Biomedical Engineering Reference
In-Depth Information
changes in ultrasonic tissue characteristics with temperature.
These changes have been investigated both theoretically and
in vitro
with an eye toward using these signals for noninva-
sive monitoring of thermal therapy. Most of these investigators
have looked at the consequences of changes in SOS and thermal
expansion with temperature that causes echo shifts or changes
in the amplitudes of backscattered ultrasonic signals.
with deionized water, which had been degassed by vacuum
pumping in an appropriate vessel. Degassing is necessary to pre-
vent the formation of bubbles during heating. Gas bubbles scat-
ter ultrasound and corrupt quantitative assessment of ultrasonic
parameters.
Most ultrasonic imaging systems are based on notebook com-
puters or other forms of computer-based systems. The computer
in these imaging systems can be used to control tissue heating
by setting the temperature of a circulating heater. The tempera-
ture in the tissue, monitored by a thermocouple, can be reported
to a routine in the computer of the imaging system. When the
desired temperature is reached the heater can be turned off and
an image frame or image loop acquired and saved. The computer
can also be used to automatically control the position of the
transducer array position in the elevation direction via stepper
motor to acquire 3D image sets. This process can be fully auto-
mated. For example, all keystrokes needed to switch between
peripheral control and imaging can be administered using key-
stroke-emulation software, such as AutoIT (hiddensoft.com).
An alternate for some imaging system is to control all of the
operations via a proprietary software developers kit.
13.4.1 Experimental Concerns
Among the inherent advantages of using ultrasound for tem-
perature imaging is that it is nonionizing, convenient, and rela-
tively inexpensive. A-mode signals and phased-array images
are routinely produced in real time. To take advantage of these
attributes, effective temperature imaging should be expected to
operate in real time with relatively simple additional signal pro-
cessing requirements that employ existing ultrasonic equipment
using ultrasonic properties extracted, if possible, from a single
backscatter view.
To obtain sufficiently accurate measures of ultrasonic proper-
ties for temperature imaging, it may be necessary to compen-
sate for known limitations of image generation in conventional
imaging systems. These include effects of the measurement
system itself, insertion loss, reflection and transmission losses,
attenuation in tissue, and effects of beam diffraction.92-99
92-99
13.4.1.3 temperature Imaging during
Non-Uniform Heating
Heat sources in thermal therapies produce nonuniform tem-
perature distributions because of the nature of the source
itself or because of inhomogeneity in the tissue being heated.
Hyperthermia heating is more nearly uniform than ablation
therapy, which is concentrated to destroy a tumor or aberrant
pathway in the heart, for example. The computer of the imaging
system can be used to automate heating and image acquisition,
as noted previously.
A nonuniform heat source may have many forms depending
on the therapy. Heat sources include microwave antennas and
low-frequency ultrasonic arrays for hyperthermia, and high-
intensity ultrasound and RF electrodes for high-temperature
ablation. These sources may be used in experiments to test
ultrasonic thermometry for clinical settings, but other sources
may be used to develop ultrasonic temperature imaging in the
laboratory.
We created the fixture shown in Figure 13.5 to heat gelatin
phantoms and tissue specimens
102
from a central hot-water
source, while the phantom or tissue is surrounded by water
at fixed reference temperature, usually 37°C. This fixture uses
a thermocouple grid to corroborate ultrasonic temperature
images estimates. A 3D image set can be taken in which the
tips of the thermocouples on each side of the specimen can be
seen in the first and last images of the set, which are not used
in generating temperature images, but can be used to place
the thermocouple reading in the temperature images to assess
their accuracy. Furthermore, with the metal of the thermo-
couples and their holders removed, the fixture can be used to
perform MRI temperature imaging with the same hot-water
heating source.
13.4.1.1 temperature Standards
The conventional standard for temperature measurements in
tissue is a reading from a thermocouple, which can fit into a
hypodermic needle. Because of their invasive nature, however,
thermocouple grids can only be used sparsely in tissue. For
in
vivo
temperature imaging MRI is the standard modality dur-
ing thermal therapy.
38-40,100,101
Nevertheless, MRI volumetric
measurements themselves are corroborated with thermocouple
readings.
Thermocouples are calibrated with a National Institute of
Standards and Technology traceable thermometer. Typically,
individual thermocouples are accurate to within ±0.1°C.
To calibrate the thermocouples and the system that monitors
them, they may be placed in a water bath with a heater. At equi-
librium for a temperature of interest in the water bath, the NIST
traceable standard reading and the thermocouple value are taken
simultaneously. This process is repeated over the temperature
range of interest. Typically, multiple calibration experiments
must be conducted to assess the mean thermocouple errors and
their standard deviations. The thermocouple monitoring system
may employ an internal reference that varies each time the unit
is used. This offset must be measured and used to correct the
thermocouple reading during each experiment to calibrate ther-
mocouple temperatures to within 0.2°C.
13.4.1.2 Calibration of tissue properties
during Uniform Heating
Measurements of temperature-dependent ultrasonic properties
can be made by heating specimens in an insulated tank filled
