JPH0323051B2 - - Google Patents

Description

[Detailed description of the invention] (Industrial application field) The present invention relates to an ultrasonic Doppler signal analyzer.

(Prior Art) An ultrasonic Doppler signal analyzer is a medical diagnostic device that uses the Doppler effect regarding acoustic signals, and it irradiates a subject with ultrasound and collects reflected Doppler signals (i signal, q signal) obtained as a result. Frequency analysis is performed on each signal using a frequency analyzer such as a fast Fourier analyzer (FFT), and the resulting spectrum image is monitored on a display device such as a CRT to obtain various information. . Here, the i signal is a reflected Doppler signal that is in phase with the reference carrier, and q is a reflected Doppler signal that is orthogonal to the reference carrier.
According to the ultrasonic Doppler signal analyzer, it is possible to obtain information regarding a moving subject, such as the heart or blood circulation system, for example.

FIG. 4 is a block diagram showing a conventional example of an ultrasonic Doppler signal analyzer. A Doppler signal obtained from the subject enters a Doppler filter 1 and unnecessary frequency components are removed. The output of the Doppler filter 1 enters a frequency analyzer (for example, FFT) 2 where frequency analysis is performed at high speed, and the analysis results are displayed as a spectrum image on the CRT 3. The operator is
By observing the spectrum image displayed on this CRT 3, or by using the Doppler filter 1
Various information can be obtained by listening to the signal after passing through the buffer amplifier 4 as an audio signal through the speaker 5. The pass frequency band, gain, etc. of the Doppler filter 1 are determined by the operator operating the operating section 6.

In this type of device, it is desirable that the frequency characteristics of the Doppler signal path input to the FFT 2 be essentially flat in a desired region. If the gains are not equal at each frequency, the FFT
This is because the analysis results in step 2 are not desirable because they do not accurately reflect reality. However, in general, when performing frequency analysis of Doppler signals, if too many slow-moving components with strong signal levels coexist, such as those generated in soft tissue other than blood flow, the signal level will be lower than this. Components that are much smaller and have different Doppler shift widths (that is, different radial velocities) conflict with the dynamic range of the hardware used and are masked, resulting in the inconvenience that they cannot be correctly recognized and analyzed. For this reason, the Doppler filter 1 includes a powerful low-cut filter for removing fixed reflections (clutter), regardless of whether it is a CW Doppler or a pulsed Doppler, and cuts out the peak components in the low frequency range described above. However, other than the low-cut characteristics, the transmission characteristics are essentially flat.

(Problem to be Solved by the Invention) Generally, when trying to obtain the flow velocity distribution of a test fluid by frequency analysis of a Doppler signal, as a general rule, the frequency analyzer has its own frequency characteristics or weighting on the frequency axis. Therefore, the frequency characteristics of the signal transmission path up to the input to the frequency analyzer should be completely flat, and this is considered to be the only way to know the correct flow velocity distribution. Ta. However, as mentioned above, there are limitations to the dynamic cleanliness of the analyzer.
When two signals with a level ratio exceeding this arrive, when the level or gain is adjusted to the larger one, the smaller one is masked and drowned out. The dynamics of an analyzer is controlled by the amount of hardware that is mobilized, and therefore cannot be increased freely in practice. Therefore, as mentioned above, it is necessary to remove at least the high-level fixed reflection component (clutter) around the zero Doppler shift, but in addition, it approaches the limit of the dynamic range of the analyzer. In order to correctly analyze signals with such signal components, it is necessary to compress this level ratio through some kind of processing in advance, or in other words, to smooth out the peaks and valleys of the frequency spectrum. Since this spectrum flattening operation must be performed while preserving the linearity of the system, some kind of variable frequency characteristic filter or equalizer is eventually used. However, if such preliminary flattening was done carelessly,
It becomes impossible to accurately measure the distribution of signal energy for each frequency contained in the original signal, that is, the distribution of fluid volume for each flow velocity. However, such a method is very useful, at least in order to avoid overlooking signal components regardless of their level.

The present invention has been made in view of the above circumstances, and when flattening the frequency spectrum in advance as described above, the amount of correction is correctly managed and understood, and the flattened signal is By inversely correcting the frequency spectrum data obtained as a result of analysis using this preliminary spectrum flattening data and outputting it, it is intended to obtain an analysis result that correctly represents the component distribution of the original signal again. It is.

However, another object of the present invention is to use an inexpensive frequency analyzer with coarse resolution on the premise that such frequency spectrum flattening pre-processing and post-correction of output data are performed. It is about urging. In other words, it is essentially a procedure that is very similar to Fourier transform and is significantly simpler than Fourier transform, but it had problems with its dynamic range and sidelobe characteristics, and it was thought that it could not be used as an alternative. The object of the present invention is to use the Walsh transform or the like for the purpose instead of the Fourier transform.

(Means for Solving the Problems) A first invention for solving the above-mentioned problems extracts Doppler components from reflected waves from a subject, and then extracts baseband Doppler components separated in each direction. Doppler signal processing means for obtaining a signal, frequency flattening means for independently flattening the frequency spectrum of each obtained baseband Doppler signal in advance, and frequency analysis means for frequency-analyzing each frequency-flattened Doppler signal. and a correction means for inversely correcting each frequency spectrum data obtained by frequency analysis with the correction amount used for flattening by the frequency flattening means, and a correction means for inversely correcting each frequency spectrum data obtained by frequency analysis, and spectral data in each direction of the Doppler signal obtained by the inverse correction. The present invention is characterized by an ultrasonic Doppler signal analysis device comprising output means for displaying or recording.

The second invention provides a Doppler signal processing means for extracting a Doppler component from a reflected wave from a subject to obtain a baseband Doppler signal, and the obtained i (in-phase) and q (quadrature) baseband Doppler signals. a frequency flattening means for flattening the frequency spectrum of the frequency spectrum in advance under common conditions; a frequency analysis means for frequency-analyzing each frequency-flattened Doppler signal; and a frequency analysis means for frequency-analyzing each frequency spectrum data obtained by frequency analysis. A correction means for inverse correction using the amount of correction used for flattening the flattening means, and a linear combination of spectrum data in each direction of the Doppler signal obtained by the inverse correction to display or record frequency spectrum data for each direction. The present invention is characterized by an ultrasonic Doppler signal analyzer comprising means for:

(Operation) The present invention performs reverse correction by adding the correction amount used for flattening by the frequency flattening means to frequency spectrum data obtained by frequency analysis, thereby obtaining a correct result.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a configuration block diagram showing one embodiment of the present invention, and shows one channel of i-direction to q-direction Doppler signals. Therefore, the circuit shown in the figure constitutes the ultrasonic Doppler signal analyzer with two systems, but since the operation of each channel is the same, only one channel shown in the figure will be explained. In the figure, reference numeral 11 denotes a frequency flattening means that receives a baseband Doppler signal from a subject and flattens the frequency spectrum of the baseband Doppler signal in advance and independently. As the frequency flattening means 11, for example, a variable filter whose gain characteristics can be externally controlled as shown in the figure is used, but a variable equalizer may also be used. The baseband Doppler signal input to the variable filter 11 is extracted from the reflected wave from the object (subject), and then
It is obtained by separating the directions in each direction (i direction, q direction). Doppler signal processing means for obtaining this baseband Doppler signal can be implemented using existing technology.

12 is a variable filter 1 that transmits a control signal that controls the gain characteristics for each frequency band.
1 is the control circuit given to 1, and 13 is the output of the variable filter 11 (frequency-flattened Doppler signal).
This is a frequency analyzer that analyzes the frequency of As the frequency analysis means of the frequency analyzer 13, for example, Fourier transform or Walsh transform is used. 14
is an averaging circuit that receives frequency spectrum data obtained as a result of frequency analysis by the frequency analyzer 13 and averages the spectrum data in each frequency band; the output of the averaging circuit 14 is given to the control circuit 12.

A correction circuit 15 reversely corrects the output (frequency spectrum data) of the frequency analyzer 13 using the output of the control circuit 12 used to flatten the variable filter 11. Specifically, the correction circuit 15 connects the control circuit 1 to the output of the frequency analyzer 13.
Perform the operation of adding the outputs of 2. A digital scan converter 16 receives the output of the correction circuit 15, and the output is displayed on the display 17.
As the display 17, for example, a CRT is used.
To explain the operation of the device configured in this way,
It is as follows.

First, to roughly explain the overall operation, the analysis results of the frequency analyzer 13 are averaged by the averaging circuit 14 so that the frequency spectrum of the Doppler signal output from the variable filter 11 is approximately flattened, and then, The variable filter 11 for pre-correction is controlled, and the control data is added to the frequency spectrum data that is the output of the frequency analyzer 13 in the correction circuit 15 (in other words, the dB value is inversely corrected) to optimize the dynamic range. This is what we are trying to achieve.

After extracting the Doppler component from the reflected wave from the object, the baseband Doppler signal separated into the i-direction and the q-direction enters the variable filter 11 . The variable filter 11 receives a frequency flattening correction signal (filter control signal) output from the control circuit 12, and its gain is individually controlled for each frequency range. As a result, the frequency spectrum of the Doppler signal output from the variable filter 11 is approximately flattened (leveled).

The flattened Doppler signal then enters a frequency analyzer 13 for frequency analysis. Since the level in each frequency range is approximately flattened, the frequency analyzer 13 can obtain an accurate frequency spectrum for the given signal without interfering with the dynamic range or causing masking or errors. Data is obtained. This frequency spectrum data enters the averaging circuit 14. The averaging circuit 14 receives the analysis results, averages (or integrates) them by a certain amount on both the time axis and the frequency axis, and provides the output to the control circuit 12. The control circuit 12 is supplied with a filter control signal which, as described above, causes the frequency spectrum of the signal at the output of the variable filter 11 to be roughly flattened as a result.

On the other hand, the filter control signal output from the control circuit 12 is simultaneously transmitted to the correction circuit 15.
The frequency spectrum data and the dB value are added together as collection data. Therefore, for example, the control circuit 12 can adjust the gain in a certain frequency band to a certain value.
Assuming that a filter control signal (correction amount) that narrows down the dB value is applied, the corresponding correction amount (dB value) will be applied to the frequency spectrum data (dB value) that is the output of the frequency analyzer 13 in the corresponding frequency band. is added as collection data. As a result, the dynamic range transfer characteristic, which was compressed at the output stage of the frequency analyzer 13, is reversely corrected by the correction circuit 15, returning to its original state, and correct linearity is restored. In this way, the frequency spectrum data output from the correction circuit 15 is output to the display 17 via the digital scan converter 16.

Note that the filter control signal and collection data are rounded by the averaging circuit 14.
It may be used after being roughly quantized in units of about 6 dB. For example, considering an example of the spectrum of a blood flow signal, the size of the averaging (rounding process) should be such that the total width is divided into several to 10 slots on the frequency axis, and about 3 to 5 slots on the time axis. do it. Further, the variable filter 11 may have a simple configuration in which a variable attenuator is attached to each filter bank (so-called 1/3 octave filter bank). Alternatively, a transversal filter may be used. Furthermore, the output device may be a printing device such as a printer instead of a display.

FIG. 2 is a block diagram showing another embodiment of the present invention. Components that are the same as those in FIG. 1 are designated by the same reference numerals. In the embodiment shown in the figure, the frequency spectrum is flattened in advance by a filter bank in which each filter is provided with an automatic level adjustment circuit (ALC circuit). The input baseband Doppler signal is automatically level adjusted.
Commonly included in G ₁ to G _o (n is an integer). The outputs of the automatic level adjustment circuits G ₁ to G _o are then input to filter banks FB ₁ to FB _o . Here, the filters in each of the filter banks FB ₁ to FB _o can be, for example, a so-called 1/3 octave filter group.

The outputs of these filter banks FB ₁ to FB _o enter rectifier circuits D ₁ to D _o , where they are rectified and converted into DC signals. And the output of each rectifier circuit D ₁ to D _o is
Automatic level adjustment circuit G ₁ via filters F ₁ to F _o
~G _o is given as a control signal for level adjustment. Due to such a negative feedback effect, the outputs of the filter banks FB ₁ to _{FB o} are approximately flattened. The outputs of each filter bank FB ₁ to FB _o enter an adder 21, are added together, and are combined into the original Doppler signal. Moreover, the level of this Doppler signal is roughly flattened for each frequency band.

The output of the adder 21 is converted into digital data by an A/D converter 22, and then input to a frequency analyzer 13 for frequency analysis. The frequency-analyzed frequency spectrum data enters the correction circuit 15. On the other hand, the level correction signal given to the automatic level adjustment circuits G ₁ to _{G o} (outputs of the rectifier circuits D ₁ to _{D o} )
is also included in the multiplexer 23. The multiplexer 23 sequentially switches the outputs of the rectifier circuits D ₁ to D _o in response to a control signal, and the switching output is converted into digital data by the A/D converter 24 and then input to the correction circuit 15 . The correction circuit 1
Similarly to the embodiment shown in FIG. 1, 5 corrects the dynamic range by adding spectrum data and correction data (correction amounts) as dB values for each corresponding frequency band. The frequency spectrum data whose dynamic range has been corrected is sent to a display 17 via a digital scan converter 16 and output and displayed.

In the above explanation, filter banks FB ₁ to
Although we used a 1/3 octave stoichiometric frequency arrangement as the FB _o , it is not limited to this.
A constant width (constant difference) frequency arrangement may also be used.

FIG. 3 is a block diagram showing another embodiment of the present invention. The baseband Doppler signal first enters an A/D converter 31 and is converted into digital data, and then enters a digital variable filter (variable coefficient transversal filter) 32. The digital variable filter 32 is connected to the control circuit 3
The gain (coefficient) for each frequency band can be varied by control data given from 3. On the other hand, the output of the A/D converter 31 is also input to a fast Fourier transformer (FFT) 34 for rough frequency analysis. As FFT34,
It may be of a simple configuration that allows coarse frequency analysis to be performed to a degree necessary and sufficient for the purpose.

The control circuit 33 inputs the analysis results of the FFT 34 and uses control data (correction amount) to flatten the output level in each frequency range.
is applied to the digital variable filter 32. That is,
The loop of A/D converter 31→FFT 34→control circuit 33→digital variable filter 32 forms a feed forward control loop. As a result, the frequency characteristic of the output of the digital variable filter 32 becomes approximately flat, and the frequency analyzer 13 can perform frequency analysis without masking or the like occurring in the results. Frequency spectrum data obtained as a result of frequency analysis
The dB value enters the subsequent correction circuit 15 and is converted into correction data (correction amount dB) from the control circuit 33.
value) and are added for each frequency band to correct the dynamic range. And the correction circuit 15
The frequency spectrum data output from the display unit 1 is sent via the digital scan converter 16.
7 is output and displayed.

By the way, when Fourier transform is used as the frequency analysis method of the frequency analyzer 13, if the Fourier transform is performed extremely sloppily, that is, if the method of reference is simply binarized, the result will be the Wallace-Hadamard transform. It is publicly known. Furthermore, since binary multiplication can be replaced with addition, calculation speed can be significantly increased. The spectrum image output at the same time is very similar to the Fourier transform spectrum image, but "side lobes" or "superior responses" that occur at a sensitivity of about 10-odd dB or less cannot be avoided. For this reason, it was considered unsuitable for analyzing blood flow Doppler signals, which requires an analysis dynamic range of at least 40 dB or more. However, according to the present invention, if frequency flattening is performed in advance, local details can be easily kept within 10-odd dB, so if correct post-correction is performed, this Walsh transform can be used instead of Fourier transform. be able to.

For the "direction separation" work that is essential for medical Doppler equipment, base hand Doppler signals in each of the i and q directions may be separated into signals using Hilbert transformation and linear combination techniques. Such signals include USB (Upper Side Band), LSB
(Low Side Band). Alternatively, each of the i and q base hand Doppler signals may be subjected to FFT, and the result may be linearly combined to obtain a Doppler spectrum image for each direction. This technique is known. In this case, since any unbalance between the i and q channels is not allowed, the variable filter used must be able to be precisely controlled with a pair of two circuits with good reproducibility. Such a highly accurate variable filter needs to be operated as a sample point sequence filter, a switched capacitor filter, or a digital filter after A/D conversion. In principle, the strength or level of the analysis results of the i and q direction Doppler signals are exactly the same, so as a preliminary spectrum flattening means, each means of detection and feedback or inverse correction after feedback analysis is one system. nothing.

(Effects of the Invention) As explained in detail above, according to the present invention,
Before performing frequency analysis on the ultrasound Doppler signal, analyze it by flattening the frequency in advance and adding the control data (correction amount dB value) used for the frequency flattening to the spectrum data (dB value) after frequency analysis. It is possible to realize an ultrasonic Doppler signal analyzer that satisfies the dynamic range required for and can perform accurate frequency analysis even if the frequency analysis means itself does not have a wide dynamic range.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, FIGS. 2 and 3 are block diagrams showing other embodiments of the present invention, and FIG. 4 is a diagram showing an example of the structure of a conventional device. be. 1... Doppler filter, 2, 34... FFT,
3...CRT, 4...Buffer amplifier, 5...Speaker, 6...Operation unit, 11...Variable filter,
12, 33...Control circuit, 13...Frequency analyzer, 14...Averaging circuit, 15...Correction circuit, 16...Digital scan converter, 1
7...Display device, 21...Adder, 22, 24, 3
1... A/D converter, 23... Multiplexer,
32...Digital variable filter, F ₁ ~F _o ...
Filter, FB ₁ ~ _{FB o} ...Filter bank, D ₁ ~
D _o ... Rectifier circuit, G ₁ ~ _{G o} ... Automatic level adjustment circuit.

Claims (1)

[Claims] 1. Doppler signal processing means for extracting Doppler components from reflected waves from a subject and then obtaining baseband Doppler signals separated in each direction; and the obtained baseband Doppler signals. A frequency flattening means for flattening the frequency spectrum of each signal independently in advance; a frequency analysis means for frequency-analyzing each frequency-flattened Doppler signal; and a frequency-flattening means for frequency-analyzing each frequency spectrum data obtained by frequency analysis. An ultrasonic Doppler signal comprising a correction means for inversely correcting with the amount of correction used for flattening the flattening means, and an output means for displaying or recording spectrum data in each direction of the Doppler signal obtained by inversely correcting. Analysis equipment. 2. The ultrasonic Doppler signal analysis apparatus according to claim 1, wherein the frequency flattening means is feedback-controlled by the output of the frequency analysis means. 3. The ultrasonic Doppler signal analysis apparatus according to claim 1, wherein the frequency flattening means is provided with a filter bank and an automatic level adjustment circuit for each frequency band. 4. The ultrasonic Doppler signal analysis device according to claim 1, wherein the frequency flattening by the frequency flattening means is performed by coarse frequency analysis and feedback or feedback control based on the result thereof. . 5. The ultrasonic Doppler signal analysis apparatus according to claim 1, wherein the frequency analysis means is a Walsh transform. 6 A Doppler signal processing means that extracts Doppler components from the reflected waves from the subject to obtain baseband Doppler signals, and a frequency spectrum of the obtained i (in-phase) and q (quadrature) baseband Doppler signals in advance. a frequency flattening means for flattening under common conditions; a frequency analysis means for frequency-analyzing each frequency-flattened Doppler signal; and a frequency analysis means for frequency-analyzing each frequency spectrum data obtained by the frequency analysis. A correction means for performing inverse correction using the amount of correction used for the correction, and a means for displaying or recording frequency spectrum data for each direction by linearly combining spectrum data in each direction of the Doppler signal obtained by inverse correction. An ultrasonic Doppler signal analysis device consisting of: 7. The ultrasonic Doppler signal analysis apparatus according to claim 6, wherein the frequency flattening means is feedback-controlled by the output of the frequency analysis means. 8. The pre-flattening process is performed by a variable series sample point filter, and a pair of sample point series filters with the same structure are used so that the same series is applied to each of i and q. An ultrasonic Doppler signal analysis device according to claim 6.