JPH08131439A - Correlation device and flow information display device - Google Patents

Abstract

PURPOSE: To obtain a correlation device capable of improving an S/N and shortening processing time by providing a Fourier transformation means which finds a frequency band waveform from a time area waveform, a cross-spectrum arithmetic means which finds a cross-spectrum from the frequency band waveform, and an autocorrelation arithmetic means which calculates a correlation value by a specific equation. CONSTITUTION: The Fourier transformation part 1 of this correlation device 100 performs Fourier transformation on two time area waveforms p1(n.T), p2(n.T) represented in H discrete data at every sampling time T, and outputs the frequency band waveforms P1(n.Δω), P2(n.Δω). A complex conjugate part 2 outputs the frequency band waveform P1-(n.Δω) conjugate with the frequency band waveform P1(n.Δω). A complex arithmetic part 3 multiplies the frequency band waveform P2(n.Δω) conjugate with the frequency band waveform P1-(n.Δω), and outputs the cross-spectrum U(n.Δω)+j.V(n.Δω). An autocorrelator 4 calculates and outputs the correlation value (τ) by an equation.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a correlation device and a flow information display device. More specifically, the present invention relates to a correlator that generates correlation values of two time-domain waveforms and a flow information display device that extracts and displays flow information from an ultrasonic echo signal from a subject obtained by an ultrasonic probe.

[0002]

2. Description of the Related Art FIG. 8 shows a conventional flow information display device 700.
It is a block diagram of an example. This flow information display device 70
At 0, the ultrasonic probe 11 and the transmitter / receiver 12 sequentially transmit M (for example, 10) ultrasonic pulses to the region of interest of the subject at a time interval Tp (for example, 1 ms), and transmit ultrasonic waves from the region of interest. The sound wave echo signal pm (t) is received and input to the quadrature detection unit 74. Here, M is called a packet size. m = 1 to M. Mixer 74 of quadrature detector 74
a and 74b multiply the reference signal from the reference signal generators 74c and 74d by the ultrasonic echo signal pm (t), and pass LPFs (Low Pass Filters) 74e and 74f to make them in phase with the quadrature component Qm. Output the component Im. The A / D conversion units 75a and 75b respectively perform A / D conversion on the quadrature component Qm and the in-phase component Im, and the memories 76a and 7b.
Write to 6b. MTI (Moving Target Indication)
The filters 17a and 17b include the memories 75a and 75b.
Then, the quadrature component Qm and the in-phase component Im are read out, the clutter component is removed, and the quadrature component Qm ′ and the in-phase component Im ′ are output. The clutter component is a Doppler component from a tissue such as a relatively slow moving heart wall. The autocorrelator 4 calculates the Doppler frequency f by the autocorrelation calculation of the following equation with respect to the quadrature component Qm 'and the in-phase component Im'.

[0003]

(Equation 3)

DSC (Digital Scan Convertor) 19
Converts the Doppler frequency f into a display value, and the monitor 20
Displayed on the screen. For example, if the Doppler frequencies f are obtained from a large number of two-dimensionally distributed regions of interest, the Doppler frequencies f are converted into luminance × color, and the two-dimensional distribution is displayed on the screen, a CFM (Color Flow Mapping) image is obtained. Becomes In the above flow information display device 700, the quadrature detection unit 74 and the autocorrelator 4 work as a correlation device.

FIG. 9 is a conceptual diagram showing the function of the autocorrelator 4. When the input of the autocorrelator 4 is discrete data obtained by sampling the time domain waveform Im ′ + j · Qm ′ of the time period To for each time Tp, the output of the autocorrelator 4 is the reciprocal of the time period To, that is, the frequency f. Become.

On the other hand, Japanese Unexamined Patent Publication No. 62-251685 discloses a correlator having no quadrature detector (correlation circuit) and a flow information display device (device for determining a transfer factor in a medium) including the correlator. Has been done.

[0007]

In the flow information display device 700, unless the packet size M is increased, it is difficult to improve the performance of the MTI filters 17a and 17b that remove clutter components. Moreover, there is only a series of short waveforms, and the frequency estimation error is large. However, when the packet size M is increased, there is a problem that the time resolution becomes poor. Also,
Since the frequency of the reference signal in the quadrature detection unit 74 is constant, the ultrasonic echo signal does not utilize frequency components other than the frequency of the reference signal, and the measurable range is limited to the transmission pulse interval. (Ie, there is a problem with aliasing).

On the other hand, the correlation device and the flow information display device disclosed in Japanese Patent Laid-Open No. 62-251685 have a problem that the hardware configuration becomes complicated. In addition, there is a problem that the time resolution is limited by the sampling time T of the discrete data (for this reason, an interpolation circuit is required).

Therefore, an object of the present invention is to improve the SN ratio even if the packet size M is reduced, the processing time is short and the hardware configuration is simple, and the sampling time T allows the time to be shortened. It is an object of the present invention to provide a correlation device and a flow information display device in which the resolution is not limited and the problem of aliasing is greatly improved.

[0010]

According to a first aspect, the present invention provides two time-domain waveforms p1 (n · T) and p2 (n · n) represented by N discrete data for each sampling time T.
A correlation device for generating a correlation value τ of T), wherein frequency domain waveforms P1 (n · Δω) and P2 (n · Δω) are converted from the time domain waveforms p1 (n · T) and p2 (n · T). Fourier transforming means to be obtained and the frequency domain waveforms P1 (n · Δω), P
2 (n · Δω) to cross spectrum U (n · Δω) +
cross spectrum calculation means for obtaining j · V (n · Δω),

[0011]

[Equation 4]

There is provided a correlating device comprising an autocorrelation calculating means for calculating a correlation value τ by the equation (4).

In a second aspect, the present invention is a flow information display device for extracting and displaying flow information from an ultrasonic echo signal from a subject obtained by an ultrasonic probe.
A first ultrasonic pulse is transmitted to the subject, ultrasonic echo signals from the region of interest are sampled at every time T, and a first time-domain waveform p1 (nT) represented by N discrete data is obtained. A second time-domain waveform p2 represented by N discrete data is obtained by transmitting a second ultrasonic pulse to the subject and sampling the ultrasonic echo signal from the region of interest at every time T.
Data collecting means for obtaining (n · T) and the time domain waveform p
Frequency range waveform P1 from 1 (nT) and p2 (nT)
Fourier transform means for obtaining (n · Δω) and P2 (n · Δω), and the frequency range waveforms P1 (n · Δω) and P2 (n
・ Δω) to cross spectrum U (n ・ Δω) + j ・ V
A cross spectrum calculation means for obtaining (n · Δω) and the cross spectrum U (n · Δω) + j · V (n · Δ
Cross spectrum U'with clutter component removed from ω)
MTI that generates (n · Δω) + j · V ′ (n · Δω)
Filter means,

[0014]

(Equation 5)

Flow information display characterized by comprising autocorrelation calculation means for calculating the correlation value τ by the equation (5) and display means for converting the correlation value τ into a display value and displaying it on the screen. Provide a device.

[0016]

In the correlator according to the first aspect, the frequency domain waveforms P1 (n · Δω) and P2 (n · Δ) are transformed from the time domain waveforms p1 (n · T) and p2 (n · T) by the Fourier transform means.
ω). Next, those frequency range waveforms P1 (n · Δ
ω), P2 (n · Δω) from the cross spectrum U (n · Δω) + j · V (n ·
Δω) is calculated. Then, the correlation value τ is calculated by the calculation of the following equation by the autocorrelation calculating means.

[0017]

(Equation 6)

According to this, it is possible to sufficiently increase the number N of discrete data (for example, 64 points). Moreover, since all the frequency components included in the time domain waveform can be utilized, the SN ratio is good. Further, the processing time is short and the hardware configuration is simple. Furthermore, the sampling time T does not limit the time resolution, and the problem of aliasing is greatly improved.

In the flow information display device according to the second aspect, the data collecting means transmits the first ultrasonic pulse to the subject, and the ultrasonic echo signal from the region of interest is transmitted at the time T.
Each time sampling is performed, a first time domain waveform p1 (n · T) represented by N discrete data is obtained. Also, a second ultrasonic pulse is transmitted to the subject, ultrasonic echo signals from the region of interest are sampled at every time T, and a second time-domain waveform p2 (n) represented by N discrete data is generated.・ Get T). Next, the time domain waveforms p1 (n · T), p2 (n · T)
Frequency domain waveform P1 (n
・ Δω) and P2 (n · Δω) are calculated. Next, the cross spectrum U (n
・ Δω) + j ・ V (n ・ Δω) is calculated. Next, those cross spectra U (n · Δω) + j · V (n · Δω)
The MTI filter means to remove the clutter component from
Cross spectrum U ′ (n · Δω) + j · V ′ (n · Δ
ω) is generated. Then, the correlation value τ is calculated by the calculation of the following equation by the autocorrelation calculating means.

[0020]

(Equation 7)

According to this, since the number N of discrete data is sufficiently large (for example, 64 points), the SN ratio does not deteriorate even with two ultrasonic pulses (packet size M = 2), and the tap of the MTI filter ( increase the number of taps (for example, 30)
Is possible and leads to high performance MTI filtering. Further, the processing time is short and the hardware configuration is simple. Furthermore, the sampling time T does not limit the time resolution, and the problem of aliasing is greatly improved.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in more detail with reference to the embodiments shown in the drawings. The present invention is not limited to this.

First Embodiment FIG. 1 is a block diagram of a correlation device 100 according to the first embodiment of the present invention. The correlator 100 includes a Fourier transform unit 1
, A complex conjugating unit 2, a complex multiplying unit 3, and an autocorrelator 4. When the two time-domain waveforms p1 (nT) and p2 (nT) represented by N discrete data for each sampling time T are input, the Fourier transform unit 1 respectively Fourier transforms them. Frequency domain waveform P1 (n
・ Δω) and P2 (n · Δω) are output. Where n =
1 to N, Δω = 2π / (N · T).

The complex conjugating unit 2 has the frequency domain waveform P1.
When (n · Δω) is input, the frequency domain waveform P1 ^* (n · Δω) conjugate with it is output. The complex multiplication unit 3 multiplies the frequency range waveform P1 ^* (n · Δω) by the conjugate frequency range waveform P2 (n · Δω) to obtain the cross spectrum U.
(N · Δω) + j · V (n · Δω) is output. When P1 (n · Δω) = Re1 (n · Δω) + j · Ig1 (n · Δω) P2 (n · Δω) = Re2 (n · Δω) + j · Ig2 (n · Δω), P1 ^* ( n · Δω) = Re1 (n · Δω) −j · Ig1 (n · Δω), and U (n · Δω) = Re1 (n · Δω) · Re2 (n · Δω) + Ig1 (n · Δ)
ω) ・ Ig2 (n ・ Δω) V (n ・ Δω) = Re1 (n ・ Δω) ・ Ig2 (n ・ Δω) -Re2 (n ・ Δ
ω) · Ig1 (n · Δω). The autocorrelator 4 calculates and outputs the correlation value τ by the following equation.

[0025]

(Equation 8)

As shown in FIG. 2, the input of the autocorrelator 4 is the frequency domain waveform U (n) + j · V (n) of the frequency period ωo.
Is discrete data sampled for each frequency Δω, the output of the autocorrelator 4 is the reciprocal of the frequency period ωo, that is, the time correlation value τ. For example, the time domain waveform p1
When the time domain waveform p2 (n · T) is obtained by delaying (n · T) by the delay time Td, the correlation value τ = Td.

According to the above correlation device 100, the number N of discrete data is sufficiently large, so that the SN ratio is good. Further, since the quadrature detector is not used, the time domain waveforms p1 (nT), p
All frequency components contained in 2 (n · T) can be utilized. Further, the processing time is short and the hardware configuration is simple. Furthermore, the sampling time T does not limit the time resolution, and the problem of aliasing is greatly improved.

Second Embodiment FIG. 3 is a flow information display device 200 according to the second embodiment of the present invention.
FIG. This flow information display device 200
Then, the ultrasonic probe 11, the transmitter / receiver 12, and the A / D converter 13 transmit the first ultrasonic pulse to the subject, sample the ultrasonic echo signal from the region of interest at each time T, The first time-domain waveform p1 (n · T) represented by N discrete data is obtained, and the first time-domain waveform p1 (n · T)
Memory 15a. Further, after the time interval Tp (for example, 1 ms), the second ultrasonic pulse is transmitted to the subject, the ultrasonic echo signal from the region of interest is sampled at every time T, and is represented by N discrete data. The second time domain waveform p2 (n · T) is obtained and stored in the second memory 15b via the changeover switch 14. Here, the first time domain waveform p1 (n · T) is set to p1 (n · T) = p (n · T), and the region of interest is the ultrasonic probe 11 during the time interval Tp.
If it moves in a direction further away, the second time domain waveform p2 (n · T) is expressed by p2 (n · T) = α · p (n · T + Td). Where α is the damping ratio according to the distance moved, T
d is the delay time according to the distance moved.

The computing unit 16 reads the first time domain waveform p1 (n · T) from the first memory 15a and Fourier transforms it to obtain a first frequency domain waveform P1 (n · Δω). Further, the second time domain waveform p2 (n · T) is read from the second memory 15B and subjected to Fourier transform to obtain the second frequency domain waveform P2 (n · Δω). Next, those frequency range waveforms P1 (n · Δω) and P2 (n · Δω)
To cross spectrum U (n · Δω) + j · V (n · Δ
ω) is calculated and output.

The MTI filters 17a and 17b remove clutter components from the cross spectrum U (n · Δω) + j · V (n · Δω) to obtain the cross spectrum U ′ (n ·
Δω) + j · V ′ (n · Δω) is output. Number of data N
By increasing the number of taps, the number of taps of the MTI filter can be increased, so that the filter performance can be improved. The autocorrelator 4 calculates and outputs the correlation value τ by the following equation. This correlation value τ becomes equal to the delay time Td.

[0031]

[Equation 9]

The DSC 19 converts the correlation value τ into a display value and displays it on the screen of the monitor 20. For example, the correlation value τ is obtained from each of a number of two-dimensionally distributed regions of interest,
A CFM image is obtained by converting the correlation value τ into luminance × color and displaying it in a two-dimensional distribution on the screen. In the above flow information display device 100, the calculator 16 and the autocorrelator 4 function as a correlator.

FIG. 4 shows the first time domain waveform p1 (n · T).
It is an illustration figure of the 2nd time domain waveform p2 (nT). The amplitude ratio of the first time-domain waveform p1 (n · T) is 1
This is a waveform in which a pulse representing a 0: 1 clutter component and a pulse representing a signal component are combined with a time shift of 0.44 μs. The second time domain waveform p2 (n · T) is a waveform in which a pulse representing a clutter component having an amplitude ratio of 10: 1 and a pulse representing a signal component are combined with a time shift of 2.22 μs. Sampling time T = 0.05μ
s, the number of data N = 128.

FIG. 5 shows the first frequency range waveform P1 (n.Δ
ω) real part Re1 (n · Δω) and imaginary part Ig1 (n ·
Δω) and the real part R of the second frequency domain waveform P2 (n · Δω)
It is an illustration figure of e2 (n * (DELTA) (omega)) and imaginary part Ig2 (n * (DELTA) (omega)). These are the time domain waveforms p1 (n
T) and p2 (n · T) are Fourier-transformed waveforms. FIG. 6 shows the cross spectrum U (n · Δω) + j
It is an illustration figure of V (n * (DELTA) (omega)). This is the first of FIG.
Is calculated from the frequency range waveform P1 (n · Δω) and the second frequency range waveform P2 (n · Δω). Figure 7
Is the cross spectrum U ′ (n · Δω) + j · V ′ (n
It is an illustration figure of ((DELTA) (omega)). This is 0.90 from the cross spectrum U (n · Δω) + j · V (n · Δω) in FIG.
The time shift component of μs or less is cut by the MTI filter. Cross spectrum U ′ (n · Δω) of FIG.
When the correlation value τ was calculated from + j · V ′ (n · Δω) by the equation (9), τ = 2.27 μs was obtained.

According to the above flow information display device 200,
Since the number N of discrete data is sufficiently large, the SN ratio does not deteriorate even with two ultrasonic pulses (packet size M = 2). Further, since the quadrature detection unit is not used, all frequency components contained in the ultrasonic echo signal can be utilized. Further, the processing time is short and the hardware configuration is simple. Furthermore, the sampling time T does not limit the time resolution, and the problem of aliasing is greatly improved.

-Other Embodiments-In Embodiments 1 and 2, it is assumed that only two ultrasonic pulses are transmitted, but three or more ultrasonic pulses are transmitted, and the correlation value τ
You may obtain the average of. For example, transmitting four shots, combining the discrete data obtained from the first and second ultrasonic pulses with the discrete data obtained from the third and fourth ultrasonic pulses, increasing the number N of data, The number of data that enters the MTI filter may be increased.

[0037]

According to the correlation device of the present invention, the number N of discrete data is sufficiently large, so that the SN ratio is good. Further, since the quadrature detection unit is not used, all frequency components included in the time domain waveform can be utilized. Further, the processing time is short and the hardware configuration is simple. Furthermore, the sampling time T does not limit the time resolution, and the problem of aliasing is greatly improved.

Further, according to the flow information display device of the present invention, the SN ratio does not deteriorate even if the packet size is reduced, so that the time resolution can be improved. Moreover, since all the frequency components included in the ultrasonic echo signal can be utilized, the measurement accuracy can be improved. Further, the processing time is short and the hardware configuration is simple. Furthermore, the sampling time T does not limit the time resolution, and the problem of aliasing is greatly improved.

[Brief description of drawings]

FIG. 1 is a block diagram of a correlation device according to a first embodiment of the present invention.

FIG. 2 is a conceptual diagram showing the function of an autocorrelator in the correlation device of the first embodiment.

FIG. 3 is a block diagram of a flow information display device according to a second embodiment of the present invention.

FIG. 4 is an exemplary diagram of a first time domain waveform and a second time domain waveform.

FIG. 5 is an exemplary diagram of a real part and an imaginary part of a first frequency range waveform and a real part and an imaginary part of a second frequency range waveform.

FIG. 6 is an exemplary diagram of a cross spectrum.

FIG. 7 is an exemplary diagram of a cross spectrum from which a clutter component has been removed.

FIG. 8 is a block diagram of a conventional flow information display device.

9 is a conceptual diagram showing the function of an autocorrelator in the flow information display device of FIG.

[Explanation of symbols]

 100 Correlator 1 Fourier Transform Part 2 Complex Conjugate Part 3 Complex Multiplier Part 4 Auto Correlation Part 200,700 Flow Information Display Device 11 Ultrasonic Probe 12 Transceiver 13, 75a, 75b A / D Converter 14 Changeover Switch 15a, 15b, 76a, 76b memory 16 arithmetic unit 17a, 17b MTI filter 19 DSC 20 CRT 74 quadrature detector

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location G01S 15/89

Claims (2)

[Claims] 1. Two time-domain waveforms p1 (n · T) and p2 represented by N discrete data for each sampling time T.
A correlation device for generating a correlation value τ of (n · T), wherein frequency domain waveforms P1 (n · Δω) and P2 (n · T) are obtained from the time domain waveforms p1 (n · T) and p2 (n · T). Fourier transform means for obtaining Δω) and the frequency domain waveform P1 (n · Δ)
ω), P2 (n · Δω) from the cross spectrum U (n ·
Cross spectrum calculation means for obtaining Δω) + j · V (n · Δω), and A correlation device comprising: an autocorrelation calculation means for calculating a correlation value τ by the formula (1). 2. A flow information display device for extracting and displaying flow information from an ultrasonic echo signal from an object obtained by an ultrasonic probe, wherein the first ultrasonic pulse is transmitted to the object. The ultrasonic echo signal from the site is sampled at every time T to obtain the first time-domain waveform p1 (n · T) represented by N discrete data, and the second ultrasonic pulse is transmitted to the subject. A second time-domain waveform p2 represented by N discrete data obtained by sampling the ultrasonic echo signal from the region of interest at every time T
Data collecting means for obtaining (n · T) and the time domain waveform p
Frequency range waveform P1 from 1 (nT) and p2 (nT)
Fourier transform means for obtaining (n · Δω) and P2 (n · Δω), and the frequency range waveforms P1 (n · Δω) and P2 (n
・ Δω) to cross spectrum U (n ・ Δω) + j ・ V
A cross spectrum calculation means for obtaining (n · Δω) and the cross spectrum U (n · Δω) + j · V (n · Δ
Cross spectrum U'with clutter component removed from ω)
MTI that generates (n · Δω) + j · V ′ (n · Δω)
Filter means, and A flow information display device comprising: an autocorrelation calculation means for calculating a correlation value τ by the equation (2) and a display means for converting the correlation value τ into a display value and displaying it on a screen.