TG 272 - O-Arm

Received: 8 April 2021 Revised: 7 December 2021 Accepted: 13 December 2021
DOI: 10.1002/mp.15429
AAPM SCIENTIFIC REPORT
AAPM Task Group Report 272: Comprehensive acceptance

testing and evaluation of fluoroscopy imaging systems
Pei-Jan Paul Lin1 Allen R. Goode2 Frank D. Corwin1 Ryan F. Fisher3

Stephen Balter4 Kevin A. Wunderle5 Beth A. Schueler6 Don-Soo Kim7
Jie Zhang8 Yifang (Jimmy) Zhou9 Peter A. Jenkins10 Usman Mahmood11
Teh Lin12 Hui Zhao13 Mi-Ae Park14 Annalisa Trianni15 Markus Lendle16
Andrew Kuhls-Gilcrist17 Jan C. Jans18 Lionel Desponds19 Gene Banasiak20
Steve Backes21 Carl Snyder21 Angela Snyder21 Minghui Lu22
Scott Gonzalez23
1
Department of Radiology, Virginia Commonwealth University, Richmond, Virginia, USA
2
Department of Radiology, University of Virginia, Charlottesville, Virginia, USA
3
Department of Radiology, The MetroHealth System, Cleveland, Ohio, USA
4
Departments of Medicine and Radiology, Columbia University Medical Center, New York, New York, USA
5
Department of Radiology, Cleveland Clinic, Cleveland, Ohio, USA
6
Radiology Department, Mayo Clinic, Rochester, Minnesota, USA
7
Department of Radiology, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts, USA
8
Department of Radiology, University of Kentucky, Lexington, Kentucky, USA
9
Department of Imaging, Cedars-Sinai Medical Center, Los Angeles, California, USA
10
Department of Radiology, University of Utah Health, Salt Lake City, Utah, USA
11
Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York, USA
12
Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania, USA
13
Department of Radiation Oncology, University of Utah, Salt Lake City, Utah, USA
14
Department of Radiology, University of Texas Southwestern, Dallas, Texas, USA
15
Medical Physics Department, Udine University Hospital, Udine, Italy
16
Siemens Healthcare, Forchheim, Germany
17
Canon Medical Systems USA, Inc., Tustin, California, USA
18
Philips Healthcare, Best, The Netherlands
Acronyms: AAPM, American Association of Physicists in Medicine; ACR, American College of Radiology; ADRC, Automatic dose rate control; AKR, Air kerma rate;
AP, Anteroposterior (anterior–posterior); AR, Ambience ratio (Lamb /Lmin ); CA, Caudal; CNR, Contrast-to-noise ratio; CR, Cranial; CT, Computed tomography; CVIA,
Cardiovascular-visceral interventional angiography; DAP, Dose-area product (see KAP); DICOM® , Digital imaging and communications in medicine; FGI,
Fluoroscopically-guided intervention; FOV, Field of view; FPIR, Flat panel image receptor; GI, Gastrointestinal; HT, High tension; HVAC, Heating, ventilation air
conditioning; HVL, Half -value layer (mm Al); IDRIR, Input dose rate to the image receptor; IEC, International Electrotechnical Commission; II, Image intensified, image
intensifier; IPEM, Institute of Physics and Engineering in Medicine; IR, Image receptor; IR, Interventional radiology; IRID, Image receptor input dose; IRP,
Interventional reference point (see PERP); IXE, Interventional X-ray equipment; Kair,PERP , Air kerma at PERP; KAP, Air-kerma-area product (also see DAP); kVanode ,
Tube potential measured at anode versus ground; kVcathod , Tube potential measured at cathode versus ground; KVD, kV detector; KVS, kV source; kVtotal , Tube
potential measured across anode and cathode; L’max , Total maximum combined luminance; Lmax + Lamb ; L’min , Total minimum combined luminance; Lmin + Lamb ; Lamb ,
Ambient luminance; LAO, Left anterior oblique; LINAC, Linear accelerator; Lmax , Maximum luminance; Lmin , Minimum luminance; LR, Luminance ratio; L’max /L’min ;
LUDM, Luminance uniformity deviation from the median; MPPG, Medical physics practice guideline; NEMA, National Electrical Manufacturers Association; Non-II,
NonImage Intensified; PA, Postero-anterior (posterioranterior); PACS, Picture archiving and communications system; PERP, Patient entrance reference point; PMMA,
Polymethyl methacrylate; PPS, Pulse per second; PSD, Patient skin dose; QC, Quality control; QMP, Qualified medical physicist; RAO, Right anterior oblique; Rd ,
Diffuse reflection coefficient; RDIM, Radiation dose index monitoring; RDSR, Radiation Dose Structured Report; RFIRC, Radiation field to image receptor
congruence; RP, Reference point (see PERP); SDD, Source-to-Detector Distance; SDNR, Signal difference to noise ratio; SID, Source-to-image receptor distance;
SIIM, Society for Imaging Informatics in Medicine; SMPTE, Society of Motion Picture and Television Engineers; SNR, Signal-to-noise ratio; SSD, Source-to-skin
distance; SSDS, Solid-state detector system; SSF, Spectral shaping filter; TJC, The Joint Commission
Med Phys. 2022;49:e1–e49. wileyonlinelibrary.com/journal/mp © 2022 American Association of Physicists in Medicine e1

24734209, 2022, 4, Downloaded from https://aapm.onlinelibrary.wiley.com/doi/10.1002/mp.15429 by Univ of Sao Paulo - Brazil, Wiley Online Library on [29/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
e2 AAPM TGR-272: FLUOROSCOPY IMAGING SYSTEM TESTS
19
General Electric Healthcare, Buc, France
20
Clinical Micro Systems, Arlington, Virginia, USA
21
Atirix Medical Systems, Inc., Minneapolis, Minnesota, USA
22
Varex Imaging Corporation, San Jose, California, USA
23
Food and Drug Administration, Health and Human Services, Silver Spring, Maryland, USA
Correspondence Abstract
Pei-Jan Paul Lin, Department of Radiology,
Virginia Commonwealth University, Richmond,
Modern fluoroscopes used for image guidance have become quite com-
VA 23298, USA. plex. Adding to this complexity are the many regulatory and accreditation
Email: pei-jan.lin@vcuhealth.org requirements that must be fulfilled during acceptance testing of a new unit.
Further, some of these acceptance tests have pass/fail criteria, whereas others
[Correction added on 4 Feb, 2022 after first
online publication: the title is updated.]
do not, making acceptance testing a subjective and time-consuming task.
The AAPM Task Group 272 Report spells out the details of tests that are
required and gives visibility to some of the tests that while not yet required
are recommended as good practice. The organization of the report begins
with the most complicated fluoroscopes used in interventional radiology
or cardiology and continues with general fluoroscopy and mobile C-arms.
Finally, the appendices of the report provide useful information, an example
report form and topics that needed their own section due to the level of
detail.
KEYWORDS
acceptance tests, fluoroscopy, interventional angiography
1 INTRODUCTION
The American Association of Physicists in Medicine 1.1 Disclaimer and precautions

(AAPM) Task Group (TG) 272 was established in
November 2015 with the goal of defining test- The tests described in this report are recommended
ing procedures for fluoroscopic imaging systems, for acceptance testing to be performed by a Qualified
including conventional, mobile C-arm, and interven- Medical Physicist (QMP), as defined by AAPM Policy
tional/angiography systems, thereby establishing a and Procedure 1-j (PP 1-j) and with the appropriate
comprehensive acceptance test procedure for prac- qualifications, as described in AAPM Medical Physics
ticing medical physicists. This test procedure would Practice Guideline 10 (AAPM Medical Physics Prac-
incorporate: tice Guideline 10.a.: Scope of practice for clinical med-
ical physics published in the Journal of Applied Clin-
∙ regulatory tests and measurements such as those ical Medical Physics [JACMP], 2018). However, many
described in the National Electrical Manufacturers of these tests are beyond what might be needed for
Association (NEMA) standard XR 27–2013, “X-ray routine testing, and therefore, should not be considered
Equipment for Interventional Procedures User Qual- as recommended for routine evaluations. Furthermore,
ity Control Mode,1 ” the contents and tolerances described in this report
∙ Current applicable International Electrical should not be used as the basis for establishing regula-
Commission (IEC) standards (IEC 60601-2- tory or standards requirements for quality control (QC)
54:2009+AMD1:2015+AMD2:2018 CSV, consoli- testing.
dated version),2 and In the United States, each state has regulatory author-
∙ image quality assessment accounting for new tech- ity over X-ray imaging system performance. As such,
nological advancements in fluoroscopy equipment there is a wide range of regulatory requirements that
design. may or may not be covered within this document. All
AAPM TGR-272: FLUOROSCOPY IMAGING SYSTEM TESTS e3
physicists performing regulatory testing of fluoroscopes 1.3 Fluoroscopy systems included in

must be sure to follow the requirements of the appropri- this report
ate regulatory authority.
Digital image receptors are highly sensitive to sharp Previous Task Group efforts on fluoroscopy sys-
edges of radiation intensity in the radiation field. For tem evaluations and imaging equipment testing have
many test devices, including radiation detectors, phan- focused on cardiovascular and visceral interventional
toms, and instruments, it is strongly recommended angiography systems, or interventional X-ray equipment
that protective attenuators and/or proper collimation be (IXE) according to IEC terminology, because of the
provided for prolonged and/or repetitive exposures. If mechanical sophistication of these systems and the
such protection is not provided, permanent damage to complexity of their operation logic and electronic design.
the image receptor can occur. To this end, testers should This report addresses IXE as the starting point in Sec-
not allow a bare unattenuated beam to occur adjacent tion 2; however, the contents are expanded in the fol-
to highly attenuating objects such as lead sheets or lowing sections to address all other fluoroscopy imaging
lead-lined detectors. At a minimum, depending on the X- systems, including:
ray energy under investigation, 3–5 mm of copper (Cu)
sheets or 5–10 cm of polymethyl methacrylate (PMMA) ∙ conventional fluoroscopes such as those employed
plates should be used if the lead or other highly attenu- in upper/lower gastrointestinal studies, with the X-
ating material does not adequately cover the entire sur- ray tube assembly installed beneath the examination
face of the image receptor. table;
∙ over-table X-ray tube fluoroscopes that may be
operated remotely and may be used not only for
1.2 Acceptance testing versus upper/lower gastrointestinal studies but also for appli-
commissioning of fluoroscopy systems cations such as cystoscopy, endoscopy, and bron-
choscopy;
In the diagnostic radiology environment, imaging sys- ∙ standard mobile C-arm (or full-size C-arm and Intra-
tems are often preconfigured and programmed at the operative 2D/3D) fluoroscopes employed in various
factory for the intended application of the system. Appli- disciplines of medicine;
cation specialists from the vendors then visit the instal- ∙ mobile mini C-arm systems typically used for imaging
lation sites and fine-tune the examination protocols of extremities; and
based on the preferences of the site’s physicians. This ∙ on-board imaging systems (OBIs) of therapeutic
fine-tuning process, in which imaging parameters are accelerators.
adjusted iteratively as users become more familiar with
the new equipment,is part of the commissioning of a flu- To avoid repetitive explanations, descriptions of items
oroscopy system. TG 272 was not tasked with address- involved in testing IXE systems are provided in Sec-
ing the commissioning process, as it requires intimate tion 2; thereafter, these items are not described again
knowledge of the medical procedures involved and infor other systems. In addition, items unique to a specific
depth experience regarding specific models of imaging fluoroscopy system type are described in their respec-
equipment. tive sections. Additionally, boxes headed with “NOTE”
It is highly recommended that acceptance testing be appear throughout the text to provide further descrip-
performed before the imaging equipment is released for tions, delineate exceptions, or offer supplemental expla-
clinical use in patients. nations.
Acceptance testing refers to the evaluation of a new This report is written to provide practical guidance
piece of imaging equipment, and it is highly recom- regarding acceptance testing and periodic required reg-
mended that the evaluation occurs prior to clinical use. ulatory evaluation of fluoroscopy systems. A suggested
Typically, this process consists of tests designed to acceptance testing report is therefore presented as a
ensure the imaging equipment performs as expected, guiding document, and the physical parameters to be
according to manufacturers’ specifications and regula- tested and measurements required are discussed in
tory requirements. This includes evaluation of geometric detail when deemed necessary. However, a minimal
accuracy of components (Section 2.2), radiation output amount of instruction is provided regarding, for instance,
(Section 2.3), image quality (Section 2.7), monitor per- how to use the radiation probe-electrometer system
formance (Section 2.5), and basic safety features and to measure tube voltage and half -value layer (HVL).
includes both quantitative measurements and qualita- Additional information is provided in user manuals and
tive analyses. Furthermore, periodic equipment testing basic radiation physics textbooks. In addition, AAPM
should take place at regular intervals specified by the TG 238 has been tasked with addressing the rotational
state after the equipment has been introduced into clin- angiography aspect of fluoroscopic operation; thus, this
ical use. Periodic testing generally consists of a subset specific operational capability is not discussed in this
of tests included in acceptance testing. report.
1.4 Sample form The system microenvironment should be evaluated

for completeness and readiness for operation. The med-
An example data collection form is attached as ical physicist may want to review the report prepared
Appendix A. The sample form is only an example, and by the installation engineer describing the conditions
may be used as a starting point for data collection. The of the microenvironment in which the IXE is located.
form is generic and not comprehensive for any par- The microenvironment includes the areas surround-
ticular fluoro system. Fluoroscopy systems differ sub- ing the unit that may affect optimal operation. Many
stantially from one model to the next and are equipped IXE systems have generator cabinets installed in sep-
with various operating modes with extensive “tree”struc- arate rooms; temperature and humidity values should
tures based on the intended operation or examination be evaluated in both the system and where generator
type. Therefore, the sample form should by no means is installed. The heating, ventilation, and air conditioning
be considered a complete set of data to be acquired system may need to be adjusted if temperatures are out-
or used for analysis. The first two pages of the sam- side of the system operation range specification. Stray
ple form also include a summary of test results and magnetic fields should also be identified if the imag-
equipment-specific information such as serial numbers ing unit is near a magnetic resonance imaging system,
and available field size; when these data are readily vendor installation documents will typically provide max-
available, system performance and specifications can imum magnetic field tolerances.
be quickly assessed, and multiple similar units can For newly installed units, the assumption is made that
be compared. Additionally, the specifics of geometrical the required state and local codes regarding shield-
setup and selected system parameters used for testing plans and specifications have already been satis-
ing are recorded on the sample form. These factors are fied; however, a post build radiation protection survey
important to document, as changes in setup can lead may be required to ensure adequate shielding. The ven-
to differences in longitudinal test results, even with no dor may also supply an iso-dose map to illustrate scat-
changes to the underlying system performance. ter/stray radiation from the unit. This information can
also be checked to ensure proper assembly from a radi-
ation safety standpoint. The vendor’s installation manual
2 INTERVENTIONAL X-RAY should be consulted for further details.
EQUIPMENT Two additional pieces of information regarding
mechanical factors should be highlighted. First, vibra-
2.1 Mechanical factors tions caused by external mechanical and electrical
equipment installed in the environment must be consid-
At the time of equipment installation, the examination ered. If image quality issues are found during accep-
table/pedestal assembly is mechanically aligned with tance testing, characteristics of the microenvironment
the C-arm gantry. The alignment of the C-arm gantry should be considered, including vibrations or stray mag-
and the examination table affects all other device set- netic fields as IXE systems are often placed in areas
tings such as lateral plane alignment, as well as posi- near large heating, ventilation, and air conditioning sys-
tioning of auxiliary equipment such as video monitors tems; air handler units; elevators; pumping stations; and
and power injectors. so on. These building support utilities may transfer
unwanted vibrations to the IXE system through the floor
(table- and floor-mounted systems) or through the ceil-
ing (ceiling-suspended C-arm systems). Such vibrations
NOTE 1: The mechanical alignment should be evalu-
can be detected by placing a vessel containing a small
ated at the time of acceptance testing; further assess-
amount of water on the system and looking for waves in
ment of this alignment may not be required on an
the water surface.
annual basis. The mechanical alignment should also
If vibrations are of great enough magnitude, they
be evaluated whenever the C-arm is involved in a col-
may interfere with or add noise to the imaging chain.
lision that may have resulted in mechanical misalign-
Figure 1 provides an example of a line-pair phantom
ment and whenever corrective maintenance is per-
imaged using the digital subtraction angiography (DSA)
formed to mechanical components associated with the
function on an IXE system. Figure 1(a) shows a DSA
C-arm and the examination table/pedestal. Therefore,
image obtained in the absence of mechanical vibration,
the testing result may simply be “satisfactory” or “mis-
whereas Figure 1(b) shows a DSA image obtained in
alignment correction required.” Readers should create
the presence of a small vibration caused by an exter-
their own forms for the evaluation of mechanical align-
nal mechanical or electrical source. In the image with
ment especially when a misalignment is found. The
the vibration, the misregistration of the mask and sub-
issue can be as simple as a minor adjustment, or may
sequent subtracted image is imperfect, causing a mis-
be more complex, requiring a reset of the entire equip-
match in the subtraction.
ment layout.
∙ the X-ray source-to-image receptor distance (SID),

∙ the X-ray source-to-patient entrance distance,
∙ the longitudinal position of the patient relative to the
X-ray source (patient left/right),
∙ the lateral position of the patient relative to the X-ray
source (patient head/foot), and
∙ the manufacturer-specified patient entrance refer-
ence point (PERP) dose location between the source
and image receptor.
The accuracy of the positional parameters listed

F I G U R E 1 DSA images obtained in the absence of vibration
above depends on proper alignment of the table with
(a) and in the presence of vibration (b). Note that the line pairs are
more visible in the image on the right (b) because of errors in the the center of rotation of the C-arm. Methods for verify-
registration of the subtraction process caused by the vibration ing the accuracy of this alignment are provided in the
following sections and in Appendix E.
Because of this potential for image deterioration, any
detected vibrations should be mitigated with dampeners
or other shock-absorbing devices. 2.1.1 Alignment of the examination table
Second, a geometrical coordinate system as defined with the C-arm gantry
by Digital Imaging and Communications in Medicine
(DICOM) should be considered (Figure 2).3 According 1. Ensure that all movable parts of the pedestal and
to DICOM convention, the long axis of the examination tabletop (patient examination table) are at their
table (Z-axis) is defined as the lateral direction (Fig- default “park” location. In other words, all motions
ure 2a), and the X-axis is defined as the longitudinal available should be set such that the tabletop (and
direction (Figure 2b); the white arrows in the figure indi- pedestal if movable) is perpendicular to the primary
cate the positive direction.The Y-axis corresponds to the axis of rotation and the angulation is set to 0◦ . Refer-
table-up (negative) and table-down (positive) motions. ring to Figure 3, at the head-end of the tabletop, mea-
The origin (X,Y,Z) = (0,0,0) is located at the isocenter sure the tabletop width “W,” at two different locations
of the C-arm gantry. (P1 and P2 ) about 20 cm apart. Determine and mark
The signs of the primary and secondary rotations for the centers as P1 and P2 . Draw the centerline of the
the positioner (the C-arm gantry; often identified as the examination tabletop connecting P1 and P2 as shown
A-plane) are also shown in Figure 2. C-arm rotation in Figure 3(a) where WH refers to the width mea-
about the X-axis is defined as secondary rotation; C-arm sured at head end. A metal bar (approximately 5 mm
rotation about the Z-axis is defined as primary rotation. in diameter and 40 cm in length) is affixed to the cen-
The direction of positive rotation is a left-handed turn for terline with adhesive tape as shown in Figure 3(b).
secondary rotation and a right-handed turn for primary Similarly, repeat this process at the foot end of the
rotation, and the 0◦ is in the negative Y direction, or at table. Draw a line connecting P3 and P4 where WF
12 o’clock. These geometrical parameters are recorded is the width measured at the foot end of the tabletop
in the DICOM header for each acquired image, as well (Figure 3c).
as in the Radiation Dose Structured Report (RSDR) for 2. Move the tabletop to a location typically employed for
each acquired image and for each fluoroscopic pedal or imaging the patient (X,Y,Z) = (X,Y,ZP ).
button press (also known as an “event”) during the pro- 3. Elevate the tabletop close to the gantry rotation cen-
cedure. tral axis of (X,Y,Z) = (X,0,ZP ). With the positioner pri-
The geometric parameters recorded in the DICOM mary angle set to 0◦ , step on the fluoroscopy pedal.
header are critically important when using either As the C-arm gantry rotating about the Y = 0 central
manufacturer-provided or third-party software to calcu- axis, observe that the metal bar on the display moni-
late peak patient skin dose (PSD). An understanding tor is located in the center of the image as shown in
of the patient position on the table and precise table- Figure 4.
gantry positions is needed to determine PSD. These 4. Pan the tabletop in the XZ-plane such that the metal
table-gantry positional parameters include: bar appears to be stationary in the center of the dis-
play monitor during C-arm rotation, refer to Figure 4.
∙ the orientation of the patient on the examination table, The metal bar should be located at (X,Y,Z) = (0,0,ZP ).
∙ the primary angulation of the X-ray source to the left Conversely, the table location indices should also
or right of the patient, show (X,Y,Z) = (0,0,ZP ). If this is not the case, the X-
∙ the secondary angulation of the X-ray source in the and Y-values must be recalibrated or readjusted by
cranial or caudal plane of the patient, the service engineer. Lock the tabletop from moving
F I G U R E 2 The definitions of DICOM geometrical coordinates for the lateral (a) and longitudinal (b) directions are depicted. The Cartesian
coordinate system, including the positive direction of each axis, is also shown. Note that lowering the examination table is considered a move in
the positive direction (+Y). (Adapted from DICOM PS 3.3 2014c Part 03, Section C8.7.4.1)
F I G U R E 3 Determine the center line.

Measure the width “W” of the tabletop at the
head end at two locations, say 20 cm apart.
Mark the center points P1 and P2 and draw a
straight line. Similarly, connect center points
P3 and P4 and draw a straight line.
(Alternatively, draw a straight line connecting
P1 and P4 ). Therefore, P1 , P2, P3 , and P4 are
on one straight line running entire length of
the examination tabletop
in X- and Y-directions with the exception in the This alignment evaluation procedure may be modi-
Z-direction. fied appropriately to match the mechanical configura-
5. Move the tabletop in the positive Z-direction until tion of the system such as a ceiling-suspended C-arm
the tip of the metal bar placed on the tabletop or ceiling-suspended tabletop.
gently touches the gantry. The table location is at
(X,Y,Z) = (0,0,ZHMax ). ZHMax signifies the tabletop is
at the furthest extent of the Z-axis at the head-end. 2.1.2 Alignment of the lateral plane
The tip of the metal bar should be pointing to the cen- (B-plane) with the C-arm and examination
tral axis of rotation, that is, the mechanical rotation table
axis of the gantry. Using the laser as a guide, mark a
sign indicating the location of the mechanical rotation ∙ For a biplane system, the lateral plane may be locked
center of the C-arm. (See Figure 5.) at 90◦ to the A-plane at (X,Y,Z) = (0,0,Z). Rotate the
6. Place a self -leveling crossline laser alignment device entire system (both A- and B-planes together), and
at the foot end. The vertical laser beam should be observe in both display monitors that the metal bar
aligned to the centerline of the tabletop. is located in the middle of images from both planes.
7. Move the tabletop to the furthest point towards the ∙ Depending on the mechanical configuration of the
foot-end (X,Y,Z) = (0, 0, ZFMax ). ZFMax signifies the system, the ceiling-suspended C-arm (often refer to
tabletop is at the furthest extent of the Z-axis at the as L-arm due to its shape) may be designed to
foot end (see Appendix E for details). It should be function as a second primary positioner and can be
pointed out that a very long stiff bar can be employed employed in tandem with the main primary positioner.
in place of the laser alignment device. In these systems, the centerlines of both the A- and
F I G U R E 4 Photographs of the display monitor showing the metal bar is in the middle of the image (in PA projection on the left) and rotated
90o (in LAT projection)
neurologic and visceral systems, the installation con-

figuration of the X-ray tube and the image receptor is
reversed. Because the anatomy of interest is either cen-
trally located or spread over the length of the body, it is
advantageous to install the imaging chain in such a way
as to minimize scattered radiation to the fluoroscopist.
NOTE 2:There are biplane IXE systems in which the

lateral plane can be mechanically flipped/rotated about
the Y-axis to accommodate the clinical needs of all
cardiovascular interventional examinations.
F I G U R E 5 On the left, the metal bar is pointing to the center of 2.1.3 Verification of the proper functioning
mechanical rotation. The crosshair of the laser beam is offset in of mechanical parts and safety mechanisms
height due to the self -leveling laser beam of 40 mm from the tabletop
surface. Note that the vertical laser beam is connecting all All mechanical parts should be evaluated to ensure that
measurement points P1 , P2 , P3 , and P4 running through the
centerline of the tabletop
they operate smoothly over the entire range described in
the installation drawings and to ensure that any installed
detents and locks are functioning properly.
B-planes should be aligned such that the metal bar
appears as shown in Figure 4 for the A-plane. For the ∙ Typically, the C-arm (A-plane positioner) is equipped
B-plane, as an LAT projection in PA projection geom- with primary and secondary axes (as shown in Fig-
etry, the image will be displayed horizontally. However, ure 2). The range and accuracy of available angula-
the image of the B-plane may be rotated 90O and tion depend on the configuration of the X-ray tube
displayed vertically similar to A-plane is presented. assembly and the primary imaging axis. The colli-
Then, the metal bar will appear as one contiguous bar sion avoidance safety switches must be tested during
across two monitors. both single-plane operation and biplane operation.For
a ceiling-suspended single-plane system, the entire
Note that the X-ray tube assembly is commonly movable range of the C-arm should be tested on both
installed on the right side of the patient in cardiac sides of the examination table.
catheterization systems to minimize geometric magni- ∙ The L-arm (B-plane positioner) is equipped with angu-
fication of the heart. Conversely, the X-ray tube assem- lation capabilities corresponding to the primary axis
bly is installed on the left side of the patient in neuro- of rotation (as shown in Figure 2a) and rotation of the
logic and visceral angiography systems so that the fluo- L-arm as a whole about the Y-axis. If the L-arm can
roscopist receives less scattered radiation. For biplane be flipped, the coordinates and signs displayed (+X
and –X) should be carefully recorded for application (see NOTE 3A), and the SID is set to the maximum avail-
and/or correction for peak PSD calculations. able distance for the imaging system.Manufacturers can
∙ Depending on the design of the system, the table- specify an alternate position for the PERP, so the oper-
top may offer various motions in linear translation ator’s manual should be consulted.
and rotation about all three axes. All available table
motions and locks should be tested for proper func- NOTE 3A: The term “interventional reference point”
tion. (IRP) has been renamed by the IEC as “PERP”
∙ The video display monitor(s) (large format or multiple (IEC 60601-2-43).2 “PERP” is also used in Interna-
monitor assembly) must be assessed to ensure that tional Commission on Radiological Protection (ICRP)
the monitor(s) can be moved without too much effort reports and on the International Atomic Energy
and that the range of motion allows for proper viewing Agency (IAEA) website.
from either side of the examination table. When sta-
tionary, the entire assembly should not sag down or NOTE 3B: To be consistent with IEC 60601-2-432 and
rotate and should remain stationary yet allow for easy National Council on Radiation Protection and Mea-
movement. surements (NCRP) Report No. 168, the SID is defined
∙ Any ancillary equipment installed on overhead arm(s) as the source-to-image receptor housing distance
(e.g., surgical lights, power injector) must be evaluated rather than the source-to-image receptor flat panel
to ensure that it can be easily moved and yet remain distance (Figure 6).
stationary when released.
∙ Protective radiation shields must be installed such To evaluate the radiological parameters described in
that they are within the active range of motion of the the following sections, it is best to use manual mode
fluoroscopists, can be employed from either side of unless physics mode or user QC mode is provided within
the examination table (if so designed), and will remain the end-user mode, as specified in NEMA XR 27–2013
in a fixed position without movement or sagging when standards and guideline publications.1
released.
∙ Proper functioning of the 5-min cumulative imaging
time warning alarm must be verified 2.2.1 Tube potential (kV)
For guidance regarding the assessment of tube poten-

2.2 X-ray generator and radiological tial in fluoroscopy mode, see Appendix A, “Fluoroscopy
parameter accuracy Tube Potential Accuracy.” For guidance regarding the
assessment of tube potential in acquisition mode, see
Radiological exposure parameters can be easily mea- Appendix A, “Acquisition Mode.” For these evaluations,
sured with most commercially available solid-state an accuracy of at least ±5% is expected; alternatively,
detector systems (SSDS). Detailed instructions regard- consult the manufacturer specifications.
ing the operation of these detector-electrometer sys-
tems are provided by the manufacturers. NOTE 4A: Although accuracy better than ±5% can be
Note that for many IXE systems in clinical operating attained, some state regulatory agencies specify that
modes, X-ray generator parameter settings are under ±10% accuracy is acceptable.
the control of the automatic dose rate control (ADRC)
NOTE 4B: It may seem redundant to assess tube
system, and the end user may not be able to man-
potential accuracy with both fluoroscopy and acquisi-
ually set parameters. For acceptance testing, genera-
tion modes. However, tube current loading is substan-
tor parameter tests should be conducted within service
tially different between the two operations. Thus, at the
mode to gain control of these radiographic parameters.
time of acceptance testing, tube potential should be
If the physicist is not familiar with service mode, the ser-
measured using both modes.
vice (or installation) engineer from the specific vendor
should be asked to assist.
During these test procedures, the image intensifier
(II) or the flat panel image receptor (FPIR) should be 2.2.2 Tube current-time product (mAs)
protected from excessive radiation unless the logic of and output linearity
ADRC mode is an essential part of the testing. A piece
of lead at least 1/8 inch (3 mm) thick should be used to Tests for tube current-time product and mA linearity are
cover the II or the FPIR. A specially designed protective accomplished concurrently; see Appendix A, “Acquisi-
piece may be available from the manufacturer for this tion Mode.” The output linearity over the available tube
purpose. potential range is evaluated with the small focal spot set
A typical experimental setup is depicted in Figure 6. at, for example, 40 mA and 100 ms (4 mAs) exposure,
Note that, for consistency, the SSDS is positioned PERP and the large focal spot set at, for example, 200 mA and
F I G U R E 6 Experimental setup for measuring radiological parameters. The radiation sensor in the SSDS is located at the PERP, which is
typically 15 cm from the isocenter toward the X-ray source. Depicted on the left is the anteroposterior-projection geometry, which is obtained by
flipping the primary gantry C-arm 180◦ from its normal position. On the right, the SSDS is turned upside down to face the X-ray source. Without
this setup, the attenuation caused by the examination table could alter the measured values. The support table (described in Appendix C) is
needed to keep the SSDS at the PERP. The table has an opening in the middle so that no tabletop or patient comfort mattress is located in the
primary radiation beam. Alternatively, this geometry can be reproduced by rotating the C-arm to the lateral position to avoid effects from the
tabletop. The location “P” is the patient entrance exposure measurement point, 30 cm from the image receptor front face plate
200 ms (40 mAs). In this case, the output measured will measured semi-invasively by connecting a storage oscil-
have a ratio of 4 mAs: 40 mAs (1:10). Note that the reg- loscope to the kV test point within the generator control
ulatory requirement for output linearity is applied to two cabinet. Those who are not familiar with semi-invasive
successive mAs stations,not two different focal spots.As and/or invasive testing should request assistance of the
shown in Appendix A, item 5, the output linearity should service engineers. (The use of test points in the gener-
show a tolerance better than ±10% based on the state ator control circuit is described in Appendix B).
regulation, although, ±5% is achievable. Commonly employed clinical pulse rates (frames/s)
The radiation output linearity is evaluated as depicted should also be measured for accuracy. Most SSDS are
in Appendix A, item 7. The average ratios of mGy/mAs capable of displaying pulse rate; alternatively, the pre-
obtained at any two consecutive mAs selector settings viously mentioned oscilloscope can be used. However,
typically do not differ by more than 10% of their sum for most SSDS connected to a laptop computer or propri-
radiographic equipment. etary display unit are capable of showing the pulses in
graphics or in a display window; thus, the pulse rate can
be easily determined.
2.2.3 Pulse width and pulse rate
In general, exposure time is one of the most accurately 2.2.4 Tube current (mA)
controlled parameters in a fluoroscopic system and can
be easily measured with noninvasive SSDS. Pulse width Measurement of the tube current is no longer con-
of greater than 10 ms may be measured with SSDS as a sidered necessary with the use of inverter-type high-
single exposure or as a train of pulses with varying pulse frequency generators, as these generators have a
widths and pulse rates. In Appendix A, item 2, exposure built-in high-voltage divider and are self -calibrated.
times of 100 and 200 ms are evaluated, with the results However, the tube current can be measured indirectly
shown in item 4. using the output linearity to obtain the relative mA
For pulse widths less than 10 ms, the temporal resolu- values. In the example described in Section 2.2.2, the
tion limitation of most SSDS will result in significant inac- exposure time was shown to be accurate, with a ratio
curacies. Alternatively, the pulse width can be accurately of 10:1 in mAs between the two focal spot settings.
Therefore, a tube current ratio of 5:1 is expected and is ogy. The total filtration is, in general, a “free” measure-
seen with this example using 40 and 200 mA. However, ment when using an SSDS; this parameter is often a
only the linearity is verified with this technique, not the provided output that can be recorded or compared, but
actual tube current value. no regulations or validity assessments regarding total
A commercially available clamp-on, induction-type filtration are available.
current measurement probe (mA probe) can be used
to measure current noninvasively. Using this mA probe
at the X-ray tube end of the high-tension cable is the 2.3 Radiation output measurements
preferred method for measuring current; however, this and SID tracking of ADRC
requires disassembly of the protective shell covering the
X-ray tube assembly,as the cables are normally installed Once generator-related tests are completed, the X-ray
such that there is no room to accept the mA probe. imaging system should be brought back to the clinical
If direct measurement of the tube current is neces- mode of operation so that ADRC is in full operation, sim-
sary, the measurement can be performed at the high- ulating actual clinical conditions.
tension transformer end of the high-tension cable. This Radiation output is typically evaluated under two dif-
requires opening the generator/transformer cabinet and ferent arrangements; those tests for maximum air kerma
locating the anode side of the high-tension cable. Once rate (AKR) and those that are geared toward assess-
again, those who are not familiar with semi-invasive ment of the automatic dose control logic.
and/or invasive testing should request assistance of
the service engineers. In general, however, tube current
measurement is not recommended or necessary for the 2.3.1 Maximum air kerma rate
reasons stated above.
The FDA’s maximum AKR limits are described as mea-
sured free-in-air, without the contribution of any scatter-
2.2.5 First HVL ing medium. The presence of scattered radiation from
the table, the table pad, or any added attenuating mate-
Single-shot HVL values measured and calculated with rial such as PMMA is likely to increase the measurement
an SSDS system are generally acceptable and may recorded by an ionization chamber or radiation sensor,
be used in lieu of those obtained with the traditional which could lead to erroneous failing measurements
method using an ionization chamber and Type 1100 alu- for maximum output. As such, maximum AKR measure-
minum (Al) sheets specified by the FDA. To meet the ments must be performed free-in-air with no additional
IEC standard, sheets of 99.9% pure Al must be used. material in the beam path, with the exception of attenu-
However, an SSDS HVL measurement is quicker and ation material to drive the ADRC to its maximum power
easier to obtain, especially on imaging equipment for loading.
which direct control of generator parameters is not pos- Maximum AKR can be measured using the setup
sible. If there are doubts about an SSDS HVL measure- shown in Figure 7. In this setup, the examination table
ment, or if the measured value does not meet or is close is rotated outside of the beam path. To conform to the
but not meeting the FDA specified value, confirmation free-in-air definition, the ionization chamber or radiation
of the SSDS value using the traditional method is rec- sensor is, in turn, extended off the edge of the exam-
ommended. Sample test results for HVL are depicted in ination table and positioned in the primary beam path.
Appendix A, item 6. For equipment manufactured after Alternatively, if the examination table cannot be moved
June 10, 2006, regulations specify that the HVL must be out of the beam path, the measurement can be obtained
at least 2.9 mm Al at 80 kV, for example.4 with the image chain in a horizontal orientation.
Note that the HVL should be measured using the From a regulatory point of view, the maximum radia-
imaging mode and protocol with the least amount of tion output must be verified at both maximum and mini-
added filtration (zero-added filtration is preferred). If this mum SIDs; additionally, depending on the measurement
information is not available,the IXE manufacturer should geometry, the measured radiation output must be cor-
be contacted to ensure that the minimum HVL measured rected for the location “P,” the patient radiation entrance
is valid for any of the imaging conditions available. point, 30 cm from the image receptor. The measured
maximum radiation AKR must be ≤88 mGy/min under
normal operation and ≤176 mGy/min under high-dose
2.2.6 Total filtration (mm Al) rate operation.5,6 For systems equipped with SID track-
ing, the maximum output measurements should be con-
When using SSDS, the total filtration displayed by the ducted at the maximum SID, minimum SID, and at least
SSDS is a calculated value. The total filtration results one additional SID to ensure that the maximum radia-
should be consistent (within ±0.5 mm) over the entire tion output at location P does not exceed regulatory limit
range of tube potentials employed in diagnostic radiol- irrespective of the SID.
F I G U R E 7 Setup for measuring maximum

output. “A” is the source-to-image receptor
distance and “B” is the source-to-radiation
sensor distance located at the patient
entrance exposure reference point. In both
geometrical arrangements, the ionization
chamber or radiation sensor is mechanically
suspended/supported from the examination
table at the PERP, whereas the SID is set to
the maximum available distance. “P,” the
patient radiation entrance point (or, the FDA
defined patient dose measurement point), is
located 30 cm from the image receptor. The
attenuation material (lead and/or Cu sheets)
can be placed on the image receptor housing
in the case of anteroposterior projection, as
shown on the left. For ease of setup, we keep
measurement system at the PERP; however,
the values must be corrected to 30 cm from
the receptor. Alternatively, the attenuation
material can be suspended from the image
receptor housing, as shown on the right
2.3.2 Assessment of the ADRC views (FOVs), including data at a sample FOV of 32 cm
diagonal (item 8). The default FOV for calibration may
The actual patient entrance air kerma (or AKR) should depend on the size of the installed FPIR. Additionally,
be measured under more clinically realistic conditions, various control parameters can affect the radiation out-
including automatic dose control logic enabled, normal put, including the pulse rate, fluoroscopy curve selection,
positioning of the examination table, use of a table pad, and operation mode (low, normal, or high). The numer-
and presence of back scatter from any phantom used. ous combinations of these factors will result in an enor-
In terms of acceptance testing, this report describes mous amount of data. However, the testing process can
free-in-air AKR measurements using varying thick- be streamlined by consulting the X-ray imaging manu-
nesses of PMMA slabs as the attenuation phantom and facturer regarding the system’s default configuration for
several pieces of 1-mm-thick Cu sheets. calibration. Most likely, a default FOV is selected at the
time of calibration, and the data (including the radiation
NOTE 5A: If the fidelity with clinical patient radiation output) are scaled according to the system design. Addi-
exposure is the primary aim, a radiotransparent ioniza- tionally, when high-output mode is selected, the pulse
tion chamber can be employed in place of the SSDS rate may be increased from, for example, 10 pulses/s
that may affect the automatic exposure control. This (pps) (in low or normal mode) to 15 pps as the system
is especially true if the sensing area(s) for the fluoro- default setup.
scopic automatic control cover the entire image area Because of these factors, it is advisable to first deter-
such that the radiation sensor cannot be positioned mine the primary application of the specific IXE system
outside the sensing area(s). To minimize the influence under acceptance testing and the type of procedure
of the lead backing (∼1.0 mm Pb equivalent) of SSDS, for which the system will be employed (e.g., neuro-
the radiation detector/probe should be placed outside logical angiography, visceral angiography, and cardiac
of the sensing area(s) for the fluoroscopic automatic catheterization). In the selection menu, the most likely
control. applications are shown when the system power is turned
on. Because it would be difficult and time-consuming to
NOTE 5B: Testing dose rates for recorded/acquisition evaluate every procedure programmed into the system,
imaging modes (cine, DSA, roadmap) is problem- two or three procedures associated with the systems
atic because of (a) the exceedingly high dose rates primary function may be selected for acceptance
(often >1000 mGy/min) and (b) the absence of regula- testing.
tory limits on these rates. Care should be taken during The geometry employed for this test is depicted in
these measurements to avoid excess X-ray tube heat Figure 8. On the left of Figure 8, the primary plane
loading and possible sensor “burn-in” on the FPIR. C-arm (A-plane) is rotated 90◦ to a lateral projection
to support the PMMA plates on the examination table.
To aid in the following discussion of ADRC measure- On the right, a homemade supporting table with an
ments, a sample form is provided in Appendix A. This opening in the middle is employed. Although either
form is based on a cardiovascular system with six field of geometry will work, the geometry using the supporting
F I G U R E 8 Experimental setup for measuring radiation output. In both geometrical arrangements, the tabletop and the table pad are not
included in the measurements. The distance “A,” the SID, is set to the maximum distance available for consistency. The distance “B” is the
source-to-isocenter distance (SICD). The isocenter axis “B” in this example had already been determined mechanically. The source-to-detector
distance “C” is located at the PERP for consistency and for convenience in placing the radiation sensor for measurements of patient entrance
dose rate. Note that use of an ionization chamber results in increased entrance surface dose rates due to backscatter. This is minimized if the
chamber is placed farther from the scattering material by moving PMMA closer to the flat panel image receptor
table is a more convenient option. For institutions with a stable combination of tube potential, tube current,
a relatively high number of IXE systems, the supporting pulse width, and SSF at certain key calibration points
table is an invaluable asset as a testing tool, as this (for example, at attenuation of 18–20 cm water equiv-
table can also be used with mobile C-arm equipment, alent). This is especially important for IXEs employ-
which may be coupled with a variety of patient tables. ing dynamic ADRC logic. This will result in an unsta-
A drawing of a sample supporting table is provided in ble dose rate registered by the radiation detector as
Appendix C. has been discussed in AAPM Reports 125 and 190.
Thus, TG 272 recommends PMMA be employed for
NOTE 6A: To minimize the influence of a radiation both acceptance testing and annual equipment eval-
probe on the ADRC, one can simply place the detec- uation of IXE systems. Aluminum and/or copper filters
tor outside of the sensing area. Alternatively, the sens- may be utilized when the fidelity of fluoroscopy curves
ing area can be selected from the control console or is not the primary concern.
table-side control box, if available, such that the radia-
tion probe and the sensing area are not superimposed.
If the sensing area covers the entire image receptor,
2.3.3 Radiation output as a function
the lead-backed SSDS may not be the ideal detector
of PMMA thickness
to use. In such a case, an ionization chamber is rec-
ommended, with the metalic portion of the ionization
AAPM Report No. 125 established that the use of
chamber (if so designed) placed outside of the FPIR
SSFs is common in high radiation-level fluoroscopy
when possible.
equipment.7,8 The selection of SSFs is based on the
NOTE 6B: PMMA is employed extensively in this fluoroscopic curves that, in turn, are dependent on
report. Aluminum and copper sheets are more conve- the hardware capabilities of the equipment, including
nient to transport and may be preferred for annual or the X-ray tube type and generator power rating. How-
periodic evaluation of IXE. It should be noted, however, ever, the applications specialist must assess the default
that the ADRC fluoroscopy curves are designed based programming of the fluoroscopic curves to determine
on “water equivalent” and/or PMMA to simulate patient the image quality preferences of the fluoroscopists
attenuation by the major IXE manufacturers whether involved.
it is using static or dynamic spectral shaping filters Evaluation of radiation output versus PMMA thick-
(SSFs). For acceptance testing, the TG strongly rec- ness ensures that the SSFs are functioning and that
ommends PMMA be employed to evaluate the ADRC the data obtained are suitable as a benchmark when
functionality. Additionally, use of aluminum (or copper) assessing the clinical functioning of the equipment in
causes the ADRC to oscillate as it attempts to find terms of the logic of fluoroscopy curve or trajectory.
2.3.4 Dose and dose rate at the PERP ify the displayed PERP air kerma against the physically
measured air kerma with the FOV set to a known area
The patient entrance exposure with backscatter can at a known distance from the radiation source. The inte-
be measured at the PERP, as shown in Figure 8, or at gral mode is selected to avoid the continuously changing
any other convenient distance from the X-ray source. dosimeter reading that occurs with rate mode measure-
Regardless of how the exposure is measured, the ment due to “digital jumping” of the FPIR control circuit.
geometrical information must be recorded to allow Digital jumping is caused by the operation logic search-
for scaling. Although the PERP dose and dose rate ing for a stable point (of tube potential in particular) on
are typically not the same as the patient entrance the fluoroscopy curve. This phenomenon is observed
exposure or patient dose, both are important data with the use of Cu or Al sheets as the attenuation mate-
points that can be used for patient dose tracking at the rial.
PERP and can also be used in conjunction with fluo- Although the previous report recommended compar-
roscopy patient dose monitoring and tracking software ing the displayed PERP air kerma with a measurement
programs. of integral air kerma, this can also be done by com-
paring the measured and displayed AKRs (Appendix A,
item 11). The rate mode method reduces the need for
2.3.5 Image receptor input dose and input an additional setup of measuring equipment and may
dose rate reduce the overall time for the inspection. In-air mea-
surement conditions must be met; hence, no backscat-
FPIR input dose is defined as the radiation that impinges ter can contribute to the measurement. If the radiation
on the image receptor as opposed to the patient; this probe of the SSDS radiation detector is placed at the
parameter is thus a better indicator of image quality than PERP, the PERP dose and dose rate can be determined.
PERP. Previous generation of IXE systems used con- The dose-area product, or the air kerma-area product
stant input dose (dose rate) logic, in which this report (KAP), can also be verified if the FOV at the PERP is
will hereafter refer to as signal-to-noise ratio (SNR)- known.
optimized logic (SNR optimization). The input dose to Going beyond the PERP air kerma, some manufac-
the FPIR,the fluoroscopy frame dose rate,and the acqui- turers have made use of the knowledge of equipment
sition frame dose rate were previously measured as part geometry and radiation output to create real-time dose
of acceptance testing or a study related to SSF.9 maps (dose mapping) that can be displayed during a
Over the past few years, the newest IXE systems procedure. Although PERP air kerma tallies the total
have been equipped with contrast-to-noise ratio (CNR)- tube output during a procedure, dose maps show the
optimized logic (CNR optimization), and the input dose spatial distribution of that tube output on a stylized
per frame has become a moving target. As the PMMA patient model to aid in spreading radiation skin dose
thickness is changed, the ADRC seeks optimum con- when possible.
trast for each specific condition of attenuation provided The IEC 60601-2-43:2010AMD2 standard published
by the PMMA. The FPIR input dose will therefore vary in 2019 standardizes the units of measurement a ven-
to maintain optimum contrast. Thus, FPIR input dose dor may choose from to display the radiation quantity
(and input dose rate) must be measured with extreme of air KAP.2 Additionally, the tolerance of the KAP accu-
care. There is no established procedure to properly racy is ± 35%. The three choices are Gycm,2 cGycm2
evaluate the relationship of FPIR input dose (and [numerically identical to μGym2 ], and mGycm.2 How-
input dose rate) to image quality on CNR optimized ever, as this standard applies specifically to fluoroscopes
fluoroscopy systems, nor is there a suitable phantom intended for use in fluoroscopically-guided interventions,
available on the commercial market for image quality other types of fluoroscopes are not subject to this stan-
assessment. dard and may choose other units of measurement for
It is therefore premature, at this time, for TG 272 to the displayed air KAP.
address any specific test procedure for evaluation of This same IEC standards also recommend that dose
CNR-optimized fluoroscopy systems. CNR optimization mapping capability should be available as part of the
is briefly discussed in Appendix D. basic safety and essential performance abilities of inter-
ventional angiography equipment.2 The IEC guidance
also recommends that the accuracy of any such dose
2.3.6 Displayed patient entrance maps should be evaluated during physics testing once
reference point air kerma verification standards are established and implemented by various
manufacturers. Because standards for dose maps do
The measurement and verification of PERP dose were not currently exist, it is outside of the scope of TG 272 to
described in detail in AAPM Report No. 190.10 This define appropriate acceptance testing for the accuracy
report recommended the use of integral mode to ver- of skin dose maps.
2.4 Collimator congruence test for visual verification. With each of these techniques, the
physical size of the radiation field can be compared with
The objective of the congruence test is to ensure that the image displayed on the monitor (Figure 9).
the X-ray field is appropriately sized and aligned with In the example shown in Figure 9, an attenuator (a
the image receptor. Good congruence and correct align- lead sheet in this example; ∼1/16 inch, 1.56 mm) is
ment not only prevent unnecessary exposure to patients attached to the FPIR to drive the X-ray generator to gain
but also reduce scattered radiation to medical personnel sufficient screen brightness for visualization of the radi-
while improving image quality. To this end, the central ation field. The fluorescent screen is placed in the pri-
ray of the X-ray beam should be aligned with the center mary beam, with lead BB shots employed as the scales
of the image receptor to minimize image cutoff and to (in cm). While stepping on the fluoroscopy exposure
avoid irradiating tissues that do not contribute to image pedal, the operator then aligns the fluorescent screen
formation. to the FOV (roughly 12 cm × 12 cm in this example), as
shown on the left side of Figure 9. The two images are
then compared to determine the alignment of the radi-
2.4.1 Congruence of radiation field size ation field to the image receptor. The congruence test
with the image receptor should show that the radiation field is no larger than 2%
of the SID in any direction and no smaller than the FOV
Because of potential mechanical sag of the C-arm, of the image receptor active size. The shadow of colli-
the congruence test should be performed using various mator blades or iris should be confined within the spec-
angulations and rotations of the C-arm with respect to its ified FOV.
primary and secondary axes. It must also be conducted The simplest congruence test involves ensuring that
on all available FOVs. the collimator blades are just visible within the image
Depending on the equipment design, the image field when the collimator is opened to its fullest extent
receptor may be rotated, with respect to the Y-axis as under all FOVs available in the imaging chain. However,
shown in Figure 2, to optimize the imaging area to fit the video mask must be deactivated for this test and this
with the positioning and shape of the anatomy. The col- test is only valid if the blades are visible. If the individual
limator may then be synchronized with the rotated FPIR collimator blade cannot be visualized, further testing is
to optimize the imaging area available, whether the col- needed to determine the amount of incongruence.
limation is rectangular or circular in design. The congru- Requirements regarding X-ray field limitations and
ence test should include measurements at the following alignment of fluoroscopic systems are available in the
projections: FDA resource manual for compliance test parameters
of diagnostic X-ray systems.12 Specifically, X-ray fields
∙ posteroanterior and anteroposterior, produced by fluoroscopic equipment without an II should
∙ 45◦ and 90◦ right anterior oblique (RAO) and left ante- not extend beyond the entire visible area of the image
rior oblique (LAO), and receptor; means are provided for stepless adjustment of
∙ 45◦ cranial and caudal. the field size. The minimum field size at the greatest SID
must be ≤5 cm × 5 cm. For fluoroscopic equipment with
Of the items listed above, the posteroanterior projec- an II, neither the length nor the width of the X-ray field in
tion is often the only geometry at which the congruence the plane of the image receptor should exceed that of
test is performed. For reducing the number of congru- the visible area of the image receptor by more than 3%
ence test measurement, the left or RAO may be com- of the SID. The sum of the excess length and the excess
bined with the cranial or caudal measurements. width should be no greater than 4% of the SID.
Many approaches to congruence testing have been For rectangular X-ray fields used with circular image
described. For example, a detailed test procedure using receptors, the error alignment should be determined
a cassette or film is described in AAPM Report No. along the length and width dimensions of the X-ray field,
74.11 The congruence test is best performed using a which pass through the center of the visible area of the
fluorescent screen or radiochromic film and rulers (or image receptor. Means are provided to permit further
plate) with lead markers to identify the radiation field size limitation of the field. For beam-limiting devices manu-
placed at a distance of the users discretion. Computed factured after May 22, 1979, and used in equipment with
radiography cassettes may be used in place of the fluo- a variable SID and/or a visible area of >300 cm2 , means
rescent screen for indirect visual verification of the radia- are provided for stepless adjustment of the X-ray field.
tion field for all FOVs. Also, there are commercially avail- Equipment with a fixed SID and a visible area of ≤300
able, stand-alone digital radiography detectors that work cm2 is provided with either stepless adjustment of the
very well for this purpose. There are commercially avail- X-ray field or with means to further limit the X-ray field
able devices that retain the detected radiation and use size at the plane of the image receptor to ≤125 cm2 .
LCD bars or long-latency fluorescent screens to allow Stepless adjustment should, at the greatest SID, provide
F I G U R E 9 Congruence test. The photograph on the left shows a fluorescent screen with rulers using lead dots and lead numbers for field
size identification. The last-image-hold fluoroscopy image is shown on the right. By comparing the corresponding four sides of these two
images, the congruence of the radiation field to the image receptor can be evaluated with the naked eye
continuous field sizes from the maximum obtainable before the images are sent to higher performance diag-
field size to a field size of ≤5 cm × 5 cm. nostic displays. Additionally, fluoroscopic monitors are
often used in brightly lit suites with high ambient light-
ing, which can have negative consequences on image
2.4.2 Centering of the central beam to the quality. Thus, care must be taken to ensure the consis-
center of the image receptor tent display of images.
The reader is referred to AAPM TG Report No. 27014
The alignment of the X-ray central beam with the center and its associated update15 for a full discussion of dis-
of the image receptor can be easily confirmed with the play characteristics. TG Report No. 270 recommends
use of two metal washers of the same dimension or measuring several key display characteristics, includ-
of a similar size, as described previously.13 A simplified ing minimum/maximum luminance, luminance response,
method is provided here: using two metal washers of and luminance uniformity; these characteristics will be
identical size, place one washer on the tabletop, and briefly discussed here. The American College of Radi-
attach the other to the center of the FPIR (see left side ology (ACR)-AAPM-Society for Imaging Informatics in
of Figure 10). Pan the examination tabletop to align Medicine Technical Standard for Electronic Practice of
these two washers under a geometry of approximately Medical Imaging can also be referenced for further
2× magnification with the smallest FOV available. If information.16 Although this report is not as detailed as
the radiation beam is centered with the center of the the TG report, its recommendations largely mirror those
FPIR, an image similar to that shown on the right side provided in the TG report and are likely to be updated
of Figure 10 should be observed. In addition, if the more frequently.
X-ray tube (insert) or the tabletop is not perpendicular Fluoroscopy-guided imaging (FGI) systems typically
to the image receptor, the washer placed on the tabletop include multiple monitors located in the control room and
will appear as an elongated ring, whereas the washer in the FGI suite itself. For acceptance testing, the fol-
attached to the FPIR will appear as a normal ring. lowing evaluations should be performed for all monitors
associated with a system to ensure consistent image
presentation.
2.5 Display monitors
Fluoroscopy display monitors installed for IXE systems 2.5.1 Minimum/maximum luminance
are a critical part of the imaging chain. The images and luminance ratio
displayed on the monitors for these systems are used
to make clinical diagnoses and to guide the placement The initial configuration of display operating levels
of needles, catheters, or other devices inside the body, should be based on the ambient luminance (Lamb ),
whereas modality monitors for other types of imag- which is the ambient light reflected from the surface
ing are primarily used for quality assurance of images of the monitor. Ambient luminance can be measured
F I G U R E 1 0 Evaluation of radiation beam centering with the FPIR. By using the smallest FOV available, a more accurate centering
alignment evaluation can be attained compared to a larger FOV. When the centering is set up correctly, the centers of the metallic washers
should form concentric circles in the middle of the image displayed on the monitor
directly with a light meter or can be estimated based on be considered. In these cases, consultation with physi-
the ambient illuminance and diffuse reflection coefficient cians regarding the clinical acceptability of the loss of
(Rd ) of the monitor. A recent study found an average low luminance contrast is advised. In addition, test pat-
Rd of 0.0038 cd/m2 per lux in a sample of large- terns such as AAPM TG18-AD or TG270-pQC can be
format displays used with modern FGI equipment.17 useful for evaluating the effects of ambient lighting on
The display’s minimum luminance (Lmin ) should ideally low-contrast detectability.18 If such loss of low-contrast
be set such that the ambience ratio (AR, defined as information is clinically unacceptable, possible solutions
Lamb /Lmin is < 1/4. Once Lmin is set, the total minimum include dimming the room lights, repositioning monitors
combined luminance (L’min ) can be calculated by sum- to reduce ambient reflection, or replacing monitors with
ming Lmin and Lamb . The maximum luminance (Lmax ) higher Lmax models.
can then be set, targeting an overall luminance ratio
(LR) of 350, with LRs between 250 and 450 deemed
acceptable. Note that LR is defined as L’max /L’min , 2.5.2 Luminance response
where L’max is the total maximum combined luminance.
Note also that on many fluoroscopic displays, options TG Report No. 270 also recommends evaluating the
for independent adjustment of Lmin and Lmax are luminance response of display monitors. The luminance
limited. response describes the relationship between individual
Although the TG 270 recommendations for AR and gray levels for Lmin and Lmax ; proper calibration ensures
LR are typically easily achievable in dimly lit reading adequate image contrast across the range of displayed
rooms, maintaining the recommended AR and LR may values. The DICOM grayscale standard display func-
be unachievable in the brighter environments typical of tion has been widely adopted in medical imaging and is
many FGI suites, leading to loss of image contrast in recommended.19 Depending on the test patterns avail-
darker regions of the image. In cases of high ambi- able, users may be able to test luminance response
ent illuminance, increasing Lmin in an attempt to main- compliance using the 18-point or 11-point response
tain an AR ratio < 1/4 if Lmax cannot be set high methods described in TG Report No. 270. Deviation
enough to maintain an acceptable LR is not recom- of <20% from the DICOM grayscale standard display
mended. Increasing Lmin could cause a loss of avail- function is recommended.
able contrast if room lighting is later dimmed. Differing In the absence of quantitative assessment of DICOM
ambient light preferences among operators must also grayscale standard display function compliance,
qualitative evaluation via test patterns may be used 2.6 Fluoroscopic dose monitoring
instead. Many test patterns are available for this and tracking
purpose, the most prevalent of which is the Society
of Motion Picture and Television Engineers pattern For patient procedures, recording of the air KAP or the
(SMPTE)20 that provides multiple grayscale levels and air kerma at the PERP (Kair,PERP ) is required by the
high- and low-contrast objects. However, it should be Joint Commission,21 not only to maintain a record of
noted that this pattern was designed to test properties of the fluoroscopy dose but also to track the total accumu-
older cathode-ray tube displays; because of grayscale lated dose to assess potential skin injuries. Enterprise-
insensitivity, this pattern is recommended for use with wide radiation dose index monitoring systems (RDIM)
modern flat-panel displays only if no other patterns are software22 makes compliance with this Joint Commis-
available. TG Report No. 270 includes the TG270-sQC sion requirement much more efficient and effective. It
and TG270-pQC test patterns, both of which provide is not within the scope of TG 272 to discuss the pros
objects to better test the luminance response across and cons of specific RDIM programs. However, regard-
the entire luminance range of a flat-panel display. If the less of the program used, key physics corrections must
TG270 test patterns are unavailable or unable to be be applied to estimate the PSD as accurately as pos-
loaded, the AAPM TG18-QC or TG18-OIQ test patterns sible. These key physics corrections must be applied
are recommended over the Society of Motion Picture to every event of fluoroscopic exposures in addition to
and Television Engineers pattern. Any pattern should the acquisition exposures. Note that because the expo-
be evaluated at typical viewing distances and in typical sure events in an IXE study, it can include hundreds of
ambient lighting conditions. Readers are directed to TG event entries. A RDIM program is an essential tool in this
Report No. 270 for more in-depth discussion regard- process; application of these correction factors is time-
ing qualitative test patterns and suggested pass/fail consuming if calculations are performed manually.
criteria.
NOTE 7: It is noteworthy to point out that, in the
July 2021 issue of the Joint Commission Perspec-
2.5.3 Luminance uniformity tives, the sentinel event definition and chapter have
been revised.23 For a broader scope of radiation
Luminance uniformity describes the variation in lumi- injury and dosimetry, readers are directed to National
nance output across the display area. TG Report No. Council on Radiation Protection and Measurements
270 recommends that users primarily focus on qualita- (NCRP) Report No.168, Radiation dose manage-
tive evaluation of local nonuniformities rather than on ment for fluoroscopically guided interventional medical
global quantitative uniformity, although for acceptance procedures.24
testing of new monitors, quantitative methods are rec-
ommended. TG Report No. 270 introduced the lumi-
To apply these corrections, the four values described
nance uniformity deviation from the median (LUDM)
in the following sections may be acquired at the time of
metric, which is easily calculated using standard pho-
acceptance testing. Additionally, further details regard-
tometer measurements from nine locations on a uni-
ing these parameters for calculating PSD, including the
form field. For cases in which a uniform test pattern is
addition of a composite factor that includes back scat-
not available, simple window width and leveling to pro-
ter factor (BSF), can be found in a report by DeLorenzo
duce a uniform display output can suffice. Displays with
et al.25 This paper also discusses methods for develop-
an LUDM > 30% should be considered for replacement,
ing fitting equations for various fluoroscopes.
and an LUDM > 15% should be investigated for possible
clinical effects.
Note that many large-format monitors commonly used NOTE 8A: RDIM programs available on the commer-
in FGI can be customized for various arrangements of cial market vary in their level of sophistication and may
displayed images. For the purpose of luminance unifor- or may not include the essential corrections required to
mity or test pattern observation, the most common con- accurately estimate PSD. The corrections described in
figuration should be used, as attempting to test all pos- Sections 2.6.1 and 2.6.2 are typically not built into the
sible arrangements of displays from all possible video current generation of RDIM programs; manual inter-
input devices may not be practical. Also note that many vention is therefore required to apply these corrections.
displays include protective covers that may be in place In the future, perhaps, these corrections will be built
during clinical use; these covers may alter the reflective into these programs.
or transmission properties of the displays. Display test-
NOTE 8B: Additionally, NEMA XR-27 standards1 pro-
ing conditions should therefore mirror clinical operating
vide TG 190 correction (see Section 2.6.1 below), in
conditions if protective covers are in place during clinical
the DICOM header so that the correction factor can
use.
be read by RDIM programs.
2.6.1 TG 190 dose correction/calibration

(CTG190 )
The correction factor obtained from the dose-rate mode

or the integral dose (as described in Section 2.3.6) is
needed to correct the displayed PERP dose (DPERP ) to
the externally measured dose (DMEASURED ). The dose
correction factor (CTG190 ) is defined as:
CTG190 = DMEASURED ∕DPERP . (1)
Empirically speaking, DPERP is often higher than

DMEASURED for most systems; thus, CTG190 is usually < 1.
Values less than 1 provide a margin of safety, and will
alert fluoroscopists earlier. The IEC standards allow for
a tolerance of 35% in the accuracy of the reported
DPERP ,26 but 10% to 15% (CTG190 = 0.87–0.91) over-
reporting is usually observed.
2.6.2 Examination table/mattress

attenuation correction (CTable and CMattress )
F I G U R E 1 1 Geometry for measuring the transmission ratio (TR )
To more accurately assess PSD, the attenuation caused
of the patient examination table and mattress. For measurements of
by the presence of the fluoroscopic examination table- air kerma dose for this TR calculation, the in-air dose (Dair ) is
top (CTable ) and the patient mattress (CMattress ) must measured in the absence of the examination table and mattress
be measured. The total correction (CTable_Mattress ) is the (moved out). DTable is measured in the presence of the examination
product of CTable and CMattress and can range from 0.6 to table, and DMattress is measured in the presence of the mattress
(moved in). DTable_Mattress is measured with both the examination
0.75 depending on the tube potential (kV), the SSF, and
table and the patient mattress in the primary radiation beam
the beam quality (e.g., mm Cu, mm Al).27 Hence,
CTable_Mattress = CTable × CMattress . (2) IXE systems employ SSFs whether using the dynamic
filter selection of the Seissl method29 or traditional static
For accurate patient dose assessment, CTable_Mattress filter selection, radiation beam qualities encountered in
should be measured with the examination tabletop and IXE systems should be evaluated along with the varying
the mattress initially as delivered and remeasured when tube potential.
either the tabletop or the mattress is replaced. The transmission ratio (TR[Table_Mattress] ) or correction
The total manual correction (CTotal_Manual ) that can factor (CTable_Mattress ) is defined as:
be predetermined at the time of acceptance testing is
therefore: TR(Table_Mattress) = CTable_Mattress
= DTable_Mattress ∕Dair . (4)

CTotal_Manual = CTG190 × CTable_Mattress
= CTG190 × CTable × CMattress . (3) And similarly:
Note that Equation (3) is correct only when the X-ray TR(Table) = CTable = DTable ∕Dair , (5)
beam is in the posteroanterior projection. Different cor-
rections must be applied when the projection is oblique,
lateral, or anteroposterior. TR(Mattress) = CMattress = DMattress ∕Dair , (6)
Depicted in Figure 11 is the test geometry for mea-
suring the attenuation of the examination tabletop and where DTable_Mattress is the air kerma measured with
mattress. This geometry is similar to the narrow-beam the examination table and the mattress together in the
geometry described in the American Society for Test- beam. DTable and DMattress are the air kerma values mea-
ing and Materials International designation F3094-14.28 sured in the presence of the table or mattress, respec-
However,the geometry shown here has been modified to tively. Dair is the air kerma measured in air with both the
meet the constraints of typical IXE equipment. Because examination table and the mattress positioned at least
TA B L E 1 Correction factors for TR through table and mattress
Tube potential (kVp)

60 70 80 90 100 110 120 Average
SSF (mmCu) 0 0.587 0.611 0.628 0.64 0.653 0.661 0.669 0.635
0.1 0.635 0.659 0.675 0.683 0.691 0.699 0.704 0.678
0.2 0.651 0.676 0.687 0.697 0.704 0.711 0.715 0.691
0.3 0.657 0.684 0.697 0.707 0.713 0.719 0.722 0.7
0.6 0.679 0.701 0.711 0.718 0.723 0.729 0.732 0.713
0.9 0.692 0.708 0.726 0.723 0.729 0.733 0.736 0.721
Average 0.65 0.673 0.688 0.695 0.702 0.709 0.713
50 cm from the primary beam (or as far from the beam as all events if the table height is kept more or less at a
the mechanical configuration of the IXE system allows). given location for the duration of a case.
Thus, the total TR(Table_Mattress) is defined as: Similar to TR , CGeo is also a function of the geometry
employed. Thus, CGeo = CGeo (X, Y, Z, θ, φ). Although
TR(Table_Mattress) = TR(Table) × TR(Mattress) the application of CGeo is a straightforward correction,
the process can be tedious, as the exposure event
= CTable × CMattress can include hundreds of occurrences. CGeo corrections
should be applied to all events.
= CTable_Mattress . (7)
For this assessment, the radiation field size is set to 2.6.4 FOV correction (CFOV )
10 cm × 10 cm at the radiation probe to simulate typi-
cal clinical conditions. TR is most easily measured using Even if tabletop panning is minimal and the radiation
the service mode or the end-user measurement mode beam entrance point remains the same, the FOV may
(physics mode) if available. If the service mode cannot be as large as 30 cm × 30 cm at the beginning of the
be accessed, the automatic feature must be locked to examination and collimated to 10 cm × 10 cm for most
obtain meaningful measurements. The technique lock of the examination duration. An FOV correction (CFOV )
feature of the ADRC must be employed, in effect, to dis- may need to be applied in conjunction with CGeo , and
able, the ADRC temporarily after the desired combina- CTable_Mattress may also need to be included depending
tion of tube potential and SSF is achieved. on whether the angles (θ, φ) caused the primary radia-
A selection of sample data obtained from an IXE sys- tion beam to traverse the table.
tem is shown in Table 1. Measurement of TR was per-
formed on 14 IXE systems using the Seissl method of
filter selection. The values listed are typical but do vary 2.7 Image quality assessment
with changes in imaging equipment or patient mattress.
Correction of the PSD calculation can be simplified by With the advent of new imaging technology such as
selecting the most typical combinations of tube poten- modern fluoroscopy curve programming and image pro-
tial and SSF (gray cells in Table 1). When only one cor- cessing that can be applied both during and after image
rection factor can be applied, based on Table 1, a cor- acquisition, a more rigorous approach to image quality
rection of (TR = ) 0.7 is suggested for the beam quality assessment is necessary.
correction.25 It should also be noted that TR is also a Physical parameters such as input dose rate to the
function of TR = TR (X, Y, Z, θ, φ) where θ is the primary image receptor, SNR, and CNR are of great interest
rotation angle and φ is the secondary rotation angle of when characterizing the imaging chain,but from a practi-
the C-arm. cal and clinical point of view, these parameters are more
suitable for troubleshooting image quality problems.
Image quality assessment on fluoroscopes poses a
2.6.3 Geometrical correction (CGeo ) substantial challenge for several reasons. First, there
are multiple operational modes that can and should
To more accurately assess PSD, a geometrical correc- be tested, including multiple dose levels of fluoroscopy
tion is required to account for panning of the patient and various subtypes of acquisitions such as cine,
in 3D orientation. Without an RDIM program, this task DSA, single-shot images, and cone-beam rotational
cannot be easily accomplished. In practical terms, the acquisitions. Second, clinical images are viewed live,
table up-and-down correction CGeo-Y may be similar for and processing often involves frame averaging from
multiple images. Viewing single fluoroscopic or acqui-

sition images can provide insight into image quality but
does not test the full clinical use of the system. Dynamic
live image assessment is needed, although equipment
may not have the capability to easily store and review
sequences of fluoroscopic images for quality evaluation
purposes. Third, the type of image that is provided for
analysis of image quality may affect the assessment.
Often, only “for presentation” image types are avail-
able, and these images have already been processed
or manipulated. Optimally, following the guidelines
proposed in NEMA standard XR 27–2013 will result
in “for processing” images that have not undergone
image manipulations other than detector uniformity
corrections.1 Finally, many of the image quality metrics
may not have benchmarks or passing values, so it
is imperative to check vendor literature for guidance.
Benchmarks can also be established at acceptance
testing, and trends can be monitored or compared from
year to year.
Although the above reasons are hurdles to assess-
ing image quality on fluoroscopes, the real challenge
lies in understanding how the image formation and dose
logic operation function. Without some knowledge from F I G U R E 1 2 Stack of PMMA slabs on patient support table,
the vendor, image quality assessment is a daunting task. used to assess image quality
Many vendors perform localized processing within each
image or frequency-based filtering such that a traditional
test object will confound the system or at least not pro- vice. Phantoms for image quality assessment can also
vide useful information.30 be the same as those used for radiation output mea-
The following sections describe the tools used for test- surements (as described in Section 2.3).Depicted in Fig-
ing image quality, the methods and rationale for some of ure 12 is a set of 35 cm × 43 cm × 2.5 cm PMMA slabs
the tests that use these tools, and examples of record- large enough to be imaged with the large FOV image
ing methodology and QC that should be considered dur- receptors that are used in interventional radiology.
ing acceptance testing and during monitoring of sys- Another type of phantom is multipurpose in design
tem performance thereafter. Note that many of the tests and can be used to test multiple image quality parame-
described may not have pass/fail criteria but should pro- ters simultaneously. These phantoms often have targets
vide useful information regarding how the system oper- for low-contrast determination, high-contrast assess-
ates and serve as reference information from year to ment, and dynamic range. Some of these phantoms can
year. Sample results from some of these tests may be also be used to assess both positive contrast (simulated
found in Appendix A. Finally, any thorough image quality iodine or barium) and negative contrast (air). The ben-
evaluation must also be accompanied by dose analysis. efits of these phantoms are that they can be used for
multiple tests and often include Cu as an integral part of
their makeup, ensuring that the system must respond to
2.7.1 Phantoms a load that is greater than air. Lastly, these phantoms can
be used in conjunction with additional patient-equivalent
Several types of phantoms are available for routine QC material such as PMMA or Cu to test systems under
testing and acceptance testing. Note that the vendor increased load.
may require the use of a specific phantom for routine In Figure 13, three well-known multipurpose phan-
physics testing or continual QC. toms are depicted: the IEC 61223-3-1 standard
One type of phantom used for image assessment phantom,31 the Leeds TOR 18FG phantom,32 and the
consists of simple sheets of Cu or PMMA. This type NEMA XR-21 phantom.33 All three phantoms are read-
of phantom can be used alone to identify artifacts or ily available from various vendors.
nonuniformities in the image or can be used in conjunc- The NEMA XR-21 phantom was originally developed
tion with other image quality phantoms to drive the fluo- for II-based imaging chains and was designed for use
roscopy equipment to clinical levels of operation. Some with analog systems.33 The central plate of the NEMA
manufacturers may even supply high-quality Al or Cu XR-21 phantom, which includes iodine-based low-
sheets (or a combination thereof) for testing or local ser- contrast objects and high-contrast line-pair resolution
F I G U R E 1 3 Examples of multipurpose phantoms. (a) Photograph of the IEC 61223-3-1 phantom. (b) Photograph of the Leeds TOR-18FG
phantom. (c) Photograph of the central plate of the NEMA XR-21 phantom
F I G U R E 1 4 Screen-captured images of the NEMA XR-21 phantom central plate both native and inverted. Corresponding iodine
concentrations of 200,100, 50, and 25 mg/cc targets are shown in the images
targets, is nevertheless still a valid QC imaging phan- of the AAPM DSA phantom.35 This artery block phan-
tom for digital FPIR systems. Design and fabrication of tom can be used to assess iodine contrast under flu-
new image quality phantoms were outside the scope of oroscopy or recorded imaging. The phantom includes
TG 272; however, the central plate of NEMA XR-21 was multiple lesions and aneurysms and multiple sizes to
consistently used during TG 272 assessments and was gauge detectability. As is the case with the other phan-
used successfully for multiple acceptance tests. Further toms described above, this phantom can be used with
research is needed to develop a phantom specifically additional PMMA to increase the machine output to clin-
designed to evaluate the CNR-optimized fluoroscopic ically relevant conditions.
imaging chain (see Appendix D). For this report, the cen- Finally, to assess motion, a rotating wheel or spinning
tral plate of NEMA XR-21 was used not as a tool to motion phantom can be used to visualize image lag or
assess performance but as a tool to test benchmark ghosting. These phantoms can be used for qualitative
consistency for future image quality comparisons. assessment, the results of which can be used as ini-
In the middle of the NEMA XR-21, central plate is a tial benchmarks during acceptance testing and for later
line-pair resolution target consisting of lead foil that is image quality optimization.
0.1 mm thick (0.6-line pairs/mm to 5.0-line pairs/mm res-
olution test patterns) (Figure 14). There are four pairs of
circular holes (4, 3, 2, and 1 mm in diameter) per set in 2.7.2 Methods of image quality
each of the four quadrants. These holes are filled with assessment
iodine concentrations ranging from 25 to 200 mg/cc, with
the concentration doubling in each object group. Image uniformity
Depicted in Figure 15 is an artery block phantom34 Image uniformity should be evaluated early in accep-
developed by Mayo Clinic to enhance the functionality tance testing to ensure that the FOV and detector are
TA B L E 2 Integral uniformity. Single-shot image using central ROI

off grid focus (SID = 95 cm) and at grid focus (SID = 105 cm)
Maximum Minimum Integral

SID (cm) pixel value pixel value uniformity
95 2069 1797 7%
105 2080 1876 5%
F I G U R E 1 5 A photographically enhanced image of the artery

block phantom. Note that the center circular aneurysm areas are
excellent spots for region of interest (ROI) placement to determine
the CNR for various programs and dose levels. Each blood vessel
section is filled with iodine, at concentrations of 6 mg/cm2 (top),
1.5 mg/cm2 (middle), or 3 mg/cm2 (bottom)
free from artifacts, dropouts, or nonuniformities cre-

ated during the installation process. Uniformity can be
assessed both qualitatively and quantitatively. A stack
of PMMA slabs large enough to be imaged near the
detector or Cu sheets (1 or 2 mm) placed on the X- F I G U R E 1 6 Screen capture of the IEC 61223-3-1–compliant
ray tube/collimator port should provide enough load to phantom showing the dynamic range available under an examination
program and predefined imaging parameters
the system to allow image uniformity to be evaluated.
Using ROIs, sections at the periphery can be evaluated
by simply comparing the mean pixel values in the periph- ducing a small amount of grid cutoff and/or nonunifor-
ery to those in the center. If the system does not have mity. Results in Table 2 are only for example and are not
the capability to measure pixel values, images can be considered pass/fail, because those values do not exist.
exported for measurements using DICOM-compatible This test can be performed on workstations or on most
viewer software. This uniformity test should be per- picture archiving and communications systems (PACS).
formed at several SIDs, including the minimum and max-
imum SIDs, to ensure that there is no effect from grid Dynamic range
cutoff. Most grids used in fluoroscopy utilize a grid ratio Dynamic range can be assessed using one of the mul-
with the focal length focused at the midpoint of the SID tipurpose phantoms shown in Figure 13 or using a step
range. A simple integral uniformity value of a one-frame wedge (as shown in Figure 16). Although these tests of
acquisition may be calculated quantitatively from image dynamic range may not have pass/fail criteria, checking
pixel data (entire image), as shown in Equation (8):
(Maximum pixel value) − (minimum pixel value)

% Integral uniformity = × 100. (8)
(Maximum pixel value) + (minimum pixel value)
Table 2 lists sample integral uniformity data obtained several organ programs may be useful to ensure that
at two different SIDs by adjusting the height of the FPIR the default settings are appropriate. Many of the multi-
first to the minimum SID of 95 cm (off grid focus) and purpose phantoms discussed earlier incorporate a step
then to the SID of 105 cm (at grid focus), thereby intro- wedge made of varying thicknesses of Cu as part of
TA B L E 3 HCR testing results from fluoroscopy and acquisition modes
Acquisition mode
FOV size (cm) 48 42 32 22 16 11

Focal spot size (mm) 0.6 0.6 0.6 0.6 0.6 0.6
Pixel size (mm) 0.154 0.154 0.154 0.154 0.154 0.154
Nyquist frequency (1/mm) 3.247 3.247 3.247 3.247 3.247 3.247
Subjective resolution (line pairs/mm) 2.2 2.2 2.2 2.2 2.2 2.2
Fluoroscopy mode
FOV size (cm) 48 42 32 22 16 11

Focal spot size (mm) 0.6 0.6 0.6 0.6 0.6 0.6
Pixel size (mm) 0.308 0.308 0.154 0.154 0.154 0.154
Nyquist frequency (1/mm) 1.623 1.623 3.247 3.247 3.247 3.247
Subjective resolution (line pairs/mm) 1.4 1.4 2.2 2.2 2.2 2.2
their test patterns. On today’s modern systems, visu- an example of the characterization of the imaging res-
alizing most if not all of the steps on these phantoms olution in both fluoroscopy and acquisition modes. Note
should be an easy task. To quantify the dynamic range, that the pixel size and focal spot size were determined
the number of unique range steps can be counted after assessment using data in the DICOM header. The
(under fluoroscopy mode or acquisition mode). This underlined values show a change in resolution or value
value is a useful metric to assess each year as long as observed on the in-room monitor due to a change in the
setup conditions are equivalent. effective pixel size. This is a result of pixel binning, where
data from multiple detector elements are combined to
High-contrast resolution improve noise characteristics of the image.
High-contrast resolution (HCR) can also be evaluated The table shows how the resolution changes with FOV.
using one of the multipurpose phantoms described In acquisition mode, on this unit, there is no apparent
above, the central plate described above, or a line-pair increase in resolution, whereas in fluoroscopy, from 42 to
resolution phantom, ideally using physics mode. Of all 32 cm, there is a change in pixel size (underlined values)
the image quality tests, the HCR test is most sensitive that may increase resolution capability.
to magnification, focal spot size, and pixel size, factors
that are integral to understanding how and why the HCR Low-contrast detectability
changes.For this test,note that care should be taken dur- Low-contrast detectability is typically used more as an
ing placement of the phantom to avoid aliasing with the optimization test than as an acceptance test. The vari-
regular pixel matrix. ety of phantom types and technical parameters avail-
At minimum, HCR should be assessed for the default able on the unit being assessed make this test difficult
calibration FOV to establish a benchmark for periodic to perform; thus, in the absence of vendor guidance or
QC measurement using the same organ program. The reference values for a particular phantom, the goal of
settings used by fluoroscopy and acquisition modes testing should be to simply establish baseline results for
are vastly different (e.g., focal spot, mA, spectral filter); future evaluations. Results of low-contrast detectability
therefore, different results are expected for the different are typically reported in a good/better/best format rather
modes. Furthermore, careful attention should be paid to than as pass/fail. For acceptance tests of units that pri-
the angle of the bar pattern such that aliasing is mini- marily use iodine as the contrast agent, the NEMA XR-
mized on the most visible targets. 21 central plate can be used for subjective assessment,
Table 3 shows HCR testing results for single-shot and the artery block phantom can be used for both sub-
acquisition and fluoroscopy modes using the NEMA XR- jective and objective assessment as the targets are large
21 phantom central plate placed in the middle of a enough for ROI analysis (see Figures 11 and 12).
20 cm (8″, inclusive of the XR-21 Phantom) of PMMA
under clinical conditions (same imaging system for both Image lag, ghosting, and motion
modes, abdominal organ program). Placing the phantom Testing single-image frames from fluoroscopy or single-
in the middle of the PMMA stack more closely approxi- shot acquisitions is useful when assessing image qual-
mates clinical conditions in the middle of the body. This ity. However, because fluoroscopy uses live images, tests
arrangement will yield a magnification of roughly 1.5 should also be performed on live images.Tests to assess
and, more importantly, is reproducible from year to year. moving image fidelity or the presence of lag (copies
The results in Table 3 are not pass/fail and are shown as of preceding images) need not be complicated and
F I G U R E 1 7 Images of the spinning wheel of the NEMA XR-21 phantom. These images were obtained at a fluoroscopic frame rate of 7.5
pps and a pulse width of 3.4 ms. (a) No lag. (b) Some lag. (c) Extreme lag. Note that the images showing some lag (b) and extreme lag (c)
demonstrate targets (lead numbers, lead BBs, and varying diameters of piano wires) with more blur or copies of the targets that are smeared
because of frame integration over a longer period than the short pulse duration (3.4 ms) of the individual fluoroscopy image (a)
should be reproducible from year to year. Phantoms with dose rate to the image receptor is straightforward, deter-
motion capability should be used; however, the phan- mining CNR (or signal difference-to-noise ratio) is more
toms should also be imaged while stationary to obtain challenging to illustrate the operation of fluoroscopes.
baseline values (Figure 17). Different results should be The measurements shown in Figure 18 are from existing
expected for static, pulsed fluoroscopy, and recorded QC phantoms or fabricated with readily available mate-
imaging data sets. rials at the time of testing. Details for saving and analyz-
Tests of motion may also assess the degree of ing fluoroscopy loops for QC or acceptance testing may
frame averaging in fluoroscopy. Often selectable on the be found in Goode et al.36
system, frame averaging may be different for differ-
ent organ/examination programs and may have a dra- System connectivity
matic effect on image quality perception in fluoroscopy. Acceptance tests of imaging systems should also
Although this motion test is only qualitative, a motion include assessment of basic interoperability. Evaluating
phantom even similar to the NEMA XR-21 phantom may image quality during acceptance testing often requires
provide useful comparisons of several fluoroscopic set- viewing images on PACS or connected workstations
tings, which could affect frame averaging and may there- to allow for image manipulation and postprocessing.
fore be a valuable tool for program optimization during Other downstream systems such as the worklist server
fluoroscope setup. or RDIM should also be tested for accuracy. Further
tests of interoperability (e.g., alignment of fields from
SNR and CNR the worklist, parameter display on PACS) and perfor-
In general, SNR and CNR are useful parameters to mance/speed should also be performed if any prob-
asses on digital imaging systems to gauge signal, noise, lems in system connectivity are identified.For more infor-
and contrast quantities. However, image processing that mation, readers should refer to AAPM Report No. 248,
is applied internally at the FPIR or downstream from which provides testing methodology.37
the FPIR makes obtaining these measurements diffi-
cult. Obtaining valid raw images for SNR/CNR mea-
surement and analysis requires access beyond that of 2.7.3 Quality control activities
the basic user level. Furthermore, image processing
affects the images presented on the display monitor, and A rigorous QC testing is recommended to ensure that
these processed images may not be suitable for accep- the system is performing consistently over time.36 A QC
tance testing or performance evaluation. For these rea- program should optimally be set up during acceptance
sons, SNR/CNR measurements are often not included testing before the system is turned over to the end users.
in acceptance testing. The manufacturer’s QC program should be followed,
If nonprocessed images are unavailable, a simplistic as well as any in-house QC programs that are used for
approach for evaluating SNR/CNR may still be applied similar units. QC often involves performing the motions
to readily available images to establish baseline values. of imaging before the clinical procedure is initiated to
Current SNR optimization ADRC methodology may use ensure that all of the pieces needed to perform the
a combination of Kair,PERP , input dose rate to the image procedure are in place. Some common mishaps that a
receptor, and localized pixel measurements of signal simple QC program may detect include spilled contrast
difference-to-noise ratio for different materials to con- on the table, removal and lack of subsequent reinstal-
trol the system. Although determining Kair,PERP and input lation of the antiscatter grid, and dead batteries in the
F I G U R E 1 8 The single static images shown in (a) and (b) are frames extracted from a fluoroscopy loop that can be analyzed to determine
the CNR under a predefined set of conditions. (a) Uniform background with a contrast target object to mimic Iodine. (b) Custom-fabricated
contrast detail phantom. (c) This plot shows the CNR between the two regions for each frame of a fluoroscopy loop obtained from (a),
demonstrating the increase in CNR and the steady state that is achieved a few frames later. The blackened squares indicate the frames used to
average and calculate a single CNR for the fluoroscopy loop36
wireless foot pedal. All of these issues are best discov- are confined to the inner ring. However, all mobile C-
ered before the patient is on the table. A QC program arm fluoroscopes have at least a primary and secondary
may also detect suboptimal imaging from the unit itself gantry rotation, with either manual or electronic lock-
if QC is performed in a reproducible way every day. ing mechanisms. These rotations should be inspected
For phantoms and devices used to drive the fluoro- to assess ease of movement and the ability of the lock-
scope (which range in complexity from a simple Al block ing mechanism to hold the gantry in a set position.
or Cu sheet to a phantom containing targets such as To assess mechanical motions for these systems, one
a contrast detail scheme), QC testing should focus on rotational locking mechanism should be unlocked at a
reproducibility and objectivity. time to verify that there is smooth and easy gantry move-
QC tests should also focus on the main use of the ment in each axis of rotation. The gantry should then be
imaging unit. For example, if the unit is used for fluo- relocked, and the lock should be checked to ensure that
roscopy 100% of the time, a single shot-based QC pro- it holds the gantry in a set position.
gram may not yield information relevant to the clinical The gantries of mobile C-arm fluoroscopes typically
use of the unit. At a minimum, tests should collect the include a visual analog indication of the gantry rotation
parameters for kV, mA, and SSF so that this information angle, which should also be assessed for accuracy. One
may be compared to baseline values established during exception is C-arm fluoroscopes used for 3D reconstruc-
acceptance testing. More detailed tests of image qual- tions, which have automated drive motors with refined
ity metrics can be performed, but these tests require the angular accuracy. These systems should be evaluated
use of additional online tools or advanced data collec- using the process described for stationary fluoroscopes
tion methods. capable of 3D image acquisitions in Section 2.
It should be pointed out that a spacer (cone) is often
provided to C-arm fluoroscopy systems to meet with the
3 MOBILE C-ARM AND minimum source-to-skin distance (SSD) requirement. It
INTRAOPERATIVE 2D/3D is suggested to ensure that the spacer is indeed avail-
FLUOROSCOPY SYSTEMS able and provide the minimum required SSD.
This section refers to both mobile C-arms and intraop-

erative 2D/3D fluoroscopy systems. However, in general, 3.2 X-ray generator and radiological
when the topics refer to both types of systems, C-arm parameter accuracy
may be used to mean both systems unless specific to
the intraoperative 2D/3D fluoroscopy systems. The generator of a mobile C-arm unit mirrors that of a
stationary C-arm unit; therefore, the tests for these units
will largely be the same, with a few exceptions. For ease
3.1 Mechanical factors of testing, mobile C-arms can generally be unlocked and
geometrically inverted from the typical clinical orienta-
Mobile C-arm fluoroscope gantries are generally manip- tion, resulting in the X-ray tube being situated at the
ulated manually, without the precise motor-controlled top of the C-arm with the image receptor at the bot-
movements that are used in fixed C-arm fluoroscopes, tom. Regardless of image receptor type, care must be
whereas Intraoperative 2D/3D mechanical movements taken to protect the image receptor from burn-in artifacts
are moved by motors and X-ray tube-detector motions during measurements. This is usually accomplished by
using lead sheets or multiple layers of protective gar- ∙ Place a minimal amount of an attenuating medium
ments. (e.g., 2–5 cm of PMMA) on the image receptor.
Generator testing of mobile fluoroscopes can often ∙ Orient the measuring device free-in-air around the pri-
be performed in either a fluoroscopic or radiographic mary beam axis of the C-arm gantry.
mode (when available). Although either operating mode ∙ Take a radiographic exposure.
can be used, radiographic mode is generally prefer- ∙ Increase the PMMA thickness and repeat the previous
able because of the higher tube output rates associated steps, ensuring that the kV is constant. Tube voltage
with this technique. Tube outputs at lower kV stations is likely to remain constant for acquisitions as long as
(<60 kV) may be too low to trigger noninvasive measure- fluoroscopy is not engaged.
ment devices, possibly necessitating a reduction in the ∙ Plot the mGy/mAs and ensure linearity to within
focal spot-to-device distance to obtain a measurement. ±10%.
Although nearly all mobile C-arms allow manual con-
trol of the tube potential (kV), tube current (mA) is often
automatically indexed to the kV setting and not individu- 3.2.3 HVL
ally adjustable, making linearity measurements difficult.
However, some models of C-arm systems do allow for Verify that the minimum regulatory required HVL is met
independent control of both tube potential and tube cur- using methods prescribed in Section 2.2.2.
rent. For C-arms without individually adjustable mA set-
tings, single-shot radiographic acquisitions sometimes
use an automatic exposure control feature that makes 3.3 Radiation output measurements
use of a fixed kV and adjusts the mAs according to
attenuation in the beam path. For these types of units, 3.3.1 Radiation output as a function of
exposure linearity can be measured by adding increas- PMMA thickness
ing amounts of attenuating material to the beam path.
Finally, if there are multiple buttons or pedals that can During acceptance testing, radiation output and varia-
initiate the X-ray beam, all buttons and pedals should be tions in technique parameters such as kV, mA, and/or
tested for proper functionality. pulse width can be measured as a function of increas-
ing PMMA thickness in various fluoroscopy protocols to
determine the operating characteristics of the mobile
3.2.1 Peak tube potential (kV) imaging system. This information allows examination
protocols and operating modes to be matched to spe-
The kV for mobile fluoroscopes should be evaluated cific clinical tasks and provides baseline data for future
at a minimum of three stations covering the span of troubleshooting.
clinically achievable values, generally from 50 to 60 kV The vendor or the user manual should be consulted to
through 110 to 120 kV. Measurements can be made in determine whether different examination protocols are
both fluoroscopy and acquisition (radiographic) mode driven by unique fluoroscopy curves or if they only indi-
if desired, although both modes should yield simi- cate changes in image processing, as this information
lar measurements. Displayed kV should correspond to may not be obvious to the user. On systems with a large
measured values within manufacturer specifications or number of available protocols, this test of radiation out-
within ±10% of the indicated kV.For more details regard- put need only be performed for protocols with unique
ing fluoroscopy mode, see Appendix A, Fluoroscopy fluoroscopy curves, as changes in image processing will
Tube Potential Accuracy; for more details regarding not change the acquisition parameters.
acquisition mode, see Appendix A, Acquisition Mode. This test should be performed using the procedure
described in Section 2.3.2. For ongoing annual testing
beyond acceptance, the entire test does not need to be
3.2.2 Tube current/mAs linearity repeated; typical patient entrance exposure rates can
be measured for selected protocols, frame rates, and/or
For mobile fluoroscopes that allow manual kV adjust- magnifications to verify constant performance over time
ment, the mAs should be evaluated at a minimum and to compare mobile C-arm systems.
of three mAs stations covering the span of clinically
achievable values. Calculated mGy/mAs values at the
various mAs stations should all be within ±10% of each 3.3.2 Maximum AKR
other.
For mobile fluoroscopes that do not have manually In fluoroscopic imaging mode, the maximum AKR is
adjustable mAs but use an ADRC system to determine defined for all mobile C-arms with a maximum SID
mAs for single-shot radiographic acquisitions, the tube greater than 45 cm at a plane 30 cm from the input of the
current/mAs linearity should be assessed as follows: fluoroscopic imaging assembly.5,6 The maximum AKR
should be tested free-in-air with narrow-beam geometry lined in Section 2.5, although adjustability of luminance
(using a well-collimated field) either in or corrected to levels for mobile monitors and the ability to load outside
the designated plane. The maximum AKR allowable by test patterns may be more limited on these mobile sys-
the FDA is 88 mGy/min in normal mode of operation and tems. All available monitors should be evaluated under
176 mGy/min in high-level or boost operating mode. Dur- expected operational ambient light conditions. Lmin and
ing testing, the continuous audible signal that is required Lmax should be measured, LR should be calculated,
to sound when air kerma limits exceed 88 mGy/min and both quantitative (LUDM) and qualitative uniformity
in high-level operating mode should be assessed for assessments should be performed.
proper function.
As with all fluoroscopic systems,there are no AKR lim-
itations in the acquisition or radiographic modes of oper- 3.6 Fluoroscopic dose monitoring
ation, although mobile C-arms generally limited in their and tracking
acquisition capabilities.
Dose monitoring and tracking are required for mobile
C-arm fluoroscopy systems. Attenuation factors of clin-
3.3.3 Dose and dose rate at the PERP ically used examination tables and mattresses can be
measured using the methods described in Section 2.6;
Follow the procedure described in Section 2.3.3. however, a given C-arm unit may use a variety of exam-
ination table/mattress combinations such that only a
general or average table/mattress factor can be deter-
3.3.4 Image receptor input dose mined.
and dose rate
On most mobile C-arm systems, the grid is fixed to the 3.7 Image quality assessment
detector behind the cover and cannot be accessed with-
out some level of disassembly.For this reason,measure- Image quality for mobile systems should be assessed
ment of the detector input dose and dose rate is not gen- as outlined in Section 2.7, using similar phantoms. At a
erally recommended. minimum, HCR and low-contrast detectability should be
evaluated, as required by Joint Commission standards.
3.3.5 Displayed PERP dose verification

3.8 Specific tests for intraoperative
Unless otherwise stated by the manufacturer,the TG 190 2D/3D fluoroscopy
PERP for all full-size mobile C-arm systems is 30 cm
from the image receptor assembly entrance surface, the There is a limited number of intraoperative 2D/3D flu-
same location specified by the FDA for determination of oroscopy systems on the commercial market. They
the maximum allowable dose rate limit.5 include: (1) Medtronic intraoperative 2D/3D system, oth-
erwise known as O-arm, (2) Ziehm Vision RFD 3D, (3)
GE-OEC 3D, and (4) Siemens Cios Spin. Systems (2)–
NOTE 9: For all but interventional C-arms, there is no
(4) are extension of C-arm fluoroscopy capability. The
IEC definition for the RP (the PERP). However, vendors
O-arm is designed to perform both 2D fluoroscopy and
report the location of the PERP in their accompany-
3D imaging for image-guided operations such as spine
ing documentation. Often, C-arm manufacturers follow
fusions and is not a C-arm based system. Typically, 2D
FDA regulations for defining the PERP (hence, 30 cm
fluoroscopy is used to position the system in preparation
from the image receptor).
for 3D imaging, and 3D imaging is used to reconstruct
images from pulsed X-ray exposures that are obtained
during a complete 360◦ rotation of the tube within the
3.4 Collimator congruence test gantry around the patient. The 3D acquisition can be
obtained with either standard mode or high-resolution
Collimator congruence tests should be conducted simi- mode, which uses twice the number of projections as
lar fashion as in Section 2.4. standard mode. All 3D acquisitions are obtained using
programmed techniques that do not automatically mod-
ulate during the scan.
3.5 Display monitors For acceptance testing, benchmark measurements
should be obtained in both 2D and 3D modes of
Evaluation of display monitors for mobile C-arm fluoro- operation to ensure proper functioning. Note that an
scopes should be performed using the procedures out- intraoperative 2D/3D is classified by the FDA as a
mobile fluoroscopy unit; thus, as of the time of this pub-

lication, only fluoroscopy FDA regulations apply to this
type of system (barring any state-specific regulations).
3.8.1 2D fluoroscopy
Tests for 2D fluoroscopy can be performed following the

protocols described previously in Section 2.3. Note that
the tube and detector assembly on these systems can
be rotated such that the tube is overhead, allowing for
easier setup of phantoms and meters.
Despite its use primarily as a 3D imaging device, the
F I G U R E 1 9 Conventional fluoroscopy system. An FPIR system
Medtronic O-arm is classified by the FDA as a mobile is shown here. The X-ray tube is installed under the tabletop and
fluoroscopy unit. Consequently, it must adhere to maxi- enclosed in the table side panels
mum tube output regulations in 2D imaging mode. The
maximum output for 2D imaging mode must be mea-
sured 30 cm from the outer plastic cover placed over
the image receptor (83 cm from the focal spot). This is using smaller chambers in air are also acceptable; how-
not the same location as the PERP for the displayed air ever, there is no reference for these values in manufac-
kerma, which the vendor reports to be 15 cm toward the turer documentation.
X-ray tube from isocenter, or 50 cm from the focal spot. Because the 3D acquisition of this system is designed
The on-screen displayed air kerma often exceeds mainly to visualize and measure high-contrast objects,
the FDA maximum air kerma limits in 2D fluoroscopy the high-contrast module of the ACR CT phantom can
mode; thus, Medtronic does not recommend extended be used to assess spatial resolution. A resolution ≥6 line
2D imaging with the patient at isocenter. pairs/cm is recommended.
3.8.2 3D acquisition 4 CONVENTIONAL FLUOROSCOPY

For rotational angiography equipment, a method sug- A conventional fluoroscopy system is built around a
gested by Medtronic is to use the computed tomography floating patient table, which can be rotated to a vertical
dose index (CTDI) body phantom to measure dosimetry position. The fluoroscopy X-ray tube is installed under
and the ACR CT phantom to measure image quality. the patient table, and the II or digital FPIR is installed
Briefly, because of the wide cone-beam projections in above the table (Figure 19).
the Intraoperative 2D/3D, the concept of CTDI for dose
profiles up to 100 mm is not valid for these systems. The
correct way to measure the dose on these units is to 4.1 Mechanical factors
position a small volume chamber at the center and at the
periphery of a phantom that is long enough to accom- The mechanical integrity and smooth operation of the
modate all scatter to the longitudinal center. The man- examination tabletop and the fluoroscopy table/pedestal
ufacturer, however, still uses the 100-mm CTDI cham- assembly should be evaluated during acceptance test-
ber to report CTDIvol on standard CT dose phantoms ing. These tests should be conducted not only at the
(16- and 32-cm CTDI phantoms), which underestimates tableside control panel (Figure 20), but also at the con-
the dose. In practice, dose phantom selection depends trol panel located at the FPIR tower assembly (see Fig-
on the anatomy protocol. For example, the 16-cm phan- ure 19). These two control panels are similar to each
tom is used for head and neck protocols, and the 32-cm other in function, but the specifics of which functionali-
phantom is used for chest, abdomen, pelvis, and leg pro- ties are available at each location vary by vendor.
tocols. The 100-mm chamber can be used to test the dif- The following factors should be assessed:
ference between the measured and displayed CTDIvol ;
a difference of ≤20% between these values is recom- ∙ Tabletop motion: The floating table can be moved left,
mended. Note that the system uses 120 kV for all 3D right, forward, and backward.
imaging but scales mAs values based on patient size. ∙ Table angulation: The floating table can smoothly
Therefore, dosimetry testing can be performed for one rotate up to 90◦ , in both counterclockwise (–) and
patient size, and other CTDI values for various patient clockwise (+) directions,although some systems have
sizes (small, medium, large, and extra large) can then be a limited angulation of –30◦ (Trendelenburg or head
linearly scaled. Other methods for evaluating constancy down position).
F I G U R E 2 0 The tableside control panel for the examination

table. Adjustments to control tabletop motion, table assembly
angulation, and angle display window are available
F I G U R E 2 2 The experimental setup for measuring radiation

output on a conventional fluoroscopy system. Note X-ray tube under
table with a fixed SSD
4.3 Radiation output measurements
4.3.1 Radiation output as a function of

PMMA thickness
The experimental setup for measuring radiation output

is shown in Figure 22. The detailed procedures are also
described below.
FIGURE 21 Fluoroscopy suite with overhead radiographic X-ray
tube
4.3.2 Maximum AKR
For conventional fluoroscopy systems in which the X-ray

∙ Alignment and centering with the overhead radio- tube is installed under the examination table, the maxi-
graphic tube: In most conventional fluoroscopy rooms, mum fluoroscopic AKR limitations for all available FOVs
there is an overhead radiographic tube that can be should be tested under normal and boost modes with
used in conjunction with the fluoroscopy table to func- the radiation detector placed 1 cm above the tabletop
tion as a radiographic system (Figure 21). Alignment in accordance with FDA regulations. Typically, the image
and centering between the fluoroscopy table and the receptor tower assembly is raised at least 30 cm from
radiographic tube should be assessed. the tabletop to simulate clinical conditions. However, the
∙ Radiation shielding apparatus: The shielding appara- maximum SID (∼100 cm) may be employed to ensure
tus should be assessed to identify any possible dam- consistency in measurements from year to year.
age. The installation of the shielding apparatus should For regulatory maximum AKR measurements, in-air
also be checked to ensure proper shielding of opera- geometry with no scatter material in the primary beam
tors at typical positions. (with the exception of the examination tabletop) should
be used.
4.2 X-ray generator and radiological

parameter accuracy 4.3.3 Dose and dose rate at PERP
Generator parameter accuracy measurements (includ- Follow the procedure described in Section 2.3.3, with the
ing X-ray tube potential, tube current, mAs linearity, and PERP placed 1 cm above the tabletop without a mat-
HVL) should be assessed as outlined in Section 2. tress.
4.3.4 Image receptor input dose and dose confirmed to conform with the specific manufacturer
rate setup.
If the grid can be removed, a geometrical correction can

be applied to the FPIR measurement to calculate the 4.7 Image quality assessment
dose rate at the level of the detector surface. For cases
in which the grid is not removable, vendors can some- Image quality should be assessed as outlined in Sec-
times supply a grid correction factor that can be applied tion 2.7, using similar phantoms. At a minimum, HCR
to input surface measurements to account for attenua- and low-contrast detectability should be evaluated, as
tion of the grid and to allow for calculation of detector required by Joint Commission standards.
input dose rates. As mentioned in Section 2.3.4, image
receptor input dose rates can be useful for measuring
baseline performance but are not absolutely necessary 5 OVERHEAD-TYPE FLUOROSCOPY
for acceptance testing.
Overhead-type fluoroscopy (OHTF) systems present
several challenges for acceptance testing. First, as
4.3.5 Displayed PERP dose verification the name implies, OHTF systems are inverted with
respect to the X-ray tube and image receptor configu-
Follow the procedure described in Section 2.3.5. ration, presenting a geometry that is unfavorable to the
operating staff from a radiation safety perspective. In
systems with the X-ray tube assembly installed under
4.4 Collimator congruence test the examination table, the scatter profile in the room is
typically directed toward the floor. With OHTF geometry,
Follow the procedure described in Sections 2.4.1 the scatter profile is directed toward the torso and
and 2.4.2 for the posteroanterior direction with the table head of the operator. Therefore, acceptance testing
angulation set at ±30◦ , 45◦ , and 90◦ . for OHTF systems should include scrutiny of available
radiation shielding devices such as moveable shields
and leaded glasses38 so that staff can avoid lens
4.5 Display monitors injury.39
Second, OHTF systems typically have both radio-
Follow the procedure described in Section 2.5. graphic and fluoroscopic capabilities, which complicate
acceptance testing. The procedures performed using
OHTF units (e.g., video urodynamics and cystoscopy)
4.6 Fluoroscopic dose monitoring require scanning through the pelvis (often while the
and tracking patient is sitting) and imaging bone, fat, air, and con-
trast, necessitating a high dynamic range. These
4.6.1 TG 190 dose correction systems may also include a micturition chair that is
attached to the table, with the gantry rotated upward at
All available displayed radiation doses (reference air 90◦ .
kerma, dose-area product, KAP) should be verified per In addition, OHTF systems are often equipped
the protocols described in AAPM Report No. 190.9 The with control panels attached to the examination table
PERP for cumulative air kerma on systems in which and on a mobile pedestal, the functionalities and
the X-ray tube is installed under the examination table proper operation of these controllers should also be
is typically the same as the FDA-specified maximum verified.
dose point: 1 cm above the tabletop. Hence, the mattress
should not include.
NOTE 9: OHTF systems designed for upper/lower GI
4.6.2 Patient comfort mattress examinations fall within this category of fluoroscopy
attenuation correction imaging systems, although these systems are often
referred to as “remote-controlled”fluoroscopy systems,
Because the X-ray tube assembly is under the table, as radiologists work in the control booth behind pro-
the patient’s skin is exposed at all times with trans- tective shielding. Some room configurations therefore
mission through the mattress and tabletop. However, present opportunities for the radiologist/operator to be
depending on how the Kair,PERP and air KAP are remoted and not receive scatter from patients being
defined, the verification of the PERP needs to be imaged on the OHTF system.
and move them as one assembly. The images in Fig-

ure 24 show translation of the X-ray tube and the image
receptor motion; these motions should be assessed to
ensure proper alignment once the assembly is locked.
The centering of the X-ray central beam with the
image receptor should also be evaluated to ensure
radiation beam alignment. Because the image recep-
tor is installed just under the tabletop, the procedure
described in Section 2.4.2 must be modified to account
for configuration differences.
5.2 X-ray generator and radiological

parameter accuracy
F I G U R E 2 3 The “park” location of the X-ray tube (retracted, on
the left) and the “operation” location of the X-ray tube (extended, on
the right) on an OHTF system
Generator parameter accuracy should be measured fol-
lowing the procedure described in Section 2.2. The
presence of radiographic and fluoroscopic functionality
5.1 Mechanical factors should ease testing.
5.1.1 Tabletop motion

5.3 Radiation output measurements
OHTF systems typically have relatively short tabletops
(compared to those used in typical fluoroscopy systems) 5.3.1 Radiation output as a function of
with multiple add-ons to the table that allow for various PMMA thickness
imaging setups. The table and add-ons should be tested
for ease of motion and secure attachment, and the room Figure 22 shows the proper setup to measure patient
should be checked for pinch points during mechanical dose rates. Note the inverted geometry and the need for
motions when add-ons are installed. fixation of ionization chamber or SSDS.
5.1.2 Verification of SID 5.3.2 Maximum AKR
OHTF systems designed for upper/lower gastrointesti- The maximum radiation output (dose and dose rate) on
nal examinations are often equipped with variable SID. OHTF systems must be measured using in-air geome-
The SID selection and the accuracy of SID should there- try in the absence of any scattering material. The output
fore be verified during acceptance testing. must be calibrated so that it does not exceed the regu-
latory allowable limits set for corresponding operational
modes. Specifically, the maximum AKR must not exceed
5.1.3 Alignment and centering with 88 mGy/min in regular mode and 176 mGy/min in high-
overhead fluoroscopic tube level control or boost mode. The FDA has stated that a
measurement point 30 cm from the tabletop should be
Because of the variety of configuration options avail- used to determine the maximum allowable output.40
able on OHTF systems, the tube head may have sev-
eral degrees of freedom, including the ability to move
out of the way to allow a patient to be loaded onto the 5.3.3 Dose and dose rate at the PERP
table. These degrees of freedom provide opportunities
for misalignment between the tube and receptor and Patient entrance dose should be measured following the
should therefore be assessed during acceptance test- procedure used for other systems; however, for OHTF,
ing. Furthermore, a safety switch check should be per- the system is inverted. Figure 25 shows an SSDS in the
formed to ensure that no X-rays can be generated if the beam with PMMA stacked on the table. Regardless of
X-ray tube assembly is not interlocked over the image PMMA thickness, corrections must be made to 30 cm
receptor. Many of these systems (e.g., urology and cys- from the tabletop.
toscopy) incorporate a fixed SID. Figure 23 shows two If an SSDS is used, selecting particular fields to mea-
of these possible tube motions/positions. sure (for fluoroscopy) away from the SSDS may avoid
As shown in Figure 24, many systems also have the driving up the ADRC. These fields can also be tested
ability to couple the X-ray tube with the image receptor by covering corresponding areas with Cu or Al and
F I G U R E 2 4 Left: Longitudinal translation

of the tube assembly is not aligned to the
image receptor. Right: The alignment is
centered with the image receptor, and the
assembly is locked for imaging in fluoroscopy
mode or radiography/acquisition mode, as
required
F I G U R E 2 6 Software control panel showing nine selections of

sensing areas for the ADRC of the image receptor for fluoroscopy. In
each of these nine selections, a black rectangular box represents an
F I G U R E 2 5 Experimental setup for measuring radiation output. active sensing area. Note that control sensing areas can be
Care must be taken to avoid driving the ADRC with SSDS. An programmed for multiple or single locations by selecting one of these
ionization chamber works well in these situations and can be left in nine setups. The setup shown in the lower right corner represents a
the same position for testing all slab thicknesses. Doses can then be situation in which the entire image receptor is active as the ADRC
corrected for 30 cm from the tabletop sensor
determining whether the area selected is still being

used as a dominant sensor for the ADRC (as shown in
Figure 26). 5.4 Collimator congruence test
5.3.4 Image receptor input dose On many of these systems, a two-step verification pro-
cess may be necessary to assess the congruence
and dose rate
between light and radiation field and between radiation
field and displayed image.Refer to Section 2.4 for further
In OHTF systems,the receptor is located under the table.
information.
Unless the vendor provides access to the area under the
table, the flat panel dose can be measured only on top
of the table.
5.4.1 Light field illuminance test
5.3.5 Displayed PERP dose verification If a light field for radiographic imaging is in place,
the light field illuminance should be assessed to
Consult the vendor’s technical manual for information ensure that it meets regulations. Refer to state
about PERP. Typically, the PERP is 30 cm above the regulations for light illuminance minimum require-
tabletop. ments.
5.4.2 Light to radiation field congruence ing details. Thus, adequate assessment of both fluo-
roscopy and recorded image quality may be necessary.
This test can be performed using a computed radiogra- Consult the image quality Section 2.7 of this report
phy plate,Gafchromic film,or other independent external for more information regarding the multipurpose image
media. Rulers or other markers should be used to com- quality phantoms that should be used for these tests.
pare the exposed area on the media with the light field
for the head, the foot, and the left and right directions.
6 MINI C-ARM FLUOROSCOPY
5.4.3 Radiation field to displayed image Mini C-arm fluoroscopy systems are C-arm systems
congruence with a maximum SID < 45 cm that are used to image
the extremities. The SSD of these systems should
Once congruence between light and radiation field has be >19 cm for general extremity use but could be as low
been established, congruence between the radiation as 10 cm for specific surgical applications that require a
field and the displayed image on the monitor should be shorter SSD.40
assessed. This test not only verifies the visual accuracy
of the displayed image but also ensures that the radi-
ation does not fall outside of the fluoroscopic image 6.1 Mechanical factors
receptor and that it tracks appropriately for all magni-
fication modes. Refer to Section 2.4. The following parameters should be verified at the time
of acceptance testing for mini C-arm systems:
5.5 Display monitors ∙ The SID should be <45 cm.

∙ The SSD should be >19 cm for systems used for gen-
Display monitors should be tested using the procedure eral extremity imaging.
described in Section 2.5. ∙ The SSD should be ≥10 cm for systems used for spe-
cific surgical applications.
∙ A label stating “For extremity use only” should be
5.6 Fluoroscopic dose monitoring and affixed to mini C-arm systems used for general
tracking extremity imaging.
∙ The accuracy of the laser alignment should be
5.6.1 TG 190 dose correction assessed.
∙ A visible mark should be placed for focus location.
Refer to Section 2.6.1 for accuracy of KAP meters. ∙ The mechanical stability of the mini C-arm gantry
should be evaluated.
∙ All pivots and locks should be assessed to ensure
5.6.2 Examination mattress attenuation smooth motion.
correction
Because the tube-receptor configuration is inverted on 6.2 X-ray generator and radiological
these units, the skin is exposed without transmission parameter accuracy
through the mattress or tabletop; therefore, no correc-
tions for tabletop or mattress are required for skin dose Most mini C-arm fluoroscopy systems allow for man-
calculations. ual control of tube potential, which enables relatively
straightforward measurement of X-ray generator param-
eter accuracy.
5.7 Image quality assessment
Dynamic range is a crucial part of image quality assess- 6.2.1 Tube potential (kV)
ment for OHTF systems. The patients imaged on these
systems can be quite large. Additionally, systems used Tube potential should be measured as follows:
for urology often do not have high kW ratings, making
high dose rate fluoroscopy (>88 mGy/min) unachiev- ∙ Place an appropriate attenuation material (>1/32-inch
able. In many cases, because of these factors, fluoro- lead or 2-mm Cu sheet) on top of the image receptor
scopic image quality may be insufficient, and in extreme assembly to protect the image receptor from multiple
cases, acquisitions may be used to obtain needed imag- exposures for kV measurements.
∙ Position the noninvasive kV meter (or SSDS radiation 6.3.2 Maximum AKR
probe) 10–20 cm from the surface of the image recep-
tor housing assembly. Because mini C-arm systems are designed for extremity
∙ Measure the tube potential under fluoroscopy using use only, the maximum output with maximum kV is easily
manual kV adjustment if available. reached with a PMMA thickness of 15 cm or a 1/32-inch
∙ Ensure that the measured kV is within the manufac- lead sheet. The PERP employed for the displayed dose
turer’s specification or within a tolerance of ±10%. and dose rate for mini C-arms is not defined uniformly
∙ Repeat the last two steps for radiographic mode. across various vendors.
6.2.2 Tube current (μA) or tube 6.3.3 Dose and dose rate at the PERP
current–time product (μAs)
Follow the procedure described in Section 2.3.3.
Most mini C-arm systems do not allow for manual adjust-
ment of tube current decoupled from changes in the kV
6.3.4 Image receptor input dose and dose
setting, so independent tube current and linearity cannot
rate
be measured.
For most mobile C-arm systems inclusive of mini C-arm
6.2.3 Exposure time fluoroscopy, the grid is fixed in place behind the protec-
tive cover of the image receptor assembly. Therefore,
Most mini C-arm systems do not allow for adjustment of the grid cannot be accessed without some level of dis-
pulse width or exposure time; however, pulse width (ms) assembly. For this reason, measurement of the image
can be measured and displayed pulse rate (pps) can be receptor input dose and dose rate is not generally rec-
verified with most modern SSDS. Follow the procedure ommended.
described in Section 2.2.3.
6.3.5 Displayed PERP dose verification
6.2.4 HVL (mm Al)
The first step in evaluating the accuracy of AKR on
Follow the procedure described in Section 2.2.5. mini C-arm systems is to obtain the PERP from the
operator’s manual, as federal regulations do not specify
a standard location for the PERP. Typical RP locations
6.2.5 Total filtration (mm Al) are 2 or 5 cm from the image receptor assembly; thus,
distance corrections may be needed to adjust the value
Follow the procedure described in Section 2.2.6. from a measured AKR to a displayed AKR. TG 272
strongly recommends that the PERP for the displayed
AKR for mini C-arm equipment should be uniformly
6.3 Radiation output measurements defined as 10 cm from the image receptor assembly.
As mentioned previously, the practical measurement of

technique parameters can be performed at any point 6.4 Collimator congruence test
on the central axis. However, 21CFR1020.32(d)(3)(iv)
states that for mini C-arm fluoroscopes with an Follow the procedure outlined in Section 2.4 for all avail-
SID < 45 cm, the maximum AKR should be measured able FOVs. Typical mini C-arm systems may not be
at the minimum SSD.5 equipped with a manual collimating feature but may
have two different FOVs.
6.3.1 Radiation output as a function of

PMMA thickness 6.5 Display monitors
On mini C-arm systems, radiation output and technique Display monitor testing should be performed following
parameters such as kV, μA, and/or pulse width can be the procedure described in Section 2.5. Mini C-arm
measured as a function of PMMA thickness in various systems are typically used in brightly lit emergency
clinical modes of operation including fluoroscopy, snap- departments, surgical suites, and cast rooms. For this
shot,and cine modes.To simulate typical extremity sizes, reason, special attention is required when attempting to
the PMMA thickness can be varied from 2 to 15 cm in match the ambient lighting conditions used in testing to
1-cm increments. the lighting conditions used clinically.
6.6 Fluoroscopic dose monitoring and included in the installation schedule for acceptance test-
tracking: TG 190 dose correction ing to be performed by the medical physicist.
The area for the displayed KAP for mini C-arm systems D I S C L O S U R E S TAT E M E N T
does not actively change but can be calculated based The members of AAPM TG 272 attest that they have no
on the size of the default FOVs from the image receptor. potential Conflicts of Interest related to the subject mat-
However, the PERP of mini C-arm systems varies from ter or materials presented in this document. The Chair
one manufacturer to another and often from one ver- of the AAPM TG 272 has reviewed the required Conflict
sion of software to another for the same unit. If patient of Interest statement on file for each member of AAPM
dose monitoring is desired, the location of the PERP TG 272 and determined that disclosure of potential con-
must therefore be verified, and the displayed PSD must flicts of interest is an adequate management plan. Dis-
be normalized/standardized if multiple models and/or closures of potential conflicts of interest for each mem-
brands of mini C-arm systems are used within an ber of AAPM TG 272 are found at the close of this doc-
institution. ument. In addition, there is no funding sources provided
to this task group or any of the task group members in
relation to this report.
6.7 Image quality assessment
REFERENCES
Follow the procedure described in Section 2.7. At a
1. National Electrical Manufacturers Association. NEMA XR 27–
minimum, HCR and low-contrast detectability should be 2013 (R2018): X-ray Equipment for Interventional Procedures
evaluated, as required by Joint Commission standards. User Quality Control Mode. National Electrical Manufacturers
Note that because of the small FOV, sections of the test Association; 2019.
tools may need to be imaged separately for different 2. International Electrotechnical Commission. IEC 60601-2-
image quality metrics. Additionally, any added attenuator 43:2010+AMD1:2017+AMD2:2019 CSV, Consolidated Version.
Medical electrical equipment – Part 2–43:particular requirements
used in conjunction with a test tool should be clinically for the basic safety and essential performance of X-ray equip-
relevant (e.g., 5 cm of PMMA phantom). ment for interventional procedures. International Electrotechnical
Commission; 2019.
3. Digital Imaging and Communications in Medicine. PS 3.3-2003:
7 DISCUSSION Part 3: Information Object Definitions. C.8.7.4.1: X-ray table
attribute description. National Electrical Manufacturers Associa-
The main focus of this report is acceptance testing; thus, tion; 2003.
no attempt has been made to recommend which param- 4. US Food and Drug Administration. Code of Federal Regulations
eters need to be evaluated on a periodic basis or how Title 21. Performance Standards for Ionizing Radiation Emitting
Products, Section 1020.30: Diagnostic x-ray systems and their
often these tests should be performed. Annual equip- major components. (m) Beam quality (1) Half-value layer (HVL).
ment testing, for instance, may be a subset of accep- Food and Drug Administration; 2019.
tance testing. However, the frequency of testing required 5. US Food and Drug Administration. Code of Federal Regulations
for the various physical parameters can differ substan- Title 21. Performance Standards for Ionizing Radiation Emit-
tially. For instance, parameters such as tube potential ting Products, Section 1020.32: Fluoroscopy Equipment. (d) Air
Kerma Rates. US Food and Drug Administration; 2019.
and HVL, which are not likely to change substantially 6. Suggested State Regulations for Control of Radiation, Part F,
over a 1-year period, may be categorized as requiring Section F.5 (e)(iv)(3). Conference of Radiation Control Program
only annual evaluation. On an annual basis, a reduced Directors, Inc. 112 East Main Street, Suite 1, Frankfort, KY 40601
number of physical parameters may be evaluated in 7. Lin PJ, Rauch P, Balter S, et al. AAPM Report No. 125: Function-
addition to the parameters that must be assessed to ality and Operation of Fluoroscopic Automatic Brightness Con-
trol/Automatic Dose Rate Control Logic in Modern Cardiovascular
meet regulations, such as the alignment and congru- and Interventional Angiography Systems. American Association
ence of the radiation field and the image receptor. Two of Physicists in Medicine; 2012.
Institute of Physics and Engineering in Medicine Top- 8. Rauch P, Lin PJ, Balter S, et al. Functionality and opera-
ical Reports discuss the parameters to be evaluated tion of fluoroscopic automatic brightness control/automatic dose
and the suggested frequency of assessments in further rate control logic in modern cardiovascular and interventional
angiography systems: a report of Task Group 125 Radiog-
detail.41,42 raphy/Fluoroscopy Subcommittee, Imaging Physics Committee,
The total time required to perform the acceptance Science Council. Med Phys. 2012;39:2826-2828.
tests described in this document will vary substantially 9. Lin PJ. The operation logic of automatic dose control of fluo-
depending on the modality. For the most intensive test- roscopy system in conjunction with spectral shaping filters. Med
ing procedures on the most complex units (e.g., cardio- Phys. 2007;34(8):3169-3172.
10. Lin PJ, Schueler BA, Balter S, et al. Accuracy and calibration of
vascular and visceral interventional units), the physicist integrated radiation output indicators in diagnostic radiology: a
should plan for 12–16 h of testing per imaging plane. report of the AAPM Imaging Physics Committee Task Group 190.
Other types of units should require less time, depend- Med Phys. 2015;42:6815-6829.
ing on the physicist’s familiarity with the control func- 11. Shepard SJ, Lin PJ, Boone JM, et al. AAPM Report No. 74: Quality
tions. Ideally, irrespective of the modality, time should be Control in Diagnostic Radiology: Report of Task Group #12 of
Diagnostic X-Ray Imaging Committee. American Association of roscopy from Sources of Scattered X-Rays. American Society for
Physicists in Medicine; 2002. Testing and Materials International; 2014.
12. US Food and Drug Administration. Resource Manual for 29. Seissl J, Eschenbacher H, inventors, Siemens Aktiengesellschaft,
Compliance Test Parameters of Diagnostic X-Ray Systems. assignee. X-ray diagnostic apparatus with a filter device. US
Field Limitation and Alignment. https://www.fda.gov/Radiation- patent 5680435A. October 21, 1997.
EmittingProducts/RadiationEmittingProductsandProcedures/ 30. Inaba Y, Chida K, Kobayashi R, Zuguchi M. A cross-sectional
MedicalImaging/MedicalX-Rays/ucm115361.htm#III. Accessed study of the radiation dose and image quality of X-ray equipment
November 3, 2020. used in IVR. J Appl Clin Med Phys. 2016;17:391-401.
13. Lin PJ. A device for checking alignment of fluoroscopic systems. 31. International Electrotechnical Commission. International Stan-
Opt Eng. 1977;16:163302. dard. IEC 61223-3-1: Evaluation and Routine Testing in Medical
14. Bevins NB, Flynn MJ, Silosky MS, et al. AAPM Report No. 270: Imaging Departments. Part 3-1: Acceptance Tests - Imaging Per-
Display Quality Assurance. American Association of Physicists formance of X-Ray Equipment for Radiographic and Radioscopic
in Medicine; 2019. Systems. International Electrotechnical Commission; 1999.
15. Bevins NB, Silosky MS, Badano A, Marsh RM, Flynn MJ, Walz- 32. Leeds TOR 18FG Fluoroscopy Phantom. Leeds Test Objects
Flannigan AI. Practical application of AAPM Report 270 in dis- website. https://www.leedstestobjects.com. Accessed November
play quality assurance: a report of Task Group 270. Med Phys. 4, 2020.
2020;47:e920-e928. 33. National Electrical Manufacturers Association. NEMA XR 21–
16. ACR–AAPM–SIIM Technical Standard for Electronic 2000: Characteristics of and Test Procedures for a Phantom
Practice of Medical Imaging. Revised 2017 (Resolu- to Benchmark Cardiac Fluoroscopic and Fluorographic Perfor-
tion 41). https://www.acr.org/-/media/ACR/Files/Practice- mance. National Electrical Manufacturers Association; 2000.
Parameters/Elec-Practice-MedImag.pdf. Accessed Novem- 34. Digital Subtraction Angiography (DSA) Phantom Model
ber 3, 2020. 76–710. Supertech website. https://www.supertechx-
17. Silosky MS, Marsh RM. Performance characteristics and quality ray.com/QualityControlPhantoms/76-710.php. Accessed Novem-
assurance considerations for displays used in interventional radi- ber 4, 2020.
ology and cardiac catheterization facilities. J Appl Clin Med Phys. 35. High M, Cohen G, Lin PJ, et al. AAPM Report No. 15: Performance
2018;19:708-717. Evaluation and Quality Assurance in Digital Subtraction Angiog-
18. Samei E, Badano Al, Chakraborty D, et al. Assessment of display raphy. American Association of Physicists in Medicine; 1985.
performance for medical imaging systems: executive summary of 36. Goode AR, Snyder C, Snyder A, Collins P, DeLorenzo M, Lin
AAPM TG18 report. Med Phys. 2005;32:1205-1225. PJ. Signal and contrast to noise ratio evaluation of fluoroscopic
19. Digital Imaging and Communications in Medicine. PS 3.14-2011: loops for interventional fluoroscope quality control. J Appl Clin
Part 14: Grayscale Standard Display Function. National Electrical Med Phys. 2019;20:172-180.
Manufacturers Association; 2011. 37. Walz-Flannigan A, Weiser J, Goode AR, et al. AAPM Report No.
20. Society of Motion Picture and Television Engineers. SMPTE RP 248: Interoperability Assessment for the Commissioning of Medi-
133–1991: Specifications for Medical Diagnostic Imaging Test cal Imaging Acquisition Systems. American Association of Physi-
Pattern for Television Monitors and Hard-Copy Recording Cam- cists in Medicine; 2019.
eras (R1999).Society of Motion Picture and Television Engineers; 38. American College of Radiology, American Association of Physi-
1991. cians in Medicine. ACR–AAPM Technical Standard for Diag-
21. The Joint Commission. Revision to Recently Issued Stan- nostic Medical Physics Performance Monitoring of Fluoroscopic
dards for Organizations Providing Fluoroscopy Services, Equipment. Updated 2016. https://www.acr.org/-/media/ACR/
Issued December 17, 2018 https://www.jointcommission. Files/Practice-Parameters/RadEquip.pdf. Accessed November 3,
org/-/media/tjc/documents/standards/prepublications/ 2020.
cah_fluoroscopy_updates_prepub_jan2019_emb. 39. Vano E, Gonzalez L, Beneytez F, Moreno F. Lens injuries induced
pdf?db=web&hash=DE2AD46A5556A11E2024CC1E002F1E48. by occupational exposure in non-optimized interventional radiol-
Accessed February 2021. ogy laboratories. Br J Radiol. 1998;71:728-733.
22. Gress DA, Dickinson RL, Erwin WD, et al. AAPM medical physics 40. US Food and Drug Administration, Code of Federal Regulations
practice guideline 6.a.: performance characteristics of radia- Title 21. Performance Standards for Ionizing Radiation Emitting
tion dose index monitoring systems. J Appl Clin Med Phys. Products, Section 1020.32: Fluoroscopy equipment. (g) Source-
2017;18(4):12-22. to-Skin Distance. US Food and Drug Administration; 2019.
23. Joint commission perspectives, Sentinel event definition and 41. Honey I, Rose A, Baker C, et al. IPEM Topical Report: an evidence
chapter revised, July 2021, Page 18, Volume 41, Number 7. and risk assessment based analysis of the efficacy of tube and
24. National Council on Radiation Protection and Measurements. generator quality assurance tests on general x-ray units. Phys
Radiation dose management for fluoroscopically-guided inter- Med Biol. 2018;63:245011.
ventional medical procedures. NCRP Report No. 168. National 42. Worrall M, Shaw D, Baker C, et al. IPEM Topical Report: an evi-
Council on Radiation Protection and Measurements, Bethesda, dence and risk assessment based analysis of the efficacy of
MD. quality assurance tests on fluoroscopy units – part I; dosimetry
25. DeLorenzo MC, Goode AR. Evaluation of skin dose calcula- and safety. Phys Med Biol. 2019;64:195011.
tion factors in interventional fluoroscopy. J Appl Clin Med Phys.
2019;20:159-168.
26. International Electrotechnical Commission (IEC2010). Medical
Electrical Equipment–Part 2–43: Particular Requirements for the
Basic Safety and Essential Performance of X-ray Equipment for
How to cite this article: Lin P-JP, Goode AR,
Interventional Procedures, IEC 60601-2-43 ed2.0, International Corwin FD, et al. AAPM Task Group Report 272:
Electrotechnical Commission, Geneva, 2010 Comprehensive acceptance testing and
27. Lin PJ. Preparation for peak skin dose estimation. J Appl Clin Med evaluation of fluoroscopy imaging systems. Med
Phys. 2018;19:370-407. Phys. 2022;49:e1–e49.
28. American Society for Testing and Materials International. ASTM
F3094-1: Standard Test Method for Determining Protection Pro-
https://doi.org/10.1002/mp.15429
vided by X-Ray Shielding Garments Used in Medical X-ray Fluo-
e37
A P P E N D I X A : S A M P L E AC C E P TA N C E T E S T I N G F O R M
AAPM TGR-272: FLUOROSCOPY IMAGING SYSTEM TESTS
e38
e39
e40
e41
e42
e43
e44
A P P E N D I X B : G E N E R ATO R T E S T P O I N T S test points varies depending on the specific generator

The purpose of this semi-invasive test is to evaluate in question. Some systems require entering into ser-
the waveforms of tube potential and tube current. In vice mode and enabling or disabling certain service key
so doing, the balance and/or difference between anode switches, whereas other may have no specific precau-
and cathode tube potentials can be assessed. In addi- tionary measures.
tion, the pulse height of the tube potential and tube Once the test points are located, the following gen-
current over a prolonged exposure time (up to ∼15 erator parameters can be evaluated: tube potentials
s) can be evaluated, and fluctuations in pulse heights (kVanode , kVcathode , kVtotal ), tube current (mA), and expo-
in the kV waveform can be observed rather than the sure time (ms). The scaling factors for the tube poten-
radiation exposure train obtained with the radiation tial (e.g., 1 volt = 10 kV) and tube current (e.g., 1
detector/probe. volt = 10 mA or 1 volt = 100 mA) must also be obtained.
Decades ago, generator-related physical parameters To use this generator test point method,the instrument
such as the tube potential (kV, kV), tube current (mA), must have at least a 25- to 50-MHz dual-beam storage
and exposure time (in single exposure or in pulse oscilloscope and appropriate probes.
width [ms]) were measured using an invasive approach,
utilizing a high-tension voltage divider and a storage A P P E N D I X C : C O N V E N I E N C E TA B L E
oscilloscope.With the introduction of medium-frequency FOR IN-AIR MEASUREMENTS
inverter-type generators, these generator parameters To facilitate various measurements and to comply with
could be more accurately delivered or produced than in-air measurements, a homemade table was fabricated
with the previous generation of generators. in a machine shop, as shown below:
On the other hand, when noninvasive SSDS became The table can be designed to have either a fixed or
available, tube voltage, exposure time, and radiation adjustable height (corresponding to the Y-axis shown in
dose could be measured in one exposure. Although Figure 1 of the main text). For most interventional X-ray
SSDS are convenient,the wave form provided with these equipment systems, an acceptable total height is a fixed
systems is not the voltage (or current) waveform. Rather, 40 inches. For the table shown above, the maximum total
it is the radiation waveform, which reflects the power height (including the wheels) is 50 inches, and the height
loading of the X-ray tube and is influenced by both the can be adjusted in 1-inch increments. The same table
tube voltage and current. The exposure time is derived can be used for mobile C-arm equipment testing; the
from the radiation waveform, and its accuracy is accept- mobile C-arm gantry can be raised to meet the required
able when the exposure time is longer than approxi- geometry to perform the measurements described in
mately 15–20 ms;for exposure times shorter than 20 ms, this report.
and especially for those shorter than 10 ms, the accu- The plate with 8-inch × 10-inch opening, shown on
racy is not at the desired level of better than ±5% toler- the left of the drawing, rides on a track in the tabletop
ance. For pulse width measurements, most SSDS may (16 inches × 44 inches) for X-axis direction movement.
not be suitable measurement devices because of the The table as a whole is moved into position in the Z-
operating frequency of SSDS and the memory available. axis. The 8-inch × 10-inch plate is equipped with a trans-
For most generators, the test points are in the equip- parency film for inkjet printers (e.g., 3M’s product num-
ment cabinet and can be accessed with assistance from ber CG3460). The transparency film is employed to hold
the service engineers. The process to access these the SSDS, the radiation probe, or an ionization chamber.
A table width of 44 inches is necessary to accommo- lation of the tube potential is used to optimize visual-
date interventional X-ray systems that have the X-ray ization of the guide wire for different materials.
tube assembly installed on the primary C-arm (the A-
plane) such that the cathode–anode direction is perpen- The protocols designed for each specific procedure
dicular to the C-arm, in parallel with the X-axis. are created and analyzed based on literally thousands
of individual cases. It would be difficult if not impossi-
A P P E N D I X D : C N R O P T I M I Z AT I O N ble to evaluate such comprehensive and complex CNR
One of the concepts involved in the logic design optimization ADRC logic design. To evaluate the perfor-
of automatic dose rate control (ADRC) fluoroscopy mance of a preprogrammed procedure, an appropriate
curves is CNR optimization, as described in US Patent phantom must therefore be designed and fabricated.
6222907B1.1 Traditional logic is based on SNR opti-
mization or constant input dose (or dose rate) to the A P P E N D I X E : P O S I T I O N A L PA R A M E T E R
image receptor, and most image quality evaluations and V E R I F I C AT I O N O F I N T E RV E N T I O N A L
phantoms are based on the SNR optimization concept X - R AY E Q U I P M E N T
that the input dose is kept constant by the fluoroscopy The radiation dose delivered to the skin of the patient
ADRC control logic. However, with innovations and tech- during an interventional radiology or cardiac catheteri-
nological advancements such as higher X-ray tube heat zation procedure is directly correlated with the potential
capacity and more sensitive image receptors, there has for deterministic skin effects. The point at which the X-
been a philosophical shift in fluoroscopy ADRC logic ray beam enters the patient is therefore of paramount
design. Many new cardiovascular (and most fluoroscopy importance for determining the localized peak skin dose.
imaging) systems are shifting away from SNR optimiza- All modern fluoroscopic equipment is capable of moni-
tion operation logic to CNR optimization, which offers toring, displaying, and recording the geometrical param-
acceptable patient exposure.2 This shift in ADRC logic eters that define the position of the image receptor dur-
design, in turn, may require a change in acceptance testing an exposure. Likewise, the equipment will calculate
ing and phantom design. an air kerma value at a manufacturer-defined point in
The CNR optimization ADRC logic is designed to a space between the X-ray source and the imaging plane,
predetermined CNR tailored for a specific clinical exam- referred to as the PERP. These parameters are all saved
ination or procedure. When the type of examination or to the RDSR file, which can be viewed on a structured
procedure is selected, the ADRC control logic is pre- report viewer.
pared with certain boundary conditions that attempt to To accurately determine the peak skin dose and the
optimize CNR for the selected procedure.Some of these location on the skin of this dose, the precise orientation
predetermined parameters are as follows: of the X-ray imaging system must be understood. The
parameters involved include:
∙ the expected motion of the anatomy (e.g.,heart walls),
∙ the orientation of the patient on the examination table,
to determine the pulse width, fluoroscopy pulse rate,
∙ the primary angulation of the X-ray source to the left
and acquisition frame rate;
∙ the anatomy itself/the organ(s) to be investigated, or right of the patient,
∙ the secondary angulation of the X-ray source in the
including the intrinsic contrast scale of organs versus
cranial or caudal plane of the patient,
surrounding anatomy (e.g., the lungs, mediastinum,
∙ the X-ray source-to-imaging plane distance (SID),
and spine in chest cavity);
∙ the X-ray source-to-patient entrance distance (SPD),
∙ the contrast agent used in angiography, which may
∙ the longitudinal position of the patient (left/right) rela-
be iodine-based (k-absorption edge of 33.17 keV)
tive to the X-ray source,
or a barium-sulfate compound (k-absorption edge of
∙ the lateral position of the patient (head/foot) relative
37.44 keV); this factor predominantly determines the
to the X-ray source, and
optimal tube potential that will be used;
∙ the manufacturer-specified PERP location between
∙ the expected flow rate of circulation of contrast media
the source and image receptor
for programmed stepping imaging (typically, stepping
is necessary to follow the imaging areas from the The purpose of this appendix is to outline a method-
abdomen to the pelvis to the lower extremities in vis- ology to determine the accuracy of the geometrical
ceral angiographic studies); parameters displayed to the user versus those recorded
∙ the size and makeup of the guide wire, catheter, and to the RDSR.
stent. These items, when intended for use in angiog- Materials: Tools for determining rotational accuracy:
raphy, are composed of medical-grade stainless steel
or chromium–cobalt alloys. Iron mixed with carbon ∙ plastic framing square (commonly used by carpen-
to produce steel is the main component of stainless ters), with BBs attached at the three corners of the
steel, along with elements such as chromium, nickel, triangle,
molybdenum, silicon, aluminum, and carbon. Manipu-
TA B L E E 1 Parameters for gantry geometrical evaluation center of rotation. (The characteristics and/or
Positional parameter Suggested value size of the laser pointer might place the coro-
nal beam a couple of cm above the center of
RAO/LAO (◦ ) 0
rotation. Note the difference.)
Cranial/caudal (◦ ) 0 c. Tape a ruler horizontally on the C-arm gantry,
If ceiling mounted, rotation around 0 centered at the cross-point of the laser sagittal
mounting point (◦ ) and coronal beams.
SID (mm) Maximum d. Move the table +10 cm in the longitudinal direc-
Table height (mm) Near isocenter tion and verify that this motion is correctly
Table travel left to right (mm) Near center recorded on the in-room display.
RAO, right anterior oblique; LAO, left anterior oblique.
e. Collect and/or save a digital fluoroscopic run at
this location (Table E2, line 2).
f. Repeat step (c), this time moving the table –
∙ cross-line laser alignment pointer (commonly used 10 cm in the longitudinal direction.
by carpenters); should include a sagittal and coronal g. Collect and/or save a fluoroscopic run at this
beam, location (Table E2, line 3).
∙ plastic ruler, and h. Return the table to the central position and tape
∙ 18-inch level. the ruler in the vertical direction on the C-arm
gantry.
Method: Assess physical/displayed positional param- i. Raise and lower the table ±10 cm and verify that
eters this motion is correctly recorded on the in-room
display.
1. On the acquisition computer, create a new patient j. Collect and/or save a digital fluoroscopic run at
exam and use a unique name that can be transferred each location (Table E2, lines 4–5).
to and later identified on the in-house PACS or other k. Reposition the laser off of the table so that the
means for reading the RSDR file. Select a standard sagittal beam will now cross the table in the lon-
digital acquisition program. gitudinal direction near the center of the imaging
2. Position the C-arm in its most basic configuration plane.
using the suggested parameters shown in Table E1. l. Tape the ruler along the axis of the table and
centered to the laser beam.
Note that the manner in which these parameters are m. Move the table ±10 cm in the lateral dimension
displayed to users will vary across systems. Users must and verify that this motion is correctly recorded
familiarize themselves with these differences in dis- on the in-room display.
play. n. Collect and/or save a digital fluoroscopic run at
each location (Table E2, lines 6–7).
3. Establish the Center of rotation. 5. Ensure center position accuracy.
a. Ensure that the tabletop is level in both the lateral a. Return the gantry setup to the parameters
(Z) and longitudinal (X) dimensions. described in Table E1.
b. Define a centerline extending the entire length of b. Place the framing square on the table so that the
the table. BBs on the vertical axis are normal to the imaging
c. Establish a center-of -rotation axis with an atten- plane.
uating rod placed in the center of the imag- c. Carefully position the triangle until an image is
ing FOW. Adjust the table height and longitudi- yielded with the BBs superimposed. Any dis-
nal position until the rod stays in the center of placement of the BBs could indicate an inaccu-
the image while the C-arm is rotated around the racy in the gantry rotation.
table. d. Collect and/or save a digital fluoroscopic run at
4. Ensure table motion accuracy. this location (Table E2, line 8).
Table E2 provides a reference list of the gantry and 6. Ensure RAO/ LAO accuracy.
table positions at which exposures should be made a. Rotate the C-arm 45◦ in the RAO direction.
for verification of the geometric parameters. Refer- b. Place the framing square on the table so that the
ence to Table E2 will be made throughout the fol- BBs on the hypotenuse are normal to the imaging
lowing text. plane.
a. Collect and/or save a digital fluoroscopic run at c. Carefully position the triangle until an image
this location (Table E2, line 1). is yielded with the BBs superimposed. Any
b. Position the laser pointer on the table such that displacement of the BBs in the Y direction
the sagittal beam is along the centerline of the could indicate an inaccuracy in the C-arm
table and the coronal beam is at the height of the rotation.
TA B L E E 2 List of exposures required for verification of gantry positioning
Physical location of the image receptor relative to the patient Table position
Rotation around
Exposure RAO/LAO Cranial/caudal mounting pointa Longitudinal Lateral Vertical
1 0 0 Center Center Center Center

2 0 0 Center +10 cm Center Center
3 0 0 Center –10 cm Center Center
4 0 0 Center Center Center +10 cm
5 0 0 Center Center Center –10 cm
6 0 0 Center Center +10 cm Center
7 0 0 Center Center –10 cm Center
8 0 0 Center Center Center Center
9 45◦ RAO 0 Center Center Center Center
10 45◦ LAO 0 Center Center Center Center
11 0 45◦ cranial Center Center Center Center
12 0 45◦ caudal Center Center Center Center
13 0 0 90◦ to left Center Center Center
14 45◦ RAO 0 90◦ to left Center Center Center
15 45◦ LAO 0 90◦ to left Center Center Center
16 0 45◦ cranial 90◦ to left Center Center Center
17 0 45◦ caudal 90◦ to left Center Center Center
18 0 0 90◦ to right Center Center Center
19 45◦ RAO 0 90◦ to right Center Center Center
20 45◦ LAO 0 90◦ to right Center Center Center
21 0 45◦ cranial 90◦ to right Center Center Center
22 0 45◦ caudal 90◦ to right Center Center Center
RAO, right anterior oblique; LAO, left anterior oblique.
a
If the unit is so equipped.
d. Collect and/or save a digital fluoroscopic run at b. Collect and/or save a digital fluoroscopic run at
this location (Table E2, line 9). this location (Table E2, line 13).
e. Repeat for C-arm positions of 45◦ LAO. c. Repeat steps 6 and 7 for determining the
f. Collect and/or save a digital fluoroscopic run at RAO/LAO and cranial/caudal positioning at this
this location (Table E2, line 10). gantry angle (Table E2, lines 14–17).
7. Ensure cranial/caudal accuracy. d. Reset RAO/LAO and cranial/caudal to 0◦ and
a. With RAO/LAO reset to 0◦ , rotate the C-arm 45◦ rotate the gantry 90◦ to the right around the
in the caudal direction. mounting point.
b. Place the framing square on the table so that the e. Collect and/or save a digital fluoroscopic run at
BBs on the hypotenuse are normal to the imaging this location (Table E2, line 18).
plane. f. Repeat steps 6 and 7 for determining the
c. Carefully position the triangle until an image is RAO/LAO and cranial/caudal positioning at this
yielded with the BBs superimposed. Any dis- gantry angle (Table E2, lines 19–22).
placement of the BBs in the Y direction could 9. Close out the exam and send the RDSR data to a
indicate an inaccuracy in the C-arm rotation. structured report reader.
d. Repeat for C-arm positions of 45◦ cranial. 10. View the RDSR data file on the structured report
e. Collect and/or save a digital fluoroscopic run at reader and compare the recorded geometrical posi-
each location (Table E2, lines 11–12). tion values with those set at the time of measure-
8. Ensure overhead gantry rotation accuracy (for ment. The DICOM parameter names are listed in
ceiling-mounted systems). Table E3. The values from the RDSR file should
a. With RAO/LAO and cranial/caudal reset to 0◦ , substantially match those recorded in the room and
rotate the gantry 90◦ to the left around the mount- those in the exam protocol file. Any discrepancies
ing point. should be noted and discussed with the system
engineer.
TA B L E E 3 DICOM parameters employed for positioning
Geometrical parameter DICOM tag Unit Values In DICOM header
Distance source to detector (0018, 1110) mm Absolute √

Distance source to patient (0018, 1111) mm Absolute √
Positioner primary angle (0018, 1510) ◦ Movement from RAO to vertical is √
positive
Positioner secondary angle (0018, 1511) ◦ Movement from caudal to vertical is √
positive
Patient position (0018, 5100) text HFS, HFP, FFS, FFP, and so on √
Distance source to isocenter (0018, 9402) mm Absolute
Tabletop vertical position (300A, 0128) mm Table motion downward is positive
Tabletop longitudinal position (300A, 0129) mm Table motion toward LAO is positive
Tabletop lateral position (300A, 012A mm Table motion toward cranial is positive
FFP, feet-first-prone; FFS, feet-first-supine; HFP, head-first-prone; HFS, head-first-supine; LAO, left anterior oblique; RAO, right anterior oblique.
REFERENCES FOR APPENDIX D 2. Dehairs M, Bosmans H, Marshall NW. Implementation

1. Gordon CL III, Relihan GF, inventors; General Elec- of a spatio-temporal figure of merit for new automatic
tric Co, assignee. Image quality optimization using an dose rate control regimes in dynamic x-ray imaging.
X-ray model based optimization. US patent 6222907 Phys Med Biol. 2019;64:045001.
B1. April 24, 2001.

TG 272 - O-Arm

Uploaded by

Copyright:

Available Formats

TG 272 - O-Arm

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TG 272 - O-Arm

Uploaded by

Copyright:

Available Formats

Received: 8 April 2021 Revised: 7 December 2021 Accepted: 13 December 2021

AAPM SCIENTIFIC REPORT

AAPM Task Group Report 272: Comprehensive acceptance

Pei-Jan Paul Lin1 Allen R. Goode2 Frank D. Corwin1 Ryan F. Fisher3

Med Phys. 2022;49:e1–e49. wileyonlinelibrary.com/journal/mp © 2022 American Association of Physicists in Medicine e1

The American Association of Physicists in Medicine 1.1 Disclaimer and precautions

physicists performing regulatory testing of fluoroscopes 1.3 Fluoroscopy systems included in

1.4 Sample form The system microenvironment should be evaluated

∙ the X-ray source-to-image receptor distance (SID),

The accuracy of the positional parameters listed

F I G U R E 3 Determine the center line.

neurologic and visceral systems, the installation con-

NOTE 2:There are biplane IXE systems in which the

For guidance regarding the assessment of tube poten-

F I G U R E 7 Setup for measuring maximum

2.6.1 TG 190 dose correction/calibration

The correction factor obtained from the dose-rate mode

CTG190 = DMEASURED ∕DPERP . (1)

Empirically speaking, DPERP is often higher than

2.6.2 Examination table/mattress

= DTable_Mattress ∕Dair . (4)

= CTG190 × CTable × CMattress . (3) And similarly:

TA B L E 1 Correction factors for TR through table and mattress

Tube potential (kVp)

multiple images. Viewing single fluoroscopic or acqui-

TA B L E 2 Integral uniformity. Single-shot image using central ROI

Maximum Minimum Integral

F I G U R E 1 5 A photographically enhanced image of the artery

free from artifacts, dropouts, or nonuniformities cre-

(Maximum pixel value) − (minimum pixel value)

TA B L E 3 HCR testing results from fluoroscopy and acquisition modes

FOV size (cm) 48 42 32 22 16 11

FOV size (cm) 48 42 32 22 16 11

This section refers to both mobile C-arms and intraop-

3.3.5 Displayed PERP dose verification

mobile fluoroscopy unit; thus, as of the time of this pub-

Tests for 2D fluoroscopy can be performed following the

3.8.2 3D acquisition 4 CONVENTIONAL FLUOROSCOPY

F I G U R E 2 0 The tableside control panel for the examination

F I G U R E 2 2 The experimental setup for measuring radiation

4.3 Radiation output measurements

4.3.1 Radiation output as a function of

The experimental setup for measuring radiation output

For conventional fluoroscopy systems in which the X-ray

4.2 X-ray generator and radiological

If the grid can be removed, a geometrical correction can

and move them as one assembly. The images in Fig-

5.2 X-ray generator and radiological

5.1.1 Tabletop motion

5.1.2 Verification of SID 5.3.2 Maximum AKR

F I G U R E 2 4 Left: Longitudinal translation

F I G U R E 2 6 Software control panel showing nine selections of

determining whether the area selected is still being

5.5 Display monitors ∙ The SID should be <45 cm.

As mentioned previously, the practical measurement of

6.3.1 Radiation output as a function of

A P P E N D I X B : G E N E R ATO R T E S T P O I N T S test points varies depending on the specific generator

TA B L E E 2 List of exposures required for verification of gantry positioning

1 0 0 Center Center Center Center