CHAPTER

14
Ultrasound Equipment Quality
Assurance
James A. Zagzebski and James Kofler

KEY TERMS
axial resolution phantom string test
depth of visualization scan image uniformity vertical distance measurement
horizontal distance measurement sensitivity
lateral resolution slice thickness

OBJECTIVES
At the completion of this chapter the reader will be able to do the following:
• Discuss the importance of quality assurance for • Explain the importance of documentation of quality
ultrasound equipment assurance testing
• Describe the various phantoms used in ultrasound • Describe the basic quality control testing for Doppler
quality assurance color flow equipment
• Identify the basic quality control tests for ultrasound

OUTLINE
Components of an Ultrasound Photography and Gray-Scale Spatial Resolution Tests 264
Quality Assurance Program 257 Hard Copy 261 Axial Resolution 264
Quality Assurance and Preventive Monitor Setup and Recording Lateral Resolution 265
Maintenance 257 Devices 261 Cautions About Resolution Tests
Tissue-Mimicking Phantoms 257 Routine Quality Assurance of with Discrete Targets 265
Tissue Properties Represented in Image Recording 261 Other Test Objects and
Phantoms 257 Scan Image Uniformity 262 Phantoms 266
Typical Quality Assurance Distance Measurement Anechoic Voids 266
Phantom Design 257 Accuracy 263 Objects of Various
Cautions About Phantom Vertical Distance Echogenicity 266
Desiccation 258 Measurements 263 Spherical Object Phantom 267
Basic Quality Control Tests 259 Horizontal Distance Doppler Testing 267
Visual Inspection 259 Measurements 264 String Test Objects 267
Transducer Choice 259 Other Important Instrument Doppler Flow Phantoms 267
System Sensitivity 260 Quality Assurance Tasks 264 Electronic Probe Tests 268
Documentation 264

In an imaging facility, quality assurance is a process car- might require service. Thus in some ways, ultrasound
ried out to ensure that equipment is operating consis- equipment quality assurance is carried out every day,
tently at its expected level of performance. During even when it is not identified as a process itself.
routine scanning each sonographer is vigilant for equip- Quality assurance steps to be discussed here go
ment changes that can lead to suboptimal imaging and beyond judgments of scanner performance that are

256
 CHAPTER 14 Ultrasound Equipment Quality Assurance 257

made during routine ultrasound imaging. They involve ultrasonic scattering level). Phantoms cannot exactly
prospective actions to identify problem situations, replicate the acoustic properties of soft tissues.
even before obvious equipment malfunctions occur. This is partially due to the complexity and variability
Quality assurance testing provides confidence that of tissues. Instead, phantom manufacturers construct
image data such as distance measurements and area esti- these objects to have acoustic properties that represent
mations are accurate and that the image is of the best the average properties of many different tissues. Some-
possible quality from the imaging instrument. times the term tissue-equivalent is used when phantoms
are described; however, this term should not be inter-
preted literally because most phantom materials are
COMPONENTS OF AN ULTRASOUND not acoustically equivalent to any specific tissue.
QUALITY ASSURANCE PROGRAM

Quality Assurance and Preventive Typical Quality Assurance Phantom Design

Maintenance An example of a general purpose ultrasound quality
Various approaches are used by ultrasound facilities assurance phantom is shown in Figure 14-1. Such phan-
when setting up a quality assurance program for their toms are examined with scanner settings that are similar
scanners. Sometimes these programs include both pre- to those used when patients are being scanned.
ventive maintenance procedures performed by trained The phantom images have gray-scale characteristics that
equipment service personnel and in-house testing of are analogous to characteristics of organs, although the
scanners with phantoms and test objects. Some facilities actual structures are not anatomically represented.
rely on only one of these measures. For preventive main- Figure 14-1, B shows the internal structure of this
tenance, emphasis is usually given to invasive electronic phantom. The tissue-mimicking material within the
testing of system components such as voltage measure- phantom consists of a water-based gelatin in which
ments at test points inside the scanner. Sometimes pre- microscopic particles are mixed uniformly throughout
ventive maintenance also involves an assessment of the the volume (Burlew et al., 1980; Madsen et al., 1978).
imaging capability by scanning a phantom. The speed of sound in this material is about 1540
In-house scanner quality assurance programs usually m/sec, the same speed assumed in the calibration of
involve imaging phantoms or test objects and assessing ultrasound instruments. The ultrasonic attenuation coef-
the results. In-house tests may be performed by sonogra- ficient versus frequency is one of two values: either 0.5
phers, physicians, medical physicists, clinical engineers, dB/cm per megahertz or 0.7 dB/cm per megahertz
or equipment maintenance personnel. Detailed recom- (Box 14-1). Some users prefer the lower-attenuating
mendations from professional organizations and experts material because they find it easier to image objects in
in ultrasound on establishing an in-house quality assur- the phantom. However, standards groups recommend
ance program are available elsewhere (ACR Ultrasound the higher attenuation because it challenges machines
Accreditation Program, 2001; Goodsitt et al., 1998; more thoroughly (Zagzebski, 2000).
Zagzebski, 2000). Attenuation in the gel-graphite material in the phan-
tom is proportional to the ultrasound frequency and
mimics the behavior in tissues (Lu et al., 1999; Madsen
Tissue-Mimicking Phantoms et al., 1978; Maklad et al., 1984). Other types of mate-
In-house scanner quality assurance tests most often are rials have been used in phantoms, but only water-based
performed with tissue-mimicking phantoms. In medical gels laced with powder have both speed of sound and
ultrasound a phantom is a device that mimics soft tissues attenuation with tissuelike properties (Madsen et al.,
in its ultrasound transmission characteristics. Phantoms 1978; Zagzebski, 2000).
represent “constant patients,” and images can be taken Small scatterers are distributed throughout the tissue-
at different times for close comparison. Image penetra- mimicking material; therefore, the phantoms appear
tion capabilities, for example, are readily evaluated for echogenic when scanned with ultrasound imaging
changes over time when images of a phantom are avail- equipment (see Fig. 14-1, C). Many phantoms have
able for comparison. Phantoms also have targets in simulated “cysts,” which are low-attenuating, nonecho-
known positions, so images can be compared closely with genic cylinders. These should appear echo free on
the region that is scanned. Examples include simulated B-mode images and should exhibit distal echo enhance-
cysts, echogenic structures, and thin “line targets.” ment. Some tissue phantoms provide additional image
contrast by having simulated masses or test objects of
varying echogenicity. Such objects are evident in Fig-
Tissue Properties Represented in Phantoms
ure 14-1, C.
Tissue characteristics mimicked in commercially avail- Most quality assurance phantoms also contain dis-
able phantoms are the speed of sound; ultrasonic atten- crete reflectors such as nylon-line targets to be used
uation; and, to some degree, echogenicity (i.e., the mainly for evaluating the distance measurement
258 CHAPTER 14 Ultrasound Equipment Quality Assurance

B C

FIGURE 14-1 Example of a general-purpose quality assurance phantom. A, Phantom being imaged with an ultrasound scanner. B, Close-up
of phantom, with diagram of interior contents. C, B-mode image of the phantom.

introduced in calibrations. These phantoms also rely

BOX 14-1 Tissue Attenuation Coefficients
on the manufacturer having defined the reflector
Attenuation coefficients are normally specified in decibels per cen-
positions accurately. With the correct speed of sound
timeter. To include the dependence of attenuation on frequency,
(1540 m/sec) and precisely known distances between
phantom manufacturers divide the attenuation coefficient by the
frequency at which the measurement is done. This yields units of pointlike reflectors, it is easy to check the accuracy
decibels per centimeter per megahertz. Strictly speaking, this of distance measurements with calipers, as described
approach should be used only when attenuation is directly propor- later.
tional to the frequency, as we often assume for tissues. The value Phantoms often contain a column of reflectors, each
of 0.7 dB/cm per megahertz is representative of the attenuation separated by 1 or 2 cm, for vertical measurement accu-
coefficient in difficult-to-penetrate fatty liver.* The depth that racy tests. One or more horizontal rows of reflectors
structures can be visualized within tissue-mimicking material hav- are used for assessing horizontal measurement accuracy.
ing this amount of attenuation more closely correlates with clinical Additional sets of reflectors may be found for assessing
penetration. the axial resolution and the lateral resolution of
*Lu ZF, Lee FT, Zagzebski JA: Ultrasonic backscatter and attenuation in scanners.
diffuse liver disease, Ultrasound Med Biol 25:1047, 1999.
Cautions About Phantom Desiccation
accuracy of a scanner. Tests of the accuracy of distance When a phantom made of water-based gels is used, loss
measurements rely on the manufacturer of the phantom of water (desiccation) may become a problem as the
to have filled the device with a material with a sound phantom ages. If this occurs, the speed of sound in
propagation speed of 1540 m/sec or at least close the phantom may have changed. A scanning surface that
enough to this speed that no appreciable errors are has become concave is an indication of severe