ULTRASONIC IMAGING 24, 246- 260 (2002)
Temperature Dependence of Ultrasonic Propagation Speed
and Attenuation in Canine Tissue
U. TECHAVIPOO, T. VARGHESE, J.A. ZAGZEBSKI, T. STILES
1, 2
1
1
1
AND G.
FRANK
1
1
Department. of Medical Physics
2
Department of Radiology
The University of Wisconsin-Madison
Madison, WI-53706
tvarghese@facstaff.wisc.edu
Previously reported data on the temperature dependence of propagation speed in tissues generally
span only temperature ranges up to 60°C. However, with the emerging use of thermal ablative therapies,
information on variation in this parameter over higher temperature ranges is needed. Measurements of
the ultrasonic propagation speed and attenuation in tissue in vitro at discrete temperatures ranging from
25 to 95°C was performed for canine liver, muscle, kidney and prostate using 3 and 5 MHz center frequencies. The objective was to produce information for calibrating temperature-monitoring algorithms
during ablative therapy. Resulting curves of the propagation speed vs. temperature for these tissues can
be divided into three regions. In the 25-40°C range, the speed of sound increases rapidly with temperature. It increases moderately with temperature in the 40-70°C range, and it then decreases with increasing temperature from 70-95°C. Attenuation coefficient behavior with temperature is different for the
various tissues. For liver, the attenuation coefficient is nearly constant with temperature. For kidney, attenuation increases approximately linearly with temperature, while for muscle and prostate tissue,
curves of attenuation vs. temperature are flat in the 25-50°C range, slowly rise at medium temperatures
(50-70°C), and level off at higher temperatures (70-90°C). Measurements were also conducted on a distilled degassed water sample and the results closely follow values from the literature.
KEY WORDS: Ablation; attenuation; imaging; speed of sound; strain; temperature; ultrasound.
INTRODUCTION
Percutaneous in situ tumor ablation using various thermal energy sources is being investigated because of its promise in the treatment of focal malignant disease.1-12 These techniques
allow treatment of patients who are not considered candidates for surgery due to age or extent of disease. 13 Tumor ablation using high-intensity focused ultrasound,8 radiofrequency
11,12
14
7,15
ablation, microwave, and laser energy has been conducted in human subjects. Because
of the requirement to destroy only tumors with negative margins, with minimal damage to
surrounding tissue, an imaging system to localize tumors16-19 and real-time temperature monitoring20-26 to estimate thermal doses are essential. Diagnostic ultrasound imaging can serve
these requirements because it is real-time and inexpensive compared with other imaging
techniques such as magnetic resonance imaging and computed tomography.
Temperature estimation using diagnostic ultrasound has been studied by many research
groups.20-26 As tissue is heated during ablation, the tissue expands along with changes in the
ultrasonic propagation speed. These changes introduce time shifts in backscattered ultrasound echo signals from the heated region. Thermal expansion and sound speed changes
may appear as opposing phenomena in terms of apparent shifts in radiofrequency (rf) echo
signals. Depending on tissue composition, thermal expansion will likely cause smaller
shifts in the echo signals than shifts due to speed of sound changes.20-26 Seip and Ebbini20 estimated the temperature changes along one dimension during heating by tracking frequency
246
0161-7346/02 $18.00
Copyright 2002, 2003 by Dynamedia, Inc.
All rights of reproduction in any form reserved.
TEMPERATURE DEPENDENCE OF PROPAGATION SPEED AND ATTENUATION
247
Pump and
Temperature
controller
Tissue holder
Tissue
Transmitter
Receiver
Degassed distilled water
Membranes
Pump and
Temperature
controller
Oscilloscope
Sample
in Saline
Function Gen.
Tissue holders
Power Amplifier
Water in heating bath
FIG. 1 Experimental apparatus used in the measurements (a) and the front and side views of a tissue holder containing a tissue sample in saline (b).
variations of the echo spectrum. Echo-arrival time shifts due to heating have also been applied for temperature estimation.21,23-25 The time shifts are differentiated along the axial direction and multiplied by a tissue dependent parameter to obtain local temperature changes.
The temperature value estimated depends on the thermal expansion coefficient and on the
variation of the propagation speed with temperature. Simon et al21 obtained this parameter
experimentally and assumed it to be constant in the temperature range from 25 to 43°C.
However, this temperature range is too low for tumor ablation therapies.1-12 Furthermore, the
tissue-dependent parameter may not be constant, particularly for high temperatures.
Data on ultrasonic propagation speeds and attenuation coefficients in human and animal
tissues have been summarized in Wells,27, Goss28 and Duck.29 Variations in the propagation
speeds and attenuation with temperature have also been reported.30-34 Bowen et al30 illustrated that the rate of change of the propagation speed with temperature is correlated with the
ultrasonic velocity near normal body temperature. Bamber and Hill33 measured the ultrasonic attenuation and the propagation speed in bovine and human soft tissues in the range of
5 to 65°C. Worthington and Sherar32 measured the same properties in porcine kidney. Both
groups measured the attenuation by using a broadband substitution method. Variations in
the ultrasonic propagation speed35-38 and attenuation36-48 have also been used extensively to
differentiate between normal and diseased tissue.
In this article, we report results of measurements of ultrasound propagation speed and attenuation in excised canine tissue over the temperature range of 25 to 95°C. Specimens of
liver, kidney, muscle and prostate tissues were evaluated using a narrow band substitution
technique.
MATERIALS AND METHODS
The apparatus used for the measurement is shown in figure 1. Two water baths were used
in the experiment, one for performing acoustical measurements and the other for heating the
samples before taking measurements. The transmitting transducer (Panametrics V309,
248
TECHAVIPOO ET AL
Waltham, MA) is 13 mm in diameter. The transducer was driven by 30-cycle duration,
narrowband pulses using a function generator (Wavetek model 81, San Diego, CA) and a
power amplifier (Amplifier Research model 75A250, Souderton, PA). Center frequencies
of 3 and 5 MHz were used in the experiment. The receiving transducer (Aerotech Delta,
Krautkramer Inc., Lewistown, PA) is 8 mm in diameter. Optimal operation of the transducer
transmitter-receiver pair was at a center frequency of 5 MHz. The received signals were recorded and averaged over 50 successive traces by a digital oscilloscope (LeCroy 9410,
Chestnut Ridge, NY), triggered by the function generator. The temperature of the degassed
distilled water in each water bath was controlled by a pump and temperature controller
(Haake, Karlsruhe, West Germany). Each tissue sample was immersed in saline and enclosed in a tissue holder (whose dimensions are 1.6 cm thick, 5 cm wide and 15 cm in height).
The sides of the tissue sample holder are made of a 25 mm membrane (Saran Wrap, Dow
Chemical Company, Midland, MI) as shown in figure 1. To verify the accuracy of the measurement method, a sample of degassed distilled water whose propagation speed and
attenuation coefficient as a function of temperature were known was used in each
experiment.
A. Tissue preparation
Four types of tissue samples, liver, psoas muscle, kidney and prostate were obtained from
approximately 10-month old male hounds available from unrelated studies in an adjoining
lab. The measurements were begun less than five hours after excision. Each sample was cut
into approximate dimensions of 1.6 cm ´5 cm ´5 cm and placed in the tissue holder, along
with degassed saline buffer. Liver tissue sections without large veins and bile ducts were selected, and placed into the sample holder without any preferred orientation. Muscle tissue
was arranged in the holder such that the ultrasound signal propagated perpendicular to its fibers. The kidney was cut in the axial direction across its center for the 1.6 cm thickness and
put into the tissue holder so the kidney medulla was in the center position for measuring. The
entire prostate gland was put into the tissue holder in a direction such that the prostatic urethra was parallel to the propagation direction. The thicknesses of the tissue samples were
measured at each temperature using calipers at several locations around the center of the
samples where the acoustic measurements were also acquired.
B. Data acquisition
The propagation speeds and the attenuation coefficients were measured at eight different
temperatures, 25, 30, 40, 50, 60, 70, 80, 90 and 95°C, in the order of increasing temperature.
The experiment was conducted by transferring a specimen from the heating bath to the measurement bath, taking a measurement immediately after the sample was in position, and restoring the specimen back to the heating bath. The time shift in the received ultrasonic pulse
after interposing the specimen was recorded for the propagation speed, while the peakto-peak voltage amplitudes with and without the specimen in place were measured for the attenuation coefficient. This process was repeated three times followed by measurement of
the sample thickness. Then the temperature of the water in each bath was changed to the next
higher temperature setting and twenty minutes was allowed for thermal equilibrium before
the next measurement. For the first three temperature settings, the water temperature in the
measurement bath was the same temperature as the heating bath. For higher temperatures,
the measurement bath was maintained at 40°C, but care was taken to ensure that time shifts
and transmitted amplitudes were recorded within seconds after insertion of the sample. This
experiment was repeated ten times, each time using tissue from a different animal.
TEMPERATURE DEPENDENCE OF PROPAGATION SPEED AND ATTENUATION
249
C. Data reduction
Propagation speed was determined using the substitution technique.49 Ignoring the propagation time in the thin membrane layers of the tissue holder, the propagation speed in tissue,
cm, can be computed by
cm = cw /(1 - cw Dt / d )
(1)
where cw is the propagation speed of water in the measurement bath; Dt is the time shift after
insertion of the sample; and d is the sample thickness. The sound propagation time with and
without the sample was measured from the position of the zero crossing of the fifth cycle of
the transmitted ultrasound pulse. The speed of sound in distilled water over the temperature
range from 0 to 100 °C at atmospheric pressure can be estimated by a fifth-order polynomial
whose coefficients from the zeroth to the fifth degree are given by 1402.736, 5.03358,
-0.0579506, 3.31636´10-4, -1.45262´10-6 and 3.0449´10-9 respectively.50
The attenuation coefficients were measured at 3 and 5 MHz using the narrowband substitution technique. Assuming that attenuation of ultrasound through the water over the thickness of the sample is negligible compared to the attenuation in tissue, the attenuation
coefficient, a, in dB/cm can be written as
a = (20 / d ) log( A0 / ATtotal )
(2)
where d is the measured thickness of the sample, A0 and A are the measured peak-to-peak signal voltages without and with the sample inserted respectively, and Ttotal is the combined amplitude transmission coefficient49, 51 through the sample windows. The latter is calculated
using
Ttotal =
4 Z1 Z 3
( Z1 + Z 3 ) cos (k 2l ) + ( Z 2 + Z1Z 3 / Z 2 ) 2 sin 2 (k 2l )
2
(3)
2
where the subscripts 1, 2 and 3 refer to the properties of the water in the measurement bath,
the thin layer material and the tissue sample respectively; Z denotes the acoustic impedance,
k the wave number and l is the thickness of the thin layer. The acoustic impedance Z was calculated from Z = rc, where r is the density and c is the sound speed in the material. The wave
number k2 was calculated indirectly from 2pf/c2, where c2 is the sound speed in the thin layer.
Assuming that the acoustic impedance Z2 (= 4.25´106 kg/m2s), the sound speed c2 (= 2504
m/s), the thickness l (= 25´10-6 m) of the thin layer and the densities of the tissue sample r3 (=
1.050´103, 1.041´103, 1.044´103 and 1.045´103 kg/m3 for liver, muscle, kidney and prostate, respectively) are known and do not vary with temperature, the combined amplitude
transmission coefficient can be estimated.49 Densities of human tissue29 were used in Eq. (3)
instead of those of canine tissue. The acoustic impedance of water in the measurement bath
at different temperatures was calculated by multiplying the sound speed50 as shown above
and the density, which was also approximated by a fifth-order polynomial whose coefficients are given by 999.842594, 6.793952´10-2, -9.095290´10-3, 1.001685´10-4, -1.120083
´10-6 and 6.536332´10-9 respectively.52
250
TECHAVIPOO ET AL
RESULTS
A. Degassed distilled water measurements
Figure 2 shows the measured propagation speeds and the attenuation coefficients in the
degassed distilled water at different temperatures for 3 and 5 MHz. Similar sample holders
as those used for tissue samples were used for the degassed distilled water measurements.
Each data point is the mean over ten experiments, and its error bar represents ±1 standard deviation of the results. In figure 2a, the measurement of the propagation speed closely follows
the values reported by Greenspan and Tschegg50 with a very small standard deviation. However, the results tend to deviate from the literature values above 80°C. This suggests that our
measurements below 80°C are more accurate than those above this temperature. However,
the largest deviation is about 7 m/s, or 0.48 percent, from the literature value. This deviation
may be caused by small temperature changes in the sample during the brief time after it was
inserted into the lower temperature measurement bath for measuring the time shift.
In figure 2b, the attenuation measurements are close to zero as expected, although results
have large error bars at higher temperatures. The largest standard deviation in the attenuation coefficient measured was 0.666 dB/cm at 3 MHz and 0.422 dB/cm at 5 MHz, assuming
that the attenuation coefficient of the water is constant at 0.0025 dB/cm/MHz2 (or 0.0225
dB/cm at 3 MHz and 0.0625 dB/cm at 5 MHz) for all of the measurement temperatures. The
errors at higher temperatures could be a result of air bubbles accumulating in the sample.
Even though the samples were degassed, there was a tendency for limited number of bubbles
to accumulate on the sample walls.
B. Propagation speed in canine tissue
Figure 3 shows the propagation speed in canine tissues at 3 and 5 MHz center frequencies
versus temperature. Note that the graphs are plotted on the same scale, and each point denotes the mean value across 10 different animals, while the error bar denotes the standard deviation. For a sample size of 10, the probability is 95% that the study will detect a
relationship between the independent and dependent variables at a two-sided 5% significance level, if the true change in the dependent variable is 1.313 standard deviations per one
standard deviation in the independent variable.
In general, the propagation speeds in canine tissue samples lie between 1545-1595 m/s.
For each tissue type, the propagation speed behavior at 3 and 5 MHz is virtually identical.
The graphs in figure 3, may be broken down into three regions. In the first region, the propagation speed increases rapidly with temperature (25 to 40°C). The approximate slopes for
this region are 1.15, 1.51, 1.46, and 1.62 m/s/°C for liver, muscle, kidney and prostate respectively. In the second region, the rate of increase in the propagation speed with temperature is less than that found below 40°C. This region extends over the temperature range of 40
to 70°C (40 to 60°C for liver). The approximate slopes of propagation speed vs. temperature
for this region are 0.250, 0.410, 0.620, and 0.355 m/s/°C for liver, muscle, kidney, and prostate respectively. Finally, in the third region, the propagation speed decreases with the temperature from 70 to 95°C (60 to 95°C for liver). The approximate slopes for this region are
-0.917, -0.629, -0.644, and -0.624 m/s/°C for liver, muscle, kidney, and prostate respectively.
C. Attenuation coefficient in canine tissue
Figure 4 shows the variation in the attenuation coefficients with temperature in canine tissues at 3 and 5 MHz center frequencies. Again the graphs are plotted using the same scale,
and each point denotes the mean value across 10 different animals, while its error bar indi-
TEMPERATURE DEPENDENCE OF PROPAGATION SPEED AND ATTENUATION
251
1605
1595
1585
1575
1565
1555
5 MHz
1545
3 MHz
1535
20
30
40
50
60
70
80
90
100
80
90
100
Temperature (°C)
(a)
0.80
Attenuation Coefficient (dB/cm)
5 MHz
0.60
3 MHz
0.40
0.20
0.00
-0.20
-0.40
20
30
40
50
60
70
Temperature (°C)
(b)
FIG. 2 The propagation speed (a) and the attenuation coefficient (b) of a degassed distilled water sample vs. sample temperature. Error bars represent ± 1 SD. Results are shown for both 3 and 5 MHz center frequency bursts.
252
TECHAVIPOO ET AL
1565
1555
1545
1535
1525
1515
5 MHz
3 MHz
1505
Ref
1495
20
30
40
50
60
70
80
90
100
Temperature (°C)
(a)
1605
1595
1585
1575
1565
1555
5 MHz
1545
3 MHz
1535
20
30
40
50
60
70
Temperature (°C)
(b)
80
90
100
TEMPERATURE DEPENDENCE OF PROPAGATION SPEED AND ATTENUATION
253
1605
1595
1585
1575
1565
1555
5 MHz
1545
3 MHz
1535
20
30
40
50
60
70
80
90
100
Temperature (°C)
(c)
1605
1595
1585
1575
1565
1555
5 MHz
1545
3 MHz
1535
20
30
40
50
60
70
80
90
100
Temperature (°C)
(d)
FIG. 3 Propagation speed vs. temperature measured at 3 and 5 MHz. Results are shown for canine (a) liver, (b)
muscle, (c) kidney and (d) prostate.
254
TECHAVIPOO ET AL
20.00
5 MHz
Attenuation Coefficient (dB/cm)
18.00
3 MHz
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
20
30
40
50
60
70
80
90
100
Temperature (°C)
(a)
20.00
18.00
16.00
14.00
12.00
10.00
8.00
6.00
4.00
5 MHz
2.00
3 MHz
0.00
20
30
40
50
60
70
Temperature (°C)
(b)
80
90
100
TEMPERATURE DEPENDENCE OF PROPAGATION SPEED AND ATTENUATION
255
20.00
5 MHz
Attenuation Coefficient (dB/cm)
18.00
3 MHz
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
20
30
40
50
60
70
80
90
100
Temperature (°C)
(c)
20.00
5 MHz
Attenuation Coefficient (dB/cm)
18.00
3 MHz
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
20
30
40
50
60
70
80
90
100
Temperature (°C)
(d)
FIG. 4 Attenuation coefficient vs. temperature for canine tissue samples, measured both at 3 and 5 MHz. Results
are shown for (a) liver, (b) muscle, (c) kidney and (d) prostate.
256
TECHAVIPOO ET AL
Table 1 Comparison between results presented in this paper and those of other studies for the propagation speed in
1
2
canine tissue. Note Bowen et al (1979) and Goldman and Richards (1954).
Tissue types
T (°C)
Our results (m/s)
Other studies (m/s)
Liver
26
37
1577
1590
1581
1
1592-1604
Muscle
26
37
1562
1579
1576
1
1589-1603
Muscle
26
37
1547
1563
15592
1
1567-1571
2
2
cates the standard deviation. In figure 4a, the attenuation coefficients at 3 and 5 MHz for
liver tissue are almost invariant with temperature. Those for kidney tissue, however, (figure
4c) exhibit a linear increase with temperature. The approximate slopes of attenuation coefficient vs. temperature for liver tissue are 0.0038 and 0.0076 dB/cm/°C at 3 and 5 MHz respectively. For kidney tissue, the slopes are 0.0801 and 0.0967 dB/cm/°C at 3 and 5 MHz
respectively. For muscle and prostate tissue (Fig. 4b, d) the curves are flat at lower temperatures (25-50°C), slowly increase with temperature at medium temperatures (50-70°C), and
level off at higher temperatures (70-90°C).
DISCUSSION
The measured values of ultrasonic propagation speed and attenuation coefficient in canine
tissues and in degassed distilled water at different temperature are compared with literature
values. In general, the results are in agreement with those reported in the literature. For the
water sample, the propagation speeds are close to the values reported in reference 50 and the
attenuation coefficient is close to zero.
Our values of the propagation speed in soft tissues generally agree with results of other
studies, though subtle differences are seen. For low temperatures (26 and 37°C), our results
are slightly lower than the results reported in reference 30 and 53. These authors measured
the propagation speed in canine liver, muscle and kidney. Detailed comparisons are shown
in table 1. Note that linear interpolation was used to approximate our propagation speed results to the same temperature as in the other studies. For other temperatures, we could not
find similar published results so we compared our results with those reported for different
animal species. Bamber and Hill33 measured the propagation speed in human and bovine livers at temperatures from 5 to 65°C. Worthington and Sherar32 did so in porcine kidneys from
37 to 65°C. Graphs of propagation speed vs. temperature for each type of tissue follow the
same trend noted in this study, i.e., they increase at low temperatures and decrease at high
temperatures. Bamber and Hill33 used only one human tissue sample and two bovine tissue
samples, so the reported values may not represent the general trend for this tissue type. The
graph of propagation speed vs. temperature reported for their kidney sample32 exhibits many
abrupt changes, in contrast to the results shown in figure 3(c). This might stem from the authors’ measurement method, which used different sets of samples for each temperature.
Thus, sample-to-sample variability may have contributed to the variations in the data. Since
the kidney is inhomogeneous, the propagation speed also depends on the section of the kidney interrogated and its orientation. In contrast, our speed of sound vs. temperature curve is
smooth, probably due to the fact that the same tissue was used at all measurement tempera-
TEMPERATURE DEPENDENCE OF PROPAGATION SPEED AND ATTENUATION
257
tures and a large sample size was used at each temperature(?). However, our method could
suffer from tissue degradation with temperature. The measurements were taken over a
10-hour period and completed 15 hours after excision.
Our results for the attenuation coefficient follow the same trends as those of Bamber and
Hill33 and Worthington and Sherar32 for the same organs but different animal species. We are
unable to compare with the same species because none of the earlier studies measured in the
same canine organs and at the same frequencies as we did. Compared with Bamber and
Hill,33 at a 3-MHz center frequency, our results are very close to reported values of one of
their bovine samples and their human sample for the temperature range from 25 to 50°C.
However, our graph does not increase with temperature as rapidly as their data for higher
temperatures. At 5 MHz, our results lie between their graphs. Compared with Worthington
and Sherar,32 our samples exhibit higher attenuation but have the same trend of increasing
with temperature. The difference between our results and others might stem from a difference in measurement methods, e.g., pulse types (broadband and narrowband) and use of different animal species (bovine, porcine, and canine). Tissue degeneration with temperature
may have an effect on our experiments as well as those of previous authors. Gas bubbles
formed in tissue samples, such as could have occurred by degeneration may have increased
the attenuation coefficient estimate in our experiments. Some researchers degas the tissue
using a vacuum pump before measurement. We did not degas the tissue since it was freshly
excised and would not contain gas bubbles at the start of the experiment. The properties of
tissue close to those in living animals are of interest.
The findings of this study are restricted to an assumption of invariance with temperature of
the tissue holder membrane properties and tissue densities. Measurement accuracies at high
temperatures are also limited by the heat loss that occurs while measuring samples in the
40oC measurement bath. Note also that whether in vitro values accurately represent the in
vivo values that could be applied to a temperature-monitoring algorithm in living bodies still
needs to be established.
CONCLUSION
Variation in the ultrasonic propagation speed and attenuation with temperature are measured for canine liver, kidney, prostate, and muscle tissue. The dependence of these acoustic
parameters on temperature varies with tissue type. For each tissue, curves of speed of sound
or attenuation exhibit three distinct behaviors over the 25-95°C temperature range.
Measurements of propagation speed and attenuation were primarily to obtain calibration
curves for ultrasonic monitoring during thermal ablative treatments. The design of this experiment, where the same tissue sample is continuously heated over the entire temperature
range of 25-95°C, differs from most approaches reported in the literature, where each tissue
sample is raised to a single set temperature. We are currently analyzing differences between
these two approaches. Results of changes in the propagation speed and attenuation with
temperature will be used for calibrating temperature distributions during ultrasound-guided
ablative therapy.
ACKNOWLEDGEMENTS
The authors thank Mr. Larry Whitesell and Ms. Jennifer Puck for providing liver tissue
samples used for the experiments. This work was supported in part by start-up grant funds
258
TECHAVIPOO ET AL
from the Department of Medical Physics, Medical School and Graduate School at the University of Wisconsin-Madison and by NIH grants R01 CA39224.
REFERENCES
1. Goldberg SN, Gazelle GS, Solbiati L, Rittman WJ, and Mueller PR, Radiofrequency tissue ablation: increased lesion diameter with a perfusion electrode, Acad Radiol 3, 636-644 (1996).
2. Lee FT, Jr., Mahvi DM, Chosy SG, Onik GM, Wong WS, Littrup PJ and Scanlan KA, Hepatic cryosurgery
with intraoperative US guidance, Radiology 202, 624-632 (1997).
3. Curley SA, Izzo F, Delrio P, Ellis LM, Granchi J, Vallone P, Fiore F, Pignata S, Daniele B and Cremona F,
Radiofrequency ablation of unresecrfle primary and metastatic hepatic malignancies: results in 123 patients, Ann
Surg 230, 1-8 (1999).
4. Rhim H and Dodd GD, 3rd, Radiofrequency thermal ablation of liver tumors, J Clin Ultrasound 27, 221-229
(1999).
5. Livraghi T, Goldberg SN, Lazzaroni S, Meloni F, Solbiati L and Gazelle GS, Small hepatocellular carcinoma:
treatment with radio-frequency ablation versus ethanol injection, Radiology 210, 655-661 (1999).
6. Heniford BT, Arca MJ, Iannitti DA, Walsh RM and Gagner M, Laparoscopic cryoablation of hepatic
metastases, Semin Surg Oncol 15, 194-201 (1998).
7. Heisterkamp J, van Hillegersberg R and Ijzermans JN, Interstitial laser coagulation for hepatic tumours, Br J
Surg 86, 293-304 (1999).
8. Malcolm AL and ter Haar GR, Ablation of tissue volumes using high intensity focused ultrasound, Ultrasound
Med Biol 22, 659-669 (1996).
9. Chung JW, Park JH, Han JK, Choi BI, Kim TK and Han MC, Transcatheter oily chemoembolization of the inferior phrenic artery in hepatocellular carcinoma: the safety and potential therapeutic role, J Vasc Interv Radiol 9,
495-500 (1998).
10. Lee FT, Jr., Chosy SG, Weber SM, Littrup PJ, Warner TF and Mahvi DM, Hepatic cryosurgery via
minilaparotomy in a porcine model: an alternative to open cryosurgery, Surg Endosc 13, 253-259 (1999).
11. Gervais DA, McGovern FJ, Wood BJ, Goldberg SN, McDougal WS and Mueller PR, Radio-frequency ablation of renal cell carcinoma: early clinical experience, Radiology 217, 665-672 (2000).
12. Livraghi T, Goldberg SN, Solbiati L, Meloni F, Ierace T and Gazelle GS, Percutaneous radio-frequency ablation of liver metastases from breast cancer: initial experience in 24 patients, Radiology 220, 145-149 (2001).
13. Gazelle GS, Goldberg SN, Solbiati L and Livraghi T, Tumor ablation with radio-frequency energy, Radiology 217, 633-646 (2000).
14. Kanaoka Y, Hirai K, Ishiko O and Ogita S, Microwave endometrial ablation at a frequency of 2.45 GHz. A
pilot study, J Reprod Med 46, 559-563 (2001).
15. Waldschmidt J, Schier F, Bein U and Soerensen M, The use of the laser in the treatment of arterio-venous
malformations and vascular tumours of the liver, Eur J Pediatr Surg 3, 217-223 (1993).
16. Ophir J, Cespedes I, Ponnekanti H, Yazdi Y and Li X, Elastography: a quantitative method for imaging the
elasticity of biological tissues, Ultrasonic Imaging 13, 111-134 (1991).
17. O’Donnell M, Skovoroda AR, Shapo BM and Emelianov SY, Internal displacement and strain imaging using
ultrasonic speckle tracking, IEEE Trans Ultrasonics Ferroelec Freq Contr 41, 314-325 (1994).
18. Varghese T, Ophir J, Konofagou E, Kallel F and Righetti R, Tradeoffs in elastographic imaging, Ultrasonic
Imaging 23, 216-248 (2001).
19. Varghese T, Zagzebski J and Lee Jr. F, Elastographic Imaging of thermal lesions in the liver in-vivo following RF ablation: Preliminary Results, Ultrasound Med Biol 28, 1467-1473 (2002).
20. Seip R and Ebbini ES, Noninvasive estimation of tissue temperature response to heating fields using diagnostic ultrasound, IEEE Trans Biomed Engin 42, 828-839 (1995).
21. Simon C, VanBaren P, and Ebbini ES, Two-dimensional temperature estimation using diagnostic ultrasound,
IEEE Trans Ultrasonics Ferroelec Freq Contr 45, 1088-1099 (1998).
22. Damianou CA, Sanghvi NT, Fry FJ and Maass-Moreno R, Dependence of ultrasonic attenuation and absorption in dog soft tissues on temperature and thermal dose, J Acoust Soc Am 102, 628-634 (1997).
TEMPERATURE DEPENDENCE OF PROPAGATION SPEED AND ATTENUATION
259
23. Maass-Moreno R and Damianou CA, Noninvasive temperature estimation in tissue via ultrasound
echo-shifts. Part I. Analytical model, J Acoust Soc Am 100, 2514-2521 (1996).
24. Maass-Moreno R, Damianou CA and Sanghvi NT, Noninvasive temperature estimation in tissue via ultrasound echo-shifts. Part II. In vitro study, J Acoust Soc Am 100, 2522-2530 (1996).
25. Varghese T, Zagzebski JA, Chen Q, Techavipoo U, Frank G, Johnson C, Wright A and Lee FT, Ultrasound
monitoring of temperature change during radiofrequency ablation: preliminary in-vivo results, Ultrasound Med
Biol 28, 321-329 (2002).
26. Varghese T, Zagzebski JA, Techavipoo U and Chen Q, (US Patent P02153US, 2002).
27. Wells PNT, Biomedical Ultrasonics (London, Academic Press, 1977).
28. Goss SA, Johnston RL and Dunn F, Comprehensive compilation of empirical ultrasonic properties of mammalian tissues, J Acoust Soc Am. 64, 423-457 (1978).
29. Duck FA, Physical Properties of Tissue: A Comprehensive Reference Book (Academic Press, San Diego,
1990).
30. Bowen T, Connor WG, Nasoni RL, Pifer AE and Sholes RR, Measurement of the temperature dependence of
the velocity of ultrasound in soft tissues, pp. xi+362 (1979).
31. Wear KA, Temperature dependence of ultrasonic attenuation in human calcaneus, Ultrasound Med Biol 26,
469-472 (2000).
32. Worthington AE and Sherar MD, Changes in ultrasound properties of porcine kidney tissue during heating,
Ultrasound Med Biol 27, 673-682 (2001).
33. Bamber JC and Hill CR, Ultrasonic attenuation and propagation speed in mammalian tissues as a function of
temperature, Ultrasound Med Biol 5, 149-157 (1979).
34. O’Donnell M, Mimbs JW, Sobel BE and Miller JG, Ultrasonic attenuation of myocardial tissue: dependence
on time after excision and on temperature, J Acoust Soc Am 62, 1054-1057 (1977).
35. Lin T, Ophir J and Potter G, Correlations of sound speed with tissue constituents in normal and diffuse liver
disease, Ultrasonic Imaging 9, 29-40 (1987).
36. Tervola KM, Gummer MA, Erdman JW, Jr. and O’Brien WD, Jr., Ultrasonic attenuation and velocity properties in rat liver as a function of fat concentration: a study at 100 MHz using a scanning laser acoustic microscope, J
Acoust Soc Am 77, 307-313 (1985).
37. Shung KK, Fei DY, Yuan YW, and Reeves WC, Ultrasonic characterization of blood during coagulation, J
Clin Ultrasond 12, 147-153 (1984).
38. Wear KA and Armstrong DW, 3rd, Relationships among calcaneal backscatter, attenuation, sound speed,
hip bone mineral density, and age in normal adult women, J Acoust Soc Am 110, 573-578 (2001).
39. Ophir J, Shawker TH, Maklad NF, Miller JG, Flax SW, Narayana PA and Jones JP, Attenuation estimation in
reflection: progress and prospects, Ultrasonic Imaging 6, 349-395 (1984).
40. Maklad NF, Ophir J and Balsara V, Attenuation of ultrasound in normal liver and diffuse liver disease in vivo,
Ultrasonic Imaging 6, 117-125 (1984).
41. Ophir J, Maklad NF and Bigelow RH, Ultrasonic attenuation measurements of in vivo human muscle, Ultrasonic Imaging 4, 290-295 (1982).
42. Parker KJ, Tuthill TA and Baggs RB, The role of glycogen and phosphate in ultrasonic attenuation of liver, J
Acoust Soc Am 83, 374-378 (1988).
43. Parker KJ, Ultrasonic attenuation and absorption in liver tissue, Ultrasound Med Biol 9, 363-369 (1983).
44. Maruvada S, Shung KK and Wang SH, High-frequency backscatter and attenuation measurements of selected bovine tissues between 10 and 30 MHz, Ultrasound Med Biol 26, 1043-1049 (2000).
45. Wear KA, Ultrasonic attenuation in human calcaneus from 0.2 to 1.7 MHz, IEEE Trans Ultrason Ferroelectr
Freq Contr 48, 602-608 (2001).
46. Mimbs JW, O’Donnell M, Miller JG and Sobel BE, Changes in ultrasonic attenuation indicative of early
myocardial ischemic injury, Am J Physiol 236, H340-344 (1979).
47. Burke TM, Madsen EL and Zagzebski JA, X-ray linear attenuation coefficients in the mammographic range
for ultrasonic breast phantom materials, Radiology 142, 755-757 (1982).
48. Madsen EL, Zagzebski JA, Insana MF, Burke TM and Frank G, Ultrasonically tissue-mimicking liver including the frequency dependence of backscatter, Med Phys 9, 703-710 (1982).
260
TECHAVIPOO ET AL
49. Madsen EL, Dong F, Frank GR, Garra BS, Wear KA, Wilson T, Zagzebski JA, Miller HL, Shung KK, Wang
SH, Feleppa EJ, Liu T, O’Brien WD, Jr., Topp KA, Sanghvi NT, Zaitsev AV, Hall TJ, Fowlkes JB, Kripfgans OD
and Miller JG, Interlaboratory comparison of ultrasonic backscatter, attenuation, and speed measurements, J Ultrasound Med 18, 615-631 (1999).
50. Greenspan M and Tschegg CE, rfles of the speed of sound in water, J Acoust Soc Amer 31, 75-76 (1959).
51. Kinsler LE, A.R. F, Coppens AB, and Sanders JV, Fundamental of Acoustics, Third Edition (John Wiley,
New York, 1982).
52. Gill AE, in Atmosphere-Ocean Dynamics; Vol. Appendix 3 (Academic Press, New York, 1982), pp.
599-602.
53. Goldman DE and Richards JR, Measurement of high-frequency sound velocity in mammalian soft tissues, J
Acoust Soc Amer 26, 981-983 (1954).