JP3544722B2 - Ultrasound diagnostic equipment - Google Patents

Description

[0001]
[Industrial applications]
The present invention relates to an ultrasonic diagnostic apparatus that scans a two-dimensional region of a subject with an ultrasonic wave, detects an obtained received signal, and generates a tissue tomographic image, and in particular, discriminates an infarcted myocardium from a normal myocardium. It relates to a new technology for imaging.
[0002]
[Prior art]
Ultrasonic waves propagate while repeating reflection and refraction at the boundary between two media having different acoustic impedances. The reflection intensity depends on the difference in acoustic impedance between the two media. Ultrasound imaging is a method in which the amplitude of a reflected signal is converted into brightness and displayed. If the amplitude of the reflected signal is small, that is, if the difference in acoustic impedance is small, the change in brightness (density change) between the two media is small. The accuracy of identifying two media having a small difference in acoustic impedance on an ultrasonic tomographic image often depends on the knowledge and experience of a diagnostic physician, and is poor in objectivity.
[0003]
Such a case applies, for example, to an ultrasound diagnosis of myocardial infarction. That is, the difference in acoustic impedance between the infarcted myocardium and the normal myocardium is small, and it is difficult to discriminate both on an ultrasonic tomographic image and the objectivity is poor.
[0004]
[Problems to be solved by the invention]
An object of the present invention is to provide an ultrasonic diagnostic apparatus capable of characteristically imaging an infarcted myocardium differently from a normal myocardium.
[0005]
[Means for Solving the Problems]
According to a first aspect of the present invention, an ultrasonic wave that scans a two-dimensional or three-dimensional region of a subject with an ultrasonic wave to repeatedly receive one frame of a received signal, and generates an image by detecting the received signal. In the diagnostic apparatus, a received signal in a first local area of a first frame and a plurality of second local areas spatially adjacent to the first local area in a second frame different in scanning timing from the first frame. A feature that calculates a cross-correlation coefficient between each of the received signals in the region, and indicates a similarity of the first local region to the second local region using a maximum value of the calculated cross-correlation coefficients. A feature quantity two-dimensional distribution generating means for defining a quantity and generating a two-dimensional distribution of the feature quantity while changing the position of the first local region; and a display means for displaying the two-dimensional distribution as a feature quantity image. Do
According to a second aspect of the present invention, an ultrasonic wave that scans a two-dimensional or three-dimensional region of a subject with an ultrasonic wave to repeatedly receive one frame of a received signal, and generates an image by detecting the received signal. In the diagnostic apparatus, a received signal in a first local area of a first frame and a plurality of second local areas spatially adjacent to the first local area in a second frame different in scanning timing from the first frame. Calculating the sum of the absolute values of the signal differences between each of the received signals in the area, and expressing the similarity of the first local area to the second local area using the minimum value of the calculated sums A feature quantity two-dimensional distribution generating means for defining a feature quantity and generating a two-dimensional distribution of the feature quantity while changing the position of the first local region; and a display means for displaying the two-dimensional distribution as a feature quantity image. Preparation And wherein the Rukoto.
[0006]
[Action]
According to the present invention, an infarct portion whose position is simply changed (metastasis) without deforming the tissue in accordance with the periodic contraction of the normal myocardium is defined by the received signal in the first local region of the first frame and the infarct portion of the second frame. The similarity between the received signals in the two local regions can be regarded as high.
[0007]
【Example】
An embodiment of the present invention will be described below with reference to the drawings. Here, a description will be given assuming that a sector electronic scanning method having a narrow shallow visual field and a wide deep visual field, which is suitable for cardiac diagnosis, is employed. However, if a two-dimensional area of a subject can be scanned, a linear electronic scanning method or the like may be used. A convex electronic scanning method or another scanning method may be used.
An ultrasonic probe 1 equipped with a vibrator array for sector scanning in which a plurality of vibrators for reversibly converting an acoustic signal and an electric signal are arranged is connected to a transmission system 2 at the time of transmission, and is connected to a transmission system 2 at the time of reception. Connected to receiving system 3. The transmission system 2 includes a clock generator, a rate pulse generator, a delay circuit provided for each transducer (for each channel), and a pulsar provided for each transducer (for each channel). The clock pulse generated from the clock generator is frequency-divided by a rate pulse generator into, for example, a 5 KHz rate pulse. This rate pulse is distributed and sent to the pulser via each delay circuit. Each delay circuit gives the rate pulse the delay time necessary to focus the ultrasound in a beam and determine the direction of the ultrasound beam (called the scan line direction or azimuth direction). The pulser applies a pulse voltage to the corresponding vibrator at the timing of receiving the rate pulse. As a result, an ultrasonic pulse is emitted from the probe 1 in the azimuth direction corresponding to the delay time. The reflected wave reflected at the boundary of the acoustic impedance in the subject is converted into an electric signal by each transducer of the probe 1 and captured by the receiver 3.
[0008]
A good example of the receiving system 3 is a digital beamformer system as disclosed in Japanese Patent Publication No. 6-14934, which comprises a preamplifier, an analog-to-digital converter, a digital delay circuit, and an adder. The analog reception signal from each transducer is amplified by a preamplifier, converted to a digital signal by an analog-to-digital converter, sent to a digital delay circuit, and used to obtain reception directivity. Is given a reverse delay time for each channel, and further added by an adder. This added digital signal is hereinafter referred to as an RF reception signal. By sequentially changing the above-described transmission / reception delay time control for each repetition of the ultrasonic pulse, a two-dimensional area of the subject can be scanned by the sector to obtain an RF reception signal group for one frame. Further, by repeating such a sector scan, an RF reception signal group can be repeatedly obtained.
[0009]
The output from the receiving system 3 is detected by the signal processing unit 4 and further subjected to logarithmic amplification that substantially expands the dynamic range. As a result, B-mode image (tissue tomographic image) data is generated. The B-mode image data is sent to a display 6 via a digital scan converter (DSC) 5 and is converted into an analog signal.
[0010]
The output from the receiving system 3 is sent to the feature image generating unit 7. The feature amount image generating unit 7 uses a two-frame RF reception signal group having different scan timings and different heartbeat time phases to use a two-dimensional distribution of feature amounts (feature amounts) whose values change according to the degree of myocardial infarction. Image). This feature amount image data is sent to the display 6 and converted into an analog signal, and then displayed as a light and shade or color display as a visual feature amount image.
[0011]
FIG. 2 is a block diagram of the feature image generating unit 7. The frame memory 11 has a capacity capable of storing a plurality of temporally continuous RF reception signal groups obtained by repeating a series of sector scans. The frame memory 11 is provided with two output ports. One output port is loaded with a group of RF reception signals of the first frame, and the other output port is provided with a second frame having a different heartbeat time phase from the first frame, for example, a second frame delayed by one frame from the first frame. An RF reception signal group is loaded.
[0012]
For example, a group of RF reception signals for a plurality of frames is managed by a frame number n in frame units. The frame number n is assigned as a serial number according to the scanning order. Similarly, in the DSC 5, the image is managed by the frame number N. The frame number n and the frame number N are associated with each other. When the operator designates a specific image displayed on the display 6 via an input device (not shown), the RF reception signal group of the frame number n as the first frame corresponding to the frame number N of the specific image is changed to one of the two. The RF reception signal group of the frame number n + 1 as the second frame is loaded on the output port, and loaded on the other output port. In this way, the operator can specify the heartbeat time phase for creating the feature quantity image by specifying the image of the desired heartbeat time phase.
[0013]
From one output port, a group of RF reception signals in a local area of a predetermined size (referred to as a kernel ROI), and from the other output port, a local area (referred to as an object ROI) near the kernel ROI having the same size. Are received by the feature value calculation unit 12. The feature value calculation unit 12 calculates a feature value indicating the similarity between the two RF reception signal groups, for example, a cross-correlation coefficient. The feature amount calculation unit 12 repeats the calculation of the feature amount while shifting the position of the object ROI within the search ROI within a predetermined range while keeping the position of the kernel ROI fixed. As a result, a plurality of feature values are calculated for the kernel ROI at a certain position, and the feature value calculation unit 12 selects the maximum value (or the minimum value) from the plurality of feature values, and selects the selected maximum value. The value is recognized as a feature value corresponding to the kernel ROI. The feature value calculation unit 12 repeats such calculation of the feature value while changing the position of the kernel ROI. Thereby, a two-dimensional distribution of the feature amount, that is, feature amount image data is obtained. This feature amount image data is sent to the display 6 via a digital scan converter (DSC) 13, converted into analog, and then displayed as a grayscale or color image as a visual feature amount image.
[0014]
Note that the control system 10 indicates that the feature amount image can be displayed when the feature amount image data for one frame has been generated and stored in the DSC 13 and the feature amount image can be displayed. Is preferably displayed on the display 6 via the DSC 13. After this message is displayed, when the operator gives an instruction to display the feature amount image via an input device (not shown), the control system 10 receives the message and sends the feature amount image data from the DSC 13 to the display 6. Read and display.
[0015]
FIG. 3 is a block diagram of the DSC 13. The feature amount image data obtained by the feature amount calculation unit 12 is transmitted to the display 6 as a digital video signal at an appropriate timing designated by, for example, an operator via the luminance conversion unit 14, the coordinate conversion unit 15, and the frame memory 16. Can be The luminance conversion unit 14 converts the characteristic amount into luminance data (or RGB hue data), and includes, for example, a ROM in which the input is the characteristic amount and the output is the luminance. The correspondence between the feature amount and the luminance (so-called setting) is summarized in an allocation table and stored in the luminance conversion unit 14. It is preferable that the brightness conversion unit 14 be provided with a plurality of types of tables having different settings as shown in FIG. 5 and to be able to select an arbitrary table by an operator's specification. FIG. 5A shows settings for luminance conversion, and FIG. 5B shows settings for hue conversion. The coordinate conversion unit 15 converts the coordinate system (θ, d) representing the feature amount into an appropriate coordinate system corresponding to the display 6, for example, a rectangular coordinate system (x, y). θ indicates the swing angle of the ultrasonic beam, and d indicates the depth. It is preferable that the DSC 13 performs spatial interpolation of the feature amount.
[0016]
FIG. 4 is a schematic explanatory diagram of the generation process of the feature amount image. As described above, the feature amount is a calculated value that can differentiate an infarcted myocardium obtained by quantifying the degree of myocardial infarction from a normal myocardium. For example, it means that the degree of infarction is stronger as the feature amount is larger, and it is closer to normal as the feature amount is smaller. By the way, the infarcted myocardium has a smaller contraction rate than the normal myocardium, and the tissue in that part is solidified (infarcted) regardless of the heart beat. Therefore, the position of the infarct portion simply changes (metastasis) in accordance with the periodic contraction of the normal myocardium. The present invention focuses on this point. That is, since the infarcted myocardial portion of the first frame and the infarcted myocardial portion of the second frame to which the portion has metastasized have little tissue change, the RF reception within that portion is not performed. The distribution of the signal group is very similar (very similar) between the two frames, and the quantification of the similarity between the RF reception signal groups is given as a feature amount. However, it is unclear where the infarcted myocardium metastasizes. As will be described later, while the position of the kernel ROI is fixed and the position of the object ROI is shifted in the search ROI, the feature amount is calculated at each position, and the maximum value among the plurality of feature amounts is calculated. Is determined as a feature amount corresponding to the kernel ROI, that is, the object ROI is moved over a wide range in which the infarcted myocardial portion may metastasize to search for the feature amount. Hereinafter, the generation process of the feature image will be specifically described.
[0017]
As shown in FIG. 4A, the first frame of the frame number n designated by the operator while observing the RF reception signal group of two frames having different scanning timings and different heartbeat time phases, preferably a B-mode image. , And the RF reception signal group of the second frame of frame number n + 1 delayed by one frame from the first frame, a feature image is generated.
[0018]
First, with reference to FIGS. 4B, 4C, and 4D, a procedure for obtaining a feature amount of a certain kernel ROI (center point (xm, ym, n)) will be described. From one output port of the frame memory 11, a group of RF reception signals in the kernel ROI of the first frame is received, and from the other output port, a group of RF reception signals in the object ROI of the second frame near the kernel ROI. Are taken into the feature value calculation unit 12. The similarity (feature amount) of the distribution of both RF reception signal groups is calculated, for example, as a cross-correlation coefficient. Details of the method of calculating the feature amount will be described later. With the position of the kernel ROI fixed, the position of the object ROI is transferred within the search ROI, and the feature amount is calculated. Further, the position of the object ROI is shifted, and the feature amount is calculated in the same manner. As described above, while the object ROI is transferred within the search ROI, a plurality of feature amounts having the same kernel ROI position and different object ROI positions are obtained. The maximum value (or the minimum value) of the plurality of feature values is recognized as the feature value corresponding to the kernel ROI, and the maximum value is given the position information of the center point of the kernel ROI.
[0019]
Then, the kernel ROI is moved to, for example, the center point (xm + 1, ym + 1, n), and the feature amount is similarly obtained. While changing the position of the kernel ROI, a feature amount is obtained at each position. Thus, a two-dimensional distribution of the feature amount, that is, a feature amount image is obtained.
[0020]
The feature amount image obtained in this manner is sent to the display 6 as a digital video signal at an appropriate timing, for example, at the timing designated by an operator via the luminance conversion unit 14, the coordinate conversion unit 15, and the frame memory 16 of the DSC 13 in order. The image is displayed in parallel with the B-mode image as shown in FIG. 6A or superimposed on the B-mode image as shown in FIG. 6B.
[0021]
Next, a method of calculating the feature amount will be described. Two methods are provided for calculating the feature amount. The feature amount calculation unit 12 may be configured to be able to execute both methods, and may be selectively used in accordance with an operator's instruction, or one of the methods may be fixedly adopted in advance. The first method is a method of calculating a cross-correlation coefficient as a feature value, and the second method is a method of calculating a sum of absolute values of differences between a kernel of both frames and a ROI object ROI as a feature value. .
[0022]
Further, as a first method, a direct method and a two-dimensional Fourier transform method are provided. The direct method is calculated based on a general expression (definition expression) of the cross-correlation coefficient, and the two-dimensional Fourier transform method is the Wiener-Hinchin theorem that the cross-correlation function is the inverse Fourier transform of the power spectrum of the original function. Is calculated by using.
[0023]
FIG. 7 is an explanatory diagram of the overall flow of the first method. c (x, y) is the cross-correlation coefficient distribution to be obtained. Since c (x, y) is a complex signal, absolute value processing is required for ultimate imaging. Here, the step (transition pitch of the kernel ROI) of the operation point X and Y loops is 1/4 of the vertical and horizontal (rx, ry) of the size of the kernel ROI (same as the object ROI) shown in FIG. It is preferable from the viewpoint that the error of c (x, y) falls within an allowable range and the calculation time is shortened. In this case, c (x, y) not calculated is interpolated.
[0024]
FIG. 9 is an explanatory view of the calculation procedure of the direct method, and FIG. 10 is an explanatory view of the calculation procedure of the two-dimensional Fourier transform method. Note that the precondition is defined as shown in FIG. ^* C is the complex conjugate of c, cc (τx, τy) is the cross-correlation coefficient, csum is the element value of x = τx, y = τy of the cross-correlation coefficient, and cc and csum are complex signals.
[0025]
FIG. 11 is an explanatory diagram of the second method. While moving the object ROI within the search ROI with respect to a certain kernel ROI (center point (x, y)), at each position, the absolute value ε (xi, yj) of the sum of the differences between the corresponding positions, and The ratio ε ′ (xi, yj) of ε (xi, yj) to the sum in the kernel ROI is calculated, and the minimum value (or maximum value) of ε ′ (xi, yj) is calculated as the kernel ROI (center point (x, y) The feature amount in the above ()). By repeating such calculation while transferring the kernel ROI, a two-dimensional distribution of the feature amount is generated.
[0026]
As described above, according to the present embodiment, the degree of myocardial deformation is quantified as the similarity (feature amount) between the distributions of the RF reception signals in the local region of each frame using two frames with different cardiac phases. The infarcted myocardium can be objectively imaged. Moreover, since the similarity searches the search area, the metastasis of the infarcted myocardium can be followed.
[0027]
The present embodiment can be modified as follows. In the above description, similarity is observed based on the distribution of RF reception signals between two-dimensional local regions. It is developed into similarity between three-dimensional local regions. This can be easily realized by adding depth information to the position information of the RF reception signal. That is, as shown in FIG. 12, a position detector 20 for detecting the position of the probe 1 is added to the configuration of FIG. 1, and the position information of the probe 1 is given to the feature amount image generation unit 7 as depth information. For example, a multi-layer RF reception signal group obtained while moving the probe 1 in parallel can be handled as one three-dimensional data. FIG. 13 shows an example of a three-dimensional feature image displayed in parallel with the B-mode image subjected to the three-dimensional image processing. According to this modified example, unknown three-dimensional metastasis of infarcted myocardium can be captured.
The present invention can be implemented with various modifications without departing from the scope of the invention. Note that the above description is also applicable to B mode and color flow mapping.
[0028]
【The invention's effect】
The present invention relates to an ultrasonic diagnostic apparatus that repeatedly obtains one frame of a received signal by scanning a two-dimensional region of a subject with an ultrasonic wave, and generates an image by detecting the received signal. Generating a two-dimensional distribution of a feature representing the similarity between the received signal in the first local region of the second frame and the received signal in the second local region of the second frame having a different scanning timing from the first frame. Infarct portion comprising a volume two-dimensional distribution generating means and a display means for displaying the two-dimensional distribution as a feature image, wherein the position is simply changed (metastasis) without deforming the tissue according to the periodic contraction of the normal myocardium Can be regarded as having high similarity between the received signal in the first local region of the first frame and the received signal in the second local region of the second frame. Myocardium It can provide an ultrasonic diagnostic apparatus capable of characteristically imaging.
[Brief description of the drawings]
FIG. 1 is a block diagram showing the configuration of an embodiment of an ultrasonic diagnostic apparatus according to the present invention.
FIG. 2 is a block diagram of a feature image generating unit of FIG. 1;
FIG. 3 is a block diagram of the DSC shown in FIG. 2;
FIG. 4 is a schematic explanatory view of a generation process of a feature image.
FIG. 5 is a diagram showing setting changes of luminance and hue conversion.
FIG. 6 is a view showing a display example of a feature image.
FIG. 7 is an explanatory diagram of a first method for obtaining a feature amount.
FIG. 8 is an explanatory diagram of parameters used for calculating a cross-correlation coefficient.
FIG. 9 is an explanatory diagram of a calculation procedure of a direct method for calculating a cross-correlation coefficient.
FIG. 10 is an explanatory diagram of a calculation procedure of a two-dimensional Fourier transform method for calculating a cross-correlation coefficient.
FIG. 11 is an explanatory diagram of a second method for obtaining a feature amount.
FIG. 12 is a block diagram showing a configuration of a modified example of the ultrasonic diagnostic apparatus of the present invention.
FIG. 13 is a diagram showing a display example of a feature amount image according to a modified example.
[Explanation of symbols]
1. Ultrasonic probe, 2. Transmission system,
3 ... Reception system 4 ... Signal processing unit
5 ... DSC, 6 ... Display,
7 ... Feature amount image generation unit.

Claims (3)

By scanning a two-dimensional or three-dimensional region of a subject with ultrasonic waves, a received signal for one frame is repeatedly obtained, and an ultrasonic diagnostic apparatus that generates an image by detecting the received signals is provided.
Received signals in a first local area of a first frame and receptions in a plurality of second local areas spatially adjacent to the first local area in a second frame different in scanning timing from the first frame. A cross-correlation coefficient is calculated between each of the signals, and a feature value indicating similarity of the first local region to the second local region is defined using a maximum value of the calculated cross-correlation coefficients. A feature amount two-dimensional distribution generating unit configured to generate a two-dimensional distribution of the feature amount while changing a position of the first local region;
Display means for displaying the two-dimensional distribution as a feature image. By scanning a two-dimensional or three-dimensional region of a subject with ultrasonic waves, a received signal for one frame is repeatedly obtained, and an ultrasonic diagnostic apparatus that generates an image by detecting the received signals is provided.
Received signals in a first local area of a first frame and receptions in a plurality of second local areas spatially adjacent to the first local area in a second frame different in scanning timing from the first frame. The sum of the absolute values of the signal differences between the signals and each of the signals is calculated, and a feature value representing the similarity of the first local region to the second local region is defined using the minimum value of the calculated sum. A feature amount two-dimensional distribution generating unit configured to generate a two-dimensional distribution of the feature amount while changing a position of the first local region;
Display means for displaying the two-dimensional distribution as a feature image. The ultrasonic diagnostic apparatus according to claim 1, wherein the display unit displays the two-dimensional distribution via one of a luminance conversion process and a hue conversion process.

Images