Ultrasonic diagnostic device

The ultrasound diagnostic apparatus employs a learning model to adapt image quality processing by region, improving image quality and user customization in ultrasound diagnostics.

JP2024169860APending Publication Date: 2024-12-06FUJIFILM CORP
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2023086675
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-26
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Existing ultrasound diagnostic devices struggle to perform adaptive image quality adjustment processing that varies by region within an ultrasound tomographic image, leading to suboptimal image quality in different anatomical areas.

Method used

An ultrasound diagnostic apparatus that utilizes a learning model to determine specific regions within an ultrasound tomographic image and applies region-specific image quality adjustment parameters, allowing for adaptive processing by using both specific and predetermined parameters, and includes smoothing at boundaries for seamless transitions.

Benefits of technology

Enables high-quality ultrasound images with region-specific adjustments, enhancing contrast, resolution, and reducing artifacts, while allowing users to customize and compare image quality settings.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024169860000001_ABST
    Figure 2024169860000001_ABST
Patent Text Reader

Abstract

To provide an ultrasonic diagnostic device capable of executing adaptive image quality adjusting processing for each region in an ultrasonic tomographic image.SOLUTION: A learning model 28 performs learning to output image adjusting parameters suitable for the ultrasonic data from the input ultrasonic data. A specific region determination unit 32 specifies a specific region in a data space of the ultrasonic data. A specific parameter determination unit 34 inputs specific region data, which is the ultrasonic data in the specific region, out of the ultrasonic data to be processed to the learning model 28 that has performed learning, and determines specific parameters on the basis of the output of the learning model 28. An image quality adjusting processing unit executes image quality adjusting processing on the basis of the specific parameters for the ultrasonic data in the specific region, and executes image quality adjusting processing using predetermined parameters for image quality adjustment different from the specific parameters for the ultrasonic data outside the specific region.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] This specification discloses an improvement to an ultrasound diagnostic device. [Background technology]

[0002] Conventionally, there is known an ultrasonic diagnostic device that transmits and receives ultrasonic waves to a subject, forms an ultrasonic tomographic image based on the received signal, and displays the formed ultrasonic tomographic image (i.e., B-mode image) on a display. Conventionally, a technology for improving the image quality of an ultrasonic tomographic image in an ultrasonic diagnostic device has been proposed.

[0003] For example, Patent Document 1 discloses an ultrasound diagnostic device capable of automatically applying an appropriate image quality improvement process based on a change in the signal amount of a subject. In this ultrasound diagnostic device, the signal-to-noise ratio (S / N ratio) is improved in response to a depth-dependent change in an echo signal from a subject. Patent Document 2 discloses an ultrasound imaging system that uses channel data and beamforming data from an ultrasound probe as input, and uses the output of a neural network trained to form an ultrasound tomographic image to adjust an ultrasound tomographic image formed from the beamforming data. Patent Document 3 discloses a medical image processing device that cuts out different image regions from a plurality of medical images having different noise levels, and synthesizes the multiple image regions to obtain a high-quality medical image. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP2020-114294 Public Relations [Patent Document 2] Special Public Relations No. 2020-503142 [Patent Document 3] Public Relations for Patent Publication No. 2022-6869 Summary of the Invention [Problem to be solved by the invention]

[0005] Before an ultrasonic tomographic image is formed, various types of processing are performed on the received signal. By adjusting parameters related to these processes to appropriate values, an ultrasonic tomographic image with high image quality can be formed. Even after an ultrasonic tomographic image is formed, image quality improvement processing may be performed on the ultrasonic tomographic image. In this case, too, by adjusting parameters related to the image quality improvement processing to appropriate values, an ultrasonic tomographic image with high image quality can be formed.

[0006] In this specification, the processing of the received signal to form an ultrasonic tomographic image and the processing of the formed ultrasonic tomographic image to improve image quality are collectively referred to as image quality adjustment processing. In addition, the parameters related to the processing of the received signal to form an ultrasonic tomographic image and the parameters related to the processing of the formed ultrasonic tomographic image to improve image quality are collectively referred to as image quality adjustment parameters.

[0007] Here, the contents of the appropriate image quality adjustment process may differ for each region in the ultrasonic tomographic image. In other words, the appropriate image quality adjustment parameters may differ for each region in the ultrasonic tomographic image. Although not limited to this, for example, a region including a kidney in an ultrasonic tomographic image generally requires high contrast, so that the appropriate image quality adjustment process for the region is a process that realizes high contrast, while a region including a liver in an ultrasonic tomographic image generally requires high spatial resolution, so that the appropriate image quality adjustment process for the region is a process that realizes high spatial resolution.

[0008] An object of the ultrasonic diagnostic apparatus disclosed in this specification is to provide an ultrasonic diagnostic apparatus capable of performing adaptive image quality adjustment processing for each region in an ultrasonic tomographic image. [Means for solving the problem]

[0009] The ultrasound diagnostic device disclosed in the present specification includes a specific region determination unit that determines a specific region corresponding to a portion of an ultrasound tomographic image corresponding to the ultrasound data in a data space of ultrasound data obtained by transmitting and receiving ultrasound to a subject, and a specific parameter determination unit that determines a specific parameter, which is an image quality adjustment parameter for the specific region, based on the output of a learning model when ultrasound data within the specific region is input to a learning model that has been trained to output image adjustment parameters suitable for the ultrasound data from the input ultrasound data using learning data including ultrasound data obtained by transmitting and receiving ultrasound to a subject and teacher image adjustment parameters to be used in image quality adjustment processing for the ultrasound data. An image quality adjustment processing unit that performs image quality adjustment processing for the ultrasound data within the specific region based on the specific parameter, to adjust the image quality within the specific region of the ultrasound tomographic image, and performs image quality adjustment processing for the ultrasound data outside the specific region using a predetermined parameter for image quality adjustment different from the specific parameter.

[0010] According to this configuration, specific parameters suitable for the specific region are determined based on the output of a learning model to which ultrasound data within the specific region is input. Then, image quality adjustment processing can be performed using the specific parameters for ultrasound data within the specific region, and image quality adjustment processing can be performed using predetermined parameters for ultrasound data outside the specific region. This realizes adaptive image quality adjustment processing for each region in an ultrasound tomographic image.

[0011] The image quality adjustment processing unit may form an ultrasonic tomographic image in the specific region of the ultrasonic tomographic image, the image quality of which has been adjusted by an image formation model in which the specific parameter is a weighting parameter.

[0012] According to this configuration, an ultrasonic tomographic image within a specific region that has been subjected to image quality adjustment processing can be formed by the image formation model.

[0013] The image quality adjustment processing unit may execute an image quality smoothing process for smoothly changing image quality at a boundary between the inside of the specific region and the outside of the specific region.

[0014] According to this configuration, it is possible to smoothly change the image quality at the boundary between the inside and outside of a specific region on an ultrasonic tomographic image.

[0015] It is preferable to further include a display control unit that displays on a display unit a first ultrasonic tomographic image in which image quality adjustment processing based on the specific parameter has been performed for the specific region and image quality adjustment processing based on the predetermined parameter has been performed for outside the specific region, and a second ultrasonic tomographic image in which image quality adjustment processing based on the predetermined parameter has been performed for the entire region.

[0016] According to this configuration, the user can compare the first ultrasonic tomographic image with the second ultrasonic tomographic image.

[0017] The specific region determining unit may determine the specific region according to an instruction from a user of the ultrasound diagnostic apparatus.

[0018] According to this configuration, the user can perform different image quality adjustment processes for a desired area and other areas.

[0019] The image quality adjustment processing unit may execute image quality adjustment processing for adjusting image quality within the specific region of the ultrasonic tomographic image in accordance with an image quality adjustment policy according to instructions from a user of the ultrasonic diagnostic device.

[0020] According to this configuration, the user can perform image quality adjustment processing on ultrasound data within a specific region according to a desired image quality adjustment policy.

[0021] The specific parameter determination unit stores in memory a combination of user identification information that uniquely identifies a user of the ultrasound diagnostic device, a part of the subject included in a specific area determined in accordance with instructions from the user, and the image quality adjustment policy instructed by the user for the specific area, and when the specific area including the part is determined in accordance with instructions from the user, the image quality adjustment processing unit executes image quality adjustment processing to adjust the image quality within the specific area of ​​the ultrasound tomographic image in accordance with the user identification information of the user and the image quality adjustment policy associated with the part in the memory.

[0022] According to this configuration, the user can perform image quality adjustment processing of ultrasound data within a specific region according to a desired image adjustment policy without having to specify an image quality adjustment policy each time. Effect of the Invention

[0023] According to the ultrasound diagnostic device disclosed in this specification, it is possible to provide an ultrasound diagnostic device capable of performing adaptive image quality adjustment processing for each region in an ultrasound tomographic image. [Brief description of the drawings]

[0024] [Figure 1] 1 is a schematic diagram illustrating the configuration of an ultrasonic diagnostic apparatus according to an embodiment of the present invention. [Diagram 2] FIG. 13 is a diagram showing a first example of a specific area determination screen. [Diagram 3] FIG. 13 is a diagram showing a second example of a specific area determination screen. [Figure 4] FIG. 13 is a conceptual diagram showing a process for determining a specific parameter using a learning model. [Diagram 5] 1 is a conceptual diagram showing a process of forming an ultrasonic tomographic image using an image generation model based on specific parameters. [Figure 6] 11 is a graph showing the weight of the image quality adjustment process at the boundary between the inside and outside of a specific region. [Figure 7]11A and 11B are diagrams showing examples of displaying an ultrasonic tomographic image in which image quality adjustment processing based on specific parameters has been performed within a specific region, and an ultrasonic tomographic image in which image quality adjustment processing based on predetermined parameters has been performed over the entire region. [Figure 8] FIG. 13 is a diagram showing an example of an image quality adjustment policy decision screen. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] 1 is a schematic diagram of the configuration of an ultrasonic diagnostic apparatus 10 according to this embodiment. The ultrasonic diagnostic apparatus 10 is a medical device that is installed in a medical institution such as a hospital and used during ultrasonic examinations.

[0026] The ultrasonic diagnostic device 10 is a device that scans an ultrasonic beam on a subject and generates an ultrasonic image based on the received signal obtained by the scanning. In particular, in this embodiment, the ultrasonic diagnostic device 10 forms an ultrasonic tomographic image (B-mode image) in which the amplitude intensity of the reflected wave from the scanning surface is converted into brightness based on the received signal. Note that the ultrasonic diagnostic device 10 can also form other ultrasonic images, such as a Doppler image that represents the motion speed of tissue in the subject and is formed based on the difference (Doppler shift) between the frequencies of the transmitted wave and the received wave.

[0027] The transmission / reception unit 14, the signal processing unit 16, the image forming unit 18, the display control unit 20, the specific region determination unit 32, and the specific parameter determination unit 34 of the ultrasound diagnostic device 10 are configured by a processor. The processor includes at least one of a general-purpose processing device (e.g., a central processing unit (CPU)) and a dedicated processing device (e.g., a graphics processing unit (GPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or a programmable logic device). The processor may not be a single processing device, but may be configured by the cooperation of multiple processing devices located at physically separate locations. Furthermore, each of the above units may be realized by the cooperation of hardware such as a processor and software.

[0028] The ultrasonic probe 12 is a device that transmits and receives ultrasonic waves to and from a subject. The ultrasonic probe 12 has a transducer array that includes a plurality of transducer elements that transmit and receive ultrasonic waves to and from the subject.

[0029] The transmitting / receiving unit 14 transmits a transmission signal to the ultrasonic probe 12 (specifically, each transducer element of the transducer element array) under the control of a control unit 30 (described later). As a result, ultrasonic waves are transmitted from each transducer element to the subject.

[0030] The transmitting / receiving unit 14 also receives a reception signal from each transducer element that has received a reflected wave from the subject. The transmitting / receiving unit 14 has an adder and a plurality of delays corresponding to each transducer element, and performs a delay-and-sum process in which the phases of the reception signals from each transducer element are aligned and added by the adder and the plurality of delays. This forms a reception beam signal in which information indicating the signal strength of the reflected wave from the subject is aligned in the depth direction of the subject. In this specification, the reception signal output from each transducer element before the delay-and-sum process is called a channel signal, and the reception signal after the delay-and-sum process is called a reception beam signal as described above. In addition, the reception signal before being converted into data on the coordinate space of an ultrasonic tomographic image by the image forming unit 18 described later may be collectively called an RF (Radio Frequency) signal.

[0031] The received beam signals before being subjected to signal processing by the signal processor 16 can be temporarily stored in the memory 26 or in a memory (not shown) provided within the ultrasound diagnostic apparatus 10 for image quality adjustment processing, which will be described later.

[0032] The signal processing unit 16 performs various signal processing on the received beam signal from the transmitting / receiving unit 14, including filtering to apply a bandpass filter and detection processing. The various signal processing in the signal processing unit 16 is processing that may affect the image quality of the ultrasonic tomographic image formed by the image forming unit 18. That is, the signal processing unit 16 performs image quality adjustment processing on the received beam signal. The image quality adjustment processing performed by the signal processing unit 16 is performed based on image quality adjustment parameters. The image quality adjustment parameters referred to by the signal processing unit 16 may be set by the user of the ultrasonic diagnostic device 10. In this embodiment, the image quality adjustment parameters referred to by the signal processing unit 16 may be determined by a specific parameter determination unit 34 described later.

[0033] The image forming unit 18 forms an ultrasonic tomographic image (B-mode image) based on the received beam signal that has been signal-processed by the signal processing unit 16. First, the image forming unit 18 converts the received beam signal into data in the coordinate space of an ultrasonic image. In this specification, the converted data is called a coordinate transformation signal. Then, the image forming unit 18 forms an ultrasonic tomographic image based on the coordinate transformation signal.

[0034] Furthermore, the image forming unit 18 performs image quality improvement processing, that is, image quality adjustment processing, on the formed ultrasonic tomographic image. The image quality adjustment processing performed by the image forming unit 18 is also executed based on image quality adjustment parameters. The image quality adjustment parameters referred to by the image forming unit 18 may also be set by the user of the ultrasonic diagnostic device 10, but in this embodiment, the image quality adjustment parameters referred to by the image forming unit 18 may also be determined by the specific parameter determination unit 34.

[0035] The ultrasonic tomographic image before the image quality adjustment processing by the image forming unit 18 can also be temporarily stored in the memory 26 or a memory (not shown) provided in the ultrasonic diagnostic device 10 for the image quality adjustment processing described later.

[0036] The display control unit 20 controls the display of the ultrasonic tomographic image formed by the image forming unit 18 and various other information on the display 22. The display 22 as a display unit is a display device constituted by, for example, a liquid crystal display or an organic EL (Electro Luminescence) display.

[0037] The input interface 24 is configured with, for example, a button, a trackball, a touch panel, etc. The input interface 24 is used to input a user's command to the ultrasound diagnostic apparatus 10.

[0038] The memory 26 includes a hard disk drive, a solid state drive (SSD), an embedded multi media card (eMMC), a read only memory (ROM), etc. An ultrasound diagnostic program for operating each unit of the ultrasound diagnostic device 10 is stored in the memory 26. The ultrasound diagnostic program can also be stored in a computer-readable non-transitory storage medium such as a universal serial bus (USB) memory or a CD-ROM. The ultrasound diagnostic device 10 can read and execute the ultrasound diagnostic program from such a storage medium.

[0039] 1, the memory 26 stores a learning model 28. The learning model 28 is, for example, a Convolutional Neural Network (CNN), but may be any model as long as it exhibits the functions described below.

[0040] The learning model 28 is trained to output image adjustment parameters suitable for input ultrasound data from the ultrasound data, using training data including ultrasound data obtained by transmitting and receiving ultrasound waves to and from a subject and teacher image adjustment parameters to be used in image quality adjustment processing for the ultrasound data. The ultrasound data included in the training data of the learning model 28 includes a channel signal, a received beam signal, a coordinate transformation signal, and an ultrasonic tomographic image. In this embodiment, the learning processing of the learning model 28 is performed by a learning processing device different from the ultrasound diagnostic device 10, and the learning model 28 that has been fully trained is stored in the memory 26. However, the ultrasound diagnostic device 10 may have a learning processing unit (not shown in FIG. 1), and the learning processing unit may perform the learning processing of the learning model 28. In this case, the ultrasound diagnostic device 10 serves as the learning processing device.

[0041] Details of the learning process of learning model 28 are as follows. The processor of the learning processing device inputs ultrasound data as learning data to learning model 28. Learning model 28 predicts and outputs image quality adjustment parameters suitable for the input ultrasound data. The processor of the learning processing device changes the parameters in learning model 28 so that the difference between the output data of learning model 28 and the teacher image adjustment parameters included in the learning data becomes smaller. By repeating such processing, learning model 28 is trained. Learning model 28 that has been sufficiently trained becomes able to predict image adjustment parameters suitable for the input ultrasound data with high accuracy.

[0042] The control unit 30 includes at least one of a general-purpose processor (e.g., a CPU (Central Processing Unit)) and a dedicated processor (e.g., a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a programmable logic device). The control unit 30 may not be a single processing device, but may be configured by the cooperation of multiple processing devices located at physically separate locations. The control unit 30 controls each part of the ultrasound diagnostic device 10 according to an ultrasound diagnostic program stored in the memory 26.

[0043] The specific region determining unit 32 determines a specific region corresponding to a portion of an ultrasonic tomographic image (in other words, an ultrasonic tomographic image formed based on the ultrasonic data, or the ultrasonic data itself when the ultrasonic data is an ultrasonic tomographic image) corresponding to the ultrasonic data obtained by transmitting and receiving ultrasonic waves to and from the subject, i.e., a channel signal, a received beam signal, a coordinate transformation signal, or a data space of an ultrasonic tomographic image. Note that, in FIG. 1, the specific region determining unit 32 is illustrated as receiving ultrasonic data (i.e., a received beam signal) from the signal processing unit 16, but as described above, the ultrasonic data may be various signals, so that the specific region determining unit 32 may receive a channel signal as ultrasonic data from the transmitting / receiving unit 14, or may receive a coordinate transformation signal or an ultrasonic tomographic image as ultrasonic data from the image forming unit 18.

[0044] As a first method for identifying a specific region, the specific region determining unit 32 can identify the specific region by analyzing ultrasound data (here, particularly, a received beam signal, a coordinate transformation signal, or an ultrasound tomographic image).

[0045] For example, a learning model for area identification is prepared in advance, which is trained using ultrasound data and training data including the positions and names of parts (such as organs) included in the ultrasound data, and is trained to predict and output the positions and names of parts included in the ultrasound data based on the input ultrasound data.The specific area determination unit 32 can then identify the specific area based on the output of the trained learning model for area identification when the ultrasound data to be processed is input.

[0046] In addition, for example, partial ultrasound data indicating a desired specific region can be prepared in advance as template data, and the similarity between each part of the ultrasound data to be treated and the template data can be calculated, and the region where the similarity is greater than or equal to a threshold and is at its maximum can be identified as the specific region.

[0047] As a second method for identifying a specific region, the specific region determining section 32 can identify a specific region according to an instruction from a user of the ultrasound diagnostic apparatus 10.

[0048] For example, the image forming unit 18 uses an existing technology such as a learning model to form a segmentation image by dividing an ultrasonic tomographic image into a plurality of regions (e.g., corresponding to each organ), and the display control unit 20 causes the display 22 to display the formed segmentation image SEI as shown in FIG. 2. In the example of FIG. 2, a segmentation image SEI including a liver region, a gallbladder region, and a kidney region is shown. Then, the user selects a desired region from the displayed segmentation image SEI with a cursor. The specific region determining unit 32 identifies the region of the segmentation image SEI selected by the user as a specific region.

[0049] Furthermore, the image forming unit 18 uses existing technology such as a learning model to identify one or more area names (e.g., organ names) included in the ultrasonic tomographic image, and the display control unit 20 causes the display 22 to display an area name list RL, which is a list of the identified area names, together with the ultrasonic tomographic image USI, as shown in Fig. 3. Then, the user selects a desired area from the displayed area name list RL with the cursor. The identified area determination unit 32 identifies the area selected by the user as the identified area.

[0050] The method of specifying a specific region according to an instruction from a user is not limited to the above method. For example, the user may directly specify a specific region on the ultrasonic tomographic image displayed on the display 22 using a cursor or the like.

[0051] Returning to FIG. 1, the specific parameter determination unit 34 uses the trained learning model 28 to determine specific parameters, which are image quality adjustment parameters for the specific area identified by the specific area determination unit 32.

[0052] Specifically, the specific parameter determination unit 34 extracts specific region data, which is ultrasound data within a specific region, from the ultrasound data to be processed (i.e., channel signals, received beam signals, coordinate transformation signals, or ultrasound tomographic images). Then, as shown in FIG. 4, the specific parameter determination unit 34 inputs the specific region data to the trained learning model 28. The learning model 28 predicts and outputs image quality adjustment parameters suitable for the input specific region data. The specific parameter determination unit 34 determines specific parameters based on the output of the learning model 28.

[0053] It should be noted here that the specific parameter determination unit 34 does not input the entire ultrasound data, but only a portion of it, the specific region data, to the learning model 28. This allows the learning model 28 to predict and output image quality adjustment parameters specialized for the specific region data, rather than the entire ultrasound data. In other words, the specific parameters can be said to be image quality adjustment parameters specialized for the specific region.

[0054] For example, when the specific region is a kidney region, high contrast is required for the kidney region, so the specific parameters are determined to reduce the dynamic range and to implement filter processing with emphasis on high contrast. For example, when the specific region is a liver region, smooth depiction of the substance is required for the liver region, so the specific parameters are determined to implement TGC (Time Gain Compensation) that makes the luminance uniform from shallow to deep parts and filter processing with emphasis on high spatial resolution. For example, when the specific region is a gallbladder region, it is required to be able to detect abnormalities such as polyps in the gallbladder region, so the specific parameters are determined to implement filter processing that reduces artifacts that hinder the discovery of abnormality detection, such as multiple reflection artifacts. For example, when the specific region is a breast region, it is required to diagnose the mammary gland region in the breast region, so the specific parameters are determined to provide a high-definition image only for the mammary gland region to be diagnosed, and reduce the luminance of other regions so as not to hinder diagnosis.

[0055] The signal processing unit 16 or the image forming unit 18 serving as an image quality adjustment processing unit executes image quality adjustment processing on the ultrasound data within the specific region based on the specific parameters determined by the specific parameter determination unit 34 .

[0056] First, the image quality adjustment process by the signal processing unit 16 will be described. The signal processing unit 16 performs image quality adjustment process on the stored reception beam data before signal processing based on the specific parameters determined by the specific parameter determination unit 34. Specifically, the signal processing unit 16 performs image quality adjustment process on the reception beam signals within the specific region based on the specific parameters, and performs image quality adjustment process on the reception beam signals outside the specific region using a predetermined parameter for image quality adjustment different from the specific parameter. The predetermined parameter for image quality adjustment may be, for example, a parameter set in advance by the user or a parameter predetermined in the ultrasound diagnostic device 10. In this way, the signal processing unit 16 can be an image quality adjustment processing unit.

[0057] Next, the image quality adjustment process by the image forming unit 18 will be described. The image forming unit 18 performs image quality adjustment process on the stored ultrasonic tomographic image before image quality adjustment process based on the specific parameters determined by the specific parameter determination unit 34. Specifically, the image forming unit 18 performs image quality adjustment process on the ultrasonic tomographic image in the specific region based on the specific parameters, and performs image quality adjustment process on the ultrasonic tomographic image outside the specific region using a predetermined parameter for image quality adjustment different from the specific parameter. Here again, the predetermined parameter for image quality adjustment may be, for example, a parameter set in advance by the user or a parameter predetermined in the ultrasonic diagnostic device 10. In this way, the image forming unit 18 can also be an image quality adjustment processing unit.

[0058] As described above, the signal processing unit 16 or the image forming unit 18 performs different image quality adjustment processes based on different image quality adjustment parameters on the ultrasound data within the specific region and the ultrasound data outside the specific region. This allows adaptive image quality adjustment processes to be performed on each region in the ultrasound tomographic image. For example, image quality adjustment processes can be performed on the kidney region, which is a specific region, to achieve high contrast, while image quality adjustment processes can be performed on the other regions to achieve appropriate contrast without being too high.

[0059] Furthermore, by having the specific area determination unit 32 determine a specific area in accordance with an instruction from the user, the user can carry out different image quality adjustment processes for a desired area and other areas.

[0060] In particular, in this embodiment, for example, even for regions corresponding to the same organ, image quality adjustment processing may be performed for each ultrasound data with different specific parameters. For example, assume that a liver region is determined as a first specific region in the first ultrasound data, and the liver region is determined as a second specific region in second ultrasound data different from the first ultrasound data. In this case, if the contents of the first ultrasound data in the first specific region and the second ultrasound data in the second specific region are different, the learning model 28 may output different image quality adjustment parameters when the first ultrasound data in the first specific region is input and when the second ultrasound data in the second specific region is input. As a result, according to this embodiment, different image quality adjustment processing may be performed for the same liver region depending on the ultrasound data.

[0061] Furthermore, the specific parameter determination unit 34 inputs only the specific region data of the ultrasound data, rather than the entire ultrasound data, to the learning model 28. This can reduce the amount of calculation required for the learning model 28 to predict and output the specific parameters.

[0062] In this embodiment, the image forming unit 18 forms an ultrasonic tomographic image in a specific region based on the received beam data, and then performs image quality adjustment processing of the ultrasonic tomographic image based on a specific parameter. However, as shown in FIG. 5, the image forming unit 18 as an image quality adjustment processing unit may form an ultrasonic tomographic image in a specific region whose image quality is adjusted by an image formation model 40 in which the specific parameter is a weight parameter. The image formation model 40 may be, for example, a learning model such as a GAN (Generative Adversarial Network). In this case, the learning model 28 may be previously trained so that the format of the output data is a format suitable for the image formation model 40.

[0063] In this embodiment, the image forming unit 18 performs image quality adjustment processing of an ultrasonic tomographic image within a specific region based on specific parameters, or performs processing of forming an ultrasonic tomographic image within a specific region using the image formation model 40, but these processes may also be performed by the specific parameter determination unit 34.

[0064] In this embodiment, different image quality adjustment processes are performed inside and outside the specific area. Therefore, the difference in image quality may become significant at the boundary between the inside and outside of the specific area. Therefore, the signal processing unit 16 or the image forming unit 18 as an image quality adjustment processing unit may perform an image quality smoothing process to smoothly change the image quality at the boundary between the inside and outside of the specific area.

[0065] Fig. 6 is a graph showing the weight of image quality adjustment processing at the boundary between inside and outside the specific region. In Fig. 6, boundary B between inside and outside the specific region is indicated by a dashed line, with the area to the left of boundary B being outside the specific region and the area to the right of boundary B being inside the specific region. A certain area near boundary B between outside and inside the specific region is called boundary region BR. The solid line graph represents the weight of image quality adjustment processing using specific parameters for inside the specific region, and the dashed line graph represents the weight of image quality adjustment processing using specific parameters for outside the specific region.

[0066] As shown in FIG. 6, in the specific region other than the boundary region BR, the weight of the image quality adjustment process using the specific parameter is "1", and the weight of the image quality adjustment process using the predetermined parameter is "0". This means that the ultrasound data in the specific region other than the boundary region BR is subjected to image quality adjustment process using only the specific parameter. In the boundary region BR, the weight of the image quality adjustment process using the specific parameter gradually decreases from inside the specific region to outside the specific region (from right to left in the graph of FIG. 6), and conversely, the weight of the image quality adjustment process using the predetermined parameter increases. This means that in the boundary region BR, the result of the image quality adjustment process using the specific parameter and the result of the image quality adjustment process using the predetermined parameter are synthesized according to their respective weights. In this embodiment, this process corresponds to image quality smoothing process. Outside the specific region other than the boundary region BR, the weight of the image quality adjustment process using the specific parameter is "0", and the weight of the image quality adjustment process using the predetermined parameter is "1". This means that the ultrasound data in the specific region other than the boundary region BR is subjected to image quality adjustment process using only the specific parameter.

[0067] In this way, in the boundary region BR, by combining the results of the image quality adjustment process using specific parameters and the results of the image quality adjustment process using predetermined parameters, it is possible to smoothly change the image quality at the boundary between the specific region and the outside of the specific region on the ultrasound tomographic image.

[0068] The display control unit 20 displays the ultrasonic tomographic image that has been subjected to image quality adjustment processing on the display 22. As shown in Fig. 7, the display control unit 20 may display a first ultrasonic tomographic image USI-h, in which image quality adjustment processing based on specific parameters has been performed on a specific region and image quality adjustment processing based on predetermined parameters has been performed on an outside of the specific region, and a second ultrasonic tomographic image USI-n, in which image quality adjustment processing based on predetermined parameters has been performed on the entire region, on the display 22. This allows the user to compare the first ultrasonic tomographic image USI-h and the second ultrasonic tomographic image USI-n.

[0069] In order to display the second ultrasonic tomographic image USI-n on the display 22, the signal processing unit 16 performs image quality adjustment processing using predetermined parameters on the entire received frame signal for one frame, and the image forming unit 18 performs image quality adjustment processing using predetermined parameters on the entire formed ultrasonic tomographic image. In this way, the second ultrasonic tomographic image USI-n is formed.

[0070] As described above, appropriate image quality adjustment processes for each region of an ultrasonic tomographic image may differ from one another, but there may be multiple appropriate image quality adjustment processes for a certain region. For example, when the specific region is a kidney region, appropriate image quality adjustment processes for the kidney region include a process for increasing resolution, a process for improving contrast, and a process for improving depth of penetration. Each of these processes is performed using different image adjustment parameters (i.e., specific parameters).

[0071] Therefore, the signal processing unit 16 or the image forming unit 18 as an image quality adjustment processing unit may execute image quality adjustment processing to adjust the image quality within a specific region of the ultrasonic tomographic image in accordance with an image quality adjustment policy in response to instructions from a user of the ultrasonic diagnostic device 10.

[0072] As a premise, the learning model 28 is trained to output image adjustment parameters suitable for the ultrasound data and in accordance with the image quality adjustment policy from the input ultrasound data and image quality adjustment policy, using learning data including ultrasound data, an image quality adjustment policy, and teacher image adjustment parameters to be used in image quality adjustment processing for the ultrasound data in accordance with the image quality adjustment policy. This may be configured so that one learning model 28 can accept the image quality adjustment policy, or may be realized by training a different learning model 28 for each image quality adjustment policy.

[0073] The display control unit 20 is a diagram showing an example of an image quality adjustment policy decision screen as shown in FIG. 8. The example of FIG. 8 is an example in which the user selects a kidney region as a specific region from the above-mentioned region name list RL. The display control unit 20 further causes the display 22 to display a policy list PL showing multiple image quality adjustment policies of image quality adjustment processing suitable for the selected specific region (kidney region in the example of FIG. 8). When the user selects a desired image adjustment policy from the policy list PL, the specific parameter decision unit 34 inputs the specific region data corresponding to the kidney region, which is the specific region, and the image adjustment policy selected by the user to the learning model 28. The learning model 28 predicts and outputs image quality adjustment parameters suitable for the input specific region data and in accordance with the input image adjustment policy. The specific parameter decision unit 34 decides the specific parameters based on the output of the learning model 28. Thereafter, the signal processing unit 16 or the image forming unit 18 executes image quality adjustment processing on the ultrasound data in the specific region based on the specific parameters.

[0074] It is considered that the same user often performs image quality adjustment processing with the same image adjustment policy for specific regions including the same part of a subject. Therefore, in this embodiment, the signal processing unit 16 or the image forming unit 18 as an image quality adjustment processing unit stores an image adjustment policy selected by a certain user for a specific region including a certain part, and when a specific region including that part is next determined according to an instruction from the user, it is preferable to determine an image adjustment policy for the newly determined specific region based on the image adjustment policy previously selected by the user for the specific region including that part.

[0075] Specifically, each time the user selects an image adjustment policy on the image quality adjustment policy decision screen shown in Fig. 8, the specific parameter decision unit 34 accumulates and stores in a policy database (not shown in Fig. 1) constructed in the memory 26 a combination of a user ID as user identification information that uniquely identifies the user, a part of the subject included in the specific region decided in accordance with an instruction from the user ("kidney" in the example of Fig. 8), and an image quality adjustment policy ("improvement of contrast" in the example of Fig. 8) instructed by the user for the specific region. The user ID of the user can be identified by the control unit 30 authenticating the user when the user starts using the ultrasound diagnostic apparatus 10.

[0076] Then, for example, for other ultrasound data, when the specific parameter determination unit 34 determines a specific region including the part ("kidney" in this example) according to the instruction of the user, the specific parameter determination unit 34 refers to the policy database and specifies the user ID of the user and the image adjustment policy associated with the part. In this example, "improve contrast" is specified. Thereafter, the specific parameter determination unit 34 inputs the specific region data corresponding to the kidney region, which is the specific region, and the specified image adjustment policy to the learning model 28, and determines the specific parameter based on the output of the learning model 28. Needless to say, the specific parameter is an image adjustment parameter in accordance with the specified image adjustment policy. Thereafter, the signal processing unit 16 or the image forming unit 18 performs image quality adjustment processing on the ultrasound data in the specific region based on the specific parameter (in other words, according to the specified image adjustment policy).

[0077] This allows the user to perform image quality adjustment processing of ultrasound data within a specific region according to a desired image adjustment policy without the need for the user to specify an image quality adjustment policy each time.

[0078] Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various modifications are possible without departing from the spirit of the present invention. [Explanation of symbols]

[0079] 10 ultrasound diagnostic device, 12 ultrasound probe, 14 transmission / reception unit, 16 signal processing unit, 18 image formation unit, 20 display control unit, 22 display, 24 input interface, 26 memory, 28 learning model, 30 control unit, 32 specific region determination unit, 34 specific parameter determination unit, 40 image formation model.

Claims

1. a specific region determining unit that determines a specific region corresponding to a portion of an ultrasonic tomographic image corresponding to the ultrasonic data in a data space of ultrasonic data obtained by transmitting and receiving ultrasonic waves to and from a subject; a specific parameter determination unit that uses learning data including ultrasound data obtained by transmitting and receiving ultrasound waves to a subject and teacher image adjustment parameters to be used in image quality adjustment processing for the ultrasound data to determine specific parameters, which are image quality adjustment parameters for the specific region, based on the output of a learning model when ultrasound data within the specific region is input to the learning model that has been trained to output image adjustment parameters suitable for the ultrasound data from the input ultrasound data; and an image quality adjustment processing unit that performs an image quality adjustment process for adjusting image quality within the specific region of the ultrasonic tomographic image for the ultrasonic data within the specific region based on the specific parameter, and performs an image quality adjustment process for adjusting image quality outside the specific region of the ultrasonic tomographic image for the ultrasonic data outside the specific region using a predetermined parameter for image quality adjustment different from the specific parameter; An ultrasonic diagnostic apparatus comprising:

2. the image quality adjustment processing unit forms the ultrasonic tomographic image in the specific region of the ultrasonic tomographic image, the image quality of which is adjusted by an image formation model in which the specific parameter is a weight parameter; 2. The ultrasonic diagnostic apparatus according to claim 1 .

3. the image quality adjustment processing unit executes an image quality smoothing process for smoothly changing image quality at a boundary between inside the specific region and outside the specific region; 2. The ultrasonic diagnostic apparatus according to claim 1 .

4. a display control unit that displays on a display unit a first ultrasonic tomographic image in which image quality adjustment processing based on the specific parameter has been performed for the specific region and image quality adjustment processing based on the predetermined parameter has been performed for the outside of the specific region, and a second ultrasonic tomographic image in which image quality adjustment processing based on the predetermined parameter has been performed for the entire region; The ultrasonic diagnostic apparatus according to claim 1 , further comprising:

5. the specific region determining unit determines the specific region in accordance with an instruction from a user of the ultrasound diagnostic device.

2. The ultrasonic diagnostic apparatus according to claim 1 .

6. the image quality adjustment processing unit executes an image quality adjustment process for adjusting image quality within the specific region of the ultrasonic tomographic image in accordance with an image quality adjustment policy according to an instruction from a user of the ultrasonic diagnostic device.

2. The ultrasonic diagnostic apparatus according to claim 1 .

7. the specific parameter determination unit stores in a memory a combination of user identification information that uniquely identifies a user of the ultrasound diagnostic device, a part of the subject included in a specific region determined in accordance with an instruction from the user, and the image quality adjustment policy instructed by the user for the specific region; When the specific region including the part is determined in accordance with an instruction from the user, the image quality adjustment processing unit executes an image quality adjustment process for adjusting the image quality within the specific region of the ultrasonic tomographic image in accordance with the user identification information of the user and the image quality adjustment policy associated with the part in the memory.

7. The ultrasonic diagnostic apparatus according to claim 6.

Citation Information

Patent Citations

  • Ultrasonic diagnostic apparatus and learning program

    JP2020114294A

  • Ultrasound imaging system using neural networks to derive imaging data and tissue information

    JP2020503142A

  • Image processing apparatus, medical imaging apparatus, and image processing program

    JP2022006869A