Ultrasonic diagnostic apparatus and echo image recognition method

Abstract

To provide an ultrasonic diagnostic apparatus capable of accurately and easily identifying a tissue visualized in an echo image.SOLUTION: The ultrasonic diagnostic apparatus comprises: a panoramic image generation part which generates a panoramic image obtained by joining and expanding a plurality of sheets of echo images acquired consecutively while moving an ultrasonic probe; and an image recognition part which recognizes a predetermined portion in the body on the basis of the generated panoramic image. The image recognition part refers to a result of the image recognition based on the panoramic image and recognizes the predetermined portion on the echo signal.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an ultrasonic diagnostic apparatus for acquiring an echo image based on ultrasonic waves reflected inside the body of a subject, and an echo image recognition method for the ultrasonic diagnostic apparatus. [Background technology]

[0002] Ultrasound diagnostic devices can obtain cross-sectional images of the inside of a living body in real time, and have the advantages of being less invasive than other medical devices such as X-ray CT, and being small and easy to carry. In recent years, ultrasound images have been widely used not only for medical diagnosis of surgical diseases, but also in acupuncture and moxibustion clinics to evaluate and record the effects of treatment.

[0003] Generally, B-mode images used in ultrasound diagnosis have an inverted sector shape due to their structure, and therefore the observable range in still images is narrow and lacks objectivity. In relation to such a problem, for example, Patent Document 1 discloses a technology for stitching together multiple echo images acquired continuously while moving an ultrasound probe to generate a panoramic image. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2019-217018 A Summary of the Invention [Problem to be solved by the invention]

[0005] As mentioned above, since an echo image is inverted fan-shaped and the observable area is relatively narrow, it is difficult to identify which location (part) in the body was photographed from a single recorded echo still image. In particular, as a record of nerve blocks or acupuncture, it is insufficient to ensure objectivity to the extent that anyone can look at the echo image at a later date and accurately identify which part was photographed.

[0006] An object of the present invention is to provide an ultrasonic diagnostic apparatus or an echo image recognition method that can easily and highly accurately identify a specific site or tissue depicted in an echo image. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems, the present invention provides an ultrasound diagnostic device that acquires an echo image by emitting ultrasound from an ultrasound probe toward the body and receiving the ultrasound reflected inside the body with the ultrasound probe, and the ultrasound diagnostic device includes a panoramic image generation unit that generates an expanded panoramic image by stitching together a plurality of echo images that are continuously acquired while moving the ultrasound probe, and an image recognition unit that recognizes a specified part inside the body based on the generated panoramic image.

[0008] The ultrasound diagnostic apparatus may be such that the image recognition section recognizes the predetermined portion on the echo image by referring to an image recognition result based on the panoramic image.

[0009] The ultrasonic diagnostic apparatus may further include a display processing unit that displays each recognized part in a different color.

[0010] The present invention also provides an echo image recognition method in an ultrasound diagnostic device that acquires an echo image by emitting ultrasound from an ultrasound probe toward the body and receiving the ultrasound reflected inside the body with the ultrasound probe, the echo image recognition method including the steps of: continuously acquiring a plurality of echo images while moving the ultrasound probe; stitching together the acquired plurality of echo images to generate an expanded panoramic image; and recognizing a specified part inside the body based on the generated panoramic image.

[0011] The echo image recognition method may further include a step of recognizing the predetermined portion on the echo image by referring to an image recognition result based on the panoramic image.

[0012] The echo image recognition method may further include a step of displaying each recognized part in a different color. Effect of the Invention

[0013] According to the present invention, it is possible to accurately and easily identify a specific part or tissue in a living body depicted in an echo image acquired by an ultrasound diagnostic device. This makes it possible to obtain objectivity that enables any doctor or technician looking at the recorded echo image to accurately identify which part or tissue is depicted, regardless of the skill level of the doctor or technician. [Brief description of the drawings]

[0014] [Figure 1] 1 is a functional block diagram for explaining a system configuration of an ultrasonic diagnostic apparatus according to an embodiment of the present invention. [Diagram 2] FIG. 11 is a flow diagram for illustrating an example of a panoramic image generating process. [Diagram 3] FIG. 11 is a diagram for explaining a correction process for an echo image. [Figure 4] FIG. 13 is a diagram for explaining a synthesis process of an echo image. [Diagram 5] FIG. 2 is a diagram showing an example of a panoramic echo image. [Figure 6] FIG. 13 is a diagram showing an example of a panoramic echo image color-coded for each tissue. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] An ultrasonic diagnostic device 1 according to one embodiment of the present invention is a device that applies an ultrasonic probe 11 having multiple transducers to the skin surface of a subject, radiates ultrasonic waves toward the body, and acquires ultrasonic tomographic images (echo images) by receiving the ultrasonic waves reflected inside the body with the ultrasonic probe 11 and processing the signals. In particular, the ultrasonic diagnostic device 1 of this embodiment has a function of stitching together multiple echo images acquired continuously while moving the ultrasonic probe 11 to generate an expanded panoramic image in order to accurately and easily identify a specific muscle site or tissue in the living body depicted in the echo image.

[0016] In the description of the embodiments in this specification, an ultrasonic tomographic image is a B-mode image, which is simply referred to as an "echo image." Furthermore, unless otherwise specified, an "echo image" refers to one image frame or its image data generated for each scan period. Needless to say, "obtaining" an image includes signal processing for generating image data visualized for display.

[0017] The configuration of an ultrasonic diagnostic device 1 according to a preferred embodiment of the present invention will be described below with reference to the drawings. Fig. 1 is a functional block diagram for explaining the system configuration of the ultrasonic diagnostic device 1. In the figure, a B-mode image generating unit 13, a display processing unit 14, a panoramic image generating unit 17, and an image recognition unit 18 provided in a main body 10 are processing units whose respective functions are realized by a computer processor such as a system control unit 12 cooperating with hardware to execute arithmetic processing according to a predetermined program. In addition, these processing units may be configured with dedicated hardware integrated circuits or digital signal processing units specialized for image signal processing and / or AI (Artificial Intelligence).

[0018] In FIG. 1, the ultrasonic probe 11 has a transducer array 111 in which multiple transducers are arranged in one direction, a switching circuit 112 connected to the transducer array 111, a transmitting unit 113 that transmits an ultrasonic drive signal, and a receiving unit 114 that receives a reflected ultrasonic signal.

[0019] The transducer array 111 may be of any of convex, linear, and sector types, but the convex type has a relatively wide contact surface and is suitable for wide-angle and deep range observation. The transducer array 111 has an acoustic lens 111a and an acoustic matching layer 111b in this order from the side facing and contacting the body surface of the subject. The acoustic lens 111a is provided to focus the ultrasonic beam emitted from the transducer array 111 and improve the resolution in the thickness direction of the slice. The acoustic matching layer 111b is provided to reduce the difference in acoustic impedance between the transducer and the living body and to efficiently transmit and receive ultrasonic waves.

[0020] The transmitting unit 113 generates a pulsed driving signal for generating an ultrasonic beam emitted by the transducer array 111 under the control of the system control unit 12, and outputs the driving signal to the switching circuit 112. The receiving unit 114 is a circuit that receives a reflected ultrasonic signal detected and converted by the transducer array 111 and converts the signal into digital data. The switching circuit 112 is configured to switch the signal connection between each transducer provided in the transducer array 111 and the transmitting unit 113 and the receiving unit 114 according to a scan control signal from the system control unit 12. Specifically, the system control unit 12 outputs a scan control signal to the switching circuit 112, and selects a group of transducers in the transducer array 111 that transmit and receive ultrasonic waves while shifting them in the array direction, thereby performing linear scanning by the ultrasonic probe 11. The scan (frame) rate is appropriately set according to the image processing speed by the following B-mode image generating unit 13.

[0021] The B-mode image generating unit 13 generates echo image data based on the reflected ultrasonic signal received by the receiving unit 114. The display processing unit 14 sets display parameters such as brightness, contrast, and scale for the generated echo image data, and generates diagnostic images by arranging measurement values, diagnostic data, and the like in a predetermined layout. In the video mode, diagnostic images are processed sequentially for each scan period, and the echo images are displayed on the monitor 20 in real time as a video.

[0022] Furthermore, the echo image data generated by the B-mode image generating unit 13 is temporarily stored in the multi-frame memory 15. The multi-frame memory 15 has a memory area for storing a plurality of image frames, and a write address pointer is incremented every time a new echo image is captured. As a result, the multi-frame memory 15 sequentially stores a plurality of image frames acquired going back a predetermined time from the most recent one.

[0023] In the video mode, the operator can perform a "record" operation at any timing while viewing the echo video image displayed on the monitor 20, thereby recording, for example, echo image data depicting tissues or regions significant for medical diagnosis in the data storage 16. The operator can also later search for the echo images recorded in the data storage 16 and play them back and display them on the monitor 20. The echo image recorded in the data storage 16 in this manner may be a still image consisting of one frame, or a video consisting of multiple frames over a specified period of time.

[0024] The panoramic image generating unit 17 is a processing unit that generates a panoramic echo image with an expanded observation range compared to the conventional art. Here, FIG. 2 is a flow chart for explaining an example of the processing by the panoramic image generating unit 17.

[0025] In the panoramic mode, the ultrasonic probe 11 is moved on the skin surface of the subject in its longitudinal direction (the direction in which the transducers are arranged) to stitch together overlapping portions of multiple echo images acquired continuously, thereby obtaining the above-mentioned panoramic echo image. To this end, the panoramic image generator 17 first evaluates the overlapping portions of two positionally adjacent images for the multiple echo images stored in the above-mentioned multi-frame memory 15 in chronological order (i.e., in the order of the scan positions; n=0, 1, 2, . . .).

[0026] In an echo image used in a general diagnosis or examination, the ultrasonic beam output from the ultrasonic probe 11 is radial, so that the field of view is inverted fan shape, and the contour lines from the ground surface of the ultrasonic probe 11 are arc-shaped. Therefore, when the matching between the ends of adjacent echo images is simply compared, an error may occur in the judgment. Therefore, in this embodiment, as an initial process, it is preferable to perform a correction process such as that shown in FIG. 3 so that the contour lines become straight lines in each echo image (step S11). In addition, in order to reduce the influence of image distortion caused by the close contact state between the ultrasonic probe 11 and the skin, for example, an effective area E is set by narrowing the area of ​​the image ends to be compared toward the center. n , E n+1 may be defined (step S12).

[0027] Next, the panoramic image generating unit 17 calculates an effective area E of the echo image as shown in FIG. n , E n+1 Let S be a matrix section S, which is a number of small regions. n,i,j , S n+1,k,l Each matrix section is divided into S n,i,j , S n+1,k,l In step S13, feature extraction of the image is performed. For example, in one matrix section S n,i,j Find one or more geometric feature points in the image, and define the feature points as a multidimensional vector (called a feature vector F n,i,j ) to extract features. The extracted feature vector F n,i,j , F n+1,k,l (If there are multiple feature points in one section, the center of gravity of each feature vector) is calculated, and if the distance is less than a predetermined value, each matrix section S n,i,j , S n+1,k,l It is determined that there is a duplication (match) (step S14).

[0028] Through the above comparison, the matrix section S that is determined to be duplicated n,i,j , S n+1,k,l By merging (composing) the echo images En , E n+1 The overlapping portions can then be joined together (step S15).

[0029] The data of the panoramic echo image generated through the above-mentioned processing by the panoramic image generating unit 17 is set with display parameters such as brightness, contrast, scale, etc. by the display processing unit 14 and is displayed on the monitor 20.

[0030] The panoramic echo image with the expanded observation range is effective for accurately and easily identifying a specific site or tissue in a living body. For example, in the video mode, the panoramic echo image can be displayed as a reference on the monitor 20 together with the echo video image (see, for example, FIG. 5).

[0031] The size of the panoramic echo image is not limited by the resolution (screen size) of the monitor 20 included in the ultrasound diagnostic apparatus 1. For example, it is also possible to generate a panoramic echo image covering all directions of 360 degrees and record it in the data storage 16.

[0032] The image recognition unit 18 is a processing unit that performs image recognition of muscle tissue from contrast and contour information of fascia, etc. in the generated echo image and panoramic echo image. The image recognition unit 18 can perform image recognition of a predetermined site or tissue in a living body depicted in the echo image and panoramic echo image based on, for example, multiple past teacher data.

[0033] The image recognition unit 18 can improve the accuracy of diagnosis by identifying a specific area or muscle tissue on one echo image with reference to the image recognition result based on the panoramic echo image. This enables objective and detailed diagnosis, for example, by identifying the location of fascial adhesion and determining to what extent each muscle is adhered.

[0034] The display processing unit 14 can also color-code the echo image and / or the panoramic echo image for each recognized or specified site or tissue, and display the diagnostic image on the monitor 20 (see, for example, FIG. 6). The echo image and the panoramic echo image processed in this manner can also be recorded in the data storage 16 in a state that can be searched and played back later. [Explanation of symbols]

[0035] 1. Ultrasound diagnostic equipment 10 Main unit 11 Ultrasound probe 12 System control section 13 B-mode image generation unit 14 Display processing section 15 Multiframe Memory 16 Data Storage 17 Panorama image generator 18 Image Recognition Unit 20 Monitor 111 Transducer Array 111a Acoustic Lens 111b Acoustic matching layer 112 Switching Circuits 113 Transmitter 114 Receiving unit

Claims

1. An ultrasound diagnostic device that emits ultrasound from an ultrasound probe into a body and receives ultrasound reflected from the body by the ultrasound probe to obtain an echo image, a panoramic image generating unit that generates an expanded panoramic image by stitching together a plurality of echo images that are continuously acquired while moving the ultrasonic probe; an image recognition unit that recognizes a predetermined part inside the body based on the generated panoramic image; An ultrasound diagnostic device comprising:

2. The ultrasound diagnostic apparatus according to claim 1 , wherein the image recognition unit recognizes the predetermined portion on the echo image by referring to an image recognition result based on the panoramic image.

3. The ultrasonic diagnostic apparatus according to claim 2 , further comprising a display processing unit that displays each recognized part in a different color.

4. 1. An echo image recognition method for an ultrasound diagnostic device that acquires an echo image by emitting ultrasound from an ultrasound probe toward a body of a patient and receiving the ultrasound reflected from the body by the ultrasound probe, comprising: A step of continuously acquiring a plurality of echo images while moving the ultrasonic probe; stitching the acquired echo images together to generate an expanded panoramic image; recognizing a predetermined part of the body based on the generated panoramic image; An echo image recognition method comprising:

5. The echo image recognition method according to claim 4 , further comprising the step of recognizing the predetermined portion on the echo image by referring to an image recognition result based on the panoramic image.

6. The echo image recognition method according to claim 5, further comprising the step of displaying each recognized region in a different color.

Images