A r t i f a c t s In t e r f e r i n g w i t h

I n t e r p ret a t i o n o f C o n e
B e a m C o m p u t e d Tom o g r a p h y
Images
Scott R. Makins, DDS

KEYWORDS
 Cone beam computed tomography  Artifacts  Beam hardening  Metal artifact
 Image noise

KEY POINTS
 Artifacts in radiographic imaging are discrepancies between the reconstructed visual
image and the actual content of the subject being studied.
 In radiographic imaging, the discrepancies mean the grayscale values in the image do not
accurately reflect the attenuation values of the subject.
 Structures may also appear that do not actually exist in the subject. Whatever the source
or appearance of image artifacts, their presence degrades the accuracy of the image in
relation to the true characteristics of the subject.
 It is beneficial to be aware of the presence of artifacts and be familiar with their character-
istic appearances in order to enhance the extraction of diagnostic information from cone
beam images.

IMAGE ARTIFACTS COMMONLY SEEN IN CONE BEAM IMAGING

Artifacts in radiographic imaging are discrepancies between the reconstructed visual

image and the actual content of the subject being studied. In radiographic imaging,
this means the grayscale values in the image do not accurately reflect the attenuation
values of the subject. Additionally, structures may appear within images that do not
actually exist in the subject. Such structures can occur because of patient motion,
the image capture and reconstruction process, or malfunctions of the imaging
system.1–4
Artifacts that are inherent in the image capture and reconstruction processes
involved in cone beam imaging can often be reduced by the use of image-
modifying algorithms or by altering the image capture process. Whatever the source
or appearance of image artifacts, their presence degrades the accuracy of the image
in relation to the true characteristics of the subject. It is therefore beneficial to be aware

Department of Comprehensive Dentistry, University of Texas Health Science Center at San

Antonio, MSC 7919, 7703 Floyd Curl Drive, San Antonio, TX 78229-3901, USA
E-mail address: makins@uthscsa.edu

Dent Clin N Am 58 (2014) 485–495

http://dx.doi.org/10.1016/j.cden.2014.04.007 dental.theclinics.com
0011-8532/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
486 Makins

of the presence of artifacts and be familiar with their characteristic appearances in or-
der to enhance the extraction of diagnostic information from cone beam images.5–8
Radiographic imaging is all about the attenuation of the x-ray beam as it passes
through the subject of interest. The x-ray photons interact with the atoms within the sub-
ject in a variety of ways and can be removed or diverted from the beam’s path before it
exits the subject. Such interactions reduce the number of x-ray photons that strike the
image detector used to generate the radiographic image. Metal restorations are
composed of materials with high atomic numbers that attenuate the x-ray beam more
than materials with low atomic numbers. Bone is composed of materials with moder-
ately high atomic numbers, and therefore it attenuates the beam more than soft tissues.
The most common and annoying artifacts seen in cone beam images are white and
dark streaks or light flares originating from dense structures within the image. Such
artifacts can originate from multiple sources and have similar appearances. They often
overlie areas of diagnostic importance within the image and therefore present image
interpretation challenges. It is important to keep in mind that the artifacts seen in cone
beam imaging are often related to the geometric process of the x-ray tube head and
detector rotating about the subject used to capture the multiple basis images from
which 3-dimensional volumes are reconstructed.1–4

BEAM HARDENING ARTIFACT

Among the most obvious image artifacts are dark bands caused by a process known
as beam hardening. The x-ray beam used in cone beam imaging is termed polychro-
matic. This means the x-ray beam is not composed of x-ray photons of a single energy
level. Instead the beam is made up of x-ray photons of many different energy levels,
the maximum energy level being equal to the kVp setting selected on the cone
beam unit. As the primary beam passes through the subject, the lower energy x-ray
photons are more likely to interact with the subject, especially with denser objects
such as bone or metal restorations, and they are attenuated or removed from the x-
ray beam. The portion of the x-ray beam that exits from the subject and strikes the de-
tector is therefore composed of a greater percentage of high-energy x-ray photons
than the beam originating at the tube head.1–6
For the sake of simplicity and to provide reduced cost with faster image reconstruc-
tion times, the mathematical algorithm that processes the information from the detec-
tor and reconstructs the cone beam image assumes the primary x-ray beam exiting
the tube head is monochromatic, (ie, composed of x-ray photons of a single energy
level). This discrepancy between the assumed and true photon energy spectrum of
the initial x-ray beam causes the algorithm to misinterpret the amount of attenuation
experienced by the beam as it passed through the subject, especially the denser ob-
jects mentioned above. The algorithm thinks little, if any, attenuation of the primary
beam occurred and it assigns incorrectly low gray scale values for the areas within
the subject. This under valuation combined with the back projection mathematical
process to construct the image volume produces the characteristic dark areas and
streaks seen in cone beam images. These artifacts may project over and mask under-
lying structures, or they may provide false information regarding the density and
morphology of those areas within the subject (Figs. 1–3).6–12

METAL ARTIFACT

White streaks, often referred to as star effect or metallic artifact, can also be seen in
association with metallic structures such as crowns, surface restorations, implants,
and surgical plates or screws. This artifact results as the algorithm attempts to
 Interferences with CBCT Image Interpretation 487

Fig. 1. Beam hardening artifact adjacent a silver point in #9 and metal artifact streaks from
the metal coping.

Fig. 2. Beam hardening artifact associated with a gutta percha root canal filling.
488 Makins

Fig. 3. Beam hardening artifact resembling a root fracture produced by a plastic carrier
without gutta percha.

reconstruct an object of high atomic number, which greatly attenuates the x-ray beam,
into the 3-dimensional image volume from the information contained in the large num-
ber of 2-dimensional basis images generated by the rotational geometry of the tube
head and detector around the subject.5–14
As with beam hardening artifacts, metal artifacts can mask or misrepresent the
character of structures within the image that are located adjacent highly attenuating
materials and can project across the image field of view to cause inaccurate grayscale
values for areas not immediately adjacent the highly attenuating material. Several
types of mental artifact reduction software have been developed in an attempt to elim-
inate or reduce the effects of mental artifact within the image volume. Most of these
software programs evaluate the grayscale values of the voxels just outside the areas
of the artifacts and postulate substitute grayscale values to be assigned to the voxels
within the affected regions. Although such processes produce a more aesthetic im-
age, the information substituted for the areas that appear as artifact are not true rep-
resentations of the attenuation values for the related areas in the subject but are
essentially best guesses.6,12–14 The masking of areas beyond the metal object can
be reduced by placing a cotton roll between the anterior teeth and raising or lowering
the patient’s chin, so the artifact streaks are not projected over the entire arch as they
are when the occlusal plane is parallel to the floor (see Fig. 1; Figs. 4–6).15

Fig. 4. Metal artifact streaks from posts in #23 and #26. Note unfilled mesiolingual canal in
#30.
 Interferences with CBCT Image Interpretation 489

Fig. 5. Metal artifact from metal copings.

RING ARTIFACTS

Ring artifacts are generated when a dexel in the detector element is defective or mis-
aligned with the x-ray source. The dexel generates a faulty attenuation value as it
travels the arc around the subject, thereby producing the ring-like artifact. These arti-
facts are commonly seen in axial projections and are centered about the rotational axis
of the imaging system (Figs. 7 and 8).16

CONE BEAM EFFECT ARTIFACT

The cone beam effect produces artifacts seen at the peripheries of the superior and
inferior portions of the cone beam image. Because of the diverging cone shape of
the x-ray beam, peripheral areas at the superior and inferior aspects of the subject
along the z-axis are covered by the beam only when the x-ray source is positioned
on the opposite side of the subject. Areas that are located close to the z-axis passing
through the subject fall within the beam during the entire rotational period. The periph-
eral areas will appear to be less dense (darker) and contain more noise (grainier) than
the centrally located areas (Fig. 9).1–3

Fig. 6. Metal artifact streaks from orthodontic hardware.

490 Makins

Fig. 7. Ring artifact caused by calibration error.

Fig. 8. Ring artifact caused by a dead dexel in the detector.

Fig. 9. Cone beam effect artifact.

 Interferences with CBCT Image Interpretation 491

MOTION ARTIFACT

Another commonly observed artifact in dental cone beam imaging is associated with
patient motion. This artifact may appear as shading or streaking in the reconstructed
image, double outlines of corticated surfaces, or double outlines of the posterior border
of the tongue. Many units incorporate built-in chairs with bands placed across the pa-
tient’s forehead to reduce motion during scans. The units based on panoramic machine
platforms usually have paddles that are positioned against the patient’s temples or an
open face plate on which the patient rests his or her forehead. Having the patient sit in a
chair and lowering the unit to the patient will reduce the chances of motion.
Additionally, some units have variable scan times. A reduction in scan time, by
decreasing the scan arc, reduces the number of basis images available to reconstruct
the 3-dimensional volume, which reduces overall image quality. However, the result-
ing image is compatible with many diagnostic tasks and is a good option for patients
with involuntary movements (Figs. 10–12).1–4

ALIASING ARTIFACT

Aliasing artifacts appear as slightly wavy lines that diverge outwards toward the
periphery of a cone beam image. Another name for their appearance is Moire pattern.
They are caused by undersampling of structures within the subject by the cone beam
unit’s detector. This process is related to the size of the dexels within the detector, the
spatial location of the image slice in relation to the x-ray source and detector, the num-
ber of basis images generated during the scan, and the mathematical algorithms used
to reconstruct the 3-dimensional images. Fortunately, these artifacts do not resemble
any naturally occurring structures, but they can be confusing when trying to detect
root fractures. Dexels are the components within the detector that measure the energy
of the incident x-ray or light photons. They are commonly referred to as pixels, but true
pixels are used to display an image, hence the designation as a picture element (Figs.
13–15).1–4

NOISE ARTIFACT

Trying to isolate the words of a friend’s conversation while one is visiting in a busy
coffee shop can be challenging. Similarly, trying to locate the boundaries of objects
within a radiographic image can be complicated by noise in the image. Image noise

Fig. 10. Blurring and double cortices caused by motion artifact.

492

Fig. 11. Double cortices caused by patient motion.

Fig. 12. Motion artifact from swallowing.

Fig. 13. Aliasing pattern artifact.

 Interferences with CBCT Image Interpretation 493

Fig. 14. Artifact from an over-the-ear hearing aid outside the field of view. Note the prom-
inent palatal torus.

can originate from several sources, but a major form of image noise is referred to as
quantum mottle or sometimes unstructured noise. Random variation in the number
of x-ray photons in the beam as it exits an object and strikes the image detector pro-
duces a grainy or mottle appearance within the image. Increasing the number of pho-
tons in the beam as it exits the tube head would allow for more photons to exit the
subject, strike the receptor, and reduce the grainy appearance. However, increasing
the mA level to increase the number of incident photons also increases the exposure
to the patient. Increasing the voxel size can reduce the grainy appearance and
improve the contrast resolution of the image. However, at the same time, spatial res-
olution and the detection of very small objects may be somewhat reduced, but the
important determining factor is the diagnostic task to be performed and the pertinent
information needed (Fig. 16).15,17–22
No imaging modality—periapical, panoramic, cephalometric, cone beam, medical
computed tomography, or magnetic resonance image—is completely free of distor-
tion or artifacts. Therefore, no system provides images that perfectly represent the
subject of interest. Each system provides diagnostic information based on its

Fig. 15. Artifact from hoop earrings outside the field of view.
494 Makins

Fig. 16. Image noise as seen in 0.076 mm3 and 1.3 mm3 voxel size reconstruction images.

strengths and in spite of its disadvantages. What is key for the dental professional is
determining what diagnostic task needs to be performed and which imaging modality
will provide optimal diagnostic information to fulfill that task. The gathering of informa-
tion must then be conducted in such a manner as to maximize the beneficial outcome
for the patient while minimizing the patient’s risk. It is incumbent on the dental profes-
sional to obtain a thorough knowledge of the strengths and weaknesses of radio-
graphic imaging modalities, the proper operation of the various systems, and the
knowledge to interpret the radiographic images to obtain the maximum information.

REFERENCES

1. Schultze R, Heil U, Grob D, et al. Artifacts in CBCT: a review. Dentomaxillofac

Radiol 2001;40:265–73.
2. Barrett JF, Keat N. Artifacts in CT: recognition and avoidance. Radiographics
2004;24(6):1679–91.
3. Katsumata A, Hirukawa A, Noujeim M, et al. Image artifact in dental cone-beam
CT. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006;101(5):652–7.
4. Miracle AC, Mukherj SK. Cone-beam CT of the head and neck, Part 1: physical
principles. AJNR Am J Neuroradiol 2009;30:1088–95.
5. Esmaeii F, Johari M, Haddad P, et al. Beam hardening artifacts: comparison
between two cone beam computed tomography scanners. J Dent Res Dent
Clin Dent Prospects 2012;6(2):49–53.
6. Hunter AK, McDavid WD. Characterization and correction of cupping artifact in
cone beam CT. Dentomaxillofac Radiol 2012;41:217–23.
7. Kataoka ML, Hochman NG, Rodriguez EK, et al. A review of factors that affect
artifact from metallic hardware on multi-row detector computed tomography.
Curr Probl Diagn Radiol 2010;39(4):125–36.
8. Naitoh M, Saburi K, Gotoh K, et al. Metal artifacts from posterior mandibular
implants as seen in CBCT. Implant Dent 2013;22(2):151–4.
9. Karimi S, Cosman P, Wald C, et al. Segmentation of artifacts and anatomy in CT
metal artifact reduction. Med Phys 2012;39(10):5857–68.
10. Benic G, Sancho-Puchades M, Jung R, et al. In vitro assessment of artifacts
induced by titanium dental implants in cone beam computed tomography. Clin
Oral Implants Res 2013;24:378–83.
11. Schulze R, Berndt D, d’Hoedt B. On cone beam computed tomography artifacts
induced by titanium implants. Clin Oral Implants Res 2010;21:100–7.
 Interferences with CBCT Image Interpretation 495

12. Bechara B, Moore WS, McMahan CA, et al. Metal artifact reduction with cone-
beam computed tomography: an in vitro study. Dentomaxillofac Radiol 2012;
41(3):248–53.
13. Pauwels R, Stamatakis H, Bosmans H, et al. Quantification of metal artifacts in
cone beam CT. Dentomaxillofac Radiol 2012;41:217–33.
14. Boas FE, Fleischmann D. Evaluation of two iterative techniques for reducing
metal artifacts in computed tomography. Radiology 2011;259(3):894–902.
15. Lucklow M, Deyhle H, Beckmann F, et al. Tilting the jaw to improve image quality
or to reduce the dose in cone-beam computed tomography. Eur J Radiol 2011;
80:e389–93.
16. Abu Anas EM, Kim JG, Lee SY, et al. High-quality 3D correction of ring and
radiant artifacts in flat panel detector-based cone beam CT imaging. Phys Med
Biol 2011;56:6495–519.
17. Katsumata A, Hirukawa A, Okumura S, et al. Relationship between density vari-
ability and imaging volume size in cone beam computerized tomographic scan-
ning of the maxillofacial region: an in vitro study. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 2009;107(3):420–5.
18. Bechara B, McMahan CA, Geha H, et al. Evaluation of a cone beam CT artifact
reduction algorithm. Dentomaxillofac Radiol 2012;41:422–8.
19. Bechara B, McMahan CA, Moore WS, et al. Contrast-to-noise ratio with different
large volumes in a cone-beam computerized tomography machine: an in vitro
study. Oral Surg Oral Med Oral Pathol Oral Radiol 2012;114(5):658–65.
20. Bechara B, McMahan CA, Moore WS, et al. Contrast-to-noise ratio difference in
small field of view cone beam computed tomography machines. J Oral Sci
2012;54(3):227–32.
21. Braun H, Kyriakon Y, Kachelriess M, et al. The influence of the heel-effect in cone-
beam computed tomography: artifacts in standard and novel geometries and
their correction. Phys Med Biol 2010;55:6005–21.
22. Pauwels R, Beinsberger J, Stamatakis H, et al. Comparison of spatial and
contrast resolution for cone-beam computed tomography scanners. Oral Surg
Oral Med Oral Pathol Oral Radiol 2012;114(1):127–35.