Guided ultrasound imaging for point-of-care staging of medical conditions

Abstract

An ultrasound system (100) includes a processor (116, 134) configured to communicate with a display (132) and a transducer array (112) of an ultrasound probe (110). The processor controls the transducer array to obtain a first ultrasound image corresponding to a first view of the patient anatomy and a second ultrasound image of the patient anatomy corresponding to a second view of the patient anatomy. The processor identifies a first image feature within the first ultrasound image and a second image feature within the second ultrasound image. The processor then determines a first sub-score of the first image feature and a second sub-score of the second image feature, and determines a staging value representative of progression of the medical condition based on the first sub-score and the second sub-score. The processor then outputs a screen display including the staging value, the ultrasound image, the indication of the image feature, and the sub-score.

Description

Technical Field

[0001] The present disclosure relates generally to ultrasound imaging. In particular, an ultrasound system provides guidance for acquiring ultrasound images of multiple views, automatically extracting features from the ultrasound images, assigning sub-scores to the features, and determining a staging value for a patient's medical condition. Background Art

[0002] Physicians use many different medical diagnostic systems and tools to monitor the health of a subject and to diagnose and treat medical conditions. Ultrasound imaging systems are widely used for medical imaging and measurement. An ultrasound transducer probe may include an array of ultrasound transducer elements that transmit sound waves into a subject and record sound waves reflected and backscattered from internal anatomical structures within the subject, which may include tissues, blood vessels, and internal organs. The transmission and reception of sound waves and various beamforming and processing techniques create images of the subject's internal anatomical structures.

[0003] Ultrasound imaging is safe, useful, and in some applications, a non-invasive tool for diagnostic examinations, interventions, and / or treatments. Ultrasound imaging can be used to diagnose non-alcoholic fatty liver disease (NAFLD), including NAFLD progression stages. Accurate diagnosis of NAFLD stages typically requires advanced quantitative ultrasound (QUS) techniques that are not available in point-of-care (POC) ultrasound systems due to computational or hardware limitations. In addition, acquiring NAFLD images of sufficient quality and analyzing such images typically requires advanced expertise from the user and optimal conditions for the procedure, including sufficient time and resources. Lack of expertise, time, or resources can easily result in poor quality ultrasound and, therefore, inaccurate measurements and diagnoses. Summary of the invention

[0004] Aspects of the present disclosure are systems, devices, and methods for bedside staging of medical conditions by guided ultrasound imaging. Aspects of the present disclosure advantageously guide non-expert users to obtain ultrasound images of patient anatomical structures that include multiple views of sufficient quality for analysis. In addition, staging values ​​associated with the progression of the medical condition are automatically determined based on the acquired ultrasound images. Aspects of the present disclosure also advantageously display image features within the acquired ultrasound images and corresponding sub-scores associated with each image feature to the user. By displaying the acquired and / or selected images and corresponding sub-scores to the user, the user can be more confident that the resulting staging values ​​are accurate. Aspects of the present disclosure also advantageously reduce the time and effort of a physician or user of the system in reporting by automatically generating reports.

[0005] The invention is defined by the independent claim. The dependent claims define advantageous embodiments.

[0006] In some aspects, an ultrasound imaging system may provide a non-expert user with bedside setup instructions on how to obtain ultrasound images of a patient's anatomy from multiple views. The system may analyze the images to ensure that they are of sufficient quality and select at least one image for each view. The ultrasound image may then extract image features corresponding to each view and assign feature sub-scores to each selected image. The ultrasound system may then display the selected images and associated sub-scores, and determine a staging value for the medical condition based on the images and sub-scores.

[0007] According to an exemplary aspect, an ultrasound system is provided. The system includes a processor configured to communicate with a display and a transducer array of an ultrasound probe, wherein the processor is configured to: control the transducer array to obtain a first ultrasound image corresponding to a first view of a patient's anatomy and a second ultrasound image of a patient's anatomy corresponding to a second view of the patient's anatomy; identify a first image feature associated with a medical condition of the patient's anatomy within the first ultrasound image and a second image feature associated with the medical condition within the second ultrasound image; determine a first sub-score of the first image feature and a second sub-score of the second image feature; determine a staging value representing the progression of the medical condition based on the first sub-score and the second sub-score; and output a screen display to the display, the screen display including: a staging value; and at least one of: a first ultrasound image, an indication of the first image feature in the first ultrasound image and the first sub-score; or a second ultrasound image, an indication of the second image feature in the second ultrasound image and the second sub-score.

[0008] In some aspects, the processor is configured to output to a display user guidance for obtaining a first ultrasound image corresponding to a first view of the patient's anatomical structure. In some aspects, the user guidance includes a graphical representation of the probe and / or an orientation of the ultrasound probe. In some aspects, the user guidance includes a reference image associated with the first view of the patient's anatomical structure. In some aspects, the user guidance includes a description of dynamic behavior associated with the first view of the patient's anatomical structure. In some aspects, the processor is configured to determine the quality associated with the first ultrasound image before identifying a first image feature within the first ultrasound image. In some aspects, the processor is configured to determine the quality based on a comparison between the first ultrasound image and a reference image, the reference image being associated with the first view of the patient's anatomical structure. In some aspects, the processor is configured to: if the quality meets a threshold, identify the first image feature within the first ultrasound image; if the quality does not meet the threshold, control the transducer array to obtain another ultrasound image corresponding to the first view of the patient's anatomical structure. In some aspects, in order to determine the first sub-score and the second sub-score, the processor is configured to implement a first machine learning algorithm. In some aspects, the first machine learning algorithm includes a multi-task learning model. In some aspects, to identify the first image feature and the second image feature, the processor is configured to implement a second machine learning algorithm different from the first machine learning algorithm. In some aspects, the patient anatomical structure includes a liver, and the medical condition includes hepatic steatosis. In some aspects, the first sub-score of the first image feature and the second sub-score of the second image feature correspond to an ultrasound fatty liver index (US-FLI). In some aspects, the first image feature and the second feature each include a different one of the following: liver-kidney contrast, posterior attenuation, vascular blur, gallbladder visualization, diaphragm attenuation visualization, or focal residual.

[0009] According to an exemplary aspect, a computer-implemented method is provided. The method includes controlling a transducer array of an ultrasound imaging probe to obtain a first ultrasound image corresponding to a first view of a patient's anatomy and a second ultrasound image of a patient's anatomy corresponding to a second view of the patient's anatomy; identifying a first image feature associated with a medical condition of the patient's anatomy within the first ultrasound image and a second image feature associated with the medical condition within the second ultrasound image; determining a first subscore of the first image feature and a second subscore of the second image feature; determining a staging value representing the progression of the medical condition based on the first subscore and the second subscore; and outputting a screen display to a display, the screen display including: a staging value; and at least one of: a first ultrasound image, an indication of the first image feature in the first ultrasound image and the first subscore; or a second ultrasound image, an indication of the second image feature in the second ultrasound image and the second subscore.

[0010] Claim 1 also covers an ultrasound system for staging a medical condition, the system comprising an ultrasound probe having a transducer array; a display; and a processor configured to communicate with the display and the transducer array, wherein the processor is configured to: output to the display user instructions for obtaining multiple ultrasound images corresponding to multiple views of a patient's anatomy; control the transducer array to obtain multiple ultrasound images corresponding to the multiple views; use a first machine learning algorithm to identify multiple image features associated with the medical condition within the multiple images; use a second machine learning algorithm to determine multiple sub-scores for the multiple image features based on a clinically accepted scale; determine a staging value representing the progression of the medical condition based on the multiple sub-scores; and output to the display a screen display comprising: a staging value; and a visual representation of how the staging value was determined, including: at least one ultrasound image from the multiple ultrasound images; an indication of corresponding image features from the multiple image features in the at least one ultrasound image; and corresponding sub-scores from the multiple sub-scores.

[0011] Additional aspects, features, and advantages of the present disclosure will become apparent from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Illustrative aspects of the present disclosure will be described with reference to the accompanying drawings, in which:

[0013] Figure 1 is a schematic diagram of an ultrasound imaging system according to aspects of the present disclosure.

[0014] Figure 2 is a schematic diagram of a processor circuit according to aspects of the present disclosure.

[0015] Figure 3 is a flow chart of a method of obtaining and analyzing ultrasound images corresponding to multiple views of a patient's anatomy in accordance with aspects of the present disclosure.

[0016] Figure 4 is a schematic diagram of a graphical user interface displaying user guidance for obtaining ultrasound images for diagnosis according to aspects of the present disclosure.

[0017] Figure 5 is a schematic diagram of a table of user guidance for obtaining ultrasound images of different views for diagnosis according to aspects of the present disclosure.

[0018] Figure 6 is a schematic diagram of acquired ultrasound images of multiple views and corresponding reference images for comparison according to aspects of the present disclosure.

[0019] Figure 7 is a schematic diagram of an acquired ultrasound image and identified features according to aspects of the present disclosure.

[0020] Figure 8 is a schematic diagram of a machine learning algorithm according to aspects of the present disclosure.

[0021] Fig. 9A is a diagrammatic view of a graphical user interface displaying a selected ultrasound image, features, and scores for a particular view in accordance with aspects of the present disclosure.

[0022] Fig. 9B is a schematic diagram of a graphical user interface displaying a selected ultrasound image, features, and scores for a particular view in accordance with aspects of the present disclosure.

[0023] Fig. 9C is a diagrammatic view of a graphical user interface displaying a selected ultrasound image, features, and scores for a particular view in accordance with aspects of the present disclosure.

[0024] Fig.9D is a diagrammatic view of a graphical user interface displaying a selected ultrasound image, features, and scores for a particular view in accordance with aspects of the present disclosure.

[0025] Fig.9E is a schematic diagram of a graphical user interface displaying a selected ultrasound image, features, and scores for a particular view in accordance with aspects of the present disclosure.

[0026] Fig.9F is a schematic diagram of a graphical user interface displaying a selected ultrasound image, features, and scores for a particular view in accordance with aspects of the present disclosure.

[0027] Fig.10 is a schematic diagram of a graphical user interface displaying ultrasound images, corresponding scores, and staging values ​​according to aspects of the present disclosure.

[0028] Fig.11 is a schematic diagram of a graphical user interface displaying ultrasound images, corresponding scores, and staging scores according to aspects of the present disclosure.

[0029] Fig.12 is a flow chart of a method of determining a staging value for a medical condition according to aspects of the present disclosure. DETAILED DESCRIPTION

[0030] In order to promote the understanding of the principles of the present disclosure, reference will now be made to the aspects shown in the accompanying drawings, and specific language will be used to describe these aspects. However, it should be understood that it is not intended to limit the scope of the present disclosure. Any changes and further modifications to the described equipment, systems and methods of the present disclosure and any further application of the principles of the present disclosure are fully expected and included in the present disclosure, as those skilled in the art in the field involved in the present disclosure will generally think of. Specifically, it is fully contemplated that the features, components and / or steps described in relation to one aspect can be combined with the features, components and / or steps described in relation to other aspects of the present disclosure. However, for the sake of brevity, multiple iterations of these combinations will not be described separately.

[0031] Figure 1 1 is a schematic diagram of an ultrasound imaging system 100 according to aspects of the present disclosure. The system 100 is used to scan a region or volume of a subject's body. The subject may include a patient undergoing an ultrasound imaging procedure, or any other person, or any suitable living or non-living organism or structure. The system 100 includes an ultrasound imaging probe 110 that communicates with a host 130 via a communication interface or link 120. The probe 110 may include a transducer array 112, a beamformer 114, a processor circuit 116, and a communication interface 118. The host 130 may include a display 132, a processor circuit 134, a communication interface 136, and a memory 138 for storing subject information.

[0032] In some aspects, the probe 110 is an external ultrasound imaging device that includes a housing 111 configured for handheld operation by a user. The transducer array 112 can be configured to obtain ultrasound data when the user grasps the housing 111 of the probe 110 so that the transducer array 112 is positioned adjacent to or in contact with the skin of the subject. The probe 110 is configured to obtain ultrasound data of anatomical structures within the subject's body when the probe 110 is located outside the subject's body for general imaging, such as for abdominal imaging, liver imaging, etc. In some aspects, the probe 110 can be an external ultrasound probe, a transthoracic probe, and / or a curved array probe.

[0033] In other aspects, the probe 110 can be an internal ultrasound imaging device and can include a housing 111 configured to be positioned within a lumen of a subject's body for general imaging, such as for abdominal imaging, liver imaging, etc. In some aspects, the probe 110 can be a curved array probe. The probe 110 can be any suitable form for any suitable ultrasound imaging application, including external and internal ultrasound imaging.

[0034] In some aspects, aspects of the present disclosure can be implemented with medical images of an object obtained using any suitable medical imaging device and / or modality. Examples of medical images and medical imaging devices include x-ray images (angiographic images, fluoroscopic images, images with or without contrast) obtained by an x-ray imaging device, computed tomography (CT) images obtained by a CT imaging device, positron emission tomography-computed tomography (PET-CT) images obtained by a PET-CT imaging device, magnetic resonance images (MRI) obtained by an MRI device, single photon emission computed tomography (SPECT) images obtained by a SPECT imaging device, optical coherence tomography (OCT) images obtained by an OCT imaging device, and intravascular photoacoustic (IVPA) images obtained by an IVPA imaging device. The medical imaging device can obtain medical images when positioned outside the subject's body, spaced apart from the subject's body, adjacent to the subject's body, in contact with the subject's body, and / or inside the subject's body.

[0035] For ultrasound imaging equipment, the transducer array 112 transmits ultrasound signals toward the anatomical target 105 of the object, and receives echo signals reflected from the target 105 back to the transducer array 112. The ultrasound transducer array 112 can include any suitable number of acoustic elements, including one or more acoustic elements and / or multiple acoustic elements. In some cases, the transducer array 112 includes a single acoustic element. In some cases, the transducer array 112 can include an acoustic element array of any number of acoustic elements with any suitable configuration. For example, the transducer array 112 can include between 1 acoustic element and 10,000 acoustic elements, including values ​​such as 2 acoustic elements, 4 acoustic elements, 36 acoustic elements, 64 acoustic elements, 128 acoustic elements, 500 acoustic elements, 812 acoustic elements, 1000 acoustic elements, 3000 acoustic elements, 8000 acoustic elements and / or other values ​​of greater and lesser values. In some cases, the transducer array 112 may include an acoustic element array of any number of acoustic elements with any suitable configuration, such as a linear array, a planar array, a curved array, a curved array, a circular array, an annular array, a phased array, a matrix array, a one-dimensional (1D) array, a 1.x-dimensional array (e.g., a 1.5D array) or a two-dimensional (2D) array. An array of acoustic elements (e.g., one or more rows, one or more columns and / or one or more orientations) can be uniformly or independently controlled and activated. The transducer array 112 can be configured to obtain a one-dimensional, two-dimensional and / or three-dimensional image of the anatomical structure of the object. In some aspects, the transducer array 112 may include a piezoelectric micromachined ultrasonic transducer (PMUT), a capacitive micromachined ultrasonic transducer (CMUT), a single crystal, lead zirconate titanate (PZT), a PZT composite material, other suitable transducer types and / or combinations thereof.

[0036] Target 105 may include any anatomical structure or anatomical feature, such as a kidney, a liver, and / or any other anatomical structure of a subject. The present disclosure can be implemented in the context of any number of anatomical locations and tissue types, including, but not limited to: organs, including liver, kidney, gallbladder, pancreas, lungs; ducts; intestines; nervous system structures, including brain, dural sac, spinal cord, and peripheral nerves; urethra; and valves within blood vessels, blood, abdominal organs, and / or other systems of the body. In some aspects, target 105 may include a malignancy, such as a tumor, cyst, lesion, hemorrhage, or blood pool within any part of the human anatomy. The anatomical structure may be a blood vessel, such as an artery or vein of the vascular system of a subject, including cardiac vasculature, peripheral vasculature, neural vasculature, renal vasculature, and / or any other suitable lumen within the body. In addition to natural structures, the present disclosure may be implemented in the context of artificial structures, such as, but not limited to, heart valves, stents, shunts, filters, implants, and other devices.

[0037] The beamformer 114 is coupled to the transducer array 112. The beamformer 114 controls the transducer array 112, for example, for transmitting ultrasound signals and receiving ultrasound echo signals. In some aspects, the beamformer 114 can apply a time delay to the signal sent to each acoustic transducer within the array in the transducer 112, so that the acoustic signal is steered in any suitable direction of propagation away from the probe 110. The beamformer 114 can also provide an image signal to the processor circuit 116 based on a response to the received ultrasound echo signal. The beamformer 114 can include multiple beamforming stages. Beamforming can reduce the number of signal lines for coupling to the processor circuit 116. In some aspects, the transducer array 112 and the beamformer 114 combination can be referred to as an ultrasound imaging component.

[0038] The processor 116 is coupled to the beamformer 114. The processor 116 can also be described as a processor circuit that can include other components that communicate with the processor 116, such as a memory, the beamformer 114, a communication interface 118, and / or other suitable components. The processor 116 can include a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a controller, a field programmable gate array (FPGA) device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein. The processor 116 can also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration. The processor 116 is configured to process beamformed image signals. For example, the processor 116 can perform filtering and / or orthogonal demodulation to adjust the image signal. The processor 116 and / or 134 can be configured to control the array 112 to obtain ultrasound data associated with the target 105.

[0039] The communication interface 118 is coupled to the processor 116. The communication interface 118 may include one or more transmitters, one or more receivers, one or more transceivers, and / or circuits for sending and / or receiving communication signals. The communication interface 118 can include hardware components and / or software components that implement a specific communication protocol suitable for transmitting signals to the host 130 via the communication link 120. The communication interface 118 can be referred to as a communication device or a communication interface module.

[0040] The communication link 120 may be any suitable communication link. For example, the communication link 120 may be a wired link, such as a Universal Serial Bus (USB) link or an Ethernet link. Alternatively, the communication link 120 may be a wireless link, such as an Ultra Wideband (UWB) link, an Institute of Electrical and Electronics Engineers (IEEE) 802.11 WiFi link, or a Bluetooth link.

[0041] At host 130, communication interface 136 may receive the image signal. Communication interface 136 may be substantially similar to communication interface 118. Host 130 may be any suitable computing and display device, such as a workstation, personal computer (PC), laptop, tablet, or mobile phone.

[0042] The processor 134 is coupled to the communication interface 136. The processor 134 can also be described as a processor circuit, which can include other components that communicate with the processor 134, such as a memory 138, a communication interface 136, and / or other suitable components. The processor 134 can be implemented as a combination of software components and hardware components. The processor 134 can include a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a controller, an FPGA device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein. The processor 134 can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration. The processor 134 can be configured to generate image data based on image signals received from the probe 110. The processor 134 can apply advanced signal processing and / or image processing techniques to the image signal. In some aspects, the processor 134 can form a three-dimensional (3D) volume image from the image data. In some aspects, processor 134 can perform real-time processing on the image data to provide a streaming video of an ultrasound image of target 105. In some aspects, host 130 includes a beamformer. For example, processor 134 can be part of such a beamformer and / or otherwise communicate with such a beamformer. The beamformer in host 130 can be a system beamformer or a main beamformer (providing one or more subsequent stages of beamforming), while beamformer 114 is a probe beamformer or a microwave beamformer (providing one or more initial stages of beamforming).

[0043] Memory 138 is coupled to processor 134. Memory 138 may be any suitable storage device, such as cache memory (e.g., cache memory of processor 134), random access memory (RAM), magnetoresistive RAM (MRAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, a solid-state memory device, a hard disk drive, a solid-state drive, other forms of volatile and nonvolatile memory, or a combination of different types of memory.

[0044] The memory 138 can be configured to store object information, measurements, data, or files related to the medical history of the object, a history of procedures performed, anatomical or biological features, characteristics, or medical conditions associated with the object, computer-readable instructions (such as code, software, or other applications), and any other suitable information or data. The memory 138 can be located within the host 130. The object information can include measurements, data, files, other forms of medical history, such as, but not limited to, ultrasound images, ultrasound videos, and / or any imaging information related to the anatomical structure of the object. The object information can include parameters related to the imaging procedure, such as the anatomical scan window, the probe orientation, and / or the position of the object during the imaging procedure. The memory 138 can also be configured to store information related to the training and implementation of a machine learning algorithm (e.g., a neural network) and / or information related to the implementation of an image recognition algorithm, an image quantification algorithm, and / or an image acquisition guidance algorithm for detecting / segmenting anatomical structures, including those described herein.

[0045] The display 132 is coupled to the processor circuit 134. The display 132 may be a monitor or any suitable display. The display 132 is configured to display an ultrasound image of the target 105, an image video, and / or any imaging information.

[0046] The system 100 can be used to assist an ultrasound examiner in performing ultrasound scanning. The scan can be performed in a bedside setting. In some cases, the host 130 is a console or a movable cart. In some cases, the host 130 can be a mobile device, such as a tablet, a mobile phone, or a portable computer. During the imaging process, the ultrasound system is able to acquire ultrasound images of specific regions of interest within the object's anatomical structure. The ultrasound system 100 can then analyze the ultrasound image to identify various parameters associated with the acquisition of the image, such as a scan window, a probe orientation, an object position, and / or other parameters. The system 100 can then store the image and these associated parameters in a memory 138. In a subsequent imaging process, the system 100 can retrieve previously acquired ultrasound images and associated parameters to display to the user, which can be used to guide the user of the system 100 to use the same or similar parameters in a subsequent imaging process, as described in more detail below.

[0047] In some aspects, the processor 134 can utilize a deep learning-based prediction network to identify parameters of the ultrasound image, including anatomical scan windows, probe orientation, object position, and / or other parameters. In some aspects, the processor 134 can receive metrics or perform various calculations related to the physiological state of the imaged region of interest or object during the imaging process. These metrics and / or calculations can also be displayed to the ultrasound examiner or other user via the display 132.

[0048] Figure 2is a schematic diagram of a processor circuit according to aspects of the present disclosure. One or more processor circuits can be configured to perform the operations described herein. The processor circuit 210 can be in the probe 110, Figure 1 The processor circuit 210 may be implemented in the host system 130 or any other suitable location of the ultrasound system 100. For example, the processor 116 of the probe 110 can be part of the processor circuit 210. For example, the processor 134 and / or the memory 138 can be part of the processor circuit 210. In one example, the processor circuit 210 can communicate with the transducer array 112, the beamformer 114, the communication interface 118, the communication interface 136 and / or the display 132 and any other suitable components or circuits within the ultrasound system 100. As shown, the processor circuit 210 may include a processor 260, a memory 264 and a communication module 268. These elements may communicate with each other directly or indirectly, for example via one or more buses.

[0049] The processor 260 may include a CPU, a GPU, a DSP, an application specific integrated circuit (ASIC), a controller, an FPGA, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein. The processor 260 may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration. The processor 260 may also include an analysis module, as will be discussed in more detail below. The analysis module may implement various machine learning algorithms and may be hardware or software implementations. The processor 260 may additionally include a preprocessor of a hardware or software implementation. The processor 260 may execute various instructions, including instructions stored on a non-temporary computer readable medium such as a memory 264.

[0050] Memory 264 may include cache memory (e.g., cache memory of processor 260), random access memory (RAM), magnetoresistive RAM (MRAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, solid-state memory devices, hard drives, other forms of volatile and non-volatile memory, or a combination of different types of memory. In some cases, memory 264 includes non-transitory computer-readable media. Memory 264 may store instructions 266. Instructions 266 may include instructions that, when executed by processor 260, cause processor 260 to perform operations described herein with reference to probe 110 and / or host 130 ( Figure 1) instructions for the operations described. Instructions 266 may also be referred to as code. The terms "instructions" and "code" should be broadly interpreted to include any type of (one or more) computer-readable statements. For example, the terms "instructions" and "code" may refer to one or more programs, routines, subroutines, functions, processes, etc. The terms "instructions" and "code" may include a single computer-readable statement or many computer-readable statements. Instructions 266 may include various aspects of a preprocessor, a machine learning algorithm, a convolutional neural network (CNN), or various other instructions or codes. In some aspects, memory 264 may be or include a non-transitory computer-readable medium.

[0051] The communication module 268 can include any electronic circuitry and / or logic circuitry to facilitate direct or indirect communication of data between the processor circuit 210, the probe 110, and / or the host 130. In this regard, the communication module 268 can be an input / output (I / O) device. In some cases, the communication module 268 facilitates the processor circuit 210 and / or the probe 110 ( Figure 1 ) and / or host 130( Figure 1 ) direct or indirect communication between various elements.

[0052] Figure 3 is a flow chart of a method 300 for obtaining and analyzing ultrasound images corresponding to multiple views of a patient's anatomical structure in accordance with aspects of the present disclosure. As shown, method 300 includes a plurality of enumerated steps, but aspects of method 300 may include additional steps before, after, or between the enumerated steps. In some aspects, one or more of the enumerated steps may be omitted, performed in a different order, or performed simultaneously. The steps of method 300 can be performed by any suitable component within system 100, and all steps need not be performed by the same component. In some aspects, one or more steps of method 300 can be performed by or under the direction of a processor circuit, including, for example, processor 116 ( Figure 1 ), processor 134 ( Figure 1 )、Processor 260( Figure 2 ) or any other suitable component.

[0053] Aspects of method 300 may describe methods for providing guidance to non-expert ultrasound imaging system users to acquire high-quality ultrasound images. As will be explained below, these ultrasound images may include multiple views of regions of the patient's anatomical structure. In addition, aspects of the present disclosure relate to extracting features within the acquired ultrasound images. The extraction may include indicating regions of interest within each ultrasound image. The extracted features may be referred to as image features and may be signs, indicators, or symptoms of how a medical condition appears in an ultrasound image. These image features may reflect the severity and / or progression of a particular medical condition. In some aspects, hepatic steatosis (HS) may be a medical condition identified by an ultrasound imaging system. HS may include alcoholic fatty liver disease and non-alcoholic fatty liver disease (NAFLD). In some aspects, NAFLD may correspond to obesity and lead to various liver diseases. In this way, diagnosing NAFLD, other forms of HS, or any other condition may be accomplished by non-expert ultrasound users in a bedside setting, thereby expanding the availability of technologies capable of diagnosing these conditions. For example, the stage of progression may be determined for other medical conditions, including such as fusion diseases in the liver and / or other organs and / or other diseases in other organs.

[0054] At step 305, method 300 includes providing instructions to a user for obtaining an ultrasound image of a view of a patient's liver. The instructions to the user for obtaining an ultrasound image of a view of a patient's liver may be in any form. For example, a processor circuit of the ultrasound system may communicate with a display. The processor circuit (e.g., processor circuit 210) may output a graphical user interface including the user instructions to the display. In some aspects, as will be described with reference to FIG. Figure 4 As described, the user guidance may include pictorial graphics of an exemplary patient and ultrasound system probe and field of view. In some aspects, the user guidance may include text describing the position of the ultrasound probe relative to the patient's body. In some aspects, a graphical user interface displayed to a non-expert user may include graphics and / or text that may inform the non-expert user where to place the probe, how to move the probe, including in which direction to move the probe and various rotational positions, or in any other manner.

[0055] In some aspects, the guidance displayed to the user may include highlighting or outlining features within the acquired image in real time. For example, a live imaging loop may be displayed to a non-expert user as the user positions the ultrasound imaging probe near the patient's body. When the live image is displayed, outlines of various features within the live image may be displayed to help the user position the ultrasound imaging probe in the correct position.

[0056] In some aspects, the user guidance may include the display of a reference image. For example, a user of the ultrasound system may acquire an ultrasound image in real time and compare the acquired ultrasound image to the reference image to determine whether the probe is positioned in the correct position relative to the patient's body.

[0057] In some aspects, the user guidance may additionally include an indication of the expected dynamic behavior of the patient's anatomy during live imaging. For example, the movement of specific structures including the liver, kidneys, diaphragm, etc. or any other features within the patient's anatomy may be identified and described. The dynamic behavior may be described in text as well as exemplary images including live images or a series of images including a video clip.

[0058] The following will refer to the attached Figure 4-5 Additional aspects of providing guidance to non-expert users for acquiring high-quality ultrasound images of multiple views of a patient's anatomy are described in greater detail.

[0059] At step 310, the method 300 includes receiving an ultrasound image. The ultrasound image received at step 310 may include any suitable ultrasound image. For example, in some aspects, the received ultrasound image may include a B-mode image. Additionally, at step 310, any suitable type of ultrasound imaging system may be used to acquire the ultrasound image. For example, the ultrasound imaging system may be similar to that described in reference Figure 1 The ultrasound imaging system 100 described herein. At step 315, the method 300 includes performing a quality check of the ultrasound image. At step 315, the ultrasound imaging system 100 can ensure that the image received at step 310 is of sufficient quality to perform subsequent steps of the method 300. Performing a quality check on the received ultrasound image can include comparing the ultrasound image with a corresponding reference image, and determining whether a feature of interest exists within the ultrasound image, including whether the entire feature exists or whether a portion of the feature exists. In some aspects, the quality check of the received ultrasound image can be performed by a user of the ultrasound imaging system or a machine learning network of the ultrasound imaging system 100. For example, the machine learning network can be trained to analyze and / or compare the received ultrasound images and assign scores related to the quality of the ultrasound images. Reference will be made below to Figures 6 to 8 Additional aspects of performing a quality check on an acquired ultrasound image are described.

[0060] It should be noted that method 300 can be configured to identify multiple features within multiple ultrasound images received. For example, if method 300 is implemented to determine the staging score of NAFLD, method 300 can help users obtain multiple images corresponding to multiple views of the liver and kidney regions of the patient's anatomical structure. In some aspects, six features of the liver and kidney regions of the patient's anatomical structure can indicate the degree of NAFLD in the patient. Method 300 can be configured to extract these six features within the obtained ultrasound image. As will be explained below, it is expected that any number of features corresponding to any suitable medical condition can be extracted. In the example of identifying six features associated with NAFLD, steps 305, 310 and 315 can be performed multiple times for each of the six features. For example, when step 305 is performed for the first time, the processor circuit 210 of the ultrasound imaging system 100 can provide guidance to the system user for obtaining ultrasound images related to the features of liver and kidney contrast. At step 310, these ultrasound images can be obtained, and at step 315, these ultrasound images can be analyzed to ensure that they have sufficient quality. After performing step 315 with respect to the first feature, liver-kidney contrast, the processor circuit 210 may revert to step 305 and perform steps 305, 310, and 315 with respect to a second feature (e.g., posterior attenuation). The processor circuit 210 may then perform steps 310 and 315 to receive additional ultrasound images associated with that feature of posterior attenuation, and then the processor circuit 210 may again revert to step 305 associated with the additional feature, and so on. In some aspects, the processor circuit 210 may perform a reference step for each identified feature before repeating any of those same steps for the additional feature. Figure 3 Any subset of the steps shown and described for method 300.

[0061] At step 320, method 300 includes determining whether the ultrasound image is of sufficient quality. Determining whether the ultrasound image is of sufficient quality may include any suitable process. For example, a guidance toolkit may be provided to help a user check whether the acquired image is of sufficient quality. In some aspects, image quality may be visually performed by comparing the currently acquired image with a reference image. Reference images adjacent to the acquired image may be provided to inexperienced users for comparison. In some examples, an experienced user may verify the image quality of the acquired image without reference to the reference image. In some aspects, the ultrasound imaging system 100 may automatically perform image quality checks using various algorithms or modules. For example, the processor circuit 210 of the ultrasound imaging system 100 may be configured to implement a machine learning algorithm or image processing technique to automatically assess the quality of a particular image.

[0062] At step 325, method 300 includes selecting one or more ultrasound images for each view. As previously described with reference to step 315, multiple features of a target region of the patient's anatomy may be required to perform accurate staging of the patient's medical condition. In this way, different views of the patient's anatomy can better illustrate different features. For example, there may be a one-to-one relationship between views and features. For purposes of the present disclosure, a view may correspond to a probe position relative to the patient's anatomy. In this way, for each feature to be extracted from the acquired ultrasound image, the ultrasound imaging system 100 can provide the user with different guidance for positioning the ultrasound imaging probe at different positions relative to the patient (e.g., at step 305).

[0063] However, multiple ultrasound images corresponding to the same view or probe position may be received at step 310. Thus, the ultrasound imaging system 100 may select one ultrasound image corresponding to a view for each of the views of the patient's anatomy. In some aspects, the ultrasound imaging system may be configured to select additional ultrasound images for a single view.

[0064] The selection of one or more ultrasound images for each view may be accomplished by comparing each ultrasound image associated with a reference image. For example, an ultrasound imaging system user may examine each ultrasound image associated with a single view and compare them with the reference image. The ultrasound imaging system user may then select the ultrasound image associated with the view that most closely corresponds to the reference image from the plurality of ultrasound images. In some aspects, the ultrasound imaging system 100 may automatically select one or more ultrasound images for each view. For example, various image processing techniques and machine learning algorithms may be implemented to select an image associated with a particular view that most closely corresponds to the reference image from the plurality of ultrasound images received. Figure 6 Additional aspects of selecting one or more ultrasound images for each view are described in more detail.

[0065] At step 330, method 300 includes determining whether the selected ultrasound image shows cirrhosis. In some embodiments, determining whether the selected ultrasound image shows cirrhosis includes analyzing a liver background within the selected image. In some aspects, elastic imaging can be used to determine whether the liver background indicates cirrhosis. In embodiments where elastic imaging is not available, the liver background of the ultrasound image can be used to identify whether the liver background corresponds to a normal or healthy state or whether the liver background indicates cirrhosis.

[0066] In some aspects, step 330 may be performed by analyzing any one of the ultrasound images selected at step 325. For example, in a specific embodiment where six features are selected at step 325 and six ultrasound images are selected, the processor circuit 210 of the ultrasound imaging system 100 may analyze each of the six ultrasound images at step 330 to determine whether the liver background indicates cirrhosis. In some aspects, only one of the six ultrasound images may be selected. In other aspects, a single ultrasound image may be selected and analyzed at step 330, including any ultrasound image received at step 310.

[0067] Note that any machine learning algorithm described herein may be configured to perform step 330. For example, the ultrasound imaging system 100 and any machine learning algorithm (such as reference Figure 8 The algorithm 800 described herein may include a cirrhosis analysis module. The cirrhosis analysis module may be configured to determine whether the ultrasound image received at step 310 and / or selected at step 325 indicates cirrhosis.

[0068] At step 335, method 300 includes outputting an indication of cirrhosis. In some examples, if the ultrasound image received at step 310 indicates cirrhosis, the staging value assigned to NAFLD or any other medical condition may not be accurate enough. In such an example, method 300 proceeds to step 335. Step 335 includes indicating to the user with ultrasound imaging system 100 and associated methods that a correct liver staging may not be provided in this case. In this manner, method 300 can end at step 335.

[0069] In some aspects, after completing step 335 , the ultrasound imaging system 100 may also generate a report for a user of the ultrasound imaging system, a patient, or another physician indicating that the ultrasound images indicate cirrhosis and that the method 300 is not completed as a whole.

[0070] At step 340, method 300 includes extracting features from the selected ultrasound image. If it is determined at step 330 that the ultrasound image obtained at step 310 does not indicate liver cirrhosis, method 300 may proceed to step 340. As previously described with reference to steps 305-315, steps 340-350 may be performed multiple times for each identified feature. In an example where method 300 is implemented as a staging value for diagnosing NAFLD and processor circuit 210 is configured to identify and extract 6 features, processor circuit 210 may identify an ultrasound image corresponding to a first feature (e.g., liver-kidney contrast), and extract the feature from the selected ultrasound image at step 340. Then after completing step 345 (described below), processor circuit 210 may identify an ultrasound image corresponding to a second feature (e.g., rear attenuation) and perform steps 340-350 again. This process may continue for each feature or view until all selected ultrasound images corresponding to different views have been analyzed according to steps 340-350.

[0071] The processor circuit 210 may extract features from the ultrasound image in any suitable manner. Figure 8 Explained in more detail, the processor circuit 210 may implement a machine learning algorithm to extract features from the selected ultrasound image. In some cases, the processor circuit 210 may apply various pre-processing techniques to the selected ultrasound image before extracting the features. For example, the processor circuit 210 may adjust various characteristics of the selected image, such as image contrast, brightness, image size, coloring, or any other characteristics. In some cases, the processor circuit 210 may additionally apply various image processing techniques to the selected ultrasound image before or after implementing the machine learning network. For example, the processor circuit 210 may perform various image processing techniques, such as edge recognition of features within the selected ultrasound image, pixel-by-pixel analysis to determine the transition between bright pixels and dark pixels, filtering, or any other suitable technology to determine the location of various organs or structures shown within the selected ultrasound image.

[0072] At step 345, the method 300 includes assigning a feature score. In some aspects, the processor circuit 210 may assign a feature score to the selected ultrasound image by implementing a machine learning algorithm. The machine learning algorithm that assigns the score to the selected ultrasound image after the corresponding feature has been extracted may be the same machine learning algorithm that may be used to extract the feature at step 340. In some aspects, the machine learning algorithm used to assign the feature score may be a separate algorithm.

[0073] The scoring criteria implemented by the processor circuit 210 can be of any suitable type. For example, the processor circuit 210 can be configured to assign scores to features and / or ultrasound images using an ultrasound fatty liver index (US-FLI) scoring system. For example, the features that the processor circuit 210 is configured to extract can include liver-kidney contrast, posterior attenuation, portal vein blur, gallbladder wall visualization, diaphragm definition, and focal fat residue. Note that any suitable number or type of features can be extracted from the selected ultrasound image, including any features associated with different medical conditions. In this example, the liver-kidney contrast feature can be assigned an integer score on a scale of 0-3, where a score of 0 corresponds to an isoechoic area and a score of 3 corresponds to a severely hyperechoic area. The remaining features, including posterior attenuation, portal vein blur, gallbladder wall visualization, diaphragm definition, and focal fat residue, can be assigned integer scores of 0 or 1. For example, if the ultrasound image shows attenuation, a score of 1 can be assigned to the posterior attenuation, and if the ultrasound image does not show attenuation, a score of 0 can be assigned to the posterior attenuation. If there is vessel blurring within the selected ultrasound image, the portal vein blurring feature may be assigned a score of 1, and if there is no vessel blurring, the portal vein blurring feature may be assigned a score of 0. If it is difficult to visualize the gallbladder wall, the gallbladder wall visualization score may be 1, and if the gallbladder wall is easy to visualize, it may be 0. If the septum is obscured in the corresponding ultrasound image, the septum definition feature may be assigned a score of 1, and if the septum is clearly defined, the septum definition feature may be assigned a score of 0. If focal fat remnants are present in the corresponding ultrasound image, the focal fat remnant feature may be assigned a score of 1, and if the focal fat remnants are not present, the focal fat remnant feature may be assigned a score of 0.

[0074] For any of the above scoring criteria associated with the six features of NAFLD, the scoring ranges may be different, including any suitable ranges. In addition, the assigned scores may be non-integers. Other scoring criteria may include scoring features with percentages, measurements or values ​​of features within the selected ultrasound image (such as estimated volume or weight of fat or other tissue or material), and any other scoring criteria.

[0075] At step 350, the method 300 includes displaying the selected ultrasound image, an indication of the extracted features, and / or a feature score. After assigning a score to each feature at step 345, the processor circuit 210 may display the result of the score assignment at step 350. For example, with respect to the kidney-liver contrast feature, multiple images related to the kidney-liver contrast may have been acquired at step 310. At step 325, one of the ultrasound images may have been selected as the ultrasound image that is most strongly associated with the liver-kidney contrast feature. At step 340, the processor circuit may extract or identify the liver-kidney contrast feature from the selected image. At step 345, a score may be assigned to the feature. At step 350, the processor circuit 210 may output to a display (e.g., display 132) the ultrasound image selected at step 325 and an indication that the ultrasound image is selected to show the liver-kidney contrast feature, as well as the score assigned to the image and / or feature at step 345. In this way, the user can confirm that the ultrasound image is correctly selected and analyzed. In some aspects, this process of displaying the selected ultrasound images and corresponding features and feature scores to the user can help non-expert users of ultrasound images trust the results of the method 300, thereby increasing the likelihood that the method 300 will be more widely adopted. Figures 9A-9F Additional details of step 350 are described in greater detail.

[0076] At step 355, method 300 includes determining an installment score or installment value. Figure 8 As described, a machine learning algorithm may be implemented to determine a staging score. In some aspects, the machine learning algorithm that may determine the stage score may be the same machine learning algorithm that may be used to extract features from the selected ultrasound image and / or the same machine learning algorithm that may be used to assign feature scores to the extracted features. The staging value may include a numeric value (e.g., an integer or value within a range) or text (e.g., normal, mild, moderate, or severe).

[0077] In some aspects, various mathematical formulas or relationships can be implemented to determine the staging score based on the various feature scores previously described. In one example, the staging score can be the sum of each of the various feature scores. For example, individual feature scores can be added to produce a staging score. Various ranges of staging scores can determine the NAFLD stage of the patient. In one example, a staging score of 0-1 can correspond to a normal liver showing no signs of NAFLD. Staging scores 2-3 can indicate mild NAFLD or stage 1 NAFLD. Staging scores 4-5 can indicate moderate or stage 2 NAFLD, and staging scores 6 or more can indicate severe or stage 3 NAFLD. Other relationships can also be established, including that the staging score is the average, median or weighted average of the individual feature scores. In addition, any suitable linear or nonlinear relationship can be established between the various feature scores and the final staging score. For example, the processor circuit 210 can weight different individual feature scores differently according to the level of correlation between any one feature and the NAFLD stage.

[0078] At step 360, the method 300 includes displaying the staging score, one or more ultrasound images, one or more indications of the extracted features, and / or one or more feature scores. After determining the staging score at step 355, the processor circuit 210 may display the information. Figure 10-11 Additional aspects of step 360 are explained in more detail. The processor circuit 210 can be configured to display each selected ultrasound image corresponding to each feature and the corresponding feature score for each feature. The processor circuit 210 can additionally show a staging score, including the stage of the patient's NAFLD (e.g., stage 1, stage 2, or stage 3). In some aspects, displaying this information can increase the level of confidence that a user (particularly a non-expert user) may have in the results of method 300.

[0079] In some aspects, the processor circuit 210 can additionally generate a report including the results of the method 300 for a user, a patient, a physician, another expert physician, or any other individual. The report can include a document including the selected ultrasound image, features, feature scores, final staging scores, and stages. In some aspects, the generated report can also describe methods for determining final staging scores based on individual feature scores and methods for extracting features and assigning feature scores. In some embodiments, the expert physician can review the report to ensure that the method 300 is successfully completed and the staging diagnosis is accurately performed. Additional features and details of the method 300 will be shown and described with reference to the following figures.

[0080] Figure 44 is a schematic diagram of a graphical user interface 400 for displaying user instructions for obtaining ultrasound images for diagnosis according to aspects of the present disclosure. As described with reference to step 305 of method 300, the user instructions may include a graphical representation of the patient's body (with the probe positioned next to the body), text describing the position of the patient and the probe, reference images providing examples of the types of images that the user needs to acquire, and a description of the expected dynamic movement of the patient's anatomical structures within the expected ultrasound image.

[0081] The graphical user interface may include a title 480. The title 480 may indicate to a user of the ultrasound image the type of ultrasound image to be obtained, including the view of the patient's anatomy to be obtained and the features to be extracted from the resulting image. Figure 4 In the example shown, the graphical user interface 400 may indicate to the user that the user guidance provided within the interface 400 corresponds to obtaining an ultrasound image of a first feature (eg, liver-kidney contrast).

[0082] The user guidance within the graphical user interface 400 includes a graphical representation 410. The graphical representation 410 may include a stylized graphic of a patient's body 412 and a stylized graphic of an ultrasound transducer probe 420.

[0083] The patient body 412 may graphically depict the position of the patient for obtaining an ultrasound image of a view (in this case, view A). In the example provided, the user instructions may show the patient in a supine position with the right arm maximally abducted. The probe 420 may be shown adjacent to the patient body 412 at a position 414 relative to the patient's anatomy, and the desired orientation of the probe may be shown. Additionally, corresponding text 405 may also indicate the probe position to the user. For example, Figure 4 The text 405 shown in the example of may indicate that the probe position should be the right intercostal position. In some examples, the text 405 may further indicate the view plane of the ultrasound imaging probe. The view plane may include the transverse plane, the sagittal plane, or the frontal plane. Figure 4 As shown, a field of view 425 extending from the distal end of the probe 420 can be displayed. The field of view 425 can help the physician guide the probe 420 into the patient's anatomy.

[0084] In some examples, to obtain view A corresponding to the liver-kidney contrast feature and / or any other view or feature described herein, the probe can be positioned so that the liver-kidney interface is positioned in the middle region of the B-mode ultrasound image. This can help avoid distortion of ultrasound propagation through the boundaries between organs (e.g., liver or kidney).

[0085] In some aspects, additional user guidance may include instructions that the patient should hold their breath during imaging, or any other instruction to the patient, physician, or user. The user guidance may include various imaging control parameters, such as intensity, gain, focus, beam steering, time gain compensation, or any other parameter.

[0086] Reference image 455 can be additionally included in graphical user interface 400. In some aspects, reference image 455 can be an image of the same region of the anatomical structure from different patients. Reference image 455 can be an image collected by an expert user of the ultrasound imaging system. Reference image can be an image showing view A (e.g., liver-kidney contrast) of the region. Reference image 455 can be a view of the liver-kidney region of any suitable stage of NAFLD. In some aspects, reference image 455 can be associated with label 450.

[0087] Also like Figure 4 As shown, the graphical user interface 400 includes an indication of the expected dynamic behavior of the ultrasound image collected at the expected view. In some aspects, the expected dynamic behavior may include text 460 describing the expected dynamic behavior. For example, the expected dynamic behavior of a particular view may include the movement of the diaphragm below the liver. In some aspects, any other dynamic behavior may be expected. In some aspects, the expected dynamic behavior may include a loop of multiple ultrasound images showing movement within the patient's anatomical structure. In some aspects, this loop may replace the reference image 455. The reference image 455 may include multiple images obtained continuously by an expert sonographer in a previous imaging process. In some aspects, the expected dynamic behavior may include a stylized loop of animated images or images showing the expected movement.

[0088] As previously described, showing each of these aspects of the graphical user interface 400 to a user of the ultrasound imaging system 100 may help assure the user that the imaging procedure is being performed correctly and that the end results (eg, NAFLD staging values) are accurate.

[0089] Figure 5 FIG. 5 is a diagram of a table 500 of user guidance for obtaining ultrasound images of different views for diagnosis according to aspects of the present disclosure. Figure 5 As shown, table 500 may provide a checklist including properties of acquired ultrasound images that should be satisfied to a user of system 100. Note that table 500 may be organized in any suitable manner, including additional rows or columns or having fewer rows or columns as shown.

[0090] The first column 502 of table 500 includes a list of views to be obtained. In some aspects, the views may include six views corresponding to liver-kidney contrast, posterior attenuation, portal vein blur, gallbladder wall visualization, diaphragm definition, and focal fat residue. However, it should be understood that any number or type of views may be included in column 502. Each of these views may define a row (e.g., rows 522, 524, and 526, etc.). Each subsequent column may provide the user with attributes or information associated with each of these views.

[0091] Column 504 may include acquired images corresponding to the views. Referring to view A in row 522, the acquired images of column 504 may be images acquired during the current procedure by a user of the ultrasound imaging system 100. The acquired images 504 may be compared to reference images 506 and / or analyzed according to any other criteria listed in table 500.

[0092] In some examples, the reference image of column 506 may be Figure 4 The image 455 may be the same image, or may be different depending on the view. The acquired image of column 504 may be compared with the reference image of column 506 to ensure that the acquired image 504 accurately captures the view of row 522 and other rows.

[0093] Column 508 may include information or directions related to the probe position and / or angle. The information may be in any suitable form, including text or visual graphics, images, or symbols. For example, the probe position and / or angle may be substantially similar to the reference position and / or angle. Figure 4 A graphical representation 410 is shown and described.

[0094] Column 510 may include the expected dynamic behavior of the one or more images collected. The expected dynamic behavior may be presented in any suitable form. For example, the expected dynamic behavior listed in column 510 may be substantially similar to the reference Figure 4 Describe the expected dynamic behavior 460.

[0095] Column 512 may include the order in which the various views are to be acquired. In some aspects, the order may be determined by a user of the ultrasound imaging system 100 prior to an imaging procedure. In some aspects, the order may be provided to the user and may be based on a recommended order. In some aspects, the order of the views may not be specified.

[0096] Table 500 may be at least part of a guidance toolkit provided to a user to help the user check whether the acquired images or sequences are of sufficient quality (with clear appearance of the kidneys and / or liver or other structures) to ensure that the determined final stage value is accurate.

[0097] Figure 61 is a schematic diagram of ultrasound images acquired for multiple views and corresponding reference images for comparison according to aspects of the present disclosure. In some aspects, multiple ultrasound images corresponding to a single view can be acquired. For example, during an ultrasound imaging procedure, a user of system 100 can specify an intended view to be captured before acquiring an ultrasound image. Based on the indication, each ultrasound image acquired after receiving the indication can be associated with a specific view.

[0098] For example, a user of the ultrasound imaging system 100 may provide user input indicating that view A is the intended view to be captured. After receiving the indication, a series of ultrasound images 620 may be received. In some aspects, one of the images 620 may be selected as the one that most accurately captures view A. For purposes of this disclosure, the image that most accurately captures the intended view may be referred to as a key image. To select a key image for view A, each image 620 associated with view A may be compared to a reference image 630, as indicated by arrows 625. The reference image 630 may be a key image that is associated with view A. Figure 4 The same image as reference image 455 described is shown.

[0099] like Figure 6 As shown, image 622 in image 620 may be selected as a key image. Image 622 may be selected according to various methods. In one aspect, image 622 may be selected by a user of ultrasound imaging system 100. In another aspect, image 622 may be automatically selected by a processor circuit of ultrasound imaging system 100. For example, processor circuit 210( Figure 2 ) Various image processing techniques may be implemented to determine a similarity score between each of the images 620 and the reference image 630. The image 620 having the highest similarity score may be selected as the key image 622. In some aspects, a machine learning network may be used to determine the similarity score for each of the images 620.

[0100] In examples where key image 622 is automatically selected by processor circuit 210 , a graphical user interface may be provided to the user that allows the user to provide user input confirming that key image 622 best represents view A and corresponds to reference image 630 .

[0101] A similar process may be performed for each of the views to select a key image for each view. Figure 6 As shown, ultrasound image 640 may correspond to view B. Each of these images 640 may be compared to reference image 650, as indicated by arrow 645. In the example shown, image 642 may be selected as a key image.

[0102] Ultrasound image 660 may correspond to view C. Each of these images 660 may be compared to reference image 670, as indicated by arrow 665. In the example shown, image 642 may be selected as a key image.

[0103] Note that the ultrasound images (eg, images 620, 640, and 660) may be in any suitable format. For example, the received ultrasound images may be in ultrasound DICOM image format or raw data in a bedside ultrasound system.

[0104] Figure 7 is a schematic diagram of an acquired ultrasound image and identified features according to aspects of the present disclosure. Figure 7 As shown, each ultrasound image may correspond to one or more features. In some aspects, each of the views may correspond to one feature. However, an ultrasound image selected as a key image for a particular view may include a feature of that view and another feature corresponding to a different view.

[0105] For example, the relationship between six different views and six corresponding features may be a one-to-one correspondence.In some aspects, an image corresponding to one view may contain multiple features corresponding to multiple views.

[0106] Figure 7 Ultrasound image 710 is included. Image 710 may be a key image of view A. In this way, image 710 may be a key image of view A. Figure 6 The image 710 may include a feature 712. For example, the feature 712 may correspond to liver-kidney contrast. As shown by arrow 714, the feature 712 may be extracted from the image 710. The feature may be stored as feature A 716 in a memory that is in communication with the processor circuit 210 and associated with the image 710.

[0107] Similarly, Figure 7 The ultrasound image 720 is included. The image 720 may be a key image of the view B. In this way, the image 720 may be a key image of the reference Figure 6 Image 720 may include feature 722. For example, feature 722 may correspond to rear attenuation. Feature 722 may be extracted from image 720, as indicated by arrow 724. The feature may be stored as feature B 726 in a memory in communication with processor circuit 210 and associated with image 720.

[0108] Figure 7 Additionally, an ultrasound image 730 is included. Image 730 may be a key image for view C. In this manner, image 730 may be a key image for view C. Figure 6Image 730 may include feature 732. For example, feature 732 may correspond to portal vein blurring. Feature 732 may be extracted from image 730, as indicated by arrow 734. The feature may be stored as feature C 736 in a memory that is in communication with processor circuit 210 and associated with image 730.

[0109] like Figure 7 As shown, image 730 may additionally include a portion of feature 722 and a portion of feature 742. Processor circuit 210 may extract all relevant features from any of the images shown. In the example of view C shown, processor circuit 210 may extract feature 722 as indicated by arrow 740. In some aspects, a relationship between feature B 726 and image 730 may be established. Processor circuit 210 may assign a score to feature 722 based on an analysis of feature 722 as shown in image 720 and image 730, as will be discussed in more detail below.

[0110] Feature 742 may be a feature corresponding to a different view (e.g., view D). Feature 742 may be associated with gallbladder wall visualization. Processor circuit 210 may extract feature 742 from image 730, as indicated by arrow 744. The feature may be stored as feature D 746 in a memory in communication with processor circuit 210 and associated with image 730. In some aspects, an additional image may correspond to view D and may also include a depiction of feature 742.

[0111] Figure 8 800 according to various aspects of the present disclosure. The machine learning algorithm 800 can perform various tasks associated with the ultrasound imaging system 100. For example, the machine learning algorithm 800 can extract features from ultrasound images, such as reference Figure 7 As described, a score is assigned to each extracted feature, and a staging value is assigned based on the feature score. In some aspects, the machine learning algorithm 800 can perform other functions, such as selecting a key image from a set of ultrasound images associated with a particular view, performing a quality check on a received ultrasound image, or any other function described herein.

[0112] The machine learning algorithm 800 may include any suitable type of machine learning algorithm. For example, the machine learning algorithm 800 may include one or more convolutional neural networks (CNNs). Aspects of the machine learning algorithm 800 may be scaled to include any suitable number of CNNs. The machine learning algorithm 800 can be trained to recognize various anatomical landmarks or features within the patient's anatomy, including liver-kidney contrast, posterior attenuation, portal vein blur, gallbladder wall visualization, septum definition, and focal fat remnants, as well as any other features.

[0113] At step 805 of the algorithm, the machine learning algorithm 800 may receive an ultrasound image. The ultrasound image may be a key image. In some aspects, the network 800 may receive six key images, each key image corresponding to at least one view.

[0114] At step 810, preprocessing of the data (e.g., the image received at step 805) can be performed. In some aspects, preprocessing can include adjusting parameters of the input image. For example, various filters can be applied to enhance the visibility of image features. For example, the filter applied to the image can increase the visibility of the liver boundary, blood vessels within the liver, diaphragms, focal fat residues (if present), or any other features. Preprocessing can also include contrast enhancement, noise reduction, or any other image changes. In some aspects, the input image can be resized or otherwise modified before extracting image features at step 815.

[0115] At step 815, the machine learning algorithm 800 may extract features or regions of interest from each image received at step 805. In some aspects, this extraction may include identifying regions within each image that most closely correspond to the view of each received image. Figure 7 As explained, any of the images of step 805 may be primarily associated with a single view of the patient's anatomy. However, any of the images of step 805 may additionally include features corresponding to other views. As shown in step 820, primary features may be extracted from each of the images of step 805. In other aspects, additional features may be extracted from each image and related to each other depending on which view each feature corresponds to.

[0116] Step 830 of the machine learning algorithm 800 may correspond to a multi-task classification step and may include several layers. These layers may be any suitable number of convolutional layers, fully connected layers, or any other type of layers. In some aspects, each convolutional layer may include a set of filters configured to extract features from an input (e.g., an image of step 805 and / or features of step 820). The number of implemented layers and filters may vary according to the embodiment. In some cases, convolutional layers and fully connected layers may utilize leaky corrected nonlinear (ReLU) activation functions and / or batch normalization. Fully connected layers may be nonlinear and may gradually reduce high-dimensional outputs to the dimensions of the predicted results (e.g., classified outputs or layers 855). Therefore, fully connected layers may also be referred to as classifiers. In some embodiments, fully convolutional layers may additionally be referred to as perception or perceptual layers.

[0117] In some aspects, the multi-task classification step 830 may include an input layer 840, a shared hidden layer 845, a task-specific hidden layer 850, and an output layer 855. The output layer 855 may include six classifications corresponding to each of the six features input at the input layer 840. Figure 8 In the example shown, the output layer 855 can determine a score 860 corresponding to feature A (e.g., kidney-liver contrast), a score 862 corresponding to feature B (e.g., posterior attenuation), a score 864 corresponding to feature B (e.g., portal vein blur), a score 866 corresponding to feature D (e.g., gallbladder wall visualization), a score 868 corresponding to feature E (e.g., septum definition), and a score 870 corresponding to feature F (e.g., focal fat residue). These scores 860, 862, 864, 866, 868, and 870 can have reference Figure 3 The machine learning algorithm 800 may be formatted in the format described in step 345 of method 300. However, any suitable format may be used, including any suitable range, percentage, ratio, or other value. In some aspects, the machine learning algorithm 800 may additionally determine a confidence level associated with each feature. This may include the network 800 determining a confidence level that a particular feature is shown within a corresponding image. For purposes of this disclosure, scores 860, 862, 864, 866, 868, and 870 may be referred to as sub-scores. In this manner, the sub-scores output from the machine learning algorithm 800 may be a grading of the extent to which a particular medical condition (e.g., NAFLD) is present in a received ultrasound image. In some aspects, as described in reference to Figure 7 As described, the input image of step 805 may contain features associated with multiple views. Thus, in some aspects, any of the sub-scores 860, 862, 864, 866, 868, and 870 may be based not only on aspects of one input image, but on aspects of any input image containing information associated with the features of the particular sub-score.

[0118] like Figure 8 As shown, the final step of the deep learning algorithm may include determining a staging value 880. Staging value 880 may be the stage of progression of the particular disease or condition being monitored. For example, a stage of 0, 1, 2, or 3 may correspond to a NAFLD condition. Staging value 880 may be of any suitable type and calculated in any suitable manner, including reference method 300 ( Figure 3 ) as described in step 355.

[0119] In some aspects, the machine learning algorithm 800 can output a feature vector at the output of the last convolutional layer. The feature vector can indicate objects or features detected from the corresponding input image or other data.

[0120] The machine learning algorithm 800 may implement or include any suitable type of learning network. For example, the machine learning algorithm 800 may include a convolutional neural network, a multi-class classification network, an encoder-decoder type network, or any suitable network or module that recognizes features within an image.

[0121] In an embodiment where the machine learning algorithm 800 includes an encoder-decoder network, the network may include two paths. One path may be a contraction path, in which a large image (such as the input image of step 805) may be convolved by several convolutional layers so that the size of the image changes in the depth of the network. The image may then be represented in a low-dimensional space or a flattened space. From the flattened space, an additional path may expand the flattened space to the original size of the image. In some embodiments, the encoder-decoder network implemented may also be referred to as a principal component analysis (PCA) method. In some embodiments, the encoder-decoder network may segment the image into pieces.

[0122] In additional embodiments of the present disclosure, the machine learning algorithm 800 may include a multi-class classification network. In such an embodiment, the multi-class classification network may include an encoder path. For example, an input image (e.g., one of the images of step 805) may be a high-dimensional image. The image may then be processed using a convolutional layer so that the size is reduced. The resulting low-dimensional representation of the image may be used to generate a feature vector. The low-dimensional representation of the image may additionally be used by a fully connected layer to regress and output one or more categories, such as the features listed in the output layer 855. In some aspects, a fully connected layer may process the output of an encoder or convolutional layer. In addition to other terms, a fully connected layer may also be referred to as a task layer or a regression layer.

[0123] It should be noted that in some aspects, more than six features may be extracted and more than six corresponding images illustrating these features may be received. For example, various types of vein blurring may be included as identified features, including blurring of the great hepatic veins, blurring of the main right portal vein, or blurring of the anterior and posterior right portal veins. Additional features may include liver echo texture and overall impression.

[0124] In some aspects, the machine learning algorithm 800 may include two separate modules for different tasks. For example, one module may be implemented to extract features from an input image. Another module may be implemented to determine a staging value 880. In some aspects, the module for extracting features from an input image may be an object detection algorithm, such as Yolov4. In some aspects, the machine learning algorithm 800 may include two separate machine learning algorithms. For example, one machine learning algorithm may be implemented to extract features from an input image. Another machine learning algorithm may be implemented to determine a sub-score associated with the extracted features and a staging value 880.

[0125] Any suitable combination or variation of the described machine learning algorithm 800 is fully contemplated. For example, the machine learning algorithm may include a fully convolutional network or layer or a fully connected network or layer, or a combination of both. Furthermore, the machine learning algorithm may include a multi-class classification network, an encoder-decoder network, or a combination of both.

[0126] Fig. 9A is a schematic diagram of a graphical user interface 900 displaying a selected ultrasound image 902 of a particular view 908 , features 904 , and scores 909 in accordance with aspects of the present disclosure.

[0127] The ultrasound image 902 included in the graphical user interface 900 may be a key image corresponding to the feature selection of view A. In some aspects, the image 902 may be a different image. Fig. 9A In the example shown, the identified feature may be liver-kidney contrast, as shown by label 908. To identify feature 904 within ultrasound image 902, processor circuit 210 may be configured to overlay a graphical element corresponding to the feature on ultrasound image 902. In some aspects, feature 904 may include a region of liver 906 and a region of kidney 907.

[0128] The graphical user interface 900 may include a label identifying view A 908 and a corresponding feature (e.g., liver-kidney contrast). An additional label provides a score 909 assigned to a particular feature. The score 909 may be the score 860 described as a result of the output layer 855 of the machine learning algorithm 800. In some aspects, the score 909 may be accompanied by a description of the score. Fig. 9A In the graphical user interface 900 provided in FIG. 9 , for example, a score 909 of three may be accompanied by the description “severely hyperechoic.” Different descriptions corresponding to other values ​​of the score 909 may be included.

[0129] In some aspects, graphical user interface 900 can additionally include an area configured to receive user input, such as a button, a text box, or any other form of user input. The user input can allow the user to confirm that feature 904 correctly identifies the corresponding view A and feature (e.g., liver-kidney contrast). In some aspects, the user can adjust the position of feature 904 within image 902, or provide other user input.

[0130] Fig. 9B is a schematic diagram of a graphical user interface 910 displaying a selected ultrasound image 912 of a particular view 916 , features 914 , and scores 918 in accordance with aspects of the present disclosure.

[0131] The ultrasound image 912 included in the graphical user interface 910 may be a key image corresponding to the view feature selection, or may be a different image. Fig. 9BIn the example shown, the identified feature may be rear falloff, as shown by label 916. Processor circuit 210 may overlay a graphical element identifying feature 914 on image 912.

[0132] The graphical user interface 910 may include a label identifying view B 916 and a corresponding feature (e.g., rear attenuation). Additional labels provide a score 918 assigned to a particular feature. The score 918 may be the score 862 described as a result of the output layer 855 of the machine learning algorithm 800. In some aspects, the score 918 may be accompanied by a description of the score, such as "attenuation". Different descriptions corresponding to other values ​​of the score 918 may be included.

[0133] Similar to Fig. 9A The graphical user interface 900, Fig. 9B Graphical user interface 910 may additionally include an area configured to receive user input confirming that feature 914 correctly identifies the corresponding view B and feature.

[0134] Fig. 9C is a schematic diagram of a graphical user interface 920 displaying a selected ultrasound image 922 of a particular view 926 , features 924 , and scores 928 in accordance with aspects of the present disclosure.

[0135] The ultrasound image 922 included in the graphical user interface 920 may be a key image corresponding to the view feature selection, or may be a different image. Fig. 9C In the example shown, the identified feature may be a portal vein blur, as shown by label 926. Processor circuit 210 may overlay a graphical element identifying feature 924 on image 922.

[0136] The graphical user interface 920 may include a label identifying view C 926 and a corresponding feature (e.g., portal vein blur). An additional label provides a score 928 assigned to a particular feature. The score 928 may be a score 864 described as a result of the output layer 855 of the machine learning algorithm 800. In some aspects, the score 928 may be accompanied by a description of the score, such as "blurry" or a description of a variation corresponding to other values ​​of the score 928. Like the graphical user interface described previously, Fig. 9C Graphical user interface 920 may include an area configured to receive user input confirming that feature 924 correctly identifies the corresponding view C and feature.

[0137] Fig.9D is a schematic diagram of a graphical user interface 930 displaying a selected ultrasound image 932 of a particular view 936 , features 934 , and scores 938 in accordance with aspects of the present disclosure.

[0138] The ultrasound image 932 included in the graphical user interface 930 may be a key image corresponding to the view feature selection, or may be a different image. Fig.9D In the example shown, the identified feature may be a gallbladder wall visualization, as shown by label 936. Processor circuit 210 may overlay a graphical element identifying feature 934 on image 932.

[0139] The graphical user interface 930 may include a label identifying the view D 936 and the corresponding feature (e.g., gallbladder wall visualization). An additional label provides a score 938 assigned to a particular feature. The score 938 may be the score 866 described as a result of the output layer 855 of the machine learning algorithm 800. In some aspects, the score 938 may be accompanied by a description of the score, such as "poor visualization" or a description of the change corresponding to other values ​​of the score 938.

[0140] Like the graphical user interface described previously, Fig.9D The graphical user interface 930 may include an area configured to receive user input confirming that the feature 934 correctly identifies the corresponding view D and feature.

[0141] Fig.9E is a schematic diagram of a graphical user interface 940 displaying a selected ultrasound image 942 of a particular view 946 , features 944 , and scores 948 in accordance with aspects of the present disclosure.

[0142] The ultrasound image 942 included in the graphical user interface 940 may be a key image corresponding to the view feature selection, or may be a different image. Fig.9E In the example shown, the identified feature may be a diaphragm definition, as shown by label 946. Processor circuit 210 may overlay a graphical element identifying feature 944 on image 942.

[0143] The graphical user interface 940 may include a label identifying the view E 946 and the corresponding feature (e.g., gallbladder wall visualization). An additional label provides a score 948 assigned to a particular feature. The score 948 may be the score 868 described as a result of the output layer 855 of the machine learning algorithm 800. In some aspects, the score 948 may be accompanied by a description of the score, such as "occlusion" or a description of the change corresponding to other values ​​of the score 948.

[0144] Like the graphical user interface described previously, Fig.9E Graphical user interface 940 may include an area configured to receive user input confirming that feature 944 correctly identifies the corresponding view E and feature.

[0145] Fig.9Fis a schematic diagram of a graphical user interface 950 displaying a selected ultrasound image 952 of a particular view 956 , features 954 , and scores 958 in accordance with aspects of the present disclosure.

[0146] The ultrasound image 952 included in the graphical user interface 950 may be the key image corresponding to the view feature selection, or may be a different image. In the example shown in FIG. 9 , the identified feature may be a focal fat residue, as shown by label 956. The processor circuit 210 may overlay a graphical element identifying feature 954 on the image 952.

[0147] The graphical user interface 950 may include a label identifying the view F 956 and the corresponding feature (e.g., focal fat residue). An additional label provides a score 958 assigned to the particular feature. The score 958 may be a score 870 described as a result of the output layer 855 of the machine learning algorithm 800. In some aspects, the score 958 may be accompanied by a description of the score, such as "occlusion" or a description of the change corresponding to other values ​​of the score 958. Like the graphical user interface described previously, Fig.9F The graphical user interface 950 may include an area configured to receive user input confirming that the feature 954 correctly identifies the corresponding view F and feature.

[0148] The machine learning algorithm 800 may be displayed to the user while executing the machine learning algorithm 800. Figures 9A-9F In this way, the user can see the images corresponding to each view, the identified features within each image, and the scores associated with each feature. As previously mentioned, the user can enter confirmation of the displayed data and / or results at different stages. By displaying references to the user during the execution of the deep learning algorithm Figures 9A-9F By describing the images, features, and scores, users are more likely to accept the final staging value as accurate and trust the results of the imaging process.

[0149] Fig.10 is a schematic diagram of a graphical user interface 1000 displaying ultrasound images, corresponding scores, and staging values ​​according to aspects of the present disclosure.

[0150] The graphical user interface 1000 can simultaneously display Figures 9A-9F Each of the six ultrasound images may correspond to the six views described previously.

[0151] A corresponding list of views, features, and scores is also displayed within graphical user interface 1000. For example, label 1031 may identify view A and may include a description of the view, including, for example, liver and kidney contrast. Fig. 9A A score 909 of the description is also displayed adjacent to the label 1031. The label 1031 and the score 909 may each correspond to the ultrasound image 902.

[0152] Similarly, tag 1032 may identify view B and may include a description of the view, including, for example, rear falloff. Fig. 9B The score 918 depicted is also displayed adjacent to the label 1032. The label 1032 and the score 918 may each correspond to the ultrasound image 912. The label 1033 similarly corresponds to view C and is associated with the score 928 and the ultrasound image 922. The label 1034 corresponds to view D and is associated with the score 938 and the ultrasound image 932. The label 1035 corresponds to view E and is associated with the score 948 and the ultrasound image 942. The label 1036 corresponds to view F and is associated with the score 958 and the ultrasound image 952.

[0153] Fig.10 Additionally included is a total score label 1050 and a total score 1052. The total score 1052 may be a combination of the scores 909, 918, 928, 938, 948, and 958 corresponding to the various views AF. For example, the total score 1052 may be the sum of the scores 909, 918, 928, 938, 948, and 958. In some aspects, the total score 1052 may be a different combination of these scores, such as an average, a median, or any other combination including various relationships or functions.

[0154] Fig.10 Also included is a NAFLD stage label 1060 and a staging value 1062. As previously described, the staging value 1062 can be based on the scores 909, 918, 928, 938, 948, and 958. Figure 3 As described in step 355 of method 300, the stage value 1062 can be additionally based on the total score 1052. For example, if the total score 1052 falls within a certain range of values, the range can correspond to one of the stage values. In the example shown, the total score 1052 can be 7, and the NAFLD stage value 1062 can be 3 corresponding to severe progression of NAFLD.

[0155] The graphical user interface 1000 may allow a user to select an ultrasound image in the provided images and a view in the provided views. For example, as shown by indicator 1020, the user may select ultrasound image 922. In response to the selection, the processor circuit 210 may be configured to modify the graphical user interface 1000 to highlight the selected ultrasound image 922. Similarly, the label 1033 and the score 928 may be highlighted because the label 1033 and the score 928 correspond to the ultrasound image 922. As additionally shown by indicator 1070, the user may alternatively select the label 1033 or the score 928. In response to the selection, the label 1033 and the score 928 and the ultrasound image 922 corresponding to the label 1033 and the score 928 may be highlighted.

[0156] In some aspects, after a user selects an ultrasound image or view within the graphical user interface 1000, the selected ultrasound image can be zoomed in, as shown in FIG. Fig.11 As shown and described.

[0157] In some aspects, the processor circuit 210 can be additionally configured to generate a report based on the information displayed in the graphical user interface 1000. The generated report can inform a physician who is not involved in the ultrasound imaging procedure of the results. The report can also be viewed by the patient or any other interested party. The report can include any information, including ultrasound images, overlays, views, features, and scores, such as Fig.10 and anything else previously shown and described.

[0158] Fig.11 is a schematic diagram of a graphical user interface 1100 displaying an ultrasound image 922 , corresponding scores, and a stage score 1062 in accordance with aspects of the present disclosure.

[0159] In some aspects, the graphical user interface 1100 can be displayed to the user in response to the user selecting an ultrasound image or view of the graphical user interface 1000 previously described. For example, Fig.11 The graphical user interface 1100 shown may correspond to the view displayed to the user upon selection of the ultrasound image 922 corresponding to view C.

[0160] As shown, the graphical user interface 1100 may include a magnified view of an ultrasound image 922. The ultrasound image 922 may include features 924 identified within the image. In some aspects, the ultrasound image 922 may include additional features, such as various metrics or automatically identified features.

[0161] The graphical user interface 1100 includes the staging value 1062 and a description 1164 of the staging value in the illustrated example, since the staging value 1062 is 3, the description 1164 may be "critical".

[0162] Graphical user interface 1100 may provide the user with a more detailed view of image 922. Fig.10 The graphical user interface 1100 may include any suitable information or detailed data associated with the ultrasound image 922, including any data obtained during the ultrasound imaging procedure. In some aspects, a user of the ultrasound imaging system 100 may view, for example, additional ultrasound images associated with view C, such as reference image C. Figure 6 Any image 660 described.

[0163] Fig.1212 is a flow chart of a method for determining a staging value for a medical condition in accordance with aspects of the present disclosure. As illustrated, method 1200 includes several enumerated steps, but aspects of method 1200 may include additional steps before, after, or between the enumerated steps. In some aspects, one or more of the enumerated steps may be omitted, performed in a different order, or performed simultaneously. The steps of method 1200 can be performed by any suitable component within system 100, and all steps need not be performed by the same component. In some aspects, one or more steps of method 1200 can be performed by or under the direction of a processor circuit, including, for example, processor 116 ( Figure 1 ), processor 134 ( Figure 1 )、Processor 260( Figure 2 ) or any other suitable component.

[0164] At step 1210, method 1200 includes outputting user instructions to a display to obtain a plurality of ultrasound images corresponding to a plurality of views of the patient's anatomy. For example, the user instructions may be of any suitable type, including text or visual graphics, including two-dimensional or three-dimensional images, still images, or animated images. The user instructions may also include a description of the expected dynamic behavior of the reference image and / or the ultrasound image acquired in real time. Various aspects of the user instructions may include reference images, Figure 4 Any features shown and described.

[0165] At step 1220, method 1200 includes controlling the transducer array to obtain a first ultrasound image corresponding to a first view of the patient's anatomy and a second ultrasound image of the patient's anatomy corresponding to a second view of the patient's anatomy. Aspects of the obtained ultrasound images may be referenced Figure 3 310 of method 300 is described below.

[0166] At step 1230, the method 1200 includes identifying, using a first machine learning algorithm, a first image feature associated with a medical condition of an anatomical structure of a patient within a first ultrasound image of the plurality of ultrasound images and a second image feature associated with a medical condition within a second ultrasound image of the plurality of ultrasound images. Image features may be extracted from the received ultrasound images in any suitable manner. For example, a machine learning network may be configured to extract features from an input image, such as by referring to Figure 7 to Figure 8 Described in more detail.

[0167] At step 1240, method 1200 includes determining a first sub-score for the first image feature and a second sub-score for the second image feature using a second machine learning algorithm. The sub-scores assigned to the corresponding image features may be in any suitable format or type. Figures 9A-9FVarious aspects of determining sub-scores associated with identified image features are shown and described.

[0168] At step 1250, method 1200 includes determining a staging value representing progression of a medical condition based on the first sub-score and the second sub-score. In some aspects, the staging value may be a staging value for progression of NAFLD within the patient. In some aspects, the staging value may be a staging value for another medical condition. Figure 3 Step 355 of method 300 describes aspects of determining the period value in further detail.

[0169] At step 1260, the method 1200 includes outputting a screen display to a display, the screen display including: a staging value and a visual representation of how the staging value was determined, including: a first ultrasound image, an indication of a first image feature in the first ultrasound image, and a first sub-score; or a second ultrasound image, an indication of a second image feature in the second ultrasound image, and a second sub-score. In some aspects, the ultrasound images displayed at step 1260 can be displayed sequentially or simultaneously. By displaying this information to a user of the ultrasound imaging system 100 during an imaging procedure, the user can be more confident that the resulting staging value is accurate.

[0170] Those skilled in the art will recognize that the above-described devices, systems, and methods can be modified in various ways. Therefore, it will be appreciated by those skilled in the art that the various aspects encompassed by the present invention are not limited to the specific exemplary aspects described above. In this regard, although illustrative aspects have been shown and described, a wide range of modifications, changes, and substitutions are contemplated in the foregoing disclosure. It should be understood that such changes may be made to the foregoing without departing from the scope of the present disclosure. Therefore, it is appropriate that the appended claims be interpreted broadly and in a manner consistent with the present disclosure.

[0171] In the claims, any reference signs placed in brackets shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps other than those listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements.

Claims

1. An ultrasound system (100), comprising: A processor (114, 134) configured to communicate with a display (132) and a transducer array (112) of an ultrasound probe (110), wherein the processor is configured to: controlling (1220) the transducer array to obtain a first ultrasound image corresponding to a first view of a patient's anatomy and a second ultrasound image of the patient's anatomy corresponding to a second view of the patient's anatomy; identifying (1230) a first image feature associated with a medical condition of the patient anatomy within the first ultrasound image and a second image feature associated with the medical condition within the second ultrasound image; determining (1240) a first sub-score for the first image characteristic and a second sub-score for the second image characteristic; determining (1250) a stage value indicative of progression of the medical condition based on the first sub-score and the second sub-score; and Outputting (1260) a screen display to the display, the screen display comprising: the installment value; and At least one of the following: the first ultrasound image, an indication of the first image feature in the first ultrasound image, and the first sub-score; or The second ultrasound image, an indication of the second image characteristic in the second ultrasound image, and the second sub-score.

2. The ultrasound system according to claim 1, wherein: The processor is further configured to: User instructions for obtaining the first ultrasound image corresponding to the first view of the patient anatomy are output (1210) to the display.

3. The ultrasound system according to claim 2, wherein: The user instructions include a graphical representation of a probe (420) and / or an orientation of the ultrasound probe.

4. The ultrasound system according to any one of claims 2 to 3, wherein: The user guidance includes a reference image (455) associated with the first view of the patient anatomy.

5. The ultrasound system according to any one of claims 2 to 4, wherein: The first user guidance includes a description of dynamic behavior associated with the first view of the patient anatomy (460).

6. The ultrasound system according to any one of claims 1 to 5, wherein: The processor is also configured to determine a quality associated with the first ultrasound image prior to identifying the first image feature within the first ultrasound image.

7. The ultrasound system according to claim 6, wherein: The processor is further configured to determine the quality based on a comparison between the first ultrasound image and a reference image associated with the first view of the patient anatomy.

8. The ultrasound system according to any one of claims 6 or 7, wherein: The processor is also configured to (320): identifying the first image feature within the first ultrasound image if the quality satisfies a threshold; If the quality does not meet the threshold, controlling the transducer array to obtain another ultrasound image corresponding to the first view of the patient anatomy.

9. The ultrasound system according to any one of claims 1 to 8, wherein: To determine the first sub-score and the second sub-score, the processor is further configured to implement a first machine learning algorithm.

10. The ultrasound system of claim 9, wherein: The first machine learning algorithm includes a multi-task learning model.

11. The ultrasound system according to any one of claims 9 or 10, wherein: To identify the first image feature and the second image feature, the processor is configured to implement a second machine learning algorithm different from the first machine learning algorithm.

12. The ultrasound system according to any one of claims 1 to 11, in, The patient anatomy includes a liver, and Wherein, the medical condition comprises hepatic steatosis.

13. The ultrasound system according to any one of claims 1 to 12, wherein: The first sub-score of the first image feature and the second sub-score of the second image feature correspond to an ultrasound fatty liver indicator.

14. The ultrasound system according to any one of claims 1 to 13, wherein: The first image feature and the second image feature each include a different one of the following: liver and kidney contrast, posterior attenuation, vessel blurring, gallbladder visualization, diaphragm attenuation visualization, or focal residual.

15. A computer-implemented method comprising: controlling (1220) a transducer array of an ultrasound imaging probe to obtain a first ultrasound image corresponding to a first view of an anatomical structure of a patient and a second ultrasound image of an anatomical structure of the patient corresponding to a second view of the anatomical structure of the patient; identifying (1230) a first image feature associated with a medical condition of the patient anatomy within the first ultrasound image and a second image feature associated with the medical condition within the second ultrasound image; determining (1240) a first sub-score for the first image characteristic and a second sub-score for the second image characteristic; determining (1250) a staging value indicative of progression of the medical condition based on the first subscore and the second subscore; as well as Outputting (1250) a screen display to a display, the screen display comprising: the installment value; as well as At least one of the following: the first ultrasound image, an indication of the first image feature in the first ultrasound image, and the first sub-score; or The second ultrasound image, an indication of the second image characteristic in the second ultrasound image, and the second sub-score.

Images