Biomedical Engineering Reference
In-Depth Information
over all 1 cm 3 volumes was <0.5°C for temperatures from 38 to
42°C over all possible estimation tests using a single, separate
cubic centimeter region for calibration and < 1°C for all tem-
peratures. These results show that CBE, whose thermal behavior
is monotonic, can serve as the basis for accurate volumetric tem-
perature imaging.
estimation using echo shifts. Motion must be tracked and correc-
tion applied before temperature can be estimated using changes
in backscattered energy. Given the importance of determining
tissue motion for estimating temperature with ultrasound, and
that its determination is the most computationally intensive part
of ultrasonic temperature imaging, it is likely that more efficient
and sophisticated methods for motion tracking will be applied to
this problem. If temperature estimation based on both echo shift
and CBE continue to show promise, the accuracy and reliability
of temperature estimation with ultrasound may be enhanced by
a technique that combines both approaches.
Echo shift, attenuation, and CBE methods depend on being
able to accurately calibrate, that is, find the thermal sensitiv-
ity to those parameters. The echo shift method, in addition to
precise knowledge of the echo shift with tissue depth, requires
a knowledge of α, the linear coefficient of thermal expansion,
and β, a descriptor of the change in SOS with temperature.
Similarly, calibration of the CBE method requires knowledge
of how backscattered energy changes for a given type of tis-
sue. The sensitivity to noise for these calibration data has yet
to be fully determined for either method. That sensitivity will
determine the temperature accuracy that is possible for a given
spatial resolution for either method. A crucial step in identi-
fying a viable ultrasonic approach to temperature estimation
remains careful evaluation of its performance during in vivo
application.
Volumetric temperature imaging is an important tool for
guiding and assessing the effects of thermal therapy. Although
MRI provides volumetric temperature information, it is expen-
sive, adds increased complexity and duration to treatments, and
may not be available in many hospitals. If ultrasound methods,
which have been successful at imaging complex temperature
patterns in vitro, can be shown to provide adequate thermal and
spatial accuracy in vivo , they would provide important advan-
tages over MRI temperature imaging. The significance of such
systems would be high and could be an important element in
increasing the utilization of thermal therapies.
13.4.4.4 temperature Imaging during
Nonuniform Heating
To generate and validate CBE temperature images during non-
uniform heating, the fixture shown in Figure 13.5 was used to
hold and heat abattoir specimens of turkey breast muscle. 102
The specimen was allowed to come to equilibrium at 37°C, then
65°C water was pumped through a silicone tube in the center of
the specimen, using ultrasonic gel as a coupling agent between
the tissue and the tube.
3D ultrasonic volumes were obtained at 7.5 MHz by moving a
transducer array in the elevation direction under stepper motor
control. The first and last images of the 3 mm thick 3D set cap-
tured the position of thermocouple tips for validating the CBE
temperature images, as shown in Figure 13.14. The CBE temper-
ature images were found after motion compensation using the
sensitivity of CBE to temperature determined in the calibration
studies, namely 0.3 dB/°C.
Figure 13.14 also shows the CBE-estimated and measured
temperatures at the location of four thermocouples at three
times after the start of heating. For this specimen, CBE tem-
peratures errors were within about 1°C over the 900 s period of
heating. Similar results were found over multiple specimens. 102
As expected, the estimated temperature patterns showed tem-
perature increasing outwards from the heat source with consis-
tent patterns over time, although not in a radial pattern as seen
in homogeneous gelatin phantoms, perhaps due to inhomoge-
neous tissue structures.
13.5 Discussion
13.6 Summary and Conclusions
Ultrasonic thermometry is based on thermal effects in soft tissue
that are manifest as changes in the speed of sound, attenuation,
and backscatter of ultrasound. The use of echo shifts and changes
in backscattered energy are the most promising techniques for
estimating temperatures in the hyperthermia (41-45°C) range.
Both methods are useful to temperatures above 60°C, which
make them attractive for temperature imaging in the border
zones of high-temperature ablations. Although attenuation is
more difficult to measure than either echo shift or change in
energy, its increased sensitivity to temperatures above 50°C also
makes it of interest for assessing high-temperature ablation.
A key processing step common to all the methods just men-
tioned is the tracking of apparent motion of scattering regions in
vitro . Motion detection takes on even more importance during
in vivo studies because of the likely additional motion of the sub-
ject. Time shift as a function of depth is the basis for temperature
Ultrasound is an attractive modality for volumetric temperature
imaging to monitor thermal therapies because it is nonioniz-
ing, portable, convenient, inexpensive, and has relatively simple
signal-processing requirements. This modality has proven use-
ful for estimation of temperatures from the hyperthermia range
(41- 45°C) to border zones of regions of high-temperature abla-
tion (>60°C).
The most prominent methods for exploiting ultrasound as a
noninvasive thermometer rely on either (1) echo shifts due to
changes in tissue thermal expansion and speed of sound (SOS),
(2) variation in the attenuation coefficient, or (3) change in back-
scattered energy from tissue inhomogeneities. Each method has
its strengths in terms of temperature range for which it yields
a useful thermal signal and how well it can handle tradeoffs
between temperature accuracy and spatial resolution.
 
Search WWH ::




Custom Search