 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to an ultrasonic diagnostic imaging system that uses
nonlinear distortion for imaging tissue inside a body.
2. Description of the Prior Art
Ultrasonic diagnostic imaging systems for imaging tissue inside a body by
utilizing nonlinear propagation distortion caused by harmonics occurring
during ultrasonic wave propagation are well known in the art. In such a
system, a transducer is driven alternately by a first and a second drive
pulse of A and 2A, respectively, in amplitude. The transducer responsively
transmits a first and a second ultrasonic wave, which are reflected by
tissue in the body and received by the transducer as a first and a second
ultrasonic echo of B and 2B, respectively, in amplitude. The first and
second echoes are amplified by a variable gain amplifier with gains of C
and C/2 to yield a first and a second signal of B.multidot.C and
2B.multidot.(C/2), respectively. Since the sidelobes of the first and
second echoes are much smaller than the main lobes and accordingly small
in distortion, the amplitudes of the sidelobes of the second echo are
substantially twice those of the sidelobes of the first echo. Thus
calculating the differences between the first and second signals, i.e.,
B.multidot.C-2B.multidot.(C/2) enables the detection of the depth of
reflection point. Since a pair of pulses is used for each analysis, such
systems as described above are called "two-pulse" systems. A first and a
second pulse in such a two-pulse system are hereinafter referred to as a
"former pulse" and a "latter pulse".
However, in order for the above imaging technique to work satisfactorily,
the reflection points or ultrasonic wave transmission directions from
which the former and later echoes are obtained must be substantially the
same. This restriction prevents high-speed scanning in conventional
nonlinear distortion-based ultrasonic diagnostic imaging system.
SUMMARY OF THE INVENTION
In light of the above, it is an object of the present invention to provide
a nonlinear distortion-based ultrasonic diagnostic imaging system which
displays a raised-resolution video of tissue inside a body at an increased
frame rate.
According to an aspect of the invention, a transducer transmits a
ultrasonic wave pulse in response to a driving pulse while scanning the
transmission direction in response to a scan control signal and receives
an echo of the ultrasonic wave pulse to provide an echo signal. A
transducer driver supplies the driving pulses and the scan control signal
to the transducer such that the transducer transmits weaker and stronger
ultrasonic wave pulses alternately while putting the same intervals
between adjacent ultrasonic wave pulses to obtain a weaker echo of the
weaker ultrasonic wave pulse and a stronger echo of the stronger
ultrasonic wave pulse from the transducer. An equalizer equalizes each
weaker echo to the stronger echo into an equalized weaker echo. An
interpolator calculates an interpolation value between the equalized
weaker echo and an equalized previous weaker echo obtained from a previous
weaker echo. For each weaker ultrasonic wave pulse, a detector finds a
value indicative of a difference between the interpolation value and a
stronger echo obtained between the weaker echo and the previous weaker
echo. An image processor generates a raised-resolution video signal of the
tissue at an increased frame rate on the basis of the values and the scan
control signal.
In one embodiment, the equalizer calculates a convolution by using each
weaker echo as one of two components.
In the embodiment, the transducer driver may supply a narrower driving
pulse and a wider driving pulse for the weaker and stronger ultrasonic
wave pulses, respectively. Alternatively, the transducer driver may supply
fewer driving pulse(s) for the weaker ultrasonic wave pulse and supply
more driving pulses for the stronger ultrasonic wave pulse. These driving
pulses have an identical width.
In the embodiment, the interpolator calculates an arithmetic means of said
equalized weaker echo and said equalized previous weaker echo.
Alternatively, an arithmetic means of the absolute values of the equalized
weaker echo and the equalized previous weaker echo may be calculated.
BRIEF DESCRIPTION OF THE DRAWING
The features and advantages of the present invention will be apparent from
the following description of an exemplary embodiment of the invention and
the accompanying drawings, in which:
FIG. 1 is a schematic block diagram showing an arrangement of an ultrasonic
diagnostic imaging system according to an illustrative embodiment of the
invention;
FIG. 2 is a diagram showing waveforms of driving pulses with respective
different pulse widths T1 and T2;
FIG. 3 is a graph showing the relationship between the fundamental wave and
the second harmonic of an ultrasonic echo;
FIG. 4 is a diagram showing the relationship between the azimuth (i.e., the
angle with a normal on the transmission surface of transducer 22) and the
amplitude of the transmitted ultrasonic wave; and
FIG. 5 is a diagram showing various signals for illustrating the operation
of the ultrasonic diagnostic imaging system of FIG. 1
Throughout the drawing, the same elements when shown in more than one
figure are designated by the same reference numerals.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
FIG. 1 is a schematic block diagram showing an arrangement of an ultrasonic
diagnostic imaging system according to an illustrative embodiment of the
invention. In FIG. 1 the ultrasonic diagnostic imaging system 1 comprises
a transducer driver 10 for alternately providing a former and a latter
driving pulse different from each other in spectral intensity and a probe
20, which includes a transducer 22 for transmitting a ultrasonic wave
pulse in response to a driving pulse and receiving an echo of the
transmitted ultrasonic wave pulse. The probe 20 has its scan data input
20a connected to the controller scan control output 100c and its
transducer terminal 20b connected to the transducer driver 10 output. The
system 1 further comprises an analog-to-digital (A/D) converter 30 having
its analog input connected to the transducer terminal 20b and its control
input connected to the controller output 100b; an equalizer 40 having its
input connected to the output of the A/D converter 30; an interpolator 50
having its input connected to the equalizer 40 output; a memory 60 for
temporary storing one pulse's worth of digital echo samples from the A/D
converter 30; a detector 70 which uses the interpolator 50 output and the
A/D converter 30 output being temporarily stored in the memory 60 to
detect a signal indicative of the depth of reflection point; an image
processor 80 having its data input connected to the detector 70 output and
its control input connected to the controller output 100c; a display
device 90 having its input connected to the image processor 80; and a
controller 100 which controls the operation of the whole system 1
especially by providing control signals 100a through 100c.
Since the driving pulses from the transducer driver 10 typically have a
high voltage, the A/D converter 30 is preferably provided with a limiter
(not shown). The interpolator 50 is preferably provided with a not-shown
memory (or interpolator memory) with a capacity enough to store one
pulse's worth of equalized digital echo samples from the equalizer 40. The
memory 60, which is shown as an independent memory in FIG. 1, may be a
part of random access memory (not shown) included in the controller 100.
The controller 100 may be any suitable microprocessor-based controller.
In operation, the transducer driver 10 alternately outputs a former and a
latter driving pulse different from each other in duty cycle in response
to a transmission control signal from the controller 100 output terminal
100a as shown in FIG. 5A. FIG. 2 shows the former driving pulse pa(t) and
the latter driving pulse pb(t), which means that the former and latter
driving pulses are expressed by respective functions of time t, i.e.,
pa(t) and pb(t). The pulses preferably have three values, i.e., 0 and
positive and negative levels of a predetermined amplitude. The pulses have
respective pulse widths T1 and T2. FIG. 5 shows various signals for
illustrating the operation of the ultrasonic diagnostic imaging system 1
of FIG. 1. In FIG. 5, a letter "j" is used to indicate the sequence of
pulses (i.e., "j" is a serial number assigned to each pair of a former and
a latter pulse in order of generation). For example, in FIG. 5A, the
current former driving pulse is denoted by pa(t, j) and the previous
former driving pulse is denoted by pa(t, j-1). In the same way, ultrasonic
echoes of the ultrasonic wave pulses transmitted in response to the
driving pulses pa(t, j) and pa(t, j-1) are denoted by ra(r, j) and ra(t,
j-1), respectively. However, if there is no need of differentiating the
sequence of the pulses, we will use simplified expressions like pa(t),
ra(t), etc., omitting the sequence ID terms in the following.
The transducer 22 alternately transmits former ultrasonic waves ga(t) and
latter ultrasonic waves gb(t)that are in a fundamental frequency band and
correspond to the former pa(t) and latter pb(t) driving pulses.
Preferably, the probe 20 is so arranged as to automatically scan the
direction of ultrasonic wave transmission in response to the scan control
data from the controller output 100c. Since the transducer 22 has a
narrower frequency band width as compared with the driving pulses,
changing the spectral intensity of the driving pulse (i.e., changing the
pulse width of the driving pulse with its amplitude kept constant) enables
the control of the amplitude of the transmitted ultrasonic waves. For this
reason, the former pa(t) and latter pb(t) driving pulses with respective
pulse widths of T1 and T2 cause the transducer 22 to transmit the former
ga(t) and latter gb(t) ultrasonic waves of respective amplitudes
responsive to T1 and T2.
FIG. 3 shows the relationship between the fundamental wave and the latter
harmonic in an echo of a transmitted ultrasonic wave pulse. As seen from
FIG. 3, the ultrasonic wave pulses ga(t) and gb(t) transmitted from the
transducer 22 increase in nonlinear distortion as they travel a longer
path within the body. The larger the amplitude of the ultrasonic waves is,
the harder the nonlinear distortion is. Since the nonlinear distortion is
due to harmonics, especially, due to the latter harmonic, the fundamental
wave component decreases in amplitude as the latter harmonic increases.
For this reason, the peak portion of the main lobe, in which the amplitude
of the beam of ultrasonic wave pulse is relatively large, is subjected to
larger nonlinear distortion, while the sidelobes, in which the amplitude
is relative small, are subjected to smaller nonlinear distortion.
The former ga(t, j) and latter gb(t, j) ultrasonic wave pulses transmitted
from transducer 22 in response to the driving pulses pa(t, j) and pb(t) is
reflected by tissue within the body, and returned to and received by
transducer 22 as a former and a latter ultrasonic echo ra(t, j) and rb(t,
j), respectively, as shown in FIG. 5B. Each of former ra(t, j) and latter
rb(t, j) echo pulses is sampled and converted by A/D converter 30 into a
series of digital echo samples (or signals), ra(k, j) and rb(k, j), as
shown in FIG. 5C, where k=1, 2, . . . , N, where N is the number of
digital echo samples for one driving or echo pulse.
In order to facilitate the understanding of the invention, it is now
assumed that the transducer driver 10 has just supplied a j-th former
driving pulse pa(t, j) and, accordingly, now is just the time to analyze
echo pulses ra(t, j-1), rb(t, j-1) and ra(t, j) to get the j-1)th result.
At the time of transmission of a j-th former ultrasonic wave pulse ga(t,
j) from the transducer 22, the digital samples of the j-1)th former echo
pulse, i.e., ra(1, j-1), ra(2, j-1), . . . , ra(N, j-1) (hereinafter,
denoted as {ra(k, j-1).vertline.k=1.about.N}) have been stored in memory
of either interpolator 50 or controller 100 (not shown), and the digital
samples of the j-1)th latter echo pulse, i.e., rb(1, j-1), rb(2, j-1), . .
. , rb(N, j-1) (hereinafter, denoted as {rb(k, j-1).vertline.k=1.about.N})
have been stored in memory 60 as shown in FIG. 1. Then, each of the
digital samples of the j-th former echo pulse ra(t, j) which are supplied
from A/D converter 30 is processed on a sample by sample basis. In the
following, we will discuss how the k-th sample ra(k, j) of the j-th former
echo pulse ra(t, j) is processed along the circuit path following A/D
converter 30.
Specifically, the k-th former echo digital sample ra(k, j) is equalized by
equalizer 40 into an equalized digital sample rb'(k, j) as detailed later.
Interpolator 50 uses the just equalized signal rb'(k, j) for interpolation
together with the corresponding one rb'(k, j-1) of the equalized digital
samples of the preceding former echo rb'(1, j-1), rb'(2, j-1), . . . ,
rb'(N, j-1). For this purpose, interpolator 50 preferably retains the
recent N equalized samples:
rb'(k, j-1), rb'(k+1, j-1), . . . , rb'(N, j-1), rb'(1, j), rb'(2, j), . .
. , rb'(k-1, j).
Then, interpolator 50 has only to use the just equalized signal rb'(k, j)
and the oldest one of the stored signals, rb'(k, j-1) to calculate and
output an interpolation value si(k, j-1).
It is noted that as shown in FIG. 1 the recent N equalized samples are
actually stored in the following order:
rb'(1, j), rb'(2, j), . . . , rb'(k-1, j), rb'(k, j-1), rb'(k+1, j-1), . .
. , rb'(N, j-1). (data 1)
This is because, on completing the calculation of interpolation value si(k,
j-1), interpolator 50 writes the newest (or just used) equalized sample
rb'(k, j) over the oldest (or just used) one rb'(k, j-1) of the equalized
digital samples (data 1) stored in the interpolation 50 memory.
The detector 70 calculates the difference between the interpolator 50
output ri(k, j-1) and the corresponding one rb(k, j-1) of the digital
samples of the preceding (i.e., j-1)th) latter which are stored in memory
60 as follows:
.DELTA.r(k, j-1)=ri(k, j-1)-rb(k, j-1).
The image processor 80 processes thus obtained differences .DELTA.r(k, j-1)
for k=1.about.N for each of j=1, 2, . . . together with the scan data from
the controller output terminal 100c to provide video images of tissue
inside the body. The video images are displayed on the display device 90.
The principles of the invention, especially, the operation of equalizer 40
and interpolator 50 will be detailed in the following. The Fourier
transforms for a former pa(t) and a latter pb(t) driving pulse are denoted
by Pa(.omega.) and Pb(.omega.), where .omega. is the angular frequency of
the former and latter driving pulses. Similarly, the Fourier transforms
for a former ga(t) and a latter gb(t) ultrasonic wave pulse are denoted by
Ga(.omega.) and Gb(.omega.). Also, assuming the impulse response of the
transducer 22 to be h(t), then the Fourier transform for the impulse
response h(t) is denoted by H(.omega.).
Then, since a transmitted ultrasonic wave pulse ga(t) is expressed by the
convolution of the impulse response h(t) and the driving pulse pa(t), it
follows:
ga(t)=h(t)*pa(t) (1)
where X*Y indicates the convolution of X and Y. This means
Ga(.omega.)=H(.omega.).times.Pa(.omega.). (2)
Multiplying the both sides of equation (2) by Pb(.omega.)/Pa(.omega.), we
obtain
##EQU1##
Expressing the equation (3) in the time domain yields
gb(t)=ga(t)*invf(Pb(.omega.)/Pa(.omega.)) (4)
where the function invf(F(.omega.)) indicates the inverse Fourier transform
for the function F(.omega.). The equation means that calculating the
convolution between the former ultrasonic wave function ga(t) of the time
when transducer 22 is a driven by a driving pulse pa(t) and the function
invf(Pb(.omega.)/Pa(.omega.)) yields the latter ultrasonic wave function
gb(t) of the time when transducer 22 is driven by a driving pulse pb(t).
Assuming that a returned echo of a transmitted ultrasonic wave is expressed
by a linear combination of the transmitted ultrasonic wave, then the
equation (4) can be written, for j-th former and latter echoes, as:
rb(t, j)=ra(t, j)*invf(Pb(.omega., j)/Pa(.omega., j)). (5)
From this equation, it is seen that if equalizer 40 calculates the
convolution of a j-th former echo ra(t, j) and the function
invf(Pb(.omega., j)/Pa(.omega.), j)), then equalizer 40 must provide a
j-th latter echo rb(t, j). However, since the ultrasonic echoes ga(t) and
gb(t) differ in amplitude, the nonlinear distortions in the ultrasonic
echoes ga(t) and gb(t) also differ in degree. Taking this difference into
account, the equation (5) should be written as:
rb(t, j)=ra(t, j)*invf(Pb(.omega., j)/Pa(.omega., j))+.DELTA.r(t, j). (6)
Since the first term of the right side of equation (6) can be calculated by
equalizer 40 as:
rb'(t, j)=ra(t, j)*invf(Pb(.omega., j)/Pa(.omega., j)). (7)
The calculation of equation (7) by equalizer 40 can be realized by, for
example, a digital filter etc.
Using rb'(t, j) in equation (6) yields
rb(t, j)=rb'(t, j)+.DELTA.r(t, j). (8)
Since the signals in a circuit path which follows A/D converter 30 are
digital samples, equation (8) can be expressed as:
rb(k, j)=rb'(k, j)+.DELTA.r(k, j). (9)
However, since the scanning directions or positions (i.e., reflection
points of transmitted latter gb(t) and former ga(t) ultrasonic wave
pulses) that caused the ultrasonic echoes rb(t) and ra(t) (i.e., rb'(k,
j)), respectively, are actually different from each other as seen from
FIG. 5C, equation (9) is not valid as it is. In order to make the signals
rb'(k, j) or ra(k, j) uniform in the scanning direction, the value rb'(k,
j) is replaced, in interpolator 50, with:
##EQU2##
By doing this, the difference .DELTA.r(k, j) in equation (9) is given, in
detector 70, by:
.DELTA.r(k, j)=rb(k, j)-ri(k, j). (11)
Considering that the pulse numbers j and j-1indicate the current pulse and
the preceding pulse, respectively, in actual operation, FIGS. 1 and 5C are
drawn such that interpolator 50 calculates:
##EQU3##
detector 70 calculates
.DELTA.r(k, j-1)=rb(k, j-1)-ri(k, j-1). (11')
The difference ri(k, j) is regarded as a value caused by the peak portion
of the main lobe in the latter or larger-amplitude ultrasonic echo rb(t,
j) and indicates the depth of reflection point.
According to the present invention, as seen from FIG. 5C, the depth of
reflection point (or tissue inside the body) in the scanning direction of
a weaker and stronger ultrasonic wave pulse pair is detected by using
three successive scanning points including one used for the preceding
pair. Since such three successive scanning points are permitted to be
specially apart from one other, this enables a high-speed scanning, i.e.,
displaying an increased number of frames per unit time, permitting the
motion of tissue to be displayed smoothly.
However, it is noted that it is preferable to place the same intervals
between adjacent driving signals.
Also, since the difference .DELTA.r(k, j-1) includes substantially no
sidelobe components, high-resolution images are obtained.
Modification
Interpolator 50 may calculate
##EQU4##
instead of equation (10').
Detector 70 may calculate
.DELTA.r(k, j-1)=.vertline.rb(k, j-1).vertline.-ri(k, j-1).vertline. (13)
instead of (11').
If equation (12) or (13) is used, then the use of absolute value eliminates
phase components, causing only amplitude information to be used. This
frees the difference .DELTA.r(k, j-1) from becoming too large due to
variation in phases of received echoes.
In the above illustrative embodiment, driving pulses of different pulse
widths are used for driving pulse pairs. Pulse pairs may be realized by
changing the number of pulses of a narrow pulse width.
A filter for compensating the spectral difference between the former and
the latter driving pulses may be used for equalizer 40.
In the above illustrative embodiment, the weaker ultrasonic echoes have
been equalized to the stronger ultrasonic echoes. Alternatively, the
stronger ultrasonic echoes may be equalized to the weaker ultrasonic
echoes.
Many widely different embodiments of the present invention may be
constructed without departing from the spirit and scope of the present
invention. It should be understood that the present invention is not
limited to the specific embodiments described in the specification, except
as defined in the appended claims.
* * * * *
|
|
 |
|
 |
Description |
|
