JP2010029281A - Ultrasonic diagnostic apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To identify an abnormal area such as an artifact by using a plurality of ultrasonic echo images, and to create a display image according to the result of identification. <P>SOLUTION: A composing part 20 composes echo images created by two ultrasonic probes 10A and 10B. The composing part 20 computes the similarity of pixels in the images in the vicinity corresponding to the same position within a living body between the two echo images to be composed. If the similarity is less than a threshold, it is determined that the pixels have abnormalities such as artifacts, and the pixel value is changed to a prescribed value such as zero. If the similarity is the threshold or above, the pixel values in the respective echo images are weighted and added by a weight set according to the similarity. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an ultrasonic diagnostic apparatus, and more particularly to image quality improvement of an ultrasonic echo tomographic image.

  As a method for improving the contrast resolution of an ultrasonic echo tomographic image, inter-frame correlation processing, spatial compound processing, or the like in which a plurality of ultrasonic echo tomographic images are superimposed is known. The inter-frame correlation process is a process of adding image frames with different time phases for each pixel, and the spatial compound process transmits and receives an ultrasonic beam multiple times while changing the sound ray direction for the same part with one ultrasonic probe, This is a process of adding echo signals obtained by a plurality of times of transmission / reception for each same part (pixel).

  In Patent Documents 1 and 2, when displaying an image by correlating between a plurality of image frames having different time phases, a correlation coefficient (weight for weighted addition) is varied according to the frame rate, A method of weighting and adding the image frames using the correlation coefficient is disclosed.

  Patent Document 3 discloses a method of dynamically changing the correlation coefficient (weight) of inter-frame correlation calculation for each pixel. In this method, paying attention to the fact that the S / N ratio is good in the portion where the pixel value is large, the coefficient of the current frame is increased as the larger pixel value of the current frame and the previous frame is larger for each pixel. The influence of the previous frame in the output image is reduced, and blurring of the image due to motion is reduced.

Japanese Patent Laid-Open No. 5-7592 JP-A-11-128222 JP 2006-271557 A

  Regardless of the inter-frame correlation process or the spatial compound process, even if an abnormal part such as an artifact is included in the image frame to be added to the reference image frame, it is added as it is. Although the influence of abnormal parts such as artifacts is reduced by the addition, it cannot be said that the abnormal parts are positively identified and dealt with. The method of controlling the correlation coefficient according to the frame rate (that is, the weight of weighted addition) is, after all, only applying the coefficient uniformly to the entire image, and selectively identifying abnormal parts such as artifacts. It will not deal with.

  Although the method of Patent Document 3 changes the correlation coefficient (weight for weighted addition) for each pixel, this method is based on the idea that the larger the pixel value, the better the S / N ratio. Since the magnitude of the pixel value has little relation to the presence or absence of an abnormality such as an artifact, it cannot be said that this method also positively identifies an abnormal part such as an artifact and copes with it.

  The present invention makes it possible to identify abnormal parts such as artifacts using a plurality of ultrasonic echo images and generate a display image according to the identification result.

  An ultrasonic diagnostic apparatus according to the present invention includes a transmission / reception unit that generates a reference echo image and an auxiliary echo image by transmitting and receiving an ultrasonic beam into a subject, the reference echo image, and the auxiliary echo image. Between each pixel, a similarity calculation means for calculating a similarity between blocks of a predetermined size centered on the pixel, and a display image obtained by weighted addition of the reference echo image and the auxiliary echo image Display image generation means for generating the weight of the reference echo image and the weight of the auxiliary echo image in the weighted addition for each pixel according to the similarity for each pixel calculated by the similarity calculation means Display image generation means for controlling.

  According to this configuration, weighted addition of both echo images is controlled between the reference echo image and the auxiliary echo image for each pixel according to the degree of similarity between blocks of a predetermined size centered on the pixel. can do.

  In one aspect, the display image generation means, for pixels whose similarity calculated by the similarity calculation means is less than a predetermined threshold, weights of the reference echo image and the auxiliary echo image for the display image Both are set to 0, and a predetermined value indicating an abnormal pixel is set as a value of a weighted addition result for the pixel in the display image.

  According to this aspect, an abnormal pixel having a low similarity of less than a threshold can be displayed separately from a pixel having a similarity equal to or higher than the threshold.

  In another aspect, the transmission / reception means is different from the first transducer array that generates the reference echo image by electronically scanning the first region with an ultrasonic beam. And a second transducer array for generating the auxiliary echo image by scanning a second region including an overlapping region at least partially overlapping the first region from the position.

  In still another aspect, the second transducer array is disposed so that the angle of the second transducer array can be changed with respect to the first transducer array, and the display image generation unit is configured to be connected to the first transducer array. The pixel at the same position in the overlapping region between the reference echo image and the auxiliary echo image is specified according to the angle formed by the second transducer array, and the value of the specified pixel at the same position is determined. The weighted addition is performed.

  In still another aspect, the display image generating means uses the overlapped area of each pixel of the reference echo image as a value obtained by the weighted addition of the reference echo image and the auxiliary echo image. Each pixel other than the region generates a display image having the value of the reference echo image.

  According to the present invention, it is possible to identify abnormal parts such as artifacts based on the similarity of images between blocks between ultrasonic echo images, and to generate a display image according to the identification result.

  With reference to FIG. 1, the structural example of the ultrasound diagnosing device of embodiment is demonstrated. The illustrated ultrasonic diagnostic apparatus includes two ultrasonic probes 10A and 10B that generate two-dimensional echo tomographic images by electronic scanning. The electronic scanning shapes of the ultrasonic probes 10A and 10B are not particularly limited, but in the following, it is assumed that both are linear scanning in order to simplify the description. The ultrasonic probe 10B is rotatably connected to the ultrasonic probe 10A by a hinge or the like. That is, the ultrasonic probes 10A and 10B can be opened and closed like a cocoon. The arrangement direction of the transducers in the transducer array of the ultrasonic probes 10A and 10B is in the same plane, and the rotation axis for opening and closing the probes 10A and 10B is a direction perpendicular to the plane. If both the probes 10A and 10B are linear scanning, the scanning ranges of the probes 10A and 10B overlap each other if the angle formed by the wave transmitting / receiving surfaces of both the probes is less than 180 degrees.

  The transmission / reception units 12A and 12B drive and control the corresponding ultrasonic probes 10A and 10B, respectively, to realize transmission / reception of ultrasonic beams and electronic scanning. The transmission / reception units 12A and 12B have a transmission unit function and a reception unit function. The transmission unit functions as a transmission beam former. That is, a plurality of transmission signals are supplied from the transmission unit to the plurality of transducers of the ultrasonic probe 10A or 10B. As a result, an ultrasonic beam pulse is emitted from the probe 10A or 10B into the living body. The reflected wave from the living body is received by the probe 10A or 10B. As a result, a plurality of reception signals are output from the plurality of transducers. These received signals are input to the receiving unit of the transmitting / receiving unit 12A or 12B. The reception unit functions as a reception beamformer and applies phasing addition processing to a plurality of reception signals.

  The received signals output from the transmission / reception units 12A and 12B are detected by the detectors 14A and 14B, respectively, and logarithmically compressed by the logarithmic compression units 16A and 16B, respectively. The received signals after logarithmic compression are scan-converted into display coordinate systems by DSCs (Digital Scan Converters) 18A and 18B, respectively. Here, for example, the angle detector 24 (for example, an angle encoder) detects the angle θ formed by the ultrasonic probes 10A and 10B, and supplies information on the angle θ to, for example, the DSC 18B. In this case, the DSC 18B uses the information on the angle θ to perform coordinate conversion so that the coordinate system of the reception signal of the ultrasonic probe 10B matches the coordinate system of the ultrasonic probe 10A. In this case, the coordinate system of the ultrasonic probe 10A is a display coordinate system. Further, the DSCs 18A and 18B obtain the data of coordinate grid points where no received signal data exists during the scan conversion from the received signal data of surrounding points by interpolation. The DSCs 18A and 18B each generate an echo image of one frame that matches the display coordinate system from the reception signal of one electronic scan. Each echo image frame is a B-mode image.

  The echo image frames respectively formed by the DSCs 18A and 18B are combined by the combining unit 20 and become one display image frame. In this example, an echo image frame generated by the ultrasonic probe 10A is used as a reference image, and the reference image is improved by a simultaneous phase echo image frame generated by the ultrasonic probe 10B. That is, the echo image frame generated by the ultrasonic probe 10B is used as an auxiliary image. For example, as illustrated in FIG. 2, a certain part in the electronic scanning range 100 </ b> A of the ultrasonic probe 10 </ b> A becomes a display range 200 displayed on the display 22. In the display range 200, a display image of an area overlapping the electronic scanning range 100B of the ultrasonic probe 10B is generated by synthesizing the echo images of the probes 10A and 10B. On the other hand, for the portion of the display range 200 that does not overlap the electronic scanning range 100B, the combining unit 20 outputs only the echo image of the probe 10A. Note that the two frames in the same phase do not have to be strictly at the same time.

  The display image frame of the display range 200 generated by the combining unit 20 is displayed on the display 22. The display 22 is a display device such as a liquid crystal display or a CRT.

  Next, image composition processing in the overlapping area of the electronic scanning ranges 100A and 100B in the composition unit 20 will be described.

  In this process, the values of pixels at the same position are synthesized between the reference image frame 300A generated by the ultrasonic probe 10A and the auxiliary image frame 300B generated by the ultrasonic probe 10B. Since the reception signal of the ultrasonic probe 10B is converted into the coordinate system of the ultrasonic probe 10A by the DSC 18B, the pixels at the same pixel position may be synthesized between the reference image frame 300A and the auxiliary image frame 300B.

  In this synthesis, in this embodiment, the similarity of images between pixel blocks (hereinafter referred to as ROI, which is referred to as ROI, where ROI is an abbreviation of Region Of Interest) centered on the pixel is calculated. The weight of each pixel value of the reference image frame 300A and the auxiliary image frame 300B at the time of synthesis is controlled according to the degree. The size of the ROI may be, for example, 3 × 3 pixels or 5 × 5 pixels.

That is, as schematically shown in FIG. 3, the image similarity between the ROI ₁ centered on the pixel in the reference image frame 300A and the ROI ₂ centered on the pixel in the auxiliary image frame 300B is determined. Thus, the weight a _{1 of} the pixel in the reference image frame 300A and the weight a ₂ of the pixel in the auxiliary image frame 300B are obtained. Then, the value corresponding to each pixel is multiplied by a corresponding weight, and the multiplication results are added together. By performing weighted addition based on such similarity for all pixels in the overlapping area, a display image 350 in the overlapping area is obtained. Note that the image of the reference image frame 300 </ b> A is used as it is for the region other than the overlapping region in the display range 200.

For example, a cross-correlation coefficient between ROIs can be used as the similarity between ROI images. The cross-correlation coefficient R _xy between the image x (t) and the image y (t) can be obtained by the following equation, for example.

Here, t is an index number representing the order of pixels in the ROI, and x (t) and y (t) are values of the t-th pixel in each image. m _{x, m y} is the average of the pixel values in each ROI. Further, Σ indicates a summation operation for all pixels in the ROI.

  Further, NSAD (Normalized Sum of Absolute Difference) represented by the following equation may be used as the similarity between images of ROIs.

  For example, when the cross-correlation coefficient is used, the similarity between both images is lower as the value is closer to 0, and the similarity between both images is higher as the value is closer to 1. The similarity may be calculated by any one of the above-described calculation formulas using the ROIs of the reference image frame 300A and the auxiliary image frame 300B as images x (t) and y (t), respectively. Note that the cross-correlation coefficient and NSAD are only examples of similarities.

  In this embodiment, a threshold value for the similarity between ROI images is determined in advance. When the similarity calculated by the synthesis unit 20 for a certain pixel is less than the threshold, an abnormality such as artifact or external noise is generated in one of the reference image frame 300A or the auxiliary image frame 300B for the pixel. Judge that Since the part at the same position in both images represents the same part in the living body, the images should be the same or close to each other, and the similarity should be high. Therefore, when the similarity is considerably low, it can be determined that some abnormality has occurred. Since the similarity threshold is used to determine abnormality, it is set to a low value. For example, a cross-correlation coefficient is used as the similarity and the threshold is set to 0.1, for example. The threshold value may be set according to how much the user who uses the ultrasonic diagnostic apparatus determines that the abnormality is detected.

  In this embodiment, for a pixel whose similarity is determined to be lower than a threshold and an abnormality has occurred, as an example, the weight assigned to the pixel in both images is set to 0, and the value of the pixel is It is a predetermined fixed value indicating abnormal. The fixed value indicating the abnormality may be 0, for example. In this case, the pixel whose similarity is less than the threshold value has a pixel value of 0, so that there is no image on the B-mode image (that is, black), but even if one pixel becomes so, the visual effect is small. As another example, a value that is not used in a gray scale of a normal B-mode image is assigned as a fixed value indicating an abnormal pixel, and the pixel having the fixed value is displayed when the display 22 displays the fixed value. Alternatively, it may be displayed in a predetermined color. In this case, the user can identify a pixel in which an abnormality has occurred, and can use it as information for diagnosis, for example.

On the other hand, for the pixels whose similarity is greater than or equal to the threshold, the weights a ₁ and a ₂ are obtained using the following equation, for example, according to the similarity.

_{Wherein, R 12} is similarity ROI between the reference image frame 300A and the auxiliary image frame 300B. In this equation, when the similarity is 1, the weights of the reference image frame 300A and the auxiliary image frame 300B are both equal to 0.5, and when the similarity is greater than or equal to the threshold and less than 1, the weight for the reference image frame 300A is larger. . The calculation formula for determining the weighting a ₁ and a ₂ from the similarity is not limited to the examples.

When the weights a ₁ and a ₂ are obtained in this way, the synthesis unit 20 multiplies the pixel values of the reference image frame 300A and the auxiliary image frame 300B by the weights a ₁ and a ₂ , respectively, and adds the multiplication results. As a result, the value of the pixel in the display image is obtained and written to the address of the pixel in the memory area for the display image.

  According to the above example, a pixel having a low similarity between ROIs between the reference image frame 300A and the auxiliary image frame 300B is regarded as an abnormal pixel, and the abnormal pixel is not displayed or displayed so that it can be clearly identified. Can be.

  In the above example, pixels whose similarity is equal to or less than the threshold value are fixed values. Instead, for such pixels, the pixel values of the reference image frame 300A may be used as they are. In this case, the following effects can be expected.

  That is, in the apparatus that generates a display image by superimposing the echo images of the plurality of ultrasonic probes 10A and 10B described above, the spatial positions of the probes are separated, and the visual field ranges of the probes are different. For this reason, it is considered that the sound velocity distribution of the medium (living body) viewed from each probe is different. If the image frames of the probes are simply superimposed (added) in this state, the images of the target organs do not overlap well, resulting in blurred or blurred images. On the other hand, according to the method of this embodiment in which the display image is configured according to the similarity between the image frames of each probe, it is possible to perform satisfactory image composition while suppressing blurring and blurring.

  The method of the embodiment described above can be applied to a probe system including three or more ultrasonic probes 10A. As an example, an ultrasonic diagnostic apparatus using a probe system including three ultrasonic probes 10A, 10B, and 10C illustrated in FIG. 4 will be described.

  In this example, the ultrasonic probe 10B is attached to one end of the ultrasonic probe 10A, and the ultrasonic probe 10C is attached to the other end so as to be rotatable about the end thereof as a rotation axis. Also in this example, the ultrasonic image of the probe 10A is set as a reference image frame, and the ultrasonic images of the probes 10B and 10C are set as auxiliary image frames. In this case, the display range 200 of the reference image frame includes an area where all of the electronic scanning ranges 100A, 100B and 100C of the three probes 10A, 10B and 10C overlap, an area where two of the three overlap, and electronic scanning. The area is classified into the range of 100A only. Only the signal of the probe 10A may be used for the region of only the electronic scanning range 100A, and the region where the two overlap each other may be processed in the same manner as in the example of FIG.

For the region where the three ranges overlap, as schematically shown in FIG. 5, the auxiliary image frame 300B (image generated by the ultrasonic probe 10B), the reference image frames 300A and 300C (images generated by the ultrasonic probe 10C). ), The pixel values at the same position (the same absolute position in the living body) are weighted by the weights a ₁ , a ₂ , and a ₃ according to the degree of similarity between the images, and added for display. An image 350 is formed.

The weights a ₁ , a ₂ , and a ₃ assigned to the pixel values of each image are obtained according to the classification represented by the table shown in FIG. In this _{table, E 12} is an evaluation value for the similarity _{R 12} in ROI between the reference image frame 300A and the auxiliary image frame _300B, 1 if _{R 12} is equal to or larger than the threshold, and 0 if it is less than the threshold value . Similarly, E ₁₃ is 1 if the similarity R ₁₃ between the ROIs of the reference image frame 300A and the auxiliary image frame 300C is equal to or greater than the threshold, and 0 if it is less than the threshold, and E ₂₃ is the auxiliary image frame 300B and the auxiliary image. If the similarity R ₂₃ between the ROIs of the frame 300C is greater than or equal to the threshold, it is 1 if it is less than the threshold. In the table of FIG. 6, the first frame is the auxiliary image frame 300B, the second frame is the reference image frame 300A, and the third frame is the auxiliary image frame 300C.

As shown in the table of FIG. 6, in this example, when all of the evaluation values E ₁₂ , E ₂₃ , and E ₁₃ are 1, the display image 350 is generated using all three image frames. At this time, the weights a ₁ , a ₂ , and a ₃ assigned to the reference image frame 300A and the auxiliary image frames 300B and 300C are calculated by the following equations.

In this formula, when all of the similarities R ₁₂ , R ₂₃ , and R ₁₃ are 1, the weight a ₂ of the reference image frame 300A is 0.5, and the weights of the auxiliary image frames 300B and 300C are both 0.25.

Similarly, when the evaluation values E ₁₂ and E ₂₃ are 1 and E ₁₃ is 0, the weights a ₁ , a ₂ , a ₃ obtained by the equations (5) to (7) using all three image frames are similarly used. Is used for weighted addition. In this case, since both the auxiliary image frames 300B and 300C have a similarity equal to or higher than the threshold with respect to the reference image frame 300A, the display image is generated using both as auxiliary.

Next, when the evaluation values E ₁₂ and E ₁₃ are 1 and E ₂₃ is 0, the auxiliary image frame 300B is similar to the reference image frame 300A with a similarity equal to or higher than the threshold, but the auxiliary image frame 300C is is not. Although the auxiliary image frames 300B and 300C are somewhat similar, the auxiliary image frame 300C that is not similar to the reference image frame 300A is used for image quality improvement from the viewpoint of improving the image quality of the reference image frame 300A with the assistance of the auxiliary image frame. Not in. In this case, the pixel values are weighted and added using the weights a ₁ , a ₂ , and a ₃ obtained by the following equations (8) to (10).

Similarly, when the evaluation value E ₁₂ is 1 and E ₂₃ and E ₁₃ are 0, the pixel values are weighted and added using the weights a ₁ , a ₂ , and a ₃ obtained in the equations (8) to (10). .

Next, when the evaluation values E ₂₃ and E ₁₃ are 1 and E ₁₂ is 0, the auxiliary image frame 300C is similar to the reference image frame 300A with a similarity equal to or higher than the threshold, but the auxiliary image frame 300B is is not. Although the auxiliary image frames 300B and 300C are somewhat similar, the auxiliary image frame 300B that is not similar to the reference image frame 300A is used for improving the image quality in terms of improving the image quality of the reference image frame 300A with the assistance of the auxiliary image frame. Not in. In this case, the pixel values are weighted and added using the weights a ₁ , a ₂ , and a ₃ obtained by the following equations (11) to (13).

Similarly, when the evaluation value E ₂₃ is 1 and E ₁₂ and E ₁₃ are 0, the pixel values are weighted and added using the weights a ₁ , a ₂ , and a ₃ obtained in the equations (11) to (13). .

In the evaluation value _{E 13} is _1, if _{the E 12} and _{E 23} is 0, although somewhat similar to the auxiliary image frame 300B and 300C, neither resemble the reference image frame 300A. In this case, it is considered that an abnormality such as an artifact or noise occurs only in the reference image frame 300A. In this case, as an exceptional case from the basic idea of improving the reference image frame 300A with the auxiliary image frame, the pixel values of the auxiliary image frames 300B and 300C are weighted as shown in the following equations (14) to (16). Is used for weighted addition.

When all of the evaluation values E ₁₂ , E ₂₃ , and E ₁₃ are 0, it is determined that the pixel has an abnormality, and all of the evaluation values E ₁₂ , E ₂₃ , and E ₁₃ are determined as shown in the following equations (17) to (19) The weights a ₁ , a ₂ , and a ₃ are set to 0, and the value of the pixel of the display image 350 is set to a predetermined fixed value.

  The synthesizer 20 implements the determination rule shown in FIG. 6 and the weight calculation formulas (5) to (19) by a program or data or a combination of both, or a combination of a hardware arithmetic circuit and them. If it is.

  In the above, the case of using an ultrasonic probe of a one-dimensional array is taken as an example, but the above-described synthesis process can also be applied to the overlapping areas of volume data respectively obtained by both when using a probe of a two-dimensional array.

  Further, inter-frame correlation processing is performed on the echo image frames of the ultrasonic probes 10A and 10B (and 10C), and the synthesis processing close to the synthesis unit 20 is performed on each frame of the processing result. Also good.

  In the above, the case of synthesizing images of a plurality of ultrasonic probes has been taken as an example. However, the above-described synthesis processing is performed by synthesizing a plurality of frames at different points in the inter-frame correlation processing with a single probe, The present invention can also be applied to a case where image frames for different beam directions are synthesized by spatial compound processing using a probe. In this case, the installation location in the signal processing system of the synthesis unit 20 that synthesizes the image frame is not limited to the latter stage of the DSC, but between the detector and the logarithmic compression unit, or between the logarithmic compression unit and the DSC. May be.

It is a functional block diagram which shows schematic structure of the ultrasonic diagnosing device of embodiment. It is a figure which shows an example of the relationship between the electronic scanning range of two ultrasonic probes, and a display range. It is a figure for demonstrating the image synthesis process in the overlapping area | region of the electronic scanning range of two ultrasonic probes. It is a figure which shows an example of the relationship between the electronic scanning range of three ultrasonic probes, and a display range. It is a figure for demonstrating the image synthesis process in the overlap area | region of the electronic scanning range of three ultrasonic probes. It is a figure for demonstrating the rule of the image synthesis | combination in the overlapping area | region of the electronic scanning range of three ultrasonic probes.

Explanation of symbols

  10A, 10B ultrasonic probe, 12A, 12B transmission / reception unit, 14A, 14B detector, 16A, 16B logarithmic compression unit, 18A, 18B DSC, 20 synthesis unit, 22 display.

Claims (5)

Transmitting / receiving means for generating a reference echo image and an auxiliary echo image by transmitting and receiving an ultrasonic beam into the subject;
A similarity calculation means for calculating a similarity between blocks of a predetermined size centered on the pixel for each pixel between the reference echo image and the auxiliary echo image;
Display image generation means for generating a display image by weighted addition of the reference echo image and the auxiliary echo image, and according to the similarity for each pixel calculated by the similarity calculation means, the weighted addition in the weighted addition Display image generating means for controlling the weight of the reference echo image and the weight of the auxiliary echo image for each pixel;
An ultrasonic diagnostic apparatus comprising:   The display image generation unit sets both the weight of the reference echo image and the weight of the auxiliary echo image to 0 for the display image for pixels whose similarity calculated by the similarity calculation unit is less than a predetermined threshold. The ultrasonic diagnostic apparatus according to claim 1, wherein a predetermined value indicating an abnormal pixel is used as a value of a weighted addition result for the pixel in the display image. The wave transmitting / receiving means includes
A first transducer array that electronically scans a first region with an ultrasonic beam to generate the reference echo image;
A second region for generating the auxiliary echo image by scanning a second region including an overlapping region at least partially overlapping with the first region from a position different from the first transducer array. A transducer array;
including,
The ultrasonic diagnostic apparatus according to claim 1 or 2. The second transducer array is disposed such that the angle can be changed with respect to the first transducer array,
The display image generation unit is configured to perform the same operation in the overlapping region between the reference echo image and the auxiliary echo image according to an angle formed by the second transducer array with respect to the first transducer array. The pixel at the position is specified, and the weighted addition is performed on the value of the pixel at the same specified position.
The ultrasonic diagnostic apparatus according to claim 3. The display image generation means uses the value obtained by the weighted addition of the reference echo image and the auxiliary echo image for each pixel in the overlap area in the reference echo image, Generating a display image as a value of the reference echo image;
The ultrasonic diagnostic apparatus according to claim 3.

Images