JP2002186615A - Ultrasonic daignostic device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a means to improve image quality of ultrasonic images by effectively reducing speckle noise, or the like, contained in the ultrasonic images. SOLUTION: The first signal component of the first frequency band and the second signal component of the second frequency band are extracted from receiving signals, whereas envelop data are obtained from these signal components before the cross-correlation is calculated by a cross-correlation unit 40 with respect to these envelop data. The filter properties of a filter 42 in relation with the receiving signals is converted in accordance with the cross-correlation levels.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an ultrasonic diagnostic apparatus, and more particularly to reduction of noise contained in an ultrasonic image.

[0002]

2. Description of the Related Art An ultrasonic diagnostic apparatus is an apparatus for displaying an ultrasonic image by transmitting and receiving ultrasonic waves to and from a living body. As an ultrasonic image, a two-dimensional black and white tomographic image B
Mode images are well known. In such a B-mode image,
It includes an artifact called speckle of a granular pattern caused by interference of an ultrasonic wavefront unrelated to a structure in a living body. Such noise is useless data that hinders observation of the B-mode image. Note that such speckles or artifacts are also observed in other ultrasonic images such as M-mode images.

[0003] The present invention has been made in view of the above-mentioned conventional problems, and an object of the present invention is to improve the image quality of an ultrasonic image. It is another object of the present invention to specify a noise component contained in a received signal itself using the received signal itself.

[0004]

(1) In order to achieve the above object, the present invention provides a transmitting and receiving means for transmitting and receiving an ultrasonic wave and outputting a received signal, and a first frequency band based on the received signal. Component extraction means for extracting a first signal component and a second signal component of a second frequency band, first detection means for obtaining first envelope data from the first signal component, and second detection means for obtaining a second envelope data from the second signal component. Second detection means for obtaining envelope data; cross-correlation calculation means for performing a cross-correlation calculation between the first envelope data and the second envelope data to calculate a cross-correlation value;
Filtering means for filtering a received signal according to the cross-correlation value.

According to the above configuration, a cross-correlation operation is performed on the first envelope data and the second envelope data as envelope signals after detection, and a received signal (echo data or image data) is calculated based on the cross-correlation value. ) Is filtered. This utilizes the tendency that the cross-correlation value is relatively large in the case of a true echo from a living tissue, and relatively small in the case of noise. Therefore, according to the present invention, noise can be selectively removed and suppressed, so that the image quality of an ultrasonic image can be improved. The filtering may be performed one-dimensionally or two-dimensionally.

Preferably, the first envelope data and the second envelope data are data on a time axis, and the cross-correlation calculating means calculates the first envelope data and the second envelope data on a time axis. Perform a cross-correlation operation.

Preferably, the first envelope data and the second envelope data are data on a frequency axis, and the cross-correlation calculating means calculates the first envelope data and the second envelope data on a frequency axis. Perform a cross-correlation operation.

Preferably, the filter means is a low-pass filter, and when the cross-correlation value is large, the cut-off frequency of the low-pass filter is set high, and when the cross-correlation value is small, the cut-off frequency of the low-pass filter is set. Filter characteristic setting means for setting the off frequency low is provided. According to this configuration, it is possible to selectively and efficiently perform filtering on noise while avoiding filtering on true echo data as much as possible. Of course, in addition to changing the cutoff frequency, changing the pass band and adjusting the gain are also modes of filtering.

(2) To achieve the above object,
The present invention relates to a transmitting and receiving means for transmitting and receiving an ultrasonic wave and outputting a received signal, and a component extracting means for extracting a first signal component of a first frequency band and a second signal component of a second frequency band from the received signal. First detection means for obtaining first envelope data from the first signal component; second detection means for obtaining second envelope data from the second signal component; a one-dimensional window along the signal time series; Cross-correlation calculation means for calculating a cross-correlation value by performing a cross-correlation calculation on the first envelope data and the second envelope data cut out at each window position while scanning And filtering means for performing filtering on.

(3) In order to achieve the above object,
The present invention provides a transmitting / receiving means for scanning an ultrasonic beam and outputting a received signal for each beam position, a first signal component in a first frequency band and a second signal component in a second frequency band from the received signal.
Component extraction means for extracting a signal component, first detection means for obtaining first envelope data from the first signal component, second detection means for obtaining second envelope data from the second signal component, First envelope data and second envelope data cut out at each window position while scanning a two-dimensional window on a two-dimensional coordinate system defined by a depth direction on the beam and a scanning direction of the ultrasonic beam. A cross-correlation calculating means for calculating a cross-correlation value by performing a cross-correlation calculation of data, and a filter means for filtering a received signal according to the cross-correlation value. According to this configuration, since the correlation calculation can be performed two-dimensionally, the noise determination accuracy can be further improved.

(4) The principle of the present invention will be described below. In order to identify noise (especially speckle) included in a reception signal obtained by transmission and reception of ultrasonic waves, signal components in a plurality of (preferably two) different frequency bands are extracted from the reception signal.

FIG. 1 shows a spectrum 300 of a received signal. Reference numeral 302 denotes a pass band characteristic of a band-pass filter when the entire area is extracted.
Reference numerals 4 and 306 denote pass bands of a band-pass filter for extracting signal components of two frequency bands as described above. The pass band characteristic indicated by reference numeral 304 is set on the lower side of the spectrum 300 of the received signal, and the pass band characteristic indicated by reference numeral 306 is set on the higher side of the spectrum 300 of the received signal. Their passband characteristics are substantially separated, with some tails overlapping each other. Of course, if there is a difference between the two pass band characteristics as a whole, various pass band characteristics can be set.

When a correlation operation is performed between the envelope data of the signal components of the respective frequency bands, the shape of the envelope of the true echo component from the in-vivo structure is similar to each other, so that the cross-correlation value is obtained. Is large, while the cross-correlation value is small because the shape of the noise envelope is random. Therefore, from the magnitude of such a cross-correlation value, the probability that each received signal is a true echo component (or noise) can be obtained, or it can be determined whether the received signal is a true echo component or noise. it can.

More specifically, for example, first, signal components (data trains) of two frequency bands are extracted from a received signal, and each signal component is detected to be envelope data (envelope) data (data train). Are individually stored in the memory in units of one line (or one frame). And from each data string,
Two sets of data strings (x _i , y _i ) composed of N data in a predetermined window (see FIG. 2) (where i = 1, 2,...)
., N) are cut out and two such data strings (x
_i , y _i ) are subjected to a cross-correlation operation. By the way,
FIG. 2 shows a scanning surface 306 formed by electronically scanning the ultrasonic beam 308. The above-described extraction of the data string is performed by setting a one-dimensional window 310 on each ultrasonic beam 308. Equivalent to. The extraction of the signal component may be performed after the data string is cut out.

The correlation operation is represented by the following equation, where the cross-correlation value is Rxy.

[0016]

(Equation 1) The cross-correlation value Rxy is 1 if the data strings are exactly the same, and 0 if they are completely different. That is, the larger the difference, the smaller the cross-correlation value. Here, in the case of speckle (noise), the cross-correlation value becomes very small. On the other hand, in the case of an echo from a structure in a living body (true echo), the cross-correlation value increases.
Therefore, it is possible to discriminate between a speckle and a true echo based on the cross-correlation value, and furthermore, if the filter characteristic is linked to the cross-correlation value, the suppression process can be selectively performed on the speckle. It can be carried out. For example, in the case of a true echo, the cutoff frequency of the low-pass filter (LPF) through which the data passes is increased to pass the data as it is, while in the case of speckle, the cutoff of the LPF is cut off The frequency is lowered so that the data is suppressed and removed.

When the above-described cross-correlation calculation is performed at a certain window position, the window position is shifted and the same cross-correlation calculation is performed. Then, when this is repeated, noise reduction processing can be performed on all data.

Although the above-described window is a one-dimensional window along the depth direction (ultrasonic beam direction), it may be a two-dimensional window as shown in FIG. In FIG. 3, J corresponds to the depth direction (ultrasonic beam direction), and I corresponds to the scanning direction of the ultrasonic beam or the horizontal direction of the B-mode image. A two-dimensional window 314 is raster-scanned on the two-dimensional matrix 312, and a data string is cut out at each scan position. Wind 3
The size of 14 is WI × WJ. In this case, the cross-correlation value Rxy is obtained by the following arithmetic expression.

[0019]

(Equation 2) In the case where the above configuration is adopted, a two-dimensional LPF may be used as a filter for processing a received signal. The cross-correlation calculation may be performed on the frequency axis in addition to the time axis.

Since the received signal spectrum changes depending on the depth of the reflection point due to the influence of frequency-dependent attenuation and the like, the passband characteristics 304 and 3 shown in FIG.
06 is desirably changed dynamically.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings.

FIG. 4 shows a preferred embodiment of the ultrasonic diagnostic apparatus according to the present invention, and FIG. 4 is a block diagram showing the overall configuration.

In FIG. 4, the array transducer 10 is provided in an ultrasonic probe (not shown).
The array vibrator is constituted by a plurality of vibrating elements. The ultrasonic beam is electronically scanned on the array transducer, whereby the scanning surface 306 as shown in FIG.
Is formed.

The transmitter 12 functions as a so-called transmission beam former, and is a circuit for supplying a transmission signal to a plurality of vibrating elements constituting an array vibrator. The receiver 13 functions as a so-called reception beamformer, executes phasing addition processing on a plurality of reception signals from a plurality of vibration elements forming an array transducer, and performs reception after phasing addition. It is a circuit that outputs a signal.
This received signal is supplied to three signal processing units 1 in this embodiment.
4, 16, and 18 are input in parallel.

First, the signal processing unit 14 will be described. This signal processing unit 14 is generally provided in a conventional ultrasonic diagnostic apparatus, and performs detection processing such as quadrature detection on a received signal. A detector 20 to execute;
An amplitude calculator 22 that executes an amplitude operation on the reception signal (complex signal) after the detection, and a line memory unit 24 that stores the calculated amplitude value, that is, echo data for each ultrasonic beam. It is configured. A specific configuration example of the detector 20 will be described later with reference to FIGS. The amplitude calculator 22 will be described later with reference to FIG. The line memory unit 24 is configured by two line memories in the present embodiment, and performs a ping-pong operation using the two memories alternately for so-called data writing and reading.

The filter 42 constitutes a low-pass filter in this embodiment, and the cutoff frequency is dynamically set by a cross-correlation value described later in detail. The data output from the filter 42 is input to a scan converter 44, where the data is displayed on a display 46 after undergoing processing such as coordinate conversion from a measurement coordinate system to a display coordinate system. Specifically, the display 46 displays an ultrasonic image such as a B-mode image. By the way, the scan converter 44
Is constituted by, for example, a digital scan converter.

Next, the signal processing units 16 and 18 will be described. These signal processing units 16 and 18 have the same configuration as the signal processing unit 14 described above, that is, the detector 2
6, 28, amplitude calculators 30 and 32, and line memory units 34 and 36. However, the detectors 20, 26, and 28 have filters having pass band characteristics different from each other. More specifically, the detector 20 has a filter having a pass band characteristic covering the entire frequency band indicated by reference numeral 302 in FIG. The detector 26 includes a filter having a lower passband characteristic indicated by reference numeral 304 in FIG. The detector 28 includes a filter having a pass band characteristic on the high frequency side indicated by reference numeral 306 in FIG. Each detector has a function of converting an RF signal into an envelope signal.

Therefore, the amplitude calculator 22 calculates the amplitude of the received signal in the entire frequency band, the amplitude calculator 30 calculates the amplitude of the signal component on the low frequency side, and the amplitude calculator 32 calculates the amplitude of the signal component on the low frequency side. The amplitude of the signal component on the band side is calculated. In the line memories 34 and 36, a data string having the amplitude value calculated as described above is stored for each ultrasonic beam.

In the present embodiment, a data sequence composed of N data is extracted from the line memory units 34 and 36 as described above, and the two sets of data sequences are processed by the cross-correlator 40. A cross-correlation operation is performed. Here, N data is cut out from the line memory units 34 and 36 by setting a predetermined window as described above, and by scanning the window along the data column direction, at each window position, A cross-correlation operation is performed.

As can be understood from the above description, the cross-correlator 40 shown in FIG. 4 executes a cross-correlation operation between two sets of data strings on the time axis. Equations such as the above equation (1) are used. The cross-correlation value that is the result of the cross-correlation calculation is output to the filter 42, and the filter characteristics of the filter 42 are adaptively set to the cross-correlation value.
Thus, for echo data at a certain depth,
Filtering can be performed according to the cross-correlation value,
If the echo data is noise, the noise can be effectively removed or reduced by filtering.

FIG. 5 shows the detectors 20, 2 shown in FIG.
6, 28 show specific configuration examples. The configuration example shown in FIG. 5 shows a so-called quadrature detector.

A reception signal is input to one input terminal of the two mixers 70 and 72, and a reference signal is input to the other input terminal. The mixer 72 is a π / 2 phase shifter 68
, A reference signal having a phase angle shifted by 90 degrees is input to the mixer 72, while the input reference signal is input to the mixer 72 as it is. The quadrature detection can be performed based on the phase relationship between the two reference signals, and the signal components in the baseband region are extracted by the low-pass filters 74 and 76 from the signals after the quadrature detection. In this case, reference numeral 302 and 304 in the pass band characteristic is set, i.e. the passband characteristics of each Figure 1 of the low-pass filter 74, 76 for each detector 20,26,28 shown in FIG. 4 , 306 are set.

FIG. 6 shows another configuration example of the detectors 20, 26, and 28. The configuration example shown in FIG. 6 is a so-called quadrature sampling circuit. Band pass filter 78
Are input to the sampling circuits 82 and 84 in parallel. Here, the sampling circuit 82 performs sampling in accordance with the sampling clock, and the sampling 84 performs the sampling through the π / 2 shifter 80.
Sampling is performed according to a sampling clock whose phase is shifted by 0 degrees. As is well known, a detection signal similar to the quadrature detection can be obtained by such an orthogonal sampling method. Of course, the detectors 20, 26, 28
, Different characteristics are set as the band-pass characteristics of the band-pass filter 78. Specifically, the respective pass-band characteristics are denoted by reference numeral 3 in FIG.
02, 304, and 306 are set.

FIG. 7 shows an example of the configuration of the amplitude calculators 22, 30, and 32. The absolute value calculator 86 includes the detector 2
The absolute value is obtained by adding the square of the real part and the square of the imaginary part of the complex signal output from 0 and taking the square root of the addition result. This makes it possible to calculate the amplitude of the echo data.

The data representing the amplitude is input to a logarithmic converter 88, which performs logarithmic compression processing. Then, the coefficient β is added to the logarithmically compressed data by the adder 90, and the multiplier 92 is multiplied by the coefficient α. Here, the adder 90 functions as a gain adjustment circuit, and the multiplier 92 functions as a contrast circuit. The data output from the multiplier 92 is input to a low-pass filter 94, the band is limited for resampling by a decimator 96 immediately after the low-pass filter 94, and the data output from the low-pass filter 94 is input to a decimator 96. The decimator 96 performs a resampling process according to the display pixel rate.

FIG. 8 conceptually shows data strings stored in the line memory units 34 and 36. Here, reference numeral 304A denotes a data string stored in the line memory unit 34, and this data string corresponds to the pass band characteristic 304 in FIG.
Corresponding to the extracted signal component. In FIG. 8, reference numeral 306A denotes a data string stored in the line memory unit 36, which corresponds to a signal component extracted by the passband characteristic 306 in FIG.

As described above, two sets of data strings are partially cut out from such two data strings by a window, and a cross-correlation operation is performed on the data strings in the window.

FIG. 9 shows a configuration example of the filter 42 shown in FIG.

A cross-correlation value is input to the filter characteristic selection table 98, and a selection signal representing a filter characteristic corresponding to the cross-correlation value is input to the filter coefficient table 100.
Is output to The filter coefficient table 100 has, for example, table contents as shown in FIG. 10. Specifically, in order to change the cutoff frequency of the filter 42 as a low-pass filter according to the cross-correlation value,
A filter coefficient sequence associated with a plurality of filter characteristics is stored. When the selection signal of the filter characteristics is inputted, the filter coefficient sequence a ₁ ~a _n representing the filter characteristic selected selected from the filter coefficient table 100 by the selection signal is output, each of the filter coefficients corresponding The signals are input to one of input terminals of multipliers 104-1 to 104-n. The received signal is sequentially input to a plurality of delay lines 102-1 to 102-n connected in series, and data output from the front, rear, and intermediate points of each delay line is output to each multiplier 104-1.
Are input to the other input terminals of .about.104-n. Each of the multipliers 104-1 to 104-n multiplies each data by a filter coefficient, and the multiplication result is added in the adder 106.

FIG. 11 shows the relationship between the cross-correlation value and the cut-off frequency of the low-pass filter. In the present embodiment, the cut-off frequency decreases as the cross-correlation value decreases. The higher the correlation value, the higher the cutoff frequency. In the figure, a frequency change characteristic like an S-shaped function is shown.

FIG. 12 is a block diagram showing the overall configuration of an ultrasonic diagnostic apparatus according to another embodiment. The same components as those shown in FIG. 4 are denoted by the same reference numerals, and description thereof will be omitted.

In the configuration example shown in FIG. 12, in each of the signal processing units 50, 52, 54, the amplitude calculators 22, 3 are used.
Scan converters 56, 60, 62 and frame memory units 58, 64, 66 are provided after 0 and 32.
Here, each of the scan converters 56, 60, and 62 corresponds to the scan converter 44 shown in FIG. 4, that is, the scan converters 56, 60, and 62 convert the transmission / reception coordinate system to the display coordinate system. Coordinate transformation is performed.
The data is stored in the frame memory unit 58,
64 and 66 are stored for each frame.

When the data is stored in the frame memory units 64 and 66 in this manner, a two-dimensional window 314 shown in FIG. 3 is used, and a data string is formed at each window position while the window 314 is raster-scanned. The extracted cross-correlator 40 calculates a cross-correlation value of two sets of two-dimensional data strings. Then, according to the correlation value, the filter 42 performs filtering on the data output from the frame memory unit 58.

FIG. 13 shows a processing configuration example on the frequency axis of the cross-correlator 40 of FIG. In this configuration example, FFT circuits 110 and 1 are respectively applied to two one-dimensional data strings.
At 12, an FFT operation is performed, whereby those data strings are converted into data strings on the frequency axis. These data strings are multiplied by a multiplier 113, and the result of the multiplication (corresponding to a cross-correlation value) is input to an IFFT circuit 114, where an inverse FFT operation is performed.

FIG. 14 shows an example of the processing configuration on the frequency axis of the cross-correlator 40 of FIG. In this configuration example, FFT circuits 128 and 1 are respectively applied to two two-dimensional data strings.
At 30, a two-dimensional FFT operation is performed, whereby those data strings are converted into data strings on the frequency axis. These data strings are multiplied by a multiplier 131, and the result of the multiplication (corresponding to the cross-correlation value) is obtained by the two-dimensional IFFT circuit 1.
32, where a two-dimensional inverse FFT operation is performed.

As described above, according to each of the above embodiments, the cross-correlation value is obtained from the two frequency components of the received signal, and the filtering is performed on the received signal in accordance with the cross-correlation value to thereby reduce noise (particularly, noise). (Speckle) can be effectively removed. In the above embodiment, two pass band characteristics are set, but the respective pass band characteristics may partially overlap.

[0047]

As described above, according to the present invention,
It is possible to effectively reduce noise in the ultrasonic image and improve the image quality of the ultrasonic image.

[Brief description of the drawings]

FIG. 1 is a diagram showing the relationship between the spectrum of a received signal and passband characteristics.

FIG. 2 is a diagram showing a scanning plane and a one-dimensional window.

FIG. 3 is a diagram showing a two-dimensional window.

FIG. 4 is a block diagram showing an overall configuration of the ultrasonic diagnostic apparatus according to the embodiment.

FIG. 5 is a block diagram illustrating a configuration example of a detector.

FIG. 6 is a block diagram showing another configuration example of the detector.

FIG. 7 is a block diagram illustrating a configuration example of an amplitude calculator.

FIG. 8 is a diagram showing a data string stored in a line memory unit.

FIG. 9 is a block diagram illustrating a configuration example of a filter.

FIG. 10 is a conceptual diagram illustrating a configuration example of a filter coefficient table.

FIG. 11 is a diagram illustrating a relationship between a cross-correlation value and a cutoff frequency.

FIG. 12 is a block diagram illustrating an overall configuration of an ultrasonic diagnostic apparatus according to another embodiment.

FIG. 13 is a diagram illustrating a configuration example of a one-dimensional cross-correlation process on a frequency axis.

FIG. 14 is a diagram illustrating a configuration example of a two-dimensional cross-correlation process on a frequency axis.

[Explanation of symbols]

10 array transducer, 12 transmitter, 13 receiver, 1
4, 16, 18 signal processing unit, 20, 26, 28 detector, 22, 30, 32 amplitude calculator, 24, 34, 36
Line memory, 40 cross-correlator, 42 filter, 44 scan converter, 46 display.

Claims (6)

[Claims] 1. A transmitting / receiving means for transmitting / receiving an ultrasonic wave and outputting a received signal, and a component for extracting a first signal component of a first frequency band and a second signal component of a second frequency band from the received signal. Extraction means; first detection means for obtaining first envelope data from the first signal component; second detection means for obtaining second envelope data from the second signal component; An ultrasonic wave comprising: a cross-correlation calculating means for calculating a cross-correlation value by performing a cross-correlation calculation of the second envelope data; and a filter means for filtering a received signal according to the cross-correlation value. Diagnostic device. 2. The apparatus according to claim 1, wherein the first envelope data and the second envelope data are data on a time axis, and the cross-correlation calculating means is a first envelope data on a time axis. And performing a cross-correlation operation on the second envelope data. 3. The apparatus according to claim 1, wherein said first envelope data and said second envelope data are data on a frequency axis, and said cross-correlation calculating means is a first envelope data on a frequency axis. And performing a cross-correlation operation on the second envelope data. 4. The apparatus according to claim 1, wherein the filter means is a low-pass filter, and when the cross-correlation value is large, the cut-off frequency of the low-pass filter is set high, and when the cross-correlation value is small. Wherein an ultrasonic diagnostic apparatus is provided with filter characteristic setting means for setting a low cut-off frequency of the low-pass filter. 5. A transmission / reception means for transmitting / receiving an ultrasonic wave and outputting a reception signal, and a component extraction for extracting a first signal component in a first frequency band and a second signal component in a second frequency band from the reception signal. Means, first detection means for obtaining first envelope data from the first signal component, second detection means for obtaining second envelope data from the second signal component, one-dimensionally along a signal time series Cross-correlation calculating means for calculating a cross-correlation value by performing a cross-correlation calculation on the first envelope data and the second envelope data cut out at each window position while scanning the window; An ultrasonic diagnostic apparatus, comprising: filter means for filtering a signal. 6. A transmitting and receiving means for scanning an ultrasonic beam and outputting a received signal for each beam position, a first signal component in a first frequency band and a second signal component in a second frequency band from the received signal. , A first detector for obtaining first envelope data from the first signal component, a second detector for obtaining second envelope data from the second signal component, and On the two-dimensional coordinate system defined by the depth direction of the ultrasonic beam and the scanning direction of the ultrasonic beam, while scanning the two-dimensional window, the first envelope data and the second envelope data cut out at each window position are scanned. A cross-correlation calculating means for calculating a cross-correlation value by performing a cross-correlation calculation; and a filter means for filtering a received signal according to the cross-correlation value. Ultrasonic diagnostic equipment.