JPH0654845A - Ultrasonic diagnostic equipment - Google Patents

Abstract

(57) [Summary] [Object] The present invention relates to an ultrasonic diagnostic apparatus having a function of correcting non-uniformity of sound velocity in a subject, and has an improved function of correcting non-uniformity of sound velocity. [Structure] After a time lag is once detected based on a reception signal cut out by a gate unit, a feature amount of the time lag is obtained based on the detected time lag, and the feature amount is compared with a threshold value. A scan line having a time shift detection error is detected, and the time shift is corrected for the scan line having the time shift detection error.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an ultrasonic diagnostic apparatus for displaying a tomographic image in a subject by ultrasonic waves, and more particularly to an ultrasonic diagnostic apparatus having a function of correcting non-uniformity of sound velocity in the subject. .

[0002]

2. Description of the Related Art An ultrasonic wave is transmitted to an inside of a subject, the ultrasonic wave returned by being reflected by a tissue inside the subject is received to obtain a reception signal, and a tomographic image inside the subject is obtained based on the reception signal. An ultrasonic diagnostic apparatus that is used to diagnose diseases such as internal organs by displaying the information is used. In this ultrasonic diagnostic apparatus, a so-called reception focus method is used.

FIG. 9 is an explanatory diagram of this receiving focus method. N electroacoustic conversion elements (hereinafter simply referred to as “elements”) E ₁ arranged in a predetermined direction (left and right direction in FIG. 9),
_{E 2, ..., E i,} ..., when providing electrical signals to En, each of these elements _{E 1, E 2, ...,} E i, ..., are converted into ultrasonic waves E _n, sample the ultrasound to be It is sent inward.

Here, this ultrasonic wave is transmitted to a large number of elements E ₁ , E
Consider a case where the light is reflected at a point a on a vertical line extending from the center 0 of ₂ , ..., E _i , ..., E _n into the subject. The reflected wave from the point a is compared with the element E ₁ on the end side far from the point a and the element E _i on the center side close to the point a from the element E ₁ far away. Also reaches the nearby element E _i first. This time difference of arrival is the time when the ultrasonic wave advances by the distance difference ΔL = L ₁ −L _i when the distances from the point a to the elements E ₁ and E _i are L ₁ and L _i , respectively. Here, assuming that the sound velocity in the subject is uniform,
When this speed of sound is C, this time difference is represented by ΔL / C. Thus, point a and each element E ₁ , E ₂ , ..., E _i ,
..., because there is a distance difference between E _n and each distance,
Each distance difference is converted into each time, and each element E ₁ , E ₂ , ...,
By phasing each received signal obtained by E _i , ..., E _n by an amount corresponding to each time difference, a phasing process for aligning the arrival times of these received signals with each other is performed.
It is possible to emphasize the received signal corresponding to the ultrasonic wave reflected at the point, that is, to focus the reception at the point a. The ultrasonic waves reflected deeper in the subject reach each element later, so that each element E ₁ , E ₂ , ..., E _i ,
, E _n , the delay amounts for the respective received signals are sequentially changed and added to each other, so that not only the point a but also the points on the perpendicular line extending from the center 0 to the inside of the subject including the point a. It is also possible to perform so-called dynamic focusing, which continuously focuses. This perpendicular line is called a scanning line, and by changing each delay amount for each received signal, this scanning line is converted into an electroacoustic conversion element E ₁ , E ₂ , ..., E.
_i, ..., it can be a direction to perform a so-called sector scan to deflect the so-called linear scanning and the scanning line is moved in parallel to the (left-right direction in FIG. 9) in a fan shape lined E _n, thereby two-dimensionally in the subject A tomographic image can be obtained. Further, it is also known that a three-dimensional stereoscopic image can be obtained by arranging electroacoustic transducers also in a direction perpendicular to the paper surface of FIG. 9 and scanning in this arranging direction.

However, it is known that there are fat layers, muscles, and various other tissues in the human body, and that the sound velocity in the fat layers is different from that in other tissues. FIG. 10 is a diagram showing reception focus when the sound velocity is not uniform. As shown in FIG. 10, for example, when diagnosing the liver of the abdomen of a human body, the sound velocity is about 1470 m near the body surface.
There is a fat layer having a relatively slow sound speed of / sec, a muscle layer having a sound speed of about 1540 m / sec exists below the fat layer, and a liver having a sound speed of about 1540 m / sec exists below the muscle layer. In addition, the fat layer may be deposited in muscle or liver.

As described above, when there are portions having different sound velocities in the subject, even if each delay signal is given a delay amount determined under the assumption that the sound velocities are uniform, as shown in FIG. The arrival times of the signals are not uniform, and therefore, even if all of these received signals are added, the signal at the point a is not focused and the tomographic image is blurred. Here, the difference in arrival time of each received signal is referred to as "time difference". Moreover, since the thickness of this fat layer varies from person to person, it is not possible to uniformly correct each delay amount for each received signal.

As a method for solving this, a method using a cross-correlation calculation ([1] US Pat.
No. 4, [2] S. W. FLAX AND M. O'DO
NNEL, "Phase-Aberration Co
redirection Using Signals F
rom Point Reflectors andD
iffuse Scatterres: Basic P
rinciples ”IEEE TRANSACTI
ON ON ULTRASONICS, FERROE
L ECTRICS, AND FREQUENCY
CONTROL, VOL35, NO. 6, NOVE
MBER 1988, p758-p767) are known.

FIG. 11 is an explanatory diagram of a method for correcting a time shift using the cross correlation method. Here, first, assuming that the sound velocity in the subject is uniform, each delay amount is given to each reception signal, and a part of each reception signal after giving each delay amount is cut out as a correlation calculation region. The time lag is obtained as follows based on each received signal in the correlation calculation area.

That is, among the cut-out reception signals, the cross-correlation calculation is performed between two reception signals adjacent to each other, and the time between the adjacent elements is calculated from the position where the obtained peak value of the cross-correlation function exists. The deviation Δτ is obtained. This time shift Δτ is calculated for all of the two received signals adjacent to each other, and the calculated time shifts Δτ are sequentially integrated with reference to the arrival time of the received signal corresponding to the leftmost element in the figure, for example. As a result, the respective time lags Δt of the other received signals with respect to the reference received signal are obtained, and the respective delay amounts for the respective received signals are changed so that the respective time lags Δt are corrected. The arrival times of the received signals can be aligned.

As a conventionally proposed method for detecting the time lag of each received signal, a method called a quadrature detection method is known in addition to the above-mentioned cross correlation method ([3] US Pat. No. 4,835,689). reference). In this quadrature detection method, the phase difference between the received signals is obtained instead of the narrowly-defined time difference between the received signals. However, the present invention does not depend on this time difference detection method, so here, “time difference” is used. Is interpreted in a broad sense to include “phase difference”.

[0011]

[Problems to be Solved by the Invention] [1], [2],
Although a solution to the problem of non-uniform sound velocity in the subject has been proposed as shown in each document of [3], when an intensity reflector that strongly reflects ultrasonic waves exists near the scanning line, The problem remains and the applicant has proposed a solution.

FIG. 12 is a diagram for explaining a problem that occurs when a strong reflector exists near the scanning line. It is assumed that there is a strong reflector B in the vicinity of the point A on the scanning line, and that each received signal is given a delay amount on the assumption that the sound velocity in the subject is uniform. At this time, each received signal is a mixture of the signal component caused by the reflected wave from the point A and the signal component caused by the reflected wave from the point B, and the signal component caused by the reflected wave from the point B is Since the point B is located at an oblique position with respect to the point A which is currently focused, a time lag occurs so that it is linearly inclined as shown in the figure. Here, in order to correct the time lag due to the presence of a nonuniform sound velocity portion such as a fat layer, the time lag is detected by, for example, the above-described cross-correlation method using the signals in the time lag detection area shown in the figure. When the signal component due to the reflected wave from the strong reflector existing at the point B is detected, the time shift is detected by this signal component, and as a result, the scanning line is deflected in the direction of the point B. As described above, each delay amount with respect to each reception signal is corrected, and it is as if the reflector at the point B exists at the position of the point A on the tomographic image obtained based on the reception signal thus corrected. As a result, a virtual image appears.

As a method of solving this problem, considering that the signal component resulting from the reflected wave from the point B has a linear inclination as shown in FIG. 12, the time deviation of such an inclined linear shape is considered. When this is detected, the present applicant has proposed a method of detecting a deviation of the detected time shift from this straight line as a new time shift based on this straight line (Japanese Patent Application No. 3-66105). Issue March 29 Application
reference).

Here, when the only strong reflector B exists near the scanning line, the influence of the strong reflector B is eliminated by the method according to the above proposal, and the time lag can be corrected correctly. However, in the method according to the above proposal, when a plurality of strong reflectors are present in the vicinity of the scanning line at the same time, and therefore the reflection signals from the plurality of strong reflectors are within the time shift detection range, the time shift detection error occurs. Occurs.

FIG. 13 is a diagram schematically showing a received signal when two strong reflectors are present near the scanning line. In the case shown in FIG. 13, strong reflected waves from the point A in a section of W _a, W a strong reflected wave from the point B is _{Step b,} signals surrounded by a broken line area is dominant and becomes in this region The signal will detect the time lag. Therefore, even if the method according to the above proposal is mechanically applied to remove the slope of the detected time lag, it does not mean that the time lag dependent on the non-uniformity of sound velocity is extracted, and the time lag is corrected. When the delay amount of the received signal is controlled to
The in-vivo tomographic image was rather deteriorated than before correction, which was a problem. Such a problem is a problem not only in the cross-correlation method and the quadrature detection method but also in all the time shift detecting means.

In view of the above problems, the present invention provides an ultrasonic diagnostic apparatus having an improved correction function of sound velocity non-uniformity by detecting a time shift detection error and correcting the time shift of the detection error. The purpose is to do.

[0017]

The ultrasonic diagnostic apparatus of the present invention for achieving the above object is such that an ultrasonic wave transmitted into a subject forms a transmitting side focal point at a predetermined point in the subject. In addition, a voltage pulse delayed by each predetermined delay amount is generated by the transmission means and applied to each of the large number of electroacoustic conversion elements arranged in a predetermined direction, transmitted to the inside of the subject and reflected inside the subject. A large number of received signals are obtained by receiving the ultrasonic waves by each of a large number of electroacoustic transducers, and a large number of received signals are formed so that the receiving side focus is formed along the scanning line extending in the subject. The present invention relates to an ultrasonic diagnostic apparatus that delays each predetermined delay amount using a delay amount variable delay unit and obtains an addition signal by adding each other, and displays a tomographic image of a subject based on the addition signal. Yes, (1) A gate unit that cuts out a received signal corresponding to a predetermined depth region in the sample as a time shift detection region, and (2) a relative received signal based on the received signal in the time shift detection region cut out by the gate unit. Time lag detection unit for detecting a time lag, and (3) obtaining a feature amount of the time lag based on the time lag detected by the time lag detector and comparing the feature amount with a predetermined threshold value. By doing so, a time shift detection error detection unit for detecting a scan line having a time shift detection error, and (4) a time shift detection unit detects a scan line having a time shift detection error detected by the time shift detection error detection unit. The time shift detection error correction unit that corrects or replaces the generated time shift, and (5) the time detected by the time shift detection unit for the scanning line detected by the time shift detection error detection unit as having no time shift detection error. And the scanning line detected by the time shift detection error detection unit as having a time shift detection error is corrected by the time shift detection error correction unit or corrected by the replaced time shift. A delay amount and / or a delay amount control unit for controlling each predetermined delay amount in the delay means are provided.

The received signal used in each of the above (1) to (5) may be the received signal before passing through the delay means or the received signal after passing through the delay means. Good. Further, in the present invention, the above feature amount is not limited to a specific feature amount obtained by a specific calculation, but when the feature amounts that can be adopted in the present invention are listed,
For example, (a) the number of points in the curve representing the relative time lag of the received signal that crosses the time lag 0 (b) the average deviation of the relative time lag of the received signal from the time lag 0 (calculation process Above equivalent, including the sum of deviations). (C) Absolute value of quadratic coefficient when a curve representing relative time lag of received signals is approximated to a quadratic function. (D) Relative time lag of received signals corresponding to two scanning lines. Average deviation (including sum of deviations). (E) The correlation coefficient between the curves representing the relative time lag of the received signal between the two scanning lines can be mentioned.

Further, in the present invention, the method of setting the threshold value used in the time shift detection error detection unit is not particularly limited, and may be prepared in advance as a constant, for example, or f) An average value of the feature amount may be obtained over a plurality of scanning lines and the average value may be used as a threshold value, or (g) the feature amount of the feature amount may be spread over a plurality of scanning lines. Alternatively, the average value and the degree of variation may be obtained, and the threshold value may be set based on the average value and the degree of variation.

Further, in the present invention, the algorithm for correcting the time lag in the time lag detection error correction section is not limited to a specific algorithm, and various algorithms can be adopted, but examples thereof are shown. For example, (h) a curve representing the relative time shift of the received signal with respect to the scanning line detected as having the time shift detection error by the time shift detection error detection unit is approximated to a quadratic function, Correction is performed so that each point on the function has zero time lag. (I) A curve representing the relative time shift of the received signal is approximated to the first quadratic function for the scanning line detected by the time shift detection error detection unit as having the time shift detection error, and the time shift detection error is detected. A curve representing the relative time lag of the received signal for the scanning line detected as having no time lag detection error in the detector is approximated to a second quadratic function,
The time lag is corrected by each difference between each point on the first quadratic function and each point on the second quadratic function that correspond to each other. (J) The time shift detection error detection unit detects a relative time shift of the received signal of the first scanning line, which is detected by the time shift detection error detection unit, as being not detected by the time shift detection error detection unit. The time lag of the first scanning line is corrected by replacing it with the relative time lag of the received signal of the second scanning line. And the like.

[0021]

In the ultrasonic diagnostic apparatus of the present invention, the time lag is once detected based on the reception signal cut out by the gate section (the above (1)) (the above (2)), and then the detected time lag is detected. On the basis of the above, for example, the feature amounts of the time lags such as (a) to (e) above are obtained, and the feature amounts are obtained from the above (f), (g)
A scanning line having a time shift detection error is detected by comparing it with a threshold value obtained by the above method ((3) above).
Since the time lag is corrected by the method of (j) or the like ((4) above), the delay amount control section of (5) above causes the delay amount of the voltage pulse in the transmitting means and / or the received signal of the delay means to be The delay amount is controlled, so that even if there are a plurality of strong reflectors in the vicinity of the scanning line at the same time, the focal point on the transmitting side and / or the focal point on the receiving side are formed correctly, and high resolution and high image quality are achieved. A tomographic image of

[0022]

EXAMPLES Examples of the present invention will be described below. Figure 1
2 is a block diagram of an ultrasonic diagnostic apparatus according to an embodiment of the present invention, and FIG. 2 is a schematic diagram of a received signal waveform. A pulse-shaped voltage at each predetermined timing is applied from the transmission unit 2 toward the electroacoustic transducers 1 arranged in a predetermined direction, whereby the target is focused so as to focus on a predetermined point in the subject (not shown). Ultrasonic waves are emitted toward the inside of the specimen. The ultrasonic wave transmitted into the subject is reflected by each tissue in the subject, and the reflected wave is received by each of the many electroacoustic transducers 1 and converted into each reception signal, which is input to the receiving unit 3. To be done. The receiving section 3 is provided with an A / D converter that converts each input reception signal as an analog signal into a digital signal, and a delay unit that delays each digitized reception signal by a variable delay amount. The delay means delays and outputs each reception signal so that a reception-side focal point that sequentially moves along the scanning line extending into the subject is formed. The respective reception signals delayed by the respective predetermined delay amounts are input to the addition unit 4 and added to each other, and the addition signal after the addition is input to the display unit 10, and the display unit 10 is based on the addition signal. A tomographic image of the subject is displayed. Further, the respective reception signals delayed by the delay means provided in the receiving unit 3 are also input to the gate unit 5. In the gate unit 5, among the received signals as shown in FIG. 2, a received signal in a certain area around the time when the reflected wave from a certain depth in the subject reaches the electroacoustic conversion element 1 is time-shifted. It is cut out as a detection area. The gate unit 5 may detect the time lag based on all the received signals obtained by each electroacoustic transducer 1, but for example, a change in the speed of sound of the human body is caused by the pitch of the acoustic transducers 1 (about 0). Since it is sufficiently larger than 0.5 mm), for example, the time lag may be detected based on the reception signal obtained by every other electroacoustic conversion element 1.

The received signal in the time shift detection area cut out by the gate unit 5 is input to the time shift detection unit 6. The time shift detecting unit 6 detects the time shift using, for example, the cross-correlation method. However, when a tilt amount as shown in FIG. 12 is detected, the tilt is approximated by a straight line and the detected time shift is detected. Of these, the difference from this straight line is obtained as the time shift.

FIG. 3 is a waveform diagram showing the time lag for each scanning line detected by the time lag detector 6 with eight scanning lines as one scanning line group. In which a number of electroacoustic transducer E _0, ..., E _n is the time shift based on the received signal the received signals is obtained by the respective (see FIG. 1) is detected, each block surrounded by a rectangle , each one scan line S _0, S ₁ ..., for each of S _7, the electroacoustic transducer E _0, ..., arrangement of each received signal obtained by the E _n (hereinafter referred to as "device number" .) On the horizontal axis and the relative time lag of each received signal on the vertical axis.

Here, the scanning line group used for detecting an error in the time shift detection shown below is composed of at least two scanning lines. The number of scanning lines in the scanning line group is preferably about several to ten and several so that a large difference does not occur in the time lag between the scanning lines. The time lag shown in FIG. 3 is the time lag after the tilt component as shown in FIG. 12 is removed by the time lag detector 6.

In this figure, a large time lag detection error is seen in the scanning line S ₄ . The time lag found by the time lag detector 6 is input to the time lag detection error detector 8. The time shift detection error detection unit 8 detects a scanning line having a time shift detection error as described below, and the information is input to the time shift detection error correction unit 9.

FIG. 4 is a detailed block diagram of the time shift detection error detection section and the time shift detection error correction section. The information on the time lag for each scanning line, which is obtained by the time lag detector 6 as shown in FIG. 3, is input to the feature amount computing means 81 in the time lag detection error detector 8. This feature amount calculation means 8
In No. 1, the time-lag characteristic amount is obtained based on the input time-lag information. Here, as the feature amount, (a) the number of points at which the curve of the time lag for each scanning line shown in FIG. 3 crosses the time lag 0 (b) the time lag 0 according to the present invention
The sum of absolute values of time lags for each scanning line, which is an example of the average deviation from, or the sum of squares (c) The time lag curve for each scanning line shown in FIG. 3 is approximated to a quadratic function. Absolute value of slope (second-order coefficient) in the case of (d) An example of the average deviation of the relative time lags of the reception signals corresponding to each other between the scanning lines according to the present invention. Difference in time lag (e) Any one of the correlation coefficients between the curves representing the time lag or a combination thereof is adopted. Hereinafter, each of these (a) to (e) will be described in detail.

(A) When the number of points crossing the time lag 0 is used as the characteristic amount, the characteristic amount is obtained by the following method. In the feature amount calculation means 81, the time shifts TE _k , T of the element numbers E _k , E _{k + 1} (0 ≦ k ≦ n) for each scanning line.
The number of times that E _{k + 1} changes from positive to negative or from negative to positive or 0 is counted, and the number (number of times) is set as the feature amount.
Next, the threshold value setting means 82 obtains the average of the number of points that cross the time lag 0, which is the feature amount, across the eight scanning lines S ₀ , S ₁ , ..., S ₇ shown in FIG. , This is the threshold. The obtained feature amount and threshold value are input to the detection error determination means 83, and the detection error determination means 83 determines that the time shift of the scanning line in which the feature amount does not exceed the threshold value is a detection error.

For example, in the scanning lines (other than S ₄ ) in which no detection error occurs in the time lag of FIG. 3, the number of points (6 points) crossing the time lag 0 does not change. However, in the case of the scanning line S _{4 in} which a detection error has occurred, the number of points (3 points) crossing the time shift 0 is reduced. Therefore, the feature amount of the scanning line S ₄ is 3, the feature amount of the other scanning lines is 6, and the threshold value is 5.625. Since the feature amounts of the scan lines S ₄ does not exceed the threshold value, the time shift of the scan line S ₄ is determined that the detection error. (B) When the sum of absolute values of time lag is used as the feature amount, this feature amount is obtained by the following method.

In the feature quantity computing means 81, the following (1)
The sum of the time lags of the scanning lines S _i (0 ≦ i ≦ 7) or the sum of squares TW _i is obtained according to the formula, and the sum or the sum of squares TW _i is used as the feature amount.

[0031]

[Equation 1]

Here, TE _k is the element number E _k (0 ≦ k ≦
It is the time lag of n). Next, the threshold value setting unit 82 obtains the average <T> of the sum of time lags or the sum of squares TW _i , which is the feature amount, and sets the average value <T> as the threshold value.

[0033]

[Equation 2]

The feature amount TW thus obtained
_i (i = 0, 1, ..., 7) and the threshold value <T> are input to the detection error determination means 83, and this detection error determination means 8 is input.
In 3, the time lag of the scanning line where the feature quantity TW _i exceeds the threshold value <T> is determined to be a detection error. If there is a detection error time lag like the scan line S ₄ , this scan line S
Feature value TW ₄ of ₄ becomes a value offset is added as shown by broken lines in FIG. 3, it would exceed the average <T>.
Therefore, the time shift of the scanning line S ₄ is determined to be a detection error.

(C) When the time difference is approximated by a quadratic function
When the absolute value of the quadratic coefficient (slope) of is used as the feature quantity,
The characteristic amount is obtained by the following method. First, the characteristic amount performance
In the calculating means 81, the scanning line S_iTime (0 ≦ i ≦ 7)
Approximate deviation quadratic function Y_iIs required. Y_i= A_iX_i ²+ B_iX_i+ C_i (3) Where A_i, B_i, C_iIs a coefficient. Close to a quadratic function
An example of a similar method is the least squares method. This
Absolute value of the slope of the quadratic function of | A_i| Is set as the feature amount.
Next, the threshold value setting means 82
Absolute value | A_iThe average of <|
It The feature amount thus obtained | A _i│ (i =
0, 1, ..., 7) and the threshold value <A> are detection error determining means.
83 is input to the detection error determining means 83.
Feature amount | A_iTime of scan line where | exceeds threshold value <A>
The shift is determined to be a detection error.

[0036] When the time shift of the detected errors, such as the scanning line S ₄ is present, the absolute value of the slope in the scanning line S ₄ | A ₄ |
Becomes larger and exceeds the threshold value <A>. Therefore, the time shift of the scanning line S ₄ is determined to be a detection error. (D) In the case where the difference in scanning time difference is used as the feature amount,
The feature amount is obtained by the following method.

First, in the feature quantity computing means 81,
According to equation (4), the scanning line spacing S_i, S _{i + 1}(0 ≦ i ≦
6) Time-lag difference TS_iIs calculated and this difference TS
_iIs the feature amount.

[0038]

[Equation 3]

Here, TE _k and TE _{k + 1} are time shifts of the element numbers E _k and E _{k + 1} (0 ≦ k ≦ n). Next, the threshold value setting unit 82 obtains the average <T> of the difference TS _i of the time lag between the scanning lines, which is the feature amount, and sets this average <T> as the threshold value.

[0040]

[Equation 4]

The feature amount TS thus obtained
_i (i = 0, 1, ..., 6) and the threshold value <T> are input to the detection error determination means 83, and in this detection error determination means 83, the scan in which the feature amount TS _i exceeds the threshold value <T>. The time shift of the line is determined to be a detection error. If there is a detection error time lag like the scanning line S _4, the feature values TW ₃ , TW having a very large difference from the scanning lines S ₃ , S ₅ before and after the detection error.
₄ , which exceeds the average <T>. Therefore, the scan line S ₄
The time lag of is determined to be a detection error.

(E) When the correlation coefficient of the time lag between scanning lines is used as the characteristic amount, the characteristic amount is obtained by the following method. First, in the feature amount calculation means 81, the time shift correlation coefficient Γ _i between the scanning lines S _i and S _{i + 1} (0 ≦ i ≦ 6) is calculated by the following equation (6), and this correlation coefficient Γ _i is obtained. It is regarded as a feature amount.

[0043]

[Equation 5]

Here, TE _k and TE _{k + 1} are time shifts of the element numbers E _k and E _{k + 1} (0 ≦ k ≦ n), <TE> and <TE>.
TE ₁ > is an average of TE _k and TE _{k + 1} . Also, S
T and ST ₁ are standard deviations of TE _k and TE _{k + 1} . Next, in the threshold value setting means 82, the average <Γ> of the correlation coefficient Γ _i of the time lag between the scanning lines, which is the feature amount, is obtained by the following equation (7) and set as the threshold value.

[0045]

[Equation 6]

The feature amount Γ _i (i
0, 1, ..., 6) and the threshold value <Γ> are input to the detection error determination means 83, and the detection error determination means 83 scans the feature quantity Γ _i so as not to exceed the threshold value <Γ>. The time shift of the line is determined to be a detection error. If there is a detection error time lag like the scanning line S ₄ , the correlations with the scanning lines S ₃ and S ₅ before and after that are very small feature amounts Γ ₃ and Γ ₄ , and the average <Γ> is not exceeded. Become. Therefore, the time shift of the scanning line S ₄ is determined to be a detection error.

Here, the threshold value setting means 82 in the above (a) to (e) finds the average of the feature amounts, and the average is used as a threshold, but the average of the feature amounts is used as the threshold value. Alternatively, the threshold value setting means 82 may obtain the standard deviation σ of the feature values together with the average of the feature values, and use the average + standard deviation σ (or the average−standard deviation σ) as the threshold value. In this way, when using the threshold value with standard deviation σ added, not only when there is a time shift detection error,
This is particularly effective when there is no time shift detection error.
That is, when there is no detection error in the time lag, each feature amount has variations near the threshold value, and therefore a time lag that is not a detection error may be erroneously determined to be a detection error. Therefore, by setting the average + standard deviation σ (or the average−standard deviation σ) as the threshold value, it is possible to prevent the time lag near the average from being determined as a detection error. The threshold value may be other than the average + standard deviation σ (or average−standard deviation σ), and may be, for example, average + σ / 2 (or average−σ / 2).

Further, the threshold value setting means 82 in the above (a) to (e) may store a fixed initial threshold value. There is almost no difference between the respective feature amounts when the time shift detection error does not occur, and only the feature amounts when the time shift detection error occurs greatly deviate. Therefore, it is possible to save the trouble of setting the threshold value by calculation by setting the threshold value that is larger than each feature amount when the time shift detection error does not occur. For example, when the correlation coefficient is used as the feature amount, it is considered that there is almost no correlation if the correlation coefficient is | Γ | <0.3. Therefore, | Γ | = 0.3 is stored as the threshold value. It is possible to leave it.

[0049] As described above, the detection error detecting unit 8 shift time, the scanning line S ₄ with a detection error deviation time is detected, that the scanning lines S ₄ are shift detection error time, and Information necessary for correcting the time shift of the scanning line S ₄ is input to the time shift detection error correction unit 9. Figure 4
In the time shift detection error correction unit 9 in the embodiment shown in
The time shift replacement unit 91 replaces the time shift of the scanning line S ₄ that is determined to be a detection error by the time shift detection error detection unit 8 with the time shift of the scan line that is not a detection error. Here, it is preferable that the time lag to be replaced is the time lag of the scan line that is closest to the scanning line in which the detection error has occurred and is not the time lag detection error. For example, in FIG. 3, the scan line S
The time lag of ₄ is replaced with the time lag of scan line S ₃ , and so on. Since the time lag between adjacent scanning lines does not change significantly, by replacing it, it is possible to obtain an effect substantially equivalent to that when the time lag of the detection error is directly corrected.

FIG. 5 shows the time shift detection shown by the block in FIG.
FIG. 6 is a detailed block diagram of another example of the output error correction unit.
6 is a supplementary explanatory diagram of the time shift detection error correction unit shown in FIG.
The graph on the upper left of FIG. 6 shows the time lag of the detection error.
The graph shows the time lag with no detection error.
It The vertical axis is the time difference, and the horizontal axis is the element number (E₀~ E _n)
It represents. The approximate curve calculating means 92A shown in FIG.
Then, the time shift detection error detection unit 8 described above.
Judgment as a time shift detection error using the methods of (a) to (e)
The time lag is calculated by the least squares method,
Next function Y_aiIs approximated by.

Y _ai = A _a (E _i ) ² + B _a E _i + C _a (0 ≦ i ≦ n) (8) Similarly, the approximate curve calculation means 92B detects a scan line in which no time shift detection error has occurred. The time shift is approximated to the quadratic function Y _bi by the method of least squares. Y _bi = A _b (E _i ) ² + B _a X + C _b (0 ≦ i ≦ n) (9) In the time shift correction means 93, the difference between the quadratic function Y _a and the quadratic function Y _b shown in FIG. ΔY _i ΔY _i = Y _bi −Y _ai (10) is removed from the detection error time shift. That is, the time lag of the detection error is corrected so that the quadratic function Y _ai matches the quadratic function Y _bi , and the corrected time lag is adopted as a new time lag.

NTE _i = TE _i −ΔY _i (11) Here, TE _i is the time shift of the element number E _i (0 ≦ i ≦ n), and NTE _i is the time shift of the corrected element number E _i. Shows. The example shown in FIGS. 5 and 6 is an example of a method of correcting the detection error time shift, and the time shift after the correction is substantially equal to the time shift caused by the non-uniformity of the sound velocity.

Here, the time shift of the scanning line in which the time shift detection error is not used, which is used in the time shift detection error correction unit 9, is the time of the scan line closest to the scanning line in which the time shift detection error has occurred. It is preferable to use the shift. This is because the time lag between adjacent scan lines does not change significantly as described above. The method shown in FIGS. 5 and 6 approximates both the time lag of the scanning line determined as the time lag detection error and the time lag of the scan line in which the detection error does not occur to a quadratic function, and Although the quadratic functions were corrected so as to match, for the sake of simplicity, only the time lag of the scanning line determined as the time lag detection error is approximated to a quadratic function, and this quadratic function is a straight line with no time lag You may correct so that it may correspond. In this way, after being corrected (including replacement) as necessary in the time shift detection error correction unit, the time shift information corrected as necessary is input to the delay amount control unit 7 (see FIG. 1). Then, the delay amount control unit 7 delays the voltage pulse generated by the transmission unit 2 so that the transmission-side focus and the reception-side focus are correctly formed in the subject based on the input time deviation information. , And the delay amount of the received signal in the delay means in the receiver 3 is controlled.

In the embodiment shown in FIG. 1, both the focal point on the transmitting side and the focal point on the receiving side are corrected and controlled, but the present invention is applied to either the focal point on the transmitting side or the focal point on the receiving side. can do. FIG. 7 is a block diagram of an ultrasonic diagnostic apparatus according to another embodiment of the present invention. Only differences from the embodiment shown in FIG. 1 will be described.

Each received signal obtained by receiving the reflected ultrasonic waves at each of the electroacoustic transducers 1 passes through the matrix switch 11 as an analog signal and is input to the reception adder 12. The reception addition unit 12 is provided with a delay line having a large number of taps, and the matrix switch controlled by the delay amount control unit 7 determines from which tap each reception signal is input to the delay line. The respective reception signals input to the delay line are delayed by this delay line by a delay amount corresponding to the input taps, and currents are added to each other. The addition signal obtained by this current addition is input to the display unit 10.

In the configuration shown in FIG. 7, the reception signal before the phasing process, that is, the reception signal before being delayed by the reception addition unit 12 is input to the gate unit 5. Therefore, in the gate unit 5, as shown in FIG. 8, a parabolic region is cut out as a region for performing correlation calculation for detecting a time shift.

[0057]

As described above in detail, the ultrasonic diagnostic apparatus of the present invention determines the time lag characteristic amount based on the time lag by using, for example, the methods (a) to (e). Since the characteristic amount is obtained and compared with the threshold value obtained by the method such as (f) or (g), a scanning line having a time shift detection error is detected. Further, for the scanning line having the detected time shift detection error, the time shift is corrected by the methods (h) to (j) or the like, so that, for example, a plurality of strong reflectors may be present near the scanning line at the same time. Even in such a case, the focal point on the transmitting side and / or the focal point on the receiving side are correctly connected, and therefore, a high-resolution and high-quality tomographic image can be obtained.

[Brief description of drawings]

FIG. 1 is a block diagram of an ultrasonic diagnostic apparatus according to an embodiment of the present invention.

FIG. 2 is a schematic diagram of a received signal waveform.

FIG. 3 is a waveform diagram showing a time lag for each scanning line detected by a time lag detecting unit, with eight scanning lines as one scanning line group.

FIG. 4 is a detailed block diagram of a time shift detection error detection unit and a time shift detection error correction unit.

5 is a detailed block diagram of another example of the time shift detection error correction unit indicated by a block in FIG.

FIG. 6 is a supplementary explanatory diagram of the time shift detection error correction unit shown in FIG.

FIG. 7 is a block diagram of an ultrasonic diagnostic apparatus according to another embodiment of the present invention.

FIG. 8 is a diagram showing a received signal before a phasing process.

FIG. 9 is an explanatory diagram of a method of receiving focus.

FIG. 10 is a diagram showing reception focus when the sound velocity is not uniform.

FIG. 11 is an explanatory diagram of a method of correcting a time shift using a cross correlation method.

FIG. 12 is a diagram for explaining a problem that occurs when a strong reflector exists near a scanning line.

FIG. 13 is a diagram schematically showing a received signal when two strong reflectors are present near the scanning line.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 electroacoustic conversion element 2 transmitter 3 receiver 4 adder 5 gate 6 time lag detector 7 delay amount controller 8 time lag detection error detector 81 feature amount calculator 82 threshold value setter 83 detection error determiner 9 time deviation detection error correction section 91 time deviation replacement means 92A approximate curve calculation means 92B approximate curve calculation means 93 time deviation correction means 10 display section 11 matrix switch 12 reception addition section

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Shijo Menjo 1015 Kamiodanaka, Nakahara-ku, Kawasaki-shi, Kanagawa Fujitsu Limited

Claims (11)

[Claims] 1. A transmission unit generates a voltage pulse delayed by a predetermined delay amount so that an ultrasonic wave transmitted into a subject forms a transmission-side focal point at a predetermined point in the subject. Applying to each of a number of electroacoustic transducers arranged in a predetermined direction, by receiving at each of the plurality of electroacoustic transducers ultrasonic waves transmitted in the subject and reflected in the subject Obtaining a reception signal, so that the reception side focus is formed along the scanning line extending in the subject,
The multiple reception signals are each delayed by a predetermined delay amount using a delay amount variable delay unit and added together to obtain an addition signal, and a tomographic image of the subject is displayed based on the addition signal. In the ultrasonic diagnostic apparatus, a gate unit that cuts out the reception signal corresponding to a predetermined depth region in the subject as a time shift detection region, and the reception signal in the time shift detection region cut out by the gate unit On the basis of the time lag detection unit that detects a relative time lag of the received signal, a feature amount of the time lag is calculated based on the time lag detected by the time lag detector, and the feature amount is set to a predetermined value. A time shift detection error detection unit that detects a scanning line having a time shift detection error by comparing with a threshold value, and a scan line that has a time shift detection error detected by the time shift detection error detection unit. Regarding the inspection line, regarding the time deviation detection error correction unit that corrects or replaces the time deviation detected by the time deviation detection unit, and the scanning line detected by the time deviation detection error detection unit as having no time deviation detection error, The time lag detected by the time lag detector and the scanning line detected by the time lag detection error detector as having the time lag detection error are corrected or replaced by the time lag detection error corrector. A delay amount control unit that controls each of the predetermined delay amounts in the transmitting unit and / or each of the predetermined delay amounts in the delay unit so as to correct with a time lag. Sound wave diagnostic equipment. 2. The ultrasonic diagnostic apparatus according to claim 1, wherein the characteristic amount is the number of points across a time lag of 0 in a curve representing a relative time lag of the received signal. 3. The ultrasonic diagnostic apparatus according to claim 1, wherein the characteristic amount is an average deviation of relative time lags of the received signals from zero time lag. 4. The ultrasonic diagnostic apparatus according to claim 4, wherein the characteristic amount is an absolute value of a quadratic coefficient when a curve representing a relative time shift of the received signal is approximated to a quadratic function. 5. The ultrasonic diagnostic apparatus according to claim 1, wherein the characteristic amount is an average deviation of relative time lags of the received signals corresponding to two scanning lines. 6. The ultrasonic diagnostic apparatus according to claim 1, wherein the characteristic amount is a correlation coefficient between curves representing relative time lags of the reception signals of two scanning lines. 7. The time shift detection error detection unit obtains an average value of the feature amount over a plurality of scanning lines and uses the average value as the threshold value. The ultrasonic diagnostic apparatus according to claim 1. 8. The time shift detection error detection unit obtains an average value and a degree of variation of the characteristic amount over a plurality of scanning lines, and based on the average value and the degree of variation, the average value and the degree of variation are calculated. The ultrasonic diagnostic apparatus according to claim 1, wherein a threshold value is determined. 9. A quadratic curve representing a relative time lag of the received signal with respect to a scanning line detected by the time lag detection error detector as having a time lag detection error by the time lag detection error correction unit. The ultrasonic diagnostic apparatus according to claim 1, wherein the ultrasonic diagnostic apparatus approximates a function and performs correction such that each point on the quadratic function has zero time lag. 10. A first curve representing a relative time shift of the received signal with respect to a scanning line in which the time shift detection error correction unit detects a time shift detection error in the time shift detection error detection unit. To a second quadratic function which is a curve representing the relative time lag of the received signal with respect to the scanning line detected by the time lag detection error detection unit as having no time lag detection error. The time shift is corrected by each difference between each point on the first quadratic function and each point on the second quadratic function which are approximated and correspond to each other. 1. The ultrasonic diagnostic apparatus according to 1. 11. The relative time shift of the received signal of the first scanning line, which is detected by the time shift detection error detection unit as having the time shift detection error, by the time shift detection error correction unit. A method for correcting the time shift of the first scanning line by replacing it with the relative time shift of the received signal of the second scanning line detected by the shift detection error detection unit as having no time shift detection error. The ultrasonic diagnostic apparatus according to claim 1, wherein: