 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to methods and apparatus for performing
ultrasonic diagnosis of a target body. More particularly, the invention
pertains to methods and apparatus for the measurement of sound speed in a
target body. The invention is especially concerned with techniques for
enhancing the accuracy of sound velocity measurements in compressible
targets using one or more ultrasonic transducers in pulse-echo mode.
2. Description of Related Art
Traditional ultrasonic diagnosis is achieved by transmitting ultrasonic
energy into a target body and generating an image from the resulting echo
signals to survey anatomical structures. A transducer is used to both
transmit the ultrasound energy and to receive the echo signals. During
transmission, the transducer converts electrical energy into mechanical
vibrations. Acquired echo signals produce mechanical oscillations in the
transducer which are reconverted to electrical signals for amplification
and recognition.
A plot or display (e.g., on an oscilloscope, etc.) of the electrical signal
amplitude vs. echo arrival time yields the amplitude line (A-line) or echo
sequence corresponding to a particular ultrasonic transmission. When the
A-line is displayed directly as a sinusoidal pattern modulating at radio
frequency (RF) it is referred to as an RF or "undetected" signal. For
imaging, the A-line is often demodulated to a non-RF or "detected" signal.
Ultrasound techniques have been extensively used in the field of diagnostic
medicine as a non-invasive means of analyzing the properties of tissue in
vivo (i.e., living). A human or animal body represents a non-homogenous
medium for the propagation of ultrasound energy. Acoustic impedance
changes at boundaries of varying density and/or sound speed within a
target body. A portion of the incident ultrasonic beam is reflected at
these boundaries. Inhomogeneities within the tissue form lower-level
scatter sites that result in additional echo signals. Images may be
generated from this information by modulating the intensity of pixels on a
video display in proportion to the intensity of echo sequence segments
from corresponding points within the target body.
Conventional imaging techniques are widely used to evaluate various
diseases within organic tissue. Imaging provides information concerning
the size, shape and position of soft tissue structures using the
assumption that sound velocity within the target is constant. Qualitative
tissue characterization is carried out by interpretation of the grey scale
appearance of the echograms. Qualitative diagnosis largely depends on the
skill and experience of the examiner as well as system characteristics.
However, images based only on relative tissue reflectivity cannot be used
for a quantitative assessment of disease states.
Techniques for quantitative tissue characterization using ultrasound are
needed for more accurate diagnosis of disease. One of the most promising
parameters for quantitative measurement is sound speed. Speed of sound
changes within regions of varying density and/or molecular compressibility
within the tissue. Thus, it is expected that changes in tissue density due
to disease will result in changes in the speed of sound. Indeed, it has
been shown that changes in the speed of sound in tissue often correlate
with tissue pathology. For example, cirrhotic liver tissue has been
observed to contain more fat than normal liver tissue. The velocity of
sound in cirrhotic tissue would therefore be expected to be lower than in
normal tissue. Similarly, changes in tissue density in the region of
tumors may result in changes in sound velocity in the tumor region.
Unfortunately, however, such changes are relatively small and account for
up to only 10% of the speed of sound in normal tissue. Therefore, accuracy
in sound velocity estimation is extremely important in the analysis of
tissue for pathological conditions. Usually, the accuracy of sound
velocity estimations must be at least 1.0% to have specific value for
quantitative tissue characterization. Hence, a need exists for the
accurate measurement of sound velocity in organic tissue for clinical
diagnosis.
Traditionally, measurement of sound speed has been conducted with
transmission techniques. A first method of sound velocity measurement
involves the transmission of sound pulses through tissue regions of known
dimension and recording the time required for the pulse to traverse the
region. The quotient of travel distance and travel time is computed to
yield the velocity. However, due to the softness of most tissues, the
dimensions of the tissue sample cannot be accurately measured which
results in an error-prone measurement of sound velocity. Moreover, a
reference liquid with a known speed of sound may be required to calibrate
the apparatus.
A second transmission technique that has been used in medical diagnostics
involves a transmitting transducer and a separate receiving transducer
arranged so that they are aimed at one another with their respective axes
of radiation coincident. The body of the subject is placed between the
transmitting and receiving transducers. However, in vivo application of
this technique has been limited to accessible organs like the breast or
testes; other in vivo applications can be adversely affected by such
factors as bowel gas, bone and inaccessibility.
A third transmission technique is disclosed by Ophir and Lin, "A
Calibration-Free Method for Measurement of Sound Speed in Biological
Tissue Samples", IEEE Transactions on Ultrasonics, FerroeIectrics, and
Frequency Control, Vol. 35, No. 5, (1988) 573-577. This method allows
accurate measurement of the speed of sound in soft tissue samples, while
overcoming the limitations of initial techniques. The method employs a
receiving hydrophone and a transmitting transducer that are coaxially
aligned opposite each other. The transmitting transducer is in contact
with the tissue sample, while the hydrophone penetrates the tissue sample
at well-controlled incremental depths. The transit times of the pulse are
recorded for all penetration depths of the hydrophone. These transit times
are then plotted against the relative depths of the hydrophone, and a
linear regression fit is made to the data. The slope of the fitted line is
c.sup.-1, where c is the estimated speed of sound in the tissue sample.
The technique requires neither calibration involving a reference medium,
nor the knowledge of the thickness of the tissue sample. Yet, while this
technique is capable of accurate measurements of tissue in vitro, it is
clearly not suitable for speed of sound estimations in vivo.
Several techniques have been proposed for the measurement of sound velocity
in vivo using ultrasonic transducers in pulse-echo mode. In one method,
sound speed is measured using misregistration between pulse-echo images of
the same structure obtained with two different sound beams. Sound velocity
is determined from the difference in position of the same feature in
different images. This method works best when a well-defined feature is
available. In simulated tissue regions, known as "phantoms", thin wire
added to the region will provide such a well-defined feature. However,
well defined features are not easy to find in living tissue and the
resulting sound speed measurement is therefore not as accurate. See
Robinson et al., "Measurement of Velocity of Propagation from Ultrasonic
Pulse-Echo Data", Ultrasound in Med. & Biol., Vol. 8, No. 4, (1982)
413-320.
In another pulse-echo technique called the "focus adjustment method", the
mean sound speed between a reflector and linear array transducer is
measured using the following three parameters: time of flight, time of
flight difference, and distance between two receiver elements. To detect
time of flight, the system delay-line time compensator is adjusted to
obtain the sharpest reflector image. Thus, the sharpness of the target is
maximized by interactive user control of signal delays at the transducer
aperture. However, irregular tissue structures cause random refractions of
the ultrasonic beams and make sharp focusing difficult. Also, the method
is highly dependent on qualitative judgment. See Hayashi et al., "A New
Method of Measuring In vivo Sound Speed in the Reflection Mode", J. Clin.
Ultrasound, Vol. 16, (1988) 87-93.
A third pulse-echo method described in U.S. Pat. No. 4,669,482, involves in
vivo sound velocity estimation by identifying segments of different sound
velocity along a tracked ultrasonic beam using at least two
widely-separated acoustic vantage points. The tracked beam is partitioned
into at least two contiguous segments, the boundary between the two
segments being the inner body of the body wall fat. A plurality of
ultrasound pulse travel time measurements are made, each with a different
apparent angle of intersection between the tracked beam and the tracking
beam. For each measurement, techniques are employed for correcting
refraction occurring in a transverse plane. Data pairs collected in the
plurality of measurements are fitted to an appropriate equation using
curve-fitting techniques well known in the art, by which the index of
refraction at the body wall inner boundary, the inclination of the inner
boundary, and the speed of sound in the internal tissue are derived. This
technique, however, is not desirable in clinical settings because of the
large "footprint" of the apparatus on the patient that results in a
cumbersome examination procedure. Also, inaccuracies due to bone and/or
bowel gases are common because of the wide spacing between transmitting
and receiving transducers.
Hence, all the above pulse-echo techniques are clinically limited due to
the need to use two widely separated acoustic vantage points and/or by the
requirement that an identifiable, discrete target be available in the
tissue. The use of two widely separated vantage points makes the apparatus
and the examination procedure cumbersome, while the existence of a
discrete target cannot always be guaranteed. Another potential problem is
due to the effects of the overlying fat layer of the body on the
estimation.
SUMMARY OF THE INVENTION
The present invention provides an improved pulse-echo method and apparatus
that has particular application in estimating sound velocity in organic
tissue. The present invention addresses the problems of prior pulse-echo
techniques by providing a relatively small footprint and obviating the
need for a readily identifiable, discrete target within the tissue.
According to the present invention, a standard transducer or transducer
containing device is translated transaxially, thereby compressing or
displacing a proximal region of a target body in small known increments.
At each increment, a pulse is emitted and an echo sequence (A-line) is
acquired from regions within the target along the sonic travel path or
beam of the transducer. Segments of the echo sequence corresponding to a
distal region within the target are selected as a reference to estimate
the incremental change in echo arrival time. A plot of these arrival time
estimates versus the target compression depth is then generated and a
least squares linear fit is made. The slope of the linear fit is c.sup.-1,
where c is an estimate of the speed of sound in the tissue.
The present invention takes advantage of the acoustical properties of
physically compressible or displaceable materials. These materials often
contain a large number of acoustic "scatterers." These scatterers, being
small compared to the wavelength of the sound frequencies involved, tend
to reflect incident sound energy in all directions. For example, in
homogeneous tissue regions, the scatterers may comprise a collection of
nearly identical reticulated cells. The combined reflections from each
scatterer create a background echo signal called speckle. The present
invention employs standard pattern matching techniques to track a
reference echo sequence segment corresponding either to a reflector or
other echo source, such as speckle, in a distal tissue region within the
target body. See, e.g., J. S. Bandat and A. G. Piersol, "Random Data:
Analysis and Measurement Procedures," Wiley Interscience, New York 1971,
pp. 30-31. A discrete reflector, like a bone or blood vessel, may be used
as a reference if desired, but is not necessary; any arbitrary segment of
the backscattered echo sequence may be used as a reference.
Bias occasioned by distal deformation of the reference echo source due to
the proximal compression or displacement of the target may be corrected by
using a second stationary transducer. The second transducer is oriented
such that its beam intersects the beam of the first transducer at a small
angle within the region of the reference echo source. The echo time delay
due to the distal deformation is detected by the second transducer and is
used to unbias the sound velocity estimate. While two acoustic vantage
points are used, they are maintained at close proximity to each other, so
that the total transducer "footprint" on the target is no larger than that
which is due to a standard transducer array.
The present invention is of particular interest in interrogating organic
tissue, especially human and other animal tissue. A principal object of
such interrogation is to detect echo signals in the tissue that may
suggest the presence of abnormalities. More specifically, the effect of
compression or displacement of the tissue on the characteristics of the
echo signals becomes a possible key to such detection. It will be noted at
this point that the invention is contemplated to have significant
applications other than in the study of tissue. One such application, for
example, may be materials and products such as cheese or crude oil that
are physically compressible or displaceable by movement of a transducer.
Thus, as a transducer is pressed against such a material, particles within
the material are displaced from one position to another. For elastic
materials, release of the pressure enables the particles to return to
their original position.
It will be noted that the transducers employed in the present invention
need not be in direct contact with the materials to which they are
applied. It is necessary, however, that transducers be sonically coupled
to the materials. Sonic coupling methods and agents are well known in the
art.
It will also be noted that a material may be interrogated according to the
invention either (a) by advancing a transducer against a material to
increase compression, or (2) by retracting a transducer from a compressed
position within the material.
As noted above, it is not necessary that an echo from a discrete feature in
a tissue or other compressible material be employed. It is sufficient that
an identifiable echo segment be present in the echo signal resulting from
a transmittal signal. Even though the physical feature within a material
responsible for a selected echo sequence segment may not be clearly known,
the selected echo segment is an adequate reference for the purposes of the
invention. Thus, compression of the material and the signal travel times
determined before and after such compression may be based on such echo
segments.
As stated above, the invention may be practiced either by compressing a
transducer against a compressible material from an initially
non-compressed condition, or by retracting a transducer from an initially
compressed condition. In either case, however, it is preferable that the
distance traveled by the transducer be less than the wavelength of the
ultrasonic signal produced or received by the transducer.
The present invention may also be employed for localized estimation of
sound speed in targets having multiple layers. The speed of sound in each
of progressively deeper layers is sequentially estimated by employing the
same techniques discussed above. Distal regions at layer boundaries are
used as the echo source for arrival time estimates. According to the
present invention, the speed of sound can be estimated in each layer from
only two echo sequences along the axis of radiation. Thus, imaging of the
speed of sound parameter in a plane or volume of a target body can also be
accomplished by appropriate lateral translation of the transducers.
Other objects and advantages of the invention will become readily apparent
from the ensuing description.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1a shows an embodiment of the invention where one transducer is
sonically coupled to a target body to interrogate a distal tissue region
within the target body;
FIG. 1b shows a plot of the RF echo signal originating from the distal
tissue region interrogated in FIG. 1a;
FIG. 2a shows the transducer of FIG. 1a imparting a small compression to a
proximal region of the target body;
FIG. 2b shows a plot of the time shifted RF echo signal originating from
the distal tissue region interrogated in FIG. 2a;
FIG. 3a shows the transducer of FIG. 1a imparting a further compression to
a proximal region of the target body;
FIG. 3b shows a plot of the further time shifted RF signal from the tissue
region interrogated in FIG. 3a;
FIG. 4a shows an embodiment of the invention where both compressing and
noncompressing transducers are acoustically coupled to a target body to
interrogate a distal tissue region within the target body;
FIG. 4b shows a plot of the RF signal originating from the distal tissue
region interrogated in FIG. 4a from the vantage point of the
noncompressing transducer;
FIG. 4c shows a plot of a time shifted RF signal originating from the
distal region interrogated in FIG. 4a; from the vantage point of the
noncompressing transducer;
FIG. 5 shows an embodiment of the invention where two transducers are used
to interrogate multiple tissue layers;
FIG. 6 shows an embodiment in which a transducer is sonically coupled to a
target via a stand-off device containing an acoustic coupling fluid; and
FIG. 7 is a plot comparing corrected and uncorrected speed of sound
estimations in simulated tissue.
DETAILED DESCRIPTION
The basic method resembles the penetrating hydrophone transmission
technique discussed above. An adaptation of this technique to the
pulse-echo mode is used. A transducer is positioned on or otherwise
coupled to a target body and advanced axially toward the target in small
known increments. As noted earlier, the invention may also be practiced by
incrementally retracting a transducer from a previously compressed
position. Since the relatively large aperture size precludes penetration
of the tissue, small tissue compressions occur instead. At each increment,
a pulse is emitted and echo sequence (A-line) segments from one or more
selected distal tissue regions are used as a reference. Any arbitrary
segment of the backscattered RF echo signal from within the tissue may be
identified and used as a reference. The selected segment --wavelet--of the
RF signal corresponds to a particular echo source within the tissue along
the beam axis of the transducer. As the transducer compresses the tissue,
it moves closer to the echo source, thereby shortening the travel path of
the pulse and corresponding echo. The change in arrival times for echoes
originating from the echo source as the transducer is incrementally
advanced (or retracted) is related to the speed of sound in the tissue.
Thus, the speed of sound may be determined even though the distance
between the transducer aperture and the selected echo source are unknown.
The present invention contemplates transducers that may be piezoelectric,
ferroelectric or magnetostrictive in nature. The present invention is not
limited by the size, focusing properties or bandwidth of the transducer to
be employed.
FIG. 1a shows the transducer 10 sonically coupled to a target body 15. An
ultrasonic pulse 18 is shown propagating within beam 20 toward a echo
source 25 on beam axis 12. As the pulse 18 propagates through the target
15, corresponding echoes are generated and arrival times noted at the
transducer aperture 11. The combination of all echoes generated from
reflections within the beam 20 is the echo sequence or A-line
corresponding to pulse 18. A radio frequency ("RF") signal plot of the
A-line acquired from pulse 18 is shown in FIG. 1b. The amplitude of the
signal in millivolts is plotted against echo arrival times in microseconds
(.mu.s). Latter arrival times correspond to progressively deeper regions
within the target body 15. An echo wavelet 30, within a chosen arrival
time window, is selected as a reference. The time window may be selected
based on anatomical data from ultrasound imaging, or may be arbitrary,
e.g., every x micro seconds. The wavelet 30 originates from the echo
source 25 that is at an unknown distance from the transducer aperture 11.
FIG. 2a shows the transducer 10 being translated along axis 12 to impart a
small compression (y.sub.1) to the tissue. Alternatively, as shown in FIG.
6, a transducer 80 may be associated with a stand-off device 85 which
allows the transducer 80 to be acoustically or sonically coupled to the
target body 90 without being in direct contact with the target body. In
this case the stand-off 85, and not the transducer, compresses the target.
In either case, however, the incremental compressions of the transducer or
transducer containing device are dependent on the frequency of the
transducer employed. More specifically, the magnitude of the incremental
compressions are based on the wavelength which is a function of transducer
frequency. In general, incremental shifts of less than about one
wavelength are employed unless a discrete target is used as a reference.
Otherwise, tracking the reference signal segment will be complicated by
phase wrap. For example, in ophthalmic diagnosis a transducer of about 20
mHz may be employed, whereas a transducer of 3-5 mHz would be suitable for
interrogating abdominal tissue. When a transducer of 3-5 mHz is used, the
compressions are generally on the order of several mm, preferably between
0.1-2 mm.
After the transducer 10 compresses the target, a second pulse 22 is emitted
and the corresponding A-line segment is acquired from a desired depth
within the tissue. FIG. 2b shows the RF plot of a time shifted A-line
corresponding to pulse 22. The wavelet segment or 32 associated with echo
source 25 is also time shifted. The time shifted wavelet 32 is tracked
within the selected time window using standard pattern matching
techniques. The arrival time of wavelet 32 is prior to that of wavelet 30
above, since the distance between aperture 11 and feature 25 was shortened
by the compression Y.sub.2.
FIG. 3a shows further tissue compression (y.sub.1) and a third pulse 24
emitted after the compression. The RF plot of the A-line in FIG. 3b shows
an additional time shift in the signal. The wavelet 35 is tracked within
the selected time window and is used to note the signal time shift.
Assuming uniform sound speed and no displacement of the echo source
involved in producing the RF signal wavelet of interest, the sound speed
estimate in the tissue contained between the transducer and the location
of these scatterers is:
##EQU1##
where n is the number of uniform transducer compressional displacements,
y.sub.i is the ith compression, and t.sub.i is the ith measured temporal
shift in the reference echo signal wavelet. The factor of 2 in the
numerator accounts for the pulse-echo nature of the technique in which
ultrasound (pulses) travels to and returns (echoes) from the echo source
in the selected distal region. However, the method of the present
invention is not limited to a particular algorithm for calculating the
sound speed characteristics of a target body.
According to the present invention, the one transducer embodiment discussed
above may be conveniently employed in instances in which the target body
being interrogated contains very compressible materials. Also, the method
may be adapted to compress the tissue and acquire an A-line segment prior
to the arrival of an elastic wave associated with the proximal
compression. This is possible because, although the elastic wave travels
at about 20 meters per second (m/s), the ultrasonic pulse travels at about
1540 m/s. Thus, the A-line is obtained from the selected time window prior
to the arrival of the elastic wave. However, this is not feasible in some
instances. In these cases, the assumption of no distal feature
displacement is inadequate. Although the displacements of echo sources
within the target will generally fall off asymptotically with range,
minute displacements may occasionally be detected even far from the
transducer. When this occurs, it is necessary to make a correction for the
distal displacements.
To correct the estimate, the expression of eq. (1) is modified to reflect
the presence of additional, unknown time delays t.sub.d,i due to such
displacements indicated by the subscript d. Therefore, the resulting
modified estimate of the speed of sound is:
##EQU2##
Since the quantities (t.sub.i -t.sub.d,i).ltoreq. t.sub.i are the actual
time delays that are measurable, the estimate is always positively biased
unless the t.sub.d,i =0.
Fortunately, the quantities t.sub.d,i can be independently estimated using
a second transducer. This is shown in FIG. 4a. In addition to the
compressing transducer 38, a stationary noncompressing transducer 40 is
used, whose beam axis 42 is directed such that it intersects the beam axis
52 of the compressing transducer 38 at the range that corresponds to the
echo source 50. The noncompressing transducer 40 operates in the
pulse-echo mode and detects minute displacements of the echo source 50 in
the region of beam intersection that appear at time shif | | |
