JPH11142425A - Flow velocity measuring device and ultrasonograph - Google Patents

Abstract

PROBLEM TO BE SOLVED: To process signals accurately with a simple circuit or Doppler flowmeter using high-frequency ultrasonic waves. SOLUTION: A receiving transducer 5 receives ultrasonic signals or reflected wave signals that have been reflected by a fluid. The reflected wave signals are amplified in a radio-frequency amplifier 10, and only a target domain is chopped out by a band-pass filter 11 and then subjected to under-sampling in an A/D converter 13. If the carrier frequency is 998 kHz and the spectrum of the reflected wave signals lies within the range of ±2 kHz, for example, the under-sampling is carried out with the sampling frequency of 8 kHz, with the results that reflected mappings of a multiple order are formed within the range of 0 to 4 kHz and the carrier frequency is transferred to 2 kHz. The mappings are chopped out and fed into a DSP 14 for the formation of spectra lying near the base band through direct sampling, and not through down- conversion using a mixer.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a flow velocity measuring device for measuring the flow velocity of a fluid using the Doppler effect of ultrasonic waves, and an ultrasonic diagnostic apparatus for diagnosing a human body using ultrasonic waves.

[0002]

2. Description of the Related Art A heart-lung machine is provided with a blood flow meter for measuring the amount of blood supplied. A commonly used blood flow meter irradiates ultrasonic waves to a tube through which blood flows, and measures the flow velocity by Doppler shift of a reflected wave reflected by the blood. In a heart-lung machine, it is necessary to accurately supply a predetermined amount of blood to a patient, and therefore, measurement of the flow rate requires high accuracy.

[0003]

In an artificial heart-lung machine, there are cases where blood is supplied in a steady flow, and blood is supplied in a pulsatile flow resembling a heartbeat, and FFT sampling for analyzing the Doppler shift frequency is performed. If the time is long, the flow velocity that changes with respect to the time of the pulsating flow cannot be measured accurately, and if the sampling time of the FFT is short, the flow velocity of the steady flow can be measured with high accuracy. Can not.

In order to obtain a reflected wave of a fluid such as blood,
In the current meter, a high frequency of about several MHz is used. To process a signal of this frequency, a down converter using a mixer has conventionally been used. However, when the mixer is used, there is a problem that noise is mixed in or a signal is attenuated, and measurement accuracy is deteriorated. Ultrasound diagnostic devices that irradiate the human body with ultrasonic waves and diagnose the human body with the reflected waves have been put to practical use, but this ultrasonic diagnostic device also uses ultrasonic waves of several MHz as carrier signals to be transmitted. Therefore, there is a similar problem in processing of the received reflected wave signal.

An object of the present invention is to realize a high frequency resolution and a time resolution with a simple circuit in a Doppler flow velocity measuring apparatus using a high frequency, and to further simplify the circuit to reduce the influence of noise and the like. It is an object of the present invention to provide an ultrasonic diagnostic apparatus which has been removed.

[0006]

According to a first aspect of the present invention, there is provided a transmitting ultrasonic vibrator and a receiving ultrasonic vibrator which are obliquely disposed with respect to a fluid flowing in a predetermined direction, Means for receiving a reflected wave signal of the ultrasonic signal transmitted by the ultrasonic transducer from the fluid by the receiving ultrasonic transducer, and measuring a flow velocity of the fluid by a Doppler shift amount of the reflected wave signal. In the flow velocity measuring device, sampling means for sampling the reflected wave signal,
While obtaining the flow velocity by Fourier analysis using short-time sampling data, and when it is determined that the flow velocity is substantially constant based on the result of the analysis, the flow velocity is determined by Fourier analysis using longer-time sampling data. Analysis means to be obtained.

The invention according to claim 2 of the present application is characterized in that when the flow velocity obtained by the analysis means becomes a negative value in the predetermined direction, a warning is issued as a backflow. .

According to a third aspect of the present invention, the analyzing means includes a frequency shifting means for shifting the center frequency of the reflected wave signal to 0 Hertz by multiplying the sampling data sequence by a discrete complex exponential function. Means for performing a Fourier analysis using data obtained by thinning out the shifted sampling data at a predetermined decimation rate.

The invention of claim 4 of the present application is characterized in that the analysis means includes interpolation means for interpolating a discrete spectrum obtained by Fourier analysis to estimate a true peak value.

According to a fifth aspect of the present invention, the sampling means samples the reflected wave signal at a frequency lower than twice the frequency of the reflected wave signal. The present invention is characterized in that a return map generated near 0 Hz is cut out and supplied to the analysis means.

According to a sixth aspect of the present invention, in the third aspect of the present invention, the sampling means is configured to set a center frequency of the reflected wave signal to a quarter or four of a sampling frequency.
Means for sampling the reflected wave signal at a sampling frequency to be converted to three-thirds of the frequency, wherein the frequency shifting means multiplies the sampled data sequence by a complex unit multiplier data sequence to obtain the sampling data. The frequency spectrum of the column is shifted.

Here, the complex unit multiplier data sequence is a sequence in which coordinate values of intersections of a unit circle and a real axis / imaginary axis on a complex plane are sequentially arranged in a counterclockwise or clockwise direction.
→ −j → −1 → + j → repetition, −j → −1 → + j →
+ 1 → repetition, −1 → + j → + 1 → −j → repetition, + j → + 1 → −j → −1 → repetition, or +
Repeat 1 → + j → −1 → −j →, + j → −1 → −j
→ + 1 → repetition, −1 → −j → + 1 → + j → repetition, −j → + 1 → + j → −1 → repetition.

According to a seventh aspect of the present invention, in the sixth aspect, each tap is defined as a filter coefficient obtained by multiplying a coefficient of a low-pass filter for limiting a band of sampling data and the complex unit multiplier data sequence. Each time sampling data sequentially input to the FIR filter advances by a decimation rate set to a multiple of 4, the filter coefficient is calculated for each tap output of the FIR filter. A means for multiplying and outputting filter output data is provided.

The invention of claim 8 of the present application is characterized in that the sampling means is means for supplying sampling data consisting of a real number term and an imaginary number term to the analysis means.

According to a ninth aspect of the present invention, there is provided an ultrasonic diagnostic apparatus for diagnosing a human body by irradiating an ultrasonic wave to a human body and receiving a reflected wave of the ultrasonic wave. Sampling means for sampling at a lower frequency than that, and among the aliasing maps generated by the sampling, extracting an aliasing map generated near 0 Hz and outputting as a sampling data stream; and a sampling data stream supplied from the sampling means. Analyzing means for diagnosing the human body by performing analysis.

The invention of claim 10 of the present application is the invention of claim 9
In the invention, a band rejection filter that blocks a frequency band near a carrier frequency of the ultrasonic wave applied to the human body is provided at a stage preceding the sampling unit.

The invention of claim 11 of the present application is based on claim 9
In the invention, the analyzing means comprises a frequency shift means for shifting the center frequency of the reflected wave signal to 0 Hertz by multiplying the sampling data sequence by a discrete complex exponential function, and Means for performing a Fourier analysis using the data decimated at the decimation rate.

The invention of claim 12 of the present application is based on claim 9
In the invention, the analyzing means includes an interpolating means for estimating a true peak value by interpolating a discrete spectrum obtained by Fourier analysis.

The invention of claim 13 of the present application is claim 9
In the invention, the sampling means is means for sampling the reflected wave signal at a sampling frequency for converting a center frequency of the reflected wave signal to a quarter or three-fourths of the sampling frequency; The means is a means for shifting the frequency spectrum of the sampled data sequence by multiplying the sampled data sequence by a complex unit multiplier data sequence.

The invention of claim 14 of the present application is the invention of claim 1
According to the third aspect of the present invention, an FIR filter in which a coefficient obtained by multiplying a coefficient of a low-pass filter for limiting a band of sampling data and a coefficient obtained by multiplying the complex unit multiplier data sequence for each tap is provided as a filter coefficient, and is sequentially input to the FIR filter Each time the sampling data to be advanced by the decimation rate set to a multiple of 4, the FIR
Means for multiplying each tap output of the filter by the filter coefficient to output filter output data.

In the above invention, by performing FFT (Fast Fourier Analysis) on sampling data for a short time, it is possible to measure a flow velocity with a high time resolution,
In addition to being able to accurately measure the pulsating flow velocity,
Reflux can be detected quickly. On the other hand, by performing FFT on sampling data for a long time, analysis with high frequency resolution becomes possible, and the average flow velocity of a pulsating flow or the flow velocity of a steady flow can be measured with high accuracy.

In this case, a high frequency ultrasonic wave of about several MHz is used to reflect the ultrasonic wave with the fluid.
When such a high frequency is sampled, the down-conversion is usually performed using a mixer, but the down-conversion causes problems such as generation of noise and attenuation of the signal level. In order to solve this, according to the present invention, direct sampling is performed at a low frequency as described below, and down-conversion is realized at the same time by using aliasing mapping.

When an A / D change occurs at a sampling frequency less than twice the input signal, aliasing (aliasing: a
However, in the case of the reflected wave of the Doppler velocity measurement device, the signal spectrum is distributed only near the carrier frequency of the transmission signal, and the aliasing occurs by appropriately selecting the sampling frequency and the like. Even if the regions of interest in which the spectra are distributed do not overlap, the original signal can be reproduced. Therefore, by appropriately selecting the bandwidth of the reflected wave signal, the highest frequency of the reflected wave signal, and the sampling frequency, it is possible to prevent the folded images of the attention area from overlapping as shown in FIG. It is possible to directly sample at a low sampling frequency without down-converting a high-frequency reflected wave signal to a low frequency using a mixer. Furthermore, by cutting out the aliasing map around the baseband generated by this sampling, down-conversion of the high-frequency spectrum can be realized at the same time.

Then, as shown in FIG. 1B, the sampling frequency fs at which the undersampling is performed is changed to fs / 4 after the center frequency fc (the carrier frequency of the transmission signal in the case of the reflected wave signal) of the target area is sampled. Alternatively, the frequency is set to be 3 fs / 4. However, fs is set so that fs / 2 is equal to or larger than the bandwidth B of the attention area.

When the reflected wave signal is sampled at such a sampling frequency fs, the frequency spectrum is developed in a positive / negative range across the center frequency fc, but the spectra do not overlap due to aliasing. Also,
At this time, the mapping generated near 0 Hz is not limited to the primary mapping and may be a multi-dimensional mapping such as a secondary mapping. However, since the center frequency fc is known in advance, which mapping is located at which location. This can be predicted, and the most usable mapping around 0 Hz can be used.

When the FFT is performed on sampling data for a long time in order to increase the frequency resolution, the number of sampling points increases and the amount of calculation of the FFT increases exponentially. Focusing on the fact that the frequency band in which the signal is distributed is a part, and by thinning out (decimating) the sampling data, the spectrum is expanded and the frequency resolution is improved without increasing the number of sample points. . In this decimation, it is necessary to shift the center frequency fc of the spectrum to 0 Hz. A specific method for simplifying these will be described below.

As described above, a signal that has been sampled at the sampling frequency fs and has become a discrete-time data string with the center frequency fc is:

[0028]

(Equation 1)

By multiplying the exponential function sequence, the frequency spectrum can be shifted so that the center frequency fc becomes 0 Hz (DC). That is, the frequency spectrum obtained from the DFT transform of the data sequence x (n) is

[0030]

(Equation 2)

The data sequence x
The frequency spectrum Xshift (k) obtained by multiplying (n) by the discrete complex exponential function c (n) is

[0032]

(Equation 3)

The frequency spectrum X (k)
Is shifted along the frequency axis. That is,

[0034]

(Equation 4)

As a result, the entire spectrum can be shifted along the frequency axis so that the center frequency fc of the region of interest in the spectrum becomes zero.

Further, the exponent part of the above c (n) (-jΩ)
_c n) by adding a natural number M to (n +
M), that is, even if the data sequence is multiplied by shifting the discrete complex exponential function by M times,

[0037]

(Equation 5)

As is apparent from the above, the frequency power spectrum is similarly shifted without being affected by this shift.

Then, as described above, the sampling frequency fs and the center frequency fc of the region of interest are fc = fs / 4.
Alternatively, since sampling is performed in such a relationship that fc = 3fs / 4, Ω _c = 2π (fc / fs) = π / 2 or Ω _c = 2π (fc / fs) = 3π / 2. The discrete complex exponential function c (n) is

[0040]

(Equation 6)

## EQU1 ## Here, considering the case of Ω _c = π / 2, for an arbitrary integer value n,

[0042]

(Equation 7)

Thus, it can be seen that only four kinds of values of +1, -j, -1, + j are taken.

Therefore, it is not necessary to actually multiply x (n) and c (n) in order to shift the frequency spectrum.
+1, −j, −1, +
It is only necessary to perform the sign control and the real number imaginary number control in accordance with the case where j is multiplied. That is, when c (n) is a minus sign, only the sign inversion calculation is performed, and when c (n) is a real number, all values of x (n) are processed as real parts,
When c (n) is an imaginary number, all values of x (n) may be processed as imaginary parts. When Ω _c = 3π / 2,
Ω _c = π / 2, the direction is reversed, and +1, + j, −
1, -j. Note that the head of the repetition of + 1 → −j → −1 → + j or + 1 → + j → −1 → −j is +1, +
Any one of j, -1, and -j may be used.

Thus, fc = fs / 2 or fc =
By sampling at a sampling frequency fs of 3 fs / 2, the frequency spectrum can be shifted only by sign control and processing of allocating to real and imaginary numbers based on the sampling number of the sampling data. Since it is not necessary to actually multiply and calculate, the processing can be greatly simplified.

In the sampling data sequence frequency-shifted by this process, only the real part data and the imaginary part data appear alternately. Therefore, in the operation such as the filter operation at the subsequent stage, the real part operation / imaginary number operation is performed. The operation of each part can be reduced by half the amount of the normal operation. That is, the operation of the real part is performed on the real part of the sampling data multiplied by +1 or -1, and the operation of the imaginary part is performed on the imaginary part of the sampling data multiplied by + j or -j. This can be realized by an arithmetic processing unit having a data length that is half the length.

Further, by shifting the center frequency to 0 as described above, the sampling data is thinned (decimated), and as shown in FIG.
The frequency spectrum can be elongated around z (DC). By thinning out the sampling data, long-time sampling data can be handled with a small number of sampling points (data points).

Here, the fact that the spectrum is expanded by the decimation will be described using mathematical expressions. The original discrete sequence data is x (n), and a sequence obtained by replacing every other data with zero except for each D sequence of the sequence is x _p (n), x
Let x _d (n) be a sequence extracted from (n) or x _p (n) for each D, that is, decimated.

These sequences x _d (n), x (n), x _p
The relation of (n) is such that x _d (n) = x (nD) = x _p (nD) holds, and the Fourier transforms are x _d (Ω) and x
(Ω) and x _p (Ω). Using the above relational expression x
Looking for the Fourier transform of _d (n),

[0050]

(Equation 8)

The following holds.

Further, an impulse train p (n) having a period D
Is

[0053]

(Equation 9)

And the Fourier transform P (Ω) of p (n)
Is

[0055]

(Equation 10)

Is as follows. On the other hand, p (n) is given by the following equation:

[0057]

[Equation 11]

This Fourier transform transform is as follows:

[0059]

(Equation 12)

,

[0061]

(Equation 13)

Is established.

From this,

[0064]

[Equation 14]

Where the spectrum is D with respect to the frequency axis.
It can be seen that the magnification was doubled.

Further, in the present invention, 4n (n: a positive integer) is used as the decimation rate, and the FIR filter which is a low-pass filter for preventing overlapping (aliasing) of aliasing spectra generated by the decimation is used. The filter coefficients are previously multiplied by multipliers for the above-mentioned frequency shift, +1, -j, -1, + j or +1, + j, -1, -j. Since the decimation rate is 4n, a filter operation is performed every time the input sampling data sequence advances by 4n,
+1, -j, -1, + j or + repeated in four cycles
Since the same sampling data is always multiplied by the same multiplier in synchronization with the multipliers of 1, + j, -1, and -j, the sampling data must not be multiplied by this multiplier before being input to the FIR filter. However, the frequency spectrum can be simultaneously shifted in the filter operation. Further, as described above, the multipliers are +1 and −
By using j, -1, + j or +1, + j, -1, -j, either the real part or the imaginary number of the sampling data always becomes 0, so the tap becomes 0 (coefficient operation).
Is omitted, the filter length can be reduced to about 1/2.

As a result, the frequency shift, the FIR filter processing, and the decimation processing at the decimation rate 4n can be collectively executed, and the filter length is reduced to （(the amount of filter operation is reduced to １ ／). And the processing can be greatly simplified.

Even if the frequency resolution is improved in this manner, the discrete spectrum is obtained by the FFT.
By estimating the envelope of the continuous spectrum by interpolating the discrete spectrum, it is possible to detect the peak frequency with higher accuracy. As the interpolation method, for example, the Tabei-Ueda method may be used.

On the other hand, when cutting out finite-length sampling data, it is common practice to cut out data by multiplying by a window function such as a Hanning window in order to preserve the frequency spectrum. Exponential function sequence c for shifting the frequency spectrum
By multiplying (n) in advance and multiplying it by the sampled data sequence x (n), it is possible to simultaneously cut out sampling data and shift the frequency spectrum by the window function.

Further, also in this case, as shown in FIG. 12, similarly to the shift of the frequency spectrum, the sampling frequency fs and the center frequency fc of the region of interest are fc
= Fs / 2 or fc = 3fs / 2, sampling is performed by adding +1,
-J, -1, + j or +1, + j, -1, -j are simply multiplied by the data sequence multiplied by the data sequence.
Simply, sampling data can be cut out, the frequency spectrum can be shifted, and if necessary, the spectrum can be inverted.

[0071]

FIG. 2 is a block diagram of a Doppler flowmeter according to an embodiment of the present invention, FIG. 3 is an external view of a transducer section of the Doppler flowmeter, and FIG. 4 is a view of a receiving section of the Doppler flowmeter. It is a block diagram. This flow meter is applied to a blood flow meter that measures the blood flow of a heart-lung machine. An artificial heart-lung machine may supply a constant amount of blood in a constant flow, or may supply the blood in a pulsatile flow at a constant cycle like an actual arterial flow. It is necessary to detect the flow rate. In addition, blood may flow backward due to abnormalities in blood clots or joints during the operation, but this is a situation that affects the life of the patient, and therefore it is necessary to detect this immediately. This flow meter meets such a requirement.

The Doppler flow meter comprises a transmitting section 1, a receiving section 2 and a control section 3. The transmitting unit 1 includes a tube 6
It has a transmitting transducer 4 for transmitting ultrasonic waves to a fluid (blood) flowing in the inside. Also, the receiving unit 2
Has a receiving transducer 5 for receiving a reflected wave signal from the fluid. These transducers 4 and 5 are fixed at an angle α to the tube 6 by jigs 7 and 8, respectively. The jigs 7 and 8 are filled with a gel-like substance so as not to attenuate the ultrasonic waves. Since the Doppler velocimeter reflects ultrasonic waves from particles in a fluid and measures the flow velocity by Doppler shift of the reflected wave signal, a high frequency of about 1 MHz is used for the ultrasonic waves.

In FIG. 4, the receiving transducer 5
Are connected to the high-frequency amplifier 10. The reflected wave signal received by the receiving transducer 5 is amplified by the high-frequency amplifier 10. The amplified reflected wave signal is input to the band pass filter 11. Bandpass filter 1
1 passes only the frequency band (region of interest) in which the spectrum of the reflected wave signal is distributed among the reflected wave signals, and removes noise in other frequency regions. The removal of noise outside the region of interest is indispensable to prevent destruction of the spectrum due to aliasing noise in the subsequent undersampling. In the case of a Doppler flow meter, the above-mentioned region of interest (that is, the transmission band of the bandpass filter 11) is 2 kHz.
z is sufficient. Details of this bandwidth will be described later. The reflected wave signal whose band is limited by the band-pass filter 11 is amplified to a level that can be sampled by the amplifier 12 and input to the A / D converter 13.

The A / D converter 13 samples this reflected wave signal at a sampling frequency fs that is at least twice the bandwidth Δf of the region of interest. This sampling frequency fs
May be lower than the carrier frequency fc. The details of the relationship between the sampling frequency fs, the carrier frequency fc, and the bandwidth Δf (= B) will be described later.

Digital data (sampling data) undersampled by the A / D converter 13
Is input to the DSP 14. D
The SP performs an FFT analysis on the sampling data to determine a Doppler shift frequency, and determines a blood flow velocity and flow rate based on the Doppler shift frequency. As the A / D converter 13, a wideband high-speed A / D having a built-in sample hold circuit is used.
A converter (sampling A / D converter) may be used.

As described above, by using undersampling at a frequency of about 1/250 of the input signal frequency, it is possible to reduce the number of front-end analog circuits and reduce signal distortion.
1 / f generated by IF sampling by mixer
Low noise. Loss of signal by mixer (usually -3d
B to about -7 dB). There is no induction of DC offset. There is no fear that the local frequency component leaks to other circuits. The transfer rate of sample data can be reduced (the bandwidth can be narrowed). Eliminates the need for a large data memory. Data can be directly processed by the DSP, which facilitates real-time processing. And other advantages.

An example of the set value of each part when measuring the flow rate of the fluid (blood) with the configuration shown in FIG. 3 will be described. First, when a resin tube having an outer diameter of about 1 cm is used as the tube 6, the inner diameter is 0.953 cm, and the cross-sectional area is 0.713 cm.
² The flow rate is calculated by multiplying this cross-sectional area by the flow velocity obtained from the Doppler shift frequency. For example, when measuring a flow rate of 100 ml / min, the flow rate is 2.34 cm / sec. The angle α of the transducers 4 and 5 is 45 °, and the transmission frequency is about 1M.
In Hz, the Doppler shift frequency corresponding to the above flow rate is about 22 Hertz. Under these conditions, 5 ml / sec
In order to obtain the speed resolution, a frequency resolution of approximately 1 Hertz is required.

Assume that the measurement range is 5 ml / min to 100
Assuming 0 ml / min and a transmission frequency of 1.0 MHz, the maximum Doppler shift frequency is ± 220 Hertz.

Therefore, in this embodiment, as shown in FIG. 5, a band of ± 2 kHz from the transmission frequency is cut out as a region of interest, and the transmission frequency fc (the center of the region of interest) is selected so that the undersampling process can be easily performed. frequency)
Was set to 998 kHz. That is, the maximum frequency Fmax of the reflected wave signal is allowed up to 1000 kHz.

Therefore, in this embodiment, under-sampling is performed with the transmission carrier frequency set to 998 kHz and the sampling frequency set to 8 kHz under the above conditions. In this undersampling, as shown in FIG. 5B, N = Fmax / B = 1000/4 = 250,
Since N is an even number, the inverted spectrum is
Appears between 4 kHz.

The DSP 14 performs an FFT on the spectrum appearing between 0 and 4 kHz and analyzes the Doppler shift frequency. The DSP 14 operates according to a microprogram. FIG. 6 shows a flowchart of a schematic operation of the DSP 14. This operation is repeatedly executed at regular intervals (about one second). First, 1024 sampling points are cut out and FFT is performed to obtain a spectrum distribution (s
1). Since the sampling frequency fs = 8 kHz,
The 1024 sampling points correspond to a sampling time of 0.128 seconds, and the frequency resolution is 8000/102.
4 = 7.8 Hz. It is determined whether this flow is a steady flow or a pulsating flow by comparing with the past analysis results, such as the last time and the last two times (s2). At this time, whether or not the flow is flowing backward is also determined at the same time. In the case of a pulsatile flow, since the time variation of the flow velocity is large, accurate analysis cannot be performed even if analysis is performed with a longer sampling time. In this case, the process proceeds to s5, and the actual peak is estimated by interpolating the discrete spectrum obtained by the analysis (s5). As this estimation method, the Tabei-Ueda method or the like may be applied. If this interpolation method is used, the frequency resolution can be increased by a factor of 10 to 100 (in the absence of noise), so that even an FFT with 1024 sampling points can achieve 1H.
A frequency resolution of z or less can be realized. Also,
It is also possible to obtain the average Doppler frequency from the center of the spectrum or the like.

On the other hand, if it is determined in s2 that the flow is a steady flow, a decimation process of thinning out the sampling points to 1/8 is performed (s3). In order to perform this decimation process, it is necessary to simultaneously perform a demodulation process for shifting the center frequency of the region of interest from 2 kHz to 0 Hz and a low-pass filter process for removing noise components other than the frequency band in which the spectrum is distributed. There is. In the DSP 14, these processes are collectively performed by an FIR filter. Details of these will be described later. After this decimation processing, the FFT of 1024 points is executed, and a spectrum distribution is obtained with a frequency resolution of 0.98 Hz (s4). In other words, even if there are 1024 points, the sampling time is extended to 1.024 seconds due to the 8 times decimation, so that the above frequency resolution can be obtained.

Then, the spectrum distribution is further interpolated (s5), and a Doppler shift frequency is obtained with a frequency resolution of 0.1 Hz or less.

After obtaining the Doppler shift frequency, the flow velocity and the flow rate of the fluid (blood) are calculated (s6, s7).

On the other hand, if the frequency spectrum distribution is clearly negative in s2, it is determined that the current is flowing backward, and an alarm is issued (s8). The alarm may be an acoustic alarm or a display alarm, and these alarms may be used in combination.

Here, the decimation process will be described in detail. If an FFT with a sampling frequency of about 1 second is performed at a sampling frequency of 8 kHz, about 800
It is necessary to perform FFT on zero sampling points, which is not practical because the amount of calculation is enormous. But,
Since it is known that the spectrum distribution of the reflected wave signal is narrower than the sampling frequency, only the region where the spectrum is actually distributed is enlarged (zoomed) while maintaining the same frequency resolution as the 8192-point FFT. Perform FFT with fewer points.

Decimation
When the number of FFT points is reduced by, the spectrum is expanded around the zero frequency. For this reason, it can be seen that if 2.0 kHz corresponding to the zero frequency of the Doppler shift is translated in parallel to the zero frequency of the spectrum (frequency shift) and decimated, the Doppler shift component is enlarged. On the other hand, if a frequency shift of 6 kHz is performed, a mapping that is not inverted from the original spectrum can be shifted to zero frequency, and the inversion of the spectrum can be corrected by the shift. Even if the frequency spectrum is inverted, it can be easily corrected by inverting ± of the frequency in the operation of the DSP 14.

FIG. 7 shows a functional block of processing for zooming the frequency spectrum in the DSP 14. FIG. 8 shows how the spectrum is converted by this functional block.

In FIG. 7, the mixer 20 multiplies the sampling data sequence by the complex function sequence of [Equation 1].
As a result, the center frequency of the attention area is shifted to 0 Hz as shown in FIG. Then, a portion other than the spectrum distribution region is cut from this data string by the low-pass filter 21. Thus, the inverted spectrum is cut as shown in FIG. 8B, and the spectrum distribution becomes similar to the spectrum distribution represented by the complex number. Next, the data sequence sampled at the sampling frequency fs by the decimation process 22 is x (n, data is extracted from the sampled data every eight data items. That is, 1024 data items are extracted from 8,192 data items. 8C, the spectrum is expanded by a factor of 8. The FFT 23 is performed using the expanded data.

Here, actually, the DSP 14 does not sequentially perform demodulation, filtering, and decimation as shown in FIG. 7, but performs them collectively using an FIR filter. Hereinafter, the simplified arithmetic processing method will be described.

A region of interest of the frequency spectrum obtained from the data sequence x (n) A / D converted by the sampling frequency fs (frequency band where the spectrum is developed)
Decimation processing is performed to enlarge.

To enable this decimation process, the center frequency (2 kHz) of the region of interest is shifted to zero frequency (DC). That is, the spectrum is shifted in parallel with respect to the frequency axis. This operation is performed by multiplying each data of the sampling data sequence x (n) by the discrete multiple exponential function sequence c (n) as described above.

A signal sampled at a sampling frequency of 8 kHz and converted to a discrete-time data string having a center frequency of 2 kHz is multiplied by an exponential function string of [Formula 7] so that the center frequency 2 kHz becomes 0 (DC). To shift the frequency spectrum. This discrete complex exponential function sequence is a complex unit multiplier data sequence (+1, -j, -1, + j
Or only four types of values (sequences starting from arbitrary values of +1, + j, −1, and −j), there is no need to perform multiplication in actual processing, and when c (n) is a minus sign, Performs only the sign inversion calculation, and if c (n) is a real number, x
If all the values of (n) are real parts, and if c (n) is an imaginary part, only the values of x (n) are all imaginary parts.

FIG. 9 is a diagram for explaining a method of transferring data from the A / D data buffer x (n) to the complex number buffer X (n). The A / D data buffer x (n) is a buffer for storing an undersampled sampling data (A / D data) sequence, and is a complex number buffer X (n).
Is a buffer that stores a sampling data string including an imaginary part whose frequency spectrum has been shifted to the center frequency = 0.

In this figure, x (0) is X
The data is transcribed to the real part of (0), and 0 is written to the imaginary part of X (0). After the sign of x (1) is inverted, it is transcribed to the imaginary part of X (1), and 0 is written to the real part of X (1). After the sign of x (2) is inverted, it is transcribed to the real part of X (2), and 0 is written to the imaginary part of X (2). x (3) is X as it is
The imaginary part of (3) is transcribed, and 0 is written in the real part of X (3). As described above, the operation result of the discrete complex exponential function is converted into a complex unit multiplier data string (+1, -j, -1, +
This frequency shift can be performed only by sequentially repeating the sign inversion and the transcription in accordance with the value of j or any sequence of +1, + j, -1, -j), and the exponential function is actually multiplied. There is no need to perform calculations, and processing can be greatly simplified.

Further, since one of the real parts Real (n) and Imaginary (n) of the complex number buffer X (n) is always 0, 0 is stored if the side which becomes 0 based on the above rule is known. By omitting the buffer, the storage area of the buffer can be substantially halved.

Although the processing has been simplified in this way,
Performing the frequency spectrum shift calculation of the total number N of the A / D conversion data before decimation every time the FFT processing is performed in the continuous wave Doppler mode takes time because of the large number of data points. For example, FFT score 10
If N is 24 and the decimation rate is 8, N is 8192.

An FIR filter is generally used as a filter for limiting a spectrum band.
Even if all the sampling data x (n) are subjected to the FIR filtering processing, only 1 / D data selected by the decimation of the decimation rate D is used, which is wasteful (x (n in this description). ) Represents the frequency-shifted X (n).) Therefore, the filtering process is not performed for each data, the data discarded by the decimation is skipped, and the filtering process is performed for each D-th data. FI to skip this discarded data
FIG. 10 is a block diagram showing the R filter processing. In this figure, the FIR filter functions only when the data x (Dn) selected by the decimation is input among the sampling data x (n) which is input one by one continuously from the left end input terminal, The operation unit of each tap outputs data multiplied by the filter coefficient.
By adding these data, the filtered x
(Dn) data can be obtained. When data to be discarded by the decimation is input from the input terminal, only the upper delay circuit (shift register) functions and the operation of the arithmetic unit of each tap only needs to be stopped. Thus, filtering and decimation can be performed simultaneously.

Here, the FIR filter requires two systems corresponding to the real part and the imaginary part of x (n).
Each term in (n) has a significant value only in the real part or imaginary part, and the other is 0 (see FIG. 9). Therefore, D
Is a multiple of 2, when outputting a value from the FIR filter, a filter coefficient to which a significant real part and a significant imaginary part are given, and a filter coefficient to which a real part of 0 and an imaginary part of 0 are given Is decided. Therefore, D
Is set to be a multiple of 2, it is possible to omit a filter coefficient to which 0 is always given and its calculation processing.

Further, as described above, the frequency shift is
This is a process of transferring the input sampling data to a real part or an imaginary part and controlling the sign thereof. For every four data, (+1, -j, -1, + j) or (+1,
+ J, -1, -j) are repeatedly performed.
Therefore, by incorporating the processing for extracting the real or imaginary part and the sign as a filter coefficient, frequency shift, filtering, and decimation can be simultaneously performed.

FIG. 11 shows frequency shift, filtering,
FIG. 4 shows a block diagram of an FFT pre-processing process which is a batch process of decimation. This figure shows a case where the filter length is odd and (L + 1) / 2 is even.
As a condition of this processing, the decimation rate D is required to be a multiple of 4 which is a repetition step of the frequency shift processing. When data x (Dn) selected at a rate of one per D by decimation is input,
Data x that is inverted from this x (Dn) by the filter length
For data up to (Dn- (L-1)), filter coefficients h (0), h (1), -h (2), -h (3),.
.., −h (L−1), and x (Dn) · h (0) +
x (Dn−2) · (−h (2)) +... + x (Dn (L
-1)) · (-h (L-1)) is output as the real part of the filter output. And x (Dn−1) · h (1) + x
(Dn−3) · (−h (3)) +... + X (Dn (L−
1)) Output h (L-1) as the imaginary part of the filter output. Thereby, frequency shift and filtering can be performed simultaneously. Then, when the A / D data advances by D and the next selection data x (D (n + 1)) is input, the same processing is performed. Decimation is performed by the processing for each D. As described above, the A / D data is advanced one data at a time, and when data selected by decimation is input, filtering and frequency shifting are performed at a stretch, thereby simplifying the processing and usinglessly. Since no arithmetic processing is performed, the time required for the arithmetic operation can be greatly reduced.

As described above, this example shows a case where the filter length is an odd number and (L + 1) / 2 is an even number. However, even when (L + 1) / 2 is an odd number, The same processing can be performed even when the filter length is even and L / 2 is odd / even.

In this embodiment, the Doppler velocimeter for measuring the velocity of a fluid flowing in a tube such as a blood flow has been described. , Can be applied to anything. Further, since negative frequency detection is also possible, it is possible to measure the flow velocity in both directions.

In the above embodiment, the processing is performed only with the real-valued data until the frequency shift processing of the DSP 14 is performed. However, the reflected wave signal is separated into complex signals (I signal, Q signal) in advance. You may do so. The stage at which separation is performed is arbitrary. As shown in FIG. 13A, the local signal to be multiplied at the time of down-conversion by the mixer may be separated by shifting the phase by π / 2. Alternatively, the sampling may be performed as shown in FIG. 3B and then separated by a filter (a band-pass filter and a Hilbert filter) in the DSP 14.

As described above, I and Q are obtained by using two mixers.
Undersampling can also be applied to a circuit that performs quadrature detection that separates signals.

Further, if the separation of the I and Q signals from the undersampled signal of one channel is realized by a digital filter or a DSP, there is no mismatch between I and Q. There is no uneven output amplitude. The phase shift due to the fluctuation of the π / 2 phase shift between I and Q can be ignored. Two A
It can be realized without considering synchronization between the / D converters at all. This can be realized with only one A / D converter.
There is no amplitude mismatch between I and Q, and no phase shift occurs. And other advantages.

FIG. 14 is a schematic block diagram of an ultrasonic diagnostic apparatus according to a second embodiment of the present application. The ultrasonic diagnostic apparatus transmits an ultrasonic carrier signal from an ultrasonic transducer element provided on a probe, and receives a signal obtained by reflecting the ultrasonic carrier signal on each part of a human body by the ultrasonic transducer element. By analyzing the reflected wave signal, the state inside the human body is output as an image or an acoustic signal. The transmitted ultrasonic carrier signal is a signal of several megahertz. This figure shows only the configuration of the receiving unit.

A high-frequency amplifier 32 and a delay 33 are connected to each ultrasonic transducer element 31. The high frequency amplifier 32 amplifies the received reflected wave signal. The delay 33 forms a beam by aligning the phase of the signal received by each ultrasonic transducer element 31. The reflected wave signal output from the delay 33 corresponding to each ultrasonic transducer element 31 is added by the adder 34. The added reflected wave signal is a signal having directivity in a predetermined direction with respect to the probe. This reflected wave signal has an extremely high peak near the carrier frequency as shown in FIG. 15, and the reflected wave signal spectrum from a blood flow or the like to be diagnosed is about 1.
There is a level difference of 00 dB. Therefore, a signal component near the carrier frequency is removed from this signal by the notch filter 35. Thereafter, the data is sampled by the A / D converter 35 and converted into a digital data string. The A / D converter 35 also performs undersampling as in the first embodiment. When operating under the same conditions as in the first embodiment, as shown in FIG.
The carrier frequency may be set to 998 kHz and sampling may be performed at 8 kHz. By doing so, 0-4k
An inverted folding map appears in the frequency band of Hz, and this can be used as sampling data. The sampled data is supplied to a bandpass filter (BP
F) 37 and the Hilbert filter 38 separate the data into real value data I and imaginary value data Q. The separated complex data is input to the image processing unit 40 and the DSP 50.

The image processing section 40 is a processing section for processing a scanned signal into a two-dimensional image and displaying the processed signal on the monitor 44. The B-mode image 44a, which is a normal scan image, and the pulsation state of a plane being scanned. A color flow image 44d or the like that expresses by color is created. The created image data is input to the monitor 44 via the scan converter 41 → the color processor 42 → the D / A converter 43.

On the other hand, the DSP 50 performs the same processing as that of the DSP 14. That is, the input I and Q data are shifted to the center frequency of 0 Hz by demodulation 51, the band is limited by the LPF 52, the number of sample points is reduced by the decimation 53, and the frequency resolution is increased to execute the FFT 54. .
Thus, only the frequency band that has been Doppler shifted by the blood flow, which is the region of interest, can be expanded to accurately determine the blood flow state. In this case, FIGS.
2 can be applied. Further, by interpolation processing, 1 of the discrete spectrum by FFT is obtained.
A frequency resolution of 0 to 100 times can be obtained. This data is input to the scan converter 41. This data is displayed on the monitor 44 as an M-mode Doppler image 44b or a spectral Doppler image 44c. Each of the images 44a to 44d can be switched to one of the modes by switching the mode.
Are displayed on the monitor 44. The DSP 50 can be used in both the pulse Doppler mode and the continuous wave Doppler mode. On the other hand, the decimated data is input to the audio processing unit. The audio processing unit 55 cuts out only the audible band from the positive frequency component, that is, the data on the near side, and outputs it as an audio signal of the right (left) channel, and outputs only the audible band from the negative frequency component, that is, the data on the far side. Cut out and output as audio signal of left (right) channel. The output data is converted into an analog audio signal by a D / A converter 56 and output to a stereo speaker or stereo headphones 58 via an amplifier 57. The person making the diagnosis can judge the intensity and direction of the pulsation based on the sound quality of the audio signal and the left and right localization of the sound, and can make an auditory diagnosis in addition to the visual sense.

As described above, in this ultrasonic diagnostic apparatus, since the received reflected wave signal is directly sampled (A / D converted) without being down-converted by the mixer, noise is mixed or the signal is attenuated. Nothing. A /
Since the I and Q separations are performed immediately after the D conversion, this can be directly converted into an audio signal, and the circuit can be simplified.

When an A / D converter with a high speed and a wide band and a high dynamic range (multi-bit) is used, even a weak target signal (Doppler shift signal due to blood flow) can be A / D converted uniformly. Since it is possible, the reflected wave signal having a high peak in the carrier frequency may be subjected to A / D conversion without passing through the notch filter.

[0113]

According to the first aspect of the present invention, the pulsating flow and the backflow can be measured with high accuracy by obtaining the flow velocity in a short sampling time, and when the flow is determined to be a steady flow, the pulsating flow and the backflow can be measured long. Since the flow velocity is determined by the sampling time, both time resolution and velocity resolution can be achieved.

According to the second aspect of the present invention, a warning is issued when it is determined that the flow velocity is negative and the flow is reverse, so that it is possible to quickly respond when the flow velocity measurement apparatus is used in a heart-lung machine or the like. Become.

According to the third aspect of the present invention, by analyzing the long sampling time by reducing the number of sampling points by decimation, it is possible to improve the speed resolution without increasing the amount of calculation.

According to the fourth aspect of the present invention, by estimating the true peak value by interpolation, it is possible to realize a frequency resolution of 10 to 100 times that in the case where only the discrete spectrum is used.

According to the fifth aspect of the present invention, 2 of the reflected wave signal
Sampling (undersampling) at a sampling frequency lower than twice eliminates the need to downconvert by multiplying the local signal by the mixer, prevents noise contamination and signal attenuation, and directs aliasing mapping near the baseband. Can be used.

According to the sixth aspect of the present invention, the frequency shift for the decimation can be performed by a simple multiplication of 1, j, -1, and -j, so that the calculation amount is reduced and the processing speed is improved. can do.

According to the seventh aspect of the present invention, since the shift of the frequency spectrum, the restriction of the frequency band, and the decimation can be performed in a lump by incorporating the filter coefficients of the FIR filter, the frequency spectrum can be obtained. The signal processing steps can be simplified.

According to the eighth aspect of the present invention, by supplying complex number data consisting of a real value and an imaginary value to the analyzing means, it is possible to process the data as baseband data including a spectrum on the negative side.

According to the ninth aspect of the present invention, 2 of the reflected wave signal
Sampling (undersampling) at a sampling frequency lower than twice eliminates the need to downconvert by multiplying the local signal by the mixer, prevents noise contamination and signal attenuation, and directs aliasing mapping near the baseband. Can be used.

According to the tenth aspect of the present invention, by removing the carrier component by the band rejection filter, a weak reflected signal can be digitized at a large level, and a highly accurate analysis can be performed.

According to the eleventh aspect, by analyzing a long sampling time by reducing the number of sampling points by decimation, it is possible to improve the speed resolution without increasing the calculation amount.

According to the twelfth aspect, by estimating the true peak value by interpolation, it is possible to realize a frequency resolution of 10 to 100 times that in the case of using only discrete spectra.

According to the thirteenth aspect, since the frequency shift for the decimation can be performed by a simple multiplication of 1, j, -1, and -j, the amount of calculation is reduced and the processing speed is improved. can do.

According to the fourteenth aspect of the present invention, the shift of the frequency spectrum, the limitation of the frequency band, and the decimation can be performed in a lump by incorporating the filter coefficients of the FIR filter into a single process. The signal processing steps can be simplified.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a signal processing method according to the present invention.

FIG. 2 is a schematic configuration diagram of a Doppler velocimeter which is an embodiment of the present invention.

FIG. 3 is a structural diagram around a transducer of the Doppler velocimeter.

FIG. 4 is a diagram showing a configuration of a receiving unit of the Doppler velocimeter.

FIG. 5 is a diagram illustrating a method of processing a reflected wave signal in the Doppler velocimeter.

FIG. 6 is a flowchart showing the operation of the DSP of the Doppler velocimeter.

FIG. 7 is a functional block diagram of the DSP.

FIG. 8 is a diagram illustrating conversion of a spectrum of sampling data processed by the DSP.

FIG. 9 is a block diagram showing a frequency shift process in the DSP.

FIG. 10 is a block diagram showing an FFT pre-processing for simultaneously performing a frequency shift, an FIR filter, and a decimation in the DSP.

FIG. 11 is a block diagram illustrating a processing method for an input signal in the FFT preprocessing.

FIG. 12 is a diagram showing processing blocks when multiplication of a window function and frequency shift processing are performed simultaneously.

FIG. 13 is a diagram illustrating a configuration of a receiving unit when a reflected wave signal is separated into I and Q.

FIG. 14 is a schematic block diagram of an ultrasonic diagnostic apparatus according to another embodiment of the present invention.

FIG. 15 is a diagram showing a frequency spectrum of a reflected wave signal received by the ultrasonic diagnostic apparatus.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Transmission part, 2 ... Reception part, 3 ... Control part, 4 ... Transmission transducer, 5 ... Reception transducer, 6 ... Tube (flowing blood), 7, 8 ... Jig (supporting a transducer), 10 ... High frequency amplifier, 11 ... Band pass filter, 13 ... A / D converter, 14 ... DSP,
31 ultrasonic transducer element, 35 notch filter, 36 A / D converter, 37 bandpass filter, 38 Hilbert filter, 50 DSP

Claims (14)

[Claims] An ultrasonic transducer for transmission and a ultrasonic transducer for reception installed at an angle with respect to a fluid flowing in a predetermined direction, and an ultrasonic signal transmitted by the ultrasonic transducer for transmission is transmitted by the fluid. Means for receiving the reflected wave signal by the receiving ultrasonic transducer and measuring the flow velocity of the fluid based on the Doppler shift amount of the reflected wave signal, wherein the sampling for sampling the reflected wave signal is performed. Means, and the flow velocity is obtained by Fourier analysis using short-time sampling data, and when it is determined that the flow velocity is substantially constant based on the result of the analysis, by Fourier analysis using longer-time sampling data. A flow rate measuring device, comprising: an analysis unit for obtaining the flow velocity. 2. The flow velocity measurement according to claim 1, further comprising: a means for issuing a warning as a reverse flow when the flow velocity obtained by the analysis means becomes a negative value in the predetermined direction. apparatus. 3. A frequency shifter for shifting the center frequency of the reflected wave signal to 0 Hertz by multiplying the sampling data sequence by a discrete complex exponential function, and analyzing the shifted sampling data. The flow velocity measuring apparatus according to claim 1, further comprising: means for performing a Fourier analysis using data thinned out at a predetermined decimation rate. 4. The flow velocity measuring apparatus according to claim 1, wherein said analysis means includes interpolation means for interpolating a discrete spectrum obtained by Fourier analysis to estimate a true peak value. 5. The sampling means samples the reflected wave signal at a frequency lower than twice the frequency of the reflected wave signal, and among the aliased maps generated by the sampling, an aliased map generated near 0 Hz. 2. The flow velocity measuring device according to claim 1, wherein the flow rate measuring device is a means for cutting out and supplying the analysis means to the analysis means. 6. The sampling means is a means for sampling the reflected wave signal at a sampling frequency for converting a center frequency of the reflected wave signal to a quarter or three-fourths of a sampling frequency. The frequency shift means repeats the sampled data sequence starting from any value of the sequence of +1, -j, -1, + j or any value of the sequence of +1, + j, -1, -j 4. The flow velocity measuring apparatus according to claim 3, wherein the means is a means for shifting the frequency spectrum of the sampling data sequence by multiplying a multiplier data sequence consisting of repetitions starting from (hereinafter referred to as "complex unit multiplier data sequence"). 7. An FIR filter in which each tap is provided with a coefficient obtained by multiplying a coefficient of a low-pass filter for limiting a band of sampling data and the complex unit multiplier data sequence as a filter coefficient, is sequentially input to the FIR filter. Means for multiplying each tap output of the FIR filter by the filter coefficient and outputting filter output data every time the sampling data advances by a decimation rate set to a multiple of four. The flow velocity measuring device according to claim 6, wherein: 8. The flow velocity measuring apparatus according to claim 1, wherein said sampling means is means for supplying sampling data comprising a real number term and an imaginary number term to said analysis means. 9. An ultrasonic diagnostic apparatus for diagnosing a human body by irradiating an ultrasonic wave to a human body and receiving a reflected wave thereof, wherein the received reflected wave signal is sampled at a frequency lower than twice the frequency, Sampling means for cutting out a return map generated near 0 Hertz out of the return maps generated by the sampling and outputting it as a sampling data sequence, and diagnosing the human body by analyzing the sampling data sequence supplied from the sampling means. An ultrasonic diagnostic apparatus comprising: an analyzing unit that performs analysis. 10. The ultrasonic diagnostic apparatus according to claim 9, wherein a band cut filter for cutting off a frequency band near a carrier frequency of the ultrasonic wave applied to the human body is provided at a stage preceding the sampling unit. . 11. A frequency shifter for shifting the center frequency of the reflected wave signal to 0 Hertz by multiplying the sampling data sequence by a discrete complex exponential function, and analyzing the shifted sampling data. The ultrasonic diagnostic apparatus according to claim 9, further comprising: means for performing a Fourier analysis using data thinned out at a predetermined decimation rate. 12. The ultrasonic diagnostic apparatus according to claim 9, wherein said analysis means includes interpolation means for interpolating a discrete spectrum obtained by Fourier analysis and estimating a true peak value. 13. The sampling means is a means for sampling the reflected wave signal at a sampling frequency for converting a center frequency of the reflected wave signal to a quarter or three quarters of a sampling frequency. 10. The ultrasonic diagnostic apparatus according to claim 9, wherein the frequency shifting means shifts the frequency spectrum of the sampled data sequence by multiplying the sampled data sequence by a complex unit multiplier data sequence. 14. An FIR filter in which each tap is provided with a coefficient of a low-pass filter for limiting a band of sampling data and a coefficient obtained by multiplying the complex unit multiplier data string as a filter coefficient, and sequentially inputted to the FIR filter. Means for multiplying each tap output of the FIR filter by the filter coefficient and outputting filter output data each time the sampling data to be advanced by a decimation rate set to a multiple of 4 The ultrasonic diagnostic apparatus according to claim 13, wherein: