Ultrasound system with adaptive nonlinear axial scale

Abstract

An ultrasound system can acquire ultrasound data of a region of interest of a subject. The ultrasound system can generate an ultrasound image of the region of interest of the subject based on acquiring the ultrasound data of the region of interest of the subject, where the ultrasound image has a non-linear axial scale. The ultrasound system can display the ultrasound image with the non-linear axial scale based on generating the ultrasound image.

Description

Statement on Federal Government Funding for Research and Development

[0001] This invention was made with government support under license number 75A50123C00035 granted by the Biomedical Advanced Research and Development Authority (BARDA). The government owns certain rights to the invention. Technical Field

[0002] This disclosure relates to an ultrasound system configured to generate and display ultrasound images with a nonlinear axial scale. Background Technology

[0003] Lung ultrasound is a minimally invasive imaging modality that enables real-time visualization of lung structures. It is particularly effective in assessing pleural pathologies such as pleurisy, pleural effusion, empyema, and pleural tumors. To perform lung ultrasound imaging, the ultrasound probe of the system emits ultrasound signals towards the patient's lungs and then receives the echo signals reflected or backscattered from the lung structures. The system then generates ultrasound images based on these echo signals. Lung ultrasound images can include elements such as pleural lines, lung lines, fractal lines, A-lines, B-lines, and sub-B-lines. The system displays these images for clinicians to review and evaluate. Clinicians can use the various elements in the lung ultrasound images to assess any pathology in the patient.

[0004] Ultrasound images may include a scale indicating the relationship between distances on the ultrasound image and their corresponding actual distances within the patient's body. Generally, ultrasound images have a linear axial scale. In this case, a specific distance measured anywhere on the ultrasound image corresponds to the same actual distance within the patient's body. The scale may also refer to a series of visual indicators at specific intervals used to measure distances in the ultrasound image. In this case, the ultrasound image may include visual indicators provided at equidistant intervals.

[0005] In some cases, the content of a near-field ultrasound image may be relatively more clinically diagnostic than that of a far-field ultrasound image. In such cases, clinicians may be relatively more interested in viewing the near-field content at a higher resolution than in the far-field content. However, in ultrasound images with a linear axial scale, the near-field resolution can be the same as the far-field resolution. As a specific example, in lung ultrasound, clinicians may be relatively more interested in visualizing pleural pathology, which is typically located at about 2 cm to 3 cm, than visualizing B-lines, which typically extend from 10 cm to 18 cm. Nevertheless, for the purpose of performing an assessment, clinicians may still be interested in visualizing both the pleura and the B-lines. These two interests can be contradictory in some scenarios, especially with fan-shaped two-dimensional (2D) or three-dimensional (3D) probes, because they have a narrow field of view at shallow depths. This can lead to suboptimal pleural visualization when the image depth is set for optimal B-line visualization. Summary of the Invention

[0006] This invention describes concepts in more detail in specific embodiments. It should not be used to identify essential features of the claimed subject matter, nor should it be used to limit the scope of the claimed subject matter.

[0007] In one aspect, an ultrasound system may include: a lens, an acoustic matching layer, an acoustic dematching layer, and a plurality of transducer elements; a memory configured to store instructions; and one or more processors configured to execute instructions to: acquire ultrasound data of a region of interest (ROI) of a subject; generate an ultrasound image of the RIO of the subject based on the acquired ultrasound data of the RIO, wherein the ultrasound image has a non-linear axial scale; and display the ultrasound image with the non-linear axial scale based on the generated ultrasound image.

[0008] In one aspect, one or more processors may be further configured to transform the original ultrasound data before performing a scan conversion, such that a first region of the ultrasound image has a first linear axial scale, and a second region of the ultrasound image has a second linear axial scale different from the first linear axial scale.

[0009] In one aspect, one or more processors may be further configured to scale and transform the raw ultrasound data by applying the μ law prior to performing the scan transformation.

[0010] In one aspect, one or more processors may be further configured to determine the depth of a scale used to transform ultrasound images.

[0011] In one aspect, a first region of the ultrasound image includes a first linear axial scale, and a second region of the ultrasound image includes a second linear axial scale that is different from the first linear axial scale.

[0012] In one aspect, the ultrasound image includes a first region corresponding to a first depth range within the body of the subject, wherein the ultrasound image includes a second region corresponding to a second depth range within the body of the subject, wherein the first depth range is the same as the second depth range, and wherein the first region includes a first number of pixels, the first number of pixels being greater than a second number of pixels in the second region.

[0013] On the one hand, ultrasound images are single, continuous ultrasound images.

[0014] In another aspect, a method may include: acquiring ultrasound data of a region of interest (ROI) of a subject; generating an ultrasound image of the RIO based on the acquired ultrasound data of the RIO, wherein the ultrasound image has a non-linear axial scale; and displaying the ultrasound image with the non-linear axial scale based on the generated ultrasound image.

[0015] In one aspect, the method may include transforming the original ultrasound data before performing a scan conversion, such that a first region of the ultrasound image has a first linear axial scale, and a second region of the ultrasound image has a second linear axial scale different from the first linear axial scale.

[0016] In one aspect, the method may include scaling the raw ultrasound data by applying the μ law before performing the scan transformation.

[0017] In one aspect, the method may include determining the depth of a scale used to convert ultrasound images.

[0018] In one aspect, a first region of the ultrasound image includes a first linear axial scale, and a second region of the ultrasound image includes a second linear axial scale that is different from the first linear axial scale.

[0019] In one aspect, the ultrasound image includes a first region corresponding to a first depth range within the body of the subject, wherein the ultrasound image includes a second region corresponding to a second depth range within the body of the subject, wherein the first depth range is the same as the second depth range, and wherein the first region includes a first number of pixels, the first number of pixels being greater than a second number of pixels in the second region.

[0020] On the one hand, ultrasound images are single, continuous ultrasound images.

[0021] In another aspect, a non-transitory computer-readable medium stores instructions that, when executed by one or more processors, cause one or more processors to: acquire ultrasound data of a region of interest of a subject; generate an ultrasound image of the region of interest of the subject based on the acquired ultrasound data of the region of interest of the subject, wherein the ultrasound image has a non-linear axial scale; and display the ultrasound image with the non-linear axial scale based on the generated ultrasound image.

[0022] On the one hand, the instructions may also cause one or more processors to transform the original ultrasound data before performing the scan conversion, such that a first region of the ultrasound image has a first linear axial scale and a second region of the ultrasound image has a second linear axial scale different from the first linear axial scale.

[0023] On the other hand, the instructions also enable one or more processors to scale and transform the raw ultrasound data by applying the μ law before performing a scan transformation.

[0024] On the other hand, the instructions can also enable one or more processors to determine the depth of the scale used to transform ultrasound images.

[0025] In one aspect, a first region of the ultrasound image includes a first linear axial scale, and a second region of the ultrasound image includes a second linear axial scale that is different from the first linear axial scale.

[0026] In one aspect, the ultrasound image includes a first region corresponding to a first depth range within the body of the subject, wherein the ultrasound image includes a second region corresponding to a second depth range within the body of the subject, wherein the first depth range is the same as the second depth range, and wherein the first region includes a first number of pixels, the first number of pixels being greater than a second number of pixels in the second region. Attached Figure Description

[0027] Figure 1 This is a diagram of exemplary components of an ultrasound system used to generate and display ultrasound images with a non-linear axial scale.

[0028] Figure 2 This is a diagram of an exemplary ultrasound probe used in an ultrasound system for generating and displaying ultrasound images with a non-linear axial scale.

[0029] Figure 3 This is a flowchart of an exemplary process for generating and displaying ultrasound images with a non-linear axial scale.

[0030] Figure 4A This is a diagram of an exemplary ultrasound image with a linear axial scale.

[0031] Figure 4B This is a diagram of an exemplary ultrasound image with a non-linear axial scale.

[0032] Figure 5 This is a diagram of an exemplary 3D ultrasound image with a non-linear axial scale.

[0033] Figure 6 This is a flowchart of an exemplary process for determining the depth of a scale used to convert ultrasound images.

[0034] Figures 7A to 7C This is a diagram of a scaling technique used to generate ultrasound images with a non-linear axial scale.

[0035] Figure 8 This is a flowchart of an exemplary process for displaying ultrasound images with linear or non-linear axial scales based on user selection.

[0036] Figure 9 This is a flowchart of an exemplary process for displaying distances related to ultrasound images based on user selection.

[0037] Figure 10 This is a diagram illustrating an exemplary process for training, deploying, and monitoring artificial intelligence (AI) models. Detailed Implementation

[0038] As mentioned above, in some cases, the content of near-field ultrasound images may be relatively more clinically diagnostic than that of far-field ultrasound images. In such cases, clinicians may be more interested in viewing the near-field content at a higher resolution than in the far-field content. However, in ultrasound images with a linear axial scale, the resolution of the near field can be the same as that of the far field.

[0039] Some embodiments described herein provide an ultrasound system that acquires ultrasound data of a patient's region of interest (ROI); generates an ultrasound image of the RIO based on the acquired RIO data, the image having a non-linear axial scale; and displays the generated ultrasound image with the non-linear axial scale. For example, in the context of lung ultrasound, the system can transform the raw ultrasound data along an axial axis based on the pleural location, so that the pleura is presented at high resolution, while the B-line extends as a straight line at lower resolution to the end of the ultrasound image. In this way, the embodiments described herein enable clinicians to observe near-field details in a single continuous ultrasound image while maintaining full imaging depth.

[0040] Therefore, some embodiments of this paper provide improvements to the field of ultrasound imaging technology by offering a novel method of generating single, continuous ultrasound images with a nonlinear axial scale. Furthermore, some embodiments of this paper provide improvements to ultrasound systems by enabling the ultrasound system to generate single, continuous images with a nonlinear axial scale. The nonlinear axial scale allows for simultaneous observation of the near and far fields in a manner that provides enhanced detail and resolution in the near field. Thus, clinicians can observe near-field details while maintaining full imaging depth in a single, continuous ultrasound image.

[0041] Figure 1 This is a diagram of exemplary components of an ultrasound system 100 used to generate and display ultrasound images with a non-linear axial scale. (See diagram for example.) Figure 1 As shown, the ultrasound system 100 may include an ultrasound probe 102, a transmit beamformer 104, a transmitter 106, a receiver 108, a receive beamformer 110, a user input device 112, a processor 114, a display 116, a memory 118, and a communication interface 120. The aforementioned components may be connected via wired or wireless connections.

[0042] The ultrasound probe 102 can be configured to acquire ultrasound data. For example, the ultrasound probe 102 can be a linear probe, a phased array probe, a curved linear probe coupled to a positioning and tracking system, a mechanically manipulated linear array transducer, a phased array transducer, a curved linear array transducer, an electronically manipulated 2D transducer array, an electronic 3D (e3D) probe, an electronic 4D (e4D) probe, or a low-profile wearable patch version of any of the aforementioned probes. According to one embodiment, the ultrasound probe 102 can be configured to generate ultrasound signals, emit ultrasound signals toward the region of interest of the subject, receive echo ultrasound signals backscattered from the region of interest of the subject, generate ultrasound data based on the echo ultrasound signals, and output ultrasound data.

[0043] Transmit beamformer 104 can be configured to apply a delay time to the electrical signal provided to the elements of ultrasound probe 102 to focus the corresponding ultrasound signal at the region of interest. Transmitter 106 can be configured to send an electrical signal to the elements of ultrasound probe 102 to drive the elements of ultrasound probe 102 to emit an ultrasound signal toward the region of interest. Elements of ultrasound probe 102 can be configured to receive the electrical signal from transmitter 106, convert the electrical signal into an ultrasound signal, and emit the ultrasound signal toward the region of interest. Elements of ultrasound probe 102 can be configured to receive the echo ultrasound signal backscattered from the region of interest, convert the echo ultrasound signal into an electrical signal, and provide the electrical signal to receiver 108. Receiver 108 can be configured to receive the electrical signal from the elements and provide the electrical signal to receiver beamformer 110. Receiver beamformer 110 can apply a delay time to the electrical signal received from the elements of ultrasound probe 102.

[0044] User input device 112 can be configured to receive user input and provide the user input to processor 114. User input device 112 can be a user interface. For example, user input device 112 can be a touch screen display, keyboard, keypad, mouse, button, switch, microphone, etc. Additionally or alternatively, user input device 112 can be configured to sense information. For example, user input device 112 can sense information from an electromagnetic positioning system, inertial measurement system, accelerometer, gyroscope, actuator, etc.

[0045] Processor 114 may be configured to perform the operations described herein. For example, processor 114 may be a central processing unit (CPU), graphics processing unit (GPU), accelerated processing unit (APU), microprocessor, microcontroller, digital signal processor (DSP), field-programmable gate array (FPGA), application-specific integrated circuit (ASIC), or another type of processing component. Processor 114 may be implemented in hardware, firmware, or a combination of hardware and software. Processor 114 may include one or more processors 114 configured to perform the operations described herein. For example, a single processor 114 may be configured to perform all the operations described herein. Alternatively, multiple processors 114 may be collectively configured to perform all the operations described herein, and each of the multiple processors 114 may be configured to perform a subset of the operations described herein. For example, a first processor 114 may perform a first subset of the operations described herein, a second processor 114 may be configured to perform a second subset of the operations described herein, and so on.

[0046] Processor 114 can be configured to control ultrasound probe 102 to acquire ultrasound data. Processor 114 can be configured to control which elements of the ultrasound probe 102 are active and to control the shape of the beam emitted from the ultrasound probe 102. Processor 114 can generate ultrasound images for display. For example, processor 114 can generate B-mode images, color Doppler images, M-mode images, color M-mode images, etc. Ultrasound images can be 3D images, 2D images, single-plane images, dual-plane images, tri-plane images, multi-plane images, etc. Ultrasound images can correspond to various anatomical planes (e.g., sagittal, coronal, and transverse) of the region of interest.

[0047] Processor 114 can perform scan conversion. For example, processor 114 can perform scan conversion by converting raw ultrasound data in scan format to display format. To reiterate, processor 114 can convert raw ultrasound data from polar coordinate format to ultrasound data for display in Cartesian coordinate format.

[0048] Display 116 can be configured to display information. For example, display 116 can be a monitor, LED display, cathode ray tube, projection display, touch screen, tablet computer, mobile phone, etc. Display 116 can display ultrasound images based on ultrasound data in real time. For example, display 116 can display ultrasound images within one second, two seconds, five seconds, etc., after ultrasound probe 102 acquires ultrasound data.

[0049] Memory 118 may be configured to store information and / or instructions for use by processor 114. Memory 118 may be a non-transitory computer-readable medium. For example, memory 118 may be random access memory (RAM), read-only memory (ROM), and / or another type of dynamic or static storage device (e.g., flash memory, magnetic memory, and / or optical memory) for storing information and / or instructions for use by processor 114. Memory 118 may be configured to store instructions that, when executed by processor 114, cause processor 114 to perform the operations described herein.

[0050] The communication interface 120 can be configured to enable the processor 114 to communicate with other systems, such as via a wired connection, a wireless connection, or a combination of wired and wireless connections. For example, the communication interface 120 may include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi interface, a cellular network interface, etc.

[0051] Figure 1 The number and arrangement of components in the ultrasound system 100 shown are provided as an example. In practice, the ultrasound system 100 may include additional components, fewer components, different components, or components related to... Figure 1 The components shown are arranged differently. Additionally or alternatively, the set of components of the ultrasound system 100 (e.g., one or more components) may perform one or more functions described as being performed by another set of components of the ultrasound system 100.

[0052] Figure 2 This is a diagram of an exemplary ultrasound probe 102 used in an ultrasound system 100 for generating and displaying ultrasound images with a non-linear axial scale. (See diagram for example.) Figure 2 As shown, the ultrasonic probe 102 may include a lens 202, an acoustic matching layer 204, a transducer element 206, an acoustic dematching layer 208, and a backing layer 210.

[0053] According to one embodiment, lens 202 can be configured to guide an ultrasonic signal toward the region of interest (ROI) of a subject. For example, lens 202 can be made of silicone, epoxy, or rubber. According to one embodiment, acoustic matching layer 204 can be configured to facilitate matching of impedance differences that may exist between the higher impedance transducer element 206 and the lower impedance subject. For example, acoustic matching layer 204 can be made of graphite, plastic, or resin. According to one embodiment, transducer element 206 can be configured to receive element-specific transmitted signals, deform based on element-specific transmitted signals, generate ultrasonic signals based on deformation, and transmit ultrasonic signals toward the RIO. Additionally or alternatively, transducer element 206 can be configured to receive echo signals reflected or backscattered from the RIO, deform based on the echo signals, generate electrical signals based on deformation, and transmit electrical signals. For example, transducer element 206 can be a piezoelectric material, such as Pb(Mg) 1 / 3Nb 2 / 3 )O3-PbTiO3 ("PMN-PT"), Pb(In 1 / 2 Nb 1 / 2 O3-Pb(Mg) 1 / 3 Nb 2 / 3 The acoustic dematching layer 208 can be configured to reduce insertion loss and extend the frequency bandwidth of the transducer element 206, such as tungsten carbide or silicon carbide. According to one embodiment, the backing layer 210 can be configured to attenuate ultrasonic signals directed from the transducer element 206 in the opposite direction to the subject, and to attenuate ultrasonic signals deflected by the housing of the ultrasonic probe 102. For example, the backing layer 210 can be epoxy resin or metal.

[0054] Figure 2The number and arrangement of components of the ultrasonic probe 102 shown are provided as an example. In practice, the ultrasonic probe 102 may include additional components, fewer components, different components, or components similar to those in the original model. Figure 2 The components shown are arranged differently. Additionally or alternatively, the assembly of components of the ultrasonic probe 102 (e.g., one or more components) may perform one or more functions described as being performed by another assembly of components of the ultrasonic probe 102.

[0055] Figure 3 This is a flowchart of an exemplary process 300 for generating and displaying ultrasound images with a non-linear axial scale. (Example:) Figure 3 As shown, process 300 may include acquiring ultrasound data of the region of interest (ROI) of the subject (operation 310). For example, ultrasound imaging system 100 may acquire ultrasound data of the ROI of the subject. Ultrasound system 100 may control ultrasound probe 102 to send ultrasound signals toward the ROI of the subject and control ultrasound probe 102 to receive echo signals reflected from or backscattered from the ROI of the subject. The subject may be a person, animal, phantom, object, etc. The ROI may be any region of the subject. For example, the ROI may be the lungs, heart, liver, etc. In the context of lung ultrasound, the ROI may depict the pleura.

[0056] like Figure 3 As further shown, process 300 may include generating an ultrasound image with a non-linear axial scale based on ultrasound data acquired from the region of interest of the subject (operation 320). For example, ultrasound system 100 may generate an ultrasound image with a non-linear axial scale based on ultrasound data acquired from the region of interest of the subject.

[0057] As used herein, a “scale” can refer to an indication of the relationship between a distance on an ultrasound image and its corresponding actual distance within the patient’s body. In this context, and as used herein, a “linear axial scale” can refer to a scale such that a specific axial distance on an ultrasound image measured at any point on the ultrasound image corresponds to the same actual axial distance within the patient’s body. For example, suppose a user interface element (e.g., a line) has a fixed length (e.g., 2 cm). If the user interface element is positioned at any point on the ultrasound image, the actual axial distance within the patient’s body (e.g., 3 cm) is the same. For example, if the line is positioned at a first location, the actual axial distance within the patient’s body is 3 cm. If the line is positioned at a second location (or any other location) on the ultrasound image, the actual axial distance within the patient’s body is 3 cm.

[0058] Furthermore, as used herein, a "non-linear axial scale" can refer to a scale that allows a specific axial distance on an ultrasound image measured at various points to correspond to different actual axial distances within the subject's body. For example, suppose a user interface element (e.g., a line) has a fixed length (e.g., 2 cm). The subject's actual axial distance depends on where the line is positioned. For example, if the line is positioned at a first location, the actual axial distance within the subject's body is 3 cm. If the line is positioned at a second location on the ultrasound image, the actual axial distance within the subject's body could be 5 cm.

[0059] As used herein, "scale" can also refer to a series of visual indicators (e.g., marks, dots, lines, points, etc.) at specific intervals used to measure distances within a subject's body. In this case, and as used herein, "linear axial scale" can refer to a series of visual indicators provided at the same constant intervals. To reiterate, the corresponding distances between any two visual indicators are the same. Furthermore, and as used herein, "non-linear axial scale" can refer to a series of visual indicators provided at varying intervals. To reiterate, the corresponding distances between any two visual indicators can be the same or different.

[0060] According to one embodiment, the ultrasound system 100 can generate an ultrasound image with a non-linear axial scale by transforming the original ultrasound data such that the resolution of a first region of the ultrasound image is greater than the resolution of a second region of the ultrasound image. Resolution may refer to the number of pixels in the ultrasound image or the number of pixels on the display 116 of the ultrasound system 100. The original ultrasound data may refer to ultrasound data acquired via the ultrasound probe 102 and not yet processed by a scan conversion process. According to the non-linear axial scale, the resolution of the first region may be greater than the resolution of the second region. For example, the first region may have a first number of pixels, greater than the second number of pixels in the second region. The first region may correspond to a first depth range (e.g., 0cm-2cm, 0cm-3cm, 1cm-2cm, etc.), and the second region may correspond to a second depth range (e.g., 2cm-4cm, 4cm-6cm, 7cm-8cm, 10cm-14cm, etc.).

[0061] According to one embodiment, the ultrasound system 100 can transform raw ultrasound data such that a first region of the ultrasound image has a first linear axial scale, and a second region of the ultrasound image has a second linear axial scale different from the first linear axial scale. In this way, the ultrasound image can have a non-linear axial scale because the ultrasound image has at least two different linear axial scales. The ultrasound image can have any number of different linear axial scales. For example, the ultrasound image can have n different regions, each with n different linear axial scales. According to another embodiment, the ultrasound system 100 can transform the raw ultrasound data by applying μ-law scaling. The ultrasound system 100 can use different compression values ​​(e.g., k values). The ultrasound system 100 can apply μ-law scaling over the entire ultrasound image. Alternatively, the ultrasound system 100 can apply μ-law scaling at a specific depth of the ultrasound image. It should be understood that the ultrasound system 100 can use any other type of technique to apply a non-linear axial scale to the ultrasound image.

[0062] In the context of lung ultrasound, and by transforming the original ultrasound data before performing scan conversion, the ultrasound system 100 can maintain the appearance of the B-line compared to applying a nonlinear axial scale to the ultrasound data that has already been scanned and converted. For example, the B-line is an acoustic artifact that occurs along the direction of the ultrasound transmission beam in the axial direction. By transforming the original ultrasound data, the ultrasound system 100 can maintain a straight B-line. Conversely, if the ultrasound system 100 transforms the scanned and converted ultrasound data, the nonlinearity in the Cartesian direction will result in the appearance of a curved B-line.

[0063] According to one implementation, the ultrasound image can be a single continuous ultrasound image exhibiting a non-linear axial scale. That is, the ultrasound image may not include magnified portions, reduced portions, discontinuous portions, etc.

[0064] like Figure 3 As further shown, process 300 may include displaying an ultrasound image with a non-linear axial scale based on the generated ultrasound image (operation 330). For example, ultrasound system 100 may display an ultrasound image with a non-linear axial scale based on the generated ultrasound image. In this way, the ultrasound image can display a first region with higher resolution and a second region with lower resolution, which allows clinicians to assess the first region in more detail while still being able to visualize the second region. For example, in the context of lung ultrasound, clinicians can visualize the pleura in more detail while still being able to visualize B-lines.

[0065] Figure 4A Figure 400 shows an exemplary ultrasound image 402 with a linear axial scale. (See Figure 400.) Figure 4AAs shown, ultrasound image 402 can display the pleural line 404 and the B-line 406. Ultrasound image 402 has a linear axial scale, meaning that a specific axial distance measured at any point on ultrasound image 402 corresponds to the same actual axial distance within the subject's body. Furthermore, ultrasound image 402 has a linear axial scale, meaning that a series of visual indicators 408 are provided at identical, constant intervals. To reiterate, the distance between the visual indicators "2cm" and "4cm" is the same as the distance between the visual indicators "4cm" and "6cm".

[0066] Figure 4B Figure 400 shows an exemplary ultrasound image 410 with a non-linear axial scale. Figure 4B As shown, ultrasound image 410 can display the pleural line 412 and the B-line 414. Ultrasound image 410 has a non-linear axial scale, meaning that a specific axial distance measured at any point on ultrasound image 410 can correspond to the same actual axial distance within the subject's body or different axial distances within the subject's body. Furthermore, ultrasound image 410 has a non-linear axial scale, meaning a series of visual indicators 416 are provided at varying intervals. To reiterate, the distance between visual indicators "2cm" and "4cm" is the same as the distance between visual indicators "4cm" and "6cm," however, the distance between visual indicators "8cm" and "10cm" is different from, for example, the distance between visual indicators "2cm" and "4cm."

[0067] In this way, the resolution of the region between 0 cm and 4 cm in ultrasound image 410 can be greater than that of the region between 0 cm and 4 cm in ultrasound image 402. Therefore, clinicians can more easily assess the pleural line 412. Furthermore, the resolution of the region between 6 cm and 10 cm in ultrasound image 410 can be lower than that of the region between 6 cm and 10 cm in ultrasound image 402. However, clinicians can still visualize the B-line 414 in ultrasound image 410.

[0068] Figure 5 Figure 500 shows an exemplary three-dimensional (3D) ultrasound image 502 with a non-linear axial scale. Figure 5As shown, the 3D ultrasound image 502 can display the pleural line 504 and the B-line 506. The 3D ultrasound image 502 has a non-linear axial scale, meaning that a specific axial distance measured at any point on the 3D ultrasound image 502 can correspond to the same actual axial distance within the subject's body or different axial distances within the subject's body. Furthermore, the 3D ultrasound image 502 has a non-linear axial scale, meaning that a series of visual indicators 508 are provided at varying intervals. To reiterate, the distance between the visual indicators "2cm" and "4cm" is the same as the distance between the visual indicators "4cm" and "6cm," however, the distance between the visual indicators "8cm" and "10cm" is different from, for example, the distance between the visual indicators "2cm" and "4cm." Figure 5 As further shown, the ultrasound system 100 can also display a 2D ultrasound image 510 corresponding to a specific layer of the 3D ultrasound image 502. The 2D ultrasound image 510 can display the pleural line 512 and the B-line 514. The 2D ultrasound image 510 has a non-linear axial scale, meaning that a specific axial distance measured at any point on the 2D ultrasound image 510 can correspond to the same actual axial distance within the subject's body or different axial distances within the subject's body. Furthermore, the 2D ultrasound image 510 has a non-linear axial scale, meaning that a series of visual indicators 516 are provided at varying intervals. To reiterate, the distance between the visual indicators "2cm" and "4cm" is the same as the distance between the visual indicators "4cm" and "6cm," however, the distance between the visual indicators "8cm" and "10cm" is different from, for example, the distance between the visual indicators "2cm" and "4cm."

[0069] Figure 6 This is a flowchart of an exemplary process 600 for determining the depth of a scale used to convert ultrasound images. (Example:) Figure 6 As shown, process 600 may include determining the depth of a scale used to convert ultrasound images (operation 610). For example, ultrasound system 100 may determine the depth of a scale used to convert ultrasound images.

[0070] According to one embodiment, the ultrasound system 100 can use image processing techniques to determine the depth of a scale used for converting ultrasound images. For example, the image processing techniques may be segmentation techniques, pattern matching techniques, feature extraction techniques, image analysis techniques, edge detection techniques, image registration techniques, etc. In this case, the ultrasound system 100 can use image processing techniques to analyze the ultrasound image and determine the depth for converting the scale based on the analysis. For example, the ultrasound system 100 can analyze the ultrasound image to identify one or more structures that can be used to delineate the location for converting the scale. In the context of lung ultrasound, the one or more structures may be the pleura.

[0071] According to another embodiment, the ultrasound system 100 can use an AI model to determine the depth of the scale used for converting ultrasound images. For example, the ultrasound system 100 can input an ultrasound image into the AI ​​model and determine the depth of the scale used for converting the ultrasound image based on the output of the AI ​​model. The AI ​​model can be a convolutional neural network (CNN) model, a residual neural network, a random forest model, a decision tree model, an artificial neural network (ANN), a naive Bayes model, a decision tree, a recurrent neural network (RNN), a logistic regression model, a support vector machine, etc. In this case, the AI ​​model can be trained to receive ultrasound images, analyze the ultrasound images to determine the depth of the scale used for converting the ultrasound image, and output information identifying the depth of the scale used for converting the ultrasound image. Training data may include known ultrasound images associated with known depths at which the scale of the ultrasound image will be converted.

[0072] According to one implementation, the ultrasound system 100 can use predetermined information corresponding to the subject to determine the depth of a scale used for converting ultrasound images. For example, the predetermined information may identify the depth and may be associated with the subject's gender, physical characteristics (e.g., height, weight, body mass index, etc.), age, etc. In the context of lung ultrasound, the depth may correspond to the depth of the pleura, and may be the average depth of the pleura in a population, etc.

[0073] According to one implementation, the ultrasound system 100 can determine the depth of a scale used to transform the ultrasound image based on a specific region of interest being scanned. For example, the ultrasound system 100 can determine the specific region of interest based on analyzing the ultrasound image, tracking data based on the ultrasound probe 102, or the subject's position data. In the context of lung ultrasound, the depth can be shallower for the upper region of the lung and deeper for the lower region of the lung.

[0074] According to one embodiment, the ultrasound system 100 can determine the depth of a scale for converting ultrasound images using preoperative imaging data of a region of interest acquired by a preoperative imaging system (e.g., a computed tomography (CT) system, a magnetic resonance imaging (MRI) system, an ultrasound system, an X-ray system, a positron emission tomography (PET) device, etc.). For example, the ultrasound system 100 can register ultrasound images with preoperative images and determine the depth of a scale for converting ultrasound images based on this registration. In this case, the preoperative imaging data can identify the depth of a scale for converting ultrasound images.

[0075] According to one embodiment, ultrasound system 100 can determine the depth of a scale used for converting ultrasound images based on user input received via user input device 112. For example, a user of ultrasound system 100 can interact with user input device 112 to select the depth of a scale used for converting ultrasound images. Ultrasound system 100 can determine the depth of a scale used for converting ultrasound images based on this selection.

[0076] like Figure 6 As further shown, process 600 may include generating an ultrasound image with a non-linear axial scale based on a depth determined by a scale used for converting the ultrasound image (operation 620). For example, ultrasound system 100 may be combined with the above. Figure 3 The method described in operation 320 is similar to generating an ultrasound image with a non-linear axial scale based on ultrasound data acquired from the subject's region of interest. The ultrasound system 100 can transform the original ultrasound data using a determined depth, such that the ultrasound image includes a non-linear axial scale. For example, the ultrasound system 100 can transform the original ultrasound data at a depth shallower than the determined depth, resulting in higher resolution compared to transforming the original ultrasound data at a depth deeper than the determined depth.

[0077] Figures 7A to 7C Figure 700 shows a scaling technique used to generate ultrasound images with a non-linear axial scale. (As shown...) Figures 7A to 7C As shown, the ultrasound system 100 can use μ-law scaling techniques to apply various compression values ​​(e.g., k-values) to generate ultrasound images with a non-linear axial scale. Figure 7A As shown, the ultrasound system 100 can use a k value of 0.0001, which generates an ultrasound image with a substantially linear axial scale. For example, as shown, a first depth range of 0 mm to 20 mm within the patient's body may include 100 pixels in the ultrasound image, a second depth range of 20 mm to 40 mm may include 100 pixels in the ultrasound image, a third depth range of 40 mm to 60 mm within the patient's body may include 100 pixels in the ultrasound image, and so on. Figure 7B As shown, the ultrasound system 100 can use a k-value of 0.0200, which generates an ultrasound image with a non-linear axial scale. For example, as shown, a first depth range of 0mm-20mm within the patient's body may include 200 pixels in the ultrasound image, a second depth range of 20mm-40mm may include 120 pixels in the ultrasound image, and a third depth range of 40mm-60mm within the patient's body may include 110 pixels in the ultrasound image, and so on. Figure 7CAs shown, the ultrasound system 100 can use a k value of 0.1000, which generates an ultrasound image with a non-linear axial scale. For example, as shown, a first depth range of 0mm-20mm within the patient's body may include 310 pixels in the ultrasound image, a second depth range of 20mm-40mm may include 140 pixels in the ultrasound image, and a third depth range of 40mm-60mm within the patient's body may include 100 pixels in the ultrasound image, and so on.

[0078] Figure 8 This is a flowchart of an exemplary process 800 for displaying ultrasound images with linear or non-linear axial scales based on user selection. Figure 8 As shown, process 800 may include displaying a prompt allowing the user to select a linear or non-linear axis scale (operation 810). For example, ultrasound system 100 may display a prompt allowing the user to select either a linear or non-linear axis scale for displaying ultrasound images. Ultrasound system 100 may display prompts based on the acquisition of ultrasound data of the subject's region of interest, based on the display of ultrasound images, etc. Figure 8 As further shown, process 800 may include determining whether the user has selected a nonlinear axial scale (operation 820). For example, ultrasound system 100 may determine the user selection and whether that selection corresponds to a linear or nonlinear axial scale. If ultrasound system 100 determines that the user has selected a linear axial scale (operation 820 - No), then as... Figure 8 As further shown, process 800 may include displaying an ultrasound image with a linear axial scale (operation 830). For example, ultrasound system 100 may display an ultrasound image with a linear axial scale. Alternatively, if ultrasound system 100 determines that the user has selected a non-linear axial scale (operation 820 - yes), then as Figure 8 As further shown, process 800 may include displaying an ultrasound image with a non-linear axial scale (operation 840). For example, ultrasound system 100 may display an ultrasound image with a non-linear axial scale. In some cases, ultrasound system 100 may switch between displaying an ultrasound image with a linear axial scale and displaying an ultrasound image with a non-linear axial scale, or vice versa. That is, ultrasound system 100 may dynamically switch between displaying a specific ultrasound image with a linear axial scale or a non-linear axial scale based on user selection.

[0079] Figure 9 This is a flowchart of an exemplary process 900 for displaying distances related to ultrasound images based on user selection. (Example:) Figure 9As shown, process 900 may include an ultrasound image with a non-linear axial scale displaying the region of interest of the subject (operation 910). For example, ultrasound system 100 may display an ultrasound image with a non-linear axial scale of the region of interest of the subject. Figure 9 As further shown, process 900 may include receiving a selection of a distance to be measured in an ultrasound image having a non-linear axial scale (operation 920). For example, ultrasound system 100 may receive a selection of a distance to be measured in an ultrasound image. Ultrasound system 100 may receive the selection based on user interaction with user interface elements of ultrasound system 100. For example, user interface elements may be lines, circles, etc., used to measure distances in ultrasound images. Users may interact with user interface elements to locate and / or orient the user interface elements relative to specific anatomical features or content of the ultrasound image. Figure 9 As further shown, process 900 may include measuring distances within the subject's body (operation 930). For example, ultrasound system 100 may measure distances within the subject's body based on distances to be measured in ultrasound images. Ultrasound system 100 may measure distances within the subject's body based on a scale that indicates the relationship between distances on ultrasound images and corresponding actual distances within the subject's body. Figure 9 As further shown, process 900 may include displaying distances related to an ultrasound image (operation 940). For example, ultrasound system 100 may display distances related to an ultrasound image. Ultrasound system 100 may display distances related to an ultrasound image by displaying a numerical value of the distance superimposed on an ultrasound image adjacent to a user interface element. Alternatively, ultrasound system 100 may display distances related to an ultrasound image by displaying a numerical value of the distance at a predetermined location on display 116 of ultrasound system 100.

[0080] Figure 10 Figure 1000 illustrates an exemplary process for training, deploying, and monitoring an AI model 1020. According to one embodiment, the ultrasound system 100 may use one or more AI models 1020. One or more AI models 1020 may be associated with a training phase 1002, a deployment phase 1008, and a monitoring phase 1014. In the training phase 1002, the ultrasound system 100 (or an external system) may receive and process training data 1004 to generate a trained model 1020. The training data 1004 may be generated, received, or otherwise obtained from internal and / or external resources.

[0081] Generally, the trained model 1020 may include a set of variables (e.g., nodes, neurons, filters, etc.) tuned (e.g., weighted, biased, etc.) to different values ​​through the application of training data 1004. According to one embodiment, the training process may employ supervised, unsupervised, semi-supervised, and / or reinforcement learning processes to train model 1020. According to one embodiment, a portion of the training data 1004 may be retained during training and / or used to validate the trained model 1020.

[0082] For a supervised learning process, training data 1004 may include labels or scores that can facilitate the training process by providing a baseline ground truth. For example, the labels or scores may indicate the output of AI model 1020. Training can continue by feeding training data 1004 into AI model 1020. AI model 1020 may have variables set to initial values ​​(e.g., randomly, based on Gaussian noise, based on pre-trained values, etc.). AI model 1020 may generate an output based on the training data 1004 being fed into AI model 1020. The output may be compared with a corresponding label or score (e.g., a baseline ground truth) indicating a known output, and then backpropagated through AI model 1020 to adjust the values ​​of the variables. This process may be repeated for multiple samples, at least until a determined loss or error falls below a predefined threshold. According to one embodiment, some of the training data in training data 1004 may be retained and used for further validation or testing of the trained AI model 1020.

[0083] For unsupervised learning processes, training data 1004 may not include pre-assigned labels or scores to aid the training process. Instead, unsupervised learning processes may include clustering, classification, etc., to identify naturally occurring patterns in training data 1004. For example, training data 1004 may be clustered into groups based on the similarity and / or patterns of the labels. K-means clustering or K-nearest neighbors, which can be supervised or unsupervised, may also be used. A combination of K-nearest neighbors and unsupervised clustering techniques may also be used. For semi-supervised learning, a combination of training data 1004 with pre-assigned labels or scores and training data without pre-assigned labels or scores can be used to train AI model 1020.

[0084] When reinforcement learning is employed, an agent (e.g., an algorithm) can be trained to make decisions from training data 1004 through trial and error. For example, based on the decisions made, the agent can then receive feedback (e.g., a positive reward if the prediction is higher than a predetermined threshold), adjust its next decision to maximize the reward, and repeat until the loss function is optimized.

[0085] After being trained, the trained model 1020 can be stored and subsequently applied by the ultrasound system 100 during deployment phase 1008. For example, during deployment phase 1008, the trained AI model 1020, executed by the ultrasound system 100, can receive input data. During deployment phase 1008, the trained AI model 1020 can perform actions such as combining... Figure 3 , Figure 6 , Figure 8 and / or Figure 9 The one or more operations described above.

[0086] After deployment, the trained AI model 1020 can be monitored during monitoring phase 1014. For example, during monitoring phase 1014, the AI ​​model 1020 can generate monitoring data for monitoring the trained AI model 1020. Monitoring data may include data identifying outputs, as determined by the operator. During monitoring phase 1014, the monitoring data, along with predicted output and input data, can be analyzed to determine the accuracy of the trained AI model 1020. According to one implementation, based on the analysis, the process can return to training phase 1002, where the values ​​of one or more variables of the AI ​​model 1020 can be adjusted to improve the accuracy of the AI ​​model 1020.

[0087] Although the embodiments described herein are in conjunction with ultrasound systems and ultrasound images, it should be understood that the embodiments described herein are applicable to other types of medical imaging modalities and other types of medical images (e.g., CT images, MRI images, etc.). Furthermore, although some embodiments described herein are in conjunction with lung ultrasound, it should be understood that the embodiments described herein are applicable to any other region of interest of the subject (e.g., heart, liver, etc.). Additionally, although some embodiments described herein use nonlinear axial scales, it should be understood that the embodiments described herein are applicable to nonlinear horizontal scales.

[0088] In this way, some embodiments of this paper provide improvements to the field of ultrasound imaging technology by offering the generation of novel ultrasound images—single, continuous images with a non-linear axial scale. Furthermore, some embodiments of this paper provide improvements to ultrasound systems by enabling the ultrasound system to generate single, continuous images with a non-linear axial scale. The non-linear axial scale allows for simultaneous observation of the near and far fields in a manner that provides enhanced detail and resolution in the near field. Therefore, clinicians can observe near-field details while maintaining full imaging depth in a single, continuous ultrasound image.

[0089] The embodiments shown in the accompanying drawings and described above are merely illustrative embodiments and are not intended to limit the scope of the appended claims, including any equivalents included within the scope of the claims. Various modifications are possible and will be apparent to those skilled in the art. Any combination of non-mutually exclusive features described herein is intended to be within the scope of the invention. That is, features of the described embodiments may be combined with any suitable aspect described above, and optional features of any aspect may be combined with any other suitable aspect. Similarly, features listed in dependent claims may be combined with non-mutually exclusive features of other dependent claims, particularly where the dependent claims are subordinate to the same independent claim. In some jurisdictions that claim single-claim dependents, such single-claim dependents may have been used in practice, but this should not be construed as meaning that features in dependent claims are mutually exclusive.

Claims

1. An ultrasonic method, the ultrasonic method comprising: Collect ultrasound data of the region of interest from the subject; An ultrasound image of the region of interest of the subject is generated based on the ultrasound data acquired from the subject, wherein the ultrasound image has a non-linear axial scale; and The ultrasound image is displayed based on the generated ultrasound image, which has the nonlinear axial scale.

2. The ultrasound method according to claim 1, wherein the first region of the ultrasound image includes a linear axial scale, and wherein the second region of the ultrasound image includes the nonlinear axial scale.

3. The ultrasound method according to claim 1, further comprising: The original ultrasound data is scaled and transformed by applying the μ law before performing the scan transformation.

4. The ultrasound method according to claim 1, further comprising: Determine the depth of the scale used to convert the ultrasound images.

5. The ultrasound method according to claim 1, wherein the first region of the ultrasound image includes a first linear axis scale, and wherein the second region of the ultrasound image includes a second linear axis scale different from the first linear axis scale.

6. The ultrasound method of claim 1, wherein the ultrasound image includes a first region corresponding to a first depth range within the body of the subject, wherein the ultrasound image includes a second region corresponding to a second depth range within the body of the subject, wherein the first depth range is the same as the second depth range, and wherein the first region includes a first number of pixels, the first number of pixels being greater than a second number of pixels in the second region.

7. The ultrasound method according to claim 1, wherein the ultrasound image is a single continuous ultrasound image.

8. The ultrasonic method according to claim 1, further comprising: Image processing techniques or artificial intelligence models are used to determine the depth of the pleura in the region of interest of the subject. The one or more processors are configured to generate an ultrasound image with the nonlinear axial scale based on determining the depth of the pleura, and The first region of the ultrasound image above the depth of the pleura includes a first axial scale, and the second region of the ultrasound image below the depth of the pleura includes a second axial scale different from the first axial scale.

9. The ultrasonic method according to claim 1, further comprising: The system displays a prompt allowing the user to select either a linear axial scale or the non-linear axial scale. as well as Receive user selection of the nonlinear axial scale. The one or more processors are configured to display the ultrasound image with the nonlinear axial scale based on the user's selection.

10. The ultrasound method according to claim 1, wherein the ultrasound image is a three-dimensional (3D) ultrasound image having the nonlinear axial scale.

11. An ultrasound system, the ultrasound system comprising: Lens, acoustic matching layer, acoustic dematching layer, and multiple transducer elements; A memory configured to store instructions; and One or more processors configured to execute the instructions to implement the method according to any one of claims 1 to 10.

12. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform the method according to any one of claims 1 to 10.

Images