J Mammary Gland Biol Neoplasia (2006) 11: 113123

DOI 10.1007/s10911-006-9018-0

A Review of Breast Ultrasound

Chandra M. Sehgal & Susan P. Weinstein &
Peter H. Arger & Emily F. Conant

Published online: 4 November 2006

# Springer Science + Business Media, Inc. 2006

Abstract Frequent advances in transducer design, electronics, computers, and signal processing have improved
the quality of ultrasound images to the extent that
sonography is now a major mode of imaging for the
clinical diagnosis of breast cancer. Breast ultrasound is
routinely used for differentiating cysts and solid nodules
with high specificity. In combination with mammography,
ultrasound is used to characterize solid masses as benign or
malignant. There is growing interest in using Doppler
ultrasound and contrast agents for measuring tumor blood
flow and for imaging tumor vascularity. Ease of use and
real-time imaging capability make breast ultrasound a
method of choice for guiding breast biopsies and other
interventional procedures. Breast ultrasound is used in
many forms. B-mode is the most common form of imaging
for the breast, although compound imaging and harmonic
imaging are being increasingly applied to better visualize
breast lesions and to reduce image artifacts. These developments, together with the formulation of a standardized
lexicon of solid mass features, have improved the diagnostic performance of breast ultrasound. Several approaches
that are currently being investigated to further improve
performance include: (1) computer-aided-diagnosis; (2) the
assessment of tumor vascularity and tumor blood flow with
Doppler ultrasound and contrast agents; and (3) tissue
elasticity imaging. In the future, ultrasound will play a
greater role in differentiating benign from malignant masses
and in the diagnosis of breast cancer.

C. M. Sehgal (*) : S. P. Weinstein : P. H. Arger : E. F. Conant

Silverstein, Department of Radiology, University of Pennsylvania,
3400 Spruce Street, Philadelphia, PA 19104, USA
e-mail: sehgalc@uphs.upenn.edu

Keywords Breast cancer . Ultrasound imaging . Neoplasm .

Sonography
Abbreviations
CAD computer-aided-diagnosis
DCIS ductal carcinoma in situ
ROC receiver operating characteristics

Introduction
Since early attempts in 1953 by Wild and Reid to build a
real-time handheld ultrasound scanner to image breast
lesions [1], ultrasonic imaging has undergone several
transformations, both in instrument design as well as in
clinical applications. In the 1980s, dedicated water bath
scanners using ultrasound-computed tomography (UCT)
gained popularity for imaging the breast [27]. These
scanners provided high resolution images in the reflection
and transmission modes, but they did not prove to offer
significant clinical advantages over real-time sonography
due to several factors [8], including the inconvenience of
using water baths for scanning and significant refraction
artifacts in the transmission mode. Today, handheld systems
are predominantly used in clinics for scanning patients.
Modern ultrasound scanners are digital systems. The
scanners digitize the acoustic signal immediately after it is
received by the transducer, and focus and steer the beam
using high speed digital electronics. Unlike early scanners,
which constructed images by a moving single-element transducer, modern sonographs use a multi-element ultrasound
transducer that does not involve physical movement of the
imaging transducer. The beam is focused and steered
electronically by broadcasting ultrasound from the member
elements of the multi-element array at different times, so that

114

the sound wave from each element arrives at the intended

focal point in the tissue simultaneously.
Modern ultrasound scanners offer various modes of
imaging. The most common mode is grayscale or B-mode
imaging, which uses ultrasound energy scattered by the
tissue back to the transducer for constructing images.
Doppler vascular imaging, on the other hand, uses the
acoustic signal scattered by moving blood to construct
images. There are also modes of imaging that use harmonics
of the echo signal or the shift in echo signal from tissue
motion to construct images. All these imaging modes are
integrated on a single scanner, and users are able to switch
from one mode to another easily. Information on both tissue
architecture and blood flow can thus be obtained in a single
study.
B-Mode Ultrasonography
The most common form of ultrasound imaging used in
breast diagnosis is B-mode ultrasonography [911]. It is
based on broadcasting short bursts of ultrasound energy
from transducer elements, then listening to the echoes from
tissues at different depths along the path of ultrasound
propagation. The strength (amplitude) of the echo received
from the tissue is used to control the brightness (B-mode)
of the display monitor. The process of ultrasound pulse
transmission and echo reception is repeated by aiming the
ultrasound pulses in different directions. The echoes
received from all directions are combined to construct a
2D image of the breast. Since each pulse transmission and
echo reception occurs in less than a millisecond, as many as
30 images can be obtained in each second. These images,
when displayed in rapid succession, allow real-time
imaging of the breast.
The brightness of an ultrasonic image represents a
complex combination of sonographic properties of the
tissue and of the ultrasound equipment. In essence, the
equipment adds its own signature to the displayed images.
If we neglect the influence of the equipment, the brightness
of the sonographic image is determined by the size and by
shape of the acoustic inhomogeneities, as well as the
mechanical properties (tissue density and tissue compressibility) of the tissue.
Many factors affect image quality and the ability of an
ultrasound system to display a lesion. They include design
of the transducer (frequency, bandwidth, and aperture),
acoustic properties of the tissues (frequency dependent
attenuation, regional variation in sound speed and tissue
density, and speckle caused by the interference of waves),
signal processing, and the nature of the display monitor.
Because of the complex interactions between various
factors, highly specialized skills are required for interpreting B-mode images.

J Mammary Gland Biol Neoplasia (2006) 11: 113123

Ultrasound images can reveal different soft tissue layers

within the breast. The skin is highly reflective and appears as a
bright line in the image. Lactiferous ducts are easily seen,
especially when they are dilated. Coopers ligaments, supporting glandular structures of the breast, are also commonly
seen as thin echogenic fibers. Fatty tissues cause low-level
echoes and appear dark in the image. Conversely, parenchymal and fibrous tissues reflect ultrasound strongly and appear
bright. The parenchyma to fat ratio varies considerably among
patients, as well as within a patient, depending on her age and
hormone levels. Breast lesions generally appear darker
(hypoechoic) than surrounding tissues.
The main applications of breast ultrasound are (1) to
differentiate between cystic and solid lesions, (2) to
evaluate palpable masses not visible mammographically,
and (3) to evaluate young and pregnant patients with
palpable masses. Some reports have also suggested the use
of ultrasound to determine lymph node status [1218].
Although simple cysts can be diagnosed with a high
sensitivity of 98100% [19], the characterization of
complex cysts filled with echogenic fluid can be difficult.
Hogg et al. found that as many as 33% of the lesions
considered to be inderminate were cystic on aspiration [20].
Nightingale et al. proposed the use of ultrasonically
induced acoustic streaming to differentiate fluid-filled and
solid lesions in the breast [21].
Until recently, the use of ultrasound to differentiate
between benign and malignant solid masses and for
screening asymptomatic patients was not recommended.
While ultrasound is still not recommended for screening,
there is growing evidence that it can detect clinically and
mammographically occult cancers [2224]. These findings
have rekindled interest in using breast ultrasound for
screening breast masses, and there has been a call from
investigators to conduct clinical trials evaluating the efficacy
of breast ultrasound as a screening tool [25]. Furthermore,
computer methods are being developed that analyze acoustic
shadowing for detecting lesions on breast sonograms. Early
results are promising, and, in the future, computerized
approaches may facilitate ultrasound screening [26].
Since the early 1990s, there has been significant progress
in the use of ultrasound imaging for distinguishing
malignant and benign masses [2730]. It is now common
practice to use sonography to aid in mammographic
diagnosis of solid masses. Several sonographic features
based on margin, shape, and echotexture have been proposed
for the diagnosis of breast nodules [12, 13, 28]. Malignant
nodules are often characterized by poorly defined margins
and irregular borders. Spiculation, angular margins, microlobulations, marked hypoechogenicity, shadowing, duct
extension, and tissue architectural distortion are some of
the common features associated with malignant masses.
Benign lesions, on the other hand, are often well differen-

J Mammary Gland Biol Neoplasia (2006) 11: 113123

115

tiated from the surrounding tissue by a well-defined,

circumscribed margin. They tend be round or oval with
gentle bi- or tri-lobulations. Figure 1 shows sonographic
images of malignant and benign breast lesions.
Table 1 describes the performance of individual sonographic features in two different studies using the criteria
originally outlined by Stavros et al. [28]. The results show
that the individual features have high levels of sensitivity,
specificity, positive predictive value (PPV), and accuracy.
Using some of the same features, Arger et al. [31] evaluated
inter-reader variability in diagnosis. Arger et al. concluded
that the interpretations of four different readers were very
similar when standardized descriptions of breast masses
were used for differentiating malignant and benign masses;
the diagnostic performance of individual features was
comparable to the performance reported in the earlier study
by Stavros et al. (Table 1). In another study, Paulinell et al.
added the thickness of the Coopers ligament to the
conventional margin, internal echo, and shadowing features, concluding that poorly circumscribed margins,
heterogeneous echo pattern, and thickened Coopers ligaments indicate higher probability of malignancy [30].
More recently, the American College of Radiology has
instituted a lexicon, US-BIRADS, to standardize the
process for reporting breast masses [3235]. Using the
BI-RADS lexicon, Hong et al. demonstrated that the BIRADS features of margin, shape, orientation, lesion
boundary, echo pattern, and posterior acoustic shadowing
were significantly different for malignant and benign
masses [35].
The clinical analysis of breast images is currently almost
exclusively done by a qualitative assessment of lesion features

within an image. Variations in human perception of the

images, differences in features used for diagnosis, and, more
fundamentally, a lack of quantitative measurements of the
features used for image analysis are some of the factors that
cause variability in diagnosis by individuals. Consequently,
the final confirmation of benign or malignant diagnosis
usually requires a biopsy. The large number of negative
biopsies results in unnecessary stress and cost to the patient
and the system.
While computer technology has had a tremendous
impact on the design and performance of medical imaging
systems, this technology has not yet been used to improve
the diagnostic performance of individuals reviewing the
images. The development of computer-aided diagnosis
(CAD) could address this limitation and, when used with
routine clinical evaluation, could improve the specificity of
diagnosis. CAD systems apply intelligent algorithms that
use mathematically sophisticated classifiers and learning
algorithms to recognize complex patterns in images. When
such systems are successfully implemented, computers can
be increasingly used to help differentiate benign and
malignant tumors. With the wide acceptance of mammographic CAD and the growing acceptance of magnetic
resonance CAD, interest in ultrasound CAD has also
increased. Several studies on sonographic CAD have
shown that computerized analysis can aid in the interpretation of the images [3651].
CAD is a multi-step process. It involves identification of
lesions on the image either manually or by computerized
segmentation. Various echo texture and morphometric
(shape and margin) features are extracted from the
segmented masses. Through sophisticated mathematical

Figure 1 Sonograms of biopsy proven malignant and benign masses.

(a) The infiltrating ductal carcinoma on the left has an indistinct
margin with an irregular border. The nodule is markedly hypoechoic,
taller-than-wide, and casts a strong shadow. (b) The fibroadenoma on

the right is well encapsulated and has a sharp, a well-defined margin.

The interior has weak heterogeneous echogenicity. The mass is
shorter-than-wide. Except for the weak shadowing at the edges, due to
refraction, the mass does not have posterior shadowing.

116

J Mammary Gland Biol Neoplasia (2006) 11: 113123

Table 1 Sensitivity, specificity, PPV, and accuracy of nodule features measured by two independent studies [28, 31].
Feature

Spiculation
Taller than wider
Angular margins
Shadowing
Branching pattern
Hypoechogenicity
Calcifications
Duct extentions
Microlobulation

Sensitivity

Specificity

PPV (%)

Accuracy

[28]

[31]

[28]

[31]

[28]

[31]

[28]

[31]

36.0
41.6
83.2
48.8
29.6
68.8
27.2
24.8
75.2

59 11
49 18

99.4
98.1
92.0
94.7
96.6
90.1
96.3
95.2
83.8

97 5
91 4

91.8
81.2
67.5
64.9
64.0
60.1
59.6
50.8
48.2

97 5
90 4

88.8
88.7
90.5
87.1
85.5
87.2
84.8
79.3
82.4

73 8
64 12

59 8
72 10
48 17
51 20

77 1
86 5
95 5
84 3

82 2
91 2
95 5
82 10

65 5
77 6
65 12
63 12

In [28] the term markedly hypoechoic instead of hypoechogenicity. Values quoted in [28] represent mean STD of four observers.
PPV Positive predictive value, STD standard deviation

pattern recognition techniques, computers are trained to

assign a probability of malignancy. While there are
considerable differences between studies, the various
approaches described in the literature can be classified into
those that emphasize echo texture [38, 4044] and those that
use shape and margin features [4448]. Both approaches
yield comparable diagnostic performance for characterizing
solid masses. The area under the ROC curve, Az, is a metric
that measures diagnostic performance. An Az of 1 represents
perfect performance. For sonographic CAD, Az is between
0.83 and 0.87 for the majority of studies [37, 4345, 47,
48, 51], with some studies reporting a higher performance
of 0.92 [44] and near perfect values of 0.950.98 [46].
Only limited data are available on the diagnostic performance of sonographers. In the study by Arger et al. the Az
value for three radiologists ranged between 0.90 and 0.92,
though one radiologist had an Az of 0.97 [31]. In another
study, the Az for radiologists ranged between 0.84 and 0.92
[44]. These data suggest that sonographic CAD computer
performance is comparable to human performance.
Kuo et al. studied the effect of ultrasound scanners on
CAD performance and observed the properties of the
ultrasound scanners did not influence the CAD performance significantly [36]. Horsch et al. showed that the
performance of an expert sonographer improved when
CAD was used in the interpretation of sonographic images
of the breast [37]. While the results of sonographic CAD
are encouraging, it is important to note that all studies were
performed on static images using complex and time
consuming off-line analyses. Lack of real-time analysis
has, at least in part, made the widespread application of
sonographic CAD systems impractical. This may change as
manufacturers begin to incorporate CAD as built-in features
on ultrasound scanners.
Microcalcification imaging Mammographically detected
microcalcifications with a diameter on the order of 50 m

may represent an early cancer, ductal carcinoma in situ

(DCIS). A limitation of breast ultrasound has been its
inability to detect small calcifications. The complex
structure of breast tissue and the grainy noise in the image
due to speckle phenomena often mask microcalcifications,
and their detection with B-mode ultrasound is unreliable.
Compound imaging, discussed below, suppresses speckle
and tissue noise and improves the visibility of breast
calcifications. Also, promising new techniques based on
exciting calcifications by external vibrations [52], or by
radiation force [53], are being developed to better visualize
such microcalcifications. However, the studies to date have
been performed largely using tissue-mimicking phantoms
and excised tissues.
Interventional sonography The absence of radiation, the
real-time nature of the imaging, and the ability to visualize
masses often make B-mode imaging the method of choice
for interventional procedures. Ultrasound is commonly
used for guiding biopsy needles for both aspiration and
core biopsies. Figure 2 shows an example of biopsy
imaging. Ultrasound imaging is also being used for surgical
localization of non-palpable masses, and for the placement
of percutaneous ablation devices such as radiofrequency
electrodes [12, 54].

Compound Imaging
Despite advances, B-mode images contain inherent artifacts
which degrade image quality. Speckle due to interference of
coherent ultrasound waves [55] causes small-scale brightness fluctuations, giving a granular appearance to otherwise
homogeneous tissue. The side lobes and grating lobes of
ultrasound beams result in multi-path reverberations that
often lead to spurious echoes from the tissues. High

J Mammary Gland Biol Neoplasia (2006) 11: 113123

Figure 2 Sonogram showing an ultrasound-guided core needle

biopsy. The needle (arrow), a strong reflector of ultrasound, is seen
approaching the surface of the target nodule (dotted arrow),
represented by the hypoechoic region in the image.

attenuation of ultrasound by solid masses can cause intense

shadowing and obscure breast anatomy posterior to the
lesion. Also, breast structures like connective tissues and
lactiferous ducts act as specular reflectors and behave as
sound wave mirrors. A small change in the direction of
ultrasound propagation changes the angle of incidence from
normal to oblique, which can dramatically alter the echo
signal from very strong (causing saturation) to very weak
(causing the poor visibility). In addition, because the lateral
edges of breast lesions are parallel to the scan lines, they do
not reflect ultrasound back to the transducer. This makes
the lateral margins and border of subtle lesions often
difficult to detect. Taken together, all these factors reduce
the conspicuity of small lesions and limit the delineation of
the margins of larger lesions.
To decrease the incidence of artifacts, several investigators have proposed the use of compound imaging. In
conventional sonography, ultrasound is transmitted in a
single direction perpendicular to the ultrasound transducer
surface. In compound imaging, multiple views of a given
region of tissue are obtained from different angles. The
different views are combined to form an image. This is
demonstrated in Fig. 3. The different views are obtained by
keeping the ultrasound transducer fixed and steering the
ultrasound beam electronically in off-axis directions within
the plane of imaging. The compound image is updated as
each new frame is acquired, enabling real-time imaging.
This averaging of multiple frames suppresses the artifacts
related to speckle, spurious noise, and shadowing. In
addition, it improves the definition of lesion margins in
the direction orthogonal to the axis of the transducer.
Figure 4 compares conventional B-mode and compound
images of a breast lesion.

117

Several clinical studies have demonstrated the benefits of

compound imaging. Entrekin et al. [56] observed that the
echo amplitude and texture of fat is markedly different from
glandular tissue in compound imaging. The lateral margins
of the lesion were well delineated; the suppression of
speckle revealed punctuated calcifications; the images had
fewer artifacts; and there was better agreement among
observers in identifying lesions. Kwak et al. studied normal
breast structures and abnormal lesions by conventional and
compound imaging [57] and concluded that, compared to
conventional B-mode images, real-time compound images
have better spatial and contrast resolution. In compound
scans the margins of the lesions and their internal
architecture are better delineated. The suppression of
speckle noise by image compounding enhances the visualization of microcalcifications. Kangas et al. observed that
compound imaging increases contrast-to-noise ratio and
lesion signal-to-noise ratio for both benign and malignant
masses by as much as 3040% [58].
Despite the benefits of compound imaging, conventional
B-mode imaging continues to be the most common
clinically used method. The reason for this is not clear,
but is perhaps related to the users familiarity with the
texture of B-mode images, and the fact that the shadowing
suppressed in compound imaging is often used as one of
the features for differentiating benign and malignant
masses. Another potential drawback of compound imaging
is that if the transducer is moved rapidly during scanning,
the frame averaging during compounding can cause motion
blurring.
(a)

(b)

(c)

(d)

(e)

Figure 3 This figure illustrates the concept of spatial compounding of

ultrasound images. The conventional B-mode image is constructed
with a single scan of the beam generated at 90 to the surface of the
beam as shown in panel (b) of the figure. In compound imaging, three
to nine views are obtained by steering the beam at different angles, as
shown in panels (a), (b) and (c). The frames in panels (a), (b) and (c)
are combined to generate a single frame (e). After displaying
compound image (frame e), the frame (a) is dropped, a new frame
(d) is acquired, and frames (b), (c) and (d) are compounded to form a
new image. Since updating of each compound image requires
acquisition of only one new frame, compound imaging can be
performed in real time, similar to conventional B-mode imaging.

118

J Mammary Gland Biol Neoplasia (2006) 11: 113123

Figure 4 Images of biopsy-proven DCIS obtained using (a) conventional B-mode and (b) compound mode. The tumor margin and the
internal structure of the lesion are better delineated in the compound

image. Compound imaging also improves the visualization of

suspicious microcalcifications (arrow) which are difficult to identify
in the B-mode image.

Harmonic Imaging

[59, 60]. The technique has been used successfully for

imaging axillary lymph nodes [61] and for evaluating
neoadjuvant chemotherapy responses in breast cancer
patients [62].
Although thus far harmonic imaging has been primarily
used to improve image quality, laboratory studies show that
harmonic signals generated by nonlinear interactions between the tissues and the ultrasound waves represent tissue
properties different from those represented by B-mode echo
signals [63, 64]. Therefore, harmonic imaging could
provide important new information about tissue in clinical
exams, an area that has not yet been explored.
(a)

(c)

123456789111
12
013
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
100
99
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201

c - c

Fundamental

c + c

f0

2f0

H2

(d)

H1

(b)

3f0

Fundamental

The discussion above focuses on grayscale imaging, in

which the frequency of the signal used to construct images
is the same as the transmitted frequency. Advances in
electronics and in our understanding of tissue-ultrasound
interactions have revealed that transmission of ultrasound
through breast tissues leads to the formation of higher
frequencies. These frequencies are multiples of the transmitted frequencies. For example, ultrasound transmission at
a fundamental frequency of 3 MHz leads to the formation
of harmonics at 6, 9, 12 MHz.... In harmonic imaging, the
fundamental frequency is removed and the residual energy
generated by ultrasound-tissue interactions is used for
constructing images.
Harmonic signals are formed because the speed of sound,
c, is not constant, as commonly assumed in conventional Bmode imaging. Instead, c increases with the particle
velocity of the ultrasound wave. As a result, the high
pressure region of the wave travels faster, whereas the low
or negative pressure region of the wave lags behind. This is
demonstrated in Fig. 5. The mismatch between the rate at
which the high and low pressure components of the wave
travel distorts the ultrasound pulse and generates harmonic
frequencies. Harmonic images can be constructed either by
exploiting pulse distortion or by separating the harmonic
signal from the fundamental frequency.
Harmonic imaging makes the ultrasound beam narrower
and suppresses the side bands. The harmonic signal exhibits
less clutter and phase aberration and fewer reverberation
artifacts. All these factors improve the resolution and the
overall quality of the images. For example, harmonic
images of cysts are free of artifacts, simplifying their
diagnosis. Similarly, harmonic imaging also improves the
definition of margins and the detectibility of breast lesions

f0

2f0

3f0

Figure 5 Formation of harmonic signal by nonlinear propagation of

ultrasound. During propagation through the tissue, different points of
the ultrasound pulse (panel a) travel at different sound speeds. The
positive part of the pulse travels faster than the negative part and
distorts the pulse (panel b). The spectrum of the distorted pulse in
panel b, has harmonic signal (panel d). H1 and H2 represent first and
second harmonics.

J Mammary Gland Biol Neoplasia (2006) 11: 113123

119

Figure 6 Tumor vessels imaged by Doppler ultrasound. Biopsy proven DCIS (arrows) imaged with (a) color Doppler and (b) power Doppler
ultrasound. The two images show comparable vascularity.

Doppler Ultrasound Imaging without Contrast Agents

Both color and power Doppler imaging have been used to
characterize breast lesions [6574]. Figure 6 shows
examples of color and power Doppler images. Studies have
consistently reported that tumor blood vessels are visible
with Doppler ultrasound. Cosgrove et al. observed that
while 99% of malignant masses had blood vessels, only 3%
of benign masses showed Doppler signal [65]. Raza and
Baum achieved diagnostic sensitivity and specificity of 68
and 85% using penetrating vessels as indications of
malignancy [67]. However, according to Birdwell et al.
qualitative assessment of images showed no significant
differences in vascularity between malignant and benign
masses [68]; Wilkens et al. also found that Doppler was of
limited value in lesion analysis [73].
Carson et al. used several qualitative measures and
observed that power Doppler features were promising for
characterizing breast lesions [70]. Lee et al. also found
Doppler evaluation helpful in differentiating benign and
malignant masses [72]. Sehgal et al. derived quantitative
measures of tissue vascularity (related to the area covered
by blood vessels per unit area of tumor) from color and
power Doppler images [71]. They observed malignant
masses to be 14 to 54% more vascular than benign masses.
Both types of masses were more vascular than the
surrounding tissues. On average, the benign masses were
2.2 times more vascular than the surrounding tissue, while
malignant masses were five times more vascular. The
regional distribution of tumor vascularity for the two
groups was also different. Whereas tumor vascularity was
equivalent for the core and periphery of the tumor, the
malignant masses had greater vascularity per unit tissue
towards the center of the mass. While the average
vascularity for the malignant and benign groups were
statistically different, there was considerable overlap be-

tween the two groups. As a result, the simple assumption

that higher vascularity indicated higher probability of
malignancy resulted in a weak diagnostic performance
(ROC area Az = 0.560.65). The diagnostic performance
improved significantly Az 0:85
0:06 when the vascularity differences were used in combination with a rulebased decision tree.

Doppler Imaging with Contrast Agents

The use of contrast agents to enhance tissue structures has
changed the way cross-sectional imaging, such as CT and
MRI, is performed. The development of microbubble-based
contrast agents has raised similar hopes for ultrasound
imaging. The microbubble agents are several orders of
magnitude more echogenic than red blood cells. They do
not extravasate from the blood vessels into the neighboring
tissues. Therefore, the enhancement of the image by these
agents indicates the presence of blood vessels. As described
above, Doppler imaging can directly visualize blood vessels
without using contrast agents. However, several technical
and physiological factors prohibit the visualization of blood
vessels smaller than 100200 m. The expectation is that
the use of ultrasound contrast agents will allow the
visualization of smaller vessels (possibly at the capillary
level) with low volume blood blow [75]. Figure 7 shows a
power Doppler image of a breast lesion that appeared
originally to be avascular. The same lesion exhibited
significant vascularity when contrast agent was used.
Kedar et al. observed increased specificity of cancer
characterization from 87.5 to 88.9% [76] with the use of
ultrasound contrast agents. Huber et al. [77], Alamo et al.
[78], and Reiniskainen et al. [79] did not find significant
improvement in tumor characterization by using 2D
imaging and ultrasound contrast agents. However, Forsberg

120

J Mammary Gland Biol Neoplasia (2006) 11: 113123

Figure 7 The left panel shows a power Doppler image of a lesion with
a blood vessel (arrows) leading to the nodule. The lesion (dotted circle)
appears avascular. The right panel shows the same lesion after Optison
contrast injection. The relationship between the feeder vessel and the
mass can be better appreciated in the contrast-enhanced image. The

contrast-enhanced image shows the nodule (dotted circle) to be vascular.

Although small blood vessels are distributed over the entire mass, there
is greater concentration at the periphery. In the region distal to the feeder
vessels (arrow in the right panel), there is a color blooming artifact
commonly observed in contrast-enhanced power Doppler images.

et al. observed that the diagnostic performance, measured

by ROC area under the curve Az, improved substantially
from 0.51 to 0.76 when contrast-enhanced 3D power
Doppler imaging was used instead of 2D contrast-enhanced
power Doppler [80]. Yang et al. also found that tumor
microvessel density measured by immunohistochemical
assay correlated to the number of intratumoral vessels
measured by 2D and 3D contrast-enhanced power Doppler
sonography [81]. The varying results show that the use of
ultrasound contrast agents for differentiating breast masses
is still experimental, and there is yet no accepted methodology for either its use or the interpretation of the images.

the elasticity of breast nodules has encouraged several

investigators to develop approaches to image tissue
elasticity or the properties related to it [9299].
Several different elastography approaches have been
proposed. A common approach for breast imaging involves
sonographic imaging of the tissue before and after gentle
compression. The radiofrequency data from the pre- and postcompression images are compared to determine tissue
deformation. The compression is often applied freehand by
pressing the transducer on the breast [9294]. Other approaches using low frequency fremitus vibrations [12, 95
97], and acoustic radiation force [98100] have also been
proposed for breast elasticity imaging with promising results.
Sohn et al. [9597] used low-frequency vibrations
associated with vocal fremitus to perturb the breast tissue
and Doppler imaging to visualize the vibrations. The
hypothesis is that well-encapsulated benign masses would
displace the surrounding tissue, whereas the infiltrative
nature of the cancers would enable the transmission of
vibrations into the central portion of the mass. Thus,
depending on whether the surrounding tissue or the center
of the mass is enhanced by Doppler imaging during
fremitus, the benign and malignant masses could be
distinguished. Sohn reported 9193% accuracy by this
method. Schultz et al., however, observed enhancement of
the breast tissue to be significantly influenced by the choice
of imaging parameters [101].
Nightingale et al. proposed the use of radiation force
associated with ultrasound pulses to displace breast tissue for
elastography imaging [98]. The advantage of this approach
is that with some modifications the same transducer can be
used to generate as well as to detect local displacements.
Alizad et al. also proposed the use of radiation force to induce tissue displacement, but instead of measuring the displacement directly, Alizad et al. monitored low-frequency

Tissue Elasticity Imaging: Elastography

and Related Techniques
Over the last decade, there has been an intense effort to
improve the sensitivity and specificity of cancer detection
and of characterization using different imaging modalities.
An area of considerable interest is the use of ultrasound for
imaging tumor elasticity in patients with breast cancer.
Solid tumors are often palpable, indicating that they are
harder than the surrounding tissue. The hope is that
elasticity imaging will facilitate the detection and characterization of these masses. Several excellent reviews are
available on this topic [8287].
Tissue stiffness or tissue hardness is often described by
Youngs modulus of elasticity, which measures the amount
of longitudinal deformation tissue undergoes in response to
an applied longitudinal force. The Youngs modulus of
elasticity for soft tissue can range from 1 to 100 kPa [88
91]. While fibroadenomas are only two times stiffer than
normal breast parenchyma, breast cancers can be as much
as 15 times stiffer [88, 90, 91]. The marked difference in

J Mammary Gland Biol Neoplasia (2006) 11: 113123

sound waves generated by the impact of ultrasound pulses

on the tissue [99, 100]. The hypothesis is that the signature
of these low-frequency sound waves is directly correlated
to the viscoelastic properties of the tissue.
Although there are several appealing approaches for
elasticity imaging, elastography based on manual compression of the breast has been adopted by many investigators.
Firm masses show less deformation and appear darker on the
elastograms. Malignant masses known to be stiffer (less
deformation) appear darker in the image compared to the
benign masses. The benign masses on the other hand may
appear brighter, similar, or darker when compared to the
surrounding normal tissue [94]. Comparing the sizes of the
breast lesion on elastograms and conventional B-mode
images revealed an interesting difference between malignant
and benign lesions. Whereas cysts and fibroadenomas are the
same size in the two modes of imaging, carcinomas appear to
be larger on elastograms. This is a significant finding, since
the difference in size can be exploited to differentiate benign
and malignant masses [94]. It is believed that the larger size
of cancers in elastograms indicates that the infiltration of
cancer into the surrounding tissues increases their stiffness.

Conclusions
Breast ultrasound is multimodal and can map blood flow
and mechanical properties of the tissue. Consistent
improvements in image quality over the years have
expanded the role of ultrasound in the detection and
diagnosis of breast pathology, and sonography is routinely
used as an adjunct to X-ray mammography. There are
attempts to combine ultrasound imaging and mammography in one instrument to facilitate scanning and to improve
diagnosis [102]. Ultrasound is also a method of choice for
guiding biopsy needles and various therapeutic interventions. We believe that newer techniques will continue to
improve the sensitivity and specificity of breast ultrasound.
Acknowledgments The authors gratefully acknowledge the assistance from Susan M Schultz and Theodore W Cary in preparing this
manuscript. This work was in part supported by NIH grants CA-85424
and CA-87526.

References
1. Wild JJ. The scientific discovery of sonic reflection by soft tissues
and application of ultrasound to diagnostic medicine and tumor
screening. Third Meeting of the World Federation for Ultrasound
in Medicine and Biology (Scientific Exhibit Section) Brighton,
EnglandJuly 1982. (Cantab) Medico-Technological Research
Institute of Minneapolis, Minneapolis, Minnesota, USA
2. Sabel M, Aichinger H. Recent developments in breast imaging.
Phys Med Biol 1996;41:31568.

121
3. Wade G. Ultrasonic imaging by reconstructive tomography.
Acoust Imag 1980;9:379431.
4. Greenleaf JF, Bahn RC. Clinical imaging with transmission
ultrasonic computerized tomography. IEEE Trans Biomed Eng
BME 1981;28:17785.
5. Hiller D, Ermert H. Ultrasound computerized tomography using
transmission and reflection mode: application to medical
diagnosis. Acoust Imag 1982;12:55364.
6. Schreiman JS, Gisvold JJ, Greenleaf JF, Bahn RC. Ultrasound
transmission computed tomography of the breast. Radiology
1984;150:52330.
7. Scherzinger AL, Belgam RA, Carson PL, Meyer CR, Sutherland
JV, Bookstein FL, et al. Assessment of ultrasonic computed
tomography in symptomatic breast patients by discriminant
analysis. Ultrasound Med Biol 1989;15:218.
8. Friedrich M, Hundt E, Maderlechner G. Computerized ultrasound
echo tomography of the breast. Eur J Radiol 1982;2:7887.
9. Kremkau FW. Diagnostic ultrasound: principles and instruments.
5th ed. Philadelphia: WB Saunders; 1984.
10. Hedrick WR, Hykes DL, Starchman DE. Ultrasound physics and
instrumentation. 3rd ed. St Louis: Mosby; 1995.
11. Zagzebski JA. Essentials of ultrasound physics. St Louis: Mosby; 1996.
12. Stavros AT. Breast ultrasound. Philadelphia: Williams and Wilkins; 2004.
13. Helmut M. The practice of breast ultrasound: techniques,
findings, differential diagnosis. New York: Thieme; 2000.
14. Bassett LW, Kimme-Smith C, Sutherland LK, Gold RH, Sarti D,
King W, 3rd. Automated and hand-held breast US: effect on
patient management. Radiology 1987;165:1038.
15. Bassett LW, Ysrael M, Gold RH, Ysrael C. Usefulness of
mammography and sonography in women less than 35 years of
age. Radiology 1991;180:8315.
16. Lamas AM, Horwitz RI, Peck D. Usefulness of mammography
in the diagnosis and management of breast disease in postmenopausal women. JAMA 1984;252:29993002.
17. Venta LA, Dudiak CM, Salomon CG, Flisak ME. Sonographic
evaluation of the breast. Radiographics 1994;14:2950.
18. Jackson VP. The current role of ultrasonography in breast
imaging. Radiol clin North Am 1995;33:116170.
19. Bassett LW, Kimme-Smith C. Breast sonography. Am J Roentgenol 1991;156:44955.
20. Hogg JP, Harris KM, Skolnick ML. The role of ultrasoundguided needle aspiration of breast masses. Ultrasound Med Biol
1988;14 Suppl 1:1321.
21. Nightingale KR, Kornguth PJ, Trahey GE. The use of acoustic
streaming in breast lesion diagnosis: a clinical study. Ultrasound
Med Biol 1999;25:7587.
22. Gordon PB, Goldenberg SL. Malignant breast masses detected
only by ultrasound. A retrospective review. Cancer 1995;76:626
30.
23. Kopans DB, Feig SA, Sickles EA. Malignant breast masses
detected only by ultrasound: a retrospective review. Cancer
1996;77:2089.
24. Kolb TM, Lichy J, Newhouse JH. Occult cancer in women with
dense breasts: detection with screening US-diagnostic yield and
tumor characteristics. Radiology 1998;207:1919.
25. Berg WA. Rationale for a trial of screening breast ultrasound:
American College of Radiology Imaging Network (ACRIN).
Am J Roentgenol 2003;180:12258.
26. Drukker K, Giger ML, Mendelson EB. Computerized analysis of
shadowing on breast ultrasound for improved lesion detection.
Med Phys 2003;30:183342.
27. Jackson VP, Reynolds HE, Hawes DR. Sonography of the breast.
Semin Ultrasound CT MR 1996;17:46075.
28. Stavros AT, Thickman D, Rapp CL, Dennis MA, Parker SH, Sisney
GA. Solid breast nodules: use of sonography to distinguish between
benign and malignant lesions. Radiology 1995;196:12334.

122
29. Jackson VP. Management of solid breast nodules: what is the
role of sonography? Radiology 1995;196:145.
30. Paulinelli RR, Freitas-Junior R, Moreira MA, Moraes VA,
Bernardes-Junior JR, Vidal CS, et al. Risk of malignancy in
solid breast nodules according to their sonographic features. J
Ultrasound Med 2005;24:63541.
31. Arger PH, Sehgal CM, Conant EF, Zuckerman J, Rowling SE,
Patton JA, et al. Interreader variability and predictive value of
US descriptions of solid breast masses: pilot study. Acad Radiol
2001;8:33542.
32. American College of Radiology. BI-RADS: Ultrasound, 1st ed.
In: Breast imaging reporting and data system: BI-RADS atlas,
4th ed. Reston, Virginia: American College of Radiology; 2003.
33. Mendelson EB, Berg WA, Merritt CRB. Toward a standardized
breast ultrasound lexicon, BI-RADS: ultrasound. Semin Roentgenol 2001;36:21725.
34. Baker JA, Kornguth PJ, Soo MS, Walsh R, Mengoni P.
Sonography of solid breast lesions: observer variability of lesion
description and assessment. Am J Roentgenol 1999;172:16215.
35. Hong AS, Rosen EL, Soo MS, Baker JA. BI-RADS for
sonography: positive and negative predictive values of sonographic features. Am J Roentgenol 2005;184:12605.
36. Kuo WJ, Chang RF, Moon WK. Computer aided diagnosis of breast
tumors with different US systems. Acad Radiol 2002;9:7939.
37. Horsch K, Giger ML, Vyborny CJ, Venta LA. Performance of
computer-aided diagnosis in the interpretation of lesions on
breast sonography. Acad Radiol 2004;11:27280.
38. Goldberg V, Manduca A, Ewert DL, Gisvold JJ, Greenleaf JF.
Improvement in specificity of ultrasonography for diagnosis of
breast tumors by means of artificial intelligence. Med Phys
1992;19:147581.
39. Koyama S, Obata Y, Shimamoto K, Ishigaki T, Ishii N, Isomoto
Y, et al. Breast ultrasonography: computer-aided diagnosis using
fuzzy inference. J Ultrasound Med 1997;16:66572.
40. Chen DR, Chang RF, Huang YL. Computer-aided diagnosis
applied to US of solid breast nodules by using neural networks.
Radiology 1999;213:40712.
41. Chen DR, Kuo WJ, Chang RF, Moon WK, Lee CC. Use of the
bootstrap technique with small training sets for computer-aided
diagnosis in breast ultrasound. Ultrasound Med Biol 2002;
28:897902.
42. Kuo WJ, Chang RF, Lee CC, Moon WK, Chen DR. Retrieval
technique for the diagnosis of solid breast tumors on sonogram.
Ultrasound Med Biol 2002;28:9039.
43. Shankar PM, Dumane VA, Piccoli CW, Reid JM, Forsberg F,
Goldberg BB. Classification of breast masses in ultrasonic Bmode images using a compounding technique in the Nakagami
distribution domain. Ultrasound Med Biol 2002;28:12951300.
44. Sahiner B, Chan HP, Roubidoux MA, Helvie MA, Hadjiiski LM,
Ramachandran A, et al. Computerized characterization of breast
masses on three-dimensional ultrasound volumes. Med Phys
2004;31:74454.
45. Dumane VA, Shankar PM, Piccoli CW, Reid JM, Forsberg F,
Goldberg BB. Computer aided classification of masses in
ultrasonic mammography. Med Phys 2002;29:196873.
46. Chen CM, Chou YH, Han KC, Hung GS, Tiu CM, Chiou HJ,
et al. Breast lesions on sonograms: computer-aided diagnosis with
nearly setting-independent features and artificial neural networks.
Radiology 2003;226:50414.
47. Drukker K, Giger ML, Vyborny CJ, Mendelson EB. Computerized detection and classification of cancer on breast ultrasound.
Acad Radiol 2004;11:52635.
48. Sehgal CM, Cary TW, Kangas SA, Weinstein SP, Schultz SM,
Arger PH, et al. Computer-based margin analysis of breast
sonography for differentiating malignant and benign masses. J
Ultrasound Med 2004;23:12019.

J Mammary Gland Biol Neoplasia (2006) 11: 113123

49. Piliouras N, Kalatzis I, Dimitropoulos N, Cavouras D. Development
of the cubic least squares mapping linear-kernel support vector
machine classifier for improving the characterization of breast lesions
on ultrasound. Comput Med Imaging Graph 2004;28:24755.
50. Cary TW, Conant EF, Arger PH, Sehgal CM. Diffuse boundary
extraction of breast masses on ultrasound by leak plugging. Med
Phys 2005;32:331828.
51. Song JH, Venkatesh SS, Conant EF, Arger PH, Sehgal CM.
Comparative analysis of logistic regression and artificial neural
network for computer-aided diagnosis of breast masses. Acad
Radiol 2005;12:48795.
52. Weinstein SP, Seghal CM, Conant EF, Patton JA. Microcalcifications in breast tissue phantoms visualized with acoustic
resonance coupled with power Doppler US: initial observations.
Radiology 2002;224:2659.
53. Alizad A, Fatemi M, Wold LE, Greenleaf JF. Performance of
vibro-acoustography in detecting microcalcifications in excised
human breast tissue: a study of 74 tissue samples. IEEE Trans
Med Imag 2004;23:30712.
54. Fornage BD, Sneige N, Edeiken BS. Interventional breast
sonography. Eur J Radiol 2002;42:1731.
55. Burkhardt C. Speckle in ultrasound B-mode scans. IEEE Trans
Sonics Ultrason 1978;25:16.
56. Entrekin R, Jackson P, Jago JR, Porter BA. Real time spatial
compound imaging in breast ultrasound: technology and early
clinical experience. Medicamundi 1999;43:3543.
57. Kwak JY, Kim EK, You JK, Oh KK. Variable breast conditions:
comparison of conventional and real-time compound ultrasonography. J Ultrasound Med 2004;23:8596.
58. Kangas SA, Schultz SM, Conant EF, Weinstein S, Arger PH,
Schnall MD, et al. Differences between conventional and compound
ultrasound imaging of solid breast lesions (Abstract). Radiology
Society of North America Chicago, Nov 28Dec 3, 2004:142.
59. Rosen EL, Soo MS. Tissue harmonic imaging sonography of
breast lesions: improved margin analysis, conspicuity, and image
quality compared to conventional ultrasound. Clin Imaging
2001;25:37984.
60. Szopinski KT, Pajk AM, Wysocki M, Amy D, Szopinska M,
Jakubowski W. Tissue harmonic imaging: utility in breast
sonography. J Ultrasound Med 2003;22:47987.
61. Kubota K, Hisa N, Ogawa Y, Yoshida S. Evaluation of tissue
harmonic imaging for breast tumors and axillary lymph nodes.
Oncol Rep 2002;9:13358.
62. Kubota K, Ogawa Y, Nishigawa T, Yoshida S. Tissue harmonic
imaging sonography of the axillary lymph nodes: evaluation of
response to neoadjuvant chemotherapy in breast cancer patients.
Oncol Rep 2003;10:19114.
63. Sehgal CM, Brown GM, Bahn RC, Greenleaf JF. Measurement and
use of acoustic nonlinearity and sound speed to estimate composition of excised livers. Ultrasound Med Biol 1986;12:86574.
64. Law WK, Frizzell LA, Dunn F. Determination of the nonlinearity
parameter B/A of biological media. Ultrasound Med Biol
1985;11:138307.
65. Cosgrove DO, Kedar RP, Bamber JC, al-Murrani B, Davey JB,
Fisher C, et al. Breast diseases: color Doppler US in differential
diagnosis. Radiology 1993;189:99.
66. Huber S, Delorme S, Knopp MP, Junkermann H, Zuna I, von
Fournier D, et al. Breast tumors: computer-assisted quantitative
assessment with color Doppler US. Radiology 1994;192:797.
67. Raza S, Baum JK. Solid breast lesions: evaluation with power
Doppler US. Radiology 1997;203:164.
68. Birdwell RL, Ikeda DM, Jeffrey SS, Jeffrey RB Jr. Preliminary
experience with power Doppler imaging of solid breast masses.
AJR 1997;169:703.
69. Sehgal CM, Arger PH, Kotlar EY, Pugh CR, Kirchoffer JI,
Bovee KH. Quantitative analysis of color and power Doppler

J Mammary Gland Biol Neoplasia (2006) 11: 113123

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

images to differentiate between malignant and benign lesions. J

Ultrasound Med 1998;17:S105.
Carson PL, Fowlkes JB, Roubidoux MA, Moskalik AP, Govil A,
Normolle D, et al. 3-D color Doppler image quantification of
breast masses. Ultrasound Med Biol 1998;24:945.
Sehgal CM, Arger PH, Rowling SE. Quantitative vascularity of
breast masses by Doppler imaging: regional variations and
diagnostic implications. J Ultrasound Med 2000;19:42740.
Lee SW, Choi HY, Baek SY, Lim SM. Role of color and power
Doppler imaging in differentiating between malignant and
benign solid breast masses. J Clin Ultrasound 2002;30:45964.
Wilkens TH, Burke BJ, Cancelada DA, Jatoi I. Evaluation of
palpable breast masses with color Doppler sonography and gray
scale imaging. J Ultrasound Med 1998;17:10915.
Del Cura JL, Elizagaray E, Zabala R, Legorburu A, Grande D.
The use of unenhanced Doppler sonography in the evaluation of
solid breast lesions. AJR 2005;184:178894.
Pugh CR, Arger PH, Sehgal CM. Power, spectral, and color flow
Doppler enhancement by a new ultrasonographic contrast agent.
J Ultrasound Med 1996;15:84382.
Kedar RP, Cosgrove D, McCready VR, Bamber JC, Carter ER.
Microbubble contrast agent for color Doppler US: effect on
breast masses. Work in progress. Radiology 1996;198:67986.
Huber S, Helbich T, Kettenbach J, Dock W, Zuna I, Delorme S.
Effects of a microbubble contrast agent on breast tumors:
computer-assisted quantitative assessment with color Doppler
US-early experience. Radiology 1998;208:4859.
Alamo L, Fischer U. Contrast-enhanced color Doppler ultrasound characteristics in hypervascular breast tumors: comparison
with MRI. Eur Radiol 2001;11:9707.
Reinikainen H, Rissanen T, Paivansalo M, Paakko E, Jauhiainen
J, Suramo I. B-mode, power Doppler and contrast-enhanced
power Doppler ultrasonography in the diagnosis of breast
tumors. Acta Radiol 2001;42:10613.
Forsberg F, Goldberg BB, Merritt CR, Parker L, Maitino AJ,
Palazzo JJ, et al. Diagnosing breast lesions with contrast-enhanced
3-dimensional power Doppler imaging. J Ultrasound Med
2004;23:17382.
Yang WT, Tse GM, Lam PK, Metreweli C, Chang J. Correlation
between color power Doppler sonographic measurement of
breast tumor vasculature and immunohistochemical analysis of
microvessel density for the quantitation of angiogenesis. J
Ultrasound Med 2002;21:122735.
Hall TJ. AAPM/RSNA physics tutorial for residents: topics in
US: beyond the basics: elasticity imaging with US. Radiographics 2003;23:165771.
Gao L, Parker KJ, Lerner RM, Levinson SF. Imaging of the
elastic properties of tissuea review. Ultrasound Med Biol
1996;22:95977.
Ophir J, Alam SK, Garra B, Kallel F, Konofagou E, Krouskop T,
et al. Elastography: ultrasonic estimation and imaging of the
elastic properties of tissues. Proceedings of the Institution of
Mechanical Engineers, Part HJ. Eng Med 1999;213:20333.

123
85. Insana MF, Bamber JC. Tissue motion and elasticity imaging.
Phys Med Biol 2000;45(6):vvi (2 p preceding 1409).
86. Wilson LS, Robinson DE, Dadd MJ. Elastographythe movement begins. Phys Med Biol 2000;45:140921.
87. Fatemi M, Greenleaf JF. Probing the dynamics of tissue at low
frequencies with the radiation force of ultrasound. Phys Med
Biol 2000;45:144964.
88. Plewes DB, Bishop J, Samani A, Sciarretta J. Visualization and
quantification of breast cancer biomechanical properties with
magnetic resonance elastography. Phys Med Biol 2000;45:
15911610.
89. Fung Y. Biomechanics: mechanical properties of living tissue.
2nd ed. Berlin Heidelberg New York: Springer; 1993.
90. Sarvazyan A, et al. Elasticity imaging as a new modality of
medical imaging for cancer detection. Proc. Int. Workshop on
Interaction of Ultrasound with Biological Media, Valenciennes,
France. (1994). pp 6981.
91. Sarvazyan A, et al. Biophysical bases of elasticity imaging.
Acoust Imag 1995;21:22340.
92. Hall TJ, Zhu Y, Spalding CS. In vivo real-time freehand
palpation imaging. Ultrasound Med Biol 2003;29:42735.
93. Hiltawsky KM, Kruger M, Starke C, Heuser L, Ermert H, Jensen
A. Freehand ultrasound elastography of breast lesions: clinical
results. Ultrasound Med Biol 2001;27:14619.
94. Garra BS, Cespedes EI, Ophir J, Spratt SR, Zuurbier RA,
Magnant CM, et al. Elastography of breast lesions: initial clinical
results. Radiology 1997;202:7986.
95. Sohn C, Baudendistel A, Bastert G. Diagnosis of the breast tumor
entity vocal fremitus in ultrasound diagnosis. Bildgebung
1994;61:2914.
96. Sohn C, Baudendistel A, Kaufmann M, Bastert MG. The
positive vocal fremitus in malignant breast tumors in color
MEM ultrasound imagingan exciting artifact in confirming the
diagnosis. Geburtshilfe Frauenheilkd 1994;54:42731.
97. Sohn C, Baudendistel A. Differential diagnosis of mammary
tumors with vocal fremitus in sonography: preliminary report.
Ultrasound Obstet Gynecol 1995;6:2057.
98. Nightingale KR, Palmeri ML, Nightingale RW, Trahey GE. On
the feasibility of remote palpation using acoustic radiation force.
J Acoust Soc Am 2001;110:62534.
99. Alizad A, Whaley DH, Greenleaf JF, Fatemi M. Potential
applications of vibro-acoustography in breast imaging. Technol
Cancer Res Treat 2005;4:1518.
100. Fatemi M, Wold LE, Alizad A, Greenleaf JF. Vibro-acoustic
tissue mammography. IEEE Trans Med Imag 2002;21:18.
101. Schultz SM, Conant EF, Weinstein S, Kangas SA, Arger PH,
Schnall MD, et al. Evaluation of ultrasound imaging parameters
on the fremitus effects in breast imaging. Radiology Society of
North America. Chicago, Nov 28Dec 3, 2004;142.
102. Kapur A, Carson PL, Eberhard J, Goodsitt MM, Thomenius K,
Lokhandwalla M, et al. Combination of digital mammography
with semi-automated 3D breast ultrasound. Technol Cancer Res
Treat 2004;3(4):32534.