Open Journal of Stomatology, 2013, 3, 292-297 OJST
doi:10.4236/ojst.2013.35049 Published Online August 2013 (http://www.scirp.org/journal/ojst/)
Artefacts in cone beam CT
Prashant P. Jaju, Mayuri Jain, Ajita Singh, Akanksha Gupta
Rishi Raj College of Dental Sciences and Research Centre, Bhopal, India
Received 22 May 2013; revised 23 June 2013; accepted 17 July 2013
Copyright © 2013 Prashant P. Jaju et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Cone beam computed tomography (CBCT) is the
modern third dimension applied in the field of oral
maxillofacial region. With lower radiation dose com-
pared to conventional CT, its applications in dentistry
has increased tremendously. Artefacts can seriously
degrade the quality of computed tomographic (CBCT)
images, sometimes to the point of making them diag-
nostically unusable. To optimize image quality, it is
necessary to understand why artifacts occur and how
they can be prevented or suppressed. CT artifacts ori-
ginate from a range of sources; physical based, scan-
ner based and patient based. This article highlights
the causes of artefacts on CBCT images and methods
to avoid them.
Keywords: CBCT; Artefacts; Metal Artefacts
1. INTRODUCTION
The introduction of Cone Beam Computed Tomography
(CBCT) technology in dentistry is rapidly changing the
diagnostic landscape, allowing dentists to now diagnose
in three dimensions. Present state-of-the-art cone beam
computed tomography (CBCT) units produce excellent
high resolution, three dimensional images of oral bony
anatomy, making dental implant planning and surgical
placement simple and reliable. Also the role of CBCT in
oral & maxillofacial surgery, orthodontics, airway assess-
ment, temporomandibular joint disorders, endodontics and
periodontics is widely described lately [1-6]. The radia-
tion dose required for CBCT is lower than that of CT if
we consider images made for the same purposes [7,8].
In computed tomography (CT), the term artefact is ap-
plied to any systematic discrepancy between the CT
numbers in the reconstructed image and the true attenua-
tion coefficients of the object. CT images are inherently
more prone to artefacts than conventional radiographs be-
cause the image is reconstructed from something on the
order of a million independent detector measurements [9].
The presence of grey level non-uniformities in CBCT
contributes to artifact formation in reconstructed CBCT
images [10]. These artefacts contribute to image degrada-
tion and can lead to inaccurate or false diagnoses. Some
of these artefacts are more pronounced in CBCT units
than their CT counterparts because of the different proc-
esses in which the images are acquired (Figure 1).
Artefact relating to CBCT will be divided into three
main categories, physics-based, patient-based and scan-
ner-based. Physics-based artifacts result from the physic-
cal processes involved in the acquisition of CT data. Pa-
tient-based artifacts are caused by factors related to the
patient’s form or function. Scanner-based artifacts result
from imperfections in scanner function.
2. PHYSICAL BASED ARTEFACTS
2.1. Noise
Noise is defined as an unwanted, randomly and/or non-
randomly distributed disturbance of a signal that tends to
obscure the signal’s information content from the ob-
server. Noise affects images produced by cone beam CT
units by reducing low contrast resolution, making it dif-
ficult to differentiate low density tissues thereby reduc-
ing the ability to segment effectively. The noise in tradi-
tional projection radiography is primarily from quantum
mottle which is defined as a variation in image density
due to statistical fluctuation of photon fluency in the ra-
diation field. In well-designed X-ray systems, the quan-
tum noise is governed by the number of X-ray photons
absorbed in the detector, the higher the number of pho-
tons absorbed, the lower the quantum mottle. The num-
ber of X-ray photons emitted is directly related to the mA
of the X-ray unit. Another source of noise in computed
tomography is scattered radiation. Scattered radiation
arises from interactions of the primary radiation beam
with the atoms in the object being imaged and its magni-
tude is largely dependent on patient size, shape, and po-
sition in the scan field. It is a major source of image
degradation in X-ray imaging techniques. When X-ray
Published Online August 2013 in SciRes. http://www.scirp.org/journal/ojst

P. P. Jaju et al. / Open Journal of Stomatology 3 (2013) 292-297 293
Figure 1. Artefact causing image degradation.
radiation passes through a p atient, three types of interac-
tions can occur, including coherent scattering, photoelec-
tric absorption and Compton scattering. Compton scat-
tering is the type most seen in diagnostic radiology. In
Compton scattering, the interaction is a co llision between
a high energy X-ray photon and one of the outer shell
electrons of an atom. This outer shell electron is bound
with very little energy to the atom so when the X-ray
photon collides with it, the electron is ejected from the
atom. Because energy and momentum are both con-
served in this collision, the energy and direction of the
scattered X-ray photon depend on the energy transferred
to the electron. If the initial X-ray energy is high, the
relative amount of energy lost is small, and the scattering
angle is small relative to th e initial direction. If th e initial
X-ray energy is small, the scatter angle is large and the
ejected electron disperses in all directions. Quantum
noise is fundamentally related to image quality and is a
function of dose, tissue transmissivity, and voxel size.
Noise is, in turn, a principal determinant of contrast re-
solution and, to a lesser exten t, spatial resolution, which,
along with artifacts, constitute the major observable de-
terminants of overall image quality (Figure 2). CBCT
imaging with flat panel detector (FPD) technology typi-
cally affords excellent spatial resolution with a relatively
low patient dose. Contrast resolution suffers, however,
due to increased X-ray scatter and the reduced temporal
resolution and dynamic range of the FPDs [11]. In-
creased scatter not only amplifies patient dose but is a
principal contributor to reduced contrast resolution and
increased noise in CBCT images [11]. There is very little
noise in conventional CT machines because of the high
mA used and due to effective pre- and post patient colli-
mation which reduces the scattered radiation to a negli-
gible amount. However, in CBCT machines the noise is
high due to the lower mA used and because of the high
Figure 2. Noise reducing image contrast.
amount of scattered radiation since there is no post-pa-
tient collimation.
2.2. Beam Hardening
An X-ray beam is composed of individual photons with a
range of energies. As the beam passes through an object,
it becomes “harder,” that is to say its mean energy in-
creases, because the lower energy photons are absorbed
more rapidly than the higher-energy photons. Two types
of artifact can result from this effect: so-called cupping
artifacts and the appearance of dark bands or streaks
between dense objects in the image [9].
Beam hardening manifests as two different artifacts
within the reconstructed image, a cupping artifact and the
appearance of dark bands or streaks.
Cupping artifacts from beam hardening occur when
X-rays passing through the center of a large object be-
come harder than those passing through the edges of the
object due to the greater amount of material the beam has
to penetrate. Because the beam becomes harder in the
center of the object, the resultant profile of the linear
attenuation coefficients appears as a “cup”. The cupping
effect artifact is demonstrated when a uniform cylindrical
object is imaged. As the effects of beam hardening and
scatter are most prevalent in the centre of a cylindrical
object, it is this area that is dominated by the cupping
effect artifact [10].
The second type of artifact relating to beam hardening
are dark streaks and bands between dense objects in an
image. In dental imaging, this type of artifact can be seen
between two implants located in the same jaw that are in
close proximity to each other. This occurs because the

P. P. Jaju et al. / Open Journal of Stomatology 3 (2013) 292-297 294
portion of the beam that passes through both objects at
certain tube positions becomes harder than when it
passes through only on e of the objects at other tube posi-
tions. McDavid et al. and Brooks and Di Chiro demon-
strated that the cupping effect is caused by beam hard-
ening by reconstructing a uniform object with ideal pro-
jections and observing the absence of the cupping effect
[12,13].
It is well known that an artifact referred to as the
‘‘truncation artifact’’ or ‘‘truncated-view artifact’’ is in-
herent to CBCT imaging. This artifact occurs because the
size of the FOV used in CBCT is smaller than the size of
the object being imaged. The largest error due to the
truncated-view artifact will occur near the edge of the
FOV [14]. Lehr imaged a 52 cm water disc phantom cen-
tred in a 48 cm FOV, which resulted in an increase in CT
numbers at the edge of the FOV [14].
Bryant et al. described a similar observed effect, termed
the “exomass effect”, on the i-CAT (Imaging Sciences
International, LLC, Hatfield, PA) CBCT unit [15].
They also observed an increase in the grey level values
in the anterior to the po sterior direction of the scan field,
the posterior representing the position within the scan
field, adjacent to th e portion of the object located outside
of the FOV. Katsumata et al. evaluated the effect of the
truncation artifact on the Alphard Vega CBCT unit (Asahi
Roentgen, Kyoto, Japan), reporting improved uniformity
of the density values with th e larger FOVs used [16].
Manufacturers minimize beam hardening by using fil-
tration, antiscatter grids, calibration correctio n, and beam
hardening correction softwar e [9].
2.3. Filtration
The use of filtration to decrease beam hardening is sup-
ported by the findings of Brooks and Di Chiro, who
demonstrated a reduction of beam hardening effects from
9.2% in 20 cm of water using a 4.5 mm aluminium pre-
filter to 1.5% using a 3.5 cm aluminium pre-filter and
reported that using high atomic number (Z) materials
such as copper, tin or Thoraeus filters could produce
even better results [13]. In fact, normal aluminium filters
are approximately 10% less efficient than filters of other
materials such as copper, brass or iron [17]. Meganck et
al. reduced the cupping effect caused by beam hardening
on a cortical bone-equivalent phantom to an insignificant
level (2%) using a combination of 0.254 mm aluminium
and 0.254 mm copper filter [10,18].
The bow tie or wedge filter is the prototypical com-
pensating filter used in CBCT systems. It modulates the
beam profile by increasing photon density at the center
of the cone and decrementally reducing density at the
periphery. In the radiation therapy CBCT literature,
Graham et al. were able to demonstrate a 50% reduction
in scatter with the implementation of copper bow tie fil-
ters [19]. Image-quality improvement has been described
with bow tie filters in a CBCT system integrated into the
gantry of a conventional CT scanner as well [20].
Compensating filtration is not without criticism, how-
ever, because beam hardness hasbeen shown to nega-
tively impact detector efficiency, as demonstrated by a
decrease in the ratio of the output signal intensity-to-
noise ratio (SNR) to the entrance exposure (SNR/en-
trance exposure) [21].
2.4. Antiscatter Grids
Antiscatter grids represent an alternative method of di-
rect scatter reduction that has been used with FPDs in
digital radiographic and fluoroscopic imaging for some
time [22]. Rather than modulating the beam properties at
the source, a grid of lead leaves is fitted over the detector
to preferentially absorb off-axis radiation not contribut-
ing to primary photon fluence.
In CBCT systems, the lead leaves are arranged in a ra-
dial pattern centered on the focal spot of the FPD.
Antiscatter grids have been evaluated in several experi-
mental CBCT systems with mixed results [11,23,24]. A
reduction in both cupping artifact and overall scatter has
been observed, though there may be insufficient im-
provement in contrast and observed image quality to
warrant use except in situations of high scatter [11].
Siewerdsen et al. evaluated antiscatter grids in a linear
accelerator-coupled CBCT system and found that image
quality improved only in situations of high scatter such
as with a large FOV s covering a large anatomic site or in
input quantum-limited situations such as with high dose
or low spatial resolution [23]. To the exten t that antiscat-
ter grids improve soft-tissue contrast and artifacts, they
also increase noise, which leads to a degradation in over-
all image quality. An escalation in dose or reduction in
spatial resolution is needed to offset the increased noise
with the implementation of grids. For a relatively small
FOVs, such as that used in a targeted head and neck scan,
antiscatter grids may improve image contrast and reduce
cupping artifacts, but the increased noise requires that the
dose be increased or spatial resolution be decreased to
produce a high-quality image.
2.5. Calibration
Calibration correction: Manufactu rers calibrate their sca-
nners using phantoms in a range of sizes. This allows the
detectors to be calibrated with compensation tailored for
the beam hardening effects of different parts of the pa-
tient [9].
2.6. Software Corrections
Beam hardening correction software: An iterative correc-

P. P. Jaju et al. / Open Journal of Stomatology 3 (2013) 292-297 295
tion algorithm may be applied when images of bony re-
gions are being reconstructed. This helps minimize blur-
ring of the bone-soft tissue interface in brain scans and
also reduces the appearance of dark bands in nonho-
mogeneous cross sections [9].
2.7. Scatter Correction Algorithms
Some sort of scatter subtraction or homogenization pre-
processing algorithm is used in most clinical CBCT sys-
tems [25,26]. Several approaches have been studied, in-
cluding Monte Carlo simulations, blocker-based or beam-
stop techniques, analytic calculations, and collimator sha-
dow estimation [27]. Perhaps the most theoretically ro-
bust algorithm is that based on the Monte Carlo simula-
tion, which predicts scatter on the basis of a voxel den-
sity model of the entire acquired tissue volume during
preprocessing. The predicted scatter contribution at each
detector element is then subtracted before reconstruction.
Monte Carlo simulations still require significant compu-
tation time, however, which has fueled continued re-
search in other algorithmic approac hes.
2.8. Partial Volume Artefacts
The algorithms used in CT data reconstruction assume
that the object is completely covered by the detector at
all view angles, and that the attenuation is caused by the
object only. When this situation does not occur, recon-
structed CT images can contain a truncated-view artifact.
In conventional CT units, this is not a problem as the
entire object is always within the field of view of the unit,
however it does affect CBCT units due to their limited
FOV. This occurs because some of the cone beam data
penetrating portions of the object other than the region-
of-interest (ROI) are missing because of the insufficient
size of the detector. When the entire volume is not
covered by the detector, shading artifacts can be visual-
ized. Another consequence of the partial volume artifact
is that the true linear attenuation coefficients cannot be
calculated because some of the X-ray paths penetrate
other portions of the object as well as the region of in-
terest and the data collected no longer represent this area
exclusively but are co rrupted by structures outside of the
FOV. This issue has a greater affect in machines that
have smaller FOVs as opposed to those that have larger
FOVs. Currently, alg orithms attempt to coun ter this issue
by estimating the remaining linear attenuation coeffi-
cients for the areas that are not completely imaged. Al-
though there is improvement in HU precision, this still
has not enabled accurate calculation of Hounsfield units.
Many methods are currently being developed and tested
to alleviate this issue. Partial volume artifacts can best be
avoided by using a thin acquisition section width [9].
3. PATIENT-BASED ARTEFACTS
3.1. Metal Artefacts
The presence of metal objects in the scan field can lead
to severe streaking artifacts. They occur because the den-
sity of the metal is beyond the normal range that can be
handled by the computer, resulting in incomplete at-
tenuation profiles. Metallic objects such as dental resto-
rations, surgical plates, dental implants and pins and ra-
diographic markers can cause this type of the artifact.
Since the metal in these materials highly attenuate the
X-ray beam, the attenuation values of objects behind the
object are incorrectly high (Figure 3). Due to the recon-
struction of the cone beam image, the metal causes the
effect of bright and dark streaks in CT images which
significantly degrade the image quality. In conventional
CT images metallic artifacts traverse the object in the
direction of the gantry and only at the level of the high
attenuation object. In CBCT, the metallic streak artifacts
occur in all directions from the high attenuation object
because of the cone-s haped beam (Figure 4).
3.2. Software Corrections for Metal Artifacts
Streaking caused by overranging can be greatly reduced
by means of special software corrections. Manufacturers
use a variety of interpolation techniques to substitute the
overrange values in attenuation profiles. MARS (Metal
artifact reduction software) provided by Sirona, (Ger-
many) is one such metal reduction software which im-
proves the quality of image. The usefulness of metal ar-
tifact reduction software is sometimes limited because,
although streaking distant from the metal implants is
removed, there still remains a loss of detail around the
Figure 3. Metal restoration art e fact seen on axia l image.

P. P. Jaju et al. / Open Journal of Stomatology 3 (2013) 292-297 296
(a)
(b)
Figure 4. Implant artefact on axial and cross sec-
tional image.
metal-tissue interface, which is often the main area of
diagnostic interest. Beam hardening correction software
should also be used when scanning metal objects to
minimize the additional artifacts due to beam hardening
[9].
4. MOTION ARTEFACTS
Patient motion can cause misregistration artifacts within
the image. Because of the relatively long acquisition
times (compared to conventional radiography) and volu-
metric image acquisition, motion artefacts are common
in CBCT. These artefacts can be attributed to improper
patient stabilization . Small motions cause image blurring
and larger physical displacements.
They produce artifacts that appear as double images or
ghost images. This results in poor overall image quality.
Since the resolutions of the present CBCT are very high,
ranging from 0.08 mm - 0.4 mm, even small motions can
have a detrimental effect on image quality. The use of
positioning aids is sufficient to prevent voluntary move-
ment in most patients.
5. SCANNER-BASED ARTEFACTS
Ring Artefacts
If one of the detectors is out of calibration on a scanner,
the detector will give a consistently erroneous reading at
each angular position, resulting in a circular artefact. A
scanner with solid-state d etectors, where all the detectors
are separate entities, is in principle more susceptible to
ring artefacts. Rings visible in a uniform phantom or in
air might not be visible on a clinical image if a wide
window is used. Even if they are visible, they would
rarely be confused with disease. However, they can im-
pair the diagnostic quality of an image, and this is par-
ticularly likely when central detectors are affected, cre-
ating a dark smudge at the center of the image. Currently
there is no evidence of ring artefacts on CBCT machine
in dental radiology literature [9].
6. CONCLUSION
Artefacts originate from a range of sources and can de-
grade the quality of a CBCT images to varying degrees.
Design features incorporated into modern machines mi-
nimize some types of artifact, and some can be partially
corrected by the scanner software like MARS. However,
there are many instances where careful patient position-
ing and the optimum selection of scan parameters are the
most important factors in avoiding image artefact.
REFERENCES
[1] Guerrero, M.E., Jacobs, R., Loubele, M., et al. (2006)
State-of-the-art on cone beam CT imaging for preopera-
tive planning of implant placement. Clinical Oral Inves-
tigations, 10, 1-7. doi:10.1007/s00784-005-0031-2
[2] Drage, N.A. and Sivarajasingam, V. (2009) The use of
cone beam computed tomography inthe management of
isolated orbital floor fractures. British Journal of Oral
and Maxillofacial Surgery, 47, 65-66.
doi:10.1016/j.bjoms.2008.05.005
[3] Holberg, C., Steinhauser, S., Geis, P., et al. (2005) Cone-
beam computed tomography in orthodontics: Benefits
and limitations. Journal of Orofacial Orthopedics, 66,
434-444. doi:10.1007/s00056-005-0519-z
[4] Garrett, B.J., Caruso, J.M., Rungcharassaeng, K., et al.
(2008) Skeletal effects to the maxilla after rapid maxil-

P. P. Jaju et al. / Open Journal of Stomatology 3 (2013) 292-297 297
OJST
lary expansion assessed with cone-beam computed tomo-
graphy. American Journal of Orthodontics and Dento-
facial Orthopedics, 134, 8-9.
doi:10.1016/j.ajodo.2008.06.004
[5] Misch, K.A., Yi, E.S. and Sarment, D.P. (2006) Accuracy
of cone beam computed tomography for periodontal de-
fect measurements. Journal of Periodontology, 77, 1261-
1266. doi:10.1902/jop.2006.050367
[6] Honda, K., Arai, Y., Kashima, M., et al. (2004) Evalua-
tion of the usefulness of the limited cone-beamCT(3DX)
in the assessment of the thickness of the roof of the gle-
noid fossa of the temporomandibular joint. Dentomaxil-
lofacial Radiology, 33, 391-395.
doi:10.1259/dmfr/54316470
[7] Bianchi, S., Anglesio, S., Castellano, S., Rizzi, L. and
Ragona, R. (2001) Absorbed doses and risk in implant
planning: Comparison between spiral CT and cone beam
CT. Dentomaxillofacial Radiology, 30, S28.
[8] Tsiklakis, K., Donta, C., Gavala, S., Karayianni, K., Ka-
menopoulou, V. and Hourdakis, C.J. (2005) Dose reduc-
tion in maxillofacial imaging using low dose cone beam
CT. European Journal of Radiology, 56, 413-417.
doi:10.1016/j.ejrad.2005.05.011
[9] Barrett, J.F. and Keat, N. (2004) Artifacts in CT: Recog-
nition and avoidance. Radiographics, 24, 1679-1691.
doi:10.1148/rg.246045065
[10] Hunter, A.K. and McDavid, W.C. (2012) Characteriza-
tion and correction of cupping effect artefacts in cone
beam CT. Dentomaxillofacial Radiology, 41, 217-223.
doi:10.1259/dmfr/19015946
[11] Miracle, A.C. and Mukerji, S.K. (2009) Conebeam CT of
the head and neck, Part 1: Physical principles. American
Journal of Neuroradiology, 30, 1088-1095.
doi:10.3174/ajnr.A1653
[12] McDavid, W.D., Waggener, R.G., Payne, W.H. and Den-
nis, M.J. (1975) Spectral effects on three-dimensional re-
construction from x-rays. Medical Physics, 2, 321-324.
doi:10.1118/1.594200
[13] Brooks, R.A. and Di Chiro, G. (1976) Beam hardening in
X-ray reconstructive tomography. Physics in Medicine
and Biology, 21, 390-398.
doi:10.1088/0031-9155/21/3/004
[14] Lehr, J.L. (1983) Truncated-view artifacts: Clinical im-
portance on CT. American Journal of Roentgenology, 141,
183-191. doi:10.2214/ajr.141.1.183
[15] Bryant, J.A., Drage, N.A. and Richmond, S. (2008) Study
of the scan uniformity from an i-CAT cone beam CT
dental imaging system. Dentomaxillofacial Radiology, 37,
365-374. doi:10.1259/dmfr/13227258
[16] Katsumata, A., Hirukawa, A., Okumura, S., Naitoh, M.,
Fujishita, M., Ariji, E., et al. (2009) Relationship between
density variability and imaging volume size in cone-beam
computerized tomographic scanning of the maxillofacial
region: An in vitro study. Oral Surgery, Oral Medicine,
Oral Pathology, Oral Radiology, and Endodontics, 107,
420-425. doi:10.1016/j.tripleo.2008.05.049
[17] Jennings, R.J. (1988) A method for comparing beam-
hardening filter materials for diagnostic radiology. Me-
dical Physics, 15, 588-599. doi:10.1118/1.596210
[18] Meganck, J.A., Kozloff, K.M., Thornton, M.M., Broski,
S.M. and Goldstein, S.A. (2009) Beam hardening arti-
facts in micro-CT scanning can be reduced by X-ray
beam filtration and the resulting images can be used to
accurately measure BMD. Bone, 45, 1104-1116.
doi:10.1016/j.bone.2009.07.078
[19] Graham, S.A., Moseley, D.J., Siewerdsen, J.H., et al.
(2007) Compensators for dose and scatter management in
cone-beam computed tomography. Medical Physics, 34,
2691-2703. doi:10.1118/1.2740466
[20] Gupta, R., Grasruck, M., Suess, C., et al. (2006) Ultra-
high resolution flat-panel volume CT: Fundamental prin-
ciples, design architecture, and system characterization.
European Radiology, 16, 1191-1205.
doi:10.1007/s00330-006-0156-y
[21] Ning, R., Chen, B., Yu, R., et al. (2000) Flat panel detec-
tor-based cone-beam volume CT angiography imaging:
System evaluation. IEEE Transactions on Medical Imag-
ing, 19, 949-963. doi:10.1109/42.887842
[22] Neitzel, U. (1992) Grids or air gaps for scatter reduction
in digital radiography: A model calculation. Medical Phy-
sics, 19, 475-481. doi:10.1118/1.596836
[23] Siewerdsen, J.H., Mose ley, D. J., Bakhtiar, B., et al. (2004)
The influence of antiscatter grids on soft-tissue detec-
tability in cone-beam computed tomography with flat-
panel detectors. Medical Physics, 31, 3506-3520.
doi:10.1118/1.1819789
[24] Nickoloff, E.L., Lu, Z.F., Dutta, A., et al. (2007) Influ-
ence of flat-panel fluoroscopic equipment variables on
cardiac radiation doses. CardioVascular and Interven-
tional Radiology, 30, 169-176.
doi:10.1007/s00270-006-0096-6
[25] Orth, R.C., Wallace, M.J. and Kuo, M.D. (For the Tech-
nology Assessment Committee of the Society of Interven-
tional Radiology) (2008) C-arm cone-beam CT: General
principles and technical considerations for use in inter-
ventional radiology. Journal of Vascular and Interven-
tional Radiology, 19, 814-820.
doi:10.1016/j.jvir.2008.02.002
[26] Dorfler, A., Struffert, T., Engelhorn, T., et al. (2008) Ro-
tati onal flat-panel computed tomography in diagnostic
and interventional neuroradiology. Rofo, 180, 891-898.
[27] Jarry, G., Graham, S.A., Moseley, D.J., et al. (2006) Cha-
racterization of scattered radiation in kV CBCT images
using Monte Carlo simulations. Medical Physics, 33, 4320-
4329. doi:10.1118/1.2358324