KR100830152B1 - Method for forming ultrasound image by decreasing decorrelation of elasticity signal - Google Patents

Abstract

The present invention provides an ultrasound image forming method for realizing an image by reducing the uncorrelation of an elastic image signal that changes with time or space. In particular, the present invention provides an ultrasound image forming method for realizing an elastic image by compensating for a phase change generated by moving one signal to reduce uncorrelation between signals before and after compression in an elastic imaging method.

Ultrasound, Imaging, Uncorrelated Diagram, Delay Time, Estimation, Shift, Phase Shift, Compensation

Description

A method of forming an ultrasound image by reducing the uncorrelatedness of an elastic image signal {METHOD FOR FORMING ULTRASOUND IMAGE BY DECREASING DECORRELATION OF ELASTICITY SIGNAL}

1 is a graph showing an ultrasonic signal before and after compression.

2 is a schematic diagram showing a delay signal model for displacement calculation.

3A is a graph showing received signals before and after compression in adjacent windows.

3B is a graph showing the signal after compression shifted by the estimated delay time together with the signal before compression.

* Description of the reference numerals for the main parts of the drawings *

101: transducer

102: ultrasonic signal received before compression

103: ultrasonic signal received after compression

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an ultrasound image forming method, and more particularly, to an ultrasound image forming method for reducing displacement estimation error by reducing uncorrelation of an elastic image signal before and after compression.

The ultrasound image forming apparatus is a device for transmitting ultrasound toward an object to be diagnosed and forming and displaying an image of the object from reflected ultrasound, and is widely used in the medical field.

Ultrasound images are often represented as B-modes using a reflection coefficient that varies with impedance differences between tissues. However, it is difficult to observe the portion of the surrounding tissues, such as tumors or cancer tissues, where the reflection coefficient is not significantly different from the ultrasound images. As such, tissues having no difference in reflection coefficient may be expressed by elastic imaging to image mechanical characteristics. Elastic imaging is helpful in diagnosing lesions by imaging the mechanical properties of tissues that cannot be diagnosed in B-mode imaging.

In the elastic imaging method, cross correlation coefficients are obtained using RF received data input from an ultrasonic diagnostic apparatus, and autocorrelation coefficients are calculated using complex baseband signals, and displacements are calculated from phase differences. Can be divided into The latter has the advantage that the calculation speed can be increased by using a baseband signal to calculate a smaller amount of data than using RF received data. However, since the value calculated through autocorrelation is not a value for time but a value of phase, it is necessary to change it back to time. The center frequency of the ultrasonic transmission signal is used to change the phase value to a time value. Since the center frequency depends on the depth of the object, an error occurs when a fixed center frequency is used. In addition, when the phase is calculated by autocorrelation, aliasing occurs when the phase difference between the two signals is larger than 1/2 of the transmission ultrasound wavelength, and thus a process for compensating for this is additionally required. In addition, since the phase difference increases as the depth of the object increases, the shape of the signal to be compared is changed, and thus the error increases because data having large de-correlation is used.

In order to reduce the error caused by the use of large uncorrelated data in a relatively deep area, Korean application 2005-109477 entitled "Method of Reducing the Uncorrelation of Elastic Image Signals to Form Ultrasound" Image In the shallow region, the delay time is estimated by the phase difference between the two signals before and after compression, and then the delay time is calculated by moving one signal by the estimated delay time in the direction of decreasing the delay time in the relatively deep region. Then, a method of determining the final delay time by the sum of the estimated delay time and the delay time calculated after the movement is proposed.

However, the above-described prior art has a problem in that it does not reflect the phase change generated by the signal movement. In order to compensate for the abnormal change caused by the movement, it is possible to consider the method of calculating the displacement by repeatedly moving the signal so that the phase becomes zero, but in this case, the calculation time is increased by the repeated signal movement. .

The present invention provides an ultrasonic image forming method for reducing displacement estimation error by reducing the uncorrelation of an elastic image signal. In particular, the present invention, in the elastic imaging method for measuring the rigidity of the tissue, the ultrasound to implement the elastic image by compensating for the phase change generated by moving one signal in order to reduce the non-correlation between the signals before and after compression It provides an image forming method.

The ultrasound image forming method according to the present invention includes a first received signal obtained from an ultrasound signal reflected from an uncompressed object and a second received signal obtained from an ultrasound signal reflected from a compressed object—the first received signal and the second received signal. Receiving a delay time between the first and second received signals, wherein the first and second received signals are divided into a plurality of areas; Estimating a delay from a phase difference between the first and second received signals in a first region; Moving any one selected from the first received signal and the second received signal of a second area adjacent to the first area by the estimated delay; After the movement is completed, calculating a displacement between the moved signal and the unmoved signal using the correlation in the second region, wherein the displacement is calculated by compensating for the phase change according to the movement; Calculating a delay between the first received signal and the second received signal according to the compression in the second region based on the estimated delay and the displacement; And forming an ultrasound image of the object based on the calculated delay.

In addition, the ultrasound image forming method according to the present invention, the first received signal obtained from the ultrasonic signal reflected from the uncompressed object and the second received signal obtained from the ultrasonic signal reflected from the compressed object-the first received signal and The delay between the second received signal varies according to the depth of the object, the first received signal and the second received signal are divided into a plurality of regions, and the first received signal and the second received signal are represented by a phase function. Obtaining-; Estimating a delay from a phase difference between the first received signal and the second received signal in the first region; Moving any one selected from the first received signal and the second received signal of a second area adjacent to the first area by the estimated delay; After the movement is completed, calculating a displacement between the moved signal and the unmoved signal using the correlation in the second region, wherein the displacement is calculated by compensating for the phase change according to the movement; Obtaining an instantaneous frequency by differentiating a phase function of any one of the first and second received signals; Calculating a delay between the first received signal and the second received signal according to the compression in the second region using the estimated delay, the displacement, and the instantaneous frequency; And forming an ultrasound image of the object based on the calculated delay.

Hereinafter, embodiments of the present invention will be described.

First, before applying a stress, that is, a force applied per unit area, that is, stress, an ultrasonic wave is applied to an object to obtain an RF reference signal (first received signal), and a stress is applied to the surface of the object to compress the object while compressing the RF received signal. (Second received signal) is obtained.

1 shows the shape of a received signal before and after compression, that is, before and after applying stress. Applying compression to the object moves the reflectors in the object in the compression direction. As a result of the movement of the reflector, the movement of the received signal appears as compared with before the compression. Therefore, the displacement of the medium can be obtained by calculating the movement, or delay, between the two signals. Since the displacement varies depending on the rigidity of the object, the displacement may be used to determine the characteristics of the medium.

On the other hand, as shown in Figure 1, the movement of the signal due to stress is small in the near to the transducer 101, the displacement accumulates in the distant place, the greater the movement between signals.

When a single pressure is applied to the tissue, the degree of deformation varies depending on the rigidity of the tissue. Therefore, by calculating the displacement of the medium with respect to the force applied from the outside or the inside, and calculating the slope by differentiating the function of the displacement, the strain, that is, the strain, can be obtained. An elastic image is constructed based on this strain value.

In the embodiment of the present invention, in order to calculate the displacement, the RF signal before and after compression is demodulated and converted into an I / Q signal of a base band, and autocorrelation is obtained to calculate a phase difference.

First, a model for estimating delay time will be described. As shown in FIG. 2, the model for delay estimation models a delay time due to compression using a global pass filter having a linear delay. When the ultrasonic transmission center frequency is ω ₀ and the amplitude and phase over time are r (t) and φ (t), respectively, the received signal x ₁ (t) before compression and the signal x ₂ (t) after compression are respectively. It is defined as Equation 1 and Equation 2 below.

In Equation 2, t _g and t _p represent group delay and phase delay due to compression, respectively. Assuming that the group delay t _g and the phase delay t _p are equal, the baseband complex signal after demodulation of the received signal x ₁ (t) before compression and the signal x ₂ (t) after compression is represented by the following equations (3) and (4). same.

The phase difference Δ Φ ₀ (displacement due to compression) of two signals in a finite interval, that is, a constant magnitude correlation window, is expressed by Equation 5 from the correlation calculation.

를 테일러 급수(Taylor series)로 1차항까지 전개하여 다음의 수학식 6을 얻는다.In Equation 5, 'arg' is a function for calculating a phase, and '<·>' is a function for calculating a correlation. Of equation (5)

Is expanded to the first term in the Taylor series to obtain the following equation (6).

Putting this into Equation 5, the phase difference Φ ₀ is expressed as Equation 7 below.

Summarizing this for the group delay t _g , we obtain Equation 8.

는 위상의 미분이므로 수학식 8에 보인 바와 같이

는 기저대역 신호의 순시주파수(instantaneous frequency) ω_B(t)에 해당한다. 따라서

는 수학식 9와 같이 표현될 수 있다.

Since is the derivative of the phase, as shown in (8)

Corresponds to the instantaneous frequency ω _B (t) of the baseband signal. therefore

May be expressed as in Equation (9).

In Equation 9, 'T' is a sampling time interval. If ω ₀ ≫ ω _B (t), the group delay t _g can be expressed by the following equation (10).

However, since the ultrasonic signal has a wide frequency bandwidth and the frequency changes according to the depth of the object, an error occurs when the denominator is fixed to the constant ultrasonic transmission center frequency ω ₀ . Such an error can be reduced by considering the instantaneous frequency component of the baseband signal.

When the object is compressed, as the depth of the object increases, the displacement of the entire object increases, so that the shape of the signal to be compared varies greatly. As a result, the error is increased because the data is calculated using large uncorrelated data. In order to overcome this, the delay time is estimated by the phase difference of two signals in the region where the depth of the object is relatively low, then one signal is first moved by the estimated delay time in the direction of decreasing the delay time, and then the delay time is calculated again. . The final delay time value is given as the sum of the estimated delay time and the delay time calculated after the movement. This will be described in detail with reference to FIGS. 3A and 3B.

In FIGS. 3A and 3B, the windows w (n) and w (n + 1) respectively indicate received signals (first and second received signals) before and after compression when the object is relatively shallow and deep. This window is visible. Depth is the relative depth between the reflector (tissue) and the probe in the object. Meanwhile, the windows w (n) and w (n + 1) may be windows showing received signals before and after compression over time.

As can be seen from the comparison of the two windows, it can be seen that the phase difference becomes larger when the depth of the object is relatively deep (when time passes). If the phase difference is π or more, aliasing occurs. If the two windows are adjacent, the phase difference of the signal is not large so that aliasing can be effectively prevented. When the delay time due to the compression in the window w (n) of FIG. 3A is estimated as t _d (n), the delay time due to the compression in the window w (n + 1) adjacent to it is t _d (n + 1). Assume

As shown in FIG. 3B, when the received signal before compression is moved using linear interpolation by the estimated delay time t _d (n), the phase difference of the signal in the window w (n) becomes close to zero as the movement is performed. . Although FIG. 3B shows an example of moving the pre-compression signal, the moving signal may be both a pre-compression signal and a post-compression signal. When the time delay of the signal in w (n + 1) calculated after the movement is τ (n + 1), τ (n + 1) becomes smaller than the time delay t _d (n + 1) before the movement. Accordingly, the correlation between the received signals before and after compression is increased to obtain a noise reduction effect, and the phase difference is also reduced to prevent the occurrence of aliasing.

이 반영되었으므로, 최종 지연시간은 τ(n+1)과

의 합에 기초하여 계산된다. 즉, 다음의 수학식 11과 같이 비상관도가 큰 윈도우 w(n+1) 내의 압축에 따른 지연시간 t_d(n+1)를 그대로 반영하여 영상을 형성하지 않고, 대상체의 깊이가 상대적으로 얕은 영역인 윈도우 w(n)내의 지연시간 t_d(n)에 대한 지연시간 차이 τ(n+1)를 반영하여 영상을 형성한다.The delay time already estimated at the time delay τ (n + 1) of the window w (n + 1) after the signal shift.

Is reflected, the final delay time is τ (n + 1) and

Is calculated based on the sum of. That is, as shown in Equation 11 below, the image is not formed by reflecting the delay time t _d (n + 1) due to the compression in the large uncorrelated window w (n + 1), and the depth of the object is relatively shallow. The image is formed by reflecting the delay time difference τ (n + 1) with respect to the delay time t _d (n) in the window w (n).

As described above, when the estimated delay time t _d is approximated to δ, the displacement τ (n + 1) in the window w (n + 1) after the movement is different from the group delay t _g and δ due to compression. (Τ (n + 1) = t _g -δ).

In order to reduce the uncorrelation of the ultrasonic signal before and after compression, the delay time (phase difference) of the signal before and after compression is estimated, and then one of the two signals is moved in the direction in which the delay time (phase difference) decreases. According to the group delay tg according to the compression is obtained.

The signal shifted by δ in the direction of canceling the group delay t _g from the baseband complex signal x _b2 (t) after demodulation of the received signal after compression as expressed by Equation 4 is expressed by Equation 12 below.

The following equation (13) is obtained from the baseband complex signal x _b1 (t) represented by equation (3) and x _b2 (t + δ) expressed by equation (12) after demodulation of the pre-compression received signal.

When ΔΦ _δ is expressed as an approximation equation using a Taylor series, it is expressed by Equation 14 below.

Equation (15) below summarizes equation (14 ₎ with respect to t _g in group delay.

Equation 15 shows a group delay as the signal after compression in the window w (t + 1) is moved by the estimated delay time of the received signal before and after compression in the window w approximated by δ. In case of δ = 0, Equation 15 is equal to Equation 8, in which the signal is not moved. In addition, when δ = t _g , the group delay is expressed from the following equation (16).

이므로 이를 상관도(correlation) 계산에 반영하면, 수학식 14는 다음의 수학식 17과 같이 표현된다.That is, even if the displacement is compensated for by the movement of the signal, the phase difference does not become zero. The reason why the phase difference does not become zero is that the signal is shifted by δ in the direction of canceling the group delay t _g from the baseband complex signal x _b2 (t) after demodulation of the received signal after compression, as expressed by Equation (4). This is because they do not compensate for the changing phase. Therefore, in order for the phase difference to be zero after the movement of the signal, it is necessary to compensate for the phase change occurring in accordance with the movement of the signal. The phase change as the signal moves by δ

Since this is reflected in the correlation calculation, Equation 14 is expressed as Equation 17 below.

From Equation 17, the following Equation 18 is obtained.

이 된다. 그러나,

이면

는 0이 되지 않으므로,

로부터 (t_g-δ) 및 t_g를 계산할 수 있다. 수학식 17과 18을 이용하면 작은 변위를 계산하게 되므로 신호의 엘리어싱(aliasing)을 피할 수 있고, 종래에 위상을 보상하기 위해, 즉 위상이 0(zero)가 되도록 신호를 반복적으로 이동시키는 과정을 생략할 수 있다. 또한, 신호의 이동에 따른 신호 크기(amplitude)의 변화와 위상의 변화를 같이 고려함으로써 계산의 정확도를 높일 수 있다.When δ = t _g in Equation 18

Becomes But,

Back side

Does not become 0, so

(T _g -δ) and t _g can be calculated from By using Equations 17 and 18, small displacements are calculated, so that aliasing of signals can be avoided, and the process of repeatedly moving a signal to compensate for the phase, that is, the phase becomes zero, is conventionally used. Can be omitted. In addition, the accuracy of the calculation can be improved by considering the change in signal amplitude and the change in phase according to the movement of the signal.

Hereinafter, a process of forming an ultrasound image by reducing the uncorrelation of the elastic image signal will be described.

According to an embodiment of the present invention, the first received signal obtained from the ultrasonic signal reflected from the uncompressed object and the second received signal obtained from the ultrasonic signal reflected from the compressed object are received. The delay time between the first and second received signals varies according to the depth of the object, and the first and second received signals are divided into a plurality of regions. The first received signal and the second received signal are signals obtained from transmission ultrasonic waves whose center frequency is ω ₀ . Further, the group delay t _g is generated between the first received signal and the second received signal by the compression. The first received signal and the second received signal are represented by Equations 3 and 4, respectively.

The delay is estimated from the phase difference between the first received signal and the second received signal in the first region such as the window w (n) of FIG. 3B. The estimated delay corresponds to the phase difference δ of the signal before and after compression in the window w (n) of FIG. 3B.

를 보상하여 변위를 계산한다. 이에 따라 계산된 변위는 전술한 수학식 17과 같이 표현된다.Subsequently, any one selected from the first reception signal and the second reception signal in the second area adjacent to the first area as shown in the window w (n + 1) of FIG. 3B is moved by the estimated delay δ. After the movement is completed, the displacement between the moved signal and the unmoved signal is calculated using the correlation shown in Equation 5 in the second region. At this time, the phase change according to the movement

Compute the displacement by compensating for it. The displacement calculated according to this is represented by Equation 17 described above.

Based on the estimated delay δ and the displacement, a delay between the first received signal and the second received signal according to compression in the second region is calculated. The calculated delay between the first and second received signals due to compression corresponds to a group delay t _g obtained from equation (17).

Next, an ultrasound image of the object is formed based on the calculated delay (group delay t _g ).

The first region described above corresponds to a region having a shallower depth of the object than the second region.

After calculating the displacement according to another embodiment of the present invention, the instantaneous frequency is calculated by differentiating a phase function of any one of the first and second received signals based on Equation (9). That is, the instantaneous frequency is calculated based on the correlation between the first received signal or the second received signal during the sampling time T. In this case, the delay between the first and second received signals due to compression is calculated using the estimated delay, displacement and instantaneous frequency.

According to the present invention made as described above, by using different ultrasonic transmission frequencies according to the depth of the medium, it is possible to reduce the error occurring in the process of converting the phase to a time value. In addition, since the data is moved and the correlation of the data is increased and a small displacement is calculated, aliasing of the signal can be avoided. Accurate displacement can be calculated by considering the change in signal amplitude and the change in phase due to data movement.

Claims (7)

The first received signal obtained from the ultrasonic signal reflected from the uncompressed object and the second received signal obtained from the ultrasonic signal reflected from the compressed object—the delay time between the first received signal and the second received signal is determined by the depth of the object. Receiving the first and second received signals divided into a plurality of areas; Estimating a delay from a phase difference between the first and second received signals in a first region; Moving any one selected from the first received signal and the second received signal of a second area adjacent to the first area by the estimated delay; After the movement is completed, calculating a displacement between the moved signal and the unmoved signal using the correlation in the second region, wherein the displacement is calculated by compensating for the phase change according to the movement; Calculating a delay between the first received signal and the second received signal according to the compression in the second region based on the estimated delay and the displacement; And And forming an ultrasound image of the object based on the calculated delay. The method of claim 1, And the first region corresponds to a region having a shallower depth of the object than the second region. The method according to claim 1 or 2, The first received signal and the second received signal have a center frequency of ω ₀ and amplitude and phase according to time are r (t) and φ (t), respectively. Obtained from transmit ultrasound, The group delay t _g and the phase delay t _p between the first received signal and the second received signal are generated by the compression, The group delay t _g is equal to the phase delay t _p , The first received signal and the second received signal are represented by Equation 1 and Equation 2 as x _b1 (t) and x _b2 (t), respectively, (Equation 1)

, (Equation 2)

The estimated delay is δ, The displacement of the first received signal and the second received signal in the second region is ΔΦ ₀ represented by Equation 3 below, (Equation 3)

Phase change according to the movement

The displacement of the first received signal and the second received signal in the second region is reflected by Equation 4 below.

Represented by (Equation 4)

The group delay t _g and the estimated delay δ satisfy the following Equation 5, (Equation 5)

Ultrasound image forming method. The first received signal obtained from the ultrasonic signal reflected from the uncompressed object and the second received signal obtained from the ultrasonic signal reflected from the compressed object—the delay between the first received signal and the second received signal is the depth of the object. And the first received signal and the second received signal are divided into a plurality of areas, the first received signal and the second received signal represented by a phase function. Estimating a delay from a phase difference between the first received signal and the second received signal in the first region; Moving any one selected from the first received signal and the second received signal of a second area adjacent to the first area by the estimated delay; After the movement is completed, calculating a displacement between the moved signal and the unmoved signal using the correlation in the second region, wherein the displacement is calculated by compensating for the phase change according to the movement; Obtaining an instantaneous frequency by differentiating a phase function of any one of the first and second received signals; Calculating a delay between the first received signal and the second received signal according to the compression in the second region using the estimated delay, the displacement, and the instantaneous frequency; And And forming an ultrasound image of the object based on the calculated delay. The method of claim 4, wherein And the first region corresponds to a region having a shallower depth of the object than the second region. The method according to claim 4 or 5, The first received signal and the second received signal have a center frequency of ω ₀ and amplitude and phase according to time are r (t) and φ (t), respectively. Obtained from transmit ultrasound, The group delay t _g and the phase delay t _p between the first received signal and the second received signal are generated by the compression, The group delay t _g is equal to the phase delay t _p , The first received signal and the second received signal are represented by Equation 1 and Equation 2 as x _b1 (t) and x _b2 (t), respectively, (Equation 1)

, (Equation 2)

The estimated delay is δ, The displacement of the first received signal and the second received signal in the second region is ΔΦ ₀ represented by Equation 3 below, (Equation 3)

Phase change according to the movement

The displacement of the first received signal and the second received signal in the second region is reflected by Equation 4 below.

Represented by (Equation 4)

The group delay t _g and the estimated delay δ satisfy the following Equation 5, (Equation 5)

Ultrasound image forming method. The method of claim 6, The instantaneous frequency is calculated based on the correlation between the first received signal or the second received signal during the sampling time (T).

Images