software incorporated into the
system, the mean CT number and the standard devi-
ation of mean CT numbers were calculated in a
central and four peripheral regions of interests (ROI)
with an area of 500 mm
2
.
Patient dose
Taking into account the rotational geometry and
the large detector size, the most appropriate dose
quantity for patient dose measurements in CBCT
was decided to be the air kerma-area product
(KAP). Measurements were performed with a trans-
mission ion chamber (Diamentor E, PTW, Freiburg,
Germany) positioned at the tube exit. KAP was
measured for all available examination protocols.
Organ doses and the effective dose E were calcu-
lated with a commercial software (PCXMC v.2.0,
STUK, Finland). The examination for a typical
patient was simulated with 12 views in 308 intervals.
All available protocols for patient examinations were
simulated. The effective dose was calculated with
tissue weighting factors from Publication 60 and
Publication 103 of the ICRP.
RESULTS AND DISCUSSION
The system operates at a xed tube voltage of
120 kVp. The measured tube voltage was 126+
3 kV. The results were similar when measured in a
scout mode and in a rotational mode. For kVp
accuracy +6 kVp was applied as an RL and 15 %
as an SL. The measured value was within the
accepted limits but it was expected that for a new
system the tube voltage accuracy should be better.
The possible reason for the found inaccuracy of the
tube voltage could be the calibration of the kVp-
metre. The CBCT system operates at high kVp but
with a low tube current, and corresponding cali-
bration of kVp-metres has to be introduced by the
manufacturers of measuring instruments.
The tube voltage precision was better than 1 %.
A maximum deviation of 5 % of each measured
value from the mean is proposed as an RL for this
parameter.
Specic tube output consistency was tested with
three consecutive measurements at xed exposure
parameters and was found to be better than 1 %.
Table 1. Technical specication of the cone beam CT scanner (ILUMA
TM
Ultra, IMTEC Imaging, USA).
Tube voltage Fixed 120 kVp HF continuous mode
Tube current Selectable 1 mA or 3.8 mA
Scan time One rotation 360; selectable 20 s or 40 s
Reconstructed voxel size Isotropic; 0.4; 0.3; 0.2 or 0.1 mm
Grey scale 16 bits
X-ray detector Amorphous Si at panel with active pixel area: size 193 mm 242 mm; matrix 1516 1900
FOV Fixed
Light eld Two lasers for centring and a third for the midline
QC AND DOSIMETRY IN DENTAL CBCT
311

a
t

U
n
i
v
e
r
s
i
d
a
d
e

E
s
t
a
d
u
a
l

P
a
u
l
i
s
t
a

J

l
i
o

d
e

M
e
s
q
u
i
t
a

F
i
l
h
o

o
n

J
u
l
y

3
0
,

2
0
1
3
h
t
t
p
:
/
/
r
p
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

The proposed RL for this parameter is a deviation
of 5 % of each measured value from the mean.
The HVL measured at 120 kVp was 7.3 mm Al.
The same RL as for conventional CT was proposed
for an HVL of 3.8 mm Al at 120 kVp. The test of the
size of radiation beam and beam alignment to the
detector is important because of the strong inuence
on the patient dose. It is expected that the radiation
eld will not exceed the size of the active detector
area. The measured irradiated eld was found to be
185 mm 234 mm, which is lower than the declared
active detector area of 193 mm 242 mm.
CT number uniformity was tested by the calcu-
lation of mean CT numbers in ve regions of the
image of the water phantom. Applicability of the cri-
teria for fan-beam CT was tested: maximum differ-
ence of +8 HU between CT numbers in different
areas of the phantom. The results for the mean values
were as follows: for the central ROI: 7.7 HU and for
the peripheral ROI: 16.6; 12.0; 22.0; 14.5. The found
deviations between mean CT numbers are higher than
expected. The question under investigation is whether
CT numbers in CBCTare the same as in axial CT.
The measured value of the standard deviation of
the mean CT numbers in the central ROI, s 0.35,
was recorded as a baseline value for the image noise.
Similar to CT, remedial actions at future QC tests
will be needed when the deviation of s from the
baseline value exceeds 20 %.
The measured air KAP and calculated effective
doses for three available clinical protocols are pre-
sented in Table 2. KAP values were found to vary
from 110 to 185 mGy m
2
for available adult proto-
cols and to be 54 mGy m
2
for the paediatric proto-
col. The effective dose calculated with tissue
weighing factors from the ICRP Publication 103 was
0.05 mSv for children and 0.090.16 mSv for adults.
These values are 2427 % higher than those calcu-
lated with the factors from the ICRP Publication 60.
The reason is the inclusion of the salivary glands
and brain as individually weighted tissue with a
factor 0.01 and the addition of the oral mucosa to
the remaining organs with a factor of 0.12.
The calculated effective doses are lower than those
calculated by Ludlow for the same CBCT system on
the basis of phantom measurements of organ
doses
(4)
. Further verication of the results is needed
with a comparison of calculations and measure-
ments for the same system.
CONCLUSION
The present study demonstrated the applicability of
some of the QC tests to the CBCT for conventional
fan-beam CT, for panoramic dental systems and for
conventional radiography. The following additional
tests, not performed here, could also be included for
testing the at panel detector quality: dark noise,
image retention and resolution. In addition, image
quality tests are recommended, but for that purpose
a dedicated phantom should be available.
REFERENCES
1. Kau, C. H., Richmond, S., Palomo, J. M. and Hans,
M. G. Three-dimensional cone beam computerized tom-
ography in orthodontics. J. Orthod. 32, 28293 (2005).
2. Scarfe, W., Farman, A. and Sukovic, P. Clinical appli-
cations of cone beam computed tomography in dental
practice. J. Can. Dent. Assoc. 72(1), 7280 (2006).
3. Scarfe, W. and Farman, A. Cone beam computed tom-
ography: a paradigm shift for clinical dentistry. Aus.
Dent. Pract. 102110 (2007).
4. Ludlow, J. B. and Ivanovic, M. Comparative dosimetry
of dental CBCT devices and 64-slice CT for oral and
maxillofacial radiology. Oral Surg. Oral Med. Oral
Pathol. Oral Radiol. Endod. 106, 106114 (2008).
5. Roberts, J. A., Drage, N. A., Davies, J. and Thomas,
D. W. Effective dose from cone beam CT examinations
in dentistry. Br. J. Radiol. 82, 3540 (2009).
6. Horner, K., Islam, M., Flygare, L., Tsiklakis, T. and
Whaites, E. Basic principles for use of dental cone beam
CT: consensus guidelines of the European Academy of
Dental and Maxillofacial Radiology. Dentomaxillofac.
Radiol. 38, 187195 (2009).
7. Radiation Protection: Cone beam CT for dental and
maxillofacial radiology. Provisional guideliness. Vol. 1.
A report prepared by the SEDENTEX CT project.
(2009) Available on www.sedentexct.eu.
8. Pryor, M. and Hollaway, P. Dental cone beam CTa
review of our experience. Ninth CT users group
meeting. Birmingham (2007) Available on http://www.
ctug.org.uk/meet07-10-23/
9. Betancourt Ben tez, R., Ning, R., Conover, D., Zhang,
Y. and Cai, W. Development of a quality control
program for a cone beam CT imaging system. Proc.
SPIE. 6913, DOI:10.1117/12.770427 (2008).
10. Ministry of Health of Republic of Bulgaria. Ordinance
No 30/31.10.2005 for protection of individuals at
medical exposure. Promulgated in the State Gazette
(ofcial organ of the National Assembly of the
Republic of Bulgaria) No 91, November 15 (2005).
(In Bulgarian)
Table 2. Measured KAP and calculated effective doses.
Exposure parameters (protocol) Measured KAP, mGy m
2
E, mSv (ICRP 60) E, mSv (ICRP 103)
3.8 mA; 40 s (standard adult) 184 126 157
3.8 mA; 20 s (low-dose adult) 110 74 94
1 mA; 20 s (paediatric) 54 37 46
J. VASSILEVA AND D. STOYANOV
312

a
t

U
n
i
v
e
r
s
i
d
a
d
e

E
s
t
a
d
u
a
l

P
a
u
l
i
s
t
a

J

l
i
o

d
e

M
e
s
q
u
i
t
a

F
i
l
h
o

o
n

J
u
l
y

3
0
,

2
0
1
3
h
t
t
p
:
/
/
r
p
d
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m