ICRP Publication 129

Guest Editorial
CBCT: WIDE RANGE OF CLINICAL APPLICATIONS AND WIDE
RANGE OF DOSES

This publication on radiological protection in cone beam computed tomography

(CBCT) is both timely and practical. Previous ICRP publications have addressed mana-
ging the dose in patients undergoing computed tomography (Publication 87), and mana-
ging patient dose in multi-detector computed tomography (MDCT) (Publication 102)
(ICRP, 2000a, 2007a). While some of the same principles apply to CBCT, new chal-
lenges exist. CBCT is a technology that is becoming more prevalent in clinical practice.
These scanners extend the use of computed tomography (CT) into new clinical envir-
onments and by new practitioners, not all of whom have had the radiation safety
training that radiological technologists, medical physicists, radiologists, and radiation
oncologists have traditionally received. In addition, this technology is in evolution, and
opportunities exist for device manufacturers to standardise dose displays. Thus, there is
a need for education, guidelines, and standardisation in the industry.

Over the past decades, conventional CT has been ‘game-changing’ technology in

patient care. Abdominal CT scans have replaced more invasive surgical procedures.
Head CT scans and CT angiography have replaced many catheter-based angiograms.
In a survey of 235 internists, CT and magnetic resonance imaging ranked highest on
the list of innovations whose loss would have the greatest adverse effect on patients
(Fuchs and Sox, 2001). CT ranked ahead of several mainstream medical technologies
such as gastrointestinal endoscopy, balloon angioplasty, and coronary artery bypass
grafts. However, use of CT imaging comes with responsibility. Knowledge and tools
are needed to balance the benefits and the harms. Now, as CBCT is used in other
clinical areas by other providers, the need for education, standardisation, and guide-
lines presents an opportunity that must be met.

The use of CBCT spans a wide range of clinical specialties and procedures: radio-
therapy; orthopaedics; urology; dental/maxillofacial; neurointerventions; and vascu-
lar and non-vascular interventions. Patient dose also demonstrates a large range,
from <1 mGy for organ absorbed dose to >400 mGy for skin dose. This range could
be even wider depending on the number of CBCT scans performed and the com-
plexity of intervention. This publication provides practical guidance for dose man-
agement in general and in specific clinical settings.

Dose reduction comes with trade-offs. This will require standard measurements of
both dose and image quality across all manufacturers. For equipment used for

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

fluoroscopy and CBCT, the aggregate dose to the patient for the entire procedure
should be available. These data should be displayed on the operator console, and
should be available for incorporation into the electronic health record. There are
many ways to reduce dose. These include the design of CBCT equipment and how
the equipment is used in specific clinical settings. This publication appropriately
reinforces fundamental concepts such as ‘as low as reasonably achievable’.
However, this publication also addresses another critical source of dose reduction.
One of the best ways to reduce dose is to ensure that the imaging is appropriate or
clinically indicated. Conventional CT examinations may be performed that do not
represent the most appropriate imaging test for the clinical question being asked.
Multiple strategies have been developed to address unnecessary use of imaging over-
all (Bernardy et al., 2009). Guidelines on appropriate use of CBCT need to be widely
adopted.

In an era of population health, further development and use of CBCT should be

driven by appropriate clinical need, balanced with risk to patients and workers. This
publication provides a helpful direction for policy makers, imaging professionals,
medical physicists, and manufacturers to optimise protection of both patients and
workers, while preserving the expectation of high diagnostic yields from imaging and
excellent clinical outcomes.

JAMES V. RAWSON

REFERENCES
ICRP, 2000. Managing Patient Dose in Computed Tomography. ICRP Publication 87. Ann.
ICRP 30(4).
ICRP, 2007a. Managing patient dose in multi-detector computed tomography (MDCT).
ICRP Publication 102. Ann. ICRP 37 (1).
Fuchs, V.R., Sox, H.C. Jr., 2001. Physician s views of the relative importance of thirty medical
innovations. Health Aff. 20, 30–42.
Bernardy, M., Ullrich, C.G., Rawson, J.V., et al., 2009. Strategies for managing imaging
utilization. J. Am. Coll. Radiol. 6, 844–850.

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Radiological Protection
in Cone Beam Computed Tomography (CBCT)
ICRP PUBLICATION 129

Approved by the Commission in January 2015

Abstract–The objective of this publication is to provide guidance on radiological

protection in the new technology of cone beam computed tomography (CBCT).
Publications 87 and 102 dealt with patient dose management in computed tomog-
raphy (CT) and multi-detector CT. The new applications of CBCT and the asso-
ciated radiological protection issues are substantially different from those of
conventional CT. The perception that CBCT involves lower doses was only true in
initial applications. CBCT is now used widely by specialists who have little or no
training in radiological protection. This publication provides recommendations on
radiation dose management directed at different stakeholders, and covers principles
of radiological protection, training, and quality assurance aspects. Advice on appro-
priate use of CBCT needs to be made widely available. Advice on optimisation of
protection when using CBCT equipment needs to be strengthened, particularly with
respect to the use of newer features of the equipment. Manufacturers should stand-
ardise radiation dose displays on CBCT equipment to assist users in optimisation of
protection and comparisons of performance. Additional challenges to radiological
protection are introduced when CBCT-capable equipment is used for both fluoros-
copy and tomography during the same procedure. Standardised methods need to be
established for tracking and reporting of patient radiation doses from these proced-
ures. The recommendations provided in this publication may evolve in the future as
CBCT equipment and applications evolve. As with previous ICRP publications, the
Commission hopes that imaging professionals, medical physicists, and manufac-
turers will use the guidelines and recommendations provided in this publication
for implementation of the Commission’s principle of optimisation of protection of
patients and medical workers, with the objective of keeping exposures as low as
reasonably achievable, taking into account economic and societal factors, and con-
sistent with achieving the necessary medical outcomes.
ß 2015 ICRP. Published by SAGE.

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Keywords: Cone beam CT; C-arm CBCT; ICRP recommendations; Dose manage-
ment CBCT; Interventional CBCT; CT fluoroscopy

AUTHORS ON BEHALF OF ICRP

M.M. REHANI, R. GUPTA, S. BARTLING, G.C. SHARP,
R. PAUWELS, T. BERRIS, J.M. BOONE

10

Downloaded from ani.sagepub.com by guest on July 22, 2015

 PREFACE
The International Commission on Radiological Protection (ICRP) provides recom-
mendations and guidance on application of its principles of radiological protection.
This has been done through specific publications on the use of ionising radiation in
medicine in various imaging and therapeutic modalities. This is in addition to the
reports published by ICRP providing general advice on radiological protection and
safety in medicine through Publication 105 (ICRP, 2007c). Analysis of current tech-
nology from the point of view of radiological protection has resulted in recommen-
dations directed at manufacturers that have potential for technological developments
for safer technology. In this manner, ICRP has acted as an important resource
presaging safety issues based on current and future use of technology, and identifying
needs where technology can contribute. Of course, there are vast areas of optimisa-
tion where users can play a large role in minimising radiation doses to patients
without compromising diagnostic or clinical purpose. In recent years, there have
been evaluations of practices which have indicated that a large number of imaging
procedures have not met the appropriateness guidelines. While ICRP has provided
three levels of justification, there is increasing need to scrutinise justification at Level
3, and provide guidance on justification of an examination. The current climate of
interest in radiological protection has enhanced the audience of ICRP publications to
cover policy makers, health authorities, public health organisations, patient groups,
organisations developing criteria on the appropriateness of different techniques and
their use, and a variety of medical specialists who have now started using imaging
technology that was not available to them a decade or so ago. This publication
addresses the challenges faced with the new technology of cone beam computed
tomography (CBCT) that is being used increasingly in day-to-day practice in hos-
pitals by increasing numbers of medical specialists. The advice from ICRP is timely.
Issues of patient and worker protection are also addressed in this publication.

The Commission launched a Task Group on Radiological Protection in CBCT in

2013.

The membership of the Task Group was as follows:

M.M. Rehani (Chairman) S. Bartling R. Gupta

The corresponding members were:

T. Berris (to October 2013) J.M. Boone

G.C. Sharp R. Pauwels (from December 2013)

11

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Committee 3 critical reviewers were:

C. Martin R. Loose

Main Commission critical reviewers were:

C. Cousins H-G. Menzel

The membership of Committee 3 during the period of preparation of this report was:

(2009–2013)

E. Vañó (Chair) J-M. Cosset (Vice-Chair) M.M. Rehani (Secretary)

M.R. Baeza L.T. Dauer I. Gusev
J.W. Hopewell P-L. Khong S. Mattson
D.L. Miller P. Ortiz López K.Å. Riklund
H. Ringertz M. Rosenstein Y. Yonekura
B. Yue

(2013–2017)

E. Vañó (Chair) D.L. Miller (Vice-Chair) M.M. Rehani (Secretary)

K. Applegate M. Bourguignon L.T. Dauer
S. Demeter K. Kang P-L. Khong
R. Loose C. Martin P. Ortiz López
K.Å. Riklund P. Scalliet Y. Yonekura
B. Yue

12

Downloaded from ani.sagepub.com by guest on July 22, 2015

 MAIN POINTS

. The guidelines and recommendations on radiological protection in cone beam com-

puted tomography (CBCT) are important because CBCT extends the use of com-
puted tomography (CT) to areas that were not typically associated with CT imaging
in the past, e.g. surgery, dental and otolaryngology [ear/nose/throat (ENT)] clinics,
angiography suites, radiotherapy treatment vaults, and orthopaedic polyclinics.
. The manufacturers of CBCT scanners have invested considerable effort into meeting
the electrical and mechanical safety requirements of the users. Similar diligence is
needed for issues related to radiation dose and radiological protection.
. This publication provides a basis to develop informed decisions and to direct the use
of CBCT for optimising the trade-off between clinical benefit and radiation risk.
. The International Commission on Radiological Protection (ICRP) emphasises that
protection should be optimised not only for whole-body exposures, but also for
exposures to specific tissues, especially those of the lens of the eye, the heart, and
the cerebrovascular system.
. Equipment used for both fluoroscopy and CBCT should provide aggregate dose
indices for individual patients throughout the procedure through electronic display
on the operator console and a radiation dose structured report.
. Optimisation of both patient and worker doses, particularly when workers have to be
near the machine, is important when monitoring of doses becomes an essential tool.
Recording, reporting, and tracking of radiation dose for a single patient should be
made possible in a consistent manner across vendors.
. Low-dose protocols may be sufficient to answer diagnostic questions focused on high-
contrast structures, such as lung, bones, dental and maxillofacial scans, ENT scans
(paranasal sinuses, skull, temporal bone), interventional material, and contrast-
enhanced vessels (angiographic interventions).
. Higher-dose protocols should only be selected if visualisation of soft tissue struc-
tures, such as intracranial haemorrhage, soft tissue tumours, or abscesses, is the
primary focus.
. Most interventional and intraprocedural C-arm CBCT systems can scan an angular
range spanning 180–240o plus the cone angle of the x-ray beam. Localised critical
organs, such as the thyroid, eyes, female breasts, and gonads, should be on the
‘detector side’ of the arc whenever possible.
. Clinical need permitting, every effort should be made by users to ensure that the
volume of interest is fully incorporated in the field of view (FOV) provided by the
CBCT scanners, while radiosensitive organs should be placed outside the FOV.
. The aim of CBCT should be to answer a specific diagnostic or intra-operative ques-
tion vis-à-vis other imaging modalities, and not to obtain image quality that rivals
multi-detector CT (MDCT). The decision by the referring practitioner to use CBCT
should be made in consultation with an imaging professional.
. There is a need to provide checks and balances, such as dose check alerts imple-
mented in CT in recent years, to avoid high patient doses compared with locally
defined reference values.

13

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

. Methods that provide reliable estimates of dose to the eye under practical situations
should be established and used.
. The user of CBCT in interventions can influence the radiation dose imparted to the
patient significantly by judicious use of a ‘low-image-quality or low-dose’ scan
instead of a ‘high-image-quality or high-dose’ scan.
. In radiotherapy, justified use of CBCT has potential at different stages of therapy
such as: pretreatment verification of patient position and target volume localisation;
evaluation of non-rigid misalignments, such as flexion of the spine or anatomical
changes in soft tissue; and during or after treatment to verify that the patient position
has remained stable throughout the procedure. Low-dose CBCT protocols should be
used for pretreatment alignment of bony structures.
. Many machines were only capable of fluoroscopy initially, but can now also perform
CBCT. Due to the improved clinical information in CBCT and its ability to remove
overlying structures, the user may be tempted to over-use the CBCT mode. The
CBCT mode should be used judiciously.
. In orthopaedics, justified use of CBCT can help in assessing the position of fractures
and implants with respect to the bony anatomy, especially in situations where fluor-
oscopy alone is insufficient, and thus can help in patient dose management.
. In urology, low-dose CBCT protocols should be used when imaging high-contrast
structures, such as calcified kidney stones.
. Dental and maxillofacial CBCT scans should be justified, considering alternative
imaging modalities. Once justified, they should be optimised to obtain images with
minimal radiation dose without compromising the diagnostic information.
. The level of training in radiological protection should be commensurate with the level
of expected radiation exposure.
. All workers intending to use CBCT for diagnostic purposes should be trained in the
same manner as for diagnostic CT, and those intending to perform interventional
CBCT should be trained in the same manner as for interventional CT.

14

Downloaded from ani.sagepub.com by guest on July 22, 2015

 GLOSSARY
Absorbed dose, D

The absorbed dose, D, is the quotient of d" by dm, where d" is the mean energy
imparted by ionising radiation to matter of mass dm, thus:

d"
D¼
dm

The unit of absorbed dose is J kgl. The special name for the unit of absorbed
dose is gray (Gy); 1 Gy ¼ 1 J kg1.

Automatic exposure control (AEC)

A device that automatically determines and provides the exposure needed to

produce a preselected image quality by sampling the x-ray intensity at the
image receptor.

Collimation

Geometrical limitation of the extent of the radiation beam.

Cone-beam computed tomography (CBCT)

In this publication, the term ‘CBCT’ is used to designate a subset of computed

tomography (CT) scanners that share certain key design features, image quality
characteristics, and application domains that distinguish this set of scanners
from multi-detector CT (MDCT) scanners. The most characteristic design fea-
ture that distinguishes CBCT scanners from MDCT scanners is the use of a
two-dimensional digital flat-panel detector to yield a three-dimensional
volumetric image in one rotation. Flat-panel detectors in CBCT allow wide
cone angle, large z-coverage, and high spatial resolution at the expense of
low-contrast resolution.

Dental and maxillofacial imaging

In this publication, dental and maxillofacial imaging refers to imaging of high-

contrast structures related to the teeth and jaw bones. Visualisation of other
structures (e.g. maxillary sinus, temporomandibular joint, facial skeleton) can
be considered as dental and maxillofacial imaging if the primary indication for
imaging relates to dentistry. Ear/nose/throat imaging is considered as a separ-
ate application in this publication, although it often involves similar radio-
graphic equipment.

15

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Detector quantum efficiency (DQE)

A widely used metric that describes the quality of an x-ray detector. It measures
the efficiency (i.e. signal-to-noise performance) of the detector to produce an
image from a given incident fluence. Intuitively, it captures how well a detector
translates the fluence incident on it into an image, relative to an ideal detector.

Deterministic effect

Injury in populations of cells, characterised by a threshold dose and an increase

in the severity of the reaction as the dose is increased further. Also termed
‘tissue reaction’. In some cases, deterministic effects are modifiable by post-
irradiation procedures including biological response modifiers.

Diagnostic reference level (DRL)

Dose levels in medical radiodiagnostic practices or, in the case of radiophar-

maceuticals, levels of activity for typical examinations for groups of standard-
sized patients or standard phantoms for broadly defined types of equipment.
These levels are not expected to be exceeded for standard procedures when
good and normal practice regarding diagnostic and technical performance is
applied.

Dose limit

The value of the effective dose or the equivalent dose to an organ received by an
individual within a specified period from planned exposure situations that shall
not be exceeded. Dose limitation is one of three fundamental principles of
radiological protection originally defined by ICRP.

Effective dose, E

The tissue-weighted sum of the equivalent doses in all specified tissues and
organs of the body, given by the expression:
X
E¼ wT HT
T

where HT is the equivalent dose in a tissue or organ, T, and wT is the tissue-

weighting factor. The SI unit for effective dose is sievert (Sv), equal to J kg1.

16

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Equivalent dose, HT

The dose in a tissue or organ T given by:

X
HT ¼ wR DT,R
R

where DT,R is the mean absorbed dose from radiation R in a tissue or organ T,
and wR is the radiation weighting factor. The unit for equivalent dose is the
same as for effective dose (sievert, Sv), equal to J kg1.

Hounsfield unit (HU)

Number used to represent the mean x-ray attenuation associated with each
elemental area of the CT image. Measured values of attenuation are trans-
formed into HU (also known as CT numbers) using the Hounsfield scale:
material  water
HU ¼  1000
water

where  is the effective linear attenuation coefficient of the measured material

relative to water for the x-ray beam used. The scale is defined so that water has
a value of 0 HU and air has a value of 1000 HU.

Justification

One of three fundamental principles of radiological protection originally

defined by ICRP. The justification principle requires that the net benefit of
radiation exposure should be positive.

Multi-detector computed tomography (MDCT)

According to Publication 102 (ICRP, 2007a), ‘MDCT systems are CT scanners

with a detector array consisting of more than a single row of detectors. The
‘multi-detector-row’ configuration of MDCT scanners refers to the use of mul-
tiple detector arrays (rows) in the longitudinal direction (i.e. along the length of
the patient). MDCT scanners use third-generation CT geometry in which the
arc of detectors and the x-ray tube rotate together. All MDCT scanners use a
slip-ring gantry, allowing helical acquisition.’ An arc of detector rows used in
MDCT should be distinguished from a digital flat-panel detector typically
employed by CBCT scanners, as these two different detector technologies

17

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

have a very distinct acquisition time, latency, dynamic range, and spatial
resolution.

Noise

A fundamental statistical phenomenon that is present in all images. Noise tends

to reduce the visibility of structures and objects, especially those that have
relatively low contrast. In medical imaging, the objective is not to eliminate
the noise, but to reduce it to a clinically acceptable level. Noise is the point-to-
point variation in image brightness that does not contain useful information.
The magnitude of noise is indicated by the standard deviation of the grey
values within a region of interest in the image.

Occupational exposure

All exposure incurred by workers in the course of their work, with the excep-
tion of: (1) excluded exposures and exposures from exempt activities involving
radiation or exempt sources; (2) any medical exposure; and (3) the normal local
natural background radiation.

Optimisation of protection

The likelihood of incurring exposure, the number of people exposed, and the
magnitude of their individual doses should all be kept as low as reasonably
achievable, taking into account economic and societal factors. In medical ima-
ging, optimisation of protection implies lowest dose for the clinical purpose.

Phantom

A device that absorbs or scatters radiation in an equivalent manner to a

patient, used to estimate radiation doses and test imaging systems without
actually exposing a patient. A phantom may be an anthropomorphic or a
physical test object.

Population dose

An expression for the aggregate radiation dose incurred by a population,

defined as the product of the number of individuals exposed to a source and
their average radiation dose. The collective dose is expressed in man-sieverts
(man-Sv), and is intended solely as an instrument in the optimisation of radio-
logical protection.

18

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Scatter

Deviation of x rays from their original trajectory due to interaction with

matter.

Shielding

The placement of a high-absorption material (e.g. lead) between the source and
its environment for the purpose of reducing radiation dose to workers, patients,
or the public.

Slice

A tomographic section (defined by the position and thickness) of a test phan-

tom or patient under investigation during a single CT or CBCT exposure.

Stochastic effects

Malignant disease and heritable effects for which the probability of an effect
occurring, but not its severity, is regarded as a function of dose without
threshold.

Worker

Any person who is employed, whether full-time, part-time, or temporarily, by

an employer, and who has recognised rights and duties in relation to occupa-
tional radiological protection.

19

Downloaded from ani.sagepub.com by guest on July 22, 2015

Downloaded from ani.sagepub.com by guest on July 22, 2015
 1. INTRODUCTION

. The guidelines and recommendations on radiological protection in cone beam com-

puted tomography (CBCT) are important because CBCT extends the use of com-
puted tomography (CT) to areas that were not typically associated with CT imaging
in the past, e.g. surgery, dental and otolaryngology [ear/nose/throat (ENT)] clinics,
angiography suites, radiotherapy treatment vaults, and orthopaedic polyclinics.
. The manufacturers of CBCT scanners have invested considerable effort into meeting
the electrical and mechanical safety requirements of the users. Similar diligence is
needed for issues related to radiation dose and radiological protection.
. This publication provides a basis to develop informed decisions and to direct the use
of CBCT for optimising the trade-off between clinical benefit and radiation risk.

(1) The guidelines and recommendations on radiological protection in CBCT are

important because CBCT extends the use of CT to areas that were not typically
associated with CT imaging in the past, e.g. surgery, dental and otolaryngology
(ENT) clinics, angiography suites, radiotherapy treatment vaults, and orthopaedic
polyclinics.
(2) The International Commission on Radiological Protection’s (ICRP) radio-
logical protection principles and recommendations provided in earlier publications,
particularly Publications 87 and 102 (ICRP, 2000a, 2007a), apply to these newer
applications and should be adhered to.
(3) The cone beam nature of the radiation field presents new challenges in dose
management to ensure patient safety. The manufacturers of CBCT scanners have
invested considerable effort into meeting the electrical and mechanical safety require-
ments of the users. Similar diligence is needed for issues related to radiation dose and
radiological protection.
(4) This publication provides a basis to develop informed decisions and direct the
use of CBCT for optimising the trade-off between clinical benefit and radiation risk.
(5) Appropriate use of CBCT, including radiological protection, is a joint respon-
sibility of the referring practitioner and the imaging professionals. Imaging profes-
sionals also have responsibility towards optimisation of protection. When referring a
patient for a diagnostic CBCT examination, the referring practitioner should be aware
of the strengths and weaknesses for CBCT vis-à-vis multi-detector CT (MDCT), mag-
netic resonance imaging, and other competing imaging modalities. The decision to use
CBCT should be made in consultation with an imaging professional.
(6) In this publication, the term ‘CBCT’ is used to designate a subset of CT
scanners that share certain key design features, image quality characteristics, and
application domains that distinguish this set of scanners from MDCT scanners. The
most characteristic design feature that distinguishes CBCT scanners from MDCT
scanners is the use of a two-dimensional (2D) digital flat-panel detector (FPD) to
yield a three-dimensional (3D) volumetric image in one rotation. FPDs in CBCT
allow a wide cone angle, large z-coverage, and high spatial resolution at the expense
of low-contrast resolution. Some cone beam scanners still use an image-intensifier

21

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

tube. These scanners have a lot in common with flat-panel-based CT scanners and
should be considered as CBCT devices. All CBCT scanners have a wide cone angle
and large z-coverage. The use of a flat panel or image intensifier as a detector means that
CBCT scanners have a lower dynamic range and lower soft tissue contrast than MDCT
scanners, making them suitable for high-contrast structures such as bones and contrast-
enhanced vasculature. Generally, CBCT scanners require longer scan times.
(7) It should be noted that a wide cone angle or large z-coverage is not, in itself,
sufficient to define a CBCT scanner for the purposes of this publication. Many
traditional MDCT scanners, built using individual rows of detectors made of scin-
tillating ceramics, have a wide cone angle and large z-axis coverage. For example, the
Aquilion OneTM (Toshiba Medical Systems) is a 320-row scanner with 16 cm of z-
coverage. This scanner, and others in its class, should be designated as wide-area
MDCT scanners rather than CBCT scanners.
(8) CBCT represents an emerging technology that enables high-resolution volu-
metric scanning of the anatomy under investigation. Just as in MDCT, use of
CBCT is increasing steadily in clinical practice. Although it is a relatively new
modality, CBCT is already being used for a variety of clinical applications, such
as dental imaging, head and neck imaging (including sinus CT), paediatric ima-
ging, high-resolution bone imaging, and intra-operative and interventional
imaging.
(9) CBCT imaging is also used in radiotherapy for pretreatment verification of
patient position and target volume localisation. In this case, the CBCT system is
usually mounted on the gantry of a linear accelerator at 90 to the therapeutic beam.
For radiotherapy, CBCT imaging is often used for daily repositioning. Under clas-
sical fractionation schedules, high cumulative imaging dose to tissues outside the
exposure field can accrue.
(10) Although the concept of CBCT has existed for over 25 years, it has only
recently become possible to develop clinical CBCT systems that are both sufficiently
inexpensive and sufficiently small to be used in operating rooms, outpatient clinics,
emergency rooms, and intensive care units. Technological and application-specific
factors that have converged to make clinical CBCT possible are:

1. compact, high-quality FPD arrays;

2. computer power sufficient for timely cone beam image reconstruction;
and
3. x-ray tubes designed for cone beam scanning.

(11) Most modern CBCT systems use a digital FPD instead of an image intensifier
for image capture. By virtue of these specialised detectors, which are different from
the detectors used in conventional MDCT, CBCT is capable of ultra-high spatial
resolution and large-volume coverage in a single (or partial) rotation of the C-arm.
Digital FPDs used in CBCT scanners also enable fluoroscopy, radiography, volu-
metric CT, and dynamic imaging using a single or partial rotation. These capabilities
are extremely useful for intra-operative and vascular applications.

22

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

(12) The manufacturers of CBCT scanners have invested considerable effort into
meeting the electrical and mechanical safety requirements of the users, which are
mandated by national regulatory bodies. Similar diligence is needed for issues related
to radiation dose. In this respect, the cone beam nature of the radiation field presents
new challenges in dose management to ensure patient safety; guidelines are needed
for various stakeholders in this new modality. This publication briefly describes the
current state-of-the-art CBCT technology, reviews current dose measurement and
management approaches, provides recommendations for safe use of CBCT scanners,
and identifies gaps that relate to radiological protection where further research is
needed.
(13) CBCT systems differ from MDCT systems in several ways that affect image
quality and radiological protection. Some key differences are listed below.
(14) Due to the cone beam nature of the irradiated field and the associated non-
uniformities in the primary and scatter radiation imparted to the scan volume, the
standard dose metrics popularised by MDCT cannot be applied to CBCT, or even to
wide-area MDCT scanners.
(15) CBCT systems usually have superior spatial resolution for high-contrast
objects (e.g. bone, lung), but inferior contrast resolution for low-contrast objects
(e.g. soft tissue). A trained and skilled user of CBCT can influence the radiation
dose imparted to the patient significantly by judicious use of ‘high-dose’ and ‘low-
dose’ scans. A high-dose scan is generally required if soft tissue structures are the
main diagnostic focus, while a low-dose scan may be sufficient for angiographic scans
with arterial or venous contrast media, or for defining the position of interventional
catheters.
(16) Due to the higher spatial resolution of an FPD, CBCT slices are intrinsically
thinner and have lower signal-to-noise ratios (SNRs) for the same dose than MDCT
slices. Any attempt to match the SNR in a thin CBCT slice with a thick MDCT slice
will result in a proportionate increase in dose. Instead, increasing the slice thickness,
or other similar image processing methods, should be applied to improve the SNR in
CBCT.
(17) In many CBCT scanners, the angular span over which the projection data are
acquired can be customised. This feature may also be available in some MDCT
scanners (e.g. some systems allow tube current to be reduced when the beam is
covering radiation-sensitive organs such as the breast, thyroid, or lens of the eye).
It should be used in both types of scanner systems to minimise the dose to selected
organs.
(18) The purpose of this publication is to identify radiological protection
issues for patients and workers and, in line with other ICRP publications, recom-
mendations are set out for all stakeholders ranging from day-to-day clinical
users, auxiliary support workers, buyers, manufacturers, and policy directing
committees.
(19) The primary target audience of this publication, as for most other publica-
tions produced by the Commission related to protection in medicine, is health pro-
fessionals working with CBCT, other workers tasked with radiological protection

23

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

and image quality optimisation in CBCT, manufacturers of imaging equipment,

regulators, and policy makers in charge of radiological protection.

1.1. History of development

(20) The first CBCT scanner was built for angiography at the Mayo Clinic,
Rochester, NY, USA in 1982 (Robb, 1982). Multiple teams in the early 1990s pur-
sued the idea of multi-angle projections from a wide-area detector for medical ima-
ging. For example, Saint-Félix et al. (1994) tested a system called the ‘Morphometer’
consisting of two imaging chains, each with an x-ray tube and an image intensifier.
This CBCT system was designed for 3D angiography using the gantry of a conven-
tional CT scanner. It reconstructed vascular images from a set of digitally subtracted
angiography images. This gantry platform, which was never released clinically, was
abandoned in favour of a C-arm supporting a single imaging chain.
(21) Fahrig et al. (1997, 1998) also developed a CBCT system based on an
image intensifier and C-arm for use in angiography. Wiesent et al. (2000) devel-
oped a similar system comprising a C-arm plus an image intensifier for interven-
tional angiography. Ning et al. (2000a,b) and Wang (1997) developed a CBCT
angiography imager based on a GE 8800 CT scanner with an image intensifier–
charge-coupled device chain and later with an FPD. Schueler et al. (1997) and
Kawata et al. (1996) developed a CBCT angiography scanner based on a biplanar
C-arm system.
(22) Siewerdsen and Jaffray developed a CBCT system for radiotherapy guidance
based on an amorphous silicon FPD (Siewerdsen and Jaffray, 1999, 2001; Jaffray
and Siewerdsen, 2000). Efforts are also underway to build a dedicated CBCT-based
imaging system for mammography (O’Connell et al., 2010; Kalender et al., 2012;
Packard et al., 2012).
(23) CBCT devices were introduced in dental and maxillofacial radiology in the
late 1990s (Mozzo et al., 1998; Arai et al., 1999).

1.2. Current standards in radiological protection in CBCT

(24) The guidelines and recommendations on radiological protection in CBCT are
particularly important as CBCT extends the use of CT to areas that were not typ-
ically associated with CT imaging in the past, e.g. surgery, dental and otolaryngology
(ENT) clinics, angiography suites, and orthopaedic polyclinics. Fundamentally,
CBCT is a form of CT; as such, most facility design and quality assurance (QA)
requirements that apply to MDCT should also be applied to CBCT. This, however,
can lead to an erroneous impression that CBCT is identical to MDCT, making it
difficult to manage CBCT from operational and radiation safety points of view.
Further complications arise when a user is tempted to regard CBCT as a ‘light’ or
‘low-dose’ CT; a view that is maintained because CBCT functionality is often an
adjunct to existing capabilities, such as fluoroscopy and angiography in a C-arm or
other clinic-based system. Embedded in these user biases is the risk for potential

24

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

overuse of CBCT resulting in unnecessary radiation dose to the patients and/or

workers.
(25) Traditionally, the use of CBCT in dentistry has entailed a relatively low
radiation dose. However, this is not always the case, and many recent applications
of CBCT, especially in ENT and interventional procedures, can impart much higher
radiation doses that equal or exceed those from MDCT (Kyriakou et al., 2008a;
Dijkstra et al., 2011; Schulz et al., 2012). There are also situations in which multiple
CBCT procedures have to be performed on one patient (such as CBCT-guided inter-
ventions), enhancing the need to keep the inflicted radiation dose to a minimum.
Therefore, dose implications of CBCT pose a risk from the perspective of an indi-
vidual patient, as well as for the risk from radiation exposure of the population as
a whole.
(26) Imaging professionals and medical physicists are well aware of the radiation
dose issues in CT. This knowledge, however, does not directly translate to CBCT, for
which the trade-off between image quality and radiation dose can be quite complex.
At the same time, clinical users, as well as those undertaking QA and members of
radiation safety committees, need clear guidelines on operating and regulating these
systems. This publication, which is thought to be the first on radiological protection
in CBCT from an international source, provides a basis for developing informed
clinical decisions on the use of CBCT and guidance for optimising the trade-off
between clinical benefit and radiation risk.

1.3. Responsibilities of different stakeholders

(27) Approximately 80 million CT scans are performed every year in the USA, and
this number is increasing on a yearly basis (Sierzenski et al., 2014). Multiple recent
papers have drawn attention to the population dose from these scans (Brenner,
2010). There is also increasing realisation that a large fraction of this radiation
dose to the population is avoidable as it comes from unjustified or inappropriate
examinations. Currently, data on inappropriate use are mostly available for CT
rather than CBCT. Appropriate use of CT scanning is a joint responsibility of the
referring practitioner and the imaging professional, and most national regulations
assign this responsibility either jointly or to the imaging professional. As a referring
practitioner best understands the clinical need for the examination, he/she must
interact with an imaging professional to decide upon the radiological examination
or procedure that is in the best interests of the patient. Electronic referrals with
decision support have the potential to simplify and streamline this interaction
while making this process more evidence based (Sistrom et al., 2009). Such systems
can go a long way towards facilitating the desired radiological examination per-
formed with the lowest radiation dose, while maintaining the image quality needed
for the clinical purpose. Practitioners, technologists, and medical physicists must
understand their roles and responsibilities in this endeavour. To this end, there is
a need to further develop methods that facilitate the interaction between referring

25

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

practitioner and imaging professional to translate their joint responsibility for radio-
logical safety into practice.
(28) Over the years, manufacturers have played a vital role in technological devel-
opments to reduce patient doses from particular CT examinations. The Commission,
while acknowledging this role, hopes that manufacturers will remain at the forefront
of developing new technologies for radiological protection of patients and workers.

1.4. Why is it important to know CBCT doses?

(29) It is easy for a practitioner, not versed in the details of dose management, to
dismiss CBCT as upgraded fluoroscopy coupled with 3D reconstruction. For the
most part, the dose from CBCT is indeed lower than that from MDCT, which may
reinforce this belief. However, uncritical application of CBCT under the assumption
that it is a modality with minimal dose consequences could result in significant doses
in some circumstances, and is not appropriate for protection of the patient.
(30) CBCT is a relatively new development in clinical practice. Data on radiation
doses and possible effects of CBCT are still being gathered and analysed. Even at this
early stage, however, studies indicate that there is room for optimisation to keep the
radiation dose as low as reasonably achievable. This publication systematically sum-
marises the available dose data related to use of CBCT, and discusses radiological
protection issues for patients and workers. Given the potential of CBCT to become a
significant source of radiation dose to patients in the future, it is appropriate to be
mindful of the radiation exposure while using the full diagnostic potential of this
exciting modality. In 1999–2000, while preparing Publication 87 (ICRP, 2000a;
Rehani and Berry, 2000), the Commission had similarly presaged the need to
watch for increasing radiation doses from MDCT. Although this concern was not
well appreciated at that time, it has become a major issue in subsequent years with
multiple high-profile reports in the media. This publication provides a similar review
of the current CBCT literature, and presents the data regarding radiation doses to
patients and workers associated with the use of CBCT.

1.5. Safety in perspective

(31) Safety is achieved most readily when it is built into the system, rather than a
matter of choice for users. A good example is a collision avoidance system; an
innovation that started in the automobile industry but has been implemented in
multiple types of imaging gantries to avoid accidents. With such a system in place,
if the gantry of the imaging device comes into contact with a person or object, it
simply stops moving. In the absence of such a system, when collision avoidance has
to be accomplished primarily via user education, training, and instructions, the risk
of injury from collisions will be higher. There are instances when both detection of an
anomalous condition and its automatic avoidance cannot be implemented simultan-
eously. In such cases, detection and warning may accomplish a similar end result.

26

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

For example, radars for detection of speed limits have been shown to decrease the
incidence of speeding violations.
(32) For radiation safety in MDCT, a display of radiation exposure information
on the operator console has been present for a number of years. After a series of
accidental exposures was reported in the USA in 2007–2008, MDCT systems can
now automatically detect settings to prevent accidental exposure (NEMA, 2010).
Such systems provide an additional layer of non-intrusive checks and balances in
the conduct of a scan. Display of such information on CBCT consoles needs to be
standardised. The Commission recommends development and implementation of
safety systems that require the least amount of interaction from the operator and
workers while providing:
. regular and continuous monitoring of radiation output throughout the
examination;
. automatic comparison with reference or desired dose levels that need to be
established;
. timely feedback to the system operator;
. wide availability of automatic adjustment of the dose to a prescribed level in a
manner that is somewhat similar to automatic exposure control (AEC); and
. alerts when dose is higher than specified. Currently, dose checks do not apply
to CBCT systems (NEMA, 2010).
(33) Other technologies that many CBCT vendors need to implement uniformly
include automatic collimation control so that the x-ray beam always falls on the
detector, guidance for instruments during image-guided interventions, and minimisa-
tion of scatter dose resulting from mechanical components.

1.6. Scope of this publication

(34) As a substantial amount of information is currently available on dental and
maxillofacial CBCT, including a publication issued by the European Commission
(EC) project SEDENTEXCT (Safety and Efficacy of a New and Emerging Dental
X-ray Modality) (http://www.sedentexct.eu/) (EC, 2012a), the decision was made to
cover dental and maxillofacial CBCT briefly in this publication
(35) It should be emphasised that the main focus of this report is on doses to
patients and workers from CBCT acquisitions. CBCT acquisition can be part of
fluoroscopically guided procedures. In such cases, it is necessary to account for the
dose from fluoroscopy and relevant implications. Publication 117 included informa-
tion pertinent to radiological protection of patients and workers in fluoroscopic
procedures performed outside the imaging departments (ICRP, 2010), and
Publication 120 covered radiological protection of patients and workers during inter-
ventional fluoroscopy (ICRP, 2013). The term ‘patient dose’ has been used in this
publication in a qualitative and relative sense, and wherever quantitative figures are
provided, the appropriate dose quantity [i.e. skin dose, organ dose, kerma-area
product (KAP), effective dose, etc.] has been mentioned.

27

Downloaded from ani.sagepub.com by guest on July 22, 2015

Downloaded from ani.sagepub.com by guest on July 22, 2015
 2. CBCT TECHNOLOGY
2.1. Introduction
(36) In the past decade, development of digital FPDs for conventional x-ray radi-
ography, fluoroscopy, and mammography has propelled the use of CBCT into the
mainstream of medical imaging. Most CBCT systems in current use leverage the
power of dynamic FPDs (i.e. able to acquire several frames s1 compared with
static FPDs) to provide volumetric 3D datasets.
(37) A C-arm gantry consisting of a digital FPD and a large cone-angle x-ray tube
is the most commonly used platform for CBCT. There are a number of other imple-
mentations of CBCT that differ in the mechanical gantry used for scanning, the
detector subsystem, the type of x-ray tube and filtration, the cone angle employed
for imaging, and the algorithm used for reconstructions. The following section
describes and introduces different types of CBCT scanners.

2.2. Technological issues

(38) As far as tomographic capabilities of a CBCT scanner are concerned, in
simple terms, one can think of them as a conventional MDCT in which the rows
of detector elements (typically 16–64 rows) have been replaced by an area detector
(Ross et al., 2004; Grasruck et al., 2005; Popescu et al., 2005). In general, a CBCT
scanner consists of an x-ray source, a detector, and a gantry to move this imaging
chain around the patient. The most commonly used subsystems are described briefly
below.

2.2.1. X-ray source

(39) The x-ray source used in a CBCT scanner must provide a broad, cone-shaped
beam of radiation. Consequently, CBCT scanners use a much larger anode angle
than a tube used in an MDCT scanner. Typical operating conditions are an x-ray
tube voltage of 50–140 kVp, a tube current of 10–800 mA, and a total power of 10–
80 kW. In order to take advantage of the small detector pixel size, the focal spot size
ranges from 0.2 mm to 0.8 mm. The typical field of view (FOV) covered in one
rotation, using a single FPD, can be as much as 25 cm in the angular direction,
and 20 cm in the z-direction. Larger sizes are possible when multiple panels or
dual scans are used, such that the principal axis of the x-ray illumination is offset
from the centre of the panel to allow beam correction.

2.2.2. Detector

(40) While some older systems still use an image intensifier, most modern CBCT
scanners use a digital FPD. FPDs provide higher dose efficiency and dynamic range

29

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

than other detector technologies they replaced (x-ray film, film/screen combinations,
and image intensifiers); however, their dynamic range is lower than that of standard
MDCT detectors (Miracle and Mukherji, 2009a). FPDs also generally provide higher
spatial resolution than image intensifiers and conventional detector arrays used in
MDCT. Direct digital readout up to 30 frames s1 ensures that the data are available
in a directly usable form for both projection and 3D reconstruction.
(41) The native resolution of a flat panel is typically at or below 200 mm, although
higher resolution detector panels are available. After accounting for magnification
and x-ray focal spot size, this yields an isotropic voxel resolution of approximately
150 mm. Generally, in 3D acquisition mode, FPDs are operated in a 2  2 binning
mode (summing signals from two rows and two columns to increase the SNR and the
readout speed, and reduce the matrix size), and the isotropic resolution is of the
order of 200 mm. Therefore, compared with conventional MDCT scanners, a flat-
panel-based CBCT system improves the spatial resolution by a factor of almost
12 on a voxel-by-voxel basis. Its high spatial resolution is capable of visualising
complex human anatomy, including fine structures of the maxillofacial region and
skull base.
(42) Typically, FPDs used in CBCT are composed of a matrix of detector elements
that can span anywhere from 5  5 cm2 to 40  40 cm2. Such scanners, therefore, are
capable of producing a large number of slices spanning anywhere from 5 to 20 cm in
one rotation. The z-coverage afforded by these scanners can be large enough to
image an entire organ such as the brain, heart, liver, or kidneys in one axial scan.

2.2.3. Gantry

(43) Depending on the mechanical system of the gantry, CBCT scanners can allow
conventional fluoroscopy, angiography, and radiography in the same setup as well as
providing high spatial resolution and large-volume coverage. These facilities make
such machines especially attractive for intra-operative and vascular applications. The
various gantry platforms in common use are described below.

C-arm-based CBCT
(44) All major imaging equipment vendors now provide C-arm scanners that
employ digital FPDs integrated with a C-arm gantry (Fig. 2.1). The C-arm platform
offers open architecture and ready patient access. There are two major C-arm-based
setups that need to be distinguished: C-arm-based interventional CBCT systems and
dedicated C-arm-based CBCT systems.
(45) C-arm-based interventional CBCT systems. One can use the C-arm for fluor-
oscopy and projective angiography (including digitally subtracted angiography).
However, by putting the C-arm in a fast-spin mode while acquiring images, one
can obtain projection data that can be converted into relatively high-quality, high-
contrast CT images. Interventional procedures are usually performed using

30

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Fig. 2.1. C-arm-based cone beam computed tomography. A C-arm is used to mount the
imaging chain, and this provides the necessary amount of freedom required to revolve
around the patient. C-arm systems are used in surgical, orthopaedic, urologic, or interven-
tional environments (image provided by Rolf Kueres).

fluoroscopy. The operator can use the CBCT mode intermittently for clarification
and 3D localisation (Orth et al., 2008; Schafer et al., 2011). These machines, there-
fore, enable a seamless integration of these previously separate modalities. They are
used in angiographic, surgical, orthopaedic, urologic, and other interventional
settings.
(46) Dedicated C-arm-based CBCT systems. A number of systems dedicated for
dental, ENT, head and neck, extremity imaging, and mammography are available.
One popular variation of C-arm-based CBCT systems is the so-called ‘seat scanner’,
in which a small C-arm, with a horizontal imaging chain consisting of an FPD and
an x-ray tube, revolves around the head of the patient while they sit on a chair
(Fig. 2.2). Alternatively, for certain models, the patient is in a supine or standing
position. These scanners are dedicated to dental, maxillofacial, and temporal bone
applications because of their relatively small scan FOV. Besides weight and mech-
anical considerations, there is no fundamental reason why their FOV cannot be
increased. They are currently limited to these niche applications.

CT-gantry-based CBCT
(47) A flat-panel volume CT scanner combines a continuously rotating CT gantry
with digital FPD technology (Fig. 2.3). It is, in fact, a CT machine in which the
detector rows have been replaced by an FPD. From an operational point of view, the
main difference between a CT-gantry-based and a C-arm-based cone beam system
lies in basic engineering: the gantry-based systems are more stable and have fewer

31

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Fig. 2.2. Clinic-based cone beam computed tomography. The imaging chain is mounted on
a horizontal rotating C-arm. These systems are usually used in head and neck applications
(image provided by Rolf Kueres).

geometric inaccuracies compared with the C-arm-based systems. In addition, the

isocentre of any CT gantry, by virtue of its mechanical design, is defined much
more precisely than the best C-arm gantries. As a result, gantry-based designs
may, in most cases, offer better spatial resolution.
(48) In a C-arm system, the detector and the x-ray tube are connected to the
control hardware by an umbilical cord of cables that prevents them from spinning
continuously around the patient. This is not the case for a CT-gantry-based system,
in which a slip ring is used to take data from a rotating component. Elaborate
collision avoidance schemes have been implemented to ensure operator safety. No
such concerns exist for CT-gantry-based systems.
(49) By virtue of an FPD, CT-gantry-based CBCT systems are capable of ultra-
high spatial resolution, direct volumetric imaging, and continuous rotation around a
patient. Continuous rotation enables dynamic CT scanning; the ability to observe a
process evolving with time [e.g. perfusion of an entire organ such as the brain, liver,
or kidney (e.g. after transplant or an ischaemic event)].

32

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Fig. 2.3. Gantry-based cone beam computed tomography. The patient lies on a patient
bed, and the imaging chain revolves around the patient as in multi-detector computed tom-
ography (image provided by Rolf Kueres).

CBCT in radiotherapy
(50) In radiotherapy, CBCT is used for precise alignment of the target volume
with a therapeutic, hard x-ray beam from a linear accelerator. Two separate
arrangements (kV CBCT and MV CBCT) are popular. In kV CBCT, a separate
imaging chain consisting of an x-ray tube operated in the kV range is used as the x-
ray source, and an FPD is used for imaging. The entire imaging chain is mounted
on the linac gantry, in an orientation that is orthogonal to the therapeutic beam.

Similar to C-arm systems, a linac gantry only rotates up to 360 , after which the
gantry must be rotated back. Some systems enable a larger FOV to be imaged by

scanning target volumes asymmetrically using two 180 rotations, and shifting the
FPD laterally for the second. Such systems use separate half-bowtie filters for the
two parts of the scan. A routine CBCT scan is conducted prior to therapy for
precise alignment.
(51) MV CBCT uses the high-energy x rays from the linac itself for imaging. An
FPD that can operate at very high x-ray photon energies is used to acquire the
projection data, and a separate imaging chain is not required. Given the high
photon energy and associated decrease in photoelectric absorption, the soft tissue
contrast of MV CBCT is markedly worse than that of kV CBCT. However, it is

33

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

sufficient to visualise bony anatomy, which may be acceptable for alignment

purposes.

Co-integrated systems
(52) Co-integrated systems exist mainly in nuclear medicine (e.g. single photon
emission tomography) (Sowards-Emmerd et al., 2009). Here, a flat-panel CBCT
system is mounted on the same gantry as the nuclear imaging chain. The CBCT
data are used for attenuation correction and anatomical localisation.

2.3. Clinical scenarios where CBCT is used

(53) In current clinical practice, CBCT scanners are used for a variety of imaging
applications ranging from preclinical to clinical imaging (Table 2.1). Their use is
primarily motivated by taking advantage of the following characteristics: (1) com-
bining dynamic fluoroscopy/angiography and tomographic imaging; (2) large
z-coverage; and (3) high-resolution imaging of high-contrast structures.

34

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Table 2.1. Cone beam computed tomography (CBCT) in a variety of medical applications ranging from research to clinical imaging.

Main reason
why CBCT Common use examples
Application Setup Synonyms is used* Use cases of CBCT

Non-vascular inter- C-arm system 3D C-arm, CBCT 1, 2 Liver intervention, Spatial position of inter-
ventional abscess drainage, vention instruments
procedures skeletal and material
interventions
Vascular head/body C-arm system Angiographic CT, 1 Tumour embolisa- Spatial position of inter-
interventions rotational angiog- tion, bleeding, vention instruments,
raphy CT revascularisation rule out bleeding,
in peripheral embolisation therapy
occlusive disease control
Vascular cardiac C-arm system Rotational angiog- 1 Electrophysiological Spatial position of inter-

35
interventions raphy CT catheter ablation vention instruments
Orthopaedic Mobile C-arm/O-arm 1, 2 Osteosynthesis Spatial position of
interventions systems implants, complex
fractures
Radiation therapy Gantry or C-arm 2 Tumour therapy Patient registration,
planning/guidance (with treatment physiological motion

Downloaded from ani.sagepub.com by guest on July 22, 2015

system) control
Dental and maxillo- Over-the-head C-arm DVT 3 Dental workup, Diagnostic imaging,
facial, ENT ‘seat-scanner’/ paranasal sinus, datasets for naviga-
gantry based temporal bone tion (e.g. implant
Radiological protection in cone beam computed tomography (CBCT)

placement)
Breasty Horizontal gantry 2, 3 Rule out carcinoma,
based biopsy
(continued on next page)
 Table 2.1. (continued)

Main reason
why CBCT Common use examples
Application Setup Synonyms is used* Use cases of CBCT

Urology C-arm 2, 3 Lithotripsy, diagnos- Diagnostic imaging,

tic workup stone detection
Nuclear medicine Transmission and 2 Attenuation correc- Myocardial perfusion
hybrid imaging emission systems tion, anatomical imaging, skeletal ima-
(SPECT/CT) mounted on rotat- localisation (fused ging, oncology
ing gantry physiological and imaging
anatomical data
sets)
Peripheral bone C-arm/gantry based 3 Osteoporosis Bone microstructures,
imaging bone density

36
Animal imaging/spe- Bench-top, gantry 2, 3 Research and Experimental imaging
cimen imaging based veterinary
3D, three dimensional; CT, computed tomography; ENT, ear/nose/throat; DVT, digital volume tomography; SPECT, single photon emission computed
ICRP Publication 129

tomography.
*
1, combining dynamic fluoroscopy/angiography and tomographic imaging; 2, large z-coverage; and 3, high-resolution imaging of high-contrast structures.
y
Digital breast tomosynthesis may also be regarded as a form of limited-angle CBCT with a specialised reconstruction algorithm.

Downloaded from ani.sagepub.com by guest on July 22, 2015

 3. BIOLOGICAL EFFECTS OF RADIATION

. ICRP emphasises that protection should be optimised not only for whole-body expos-
ures, but also for exposures to specific tissues, especially those of the lens of the eye,
the heart, and the cerebrovascular system.

3.1. Introduction
(54) The health effects of ionising radiation are classified into two main categories:
tissue reactions (deterministic effects) and stochastic effects. Tissue reactions include
skin erythema, hair loss, cataracts, infertility, vascular disease, and haematopoietic
and gastroenterological effects. Stochastic effects, on the other hand, are cancer and
heritable (genetic) effects.
(55) Tissue reactions appear when the radiation dose exceeds a specific threshold.
The severity of the reaction depends on the total radiation dose received by the organ
or part of the organ. Stochastic effects are governed more by the inherent random-
ness in microscopic interactions between radiation and biological matter. In most
cancer models, the probability of cancer induction due to exposure to radiation is
considered to be proportional to the radiation dose. Moreover, for the purpose of
radiological protection, no matter how low the radiation dose, theoretically there is
always a small probability that it will induce cancer or heritable effects.

3.2. Tissue reactions

(56) For tissue reactions, the damage to cells is related directly to radiation dose,
and a dose threshold exists. Publication 103 (ICRP, 2007b) states that, ‘The reason
for the presence of this threshold dose is that radiation damage (serious malfunction
or death) of a critical population of cells in a given tissue needs to be sustained before
injury is expressed in a clinically relevant form. Above the threshold dose, the sever-
ity of the injury, including impairment of the capacity for tissue recovery, increases
with dose.’ Tissue reactions have thresholds that are typically of the order of a few
hundred mGy. Skin effects may occur at absorbed doses of 3 Gy; threshold doses for
other organs are provided in Table 3.1.
(57) As a classical example, erythematous effects commonly occurred on workers’
hands during the early days of radiology, about a century ago. Such symptoms have
occurred rarely in workers using medical x rays in the last 50 years. However, skin
injuries have been observed among patients due to fluoroscopic procedures in inter-
ventional radiology and cardiology (ICRP, 2000b, 2013; Balter et al., 2010; Rehani
and Srimahachota, 2011). Also, in interventional procedures, problems including
hair loss and chronic occupational dermatitis have been reported for radiologists
and cardiologists on body parts unprotected by the lead apron or lead table shield
(Wiper et al., 2005; Rehani and Ortiz López, 2006). To the best of the authors’
knowledge, there have been no reports, to date, of skin injuries in patients undergo-
ing CBCT. Regarding MDCT, skin injuries have been observed in the past few years

37

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Table 3.1. Estimates of threshold organ doses for tissue effects in adult human testes, ovaries,
lens of the eye, and bone marrow.

Threshold for Threshold for annual dose

total dose in a single in the case of fractionated
Tissue and effect exposure (Gy) exposure (Gy year1)

Testes
Temporary sterility 0.15 0.4
Permanent sterility 6.0 2.0
Sterility 3.0 >0.2
Lens of the eye
Cataract (visual impairment) 0.5
Bone marrow
Depression of haematopoiesis 0.5 >0.4
Heart or brain
Circulatory disease 0.5
Reproduced from ICRP (2007b; Table A.3.1) with updated information regarding the lens of the eye and
heart from ICRP (2012b).

in patients undergoing MDCT scans, mainly as a result of inappropriate use of

scanners (ICRP, 2007a). Hair loss has been reported among patients undergoing
brain perfusion CT (Bogdanich, 2009, 2010; Wintermark and Lev, 2010).
Although skin injuries related to CBCT have not been reported among patients or
workers, the technique is relatively new; as use of CBCT increases, there may be
potential for such injuries, particularly in cases of poor radiological protection
practice.
(58) Besides skin injuries, there have been recent reports of radiation effects on the
lens of the eye, which is one of the most radiosensitive tissues in the body (Rehani
et al., 2011; ICRP, 2012). Radiation-induced cataracts have been demonstrated
among workers involved in interventional procedures using x rays (Vañó et al.,
1998; ICRP, 2000b), but not with CT or CBCT. However, an earlier study by
Klein et al. (1993) and a more recent study by Yuan et al. (2013) indicated that
there may be elevated risk for damage to the lens of the eye in patients undergoing
CT scans. Similar risks can be anticipated in patients undergoing CBCT (e.g. in
neuroradiological interventions when the eye is exposed to the primary beam).
Currently, there is a paucity of data and it is difficult to judge the risk for patients.
Caution is recommended where the primary beam irradiates the eye, and thus careful
attention to optimisation is necessary.
(59) In addition to patients, there are populations exposed to low doses. For some
such groups, lens opacities have been documented, including workers in interven-
tional suites (Ciraj-Bjelac et al., 2010, 2012; Vañó et al., 2010, 2013; Rehani et al.,
2011), astronauts (Cucinotta et al., 2001; Rastegar et al., 2002), radiological tech-
nologists/radiographers (Chodick et al., 2008), atomic bomb survivors (Nakashima

38

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

et al., 2006; Neriishi et al., 2007), and people affected by the Chernobyl accident (Day
et al., 1995).
(60) Recent epidemiological data suggest that tissue reactions can occur at thresh-
old doses that are lower than considered previously (ICRP, 2010, 2012). These reac-
tions usually take a long time to manifest. For lens opacities, the threshold for
damage is now considered to be as low as an absorbed dose of 0.5 Gy, whereas it
was previously set at 2 Gy (depending upon exposure scenario). The absorbed dose
threshold for circulatory disease has been chosen as 0.5 Gy to the heart or brain, as a
precautionary value. ICRP policy has been not to set any dose limits for patients.
However, the current recommendation of ICRP for occupational exposure in
planned exposure situations is an equivalent dose limit for the lens of the eye of
20 mSv year1, averaged over a defined 5-year period, with no single year exceeding
50 mSv (ICRP, 2012). Occupational doses to the lens of the eye of a few mGy in
CBCT have been reported in the literature. Doses to the lens of the eye for patients
are a few mGy for dental and head and neck CBCT with direct exposure, but doses
are much higher for interventional CBCT. Details regarding doses to the lens of the
eye in CBCT for patients and workers are available in Sections 6 and 7.

3.3. Stochastic effects

(61) Cancer and heritable effects come into the category of stochastic effects. The
probability of carcinogenic effects is much higher than that of heritable effects. This
follows from Publication 103 (ICRP, 2007b) which states that the detriment-adjusted
nominal risk coefficient or stochastic effects for the whole population after exposure
to low doses of radiation is 5.5% Sv1 for cancer and 0.2% Sv1 for heritable effects.
The latter is a theoretical risk for humans, as all documented cases of radiation-
induced heritable effects come from observations in non-human species. Cases in
humans have not been observed, even for atomic bomb survivors. Therefore, after
careful review of many decades of literature, ICRP has reduced the tissue-weighting
factor for the gonads relating to the risk of heritable effects to less than half the
previous value, from 0.2 to 0.08 (ICRP, 2007b).

39

Downloaded from ani.sagepub.com by guest on July 22, 2015

Downloaded from ani.sagepub.com by guest on July 22, 2015
 4. PRINCIPLES OF RADIOLOGICAL PROTECTION FOR PATIENTS
AND WORKERS

(62) The system of protection recommended by ICRP consists of three fundamen-

tal principles of radiological protection: justification, optimisation of protection, and
application of dose limits (ICRP, 2007b). Dose limits are only applicable for radio-
logical protection of workers and the public; diagnostic reference levels (DRLs) are
used for the protection of patients (ICRP, 2007c).

4.1. Justification
(63) The justification principle requires that any decision that alters the exposure
situation should do more good than harm. According to ICRP, there are three levels
of justification for the use of radiation in medicine.
. At Level 1, the use of radiation in medicine is acceptable when it results in more
good than harm to the patient. It is now taken for granted that the use of x rays
in medicine is justified.
. At Level 2, a specified procedure with a specified objective is defined and
justified (e.g. a CBCT examination for patients showing relevant symptoms,
or a group of individuals at risk of a condition that can be detected and
treated).
. At Level 3, the use of radiation in an individual patient should be justified (e.g.
the particular CBCT application should be judged to do more good than harm
to the individual patient).
(64) According to Publication 87 (ICRP, 2000a), requests for a CT examination
should only be generated by properly qualified medical or dental practitioners as
defined by national educational and qualification systems. Justifying individual
exposures should include verification that the information required is not already
available from previous studies, and that the proposed study is really going to answer
the questions posed (ICRP, 2007b). The referring practitioners and imaging profes-
sionals should be skilled in the selection of, and indications for, CT, CBCT, and
angiography, and possess adequate knowledge concerning alternative techniques.
This training should also apply to non-imaging professionals who plan to use
CBCT. Further aspects of training are provided in Section 8. The availability of
resources and cost should also be considered in the justification process.
(65) Justification of CBCT is a shared responsibility between the referring practi-
tioner and the imaging professional. In the case of self-referral (e.g. practitioners in
outpatient dental and ENT clinics), wherein the referring practitioner and the ima-
ging professional are the same person, their responsibilities are combined within one
person. Referring practitioners know their patients and their medical histories, but
typically have little or no knowledge about radiation doses, or the risks and limita-
tions of diagnostic radiological examinations. On the other hand, imaging profes-
sionals have expertise regarding radiological examinations, including knowledge of
alternate imaging examinations that can provide similar information with less

41

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

radiation exposure of the patient; they, however, lack in-depth knowledge about the
individual patient’s condition. Consultation between imaging professionals and
referring practitioners is essential to make the most of their combined knowledge.
While such consultation has been emphasised before, practical constraints have
made its implementation difficult to realise in practice, and there is a need for explor-
ation of tools to make this possible.
(66) ICRP has noted that there are many reports documenting lack of justification,
particularly for CT examinations although not yet for CBCT (Rehani and Frush,
2010; Fraser and Reed, 2013). ICRP recommends the use of modern technologies
such as clinical decision support systems with electronic referral to improve
justification.

4.2. Optimisation
(67) Once an examination is justified, protection of medical workers and patients
must be optimised.
(68) The primary role for optimisation of CBCT lies with the CBCT facility, and it
should ensure that the examination is performed with the lowest radiation dose to
the patient while obtaining the image quality required for the clinical purpose.
(69) DRLs have been used to promote optimisation and have shown good results
in many countries, particularly for CT applications. They were developed to identify
examinations with doses above the 75th percentile in the dose distribution so that
corrective actions could be taken. However, as expressed in ICRP’s concept of as low
as reasonably achievable, they do not obviate the need for optimisation below the
75th percentile dose (Rehani, 2013). With modern technical equipment and opti-
mised protocols, dose levels between the 25th and 50th percentile are achievable
(NCRP, 2012), so users should aim to optimise within DRLs (Rehani, 2013). The
optimisation of patient protection in CBCT requires the application of examination-
specific scan protocols tailored to patient age or size, region of imaging, and clinical
indication. Protocols provided by the vendors of CT scanners should be evaluated
for optimisation. DRLs are just one of the practical tools to promote the assessment
of existing protocols. The ability to compare dose levels between CBCT facilities
would facilitate the development of appropriate, new, and improved protocols at
each CBCT centre.
(70) DRLs for CBCT procedures need to be established. To achieve this, doctors
performing CBCT examinations should work closely with medical physicists.

4.3. Requirements for imaging facilities

(71) Practice varies worldwide but should comply with requirements laid down by
national authorities. Typically, each CBCT scanner should be registered with the
appropriate database under the overall oversight of a national or designated author-
ity. Frequently, during the process of registration and authorisation, an authority
will examine the specifications of the machine, and the size and shielding of the room

42

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

where it is going to be used, ensuring that workers and members of the public are
sufficiently protected. The International Electrotechnical Commission (IEC, 2012)
and the International Organization for Standardization provide international level
safety requirements for x-ray machines. In many countries, national standards for
x-ray machines are also available. These requirements are intended to protect work-
ers and members of the public who may be exposed to radiation. The registration
and authorisation process will also assess the availability of qualified staff. There are
requirements for periodic quality control tests for constancy and performance evalu-
ation. Acceptance tests and periodic quality control testing of CBCT equipment can
provide confidence in equipment safety and its ability to provide images of optimal
image quality. Such periodic testing is essential because a malfunctioning machine
may expose patients unnecessarily to radiation without any other overt signs.
Nevertheless, whatever national requirements are in place, it is essential that they
are followed in order to ensure that facility design and operation are safe for
patients, workers, and the public.

43

Downloaded from ani.sagepub.com by guest on July 22, 2015

Downloaded from ani.sagepub.com by guest on July 22, 2015
 5. ASSESSING PATIENT DOSES IN CBCT

. Equipment used for both fluoroscopy and CBCT should provide aggregate dose
indices for individual patients throughout the procedure through electronic display
on the operator console and a radiation dose structured report.

5.1. Dosimetry in CBCT

(72) CBCT uses a wide x-ray beam for 3D imaging of a relatively large volume.
Since the mid-1990s, the trend in MDCT has been towards an ever-increasing
number of slices with a concomitant increase in x-ray beam width; the z-axis cover-
age of the high-end, wide-area MDCT scanners available today rivals that of CBCT.
These developments have created a drive to update CT dosimetry methods so that
they are more apropos wide-area detectors. As a result, some of the work from
MDCT dosimetry, for which established measurement methods and phantoms
exist already, can be translated to CBCT dosimetry. This section and Annex A
present the shortcomings of the standard narrow-beam MDCT formalism when it
is applied directly to CBCT. Methods to overcome these problems are described in
Annex A in order to construct a comprehensive framework for CBCT dosimetry.
(73) CT dosimetry has evolved around the concept of the CT dose index (CTDI).
In order to connect CTDI-like measurements with dose, volume CTDI (CTDIvol)
and dose length product (DLP) have been used extensively in clinical practice as
relative patient dose indicators. One of the strengths of CTDI is its relevance as a QA
metric, and its measurement in a phantom under conditions mimicking those for a
scan of a patient (e.g. attenuation by the table).
(74) The limitation of CTDI for wider beams has led to new approaches in CT
dosimetry, details of which are provided in Annex A. The CTDI paradigm is prob-
lematic when there is no helical scan or patient motion (as is the case with many
CBCT scanners). In such cases, reported CTDIvol values will overestimate the dose
significantly (Dixon and Boone, 2010a).

5.2. Point-of-care scanning and clinic-based CBCT systems

(75) Clinic-based systems include head and neck CBCT, breast CT (bCT), and
dental and maxillofacial CBCT. A particular property of dental and maxillofacial
CBCT scanners is that, depending on the system, varying FOV sizes are offered. This
allows for the scanning of localised regions (i.e. a single tooth and its immediate
surroundings) as well as maxillofacial scanning. The use of horizontal collimation, as
well as other factors, results in complicated dose distributions in the axial plane,
providing an additional challenge for dosimetry (Pauwels, 2012a). In addition, most
dental and maxillofacial CBCT scanners are seated or standing, resulting in practical
complications regarding phantom and dosimeter placement.
(76) Various possible dose indices have been proposed for dental and maxillofacial
CBCT (EC, 2012a; Pauwels, 2012a; DIN, 2013). Further validation of possible

45

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

indices is required, together with a way to translate dose index readings into patient
doses.
(77) Technically, these methods could also be applied to other clinic-based systems
including systems for head and neck imaging and possibly bCT. However, there is
currently no standardisation for the measurement of such units. This highlights the
fact that the issue of standardisation in CBCT dosimetry remains largely unresolved.

5.3. C-arm CBCT systems

(78) C-arm CBCT systems are incapable of performing a full rotation around the
patient couch. Some systems, however, can only rotate 180o plus the beam angle
(Fahrig et al., 2006), which results in a non-uniform axial dose deposition to the
patient/phantom. Certain dental and maxillofacial CBCT scanners also scan along a
180–220o trajectory. In a phantom, the maximum dose occurs at the central plane
intersecting the z-axis at z ¼ 0, on the side of the phantom closest to the x-ray tube. In
the ideal case in which the heel effect is absent, the maximum dose would occur on
the bisector of the rotation angle. When the heel effect is present, the maximum dose
occurs near the bisector.
(79) For C-arm CBCT systems, Fahrig et al. (2006) proposed a metric representing
the average dose to the phantom central plane, following similar averaging to that
applied in calculation of the weighted CTDI (CTDIw).

5.4. A unified approach to CT dosimetry

(80) In its Report 87, the International Commission on Radiation Units and
Measurements (ICRU, 2012) reviewed a considerable body of work in order to
propose a method for CT dosimetry that compensates for the shortcomings of
current CTDI-based CT dosimetry methods. In addition, earlier work by Dixon
and Boone (2010b) provided a unified formalism for dose measurements on
machines capable of helical scanning (e.g. MDCT scanners), as well as on machines
that only acquire axial images (which is the case with most CBCT scanners). A set
of metrics and the use of a new 600-mm-long polyethylene phantom are proposed.
The mathematical foundation for the method is beyond the scope of this publica-
tion, but the method is described briefly here; more details can be found in Annex
A. The method is based on a measurement of the cumulative absorbed dose at the
centre of the 600-mm phantom from a complete scan. As the width of the beam for
a CBCT scan is increased, the cumulative dose at the centre will rise. However, as
the width of a CBCT beam is increased, the distance of the outer edge of the beam
from the centre of the phantom becomes greater, and contributions from scattered
photons at the outer edge to the cumulative absorbed dose at the centre become
smaller. Beyond a certain beam width, further increases in dose become negligible,
and an equilibrium dose is attained. The quantity H(L) is defined as the ratio of
the cumulative absorbed dose for a beam width or scan length L and the equilib-
rium dose.

46

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

(81) The physical interpretation of the rise to equilibrium curve presented in

Annex A is that the scan and the phantom need to be long enough so that the
asymptote tails of the profiles are reached. The longer the scan, the closer H(L) is
to unity. This representation shows that the dose to the central CT slice in a scan
increases with scan length, demonstrating the relatively low efficiency of short scans
for collecting the actual dose; this efficiency increases with longer scans.

5.5. Tracking and reporting of radiation dose

(82) Systems used for both fluoroscopy and tomography (CBCT) face new chal-
lenges. Currently, there is no standardised way to assess the aggregate radiation dose
to a patient during a single procedure. This situation needs to be addressed, and these
imaging systems should provide a means of not only comparing but also consolidat-
ing doses from both the fluoroscopy (2D) and CT (3D) components of a procedure.
Furthermore, tracking and reporting of the radiation dose for a single patient should
be facilitated. Errors for displayed dose estimates should not exceed 20% (IAEA,
2011b; IEC, 2011; EC, 2012b). A radiation dose structured report (RDSR) can be
used to report the modality output following the existing RDSR in CT and angiog-
raphy. KAP values for the different orientation of the beam can be reported inside
the RDSR when step-and-shoot acquisition techniques are used. In addition, KAP
can be considered for CBCT in fluoroscopy and other applications to facilitate direct
comparison (e.g. between 2D and 3D fluoroscopy). Effective dose is not a suitable
dosimetric quantity for reporting.

5.6. Epilogue
(83) The unified CT dosimetry method proposed by ICRU (2012) has the potential
to standardise CBCT dosimetry. Nevertheless, the value of CTDI-based measure-
ments should not be underestimated. Although CTDI has limitations, it has been
evaluated on many systems over the years, and provides important comparisons in
output for CT scanners from different manufacturers and ages. Moreover, coeffi-
cients for patient dose estimations based on CTDIvol are already available.

47

Downloaded from ani.sagepub.com by guest on July 22, 2015

Downloaded from ani.sagepub.com by guest on July 22, 2015
 6. OPTIMISATION OF PROTECTION OF PATIENTS AND WORKERS
IN CBCT

. Optimisation of both patient and worker doses, particularly when workers have to be
near the machine, is important when monitoring of doses becomes an essential tool.
Recording, reporting, and tracking of radiation dose for a single patient should be
made possible in a consistent manner across vendors.
. Low-dose protocols may be sufficient to answer diagnostic questions focused on high-
contrast structures, such as lung, bones, dental and maxillofacial scans, ENT scans
(paranasal sinuses, skull, temporal bone), interventional material, and contrast-
enhanced vessels (angiographic interventions).
. Higher-dose protocols should only be selected if visualisation of soft tissue struc-
tures, such as intracranial haemorrhage, soft tissue tumours, or abscesses, is the
primary focus.
. Most interventional and intraprocedural C-arm CBCT systems can scan an angular
range spanning 180–240o plus the cone angle of the x-ray beam. Localised critical
organs, such as the thyroid, eyes, female breasts, and gonads, should be on the
‘detector side’ of the arc whenever possible.
. Clinical need permitting, every effort should be made by users to ensure that the
volume of interest is fully incorporated in the FOV provided by the CBCT scanners,
while radiosensitive organs should be placed outside the FOV.
. The aim of CBCT should be to answer a specific diagnostic or intra-operative ques-
tion vis-à-vis other imaging modalities, and not to obtain image quality that rivals
MDCT. The decision by the referring practitioner to use CBCT should be made in
consultation with an imaging professional.
. There is a need to provide checks and balances, such as dose check alerts imple-
mented in CT in recent years, to avoid high patient doses compared with locally
defined reference values.
. Methods that provide reliable estimates of dose to the eye under practical situations
should be established and used.

6.1. Introduction
(84) CBCT scanners are highly engineered machines, and dose optimisation is a
multi-factorial problem. The imparted radiation dose may vary by several orders
of magnitude between different scan models and different ways of using the
machine. Clinical use of CBCT requires insight into the various trade-offs in
order to maximise patient benefit and minimise risk. It is essential to understand
various technological factors and scan parameters that influence dose. Knowledge
of MDCT alone is not sufficient in this endeavour as CBCT scanner systems differ
significantly from MDCT scanners in their mode of operation. For example,
while spiral scanning is the norm with MDCT, nearly all CBCT imaging is
done using a single axial scan. In addition, several special conditions exist that
do not apply to MDCT scanners (e.g. restriction of the FOV of a typical CBCT

49

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

scanner). It is therefore essential to involve a medical physicist or another suitably

qualified expert (AAPM, 2011a; EC, 2014) early on in optimisation, as well as the
audit of patient and occupational dose levels, particularly for high-dose
procedures.

6.2. Factors influencing dose to the patient

6.2.1. Equipment-dependent factors

Knowing your equipment

(85) It is important that users understand how their equipment functions,
because each CBCT scanner has some unique features, such as the application
domain, gantry design, and detector configurations. The complexity of modern
equipment necessitates a thorough understanding of the various scan modes,
parameter settings, and dose optimisation strategies. This section deals with
equipment features that affect radiation dose, and the next section is devoted to
operator actions required to achieve optimal radiological protection in clinical
scans.

Collimation
(86) In MDCT, the region of interest is usually prescribed on one and sometimes
two orthogonal scan projection radiographs [also known as antero-posterior (AP)
and lateral (LAT) scout views or topograms]; the scanner covers this scan FOV
helically or axially and reconstructs tomographic slices. Similar AP and LAT pro-
jection views may also be acquired in CBCT scanning; however, the entire FOV
usually fits within a single circular trajectory of the scanner, and helical scanning
is not used in most applications. Although the x-ray beam will not generally extend
beyond detector dimensions in situations where the detector is movable, a portion of
the beam may fall outside the detector margins. Care should be taken to collimate
the x-ray beam so that it falls entirely within the detector margins; automatic means
for delimiting the collimation window to the detector size may or may not exist,
depending on the particular scanner manufacturer and model. Any radiation outside
the detector constitutes unnecessary radiation to the patient. The beam should be
further collimated to limit its z-extent to the FOV. The source-to-detector distance
determines the maximum lateral extent of the FOV that can be scanned, and should
be adjusted appropriately depending on the anatomy under consideration. It should
be noted that the scatter noise in the projection data increases approximately linearly
with the area of the irradiated field. In general, the x-ray beam should be collimated
tightly as it not only lowers the x-ray dose but also decreases scatter, thereby improv-
ing image quality.
(87) A poorly collimated primary beam, if it is outside the patient, may
increase the occupational dose, as well as the patient dose, significantly. It is
also desirable to exclude any adjacent sensitive organs that do not need to be

50

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

imaged from the scan FOV to address the clinical question at hand. The x-ray
beam should be collimated tightly to the scan FOV. As a CBCT scan cannot be
extended in the same way as an MDCT scan, caution must be exercised to ensure
that the volume of interest is fully incorporated in the FOV provided by the
CBCT scanner.

Collimation along the z-axis

(88) Many CBCT scanners provide a means for the user to collimate the beam.
Collimation along the z-axis to achieve as narrow a beam as possible to fulfil the
clinical purpose will reduce the patient dose and improve the image quality. Use of
the thinnest possible collimation (2.3 cm) instead of the full field (19 cm) improves the
contrast-to-noise ratio.
(89) Free-in-air geometric efficiency is a means of quantifying over-beaming
(i.e. the proportion of radiation falling outside the detector margins) (Berris
et al., 2013). In CBCT scanners, the x-ray beam is usually fully intercepted by
the receptor, so the free-in-air geometric efficiency should be 100%, and over-
beaming should not occur. Furthermore, over-scanning (also known as over-
ranging), which is required at either end of helical scans to provide additional
data for image reconstruction, is not needed for axial CBCT scans (Tzedakis
et al., 2005).
(90) An effect that always occurs in CBCT is that parts of the irradiated volume
are hit by radiation, but are not fully contained in 180 of projections. Images of
these regions, shown in Fig. 6.1, cannot be reconstructed or can only be partially
reconstructed. The region that cannot be reconstructed broadens as the cone angle
increases (Grimmer et al., 2009).

Fig. 6.1. In cone beam computed tomography, data are only available from 180
projections within the region in the hexagon that is marked with parenthesis B. A part of
the irradiated volume (parenthesis A) cannot be reconstructed (or only with reduced image
quality), because data from all 180 projections are not available. The size of this area
depends on the geometry of the scanner (qualitative depiction).

51

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Dose distribution within the scan field of view along the z-axis
(91) Ideal CT scanner systems should irradiate the examined volume along the
z-axis with a homogenous dose that should decrease rapidly outside the examined
volume. In some CBCT systems, the dose distribution is different, and the central
slices receive larger amounts of radiation (Gupta et al., 2006). Wherever possible,
radiosensitive organs should be placed outside the irradiated volume, which is nor-
mally wider than the FOV, provided the clinical requirements of the procedure
permit.

Dose distribution in cases of volume-of-interest scanning

(92) In certain situations, only a small volume may be of clinical interest. Some
CBCT scanners provide a very narrow beam collimation with a relatively small
detector. A large part of the irradiated volume will be out of the primary x-ray
beam at most angular projection positions. In general, a scan volume that is
delimited in the x–y direction to a small portion of a larger body part results in
truncation artefacts. However, small-volume CBCT of high-contrast structures
such as bones and teeth, when used in conjunction with an artefact reduction
algorithm, may well give clinically acceptable images. For example, a truncation
artefact arising from a limited FOV may not affect assessment of a transpedicular
screw. This must not be confused with retrospective, selective reconstruction of a
certain region of interest inside a larger scanned volume (Table 6.1). The dose
distribution outside the volume of interest is very different in the two scanning
modes. Therefore, the user should verify whether volume-of-interest scanning is
applicable in a certain situation.

Type of detector
(93) Most currently available CBCT systems use a digital FPD. State-of-the-art
digital FPDs are offered at several gains and effective dynamic range settings. In
general, the dynamic range of digital FPDs is narrower than for MDCT detectors,
resulting in lower soft tissue contrast for CBCT scanners. The afterglow of the cae-
sium iodide (CsI) scintillators used in FPDs limits the maximum image frame rate
that can be obtained from these detectors. Typically, 30 frames s1 can be obtained
at the full FOV; a narrower FOV can provide a faster frame rate of 100–120
frames s1 (Gupta et al., 2008). Slow frame acquisition rate is the main reason for
the relatively high acquisition times of CBCT systems (Orth et al., 2008); the fastest
clinically available CBCT had an acquisition time of 5 s compared with 80 ms for a
dual-source MDCT system. Parameters such as pixel size and scintillation crystal
thickness are usually selected based on target application (e.g. maxillofacial imaging
or C-arm angiography), and the end-user has no control over their selection. No
detector technologies in current use should be strictly avoided from a radiological
protection standpoint.

52

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Table 6.1. Volume-of-interest scanning vs standard scanning. Volume-of-interest scanning is

a valuable method to reduce the radiation exposure of in-plane structures, if imaging condi-
tions allow (high-contrast structures). It must not be confused with standard scanning for
region-of-interest reconstruction.
Irradiated volume
from all
directions (from
all angular Reconstructed Radiation
positions) volume exposure Applications

Volume-of- Limited to cylin- Limited to cylin- Only volume of Mostly dental and
interest drical volume of drical volume of interest maxillofacial ima-
scanning interest interest receives full ging and most
dose interventional C-
arm setups when
body trunk is
scanned
Standard Large cross-section Anywhere within Whole-body All other
scanning body diameter, diameter
whole-body receives full
diameter, or parts dose
of full cross-
section

(94) A minority of CBCT systems still use charge-coupled device cameras

coupled with x-ray image intensifiers. The convex input screen and image distor-
tion of image intensifier systems result in non-uniform image quality across the
output image. In addition, light and electron scattering within the image intensifier
limits the contrast resolution of the reconstructed slices. CBCT systems typically
have an 8–10-bit dynamic range and can only support a very coarse level of tissue
differentiation.

Detector quantum efficiency

(95) Detector quantum efficiency (DQE) is a widely used metric that describes the
dose efficiency of an x-ray detector. Without going into detail, it measures the quality
of the image produced by the detector from a given dose or fluence to the detector.
Intuitively, it captures how well a detector translates the signal incident on it into an
image, relative to an ideal detector. Specifically, it is the square of the ratio of input
and output SNRs of a detector. For example, a detector that reduces the SNR by
50% has DQE of 0.25. The ideal detector would have DQE of 1 and would translate
all incident x-ray photons into image information. DQE is normally given as a
function of spatial frequency, and correlates image quality with incident x-ray
dose at a detector level.
(96) Current caesium iodide hydrogenated amorphous silicon (CsI-aSi:H) FPDs
have DQEs in the range of 0.6–0.7, which are lower than those of MDCT detector

53

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

systems (Gupta et al., 2006). This is a fundamental limitation that is beyond the
control of the user, and means that CBCT images will be noisier than MDCT images
for the same input radiation.

Filtration
(97) A bowtie filter in the imaging chain hardens and attenuates the x-ray beam,
reduces the scatter-to-primary ratio, and reduces the x-ray fluence heterogeneity at
the detector. Bowtie filters decrease the scatter contribution from the object periph-
ery in MDCT imaging (Orth et al., 2008). Ning et al. (2000a) showed that the quan-
tity [SNR2/entrance exposure] decreases when kVp increases for a flat-panel-based
CBCT system. This means that there is a trade-off between decreased scatter from
the object periphery (when the bowtie filter is on) and improved detector efficiency
from the ‘softer’ beam (without a bowtie filter) (Orth et al., 2008). Use of a bowtie
filter is standard in MDCT. In CBCT, a bowtie filter is not used commonly, but its
use is increasing. Other configurations such as half-bowtie filters that enable cover-
age of a large area have also been used (Wen et al., 2007). The presence of the filter
can reduce patient dose, especially at the patient periphery, and can improve tomo-
graphic image quality by improving uniformity, CT number accuracy, and contrast-
to-noise ratio. One potential disadvantage, however, is the decrease in detector effi-
ciency due to beam hardening (Mail et al., 2009). In general, a bowtie filter should be
used when imaging a wide FOV where the anatomy under consideration only occu-
pies a small central portion. Assessment of spinal hardware would be one example
application. Special care must be taken if the bowtie filter is removable; workers can
forget to mount the bowtie filter prior to imaging, resulting in additional dose to the
patient.

Anti-scatter grid
(98) An anti-scatter grid is placed between patient and detector, and consists
of lead septa that are oriented along lines projecting radially outwards from the
focal spot. This geometry allows the primary beam to reach the detector while
the off-axis radiation is absorbed. As such, an anti-scatter grid in front of the
flat panel can prevent the scatter generated by the patient from reaching the
detector. The leaves reduce the effective detector area to a small degree. The
geometry of the anti-scatter grid, which determines its selectivity and its rejec-
tion efficiency, is optimised for the scanner and application. Anti-scatter grids
are highly sensitive to the source-to-detector distance; if the latter can be
varied, or if a choice of anti-scatter grids is provided, it is essential to match
these two parameters.
(99) The efficiency of anti-scatter grids for scatter suppression and image quality
improvement has been assessed for CBCT. Although the presence of a grid did not
seem to improve the SNR in relation to applied radiation dose (Schafer et al., 2012),
a significant decrease in cupping artefacts was observed (Kyriakou and Kalender,

54

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

2007). However, in certain high-scatter conditions, the grid could lead to a reduction
in dose of up to 50% (Kyriakou and Kalender, 2007).
(100) The anti-scatter grid, if available, is usually a fixed hardware parameter
that is optimised for a certain application and a specific geometry. Typically, the
end-user has little influence on the geometry of the anti-scatter grid. However, if a
choice of different grids and geometric distances is provided, it is essential that the
two are matched for the system to function properly.

Scatter correction algorithm

(101) Scatter intensity has a broad angular distribution around the image of the
scattering object. One can think of the projection image obtained by the detector as a
2D-smeared image of the object that includes both the primary and the scatter
radiation. At any point that can receive both the primary and scatter photons,
these two components may be difficult to separate. However, in areas that are
shielded from the primary beam by the collimator, the scattered component is
observable because of the broad distribution of the scatter. An assessment of this
can be used to estimate the amount of scatter in the rest of the image. By assuming a
scattering function, the scatter profile throughout the image can be estimated. This
can then be subtracted from the measured signal to compute the contribution from
the primary signal. If a particular CBCT scanner provides a set of steps for comput-
ing the scatter function, that protocol should be strictly followed. Besides vendor-
implemented algorithms, the user has little influence over the scatter correction
algorithms.

Data correction algorithms

(102) Multiple correction algorithms are typically applied to the raw projection
data, before they can be reconstructed into a 3D stack. The following is a partial list
of data conditioning algorithms typically employed to compensate for system imper-
fections: (1) offset subtraction; (2) afterglow correction; (3) adaptive filter mask; (4)
normalisation; (5) theta correction; (6) cross-talk difference correction; (7) air cali-
bration; (8) beam hardening correction; and (9) detector z-gain non-uniformity cor-
rection. These corrections tend to be vendor specific, over which the end-user has no
control.

6.2.2. Operator-dependent factors

Reduced arc scanning

(103) Many CBCT systems are capable of reconstruction from less than 360
angular acquisitions. In general, a coverage of 180 plus the cone angle is sufficient
for tomographic reconstruction. This gives the operator considerable flexibility in
selectivity, allowing reduction of patient exposure. For example, an appropriate

55

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

choice of starting and stopping angle can be used to limit projection images of a
patient’s head to posterior angles, reducing the dose to the lens of the eye
(Kyriakou et al., 2008a) (Fig. 6.2). Daly et al. (2006) observed a five-fold decrease
in dose to the eye when 3D images were generated using a C-arm half-cycle (178 )
rotation performed with the x-ray tube posterior to the skull rather than anterior.
Another example where this is used is in CBCT imaging of the breast, where the
imaging angles can be chosen to limit unnecessary exposure of the heart and lungs.
These manoeuvres typically have no appreciable effect on the image quality in the
central portions of the scan. Selecting an appropriate angular span for the scan arc,
a parameter that has a direct impact on the dose distribution, is a user-selectable
parameter. The user should select the scan arc so that radiosensitive organs are on
the detector side of the imaging chain.
(104) Dental and maxillofacial CBCT differs regarding the use of a reduced arc.

Firstly, the start- and endpoints of a 180 rotation cannot be selected by the user,
with the detector typically being on the anterior side of the patient. However, simu-
lations and phantom studies have shown that patient dose may be lower when the
tube is on the anterior side, although differences were 10% or lower (Morant et al.,
2013; Zhang et al., 2013; Pauwels et al., 2014). This can be explained by the anterior
placement of FOVs for dental examinations, which results in several radiosensitive
organs being posterior to the centre of rotation (e.g. salivary glands or thyroid).
More evidence is needed before a definitive recommendation can be made to
manufacturers.

Fig. 6.2. In contrast to multi-detector computed tomography, cone beam computed tomog-

raphy is mostly performed with a half scan angle (180 plus cone angle). As such, the pos-
ition of the scan angle has a significant influence on the dose distribution within the patient
(Kyriakou et al., 2008a).

56

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Setting of kVp and mAs

(105) The parameters that determine x-ray beam flux and energy spectrum (i.e.
mA and kVp settings) should be kept as low as possible without compromising
the image quality and clinical utility of the scan. kVp and mA are the main user-
selectable variables that determine the overall dose to the patient. If all other
parameters are held constant, the radiation dose is directly proportional to the
applied mAs (tube current  duration of scan rotation), and this parameter influ-
ences the noise in the image significantly. As long as the detector is not saturated,
there is a direct relationship between the level of image quality and increasing
mAs. The dependence of the radiation dose and image quality on the kVp setting
is more complex. Higher-energy photons result in less interaction with tissue;
they give poorer contrast between tissues, but a larger number of photons pass
through the tissue and reach the detector to form the image. The right kVp and
mAs setting depends heavily on the anatomy being scanned, whether or not a
contrast medium was used, and several design factors such as filter systems,
frame rate, and detector type. Therefore, it is difficult to provide absolute guide-
lines. All commercial CBCT scanners come with a manufacturer-recommended
protocol for each application. The best advice to the user is to start with this
protocol and, working in conjunction with a medical physicist or another
domain expert, adapt it to the local conditions. One should also monitor publi-
cations and guidelines dedicated to the particular scanner setup or type of
examination.

Automatic exposure control

(106) AEC systems adapt the radiation exposure to obtain a desired level of image
quality and adjust the dose to that needed for the specific body part of the patient.
Similar to MDCT, AEC modulates the tube current according to patient attenuation
in a given angular direction. Usually, AEC is implemented as a feedback loop that
controls the x-ray source based on feedback from the detector. Reductions in dose by
20–40% through the use of AEC systems have been reported in CT (McCollough,
2005). It has been found that absorbed doses vary considerably during a CBCT
acquisition, which shows the potential of using tube current modulation (He et al.,
2010).
(107) Many CBCT systems do not employ AEC, instead using a fixed tube current
setting for the entire scan. AEC is almost not available in head and neck, and dental
cone beam scanners (fixed settings protocols), and is more widely implemented in
whole-body C-arm systems. The use of tube current modulation is reduced in CBCT
due to the wide z-axis coverage. Also, the demand for AEC is less stringent when
scanning the head compared with other parts of the body. The requirements and
demands on AEC are still evolving, and general guidelines are difficult to formulate.
More details on the patient-specific factors involved in the potential application of
AEC can be found in Section 6.2.3.

57

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Scan modes: number of projections

(108) In contrast to MDCT scanning, where the user is unable to select the
number of projections explicitly, this parameter is often directly selectable in
CBCT. The most commonly used detectors in CBCT systems have a much
slower time to readout, and require a wait time after each projection in order to
account for the afterglow of the scintillator. The dose delivered in each scan is also
limited because of the number of photons that can be collected by each projection
without overexposing the detector. Optimisation of the scan time using a tight
control over each exposure is much more critical in CBCT than in MDCT. These
considerations limit the range of dwell time and dose in each projection. By con-
trolling the number of projections, for example, or changing the total scan time,
one can control the dose for a scan protocol; increasing the number of projections
increases the applied radiation dose proportionately. In CBCT, the number of
projections, together with the associated changes in the total scan time, provides
a trade-off between image quality and the delivered dose that is influenced directly
by user-selected parameters.

Scan modes: binning and spatial resolution

(109) The detector elements in angiographic C-arm CBCT systems, in con-
trast to MDCT detector systems, are much smaller in order to provide the
necessary spatial resolution for fluoroscopy and angiography modes. For
example, a common FPD for C-arm systems offers a native pixel size of
154 mm in a 1920  2480 matrix. The time to readout such a large matrix,
coupled with the afterglow of the CsI scintillators, limits the maximum frame
rate achievable on such a detector. The frame rate of a CBCT detector can be as
much as one to two orders of magnitude lower than that in MDCT. A low
readout frame rate accounts for the relatively high acquisition times of
CBCT systems. For example, the fastest available clinical CBCT had an acqui-
sition time of a few seconds compared with 80 ms for a dual-source MDCT
system (Orth et al., 2008).
(110) While it may not be possible to control the afterglow or afterlag of the
scintillator, the size of the image matrix that needs to be readout can be decreased
to make the image transfer faster. A set of binning modes is provided to accom-
plish this. Each binning mode combines neighbouring detector rows and columns
in order to reduce the matrix size and the readout time. Typical binning modes
involve a 2  2 and 3  3 area, thereby reducing the data to be streamed out by a
factor of four and nine, respectively. Despite this averaging, the spatial resolution
of CBCT is higher than that in MDCT and is often above the demands of the
clinical application. Since the image noise, spatial resolution, and radiation dose
are inter-related, the user must decide on the acceptable image quality and the
spatial resolution. This choice, in turn, determines the radiation dose. The user
should not be tempted to reduce the image noise (e.g. by increasing the tube
current or increasing the number of projections using modes such as the ‘high-

58

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

quality scan mode’ offered on some systems) to reach a noise level that is com-
parable to that of MDCT. The dose penalty associated with these scans can be
much higher than would be warranted by the clinical question at hand (Blaickner
and Neuwirth, 2013). Postprocessing techniques, such as slice averaging, thick
multi-planar reformation, and use of a softer reconstruction kernel, are preferable
when trading off among competing metrics such as image noise, low-contrast
resolution, spatial resolution, and radiation dose.

Scan modes: predefined scan protocols

(111) The use of an organ-specific protocol (e.g. ‘routine head’) or a clinical
indication-specific protocol (e.g. ‘appendicitis protocol’) is established practice in
MDCT. In routine clinical care, vast libraries of such scan protocols are avail-
able. Similar to MDCT, many CBCT systems also provide predefined scan proto-
cols that encapsulate detector settings, reconstruction kernels, and other scanner
parameters. In CBCT, however, the use is less well established, with many proto-
cols named suggestively with prefixes such as ‘low-quality’ or ‘high-quality’, the
latter unflatteringly implying that the base protocol might not provide appropri-
ate image quality in certain situations (Table 6.2). Generally, the naming of the
scan protocols refers to the well-known and, within limits, physically fixed trade-
off between image quality parameters and radiation dose. High-quality scan
protocols usually provide ‘better’ image quality at ‘higher’ radiation dose.
These simple prefixes often belie the magnitude of the change that occurs; a
high-quality protocol may entail a 6- to 10-fold increase in radiation dose com-
pared with a low-quality or standard-quality protocol. In CBCT, the selection of
the scan mode or scan protocol is one of the most significant factors influencing
radiation dose (Kyriakou et al., 2008a). A low-dose scan protocol may be suffi-
cient for high-contrast structures, such as bones, teeth, kidney stones, and con-
trast-enhanced blood vessels. The manufacturers are beginning to provide scan
protocols that are named for the diagnostic challenge they are trying to address
(e.g. ‘bone’, ‘kidney stone’, ‘rule out intracranial haemorrhage’, or ‘skull base’
protocol). There should be a dedicated section for paediatric protocols. These
have special significance when the imaging system does not have AEC (e.g. in
most dental and maxillofacial CBCT scanners) to account for the lower diameter
of children’s body parts.
(112) The user interface for CBCT scanners also deserves a special mention. The
checks and balances that are routine in MDCT scanners may be missing in CBCT
scanners. For example, two vastly different but similarly named protocols may be
adjacent to each other on the user interface, or a single mouse click may cause a
10-fold change in the delivered dose. This is in sharp contrast to MDCT where
such a large increase in radiation requires several purposeful manipulations of
scan parameters and concomitant confirmation to affect the change. The user
must understand the consequences of scan protocol selection, not only in terms
of image quality, but also in terms of applied dose. This is especially important for

59

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Table 6.2. Overview of available scanning protocols, applications, and typical protocol
names. Protocols that are only a single click away from each other have vastly different
dose consequences. In addition to patient positioning and selection of the scanning arc, appro-
priate protocol selection is the most significant user-determined factor for radiation dose
calculation.
Protocol
Protocol spatial No. of
dose resolution projections Regions Clinical indication Names (examples)

Low Low Low Abdomen, Rule out kidney stone, ‘’, ‘low-quality’,
thorax assess position of ‘low-dose’
instrument/implants,
treatment planning
Medium High Low/ Skull/bones Dental and maxillofacial ‘Dental’, ‘bone’,
medium imaging, assess bone ‘high-resolution’
structures, arterial
contrast media
angiography
High High High Abdomen, head Assess soft tissue struc- ‘+’, ‘CT angiog-
tures, intracranial raphy’, ‘high-
haemorrhage, venous quality’
contrast media
angiography

CT, computed tomography.

CBCT, where such information may be entirely (and sometimes ambiguously)

encoded in the protocol name. There has been considerable variability in terms
used in imaging, and this creates difficulty in dose registry. The Commission rec-
ommends standardisation of terminology used in imaging protocols.

Scan modes: partial panel

(113) In order to expedite readout of the panel, the detector control electronics
generally allow readout of partial panels; an arbitrary number of central rows may be
readout as needed. While most systems have built-in hardware features that ensure
effective use of the beam, it is essential, from a radiological protection point of view,
that the x-ray beam is collimated appropriately to irradiate only that portion of the
detector that is being readout.

Keep unnecessary body parts out of the x-ray beam

(114) It is good practice to limit the radiation field to the body parts that must be
imaged. Inclusion of unnecessary body parts not only has dose consequences, but
may also increase image artefacts significantly. Many CBCT scanners only have a
limited scan FOV, with a diameter lower than the body region that is being

60

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

examined. Positioning of arms or legs outside the irradiated area can reduce the level
of artefacts significantly, and therefore increase the image quality without increasing
unnecessary radiation dose.

Making judicious use of CBCT acquisitions during a procedure

(115) CBCT imaging can quickly provide 3D images intra-operatively with
minimal effort on the part of the interventionalist or surgeon. These datasets
are useful as they relieve the operators from the effort of trying to distinguish
overlapping structures in 2D fluoroscopy images. They can also save dose by
replacing multiple digitally subtracted angiography runs in different C-arm
angulations with a single CBCT run. It has been shown that 3D acquisition
provides valuable clinical information and limits the need for 2D imaging:
hence, CBCT can also lower the dose in one procedure. Given this facility,
and the ease with which 3D images can be acquired, operators may be tempted
to overuse the 3D imaging features of their equipment. Although CBCT has
the potential to decrease dose in comparison with fluoroscopy and MDCT, this
effect could be cancelled by overuse of volumetric acquisition with C-arm and
other intra-operative CBCT machines. 3D data must be acquired judiciously for
purposeful clinical problem solving when fluoroscopy is insufficient for the task
at hand.

Bismuth shielding
(116) Bismuth shielding for the eyes, thyroid, breast, or other organs in CBCT
should be used with caution. With CBCT, reduced arc scanning will be more effective
(Section 6.2.2), and such shielding must not be used in conjunction with this.
Bismuth shielding can be effective in certain situations if placed in a manner that
does not interfere with AEC of the CBCT scanner. If the shield is positioned after
AEC has adjusted the tube current to be used, this may be beneficial provided that
the image is not degraded excessively by the presence of the shield in the FOV
(AAPM, 2012a). If the bismuth shield is placed before selection of AEC, its effect
may be totally negated by the increased current from AEC. The issues surrounding
the use of bismuth shielding are similar in both MDCT and CBCT, and available
guidelines for MDCT apply to CBCT.

Reconstruction algorithms
(117) In a standard CBCT reconstruction algorithm such as the modified
Feldkamp–Davis–Kress (FDK) algorithm, the noise level is proportional to the
applied tube current. However, image filtering, compressed sensing, and iterative
reconstruction algorithms, which are becoming increasingly popular in MDCT,
have the potential to disrupt this direct relationship between the applied dose and
image quality. At the present time, such novel reconstruction algorithms are not

61

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

widely available for CBCT scanners, and it is not possible to provide specific guide-
lines on how they should be used in practice. In many circumstances, the application
of these specialised algorithms is not universal. Instead, a user-selectable mixing
parameter is provided. This percentage factor determines the level to which the
output of the specialised reconstruction algorithm should be incorporated and
added to the output of the traditional algorithm. The exact setting for this mixing
factor will depend on the algorithm and the acceptable image quality, and will have
to evolve with experience.

6.2.3. Patient-specific factors

Thickness of the body part in the beam

(118) In response to the varying thickness of the anatomy, many CBCT machines
adjust radiation exposure automatically through AEC. This electronic system has a
sensor that detects how much signal is being produced at the image receptor, and
adjusts the x-ray generator to increase or decrease exposure factors (typically tube
current and, in many cases, tube voltage) so that each projection image is of a
consistent quality. When a thicker body part is in the beam, or a thicker patient is
being imaged (compared with a thinner patient), the machine will increase the expos-
ure automatically. The result is a similar image quality but an increase in the
entrance dose.
(119) In MDCT, AEC is able to vary the tube current both in the angular as well as
the longitudinal or z-direction. As a result of the angular variation, the dose in the AP
direction is lower than that in the lateral direction for any fixed, user-selected image
quality parameters. The z-axis adaptation of the dose controls the mA value in the
superior–inferior direction, resulting in a higher dose to the abdomen and pelvis com-
pared with the chest. In CBCT, as most acquisitions are performed in an axial mode
rather than a helical mode, the angular variation of tube current is more important.
(120) Some CBCT systems lack AEC. These systems operate under the assumption
that patient size does not vary significantly in the angular direction, although
absorbed doses can vary considerably during a CBCT rotation (He et al., 2010).
The assumption can be true for dental and maxillofacial, or head and neck applica-
tions, but should be investigated further. When AEC is not available, guidance
should be provided for the user on how to adjust the exposure parameters for dif-
ferent patient sizes.

Use of CBCT in children

(121) For any given exposure settings (same tube exposure parameter settings,
collimation, amount of projections, etc.), a thinner patient will receive a higher
dose (which is energy deposited per mass) than a larger patient, although the
larger patient will absorb a greater fraction of the radiation (AAPM, 2011b). This
is because the lower attenuation in a thinner body results in a smaller range in dose

62

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

through the body tissues for the smaller patient (e.g. a child). This may also some-
times be true even when the exposure factors are adjusted for body size or are
controlled by AEC. In general, especially for large patients, a greater fraction of
the x-ray beam is absorbed in the more superficial portions of the anatomy being
imaged. In other words, the skin dose is much higher than the central dose. For
thinner patients, this dose gradient is smaller, which implies that the dose is high
throughout the entire body. Figs. 6.3 and 6.4 illustrate the absorbed radiation dose
as a function of the patient’s body habitus and size when AEC compensates for
variations in body size. Thus, it is important to pay particular attention to optimising
radiological protection for children to ensure that exposure factors are not higher
than necessary.

Monitoring of patient dose indices

(122) Unfortunately, the field of patient dose monitoring in CBCT lags behind
that in MDCT. There is a lack of standardisation in dosimetry methods for CBCT;
different manufacturers have provided different ways of measuring and reporting
dose in CBCT, and these are not adopted universally. It is hoped that if the recom-
mendations of Report 87 (ICRU, 2012) are adopted by manufacturers and clinicians,
there is a good possibility that dosimetry in CBCT will be standardised and this will
provide more coherent patient dose data in the future. Means to estimate and report
patient dose will require a collaborative effort between the manufacturers of CBCT
equipment and the regulatory bodies. Methods for storing patient dose indices and
dose reports in Picture Archiving and Communication Systems (PACS) also have to
evolve as the use of CBCT becomes more prevalent.

Fig. 6.3. Qualitative illustration of the effects of automatic exposure control (AEC) on
patient exposure. AEC keeps the image quality at a given level and adjusts for variations in
patient size. The impact of patient size on the radiation dose with AEC is shown. A shows
the smallest patient diameter, C shows the largest patient diameter, and B is in between A
and C. Radiation exposure is indicated by the darkness of the grey of the radiation fan.
The bigger the patient, the higher the applied radiation exposure.

63

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Fig. 6.4. Effect of the variation in patient diameter at different angular positions in-plane
on tube output controlled by automatic exposure control (AEC). At angles where the
patient diameter is greater, the exposure is increased by AEC. The plot of tube output is an
example derived from an actual torso scan (provided by Rolf Kueres).

(123) In view of recent cases of skin injuries to patients in CT examinations, there

is a need to provide checks and balances to avoid overexposures through alerts, and
control patient dose prospectively in comparison to locally defined reference values
(Cadet, 2010; IAEA, 2010; AAPM, 2011c; NEMA, 2013). Manufacturers need to
incorporate suitable features to facilitate this.

6.2.4. Factors influencing dose to worker

(124) Occupational radiation exposure is expected to be small in the case of clinic-

based CBCT systems. When using a C-arm or other CBCT device in an interven-
tional suite or operating theatre, physicians, technologists, and other workers can
protect themselves by using shielding devices. As required under national regulations
in most countries, radiation workers must comply with regular individual dose moni-
toring requirements for managing radiation exposure, and keep a comprehensive
dose record. Further, unless necessary, workers should move outside the fluoroscopy
room radiation area when CBCT acquisition is taking place.
(125) In one study, the unshielded CBCT exposure at 35-cm distance from the
operating table, measured over a 60-s scan, was found to be 0.26 mSv (Daly et al.,
2006). Schulz et al. (2012) measured dose to the eye ranging from 28.0 to 79.3 mSv for
CBCT hepatic arterial embolisation and biliary tube placement procedures. The
primary source of radiation is the x-ray tube, and, ideally, the patient alone

64

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

should be exposed to the primary x-ray beam. Radiation scattered from the patient,
parts of the equipment, and the patient table (so-called ‘secondary radiation’ or
‘scatter radiation’) is the main source of radiation exposure to workers. A useful
rule of thumb is that radiation dose rates are higher on the side of the patient closest
to the x-ray tube. Distance is also an important factor, and workers should increase
their distance from the x-ray source and the patient when permitted in the clinical
situation. Automatic injectors should be used, as far as possible, if contrast medium
injection is necessary.

Shielding: lead apron

(126) Clinical staff taking part in diagnostic and interventional procedures using
C-arm devices for fluoroscopy or CBCT imaging wear protective aprons containing
lead (sometimes also lined with additional x-ray-absorbent materials) to shield
tissues and organs from scattered x rays (NCRP, 1995). Transmission through
these aprons will depend on the energies of the x rays and the lead-equivalent
thickness of the aprons. If the attenuation of scattered radiation is assumed to
be equal to that of the primary (incident) beam, this provides a margin of safety
(NCRP, 2005).
(127) All workers present in the room during a CBCT scan must wear a lead
apron, as it is the most essential component of personal shielding in an x-ray
room. It should be noted that the level of protection afforded by the lead apron
depends on the x-ray energy, which is a function of the voltage applied across the
x-ray tube (kV). The thicker the part of the patient’s body falling in the x-ray beam,
the higher the kV set by the fluoroscopic system. Higher kV x-ray photons have
greater penetrative power, implying that a greater lead thickness is needed to provide
the necessary attenuation.
(128) For procedures performed on thinner patients, particularly children, an
apron of 0.25-mm lead equivalence will suffice. However, for thicker patients and
with a heavy workload, a 0.35-mm lead apron may be more suitable. The wrap-
around aprons of 0.25-mm lead equivalence are ideal; these have a thickness of
0.25 mm at the back and 0.5 mm at the front. Two-piece skirt-type aprons help to
distribute the weight, and due to their overlap in front of the abdomen, they provide
a 1-mm shielding (e.g. at the level of the uterus). Heavy aprons can pose a problem
for workers who have to wear them for long periods of time. There are reports of
back injuries due to the weight of lead aprons among workers who wear them for
many years (NCRP, 2010). Some newer aprons are lightweight while maintaining
lead equivalence, and have been designed to distribute the weight through straps and
shoulder flaps.

Ceiling-suspended shielding
(129) Ceiling-suspended screens that contain lead impregnated in plastic or glass
are very common in interventional radiology and cardiology suites. However, they

65

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

are not usually used in operating theatres. Shielding screens are very effective as they
have lead equivalences of 0.5 mm or more, and can reduce x-ray intensity by more
than 90%. Practical problems make the use of radiation shielding screens for occu-
pational protection more difficult but not impossible in operating theatres.
Manufacturers should develop shielding screens that can be used for occupational
protection without hindering the clinical task. There is a need for more than one
screen to provide protection effectively to other workers in the operating theatre in
addition to the main operator.

Mounted shielding
(130) Mounted shields include table-mounted lead rubber flaps or lead glass
screens mounted on mobile pedestals. Lead rubber flaps are very common in most
interventional radiology and cardiology suites, but are rarely seen in operating the-
atres; nevertheless, their use should be promoted. Manufacturers are encouraged to
develop detachable shielding flaps to suit practices in operating theatres. Lead rubber
flaps, normally impregnated with the equivalent of 0.5-mm lead, should be used as
they provide effective attenuation. In interventional fluoroscopy, table-mounted lead
rubber drapes protect the legs of the operator; however, for CBCT scans, these may
need to be repositioned so that they do not impede movement of the x-ray tube/
detector C-arm mount.

Room shielding
(131) Room shielding requirements for CBCT systems used in dental and max-
illofacial imaging range from 0.5- to 1.5-mm lead equivalent, depending on the scan-
ner’s specifications for scattered radiation dose and its workload (EC, 2012a). In
most cases, the image receptor intercepts the entire primary beam, as in most fluoro-
scopic units and MDCT scanners. The room shielding is for scattered radiation, as is
the case with a conventional CT scanner (Sutton et al., 2012). However, for any type
of CBCT machine, the shielding should be designed to keep doses to workers and the
public as low as reasonably achievable, and, of course, below the existing dose limits
that apply in various settings.

Lead glasses
(132) Various types of leaded glass eyewear are available, although they are hea-
vier than common glass eyewear. These include eyeglasses that can be ordered with
corrective lenses for individuals who normally wear spectacles. There are also eye
shields that can be clipped on to the spectacles of workers, and full-face shields that
also function as splash guards. Leaded eyewear should either have side shields to
reduce the radiation coming from the sides, or be of a wraparound design with
angled lenses. The use of protective devices for the eyes as well as the body is
recommended.

66

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Individual protection and monitoring

(133) The principles of radiological protection of workers from ionising radiation
are discussed in Publication 75 (ICRP, 1997) and reiterated in Paragraph 113 of
Publication 105 (ICRP, 2007c). In this section, practical points pertaining to those
who need to be monitored and what protective actions should be taken are discussed.
(134) Individual monitoring of workers exposed to ionising radiation using film
dosimeters, thermoluminescent dosimeters, optically stimulated luminescence
badges, or other appropriate devices is used to verify the effectiveness of radiological
protection practices in the workplace. The advice of a radiological protection expert/
medical physicist should be sought to determine which method is most appropriate.
An individual monitoring programme for external radiation exposure is intended to
provide information about the optimisation of protection, and to demonstrate that
the worker’s exposure has not exceeded any dose limit or the level anticipated for the
given activities (IAEA, 1999). As an effective component of a programme to main-
tain exposures as low as reasonably achievable, it is also used to detect changes in the
workplace and identify working practices that minimise dose (NCRP, 2000; IAEA,
2004). In 1990, the Commission recommended a dose limit for workers of 20 mSv
year1 (averaged over a defined 5-year period; 100 mSv in 5 years) and other limits as
given in Table 3.1; these limits were retained in the 2007 Recommendations (ICRP,
1991, 2007b). However, all reasonable efforts to reduce doses to the lowest possible
levels should be used.
(135) The Commission recommended that interventional radiology departments
should develop a policy that staff should wear two dosimeters (ICRP, 2000b). A
single dosimeter worn under the lead apron will yield a reasonable estimate of effect-
ive dose for most instances. Wearing an additional dosimeter at collar level above the
lead apron will provide an indication of the thyroid dose (if unprotected) and other
parts such as the head and lens of the eye. In view of increasing reports of radiation-
induced cataracts in those involved in interventional procedures, monitoring the dose
to the lens of the eye is important (Ciraj-Bjelac et al., 2010; Vañó et al., 2010).
Recently, dosimetry for the lens of the eye has become an active research area.
Many studies have been performed to determine which personal dose equivalent
quantity is appropriate, and how it can be used for monitoring the dose to the
lens of the eye, and to develop dosimeters to measure dose to the lens of the eye
(Domienik et al., 2011). The Commission recommends that methods which provide
reliable estimates of dose to the eye under practical situations should be established
and used.
(136) A risk-based approach to occupational radiation monitoring should be
adopted to avoid unnecessary monitoring of all workers. There is a need to raise
awareness of the requirement to use a dosimeter at all times, as there are many
examples of infrequent use in practice.
(137) The lack of use or irregular use of personal dosimeters is still one of the main
problems in many hospitals (Miller et al., 2010; Padovani et al., 2011). The protec-
tion service should provide specialist advice and arrange any necessary monitoring

67

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

provisions (ICRP, 2007b). In cases where individual monitoring is inappropriate,

inadequate, or not feasible, occupational exposure of workers should be assessed on
the basis of the results of monitoring the workplace, and information about the
locations and durations of exposure of workers (IAEA, 1996). In addition to indi-
vidual monitoring, it is recommended that indirect methods using passive or elec-
tronic dosimeters (e.g. dosimeters attached to the C-arm device) should be used in
these installations to enable the estimation of occupational doses to professionals
who do not use their personal dosimeters regularly. Active dosimeters are an asset in
the education and practice of radiological protection.

6.3. Limitations of CBCT

6.3.1. Detector dynamic range and reduced contrast resolution

(138) Compared with the detector system used in MDCT scanners, FPDs have a
lower dynamic range and lower DQE. For example, the contrast resolution of FPD-
based CBCT is approximately 10 Hounsfield units (HU), which is lower than the
1–3 HU available on MDCT. Therefore, applications that require imaging of low-
contrast structures (e.g. grey–white matter differentiation in a head CT) will perform
poorly on a CBCT scanner compared with an MDCT scanner.

6.3.2. Scatter

(139) The large FOV of these scanners implies that the entire volume generates the
scatter radiation. As an anti-scatter grid, which would further decrease the efficiency of
the imaging chain, is not used typically, scatter can degrade image quality significantly.

6.3.3. Temporal resolution

(140) FPDs usually employ CsI as the scintillator. CsI is a slow scintillator and
suffers from afterglow (i.e. a ghost of the old image is seen in the new image at fast
frame rates). As a result, sufficient time must be allowed to elapse after each projec-
tion before the next projection is recorded.

6.3.4. Artefacts

(141) CBCT images, in general, suffer from more or less the same types of artefact
that are seen in MDCT, but to different degrees. A summary of MDCT artefacts has
been provided by Barrett and Keat (2004). Metal and windmill artefacts are generally
reduced in CBCT compared with MDCT, particularly for high-density metals

68

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

(Pauwels et al., 2013). Motion artefacts, on the other hand, are more prevalent in
CBCT.
(142) In MDCT, a smaller number of slices, typically four to 64 (although up to
320 slices in some scanners), is acquired in each rotation as the patient is translated
through the gantry. Therefore, any patient motion affects only those slices that were
being acquired during the motion. In CBCT, the entire dataset is constructed from
projections acquired in one rotation. Therefore, any motion, even for a very short
duration, affects the entire volumetric dataset. The rotation speed of CBCT com-
pared with MDCT is approximately 10–20 times slower: as such, CBCT is much
more sensitive to motion artefacts.

6.3.5. Hounsfield unit consistency

(143) The HU system is based on the linear attenuation coefficient of water. All CT
scanners present clinical images in this system for consistency across vendors and
scanner models. The daily calibration of MDCT scanners incorporates scanning of a
water cylinder for HU calibration and beam hardening correction. CBCT scanners
typically lack detailed radiometric calibration, and the generated HU values are
more variable than those from an MDCT scanner. In contrast to MDCT, truncation
of the body outlines and drawbacks of the reconstruction algorithm lead to cupping
artefacts. When scanning a homogeneous water phantom, the HU units are not
uniform over the entire cross-section, but decline towards the edges (Kyriakou
et al., 2011).

6.3.6. Geometric distortion

(144) Depending on the type of gantry used, a CBCT scanner is more prone to
geometric distortion than an MDCT scanner. For example, when a C-arm is used as
a CBCT scanner, the weight of the gantry may deform the unit, so that the isocentre
of the imaging chain is not as well defined. This will degrade the image quality. In
addition, flexible alignment of many of the CBCT gantries necessitates a collision-
avoidance system that may increase the complexity of a scan.

6.4. Future developments

(145) Several technical developments in the field of CBCT are expected to enable
interesting new features that will affect image quality and imparted radiation. As
these features are only at an early stage of development, and mature implementations
are not available in the scanner systems in current use, only general guidance about
their efficacy and application can be given at this point in time.

69

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

6.4.1. Novel scan trajectories

(146) For tomographic reconstruction, projective data from a rotation of at least

180 plus the cone angle are necessary. This requirement imposes several constraints
on the design and operation of CBCT in practice. For example, C-arm systems need
to have large clearance in the operating room to complete the scan trajectory, and
lack of space may limit the use of certain scan modes of the C-arm CBCT in practice.
Novel scanning trajectories, such as eccentric rotation and/or parallel shifting of the
imaging chain, may relieve some of these constraints and be useful in extending the
scan FOV. These newer, non-traditional scan trajectories lead to a much more com-
plex distribution of the applied dose in the examined volume. Currently, only one
commercial robot CBCT system uses these alternative trajectories. However, the
dose estimation systems are not designed to handle such systems. In the future,
radiological protection measurements will have to account for these non-traditional
trajectories, and factor in the associated non-uniform dose deposition.

6.4.2. Advanced methods for exposure control

(147) As mentioned above, AEC is a means to adapt the scan parameters to an

individual patient’s anatomy and its variations. Usually, AEC is provided by a feed-
back loop between the radiation measured at the detector side and the x-ray tube
exposure settings. In its simplest form, the tube current is varied so as to keep the
total radiation measured at the detector constant. This compensatory mechanism
can fail when the patient size increases beyond a certain point. After that point, for a
given kV setting, the x-ray tube may not be able to deliver any further increase in mA
without overheating or causing damage to the x-ray tube anode. Sometimes, in order
to accommodate such large variations in photon flux when current modulation alone
is not able to meet the demand, the x-ray tube voltage setting is also changed by AEC
in CBCT. In order to make this practice workable for CBCT, manufacturers use
experimentally measured correlation curves to map the signal measured at one tube
voltage to a corresponding reading at another tube voltage. This is clearly an engin-
eering approximation derived empirically; a practice rare in MDCT. In fact, a change
of the tube voltage in the middle of a single-energy scan may interfere with the
fidelity of HU calibration.
(148) If tube voltage is changed during a scan, inconsistencies in the measured CT
values with respect to the Hounsfield scale definition have to be taken into account
and corrected. AEC with tube current as well as voltage variations make actual
patient dose estimations from tube parameters and phantom experiments very com-
plex. The difficulty is compounded by the fact that the mapping between signals at
different kV settings is proprietary information. As this practice becomes more
prevalent, further research will be needed in this area of dose measurement practice
in order to account for this non-traditional use of AEC systems.

70

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

6.4.3. Novel reconstruction algorithms and compressed sensing

(149) Analytical reconstruction algorithms, such as the filtered back projec-

tion, have been the mainstay for MDCT. These algorithms provide a single pass
solution that is available on nearly all CT scanners. Although they are generally
fast and provide good image quality, they are prone to noise and artefacts. In
the past decade, a new class of iterative reconstruction algorithms has been
introduced for MDCT by various vendors. Instead of using an analytical
approach, these algorithms attempt to minimise the error between the projec-
tions and the reconstructed slices. Typically, one to 30 iterations are required
for the solution to converge. These algorithms generally provide better image
quality, and are more robust in minimising noise and artefacts. Their main
drawback, besides their complexity, is their slow computational speed. They
are generally associated with increased image resolution, decreased radiation
dose, and metal artefact reduction. They can also be used for region-of-interest
reconstruction.
(150) Currently, a non-iterative, modified FDK algorithm is the industry standard
for image reconstruction in CBCT. Similar to the reconstruction algorithms for the
MDCT systems, where the use of iterative reconstruction algorithms is now gaining
in popularity, a shift in CBCT reconstruction from a modified FDK to an iterative
technique is expected. These reconstruction methods have the ability to incorporate
prior knowledge in the form of radiation and scatter distribution, as well as know-
ledge of the anatomy. They also minimise the error between the projections and the
reconstructed image in a global sense. These features would be advantageous for
CBCT, as it is often performed in situations where repetitive scanning of the same
anatomical region is necessary (e.g. to observe the evolution of a contrast bolus
through the vasculature and the tissue). Another example of repetitive scanning
would be angiographic interventions to deploy interventional devices (e.g. aneurysm
coils) and confirm their position. Often, changes in the successive 3D volumes are
relatively minor. Iterative algorithms can accommodate these requirements more
readily, and thus minimise the number of projections required for 3D or four-dimen-
sional reconstruction.
(151) In order to reconstruct a volume of interest or a slice, a minimum number of
data points is needed, in a strict mathematical sense, for the reconstruction task. If
the dose per projection is fixed, this minimum number of projections determines the
overall patient dose. If certain assumptions can be made about the object, and the
requirement that projection images be equally spaced is relaxed, an image can be
reconstructed under conditions that contravene the Nyquist–Shannon limit (i.e. the
theoretical minimal sampling rate required for reconstruction). These methods,
which are generally called ‘compressed sensing’, can reduce the dose by reducing
the number of input projections required for reconstruction. Sparse angular sensing,
where projections are only acquired from certain angular directions, is one method
for reducing dose using compressed sensing.

71

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

(152) Both iterative reconstruction techniques and compressed sensing are in their
infancy in CBCT. However, these novel techniques are expected to have a major
impact on image quality and the associated radiation dose in CBCT in the future.
The user has to be aware that long established relationships between radiation dose
and image quality may undergo fundamental changes with the use of novel, iterative
reconstruction algorithms.

72

Downloaded from ani.sagepub.com by guest on July 22, 2015

 7. RADIATION DOSE MANAGEMENT IN SPECIFIC
APPLICATIONS OF CBCT

. The user of CBCT in interventions can influence the radiation dose imparted to the
patient significantly by judicious use of a ‘low-image-quality or low-dose’ scan
instead of a ‘high-image-quality or high-dose’ scan.
. In radiotherapy, justified use of CBCT has potential at different stages of therapy
such as: pretreatment verification of patient position and target volume localisation;
evaluation of non-rigid misalignments, such as flexion of the spine or anatomical
changes in soft tissue; and during or after treatment to verify that the patient position
has remained stable throughout the procedure. Low-dose CBCT protocols should be
used for pretreatment alignment of bony structures.
. Many machines were only capable of fluoroscopy initially, but can now also perform
CBCT. Due to the improved clinical information in CBCT and its ability to remove
overlying structures, the user may be tempted to over-use the CBCT mode. The
CBCT mode should be used judiciously.
. In orthopaedics, justified use of CBCT can help in assessing the position of fractures
and implants with respect to the bony anatomy, especially in situations where fluor-
oscopy alone is insufficient, and thus can help in patient dose management.
. In urology, low-dose CBCT protocols should be used when imaging high-contrast
structures, such as calcified kidney stones.
. Dental and maxillofacial CBCT scans should be justified, considering alternative
imaging modalities. Once justified, they should be optimised to obtain images with
minimal radiation dose without compromising the diagnostic information.

7.1. Introduction
(153) It is assumed that readers would have gone through Section 4 of this report
which deals with the principles of radiological protection. CBCT is used in a multi-
tude of clinical applications. To maximise the practical utility of this publication, this
section is organised according to different clinical application domains that use
CBCT rather than design considerations, as they tend to be very similar across
different applications. For example, a C-arm system used in interventional radiology
(neuro, non-vascular, vascular) differs only marginally, if at all, from that used in
orthopaedics or urology. However, application-specific radiation varies considerably
across these domains, primarily because of patient-related and use-related factors. At
the end of each section, practical tips on the use of CBCT are provided that are
germane to that application domain.
(154) This section also cites and summarises various published studies that provide
typical ranges of CBCT dose values for each clinical application domain. Absolute
dose values are provided and may be used by a practitioner as a reasonable starting
point.
(155) It should be stressed that disparate methods have been used in the literature
to measure and quantify dose. Many manufacturers provide concise dose values for

73

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

their machines under varying scanning conditions and protocols. Such data are often
required for the regulatory approval process. It is recommended that the user should
consult these documents and dose databases. However, documents that have been
submitted to regulatory agencies for licensing suffer from a lack of standardisation in
dose measurement techniques and units.
(156) The drawing of conclusions from published studies and vendor documents,
especially when absolute dose values are compared, should be done with care, keep-
ing in mind the limitations of such comparisons because of variations in the meas-
urement methodology. It is expected that future literature on CBCT will use dose
measurement guidelines similar to those discussed in Section 5 and described in more
detail in Annex A. Such standardised and consistent dose figures will enable direct
comparisons among different machines, protocols, and imaging practices. In parallel,
standardisation of digital imaging and communication in medicine (DICOM) dose
reporting for CBCT is needed in order to enable retrospective retrieval and review of
patient exposure from stored PACS images.

7.2. CBCT in radiotherapy

(157) The primary role of CBCT in radiation therapy is pretreatment verification
of patient position and target volume localisation. In the most common workflow
pattern, a patient lies on the treatment couch, is positioned approximately for treat-
ment using wall-mounted lasers, and precise positioning is based on CBCT imaging.
In addition to correcting the position of the patient, the images are examined for
non-rigid misalignments, such as flexion of the spine or anatomical changes in soft
tissue. CBCT imaging is also sometimes acquired during or after treatment to verify
that the patient position has remained stable throughout the procedure. CBCT can
also be used in treatment simulation, prior to the start of a course of treatment.
(158) Most radiation therapy centres use gantry-mounted kV CBCT, with an x-ray
tube as the source and amorphous silicon flat-panel imagers as detectors (Jaffray
et al., 1999). Typical energies are between 80 and 125 kVp, with typical absorbed
doses between 1 and 40 mGy. A less-common modality is MV CBCT, using the
treatment accelerator as an x-ray source and a portal imaging FPD (Pouliot et al.,
2005). MV CBCT generally uses energies of up to 6 MV, with typical absorbed doses
between 20 and 100 mGy. Compared with kV CBCT, the images produced with MV
CBCT generally have lower soft tissue contrast due to the lack of photoelectric
absorption at higher photon energies. However, these systems do have some advan-
tages, including better geometric alignment of imaging and treatment isocentres, and
better imaging for large patients or patients with metallic prostheses.
(159) The choice of imaging technique is based on the treatment site and therapy
goals. For cranial or head and neck targets, the treatment site is well accounted for
by alignment of bony anatomy. Therefore, a low-dose CBCT technique is appropri-
ate. Similarly, when the treatment target can be aligned using implanted fiducial
markers, a low-dose technique is warranted. In these cases, accurate positioning
with CBCT can be performed with absorbed doses within irradiated tissues of less

74

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

than 10 mGy. Accurate positioning in the pelvis and abdomen, however, may require
differentiation of soft tissue boundaries. In these cases, the number of photons used
for imaging should be increased and may require imaging at absorbed doses between
10 and 40 mGy.
(160) The overall absorbed doses to tissues of a patient within the field imaged
by CBCT are small compared with the prescribed treatment dose (Table 7.1).
However, the treatment dose is localised to the disease site, whereas the CBCT
imaging dose is spread across the entire imaging volume. When compared with
other pretreatment imaging modalities, CBCT can provide better setup accuracy
doses with dose equal to or lower than MV port films (Korreman et al., 2010), but
uses more dose than orthogonal planar kV x-ray imaging (Kry et al., 2005) or non-
ionising setup methods such as optical imaging or ultrasound. Furthermore, one
must keep in mind that the primary radiation fields produce Compton scattered x
rays that deposit dose in the area around the treatment site. The magnitude of the
scattered dose depends upon the distance from the treatment field, and ranges from
approximately 0.05% to 0.5% of the maximum dose at the target (dmax). The
radiation dose at dmax is defined as 100%, and decreases as penetration through
tissue increases, with the decrease primarily due to energy absorbed within the
tissue.

7.2.1. Accounting for imaging dose in radiotherapy

(161) When x-ray imaging is used in a radiotherapy setting, the patient receives
radiation from both imaging and therapy. CBCT imaging, especially when employed
daily, causes additional accumulated dose which should be considered in the context
of the patient’s treatment. For this reason, the use of daily CBCT imaging should be
evaluated for each patient for sparing sensitive organs that have low thresholds for
tissue reactions (deterministic effects), and for paediatric patients who have a higher
sensitivity to radiation.
(162) With first-generation linac-mounted kV CBCT systems, imaging doses can
account for 2% or more of the prescribed target dose (Amer et al., 2007; Ding et al.,
2008; Ding and Coffey, 2009). However, the current trend is towards dose reduction,
and second-generation systems have achieved significant dose savings in kV CBCT
(Ding and Munro, 2013). When the imaging dose constitutes a significant fraction of
the prescription dose (ICRU, 2010), it should be reflected in the patient’s prescription
dose. For example, the prescription dose can be adjusted to include the imaging dose.
A more advanced accounting procedure is to perform patient-specific CBCT dose
calculation in the radiotherapy treatment planning system (Alaei et al., 2010). If this
technology is available, the patient organ doses that combine the imaging dose and
the radiotherapy dose can be optimised in 3D to create a more precise estimate of the
patient’s total radiation burden.
(163) In summary, for most radiation oncology applications of CBCT, accurate delin-
eation and alignment of the treatment target and critical organs should be a practitioner’s

75

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Table 7.1. Doses in cone beam computed tomography (CBCT) procedures in radiotherapy.
Listed values are for a single CBCT acquisition and should be multiplied by the number of
CBCT scans performed to compute the total dose.
Procedure Reported value Method Reference

MV CBCT
Head and neck 50–150 mGy Absorbed dose at isocentre Pouliot et al., 2005
MV CBCT
Head and neck 60–73 mGy TLDs, film and ion chamber Gayou et al., 2007
Pelvis 99–121 mGy measurements in cylin
drical and anthropo-
morphic phantoms
kV CBCT
Head and neck 1–17 mGy CTDIw Song et al., 2008
Chest 11–18 mGy
Pelvis 24–54 mGy
kV CBCT
Head and neck 36.6 mGy CTDIw Cheng et al., 2011
Pelvis 29.4 mGy Effective dose, TLDs in
Head and neck 1.7 mSv female phantom,
Pelvis 8.2 mSv absorbed dose to the lens of
Head and neck 3.8 mGy (new protocol) the eye
59.4 mGy (old protocol)
kV CBCT
Head and neck 2.1–10.3 mSv Effective dose, TLDs in Kan et al., 2008
Chest 5.2–23.6 mSv female phantom
Pelvis 4.9–22.7 mSv Mean skin dose at
Head and neck 13–67 mGy irradiated site, TLDs
Chest 14–64 mGy in female phantom
Pelvis 12–54 mGy
kV CBCT
Head and neck 7  0.5 mGy (at simulator) Average absorbed dose, Stock et al., 2012
Pelvis 1  0.05 mGy (at linac) TLD measurements in
12  3 mGy (at linac) anthropomorphic phantom
36  12 mGy (at linac)
kV CBCT
Chest Spinal cord: 8–22 mGy Absorbed doses from Spezi et al., 2012
Left lung: 12–29 mGy Monte Carlo simulation
Right lung: 16–40 mGy
Heart: 17–30 mGy
Body: 12–31 mGy
kV CBCT
Head and neck Spinal cord: 1.3–1.7 mGy Absorbed doses from Spezi et al., 2012
Mandible: 4.5–8.3 mGy Monte Carlo simulation
Right parotid: 0.3–2.7 mGy
Left parotid: 0.5–2.7 mGy
Left eye: 0.1–1.8 mGy
Right eye: 0.1–1.8 mGy
Oral cavity: 1.7–3.8 mGy
(continued on next page)

76

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Table 7.1. (continued)

Procedure Reported value Method Reference

Body: 1.0–2.3 mGy

Brainstem: 0.3–1.5 mGy
Larynx: 2.6–2.8 mGy
kV CBCT
Pelvis Rectum: 11–21 mGy Absorbed doses from Spezi et al., 2012
Left femoral head: 20–47 mGy Monte Carlo simulation
Right femoral head:
25–62 mGy
Body: 11–33 mGy
kV CBCT
Thorax 0.9–21 mGy Absorbed dose, TLD meas- Shah et al., 2012
MVCT urements in thorax
Thorax 0.3–9 mGy phantom
(non-CBCT)
kV CBCT Absorbed dose, TLD meas- Shah et al., 2012
Pelvis 18–51 mGy urements in pelvis
MV CBCT phantom
Pelvis 0.9–8.0 mGy
kV CBCT Absorbed dose, IMRT Kouno et al., 2013
Pelvis 25–40 mGy phantom measurements
MV CBCT with radio-photolumi-
Pelvis 40–80 mGy nescent glass dosimeter
kV CBCT
Head 1–7 mGy
MV CBCT
Head 30–50 mGy
TomoTherapy
Pelvis 13 mGy
kV CBCT
Head and neck 19 mGy Primary absorbed doses at Kim et al., 2013
Chest 51 mGy the centre of custom-
Pelvis 167 mGy made phantom using a
glass dosimeter
kV CBCT
Pelvis 0.2–7 mGy Secondary absorbed doses Kim et al., 2013
Head and neck 0.03–0.7 mGy (20–50 cm from isocen-
tre) measured on
custom-made phantom
using a glass dosimeter
kV CBCT
Thorax
Standard low-dose 5.0  0.3 mSv Effective dose, radiochro- Alvarado et al., 2013
Reduced mAs 2.4  0.2 mSv mic film in female phan-
Reduced rotation 1.2  0.3 mSv tom and Monte Carlo
1.2  0.3 mSv simulation

CTDIw, weighted computed tomography dose index; IMRT, intensity-modulated radiation therapy;
TLD, thermoluminescent dosimeter). MVCT, megavoltage CT; TLD, thermoluminescent dosimeter;
IMRT, intensity-modulated radiation therapy; CTDIw, weighted computed tomography dose index.

77

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

primary concern. Radiation dose arising from CBCT must be weighed within the context
of therapy doses that are one to two orders of magnitude higher than the imaging doses.
Imaging technique should be chosen to match treatment goals, such as the use of low-
dose techniques for alignment of bony structures. In situations where the cumulative
CBCT dose adds up to be a non-negligible fraction, it may be reflected in the overall dose
schedule and subtracted from the therapeutic dose.
(164) Imaging technique should be chosen to match treatment goals, such as the
use of low-dose techniques for alignment of bony structures.

7.3. Neurointerventions
(165) Intraprocedural CT capability in a C-arm, a form of CBCT, has been found to
be useful in both diagnostic and therapeutic interventions. In C-arm CT, the same
imaging chain that is used for fluoroscopic as well as angiographic imaging is also
used for collecting the projection data needed for tomographic reconstruction.
(166) CBCT is used in neurointerventions to acquire 3D angiographic images to
assess potential intracranial haemorrhage, and during vertebral augmentation pro-
cedures (Psychogios et al., 2010). CBCT may also be used to guide complex, 3D
positioning of coils within an aneurysm (Levitt et al., 2011). Some systems also allow
over-laying of 3D images on fluoroscopic images (Racadio et al., 2007). It is even
possible to create a blood-volume map with data from CT perfusion using CBCT
(Fiorella et al., 2014).
(167) Manufacturers may provide high- and low-quality protocols for these applica-
tions. Low-quality scan protocols, which typically use fewer projections, are usually
sufficient for high-contrast structures such as contrast-enhanced vessels or bony anat-
omy. Furthermore, the position of intervention instruments can be assessed by low-dose
scans. A high-quality imaging protocol is recommended for soft tissue evaluation, such
as assessment of intracranial parenchymal or subarachnoid haemorrhage.
(168) The image quality of neurointerventional CT with respect to radiation dose
using phantoms was described by Fahrig et al. (2006).
(169) In many neurointerventional scans, the radiosensitive thyroid and lenses of
the eyes lie within the scan FOV. To minimise the dose to these organs, the user can
take advantage of a feature of CBCT that is only available in some MDCT scanners
as an add-on feature. CBCT projections acquired over an angular span of (180 + ’),
where ’ is the cone angle of the x-ray tube, are sufficient for image reconstruction.
Depending on the starting position of the (180 + ’) rotation arc, a significant reduc-
tion in the exposure of the eyes and thyroid can be realised with ‘tube under’ scan
arcs. Shielding of the thyroid (when not in the scan FOV) provides moderate dose
reduction (Daly et al., 2006).
(170) A neurointerventionalist can influence the radiation dose from CBCT sig-
nificantly by:
. Deciding whether or not a ‘high-dose’ soft tissue scan is needed. This would be
required to rule out intracranial haemorrhage or assess a soft tissue structure in a
diagnostic scan. For angiographic scans, for which contrast media have been

78

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Table 7.2. Doses in cone beam computed tomography procedures in neurointerventions.

Reported
Procedure value Method Reference

Head 2–37 mGy Absorbed organ Koyama et al., 2010

doses
1.2 mSv Effective dose, photo-
diodes in anthropo-
morphic phantom

Neurointerventions 40–48 mGy Modified CTDI Fahrig et al., 2006

(soft tissue/‘rule out (small-volume ion
haemorrhage’) chamber)

Neurointerventions 75 mGy Modified CTDI (250- Kyriakou et al., 2008a

(soft tissue/‘rule out mm-long ion
haemorrhage’) chamber)
Interventional head and
neck surgery
Soft tissue of head and 10 mGy Modified CTDI (using Daly et al., 2006
neck customised 16-cm
cylindrical head
phantom)
Neurointerventions 9 mGy Modified CTDI (250- Kyriakou et al., 2008a
(angiograms, interar- mm-long ion
terial contrast media chamber)
injections)
Spine
Thoracic bone visual- 1.8 mGy Modified CTDI using Schafer et al., 2011
isation CTDI (head/body)
Lumbar bone visual- 3.2 mGy and other (abdo-
isation men/thorax) phan-
Thoracic soft tissue 4.3 mGy toms, small-volume
visualisation ionisation chamber
Thoracolumbar spine
Small patient setting 3.2 mSv Effective dose from Lange et al., 2013
Large patient setting 8.1 mSv thoracolumbar
spine model, using
conversion factors
based on dose
length product
Neurointerventions
High-dose setting 32 mGy Mathematical model Sanchez et al., 2014
(brain dose) of an adult standard
anthropomorphic
phantom

CTDI, computed tomography dose index.

79

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

injected, a ‘low-dose’ scan that displays high-contrast structures is sufficient to

image vessels. A low-dose scan is also sufficient for defining the position of high-
contrast interventional materials, such as coils, clips, and OnyxTM. The choice of
low dose vs high dose may alter the applied dose considerably (Table 7.2).
. Using ‘tube under’ scans, meaning scans in which the x-ray tube is positioned
on the opposite side of the body from radiosensitive organs such as the thyroid
and the eyes for the majority of the time, whenever possible in practical situ-
ations. This decreases the dose to the radiosensitive organs without any appre-
ciable consequence for image quality or diagnostic power of the examination.

7.3.1. Dose to workers from CBCT in neuroradiology procedures

(171) Workers can reduce their radiation exposure drastically by maintaining suf-
ficient distance from the x-ray source, and should use shielding whenever possible.
For example, the in-room unshielded effective dose from a typical intra-interven-
tional CBCT scan (10 mGy to isocentre) is <0.005 mSv at 2 m from the isocentre
(Daly et al., 2006). Nottmeier et al. (2013) reported doses ranging between 0 and
1.8 mGy spin1 in badges located at different places around the O-arm under
investigation.
(172) Workers should leave the room whenever permitted by the status of the
patient during CBCT.

7.4. Vascular interventions

(173) Vascular interventions include a range of procedures, such as angioplasty in
peripheral artery disease, (fenestrated branched) endovascular aneurysm repair, vessel
occlusion for controlling acute bleeding, treatment of arterio-venous malformations,
and tumour embolisation [either bland (such as in uterine fibroid embolisation), with
chemotherapy (such as in chemoembolisation of many liver tumours), or with radio-
active particles (selective internal radiotherapy treatment)]. Other examples of such
interventions include placement of intravascular devices such as vena caval filters,
transjugular intrahepatic portosystemic shunt (TIPSS), and catheter-directed thromb-
olysis. CBCT may be used in these procedures to acquire tomographic images of the
vasculature for 3D roadmapping. CBCT is also helpful in verifying the spatial rela-
tionship of instruments and surrounding anatomy in situations where relative position
or orientation cannot be resolved sufficiently using projective imaging alone. CBCT is
being used increasingly for procedural planning (e.g. in transcatheter aortic valve
implantation) or image guidance and navigation [e.g. in atrial catheter ablation or
TIPSS (Adamus et al., 2009)]. Some of the newer machines also allow acquisition of
3D vascular roadmaps that can be overlaid on fluoroscopic images. Both intra-arterial
and intravenous contrast media injections are used. It can be expected that CBCT will
play a growing role in vascular interventions.

80

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Table 7.3. Doses in vascular cone beam computed tomography (CBCT) interventions.

Procedure Reported value Method Reference

Cardiac angiography Median: 2.4 Gy cm2 (range: KAP for sample of 756 paedi- Corredoira et al., 2015
0.35–42 Gy cm2) atric patients (age 0–19
years)
Fenestrated branched endo-
vascular aneurysm repair
Pre-operative 0.27 Gy Skin dose Dijkstra et al., 2011
Postoperative 0.55 Gy
Catheter ablation (CBCT part) 7.9  0.6 mSv Effective dose derived from Ejima et al., 2010
total KAP
Catheter ablation (CBCT part) 5.5  1.4 mSv* Effective dose from simulation Wielandts et al., 2010

81
6.6  1.8 mSvy
Liver (in hepatic arterial 8.2  1.4 mSv (male) and Effective dose from KAP of Tyan et al., 2013
embolisation therapy) 5.6  1.2 mSv (female) anthropomorphic male and
female phantoms
61 Gy cm2 (male) and KAP from 125 patients

Downloaded from ani.sagepub.com by guest on July 22, 2015

52 Gy cm2 (female)
11.5  2.3 mSv (male) and Effective dose corresponding
11.3  3.0 mSv (female) to patients’ KAP, using
conversion factors based on
anthropomorphic phantoms
Radiological protection in cone beam computed tomography (CBCT)

Hepatic arterial embolisation 75–175 mGy Skin entry dose Paul et al., 2013a,b
therapy 16–52 Gy cm2 KAP
Retrospective analysis of 126
procedures
(continued on next page)
 Table 7.3. (continued)

Procedure Reported value Method Reference

Abdominal CBCT scan 2–37 mGy Absorbed organ doses Koyama et al., 2010
4–5 mSv Effective dose, photodiodes in
anthropomorphic phantom
Abdominal CBCT 2.1–4.2 mSv Effective dose using ‘small’ Suzuki et al., 2011
anthropomorphic phantom
and Monte Carlo
simulations
Hepatic artery embolisation 238 mGy Skin entry dose, readout from Schulz et al., 2012
examination protocol
KAP, kerma-area product.
*
Using Publication 60 weighting factors (ICRP, 1991).
y
Using Publication 103 weighting factors (ICRP, 2007b).

82
ICRP Publication 129

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Table 7.4. Worker doses in vascular cone beam computed tomography (CBCT)
interventions.

Procedure Reported value to worker Method Reference

Abdominal CBCT Eye level: Digital dose Schulz et al.,

8 s per rotation: 28.0 mSv rate meter 2012
20 s per rotation: 79.3 mSv
5 s per two rotations: 32.5 mSv
Hepatic angiography Large field of view: 37.6 mSv
Eye level: 28–79 mSv per procedure

(174) The user of CBCT in vascular interventions can influence the radiation
dose imparted to the patient significantly by judicious use of protocols with ade-
quate image quality but lower dose if high-contrast objects are visualised (stents,
coils, guide wires, or high intravascular iodine contrast), or higher dose if low-
contrast objects are visualised (soft tissue or low parenchymal iodine contrast)
(Table 7.3).

7.4.1. Dose to workers in vascular interventions

(175) Paul et al. (2013b) found that the dose to the hands and the left knee of the
interventionalist was higher than the dose to the assistant physician when using
volume imaging. Mean doses received by the interventionalist ranged from
0.01 mGy (shielded thyroid, chest, and gonads) to 0.37 mGy (left finger). Mean
doses received by the assistant physician were 0.01 mGy to the shielded thyroid,
chest, and gonads, and 0.08 mGy to the left and right eyes. The mean dose to the
eye for the interventionalist was 0.11 mGy. Doses associated with the use of CBCT
were higher compared with catheter angiography and digitally subtracted angiog-
raphy. In guided needle interventions, operator hand doses in free-hand procedures
ranged from 20 to 603 mSv. Laser guidance alone or in combination with needle
holders resulted in a reduction of the hand dose to <36 mSv (5–82 mSv) per procedure
(Kroes et al., 2013). Additional worker doses for abdominal CBCT and hepatic
angiography can be found in Table 7.4.
(176) Workers should leave the room whenever permitted by the clinical
situation during a CBCT scan. For injecting contrast media, an automatic
injector should be used whenever possible. Workers who remain in the proced-
ure room during the CBCT exposure should be protected by fixed or mobile
shields.

7.5. Non-vascular interventions

(177) Non-vascular interventions include procedures such as vertebroplasty (treat-
ment of vertebral fractures, osteoporosis, or metastases), drainage of abscesses or

83

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

fluid collections, image-guided biopsies, percutaneous transhepatic cholangiography

drainage (PTCD), and tumour ablation (e.g. liver tumour microwave ablation)
(Wallace et al., 2008). These procedures are currently performed either under fluoro-
scopic guidance or MDCT guidance, with C-arm CBCT becoming increasingly
popular as it combines the advantages of both (Orth et al., 2008). Modern C-arm
systems allow the planning of percutaneous instrument insertion via preprocedural
CBCT, with fluoroscopy as the main modality for intraprocedural instrument guid-
ance. Repeated CBCT may be used for intraprocedural quality control; however, the
user should minimise the number of CBCT scans acquired during a given procedure.
(178) The user of CBCT in non-vascular interventions can influence the radiation
dose that is applied to the patient significantly by:
. appropriately choosing between a ‘high-dose’ scan and a ‘low-dose’ scan; and
. using the CBCT mode judiciously, relying on the fluoroscopy mode as much as
possible.
(179) Table 7.5 provides an overview of patient doses in non-vascular interven-
tions. Doses vary considerably depending on the diagnostic application and corres-
ponding exposure settings. Effective doses measured in phantoms were a few mSv for
each study. Various other dose quantities are also included. Reported CTDI values
were generally a few mGy, but some values >20 mGy have been measured. At the
skin and eye level, doses up to a few hundred mGy were found.

7.5.1. Dose to workers in non-vascular interventions

(180) In certain procedures, some dose to the interventionalist cannot be avoided. For
example, PTCD, or other biliary drainage procedures often require that one or both
hands/fingers are very close to the radiation field. For a short time, these procedures
may even require that these organs be in the radiation field, especially in punctures of
the left lobe of the liver. The practitioner should be cognisant of these small but potentially
repeated exposures. In a long procedure, the dose to the fingers may exceed a few mSv.
Protective gloves reduce the exposure of hands or fingers, but increase the dose to
the practitioner and patient if the hands with gloves are placed in the primary beam.
Auxiliary instrumentation for handling needles and probes in the radiation field should be
used whenever possible. Examples of doses to workers from interventional procedures are
given in Section 7.4.1; radiation doses in vascular and non-vascular interventions are
similar.

7.6. Orthopaedics/surgery
(181) In orthopaedics or trauma surgery, CBCT is used mainly to assess the position
of fractures and implants with respect to the bony anatomy, especially in situations
where fluoroscopy alone is insufficient to disambiguate the position of an implant
with respect to the bony anatomy (Table 7.6). For example, with fluoroscopy alone,
the critical relationship of a screw with respect to an articular surface may sometimes

84

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Table 7.5. Doses in non-vascular cone beam computed tomography (CBCT) interventions.

Procedure Reported value Method Reference

Lumbar spine (bone protocol) 3.7 mGy Modified CTDI* Schafer et al., 2011
Thoracic spine (bone protocol) 1.9 mGy
Lumbar spine, low resolution 6.0 mGy
(soft tissue protocol)
Lumbar spine, high resolution 12.5 mGy
(soft tissue protocol)
Thoracic spine (soft tissue 4.6 mGy
protocol)
CBCT-guided vertebroplasty 11.5 mGy (total procedure
of the thoracic spine dose)
CBCT-guided vertebroplasty 23 mGy
of the lumbar spine

85
Renal biopsy 44.0 Gy cm2 Mean KAP Braak et al., 2012
Biliary tube placement (PTCD) 413 mGy Skin entrance dose Schulz et al., 2012
Biliary protocol 4.2–8.4 mSv Effective dose, female Kim et al., 2011
anthropomorphic phantom
with MOSFET detectors
Phantom study Head: 1.2 mSv TLDs in anthropomorphic Bai et al., 2011

Downloaded from ani.sagepub.com by guest on July 22, 2015

Chest: 7.3 mSv phantom
Abdomen: 7.5 mSv
Head 4.4–5.4 mSv (absorbed dose to Effective dose, TLDs in Kwok et al., 2013
lens of the eye: 44–174 mGy) anthropomorphic phantom
Radiological protection in cone beam computed tomography (CBCT)

Abdominal 15–37 mSv

CTDI, computed tomography dose index; KAP, kerma-area product; TLD, thermoluminescent dosimeter; PTCD, percutaneous transhepatic cholangiog-
raphy drainage.
*
Using CTDI (head/body) and oblate (abdomen/thorax) phantoms, measuring at the centre and at four peripheral points with a small-volume ionisation
chamber.
 ICRP Publication 129

Table 7.6. Doses in orthopaedics/surgery cone beam computed tomography (CBCT) inter-
ventions. Certain values in Table 7.5 may also be applicable.

Procedure Reported value Method Reference

Extremity scan 6.4–15 mGy CTDI phantom, Zbijewski et al., 2011

small-volume ion
chamber in
isocentre
CBCT wrist 1.7–2.2 mGy CTDI Ramdhian-Wihlm
arthrography et al., 2012
Evaluation of finger 0.8 mSv Absorbed dose, Faccioli et al., 2010
fractures derived from prior
study using TLDs
in anthropo-
morphic phantom
Volumetric scan of 133 mGy cm Dose length product, Reichardt et al., 2008
wrist joint and the 0.11 mSv effective dose
distal radius derived from dose
length product
CTDI, computed tomography dose index; TLD, thermoluminescent dosimeter.

remain unclear. CBCT may help to clarify this relationship. CBCT is also very helpful in
spine surgery where interventions are being performed in close proximity to critical
structures such as spinal nerves. CBCT datasets are also used to confirm the position
of implants interprocedurally or to acquire datasets for intra-operative navigation.
Dedicated extremity CBCT systems are based on the same principle as other CBCT
systems used in interventional radiology or elsewhere, with the C-arm being the most
popular platform. Another system (O-arm) is becoming increasingly popular for extrem-
ity and spinal fixation procedures. An O-arm system combines the advantages of a CT-
gantry-based design with the flexibility of a C-arm-based design. It is essentially a C-arm
system with a telescopic gantry that extends to complete the ring and become an O-arm
for CT operation. As such, the gantry can function as a standard C-arm, or one can
complete the O-ring and turn the system into a CT-like gantry where the FPD and the x-
ray tube rotate freely. Usually, CBCT scanning is performed intra-operatively in a prone
or supine position. A standing position for imaging of a weight-bearing knee and a
sitting position with the upper or lower extremities extended (Zbijewski et al., 2011) have
been described (Tuominen et al., 2013).

7.7. Urology
(182) CBCT on a C-arm also enables cross-sectional imaging to be performed in a
urological operating room. Apart from standard pulsed fluoroscopy, 3D reconstruction

86

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

can be performed intra-operatively during urological procedures. Different operating

modes are available. A low-dose protocol may be appropriate when imaging high-con-
trast structures. For example, when imaging calcified stones or other calcifications
during percutaneous nephrolithotomy, a low-dose protocol should be employed because
kidney stones should be visible despite high noise in the images obtained. The same
reasoning holds true for CBCT imaging of retrograde flow of contrast in the urinary
tract and collecting system (Roy et al., 2012; Michel et al., 2014).
(183) The user should use low-dose protocols that are sufficient to detect kidney
stones, pelvic calcifications, metallic instrumentation, and contrast-media-filled effer-
ent urinary tract.

7.8. Ear/nose/throat and head diagnostics or surgery

(184) Similar to other applications in the head and neck area, applications of
CBCT in ENT take advantage of the fact that this region includes structures, such
as the paranasal sinuses, the temporal bone, and the skull base, that have high
intrinsic contrast, being composed primarily of bone, air, and soft tissue.
Therefore, relatively high noise in the images can be tolerated without compromising
the diagnostic utility of the CBCT scans. The high resolution of CBCT systems is
ideally suited for the small structures of the skull base and middle ear. In addition,
only a relatively small FOV is required to cover the necessary anatomy. In ENT
scans, the position of the scan arc is a significant factor that influences radiation
exposure of sensitive organs such as the lens of the eye and thyroid (Daly et al., 2006)
(Table 7.7). Other applications of CBCT in ENT are described in Hodez et al. (2011),
and Miracle and Mukherji (2009b). For most diagnostic ENT procedures such as
imaging of the temporal bone and paranasal sinuses, dedicated scanners with the
patient in a sitting position are used. Besides low dose and patient comfort,
high spatial resolution is another major advantage of these scanners. As a result,
these scanners are being used increasingly for surgical planning of temporal
bone interventions such as cochlear implantation. There has been a rapid adoption
of this technology in routine clinical practice; a trend that is likely to accelerate in
future.

7.9. Dental and maxillofacial

(185) CBCT has been used in dental and maxillofacial imaging for several years,
and its use is increasing. It is primarily used to acquire images of the teeth and
periodontium, their placement within the alveolus of the mandible and maxilla,
and their relationship with the adjacent nerves and other structures. The high spatial
resolution of CBCT is ideally suited for these high-contrast structures, and generally
provides excellent image quality in this field. The images are used for diagnostic
purposes, pre-operative planning, postoperative evaluation, and image guidance
during navigated surgery in this region. Pathological changes such as fractures,

87

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Table 7.7. Cone beam computed tomography doses in ear/nose/throat and head surgery.
Certain values in Table 7.5 may also be applicable.

Procedure Reported value Method Reference

‘Head scan mode’ – 10 mGy Modified CTDI (custom Daly et al., 2006
soft tissue mode 16-cm cylindrical head
Sinus imaging (bone 3 mGy phantom)
mode)
Endoscopic sinus Manarey and Anand,
surgery 2006
Continuous fluoro- 0.9 mGy Centre
scopy 1.9 mGy Maximum surface dose
1.5 mGy Centre
High-level fluoro- 3.4 mGy Maximum surface dose
scopy 4.1 mGy Centre
11 mGy Maximum surface dose,
Digital ciné CTDI head phantom,
ion chamber
CTDI, computed tomography dose index.

periapical lesions, or periodontal disease affect high-contrast structures and can

therefore be imaged precisely using CBCT. The FOV is usually large enough to
cover the maxillofacial region with one orbit around the patient. In addition, dedi-
cated small volumes (e.g. 4  4 cm) allow for high-resolution imaging of a small
region of interest, such as a single tooth root. Earlier scanners employed image
intensifiers, but in the current systems, FPDs are used almost exclusively. Most
systems are seated or standing scanners consisting of a small C-arm that rotates in
a horizontal plane along a vertical axis with the patient sitting upright. Applications
of dental and maxillofacial CBCT are described in De Vos et al. (2009).
(186) Due to the wide dose range found in dental and maxillofacial CBCT, and the
variety of diagnostic needs in dental radiology, proper application of this technique
among alternative 2D and 3D dental imaging modalities has been of great concern
since its introduction in dentistry in 1998. Due to its relatively low radiation dose and
high spatial resolution compared with MDCT, dental and maxillofacial CBCT is
considered as a suitable substitute for MDCT for applications requiring 3D imaging
of hard tissues. However, its application as a complement or substitute for 2D
imaging modalities (e.g. panoramic or cephalometric radiographic) increases the
population dose. In many cases, the 3D nature of CBCT results in superior diag-
nostic efficacy compared with 2D radiographs, but for certain applications, 2D
radiographs often suffice. Detailed evidence-based guidelines have been determined
during the SEDENTEXCT project (EC, 2012a). The guidelines encompass a variety
of topics, covering justification, referral criteria, optimisation, training, QA, and staff
protection aspects. Twenty ‘basic principles’ were defined based on a thorough

88

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Table 7.8. Effective dose ranges in dental and maxillofacial cone beam computed tomog-
raphy (CBCT).

CBCT type (volume coverage) Effective dose (mSv)*

Dento-alveolar 11–674 (median: 61)

Craniofacial 30–1073 (median: 87)
*
Full list of studies can be found in EC (2012a).
Source: EC (2012a).

literature review in combination with the experimental work performed in

SEDENTEXCT on radiation dose, diagnostic use, and other CBCT-related topics.
(187) Several of these basic principles relate to justification, as the excessive use of
CBCT in dentistry would increase the population dose. The use of CBCT in dentistry
can only be considered as justified, if a patient history and clinical information are
available, if it is expected to add new information, and if 2D radiographs do not (or
are not expected to) answer the diagnostic question. Repeated CBCT examinations
should be avoided unless each examination can be justified individually. In addition,
dental and maxillofacial CBCT should not be used in diagnostic soft tissue imaging.
(188) An important optimisation principle in dental and maxillofacial CBCT
relates to the choice of the appropriate volume size for each examination. In many
cases, the region of interest is known exactly before scanning; in other cases, the
required volume is revealed after acquisition of a frontal and lateral scout image. The
smallest available volume size should always be chosen, as this could reduce patient
dose considerably. The choice between high- and low-dose settings should be made
according to the optimisation principle, ensuring adequate image quality for diag-
nosis at the lowest achievable dose.
(189) As CBCT images often contain structures that are not part of the diagnostic
region of interest (although this should be limited as much as possible through FOV
reduction), the EC guidelines also state that the entire image should be examined and
reported, not just the region of interest. Depending on the scanning region, the
involvement of an oral or medical radiologist can be warranted.
(190) Table 7.8 provides an overview of the effective dose range in dental and
maxillofacial CBCT, measured using anthropomorphic phantoms (EC, 2012a). An
updated overview of effective doses measured in dental and maxillofacial CBCT can
be found in the systematic reviews by Bornstein et al. (2014) and Al-Okshi et al.
(2015). However, effective dose is not a good quantity to use for describing doses to
tissues in the head, as the tissue-weighting factors are only indicative of risk, and use
of equivalent doses to specific tissues or absorbed doses within them are generally
recommended. Although accuracy and intercomparability of several dosimetric stu-
dies are limited due to the varying measurement methodology (e.g. placement of
thermoluminescent dosimeter), Table 7.8 shows that patient doses vary considerably,
which reflects the wide variation of exposure parameters being applied. Volume sizes
range between a few cm3, sufficient for scanning of a single tooth area, and a few

89

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

thousand cm3, covering most of the head. In addition, there is no standardisation

regarding the kVp used in dental and maxillofacial CBCT, with values ranging
between 70 and 120 kV. Clinically applied mAs values range more than 20-fold
but are generally found between 25 and 150 mAs. The Commission recommends
standardisation of exposure parameters in dental and maxillofacial CBCT for each
imaging task.
(191) The application of dental and maxillofacial CBCT for paediatric patients is
of particular concern due to their higher radiosensitivity and smaller size. Similar to
its adult applications, paediatric use of CBCT could lead to considerable dose reduc-
tion when used as a replacement for MDCT (e.g. cleft palate), providing that FOV
limitation is applied and that exposure factors are optimised. However, its use as a
complement to, or replacement for, 2D radiography could lead to patient doses that
are disproportionate to the diagnostic benefit, especially when large-volume coverage
is required (e.g. orthodontic planning). For most paediatric applications, more evi-
dence regarding diagnostic efficacy of CBCT is needed before widespread application
can be considered. Table 7.9 contains effective dose measurements for 10-year-old
and adolescent anthropomorphic phantoms. Due to the larger relative coverage of
the child’s head, effective doses are higher compared with adults if exposure factors
are not adapted. For some CBCT models, preset ‘child dose’ exposure parameters
are available, typically corresponding to a reduction in mAs. For other models,
exposure factors can be modified by the operator. AEC is largely absent in dental
and maxillofacial CBCT, with one manufacturer having applied it for several years.
(192) Corresponding with the wide range in effective dose, absorbed doses of 0.03–
10.0 mGy have been reported for the thyroid gland, 0.02–9.3 mGy for the brain, and
0.03–16.7 mGy for the lens of the eye (Ludlow et al., 2006; Hirsch et al., 2008;
Ludlow and Ivanovic, 2008; Pauwels et al., 2012b). Various dose indices have also
been measured in dental and maxillofacial CBCT. A 2009 report by the UK Health
Protection Agency (HPA) measured KAP for 41 dental and maxillofacial CBCT
units, and normalised the results to a 4  4 cm field size with values ranging between
<100 and >2300 mGy cm2 (HPA, 2010a).
(193) Exposure of workers is reported to be in the range of 2 to 40 mGy scan1 at
1 m. For comparison, intra-oral and panoramic radiography scatter doses are less

Table 7.9. Overview of radiation doses in dental and maxillofacial cone beam computed
tomography (CBCT) for phantoms representing different ages.

Age CBCT type (volume coverage) Effective dose (mSv)

10-year-old phantom Dento-alveolar 16–214 (median: 43)

10-year-old phantom Craniofacial 114–282 (median: 186)
Adolescent phantom Dento-alveolar 18–70 (median: 32)
Adolescent phantom Craniofacial 81–216 (median: 135)
Source: Theodorakou et al. (2012).

90

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

than 1 mGy exposure1 at 1 m (EC, 2012a). The EC guidelines on dental and

maxillofacial CBCT state that ‘for worker protection from CBCT equipment, the
guidelines detailed in Section 6 of the EC publication ‘Radiation Protection 136.
European Guidelines on Radiation Protection in Dental Radiology’ should be
followed’.

7.10. Breast
(194) Mammography has been the standard imaging method for breast cancer
screening for several decades. While digital mammography has replaced screen-film
mammography in many locations, the projection-imaging nature of mammography
did not change with the introduction of digital mammography; digital mammog-
raphy still requires compression of the breast in order to acquire a 2D projection
image of the 3D breast. Digital mammography was proven to be slightly more
effective in detection of small lesions in women under 50 years of age with radio-
graphically dense breasts (Pisano et al., 2005). Digital mammography has also been
shown to reduce breast dose in comparison to screen-film radiography. In a 2010
study, mean glandular dose per view averaged 2.37 mGy for screen-film mammog-
raphy, while it was 22% lower (1.86 mGy per view) for digital mammography
(Hendrick et al., 2010). With digital mammography, contrast can be restored
(within limits) using digital enhancement techniques. Therefore, a harder x-ray spec-
trum can be used with digital mammography compared with screen-film mammog-
raphy, and this is the primary reason that some dose reduction is possible. The
harder x-ray spectrum is achieved through the use of different anode/filter combin-
ations (e.g. tungsten/rhodium instead of molybdenum/molybdenum) and higher
average tube potentials.
(195) 2D mammography suffers from the superposition of structures that may
falsely appear normal or abnormal, and this anatomical noise created by the
normal parenchyma of the breast confounds the cancer detection task. 3D
approaches relying on the principles of CT may improve breast cancer detection,
especially in the dense breast. Two approaches for ‘3D’ imaging of the breast have
been proposed: digital breast tomosynthesis and bCT. Breast tomosynthesis is
performed using multiple (e.g. 15–30) low-dose digital 2D projection images,
acquired on a modified full-field digital mammographic system that allows limited
angular movement of the x-ray tube around the breast during acquisition
(Niklason et al., 1997; Poplack et al., 2007). Tomosynthesis is the name given
to this acquisition strategy, which is formally considered to be limited-angle
tomography.
(196) Patient dose in one breast tomosynthesis acquisition, comprising 11 low-
dose projections over 28o angular movement, is approximately 4 mSv for a breast of
average thickness. This is approximately twice the dose used for digital mammog-
raphy (Poplack et al., 2007). More recently, doses from breast tomosynthesis were
estimated to be between 1.66 and 1.90 mGy for a standard breast, based on

91

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

manufacturers’ data in the absence of a standard protocol (Michell et al., 2012).

More recent tomosynthesis systems use a number of x-ray projections whose cumu-
lative dose to the breast is comparable to conventional single-view digital
mammography.
(197) bCT is currently undergoing evaluation before it can be introduced into
clinical practice. This technology has been developed to address the shortcomings
of conventional mammography such as contrast resolution, and the problems occur-
ring from overlap of structures in 2D images (O’Connell et al., 2010). Most bCT
systems make use of FPDs and are therefore CBCT systems; however, helical CT
systems for dedicated breast imaging are also being designed (Kalender et al., 2012).
(198) In the early days of bCT, there was no established method for estimating the
mean glandular dose to the breast in the pendant geometry used for this modality.
Therefore, methods for computing the dose to the breast needed to be developed.
Monte Carlo techniques were used to develop comprehensive tables of so-called
DgNCT values, which are appropriate for 360 scanning of the pendant breast
(Boone et al., 2004, 2005).
(199) Cone-beam-based bCT systems use FPDs that acquire 2D projections
which encircle the breast completely. Typically, a complete scan of a single
breast requires 10–17 s, within which time approximately 300–500 projections are
acquired (O’Connell et al., 2010; Packard et al., 2012). These systems are designed
to be low dose, and the mean glandular dose can be as low as that of two-view
mammography for each woman. Obviously, radiation dose depends on breast size
and composition. Therefore, smaller doses will occur in smaller breasts, and larger
breasts will receive higher doses. Reported mean glandular dose values range
between 4 and 12.8 mGy (O’Connell et al., 2010) and 2.5 and 10.3 mGy
(Lindfors et al., 2008). Average doses from conventional mammography docu-
mented in the abovementioned study by O’Connell et al. (2010) were in the
range of 2.2–15 mGy.
(200) Currently, bCT technology has some limitations regarding the detection of
microcalcifications, as well as coverage of the axillary region, both of which are
performed better with conventional mammography (Lindfors et al., 2008;
O’Connell et al., 2010). Higher-resolution detector systems will likely improve spatial
resolution of bCT, and consequently improve performance in detection of micro-
calcification (Kalender et al., 2012).
(201) Worker dose concerns for bCT are minimal as workers do not need to be
near the patient during image acquisition, as with most CT settings. Of course,
proper shielding of the bCT room is considered to be essential. One issue with
regards to shielding will emerge if bCT scanners become more commonplace in
the clinical imaging environment. These systems make use of higher energy x-ray
spectra than mammography systems; therefore, it is likely that additional room
shielding will be required if a bCT system is installed in a mammography room.
More details on room shielding are given in Section 6.2.4.

92

Downloaded from ani.sagepub.com by guest on July 22, 2015

 8. TRAINING CONSIDERATIONS FOR CBCT

. The level of training in radiological protection should be commensurate with the level
of expected radiation exposure (ICRP, 2009).
. All workers intending to use CBCT for diagnostic purposes should be trained in the
same manner as for diagnostic CT, and those intending to perform interventional
CBCT should be trained in the same manner as for interventional CT.

8.1. Introduction
(202) Publication 113 (ICRP, 2009) provides substantial information and guidance
on training of healthcare professionals in radiological protection for diagnostic and
interventional procedures. Much of the information provided in this section is
derived from that publication.
(203) ICRP states that a training programme in radiological protection for health-
care professionals has to be oriented towards the type of practice in which the target
audience is involved (ICRP, 2009, 2010).
(204) The main purpose of training is to make a qualitative change in practice
that helps operators use radiological protection principles, tools, and techniques
to reduce their own exposure without cutting down on work, and to reduce
patient exposure without compromising on image quality or intended clinical
purpose. The focus has to remain on achievement of skills. Unfortunately, in
many situations, training takes the form of complying with requirements mea-
sured in terms of the number of hours. While this provides an important yard-
stick, it is also essential to require trainees to learn skills to reduce occupational
and patient exposure. In large parts of the world, clinical professionals engaged
in the use of radiation outside imaging departments have either no training or
inadequate training. The Commission has recommended that the levels of edu-
cation and training should be commensurate with the level of radiation use and
expected radiation exposure (ICRP, 2009). As the use of CBCT outside imaging
departments increases, the need for education and training of workers also
increases. Professionals who are involved directly in the operation of CBCT
for diagnosis or intervention, and the interpretation of CBCT studies should
receive education and training in radiological protection at the start of their
career, and refreshment and professional development training should continue
throughout their professional life. Continuing education should include specific
training on relevant radiological protection tools and procedures as new equip-
ment or techniques are introduced.
(205) Legislation in most countries requires, or should require (if it does not cur-
rently do so), that individuals who take responsibility for medical exposures must be
trained properly in radiological protection.
(206) Training activities in radiological protection should be followed by an evalu-
ation of the knowledge acquired from the training programme (a formal examination
system).

93

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

(207) Workers who have completed training should be able to demonstrate that
they possess the knowledge specified by the curriculum by passing an appropriate
certifying examination.
(208) Nurses and other healthcare professionals who assist during CBCT proced-
ures should be familiar with radiation risks and radiological protection principles in
order to minimise their own exposure and that of others.
(209) Medical physicists should become familiar with the clinical aspects of the
specific procedures performed at their local facility.
(210) The issue of delivery of training and assessment of competency was dealt
with in Publication 113 (ICRP, 2009).
(211) For dental and maxillofacial applications of CBCT, dedicated basic training
requirements have been developed and published by the European Academy of
Dentomaxillofacial Radiology (Brown et al., 2014).

8.2. Curriculum
(212) It is anticipated that a large proportion of professionals involved in CBCT
will be those who have prior education in medical radiation physics and radiological
protection. Thus, simple orientation training may suffice in such cases. All workers
intending to use CBCT for diagnostic purposes should be trained in the same manner
as for diagnostic CT, and those intending to perform interventional CBCT should be
trained in the same manner as for interventional CT, keeping the level of dose and
use in view as specified earlier.
(213) It has been observed that most organisations follow the relatively easy route
of requiring a certain number of hours of education and training. The Commission
gives some recommendations on the number of hours required, but this should act as
a guideline and should not be applied rigidly (ICRP, 2009). Providing guidance in
terms of the number of hours has advantages in terms of implementation of training
and monitoring the training activity, but is only a guide.
(214) Many programmes fail with regard to assessment of whether the objectives
have been achieved. Others have pre- and post-training evaluations to assess the
knowledge gained, but few training programmes assess the acquisition of practical
skills. It would be more appropriate to encourage development of questionnaires and
examination systems that assess knowledge and skills, rather than prescribing the
number of hours of training. The extent of training depends upon the level of radi-
ation employed in the work, and the likelihood of overexposure of the patient or
workers.

8.3. Who should be the trainer?

(215) In view of the importance of this issue, most of the text from Publications 113
and/or 117 (ICRP, 2009, 2010) is reproduced here. The foremost point in any suc-
cessful training is that the trainer should have a clear perception about the practi-
calities of the work that the training has to cover. The primary trainer should

94

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

normally be an expert in radiological protection (normally a medical physicist), and

should have knowledge about clinical practice involving the use of radiation. That is,
the trainer should know about the nature of radiation, the way in which it is mea-
sured, how it interacts with the tissues, what type of effects it can lead to, principles
and philosophies of radiological protection, and international and national guide-
lines. As radiological protection is covered by legislation in almost all countries of
the world, awareness of national laws and the responsibilities of individuals and
organisations are essential (ICRP, 2009).
(216) Training should deal with what people can practice in their day-to-day work.
Instead, many trainers in radiological protection cannot resist the temptation to talk
about basic topics such as definition of radiation units, interaction of radiation with
matter, and even in-depth information on structure of the atom and atomic radiation
in more detail than is appropriate for the clinical audience and for the practical
purposes of radiological protection training. Such topics, while being essential in
basic educational programmes, should only be dealt with to a level such that they
make sense in the context of radiological protection training. A successful trainer
should not be too focused on definitions that are purely academic, but should be
guided by the utility of the information to the audience. The same applies to regu-
latory requirements. The trainer should speak the language of users to convey the
necessary information without compromising on the science and regulatory require-
ments. Healthcare professionals who use radiation in day-to-day work in hospitals
and deliver the radiation dose to patients know about the practical problems in
dealing with patients who may be very sick. They understand the time constraints
for dealing with large numbers of patients. They should understand the problems
associated with the radiation equipment they deal with, and the lack of radiation
measuring and radiological protection tools, when they happen. It is recommended
that training should also include lectures from practising clinicians and imaging
specialists, who can focus on good and bad radiological protection practices. It
may be useful for the radiological protection trainer to be available during such
lectures to comment on and discuss any issues raised.

8.4. Training of service engineers

(217) In some cases, service engineers are not familiar with CBCT technology. For
example, in dental and maxillofacial CBCT, some manufacturers have no prior
experience with CT equipment, and service engineers are not adequately geared
towards CBCT technology. Manufacturers are challenged to ensure adequate train-
ing of service engineers.

95

Downloaded from ani.sagepub.com by guest on July 22, 2015

Downloaded from ani.sagepub.com by guest on July 22, 2015
 9. QUALITY ASSURANCE PROGRAMMES
9.1. Introduction
(218) The purpose of a QA programme is to maintain the safety and performance
of the equipment in conformance with specifications, and to ensure the optimisation
of protection (i.e. achieving adequate image quality while minimising the radiation
dose to the patient and medical staff). In the context of this report, the QA pro-
gramme consists of the acceptance and commissioning of CBCT equipment, as well
as periodic testing and maintenance of equipment performance, patient imaging
protocols, worker and patient dose, worker training, and adherence to policies and
procedures.

9.2. Quality control of CBCT equipment

(219) Quality control begins when the equipment is installed, and continues
throughout its lifetime. The acceptance test, commissioning, and status testing of
equipment should ensure that the system is operational according to the manufac-
turer’s specifications, which are based on national or international standards. At the
time of acceptance, baseline measurements of image quality and dosimetry should be
taken along with parameters that affect these factors. These measurements will be
used as a reference for comparison with later measurements, and can indicate if the
system performance has degraded and needs corrective action.
(220) Equipment tests fall into six categories: safety system, x-ray generator per-
formance, image quality, geometry, display, and dosimetry. Safety system tests are
used to ensure the proper operation of warning lights, door and collision interlocks,
portable shielding, and the emergency-off system. X-ray generator tests can ensure
that the x-ray system operates properly, including the accurate production of kV,
mA, exposure time, and linearity. Image quality tests, such as those that measure
noise, uniformity, contrast, and resolution, can ensure that acquired images are
suitable for clinical use. The frequency of these quantitative tests should be estab-
lished to remediate image quality degradation (IEC, 2006). In addition to quantita-
tive testing, images should be inspected visually to identify image artefacts.
Geometry tests are used to ensure proper system alignment and scaling. In radio-
therapy applications, a daily test of the CBCT image isocentre geometry ensures that
images are aligned with the treatment machine. However, dental and maxillofacial
and interventional applications may not require alignment with an external coord-
inate system, and therefore only need test image scaling. Display testing will ensure
that image presentation is consistent and faithful to avoid loss of information during
interpretation. Finally, dosimetry tests are used to assess the dose to a phantom,
using standard measurement protocols appropriate for CBCT, such as those
described earlier in this publication and in Annex A. The equipment and methods
needed to perform other tests are described in publications such as Radiation
Protection Report No. 162 (EC, 2012b) and Report 91 (IPEM, 2005), although
the latter does not cover CBCT equipment specifically.

97

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

(221) Standardisation of phantoms and tests is warranted for image quality assess-
ment as well as dosimetry. In addition, manufacturers should provide users with
appropriate test objects for routine checking of performance.
(222) The schedule and scope of routine testing of CBCT equipment depend to
some degree on the clinical application. Inspection schedules recommended by six
different organisations (three for dental applications and three for radiotherapy
applications) are shown in Table 9.1. One should be aware of national recommen-
dations on this matter (e.g. DIN, 2013, 2014). The schedules are largely in agreement,
but some special considerations are worth noting. For CBCT equipment with an
exposed moving gantry that might collide with patients or workers, a daily safety
system check is recommended. If the CBCT image coordinates are used to control a

Table 9.1. Proposed quality assurance (QA) tests for cone beam computed tomography
equipment and corresponding periodicity as recommended by international, national, and
professional societies.
QA test Daily Monthly Periodic Annual

Safety systems: collision, 142, 179, IAC

warning lights, and
interlocks
Image quality: uniformity EC, 142, 179, HPA 179, IAC
Image quality: image IAC EC, 142, 179, HPA
density
Image quality: noise EC, 142, 179, HPA 179
Image quality: 142, 179 179 EC
low-contrast detail
Image quality: high- 142, 179 179, IAC EC, HPA
contrast resolution
Image quality: image IAC EC
artefacts
Geometry: isocentre 142, 147, ACR
coincidence
Geometry: scaling and slice 142, 179 179 EC, HPA
thickness
Data storage and transfer ACR
Image registration software ACR
Image display EC HPA IAC
X-ray quality, linearity, and EC, 179, HPA, IAC
field size
Dose measurements EC, 142, 179,
HPA, IAC

142, AAPM Report 142 (Klein et al., 2009); 179, AAPM (2012b); ACR, ACR (2009); HPA, HPA (2010b);
IAC, IAC (2012); EC, EC (2012a).

98

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

radiotherapy accelerator or surgical equipment, a daily check of coordinate system

integrity is recommended. If accurate density information (such as HU) is used for
diagnosis or planning, these values should be tested at least monthly. Users should
therefore consider these general guidelines to inform a risk-based QA programme
based on their clinical aims.

9.3. Patient dose reporting

(223) The need for dose reporting in CBCT follows from the principles of opti-
misation of radiological protection. Radiation dose to the patient cannot be
optimised to as low as reasonably achievable without accurate tracking of dose. The
most straightforward method for achieving dose tracking is through the electronic
display of dose on the imaging console (ICRP, 2004), and recording of delivered
dose into the patient record as a DICOM-structured dose report (IEC, 2012).
Errors for displayed dose estimates should not exceed 20% (IAEA, 2011b; IEC,
2011; EC, 2012b).
(224) In MDCT systems, it is now standard to display estimates of delivered dose
directly on the console numerically as CTDIvol and DLP. These estimates represent
the dose to a phantom, not the dose to a patient. Methods should be developed for
estimating doses to patients based on patient size and the scanning parameters used
for individual patients. A medical physicist, as part of the QA programme, should
verify the accuracy of these numbers at least annually, or whenever equipment is
repaired in a manner that can affect dose. For CBCT systems, the system for dose
reporting is not yet standardised at an international level. HPA (2010b) and EC
(2012a) recommend that the dose estimate should be displayed as KAP in dental
and maxillofacial CBCT systems. The QA programme should be prepared to verify
dose estimates as they are reported by each device, whether it be KAP or CTDI and
DLP.
(225) Electronic transfer of patient dose to an electronic medical record greatly
facilitates the tracking of annual and lifetime radiation dose to a patient over mul-
tiple procedures. MDCT systems implement this idea using the DICOM-structured
dose report, which usually expresses dose in terms of CTDIvol and DLP. Electronic
transmission of CTDIvol and DLP to PACS is now required by California State law
in the USA (California Senate Bill SB1237, 2010), and has been proposed by EC
(2011). Electronic reporting further supports initiatives to compare recorded doses
with DRLs; a concept recommended by ICRP for optimisation (ICRP, 2007c). Dose
registries are another potential tool for facilitating evaluation of patient dose.

9.4. Diagnostic reference levels

(226) DRLs have been established through government and professional organ-
isations to guide users in optimising procedure performance for both image quality
and radiation reduction. While these efforts have matured for MDCT imaging, little

99

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

progress has been made toward setting DRLs for CBCT. Based on a preliminary
audit of KAP values on 41 dental and maxillofacial CBCT units by HPA (2010b), an
achievable dose of 250 mGy cm2 (normalised to an area corresponding to 4  4 cm at
the isocentre) was proposed for placement of an upper first molar implant in a
standard adult patient. This achievable dose value was adapted by the
SEDENTEXCT Consortium (EC, 2012a), with the remark that ‘further work invol-
ving large scale audits is needed to establish robust DRLs’ for various dental and
maxillofacial CBCT applications. This remark can be extended towards other CBCT
applications.
(227) For centres that use standardised imaging protocols, the protocols should be
established within published DRLs. Until international or national DRLs are estab-
lished, local DRLs (LDRLs) should be established as part of the QA programme to
inform local policy for common procedures. LDRLs are established from mean
doses delivered to average-sized patients, with separate LDRLs established for chil-
dren (IPEM, 2004). Audits of standardised protocols should be performed period-
ically to ensure compliance. Currently, there is a dearth of data on DRLs.

9.5. Audit
(228) Periodic audits of patient imaging studies are recommended to ensure opti-
mal use of the imaging system. The audit should consider image quality, positioning,
FOV, patient motion, and radiation dose metrics. In particular, the audit should
evaluate high-dose CBCT procedures, and repeat CBCT scans. The SEDENTEXCT
Consortium report recommends two audits per year for reject analysis, and a patient
dose audit every 3 years (EC, 2012a).

100

Downloaded from ani.sagepub.com by guest on July 22, 2015

 10. RECOMMENDATIONS

(229) Expanded availability and newer applications have put CBCT technology in
the hands of medical professionals who do not traditionally use CT. ICRP’s radio-
logical protection principles and recommendations provided in earlier publications,
particularly Publications 87 and 102 (ICRP, 2000a, 2007a), apply to these newer
applications and should be adhered to.
(230) As many applications of CBCT involve patient doses similar to MDCT, the
room layout and shielding requirements in such cases need to be similar to protect
workers adequately.
(231) Medical practitioners bear the responsibility for making sure that each
CBCT examination is justified and appropriate.
(232) When referring a patient for a diagnostic CBCT examination, the referring
practitioner should be aware of the strengths and weaknesses for CBCT vis-à-vis
MDCT, magnetic resonance imaging, and other competing imaging modalities. The
decision to use CBCT should be made in consultation with an imaging professional.
(233) Manufacturers are challenged to practice standardised methods for dosim-
etry and dose display in CBCT in conformance with international recommendations
such as ICRU Report 87 (ICRU, 2012). Unfortunately, at present, there is wide
variation in dose quantities being displayed on CBCT machines. The users are
unable to compare doses between scanners or protocols.
(234) Use of CBCT systems for both fluoroscopy and tomography poses new
challenges in quantitating radiation dose. There is a need to develop methods that
aggregate exposures to individual patients during the entire procedure that may use a
combination of fluoroscopy and CBCT during a given examination.
(235) Recording, reporting, and tracking of radiation dose for a single patient
should be made possible in a consistent manner across vendors.
(236) There is a need to provide checks and balances (e.g. dose check alerts imple-
mented in CT in recent years) to avoid high patient doses compared with locally
defined reference values.
(237) Positioning radiosensitive organs such as the thyroid, lens of the eye, breasts,
and gonads on the detector side during the partial rotation scan is a useful feature in
CBCT that needs to be used for radiological protection of these organs.
(238) Many machines were only capable of fluoroscopy initially, but can now also
perform CBCT. Due to the improved clinical information on CBCT, and its ability
to remove overlying structures, a user may be tempted to over-use the CBCT mode.
Users must understand that the CBCT function of their system is not a low-dose
‘fluoroscopy run’, and should use this mode judiciously.

101

Downloaded from ani.sagepub.com by guest on July 22, 2015

Downloaded from ani.sagepub.com by guest on July 22, 2015
 REFERENCES
AAPM, 2010. Comprehensive Methodology for the Evaluation of Radiation Dose in X-ray
Computed Tomography. AAPM Report 111. American Association of Physicists in
Medicine, New York.
AAPM, 2011a. Scope of Practice of Clinical Medical Physics. Policy Number PP 17-B.
American Association of Physicists in Medicine, New York. Available at: http://
www.aapm.org/org/policies/details.asp?id=317&type=PP (last accessed 5 May
2015).
AAPM, 2011b. Size-specific Dose Estimates (SSDE) in Pediatric and Adult Body CT
Examinations. AAPM Report No. 204. American Association of Physicists in Medicine,
New York. Available at: http://www.aapm.org/pubs/reports/rpt_204.pdf (last accessed 5
May 2015).
AAPM, 2011c. AAPM Recommendations Regarding Notification and Alert Values for CT
Scanners: Guidelines for Use of the NEMA XR 25 CT Dose-check Standard. AAPM Dose
Check Guidelines Version 1.0. American Association of Physicists in Medicine, New York.
Available at: http://www.aapm.org/pubs/CTProtocols/documents/NotificationLevels
Statement.pdf (last accessed 5 May 2015).
AAPM, 2012a. AAPM Position Statement on the Use of Bismuth Shielding for the Purpose of
Dose Reduction in CT Scanning. Statement approved by AAPM Board of Directors,
February 2012, Policy date 2 July 2012. American Association of Physicists in Medicine,
New York. Available at: http://www.aapm.org/publicgeneral/BismuthShielding.pdf (last
accessed 5 May 2015).
AAPM, 2012b. Quality Assurance for Image-guided Radiation Therapy Utilizing CT-based
Technologies: a Report of the AAPM TG-179. American Association of Physicists in
Medicine, New York. Available at: http://www.aapm.org/pubs/reports/RPT_179.pdf
(last accessed 5 May 2015).
AAPM, 2015. Task Group No. 200 – CT Dosimetry Phantoms and the Implementation of
AAPM Report Number 111. American Association of Physicists in Medicine, New York.
Available at: http://aapm.org/org/structure/default.asp?committee_code=TG200 (last
accessed 5 May 2015).
Abuhaimed, A., Martin, C.J., Sankaralingam, M., Gentle, D.J., McJury, M., 2014. An assess-
ment of the efficiency of methods for measurement of the computed tomography dose
index (CTDI) for cone beam (CBCT) dosimetry by Monte Carlo simulation. Phys. Med.
Biol. 59, 6307–6326.
ACR, 2009. ACR Technical Standard for Medical Physics Performance Monitoring of Image-
Guided External Beam Radiation Therapy (IGRT). Reston, VA: American College of
Radiology. Available at: http://www.acr.org//media/ACR/Documents/PGTS/stand-
ards/IGRT.pdf (last accessed 5 May 2015).
Adamus, R., Pfister, M., Loose, R., 2009. Transjugular intrahepatic portosystemic shunt
(TIPS) puncture by 3D path planning based on the back-projection of two 2D portogra-
phies. Radiology 251, 543–547.
Al-Okshi, A., Lindh, C., Salé, H., Gunnarsson, M., Rohlin, M., 2015. Effective dose of cone
beam CT (CBCT) of the facial skeleton: a systematic review. Br. J. Radiol. 88, 20140658.
Alaei, P., Ding, G., Guan, H., 2010. Inclusion of the dose from kilovoltage cone beam CT in
the radiation therapy treatment plans. Med. Phys. 37, 244–248.
Alvarado, R., Booth, J.T., Bromley, R.M., et al., 2013. An investigation of image guidance
dose for breast radiotherapy. J. Appl. Clin. Med. Phys. 14, 4085.

103

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Amer, A., Marchant, T., Sykes, J., Czajka, J., Moore, C., 2007. Imaging doses from the Elekta
synergy x-ray cone beam CT system. Br. J. Radiol. 80, 476–482.
Arai, Y., Tammisalo, E., Iwai, K., et al., 1999. Development of a compact computed tomo-
graphic apparatus for dental use. Dentomaxillofac. Radiol. 28, 245–248.
Araki, K., Patil, S., Endo, A., et al., 2013. Dose indices in dental cone beam CT and correl-
ation with dose area product. Dentomaxillofac. Radiol. 42, 20120362.
Bai, M., Liu, B., Mu, H., et al., 2011. The comparison of radiation dose between C-arm flat-
detector CT (DynaCT) and multi-slice CT (MSCT): a phantom study. Eur. J. Radiol. 81,
3577–3580.
Balter, S., Hopewell, J.W., Miller, D.L., et al., 2010. Fluoroscopically guided interventional
procedures: a review of radiation effects on patients’ skin and hair. Radiology 254,
326–341.
Barrett, J.F., Keat, N., 2004. Artifacts in CT: recognition and avoidance. Radiographics 24,
1679–1691.
Berris, T., Perisinakis, K., Papadakis, A.E., et al., 2013. Comparison of methods for assessing
geometric efficiency on multi-detector CT scanners. Phys. Med. 29, 312–322.
Blaickner, M., Neuwirth, J., 2013. Measurements of occupational and patient exposure as well
as image quality for two C-arms. Radiat. Prot. Dosimetry 155, 451–458.
Bogdanich, W., 2009. Radiation overdoses point up dangers of CT scans. New York Times, 16
October 2009, p. A13.
Bogdanich, W., 2010. Afterstroke scans, patients face serious health risks. New York Times, 1
August 2010, p. A1.
Boone, J.M., Shah, N., Nelson, T.R., 2004. A comprehensive analysis of DgNCT coefficients
for pendant-geometry cone-beam breast computed tomography. Med. Phys. 31, 226–235.
Boone, J.M., Kwan, A.L.C., Seibert, J.A., et al., 2005. Technique factors and their relation-
ship to radiation dose in pendant geometry breast CT. Med. Phys. 32, 3767–3776.
Boone, J.M., 2009. Dose spread functions in computed tomography: a Monte Carlo study.
Med. Phys. 36, 4547–4554.
Bornstein, M.M., Scarfe, W.C., Vaughn, V.M., et al., 2014. Cone beam computed tomog-
raphy in implant dentistry: a systematic review focusing on guidelines, indications, and
radiation dose risks. Int. J. Oral Maxillofac. Implants 29(Suppl.), 55–77.
Braak, S.J., van Melick, H.H.E., Onaca, M.G., et al., 2012. 3D cone-beam CT guidance, a
novel technique in renal biopsy – results in 41 patients with suspected renal masses. Eur.
Radiol. 22, 2547–2552.
Brenner, D.J., 2010. Slowing the increase in the population dose resulting from CT scans.
Radiat. Res. 174, 809–815.
Brown, J., Jacobs, R., Levring Jäghagen, E., et al., 2014. Basic training requirements for the
use of dental CBCT by dentists: a position paper prepared by the European Academy of
DentoMaxilloFacial Radiology. Dentomaxillofac. Radiol. 43, 20130291.
Cadet, J.V., 2010. CT Makers get Five FDA Recommendations, due to Overdose Controversy.
Clinical Innovation+Technology. Providence, RI. Available at: http://www.clinical-inno-
vation.com/topics/ehr-emr/ct-makers-get-five-fda-recommendations-due-overdose-contro-
versy (last accessed 5 May 2015).
California Senate Bill SB1237, 2010. An act to add Sections 115111, 115112, and 115113 to the
Health and Safety Code, relating to public health. Available at: http://www.leginfo.ca.gov/
pub/09-10/bill/sen/sb_1201-1250/sb_1237_bill_20100929_chaptered.html (last accessed 5
May 2015).

104

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Cheng, H.C., Wu, V.W., Liu, E.S., et al., 2011. Evaluation of radiation dose and image quality
for the Varian cone beam computed tomography system. Int. J. Radiat. Oncol. Biol. Phys.
80, 291–300.
Chodick, G., Bekiroglu, N., Hauptmann, M., et al., 2008. Risk of cataract after exposure to
low doses of ionizing radiation: a 20-year prospective cohort study among US radiologic
technologists. Am. J. Epidemiol. 168, 620–631.
Ciraj-Bjelac, O., Rehani, M.M., Sim, K.H., et al., 2010. Risk for radiation-induced cataract
for staff in interventional cardiology: is there reason for concern? Catheter. Cardiovasc.
Interv. 76, 826–834.
Ciraj-Bjelac, O., Rehani, M., Minamoto, A., et al., 2012. Radiation-induced eye lens changes
and risk for cataract in interventional cardiology. Cardiology 123, 168–171.
Corredoira, E., Vañó, E., Ubeda, C., Gutiérrez-Larraya, F., 2015. Patient doses in paediatric
interventional cardiology: impact of 3D rotational angiography. J. Radiol. Prot. 35,
179–195.
Cucinotta, F.A., Manuel, F.K., Jones, J., et al., 2001. Space radiation and cataracts in astro-
nauts. Radiat. Res. 156, 460–466.
Daly, M.J., Siewerdsen, J.H., Moseley, D.J., et al., 2006. Intraoperative cone-beam CT for
guidance of head and neck surgery: assessment of dose and image quality using a C-arm
prototype. Med. Phys. 33, 3767–3780.
Day, R., Gorin, M.B., Eller, A.W., 1995. Prevalence of lens changes in Ukrainian children
residing around Chernobyl. Health Phys. 68, 632–642.
DIN, 2013. Image Quality Assurance in Diagnostic X-ray Departments – Part 161: RöV
Acceptance Testing of Dental Radiographic Equipment for Digital Cone-beam
Computed Tomography. Berlin: Deutsches Institut für Normung.
DIN, 2014. Image Quality Assurance in Diagnostic X-ray Departments – Part 15: RöV
Constancy Testing of X-ray Installations for Dental Radiographic Equipment for Digital
Cone-beam Computed Tomography. Berlin: Deutsches Institut für Normung.
De Vos, W., Casselman, J., Swennen, G.R.J., 2009. Cone-beam computerized tomography
(CBCT) imaging of the oral and maxillofacial region: a systematic review of the literature.
Int. J. Oral Maxillofac. Surg. 38, 609–625.
Dijkstra, M.L., Eagleton, M.J., Greenberg, R.K., et al., 2011. Intraoperative C-arm cone-
beam computed tomography in fenestrated/branched aortic endografting. J. Vasc. Surg.
53, 583–590.
Ding, G., Duggan, D., Coffey, C., 2008. Accurate patient dosimetry of kilovoltage cone-beam
CT in radiation therapy. Med. Phys. 35, 1135–1144.
Ding, G., Coffey, C., 2009. Radiation dose from kilovoltage cone beam computed tomog-
raphy in an image-guided radiotherapy procedure. Int. J. Radiat. Oncol. Biol. Phys. 73,
610–617.
Ding, G., Munro, P., 2013. Radiation exposure to patients from image guidance procedures
and techniques to reduce the imaging dose. Radiother. Oncol. 108, 91–98.
Dixon, R.L., Boone, J., 2010a. The CTDI Paradigm: a Practical Explanation for Medical
Physicists. Image Wisely. Reston, VA: American College of Radiology, 2010. Available at:
http://www.imagewisely.org//media/ImageWisely%20Files/Medical%20Physicist%20
Articles/IW%20Dixon%20Boone%20CTDI%20Paradigm%202.pdf (last accessed 5 May
2015).
Dixon, R.L., Boone, J., 2010b. Cone beam CT dosimetry: a uniEed and self-consistent
approach including all scan modalities – with or without phantom motion. Med. Phys.
37, 2703–2718.

105

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Domienik, J., Brodecki, M., Carinou, E., et al., 2011. Extremity and eye lens doses in inter-
ventional radiology and cardiology procedures: first results of the ORAMED project.
Radiat. Prot. Dosimetry 144, 442–447.
EC, 2011. Proposal for a Council Directive Laying Down Basic Safety Standards
for Protection Against the Dangers Arising from Exposure to Ionising Radiation.
European Commission, Brussels. Available at: http://ec.europa.eu/energy/en/topics/nuclear-
energy/radiation-protection (last accessed 5 May 2015).
EC, 2012a. Radiation Protection No. 172: Cone Beam CT for Dental and Maxillofacial
Radiology. Evidence Based Guidelines. Directorate-General for Energy. European
Commission, Brussels. https://ec.europa.eu/energy/sites/ener/files/documents/172.pdf (last
accessed 5 May 2015).
EC, 2012b. Radiation Protection No. 162: Criteria for Acceptability of Medical Radiological
Equipment Used in Diagnostic Radiology, Nuclear Medicine and Radiotherapy and
Maxillofacial Radiology. Evidence Based Guidelines. Directorate-General for Energy.
European Commission, Brussels. https://ec.europa.eu/energy/sites/ener/files/documents/
162.pdf (last accessed 5 May 2015).
EC, 2014. Radiation Protection No. 174: European Guidelines on Medical Physics Expert.
Directorate-General for Energy. European Commission, Brussels. https://ec.europa.eu/
energy/sites/ener/files/documents/174.pdf (last accessed 5 May 2015).
Ejima, K., Shoda, M., Yagishita, D., et al., 2010. Image integration of three-dimensional cone-
beam computed tomography angiogram into electroanatomical mapping system to guide
catheter ablation of atrial fibrillation. Europace 12, 45–51.
Faccioli, N., Foti, G., Barillari, M., et al., 2010. Finger fractures imaging: accuracy of cone-
beam computed tomography and multislice computed tomography. Skeletal Radiol. 39,
1087–1095.
Fahrig, R., Fox, A.J., Lownie, S., et al., 1997. Use of a C-arm system to generate true three-
dimensional computed rotational angiograms: preliminary in vitro and in vivo results. Am.
J. Neuroradiol. 18, 1507–1514.
Fahrig, R., Holdsworth, D.W., Lownie, S., et al., 1998. Computed rotational angiography:
system performance assessment using in-vitro and in-vivo models. Proc. SPIE 3336,
Medical Imaging 1998: Physics of Medical Imaging, 305.
Fahrig, R., Dixon, R., Payne, T., et al., 2006. Dose and image quality for a cone beam C-arm
CT System. Med. Phys. 33, 4541–4550.
Fiorella, D., Turk, A., Chaudry, I., et al., 2014. A prospective, multicenter pilot study inves-
tigating the utility of flat detector derived parenchymal blood volume maps to estimate
cerebral blood volume in stroke patients. J. Neurointerv. Surg. 6, 451–456.
Fraser, J., Reed, M., 2013. Appropriateness of imaging in Canada. Can. Assoc. Radiol. J. 64,
82–84.
Gayou, O., Parda, D.S., Johnson, M., et al., 2007. Patient dose and image quality from mega-
voltage cone beam computed tomography imaging. Med. Phys. 34, 499–506.
Grasruck, M., Suess, C., Stierstorfer, K., et al., 2005. Evaluation of image quality and dose on
a flat-panel CT-scanner. Proc. SPIE 5745, 179–188.
Grimmer, R., Oelhafen, M., Elstrøm, U., et al., 2009. Cone-beam CT image reconstruction
with extended z range. Med. Phys. 36, 3363–3370.
Gupta, R., Grasruck, M., Suess, C., et al., 2006. Ultra-high resolution flat-panel volume CT:
fundamental principles, design architecture, and system characterization. Eur. Radiol. 16,
1191–1205.

106

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Gupta, R., Cheung, A.C., Bartling, S.H., et al., 2008. Flat-panel volume CT: fundamental
principles, technology, and applications. Radiographics 28, 2009–2022.
He, W., Huda, W., Magill, D., et al., 2010. Patient doses and projection angle in cone beam
CT. Med. Phys. 37, 2359–2368.
Hendrick, R.E., Pisano, E.D., Averbukh, A., et al., 2010. Comparison of acquisition param-
eters and breast dose in digital mammography and screen-film mammography in the
American College of Radiology Imaging Network Digital Mammographic Imaging
Screening Trial. AJR Am. J. Roentgenol. 194, 362–369.
Herrnsdorf, L., Söderberg, M., 2013. A method to characterize the radiation output from a
cone beam O-arm using a device for dose and dose profile scanning measurement. Proc.
SPIE 8668, Medical Imaging 2013: Physics of Medical Imaging, 86682Q.
Hirsch, E., Wolf, U., Heinicke, F., et al., 2008. Dosimetry of the cone beam computed tom-
ography Veraviewepocs 3D compared with the 3D Accuitomo in different fields of view.
Dentomaxillofac. Radiol. 37, 268–273.
Hodez, C., Griffaton-Taillandier, C., Bensimon, I., 2011. Cone-beam imaging: applications in
ENT. Eur. Ann. Otorhinolaryngol. Head Neck Dis. 128, 65–78.
HPA, 2010a. Health Protection Agency Recommendations for the Design of X-ray Facilities
and Quality Assurance of Dental Cone Beam CT (Computed Tomography) Systems. HPA-
RPD-065. Health Protection Agency, Chilton.
HPA, 2010b. Health Protection Agency Guidance on the Safe Use of Dental Cone Beam CT
(Computed Tomography) Equipment. HPA-CRCE-010. Health Protection Agency,
Chilton.
IAC, 2012. The IAC Dental CT Standards for Dental/Maxillofacial Computed Tomography
(CT) Practice Accreditation Using Cone Beam Technology. Ellicott City, MD:
Intersocietal Accreditation Commission. Available at: http://www.intersocietal.org/
dental/standards/IAC_DentalCT_Standards.pdf (last accessed 5 May 2015).
IAEA, 1996. International Basic Safety Standards for Protection Against Ionizing Radiation
and for the Safety of Radiation Sources. IAEA Safety Series No. 115. International Atomic
Energy Agency, Vienna.
IAEA, 1999. Assessment of Occupational Exposure Due to External Sources of Radiation.
IAEA Safety Guide RS-G-1.3. International Atomic Energy Agency, Vienna.
IAEA, 2004. Individual Monitoring. IAEA-PRTM-2 (Rev.1). International Atomic Energy
Agency, Vienna.
IAEA, 2010. New Era in CT Scanning. International Atomic Energy Agency, Vienna.
Available at: https://rpop.iaea.org/RPOP/RPoP/Content/News/new-era-ct-scanning.htm
(last accessed 5 May 2015).
IAEA, 2011a. Status of Computed Tomography Dosimetry for Wide Cone Beam CT
Scanners. IAEA Human Health Reports No. 5. International Atomic Energy Agency,
Vienna.
IAEA, 2011b. Quality Assurance Programme for Computed Tomography: Diagnostic and
Therapy Applications. IAEA Human Health Series 19. International Atomic Energy
Agency, Vienna.
ICRP, 1991. 1990 Recommendations of the International Commission on Radiological
Protection. ICRP Publication 60. Ann. ICRP 21(1–3).
ICRP, 1997. General principles for the radiation protection of workers. ICRP Publication 75.
Ann. ICRP 27(1).
ICRP, 2000a. Managing patient dose in computed tomography. ICRP Publication 87. Ann.
ICRP 30(4).

107

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

ICRP, 2000b. Avoidance of radiation injuries from medical interventional procedures. ICRP
Publication 85. Ann. ICRP 30(2).
ICRP, 2004. Managing patient dose in digital radiology. ICRP Publication 93. Ann. ICRP
34(1).
ICRP 2007a. Managing patient dose in multi-detector computed tomography (MDCT). ICRP
Publication 102. Ann. ICRP 37(1).
ICRP, 2007b. The 2007 Recommendations of the International Commission on Radiological
Protection. ICRP Publication 103. Ann. ICRP 37(2–4).
ICRP, 2007c. Radiological protection in medicine. ICRP Publication 105. Ann. ICRP 37(6).
ICRP, 2009. Education and training in radiological protection for diagnostic and interven-
tional procedures. ICRP Publication 113. Ann. ICRP 39(5).
ICRP, 2010. Radiological protection in fluoroscopically guided procedures performed outside
the imaging department. ICRP Publication 117. Ann. ICRP 40(6).
ICRP, 2012. ICRP statement on tissue reactions/early and late effects of radiation in normal
tissues and organs – threshold doses for tissue reactions in a radiation protection context.
ICRP Publication 118. Ann. ICRP 41(1/2).
ICRP, 2013. Radiological protection in cardiology. ICRP Publication 120. Ann. ICRP 42(1).
ICRU, 2010. Prescribing, recording, and reporting intensity-modulated photon-beam therapy
(IMRT). ICRU Report 83. J. ICRU 10, 1–106.
ICRU, 2012. Radiation dose and image quality assessment in computed tomography. ICRU
Report 87. J. ICRU 12, 1–149.
IEC, 2006. Evaluation and Routine Testing in Medical Imaging Departments – Part 26:
Constancy Tests – Imaging Performance of Computed Tomography X-ray Equipment.
IEC 61223-2-6 ed. 2.0. International Electrotechnical Commission, Geneva.
IEC, 2010. Medical Electrical Equipment – Part 2-44: Particular Requirements for Basic
Safety and Essential Performance of X-ray Equipment for Computed Tomography.
IEC-60601-2-44 ed. 3.0, Amendment 1; 62B/804/CD. International Electrotechnical
Commission, Geneva.
IEC, 2011. Evaluation and Routine Testing in Medical Imaging Departments – Part 3-5:
Acceptance Tests – Imaging Performance of Computed Tomography X-ray Equipment.
IEC 61223-3-5 ed. 1.0. International Electrotechnical Commission, Geneva.
IEC, 2012. Electromedical Equipment – Part 2-44: Particular Requirements for the Basic
Safety and Essential Performance of X-ray Equipment for Computed Tomography. IEC
60601-2-44 ed. 3.1. International Electrotechnical Commission, Geneva.
IPEM, 2004. Guidance and Use of Diagnostic Reference Levels for Medical X-ray
Examinations. Report 88. Institute of Physics and Engineering in Medicine, York.
IPEM, 2005. Recommended Standards for the Routine Performance Testing of Diagnostic
X-ray Imaging Systems. Report 91. Institute of Physics and Engineering in Medicine,
York.
Jaffray, D.A., Drake, D.G., Moreau, M., et al., 1999. A radiographic and tomographic
imaging system integrated into a medical linear accelerator for localization of bone and
soft-tissue targets. Int. J. Radiat. Oncol. Biol. Phys. 45, 773–789.
Jaffray, A.D., Siewerdsen, J.H., 2000. Cone-beam computed tomography with a flat-panel
imager: initial performance characterization. Med. Phys. 27, 1311–1323.
Kalender, W.A., Beister, M., Boone, J.M., et al., 2012. High-resolution spiral CT of the breast
at very low dose: concept and feasibility considerations. Eur. Radiol. 22, 1–8.

108

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Kan, M.W., Leung, L.H., Wong, W., et al., 2008. Radiation dose from cone beam computed
tomography for image-guided radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 70,
272–279.
Kawata, Y., Niki, N., Kumazaki, T., 1996. Measurement of blood vessel characteristics for
disease detection based on cone beam CT images. IEEE Trans. Nucl. Sci. 43, 3348–3354.
Kim, S., Sopko, D., Toncheva, G., et al., 2011. Radiation dose from 3D rotational x-ray
imaging: organ and effective dose with conversion factors. Radiat. Prot. Dosimetry 150,
50–54.
Kim, D.W., Chung, W.K., Yoon, M., 2013. Imaging doses and secondary cancer risk from
kilovoltage cone-beam CT in radiation therapy. Health Phys. 104, 499–503.
Klein, B.E., Klein, R., Linton, K.L., et al., 1993. Diagnostic x-ray exposure and lens opacities:
the Beaver Dam Eye Study. Am. J. Public Health 83, 588–590.
Klein, E.E., Hanley, J., Bayouth, J., et al., 2009. Task Group 142 report: quality assurance of
medical accelerators. Med. Phys. 36, 4197–4212.
Korreman, S., Rasch, C., McNair, H., et al., 2010. The European Society of Therapeutic
Radiology and Oncology–European Institute of Radiotherapy (ESTRO–EIR) report on
3D CT-based in-room image guidance systems: a practical and technical review and guide.
Radiother. Oncol. 94, 129–144.
Kouno, T., Araki, F., Nakaguchi, Y., et al., 2013. Dose distribution from kV-cone beam
computed tomography in image-guided radiotherapy. Nihon Hoshasen Gijutsu Gakkai
Zasshi 69, 753–760.
Koyama, S., Aoyama, T., Oda, N., et al., 2010. Radiation dose evaluation in tomosynthesis
and C-arm cone-beam CT examinations with an anthropomorphic phantom. Med. Phys.
37, 4298–4306.
Kroes, M.W., Busser, W.M.H., Futterer, J.J., et al., 2013. Assessment of needle guidance
devices for their potential to reduce fluoroscopy time and operator hand dose during C-
arm cone-beam computed tomography-guided needle interventions. J. Vasc. Interv.
Radiol. 24, 901–906.
Kry, S.F., Salehpour, M., Followill, D.S., et al., 2005. Out-of-field photon and neutron dose
equivalents from step-and-shoot intensity-modulated radiation therapy. Int. J. Radiat.
Oncol. Biol. Phys. 62, 1204–1216.
Kwok, Y.M., Irani, F.G., Tay, K.H., et al., 2013. Effective dose estimates for cone beam
computed tomography in interventional radiology. Eur. Radiol. 23, 3197–3204.
Kyriakou, Y., Kalender, W., 2007. Efficiency of anti-scatter grids for flat-detector CT. Phys.
Med. Biol. 52, 6275–6293.
Kyriakou, Y., Richter, G., Dorfler, A., et al., 2008a. Neuroradiologic applications with rou-
tine C-arm flat panel detector CT: evaluation of patient dose measurements. AJNR Am. J.
Neuroradiol. 29, 1930–1936.
Kyriakou, Y., Deak, P., Langner, O., et al., 2008b. Concepts of dose determination in flat-
detector CT. Phys. Med. Biol. 53, 3551–3566.
Kyriakou, Y., Kolditz, D., Langner, O., et al., 2011. Digitale Volumentomografie (DVT) und
Mehrschicht-Spiral-CT (MSCT): eine objektive Untersuchung von Dosis und Bildqualität.
RöFo – Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden. Verfahren
183, 144–153.
Lange, J., Karellas, A., Street, J., et al., 2013. Estimating the effective radiation dose imparted
to patients by intraoperative cone-beam computed tomography in thoracolumbar spinal
surgery. Spine (Phila Pa 1976) 38, E306–E312.

109

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Levitt, M.R., Cooke, D.L., Ghodke, B.V., et al., 2011. ‘Stent view’ flat-detector CT and stent-
assisted treatment strategies for complex intracranial aneurysms. World Neurosurg. 75,
275–278.
Lindfors, K.K., Boone, J.M., Nelson, T.R., et al., 2008. Dedicated breast CT: initial clinical
experience. Radiology 246, 725–733.
Ludlow, J.B., Davies-Ludlow, L.E., Brooks, S.L., et al., 2006. Dosimetry of 3 CBCT devices
for oral and maxillofacial radiology: CB Mercuray, NewTom 3G and i-CAT.
Dentomaxillofac. Radiol. 35, 219–226.
Ludlow, J.B., Ivanovic, M., 2008. 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, 106–114.
Mail, N., Moseley, D.J., Siewerdsen, J.H., et al., 2009. The influence of bowtie filtration on
cone-beam CT image quality. Med. Phys. 36, 22–32.
Manarey, C.R.A., Anand, V.K., 2006. Radiation dosimetry of the FluoroCAT scan for real-
time endoscopic sinus surgery. Otolaryngol. Head Neck Surg. 135, 409–412.
McCollough, C.H., 2005. Automatic exposure control in CT: are we done yet? Radiology 237,
755–756.
Michel, M.S., Ritter, M., Wertz, H., et al., 2014. Theurological Dyna-CT: ex vivo feasibility
study of interventional cross-sectional imaging in the endourological operation room.
World J. Urol. 32, 277–280.
Miller, D.L., Vañó, E., Bartal, B., et al., 2010. Occupational radiation protection in interven-
tional radiology: a joint guideline of the Cardiovascular and Interventional Radiology
Society of Europe and the Society of Interventional Radiology. J. Vasc. Interv. Radiol.
21, 607–615.
Michell, M.J., Iqbal, A., Wasan, R.K., et al., 2012. A comparison of the accuracy of film-
screen mammography, full-field digital mammography, and digital breast tomosynthesis.
Clin. Radiol. 67, 976–981.
Miracle, A.C., Mukherji, S.K., 2009a. Conebeam CT of the head and neck, part 1: physical
principles. AJNR Am. J. Neuroradiol. 30, 1088–1095.
Miracle, A.C., Mukherji, S.K., 2009b. Conebeam CT of the head and neck, Part 2: clinical
applications. AJNR Am. J. Neuroradiol. 30, 1285–1292.
Morant, J.J., Salvadó, M., Hernández-Girón, I., et al., 2013. Dosimetry of a cone beam CT
device for oral and maxillofacial radiology using Monte Carlo techniques and ICRP adult
reference computational phantoms. Dentomaxillofac. Radiol. 42, 92555893.
Mori, S., Endo, M., Nishizawa, K., et al., 2005. Enlarged longitudinal dose profiles in cone-
beam CT and the need for modified dosimetry. Med. Phys. 32, 1061–1069.
Mozzo, P., Procacci, C., Tacconi, A., et al., 1998. A new volumetric CT machine for dental
imaging based on the cone-beam technique: preliminary results. Eur. Radiol. 8, 1558–1564.
Nakashima, E., Neriishi, K., Minamoto, A., et al., 2006. A reanalysis of atomic-bomb cataract
data, 2000–2002: a threshold analysis. Health Phys. 90, 154–160.
NCRP, 1995. Use of Personal Monitors to Estimate Effective Dose Equivalent and Effective
Dose to Workers for External Exposure to Low-LET Radiation. NCRP Report No. 122.
National Council on Radiation Protection and Measurements, Bethesda, MD.
NCRP, 2000. Radiation Protection for Procedures Performed Outside the Radiology
Department. NCRP Report No. 133. National Council on Radiation Protection and
Measurements, Bethesda, MD.

110

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

NCRP, 2005. Structural Shielding Design for Medical X-ray Imaging Facilities. NCRP
Report No. 147. National Council on Radiation Protection and Measurements,
Bethesda, MD.
NCRP, 2010. Radiation Dose Management for Fluoroscopically Guided Medical Procedures.
NCRP Report No. 168. National Council on Radiation Protection and Measurements,
Bethesda, MD.
NCRP, 2012. Reference Levels and Achievable Doses in Medical and Dental Imaging:
Recommendations for the United States. NCRP Report No. 172. National Council on
Radiation Protection and Measurements, Bethesda, MD.
NEMA, 2010. Computed Tomography Dose Check. NEMA Standards Publication XR 25-
2010. National Electrical Manufacturers Association, Arlington, VA.
NEMA, 2013. New MITA Smart Dose Standard Enhances Dose Optimization and
Management in CT Equipment. National Electrical Manufacturers Association,
Arlington, VA. Available at: http://www.nema.org/News/Pages/New-MITA-Smart-Dose-
Standard-Enhances-Dose-Optimization-and-Management-in-CT-Equipment.aspx (last
accessed 5 May 2015).
Neriishi, K., Nakashima, E., Minamoto, A., et al., 2007. Postoperative cataract cases among
atomic bomb survivors: radiation dose response and threshold. Radiat. Res. 168, 404–408.
Niklason, L.T., Christian, B.T., Niklason, L.E., 1997. Digital tomosynthesis in breast ima-
ging. Radiology 205, 399–406.
Ning, R., Chen, B., Yu, R., et al., 2000a. Flat panel detector-based cone beam volume CT
angiography imaging: system evaluation. IEEE Trans. Med. Im. 19, 949–963.
Ning, R., Colbeth, R.E., Chen, B., et al., 2000b. Real time flat panel detector-based volume
tomographic angiography imaging: detector evaluation. Proc. SPIE 3977, Medical Imaging
2000: Physics of Medical Imaging, 396.
Nottmeier, E.W., Pirris, S.M., Edwards, S., et al., 2013. Operating room radiation exposure in
cone beam computed tomography-based, image-guided spinal surgery. J. Neurosurg. Spine
19, 226–231.
O’Connell, A., Conover, D.L., Zhang, Y., et al., 2010. Cone-beam CT for breast imaging:
radiation dose, breast coverage, and image quality. AJR Am. J. Roentgenol. 195, 496–509.
Orth, R.C., Wallace, M.J., Kuo, M.D., 2008. C-arm cone-beam CT: general principles and
technical considerations for use in interventional radiology. J. Vasc. Interv. Radiol. 19,
814–820.
Packard, N.J., Abbey, C.K., Yang, K., et al., 2012. Effect of slice thickness on detectability in
breast CT using a prewhitened matched filter and simulated mass lesions. Med. Phys. 39,
1818–1830.
Padovani, R., Le Heron, J., Cruz-Suarez, R., et al., 2011. International project on individual
monitoring and radiation exposure levels in interventional cardiology. Radiat. Prot.
Dosimetry 144, 437–441.
Paul, J., Jacobi, V., Farhang, M., et al., 2013a. Radiation dose and image quality of x-ray
volume imaging systems: cone-beam computed tomography, digital subtraction angiog-
raphy and digital fluoroscopy. Eur. Radiol. 23, 1582–1593.
Paul, J., Mbalisike, E.C., Vogl, T.J., 2013b. Radiation dose to procedural personnel and
patients from an x-ray volume imaging system. Eur. Radiol. 23, 3262–3270.
Pauwels, R., Theodorakou, C., Walker, A., et al., 2012a. Dose distribution for dental cone
beam CT and its implication for defining a dose index. Dentomaxillofac. Radiol. 41,
583–593.

111

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Pauwels, R., Beinsberger, J., Collaert, B., et al., 2012b. Effective dose range for dental cone
beam computed tomography scanners. Eur. J. Radiol. 81, 267–271.
Pauwels, R., Stamatakis, H., Bosmans, H., et al., 2013. Quantification of metal artifacts on
cone beam computed tomography images. Clin. Oral Impl. Res. 24(Suppl. A100), 94–99.
Pauwels, R., Zhang, G., Theodorakou, C., et al., 2014. Effective radiation dose and eye lens
dose in dental cone beam CT: effect of field of view and angle of rotation. Br. J. Radiol. 87,
20130654.
Pisano, E.D., Gatsonis, C., Hendrick, E., 2005. Diagnostic performance of digital versus film
mammography for breast-cancer screening. N. Engl. J. Med. 353, 1773–1883.
Podnieks, E.C., Negus, I.S., 2012. Practical patient dosimetry for partial rotation cone beam
CT. Br. J. Radiol. 85, 161–167.
Popescu, S., Stierstorfer, K., Flohr, T., et al., 2005. Design and evaluation of a prototype
volume CT scanner. Proc. SPIE 5745, 600–608.
Poplack, S.P., Tosteson, T.D., Kogel, C.A., et al., 2007. Digital breast tomosynthesis: initial
experience in 98 women with abnormal digital screening mammography. AJR Am. J.
Roentgenol. 189, 616–623.
Pouliot, J., Bani-Hashemi, A., Chen, J., et al., 2005. Low-dose megavoltage cone-beam CT for
radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 61, 552–560.
Psychogios, M.N., Buhk, J.H., Schramm, P., et al., 2010. Feasibility of angiographic CT in
peri-interventional diagnostic imaging: a comparative study with multidetector CT. AJNR
Am. J. Neuroradiol. 31, 1226–1231.
Racadio, J.M., Babic, D., Homan, R., et al., 2007. Live 3D guidance in the interventional
radiology suite. AJR Am. J. Roentgenol. 189, W357–W364.
Ramdhian-Wihlm, R., Le Minor, J.M., Schmittbuhl, M., et al., 2012. Cone-beam computed
tomography arthrography: an innovative modality for the evaluation of wrist ligament and
cartilage injuries. Skeletal Radiol. 41, 963–969.
Rastegar, N., Eckart, P., Mertz, M., 2002. Radiation-induced cataract in astronauts and
cosmonauts. Graefes Arch. Clin. Exp. Ophthalmol. 240, 543–547.
Rehani, M.M., 2013. Challenges in radiation protection of patients in the 21st Century. AJR
Am. J. Roentgenol. 200, 762–764.
Rehani, M.M., Berry, M., 2000. Radiation doses in computed tomography. BMJ. 320,
593–594.
Rehani, M.M., Ortiz López, P., 2006. Radiation effects in fluoroscopically guided cardiac
interventions – keeping them under control. Int. J. Cardiol. 109, 147–151.
Rehani, M., Frush, D., 2010. Tracking radiation exposure of patients. Lancet 4, 376, 754–745.
Rehani, M.M., Frush, D.P., 2011. Patient exposure tracking: the IAEA smart card project.
Radiat. Prot. Dosimetry 147, 314–316.
Rehani, M.M., Vañó, E., Ciraj-Bjelac, O., et al., 2011. Radiation and cataract. Radiat. Prot.
Dosimetry 147, 300–304.
Rehani, M.M., Srimahachota, S., 2011. Skin injuries in interventional procedures. Radiat.
Prot. Dosimetry 147, 8–12.
Reichardt, B., Sarwar, A., Bartling, S.H., et al., 2008. Musculoskeletal applications of flat-
panel volume CT. Skeletal Radiol. 37, 1069–1076.
Robb, R.A., 1982. Dynamic spatial reconstructor: an x-ray video fluoroscopic CT scanner for
dynamic volume imaging of moving organs. IEEE Trans. Med. Im. MI-1, 22–23.

112

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Ross, W.R., Dawn, C., Fitzgerald, P., et al., 2004. Performance and Pre-clinical Results from
a Flat-panel-based Volumetric CT System. Proceedings of 90th Scientific Assembly and
Annual Meeting of the Radiological Society of North America, 28 November–3 December
2004, Chicago, IL, USA, SSG18-02.
Roy, O.P., Angle, J.F., Jenkins, A.D., et al., 2012. Cone beam computed tomography for
percutaneous nephrolithotomy: initial evaluation of a new technology. J. Endourol. 26,
814–818.
Saint-Félix, D., Trousset, Y., Picard, C., et al., 1994. In vivo evaluation of a new system for 3D
computerized angiography. Phys. Med. Biol. 39, 584–595.
Sanchez, R.M., Vañó, E., Fernández, J.M., et al., 2014. Brain radiation doses to patients in an
interventional neuroradiology laboratory. AJNR Am. J. Neuroradiol. 35, 1276–1280.
Schafer, S., Nithiananthan, S., Mirota, D.J., et al., 2011. Mobile C-arm cone-beam CT for
guidance of spine surgery: image quality, radiation dose, and integration with interven-
tional guidance. Med. Phys. 38, 4563–4574.
Schafer, S., Stayman, J.W., Zbijewski, W., et al., 2012. Antiscatter grids in mobile C-arm
cone-beam CT: effect on image quality and dose. Med. Phys. 39, 153–159.
Schueler, B.A., Sen, A., Hsiung, H.H., et al., 1997. Three-dimensional vascular reconstruction
with a clinical x-ray angiography system. Acad. Radiol. 4, 693–699.
Schulz, B., Heidenreich, R., Heidenreich, M., et al., 2012. Radiation exposure to operating
staff during rotational flat-panel angiography and C-arm cone beam computed tomog-
raphy (CT) applications. Eur. J. Radiol. 81, 4138–4142.
Seuri, R., Rehani, M.M., Kortesniemi, M., 2013. How tracking radiologic procedures and
dose helps: experience from Finland. AJR Am. J. Roentgenol. 200, 771–775.
Shah, A., Aird, E., Shekhdar, J., 2012. Contribution to normal tissue dose from concomitant
radiation for two common kV-CBCT systems and one MVCT system used in radiotherapy.
Radiother. Oncol. 105, 139–144.
Shope, T.B., Gagne, R.M., Johnson, G.C., 1981. A method for describing the doses delivered
by transmission x-ray computed tomography. Med. Phys. 8, 488–495.
Sierzenski, P.R., Linton, O.W., Amis, E.S., Jr, et al., 2014. Applications of justification and
optimization in medical imaging: examples of clinical guidance for computed tomography
use in emergency medicine. J. Am. Coll. Radiol. 11, 36–44.
Siewerdsen, J.H., Jaffray, D.A, 1999. Cone-beam computed tomography with a flat-panel
imager: effects. Med. Phys. 26, 2635–2647.
Siewerdsen, J.H., Jaffray, D.A., 2001. Cone-beam computed tomography with a flat-panel
imager: magnitude and effects of x-ray scatter. Med. Phys. 28, 220–231.
Sistrom, C.L., Dang, P.A., Weilburg, J.B., et al., 2009. Effect of computerized order entry with
integrated decision support on the growth of outpatient procedure volumes: seven-year
time series analysis. Radiology 251, 147–155.
Song, W.Y., Kamath, S., Ozawa, S., et al., 2008. A dose comparison study between XVI and
OBI CBCT systems. Med. Phys. 35, 480–486.
Sowards-Emmerd, D., Balakrishnan, K., Wiener, J., Shao, L., Ye, J., 2009. CBCT-subsystem
Performance of the Multi-modality Brightview XCT System (M09-26). Nuclear Science
Symposium Conference Record (NSS/MIC), 2009 IEEE, p. 3053–3058.
Spezi, E., Downes, P., Jarvis, R., et al., 2012. Patient-specific three-dimensional concomitant
dose from cone beam computed tomography exposure in image-guided radiotherapy. Int.
J. Radiat. Oncol. Biol. Phys. 83, 419–426.

113

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Stock, M., Palm, A., Altendorfer, A., et al., 2012. IGRT induced dose burden for a
variety of imaging protocols at two different anatomical sites. Radiother. Oncol.
102, 355–363.
Sutton, D.G., Martin, C.J., Williams, J.R., et al., 2012. Radiation Shielding for Diagnostic
Radiology. British Institute of Radiology, London.
Suzuki, S., Yamaguchi, I., Kidouchi, T., et al., 2011. Evaluation of effective dose during
abdominal three-dimensional imaging for three flat-panel-detector angiography systems.
Cardiovasc. Intervent. Radiol. 34, 376–382.
Theodorakou, C., Walker, A., Horner, K., et al., 2012. Estimation of paediatric organ and
effective doses from dental cone beam CT using anthropomorphic phantoms. Br. J. Radiol.
85, 153–160.
Tuominen, E.K.J., Kankare, J., Koskinen, S.K., et al., 2013. Weight-bearing CT imaging of
the lower extremity. AJR Am. J. Roentgenol. 200, 146–148.
Tyan, Y.S., Li, Y.Y., Ku, M.C., et al., 2013. The effective dose assessment of C-arm CT in
hepatic arterial embolisation therapy. Br. J. Radiol. 86, 20120551.
Tzedakis, A., Damilakis, J., Perisinakis, K., et al., 2005. The effect of z overscanning on
patient effective dose from multidetector helical computed tomography examinations.
Med. Phys. 32, 1621–1629.
Vañó, E., González, L., Beneytez, F., et al., 1998. Lens injuries induced by occupational
exposure in nonoptimized interventional radiology laboratories. Br. J. Radiol. 71, 728–733.
Vañó, E., Kleiman, N.J., Duran, A., et al., 2010. Radiation cataract risk in interventional
cardiology personnel. Radiat. Res. 174, 490–495.
Vañó, E., Kleiman, N.J., Duran, A., et al., 2013. Radiation-associated lens opacities in cath-
eterization personnel: results of a survey and direct assessments. J. Vasc. Interv. Radiol. 24,
197–204.
Wallace, M.J., Kuo, M.D., Glaiberman, C., et al., 2008. Three-dimensional C-arm cone-beam
CT: applications in the interventional suite. J. Vasc. Interv. Radiol. 19, 799–813.
Wang, X., 1997. Volume Tomographic Angiography. PhD thesis. University of Rochester,
Rochester, NY.
Wen, N., Guan, H., Hammoud, R., et al., 2007. Dose delivered from Varian’s
CBCT to patients receiving IMRT for prostate cancer. Phys. Med. Biol. 52,
2267–2276.
Wielandts, J.Y., Smans, K., Ector, J., et al., 2010. Effective dose analysis of three-dimen-
sional rotational angiography during catheter ablation procedures. Phys. Med. Biol.
55, 563–579.
Wiesent, K., Barth, K., Navab, N., et al., 2000. Enhanced 3D reconstruction algorithm for C-
arm systems suitable for interventional procedures. IEEE Trans. Med. Im. 19, 391–403.
Wintermark, M., Lev, M.H., 2010. FDA investigates the safety of brain perfusion CT. AJNR
Am. J. Neuroradiol. 31, 2–3.
Wiper, A., Katira, A., Roberts, D.H., 2005. Interventional cardiology: it’s a hairy business.
Heart 91, 1432.
Yuan, M.K., Tsai, D.C., Chang, S.C., et al., 2013. The risk of cataract associated with
repeated head and neck CT studies: a nationwide population-based study. AJR Am. J.
Roentgenol. 201, 626–630.

114

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Zbijewski, W., De Jean, P., Prakash, P., et al., 2011. A dedicated cone-beam CT system for
musculoskeletal extremities imaging: design, optimization, and initial performance charac-
terization. Med. Phys. 38, 4700–4713.
Zhang, G., Marshall, N., Bogaerts, R., et al., 2013. Monte Carlo modeling for dose assessment
in cone beam CT for oral and maxillofacial applications. Med. Phys. 40, 072103.

115

Downloaded from ani.sagepub.com by guest on July 22, 2015

Downloaded from ani.sagepub.com by guest on July 22, 2015
 ANNEX A. ASSESSING PATIENT DOSES IN CBCT

(A1) This annex provides a more in-depth description of patient dosimetry meth-
ods and limitations in CBCT. A summarised version is given in Section 5. A more
extensive coverage of dosimetry in CBCT can be found in Report 87 (ICRU, 2012).

A.1. Dosimetry in CBCT

(A2) CBCT uses a wide x-ray beam for 3D imaging of a relatively large volume.
Since the mid-1990s, the trend in MDCT has been towards an ever-increasing
number of slices with a concomitant increase in x-ray beam width; the z-axis cover-
age of the high-end, wide-area MDCT scanners available today rivals that of CBCT.
These developments have created a drive to update CT dosimetry methods so that
they are more apropos wide area detectors. As a result, some of the work from
MDCT dosimetry, for which established measurement methods and phantoms
exist already, can be translated to CBCT dosimetry. This annex discusses the short-
comings of the standard narrow-beam MDCT formalism when it is applied directly
to CBCT. In order to construct a comprehensive framework for CBCT dosimetry,
methods to overcome these problems are described.
(A3) CT dosimetry has evolved around the concept of CTDI. From its introduc-
tion in the 1980s (Shope et al., 1981), CTDI has taken different forms depending on
the adopting organisation: the United States Food and Drug Administration, IEC,
and other similar agencies. CTDI has mainly been used to compare dose character-
istics of different CT machines, to test the stability of equipment performance
(quality control), and, in some instances, to estimate patient dose despite the fact
that CTDI does not provide a direct assessment of patient dose. An extensive
description of the CTDI concept can be found in Report 87 (ICRU, 2012).
(A4) Increasingly, wide beams in modern CT and CBCT scanners complicate
CTDI measurements (Abuhaimed et al., 2014). Even for a nominal beam width of
20 mm, it is evident that CTDI measured over a length of 100 mm (i.e. CTDI100) does
not cover the tails of the dose profile in a poly(methyl methacrylate) (PMMA)
phantom. The ratio of CTDI100/CTDI1 is called ‘CTDI measurement efficiency’.
Kyriakou et al. (2008b) have shown that for 200-mm collimation, an integration
length of >600 mm would be required to approximate CTDI1 within 1%.
(A5) An alternative method for CTDI estimation in O-arm systems has been
proposed, using a point-dose detector that is moved through the CBCT field, pre-
ferably in-air (Herrnsdorf and Söderberg, 2013).
(A6) Issues regarding efficiency of CTDI measurements have been the basis of the
new approach of wide-beam CT dosimetry. IAEA (2011a) adopted a two-step
approach proposed by IEC (2010). More details regarding this modified approach
can be found in Report 87 (ICRU, 2012).
(A7) It would be useful to mention that CTDI alone is not a useful indicator of
patient dose. In order to connect the CTDI-like measurements with dose, CTDIvol

117

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

and DLP have been used extensively in clinical practice as relative patient dose
indicators. CTDIvol and DLP are connected by the equation:

DLP ¼ L:CTDIvol

where L is the length of the scan. The CTDIvol paradigm is problematic in cases
where there is no helical scan or patient motion (as is the case with many CBCT
scanners). In such cases, reported CTDIvol values will overestimate the dose signifi-
cantly (Dixon and Boone, 2010a).

A.2. Point-of-care scanning and clinic-based CBCT systems

(A8) Clinic-based systems include head and neck CBCT, bCT, and dental and
maxillofacial CBCT. A particular property of dental and maxillofacial CBCT scan-
ners is that, depending on the system, varying FOV sizes are offered. This allows for
the scanning of localised regions (i.e. a single tooth and its immediate surroundings)
as well as maxillofacial scanning. The use of horizontal collimation, as well as other
factors, results in complicated dose distributions in the axial plane, providing an
additional challenge for dosimetry (Pauwels et al., 2012a). In addition, most
dental and maxillofacial CBCT systems are seated or standing, resulting in practical
complications regarding phantom and dosimeter placement.
(A9) For dental systems, the SEDENTEXCT Consortium report (EC, 2012a) dis-
cussed the use of KAP as well as CTDI-like measurements. It was proposed that CTDI
measurements should be performed during commissioning in cases when the machine
comes with data on such measurements from the manufacturer. On the grounds that
conventional CTDI has drawbacks for use of dental and maxillofacial CBCT (due to
wider beams and greater asymmetry of dose distribution in CBCT compared with
MDCT), the Consortium tried to define a single CBCT dose index (CBCT DI)
(Pauwels et al., 2012a). During this effort, a customised phantom (SEDENTEXCT
DI) was developed which is shown in Fig. A.1. It features suitable insets for the
placement of measuring equipment. The phantom consists of four ionisation chamber
plates (2  22 mm and 2  44 mm), one thermoluminescent dosimeter plate (22 mm
thick), and one film plate (22 mm thick). Three adapters with widths of 22, 44, and
66 mm are provided that can reduce the chamber diameter from 26 to 13 mm. Two
different measurement setups (Index 1 and Index 2) are depicted in Fig. A.1. Index 1 is
defined as the average of seven measurements along the diameter of the phantom, with
the measuring line connecting the centre of the FOV and the centre of the phantom;
this allows for central or off-axis FOV positioning. Index 2 is an adapted CTDIw,
measured using a small-volume ion chamber at the central axial plane and using ½
weighting between the central measurement and the average of four peripheral mea-
surements; the FOV should always be positioned centrally for this index.
(A10) In addition, the German standard DIN 6868-161 describes a dosimetry
method for dental and maxillofacial CBCT systems, which is also applicable for
any device with an accessible detector surface (DIN, 2013). The proposed method

118

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Fig. A.1. (a) The SEDENTEXCT dose index (DI) phantom (Leeds Test Objects,
Boroughbridge, UK) for radiation dose measurements in dental and maxillofacial cone
beam computed tomography systems. (b, c) Measuring points for the estimation of Index 1
and Index 2. Figure provided by Ruben Pauwels on behalf of the SEDENTEXCT Project
Consortium (Pauwels et al., 2012a; EC, 2012a). FOV, field of view.

is based on a phantomless dose measurement at the detector, which is corrected

through geometrical factors to estimate the dose at the isocentre.
(A11) Further validation of different possible indices is required, together with a
way to translate dose index readings into patient doses. Araki et al. (2013) concluded
that CBCT DI and KAP proposed by SEDENTEXCT could be used to establish
DRLs for dental and maxillofacial CBCT, and noted that the relationship between
these indices and patient dose remains to be determined.
(A12) While a standardised dose metric for clinic-based CBCT systems may differ
from that of other CBCT systems, it is important to consider both the practicality of
the measurements and their relevance in QA. Until conversion factors for phantom-
less CBCT dosimetry have been established, phantom measurements hold an impor-
tant advantage from a QA point of view.
(A13) It has been suggested that if the manufacturer has provided a CTDI dose
figure, this quantity should be measured during commissioning. However, not all
machines come with such initial measurements. Another dose index used for CBCT
dosimetric evaluations is KAP, which is often used in panoramic and cephalometric
radiography and, of course, is widely used in radiography and fluoroscopy. Some
machines display a KAP value on screen after the exposure. The accuracy of such
measurements should be verified by medical physicists. The use of KAP has been

119

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

proposed by HPA (2010a). The main advantage of KAP is that it is easy to calculate
by measuring dose and beam cross-section at a specific point. It is considered
suitable for auditing CBCT dose in dental practices (HPA, 2010b). The
SEDENTEXCT Consortium proposes that if such measurements are not provided,
the medical physicist should create a log of such readings in all clinically used set-
tings, so that the dentist may compare it with national and international audit levels
(EC, 2012a).
(A14) Technically, the methods described above could also be applied to other
clinic-based systems including, for example, systems for head and neck imaging, and
possibly bCT. However, there is currently no standardisation in the measurements
for such units. This highlights the fact that the issue of standardisation in CBCT
dosimetry remains largely unresolved.

A.3. C-arm CBCT systems

(A15) C-arm CBCT systems are incapable of performing a full rotation around the
patient couch. Some systems can only rotate 180 plus the beam angle (Fahrig et al.,
2006), which results in a non-uniform axial dose deposition to the patient/phantom.
In a phantom, the maximum dose occurs at the central plane intersecting the z-axis at
z ¼ 0, on the side of the phantom closest to the x-ray tube. In the ideal case in which
the heel effect is absent, the maximum dose would occur on the bisector of the
rotation angle. When the heel effect is present, the maximum dose occurs near the
bisector.
(A16) For C-arm CBCT systems, Fahrig et al. (2006) proposed a metric represent-
ing the average dose to the phantom central plane (z ¼ 0):

 1 2
Dð0Þ ¼ D0 þ D p
3 3

where D0 is the dose to the central point of the central plane (on the z-axis) and Dp is
the average peripheral dose. This equation follows a similar averaging to that used in
the calculation of CTDIw; the metric that is used for dosimetry on any conventional
CT scanner performing a rotation smaller than 360 . Fahrig et al. (2006) performed
the calculation using a Farmer ionisation chamber, and measured doses at the centre
and at eight peripheral positions at 1-cm depth from the head phantom’s surface.
Podnieks and Negus (2012) showed that effective dose can be estimated from CTDIw
and irradiated length with acceptable accuracy if the ionisation chamber positions
are considered carefully.

A.4. A unified approach to CT dosimetry

(A17) Report 87 (ICRU, 2012) reviewed a considerable body of work in order to
propose a method for CT dosimetry that compensates for the shortcomings of

120

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

current CTDI-based CT dosimetry methods. In addition, earlier work by Dixon and

Boone (2010b) provided a unified formalism for dose measurements on machines
capable of helical scanning (e.g. MDCT scanners), as well as on those that only
acquire axial images (which is the case with most CBCT scanners). A set of metrics
and the use of a new polyethylene 600-mm-long phantom are proposed. This method
has been described previously (AAPM, 2010), but in this publication, the notation as
presented in Report 87 (ICRU, 2012) was used. The mathematical foundation for the
method is beyond the scope of this publication; however, the method is discussed
briefly below.
(A18) A dosimetry quantity CTDIL is proposed, the physical meaning of which is
the dose at the centre (z ¼ 0) of the scanned length for a scan extending from
z ¼ L/2 to z ¼ L/2. This formalism provides a means to estimate the dose deposited
at the central plane of the phantom, at z ¼ 0. In the case of axial scans, such as those
performed with most CBCT machines, the quantity that corresponds intuitively to
CTDI is the dose at the central point of the beam on the z-axis. If f(z) is the dose
profile function, this dose is f(0). For a number (N) of identical axial scans centred at
z ¼ 0, the dose of interest will be equal to Nf(0).

A.4.1. Formalism

(A19) For a helical CT scan, the accumulated absorbed dose distribution at the
centre of the scan length (from L/2 to +L/2) is represented by a convolution of the
axial dose profile with a rectangular function, (z/L) of scan length L. This repre-
sentation is only valid when x-ray tube current modulation is not used. Fig. A.2
shows normalised cumulative absorbed dose distributions for a series of helical CT
scans of differing scan lengths, produced by Monte Carlo simulation (Boone, 2009).
(A20) The dose DL(0) at the central part of the beam (z ¼ 0) for a beam width L
increases as the width of the beam increases. This can be seen in Fig. A.2. DL(0)
approaches a maximum value asymptotically, when the beam width increases. This
value is called the ‘equilibrium dose’ (Deq), and could be understood as CTDI1 (i.e.
when the entire dose profile has been collected).

A.4.2. Cumulative absorbed dose distribution from a helical scan of length L

(A21) The cumulative absorbed dose distribution DL(z) for helical scans in which
the table moves by a distance b per gantry rotation can be calculated by using the
following equation; this is only applicable when tube current modulation is not used.

Z þL=2
1
DL ðzÞ ¼ f ðz  z0 Þdz0
b L=2

121

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Fig. A.2. Normalised absorbed dose as a function of z-position for a number of different
scan lengths: 10 mm, 50 mm, 100 mm, 150 mm, 200 mm, 300 mm, 400 mm, 500 mm, and
600 mm (from centre to edge of graph). These data were derived by convolving the dose
spread function computed from the Monte Carlo simulation with rectangular functions
characterising the length of the scan, for a 320-mm diameter poly(methyl methacrylate)
phantom at 120 kV, using a GE Lightspeed 16-body bowtie filter. Source: ICRU (2012).

(A22) At z ¼ 0 and taking into account that pitch (p) is defined as p ¼ b/nT, the
above equation becomes:
Z þL=2
1
DL ð0Þ ¼ f ðz0 Þdz0 ¼ p:CTDIL
b L=2

(A23) Note that for p ¼ 1, D L (0) ¼ CTDI L . Conceptually, D L (0) as a func-

tion of L uses the data points along a vertical line perpendicular to z ¼ 0 in
Fig. A.2.
(A24) DL(0) depends on L, until the asymptote Deq is reached at very long scan
lengths. A new function capable of representing this dependence needs to be intro-
duced. The mathematical synonym function h(L) ¼ DL(0) is thus:
Z þL=2
1
hðLÞ ¼ f ðz0 Þdz0
b L=2

(A25) Conceptually, h(L) is the integral of the intercepted dose profile on

the z-axis for a scan of length L by keeping the detector at the centre of the
phantom.

122

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

Fig. A.3. Graphs showing measured H(L) curves. These data were measured in a 900-mm-
long, 320-mm-diameter poly(methyl methacrylate) phantom, scanned at 120 kV. Three dif-
ferent beam collimation widths are shown in each plot for the centre position (left panel)
and the periphery (right panel). Source: Mori et al. (2005).
(A26) If the cumulative absorbed dose at z ¼ 0 is normalised to Deq, the above
equation becomes:

hðLÞ DL ð0Þ
HðLÞ ¼ ¼
Deq Deq

(A27) Fig. A.3 shows H(L) curves measured by Mori et al. (2005). The maximum
H(L) value as a function of scan length L approaches unity asymptotically for large
scan lengths. This has been referred to as the rise to dose equilibrium curve. As H(L)
is normalised to unity at L!1, this function does not contain the tube output
information that h(L) does.
(A28) The physical interpretation of the rise to equilibrium curve is that the scan
and the phantom need to be long enough so that the asymptote tails of the profiles
are reached. The longer the scan, the closer H(L) approaches a value of unity.
This representation is therefore good for showing the relatively low efficiency of
short scans for collecting the actual dose, and this efficiency increases with longer
scans.

A.4.3. Phantoms

(A29) It has been shown that a phantom with a 300-mm diameter would need to
be at least 400-mm long to capture approximately 98% of Deq (this is equivalent to

123

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

Fig. A.4. The ICRU/AAPM TG 200 phantom. The phantom is made of high-density poly-
ethylene (0.97 g cm3) with a diameter of 300 mm and a length of 600 mm, which are suffi-
cient for measuring h(L) and H(L). (a) Design of the phantom. (b) Photograph of the
phantom. The phantom is large and weighs approximately 41 kg. Therefore, it was designed
to be modular with three different sections. Source: ICRU (2012).

saying that the scan profile interception would be 98% efficient). For a phantom with
the standard 320-mm diameter, a length of 425 mm would be required for the same
measurement efficiency. To tackle this problem, the committee responsible for ICRU
Report 87 collaborated with the American Association of Physicists in Medicine
(AAPM) task group responsible for Report 200 (AAPM, 2015). As a result of this
collaboration, the phantom, ICRU/AAPM TG 200, shown in Fig. A.4, was
developed.

A.4.4. Practical measurement of rise-to-equilibrium dose curves

(A30) Methods for measuring the H(L) or h(L) curves have been well described
in Report 111 (AAPM, 2010) and Report 87 (ICRU, 2012). Here, a short and
intuitive description of the measurement methods is given.
(A31) A long phantom and an integrating thimble ionisation chamber are needed.
A series of helical scans of different lengths is performed, and the air kerma inte-
grated by the thimble chamber is recorded. The scans are centred on the position of
the chamber. The air kerma readings as measured by the chamber are plotted as a
function of the length of the helical scan.
(A32) If a real-time radiation dosimeter is available, the rise-to-equilibrium curve
may be plotted using data obtained during a single long scan. In this case, the
dosimeter can create a full dose profile along the whole length of the
phantom. Different points on the curve may be calculated by integrating the dose
profile curve using appropriate integration limits (L/2 to L/2), where L is the total

124

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

integration length centred on the real-time radiation meter at the centre of the
phantom.

A.4.5. Measurements on machines only capable of axial acquisition

(A33) The methods described above are useful for measurements in MDCT
machines that provide the option to perform helical scans. However, most CBCT
machines do not perform helical scans. When table translation during a scan is not
available, it is necessary to modify the method, based on the notion that it is neces-
sary to measure a quantity that corresponds to CTDI of helical scans. As mentioned
previously, this quantity is f(0) (Dixon and Boone, 2010b). Practically speaking, f(0)
is measured by placing the ionisation chamber at the centre of the phantom and the
beam, and varying the beam width starting from the thinnest possible collimation to
the widest available. The measurement values can then be plotted against the beam
width . The values may be normalised to Aeq which is the equilibrium value that
would be reached for f(0) if the beam width was 470 mm. Such beam widths are, of
course, not found in clinical practice. Thus, the normalised approach-to-equilibrium
curve for the axial scan is only partial, and does not reach a value of 1 asymptoti-
cally. For axial CT scans with a cone beam width , dose f(0)a ¼ H()Aeq, the
conventional CT dose DL(0) can be described as a function of scan length L, includ-
ing a common equilibrium dose constant Aeq, a common scatter equilibrium length
eq ¼ Leq, and a common function H() which describes the relative approach to
dose equilibrium for both modalities, where  ¼ , or  ¼ L, such that
f(0)a ¼ H()Aeq and DL(0) ¼ H(L)Deq ¼ H(L)(b/)Aeq. Axial scanners that do
not have the facility to collimate the beam may be equipped with a collimation
gauge that could be inserted in front of the x-ray tube for dose measurement
purposes.
(A34) It is important to note here that the integration which needs to be performed
in order to measure CTDI is a result of the existence of table movement. The
definition of CTDI implies that dose to the central area of a phantom is affected
by scatter from adjacent areas. This phenomenon is completely absent in axial scans:
therefore, CTDI consistently overestimates the dose around the central area of the
phantom.

A.4.6. ICRU Report 87 recommendations

CTDIvol and CTDIair measurements

(A35) Traditionally, CTDIvol has been related to measurements of CT dose. IEC
has also recommended that CTDIvol should be displayed on the control screen of CT
scanners. Due to its widespread use and in order to keep continuity with older
measurements on CT scanners, ICRU recommends that both CTDIvol and
CTDIvol free-in-air should be measured at acceptance testing using both 160-mm-

125

Downloaded from ani.sagepub.com by guest on July 22, 2015

 ICRP Publication 129

and 320-mm-diameter PMMA phantoms at clinically relevant mAs settings across

the range of tube potentials used clinically. Furthermore, CTDIvol is used to scale
size-speciEc dose estimates (SSDE) as well as for normalisation of rise-to-equilibrium
curves. The x-ray output of the CT scanner, which is also characterised by CTDIair,
is a fundamental measurement that should be performed during acceptance testing
and after changing major components of the scanner related to dose.

Dosimetry in phantoms
(A36) If medical physicists follow the recommendations and measure CTDIvol and
CTDIair at acceptance testing, measurements of CTDIvol in phantoms are not needed
on a routine basis if periodic CTDIair measurements are stable.
(A37) Manufacturers should measure and provide users with a comprehensive set
of data for a reasonably wide range of beam settings used in clinical practice regard-
ing the rise-to-equilibrium curves of the scanner and related metrics such as H(L)
and h(L). G(L), which is the H(L) curve normalised by CTDIvol and thus related to
patient dose, should also be provided.
(A38) A subset of CTDI measurements performed by only using the central 200-
mm section of the phantom should also be provided by manufacturers so that G(L)
measured for the full 600-mm phantom can be associated with the partial G(L)
measurement acquired with the 200-mm phantom section.

Patient dose estimations

(A39) Patient dose can be estimated by using SSDE coupled with CTDIvol. The
method has been described in Report 87 (ICRU, 2012) and Report 204 (AAPM,
2011b). It must be considered, however, that CTDIvol calculation can be different for
partial rotation axial CT scans, such as in the case of a C-arm CBCT scan. Even for
full axial scans in which there is no patient translation, CTDIvol will overestimate
patient dose (Dixon and Boone, 2010b). This fact underlines the need for new
coefficients for patient dose estimation from f(0) measurements.

A.5. Tracking and reporting of radiation dose

(A40) New challenges emerge with systems being used for both fluoroscopy and
tomography (CBCT). While fluoroscopy radiation dose figures are normally avail-
able as KAP from the machines, CBCT doses are currently provided by different
manufacturers in different units. Currently, there is no way to assess the aggregate
radiation dose to a patient during a single procedure. Further, there is a need to
facilitate comparison of radiation doses to patients between a single run of CT to one
or several digitally subtracted angiography series. This situation needs to be
addressed, and a system should provide a means of not only comparing but also
consolidating doses from both fluoroscopy and CT. Furthermore, tracking and
reporting of radiation dose for a single patient should be made possible, as it is

126

Downloaded from ani.sagepub.com by guest on July 22, 2015

 Radiological protection in cone beam computed tomography (CBCT)

becoming increasingly important to do this for strengthening the processes involved

in the justification and optimisation principles of ICRP (Rehani and Frush, 2011;
Seuri et al., 2013). Errors for displayed dose estimates should not exceed 20%
(IAEA, 2011b; IEC, 2011; EC, 2012b). An RDSR can be used to report the modality
output following the existing RDSR in CT and angiography. KAP values for the
different orientation of the beam can be reported inside the RDSR when step-and-
shoot acquisition techniques are used; in addition, KAP can be considered for CBCT
in fluoroscopy and other applications to facilitate a direct comparison with doses
from 2D examinations such as fluoroscopy. Effective dose is not a suitable dosimetric
quantity for reporting patient doses.

A.6. Epilogue
(A41) Different methods for CBCT dosimetry have been presented. However, in
order to evaluate the usefulness of CBCT in regard to its alleged dose reduction in
comparison to CT, a metric which could be used for direct comparison is needed.
The unified CT dosimetry method proposed by ICRU (2012) has the potential to
standardise CBCT dosimetry. This method can be implemented without updating
the equipment already in use in the clinical CT arena. Furthermore, the methods
discussed could be used to measure dose for many types of different CBCT systems,
including radiotherapy CBCT, clinic-based systems, dedicated breast systems, and C-
arm systems. The value of CTDI-based measurements presented in this annex should
not be underestimated. Although CTDI has limitations, it has been evaluated on
many systems over the years, and provides important comparisons in output for CT
scanners from different manufacturers and ages. Also, the coefficients for patient
dose estimations that are available today are based on CTDIvol.

127

Downloaded from ani.sagepub.com by guest on July 22, 2015

View publication stats