Systems and methods for quantifying b-lines and fused b-lines in lung ultrasound images

Abstract

This paper provides a system and method for quantifying B-lines and fused B-lines in lung ultrasound images, which improves the efficiency and accuracy of detecting such B-lines and fused B-lines, thereby facilitating the use of this imaging modality in additional point-of-care settings, such as emergency medical and intensive care settings. The system described herein includes an ultrasound imaging device and electronics communicating with the ultrasound imaging device, the electronics being configured to analyze one or more lung ultrasound video loops using a trained B-line classifier and a trained fused B-line classifier. In a specific aspect, one or more ultrasound video loops can be analyzed at two or more spatial resolution scales to extract B-line image features, which are used by the B-line classifier and the fused B-line classifier to predict the probability that a B-line candidate is a B-line or a fused B-line.

Description

Government Rights Statement

[0001] This invention was developed with the support of a grant from the U.S. Department of Health and Human Services, grant number HHS / ASPR / BARDA 75A50120C00097. The U.S. government holds specific rights to this invention. Technical Field

[0002] This disclosure generally relates to ultrasound imaging systems and methods for processing ultrasound images, and more specifically to ultrasound imaging systems and methods for improving the quantification of certain pathological features seen in lung ultrasound images. Background Technology

[0003] Lung ultrasound is an imaging technique that can be used at the point of care to assess the lungs in a variety of settings, including emergency medicine and critical care. This technique has been widely adopted as a portable, non-invasive, and radiation-free modality for evaluating lung and infectious diseases. A characteristic pathological feature seen in lung ultrasound is B-lines. B-lines are defined as discrete, vertical hyperechoic artifacts that appear as long bands originating from the pleural line and extending vertically along the length of the image. The number of B-lines (i.e., B-line count) is known to correlate with fluid accumulation in the interstitial space of the lung. As interstitial fluid accumulates horizontally and becomes more severe, more B-lines will be present, leading to the eventual fusion of B-lines. Therefore, these fused B-lines (also known as confluent B-lines) become difficult to distinguish from each other and cannot be accurately counted. Summary of the Invention

[0004] As described in more detail below, this disclosure relates to improved ultrasound imaging systems and methods for quantifying B-lines and fused B-lines, which are pathological features seen in lung ultrasound images. Because B-lines and fused B-lines can be observed under various conditions, automated evaluation of these pathological features can play an important role in the screening, diagnosis, and / or management of disease progression, for example, through standardized interpretation among experts and / or enabling use by undertrained lung ultrasound users.

[0005] According to some embodiments of this disclosure, a system for viewing and analyzing lung ultrasound images is provided, the system comprising: an ultrasound imaging device including one or more ultrasound imaging transducers configured to generate a lung ultrasound video loop of a subject, the lung ultrasound video loop including a plurality of lung ultrasound imaging frames; and an electronic device communicating with the ultrasound imaging device. The electronic device may include: a display device configured to display a graphical user interface; a computer-readable storage medium storing computer-readable instructions thereon to be executed by one or more processors; and one or more processors configured by the computer-readable instructions stored on the computer-readable storage medium to perform the following operations: (i) obtaining a lung ultrasound video loop of a subject, the lung ultrasound video loop including a plurality of lung ultrasound imaging frames; (ii) analyzing the lung ultrasound video loop using a B-line classifier and a fused B-line classifier to generate B-line data for the lung ultrasound video loop; and (iii) outputting the graphical user interface via the display device, the graphical user interface including the B-line data generated for the lung ultrasound video loop.

[0006] On one hand, B-line data for a lung ultrasound video loop can be generated by the following operations: preprocessing each imaging frame of the lung ultrasound video loop to obtain a preprocessed lung ultrasound video loop; determining a region of interest for B-line analysis for each imaging frame of the preprocessed lung ultrasound video loop; analyzing the region of interest for B-line analysis of each imaging frame to identify one or more B-line candidates; for each B-line candidate, extracting a set of B-line features from the imaging frames of the preprocessed lung ultrasound video loop; and classifying each B-line candidate among the B-line candidates based on the corresponding set of B-line features using the B-line classifier and the fused B-line classifier. The method involves: predicting the likelihood that the B-line candidate is a possible B-line and / or a possible fused B-line; identifying one or more possible B-lines and / or possible fused B-lines in each imaging frame of the preprocessed ultrasound video loop based on the classification of each of the B-line candidates; determining the maximum B-line count for the lung ultrasound video loop, wherein the maximum B-line count is the maximum number of possible B-lines appearing in any single imaging frame of the preprocessed lung ultrasound video loop; and determining whether the lung ultrasound video loop is positive for a fused B-line based on the classification of each of the B-line candidates.

[0007] On one hand, the B-line classifier can be a first trained machine learning model configured to receive multiple B-line features as input and output the probability that a B-line candidate is a possible B-line. On the other hand, the fusion B-line classifier can be a second trained machine learning model configured to receive multiple B-line features as input and output the probability that a fusion B-line candidate is a possible B-line.

[0008] On the one hand, the lung ultrasound video loop is considered positive for a fused B-line if the number of imaging frames containing possible fused B-lines in the preprocessed lung ultrasound video loop meets or exceeds a predefined minimum number of imaging frames.

[0009] In one aspect, identifying one or more B-line candidates within the region of interest for B-line analysis may include performing the following operations for each imaging frame of the preprocessed lung ultrasound imaging video loop: smoothing the intensity profile of the region of interest for the corresponding imaging frame; identifying one or more local peaks along the smoothed intensity profile of the region of interest for the corresponding imaging frame; and defining a region of interest for each of the identified one or more local peaks, wherein each region of interest corresponds to a B-line candidate.

[0010] On the one hand, a set of B-line features for each B-line candidate can be extracted from the region of interest of the B-line candidate defined in the imaging frames of the preprocessed lung ultrasound video loop.

[0011] In one aspect, identifying one or more B-line candidates within a region of interest for B-line analysis may include performing the following operations for each imaging frame of a preprocessed lung ultrasound imaging video loop: smoothing the intensity profile of the region of interest for the corresponding imaging frame using a first smoothing kernel; identifying one or more local peaks along the intensity profile smoothed using the first smoothing kernel; defining a region of interest for each of the one or more local peaks identified in the intensity profile smoothed using the first smoothing kernel, wherein each region of interest corresponds to a B-line candidate; smoothing the intensity profile of the region of interest for the corresponding imaging frame using a second smoothing kernel, wherein the size of the second smoothing kernel is different from that of the first smoothing kernel; identifying one or more local peaks along the intensity profile smoothed using the second smoothing kernel; and defining a region of interest for each of the one or more local peaks identified in the intensity profile smoothed using the second smoothing kernel, wherein each region of interest corresponds to an additional B-line candidate.

[0012] On one hand, a first set of B-line features for each B-line candidate can be extracted from the region of interest of the B-line candidate based on the intensity profile smoothed using a first smoothing kernel, and a second set of B-line features for each B-line candidate can be extracted from the region of interest of the B-line candidate based on the intensity profile smoothed using a second smoothing kernel.

[0013] On one hand, a set of B-line features extracted from the imaging frames of a preprocessed lung ultrasound video loop may include at least one B-line feature measured at two or more different spatial scales.

[0014] According to other embodiments of this disclosure, an image processing method is provided, comprising: preprocessing each imaging frame of a lung ultrasound video loop to obtain a preprocessed lung ultrasound video loop, wherein the lung ultrasound video loop includes a plurality of image frames; determining a region of interest for B-line analysis for each imaging frame of the preprocessed lung ultrasound video loop; analyzing the region of interest for B-line analysis of each imaging frame to identify one or more B-line candidates; for each B-line candidate, extracting a set of B-line features from the imaging frames of the preprocessed lung ultrasound video loop; and classifying each B-line candidate using a B-line classifier and a fused B-line classifier based on the corresponding set of B-line features. Candidates are classified to predict the likelihood that the B-line candidate is a possible B-line and / or a possible fused B-line; one or more possible B-lines and / or possible fused B-lines in each imaging frame of the preprocessed ultrasound video loop are identified based on the classification of each B-line candidate; a maximum B-line count is determined for the lung ultrasound video loop, wherein the maximum B-line count is the maximum number of possible B-lines appearing in any single imaging frame of the preprocessed lung ultrasound video loop; and the lung ultrasound video loop is determined to be positive for a fused B-line based on the classification of each B-line candidate.

[0015] On one hand, the B-line classifier can be a first trained machine learning model configured to receive multiple B-line features as input and output the probability that a B-line candidate is a possible B-line. On the other hand, the fusion B-line classifier can be a second trained machine learning model configured to receive multiple B-line features as input and output the probability that a fusion B-line candidate is a possible B-line.

[0016] On the one hand, the lung ultrasound video loop is considered positive for a fused B-line if the number of imaging frames containing possible fused B-lines in the preprocessed lung ultrasound video loop meets or exceeds a predefined minimum number of imaging frames.

[0017] In one aspect, identifying one or more B-line candidates within the region of interest for B-line analysis may include performing the following operations for each imaging frame of the preprocessed lung ultrasound imaging video loop: smoothing the intensity profile of the region of interest for the corresponding imaging frame; identifying one or more local peaks along the smoothed intensity profile of the region of interest for the corresponding imaging frame; and defining a region of interest for each of the identified one or more local peaks, wherein each region of interest corresponds to a B-line candidate.

[0018] In one aspect, identifying one or more B-line candidates within a region of interest for B-line analysis may include performing the following operations for each imaging frame of a preprocessed lung ultrasound imaging video loop: smoothing the intensity profile of the region of interest for the corresponding imaging frame using a first smoothing kernel; identifying one or more local peaks along the intensity profile smoothed using the first smoothing kernel; defining a region of interest for each of the one or more local peaks identified in the intensity profile smoothed using the first smoothing kernel, wherein each region of interest corresponds to a B-line candidate; smoothing the intensity profile of the region of interest for the corresponding imaging frame using a second smoothing kernel, wherein the size of the second smoothing kernel is different from that of the first smoothing kernel; identifying one or more local peaks along the intensity profile smoothed using the second smoothing kernel; and defining a region of interest for each of the one or more local peaks identified in the intensity profile smoothed using the second smoothing kernel, wherein each region of interest corresponds to an additional B-line candidate.

[0019] According to another embodiment of this disclosure, a computer program product is provided. The computer program product may include a non-transient computer-readable storage medium storing computer-readable instructions thereon, which, when executed by one or more processors, cause the one or more processors to perform the following operations: (i) obtaining a lung ultrasound video loop of an object, the lung ultrasound video loop including a plurality of lung ultrasound imaging frames; (ii) preprocessing each imaging frame of the lung ultrasound video loop to obtain a preprocessed lung ultrasound video loop; (iii) determining a region of interest for B-line analysis for each imaging frame of the preprocessed lung ultrasound video loop; (iv) analyzing the region of interest for B-line analysis of each imaging frame to identify one or more B-line candidates; (v) for each B-line candidate, extracting a set of B-line features from the imaging frames of the preprocessed lung ultrasound video loop; (vi) classifying the B-line candidates using a B-line classifier and a fused B-line classifier based on the corresponding set of B-line features. Each B-line candidate is classified to predict the probability that the B-line candidate is a possible B-line and / or a possible fused B-line; (vii) one or more possible B-lines and / or possible fused B-lines in each imaging frame of the preprocessed ultrasound video loop are identified based on the classification of each B-line candidate; (viii) a maximum B-line count is determined for the lung ultrasound video loop, wherein the maximum B-line count is the maximum number of possible B-lines appearing in any single imaging frame of the preprocessed lung ultrasound video loop; (ix) whether the lung ultrasound video loop is positive for fused B-lines is determined based on the classification of each B-line candidate; and (x) a graphical user interface is output via a display device, the graphical user interface including the maximum B-line count and / or the determination of whether the lung ultrasound video loop is positive for fused B-lines.

[0020] These and other aspects of the various embodiments will be apparent from the embodiments described below and will be explained with reference to the embodiments described below. Attached Figure Description

[0021] In the accompanying drawings, similar reference numerals often refer to the same parts throughout the different views. Furthermore, the drawings are not necessarily drawn to scale, but rather typically focus on illustrating the principles of various embodiments.

[0022] Figure 1 It is a series of lung ultrasound imaging frames that present the severity of a deteriorating lung condition according to various aspects of this disclosure.

[0023] Figure 2 This is a block diagram illustrating an improved system configured to quantify B-lines and fused B-lines in lung ultrasound examinations according to various aspects of this disclosure.

[0024] Figure 3 This is a flowchart illustrating a method for quantifying B-lines and fused B-lines in lung ultrasound examinations according to various aspects of this disclosure.

[0025] Figure 4 This is a block diagram illustrating components of an electronic device used in conjunction with an ultrasound imaging apparatus according to various aspects of this disclosure.

[0026] Figure 5 This is a flowchart illustrating a method for analyzing various aspects of the present disclosure, circulating ultrasound videos of B-lines and fused B-lines, and visualizing the results.

[0027] Figure 6 This is a flowchart illustrating the process of analyzing ultrasound videos of B-lines and fused B-lines according to certain aspects of this disclosure and visualizing the results.

[0028] Figure 7A Two lung ultrasound image frames, before and after intensity normalization, are shown according to various aspects of this disclosure.

[0029] Figure 7B Two additional lung ultrasound image frames, before and after intensity normalization, are shown according to another aspect of this disclosure.

[0030] Figure 8 This is a flowchart illustrating a process for detecting and tracking pleural lines that can be used to determine regions of interest in B-line analysis, according to various aspects of this disclosure.

[0031] Figure 9 Two preprocessed lung ultrasound image frames are shown, annotated to illustrate the region of interest for B-line analysis, the corresponding intensity profile, and multiple B-line candidate regions of interest according to various aspects of this disclosure.

[0032] Figure 10 The image shows a preprocessed lung ultrasound frame with a region of interest for B-line analysis and two corresponding intensity profiles smoothed using two different smoothing kernels, according to various aspects of this disclosure.

[0033] Figure 11 Preprocessed lung ultrasound image frames annotated with two narrow B-line candidate regions of interest and two extended B-line candidate regions of interest are shown according to various aspects of this disclosure.

[0034] Figure 12 This is a flowchart illustrating the process of analyzing lung ultrasound imaging frames according to various aspects of this disclosure to detect and quantify B-lines and fuse B-lines using a single smooth kernel.

[0035] Figure 13This is a flowchart illustrating the process of analyzing lung ultrasound imaging frames according to various aspects of this disclosure to detect and quantify B-lines and fuse B-lines using two different smoothing kernels.

[0036] Figure 14 It is an illustration of a first graphical user interface that includes the results of B-line analysis according to various aspects of this disclosure.

[0037] Figure 15 It is an illustration of a second graphical user interface that includes the results of B-line analysis according to various aspects of this disclosure.

[0038] Figure 16 It is an illustration of a third graphical user interface that includes the results of B-line analysis according to various aspects of this disclosure. Detailed Implementation

[0039] According to this disclosure, systems and methods for improving the quantification of B-lines and fused B-lines in lung ultrasound examinations are provided. As described above, lung ultrasound is an imaging technique that can be used at the point of care to assess the lungs in a variety of settings, including emergency medicine and intensive care. This technique has been widely used as a portable, non-invasive, and radiation-free modality for evaluating lung and infectious diseases.

[0040] However, it may be difficult or impossible to detect certain pathological features seen in lung ultrasound quickly and accurately. For example, it may be difficult to detect B-lines and fused B-lines quickly and accurately. Therefore, although automated evaluation of these pathological features can play an important role in screening, diagnosing, and / or managing disease progression, conventional systems and methods are limited in their ability to detect and differentiate B-lines and fused B-lines, leading to inconsistencies in lung ultrasound interpretation among experts and underutilization of lung ultrasound by more novice users.

[0041] As described in this article, B-lines are pathological features visible in lung ultrasound images, known to be associated with fluid accumulation in the interstitial space of the lung. B-lines are discrete, vertically echogenic artifacts that appear as long bands originating from the pleural line and extending vertically along the length of the image in lung ultrasound images. However, as interstitial fluid accumulates horizontally and becomes more severe, more B-lines will be present, eventually leading to the fusion of individual B-lines.

[0042] For example, refer to Figure 1Four lung ultrasound images are shown along a continuum of increased pulmonary fluid associated with the severity of deterioration. In the leftmost ultrasound image (labeled "A"), there are no obvious B-lines in this part of the lung. In the second leftmost lung ultrasound image (labeled "B"), two distinct discrete B-lines are present in this part of the lung, indicating some accumulation of interstitial fluid. In the third leftmost lung ultrasound image (labeled "C"), four distinct discrete B-lines are present in this part of the lung, indicating worsening of interstitial fluid accumulation. And in the rightmost lung ultrasound image (labeled "D"), fused B-lines have formed as the accumulation of interstitial fluid in this part of the lung continues to worsen. It can be seen that it is important not only to identify discrete B-lines, but also to differentiate between discrete and fused B-lines in order to properly screen, diagnose, and / or manage the progression of a patient's disease. It is also important that users can quickly visualize these pathological features, otherwise lung ultrasound imaging loses its effectiveness, for example, in evaluating pulmonary and infectious diseases in emergency medicine and intensive care settings.

[0043] Therefore, refer to Figure 2 An improved system 100 for acquiring, analyzing, and viewing lung ultrasound images is provided according to certain aspects and embodiments of this disclosure. Preferably, the system 100 is configured to detect and differentiate B-lines and fused B-lines in real time. Therefore, the system 100 can be used to evaluate pulmonary and infectious diseases in emergency medical and intensive care settings.

[0044] In many embodiments, system 100 includes an ultrasound imaging device 102 configured to generate lung ultrasound data 104 of an object (e.g., a patient) 106. The ultrasound imaging device 102 may be a handheld ultrasound device including one or more ultrasound imaging transducers (not shown). In embodiments, lung ultrasound data 104 may be one or more lung ultrasound video loops of a specific region of the lung of object 106, and each video loop may include multiple lung ultrasound imaging frames.

[0045] System 100 also includes an electronic device 108 that communicates with ultrasound imaging equipment 102. In some embodiments, electronic device 108 may include, for example, but not limited to, a display device 110, a computer-readable storage medium having computer-readable instructions stored thereon to be executed by one or more processors (e.g., Figure 4 The memory 404 shown herein, and one or more processors configured by computer-readable instructions to perform one or more steps of the methods described herein (e.g., memory 404), and one or more processors configured by computer-readable instructions to perform one or more steps of the methods described herein. Figure 4 The processor 402 shown is shown.

[0046] In a particular embodiment, one or more processors may be configured by computer-readable instructions stored on a computer-readable storage medium to perform the following operations: (i) obtaining a lung ultrasound video loop 104 of object 106, the lung ultrasound video loop 104 including a plurality of lung ultrasound imaging frames; (ii) analyzing the lung ultrasound video loop 104 using a B-line classifier and a fusion B-line classifier to generate B-line data for the lung ultrasound video loop 104; and (iii) outputting a graphical user interface via a display device 110, the graphical user interface including the B-line data generated for the lung ultrasound video loop 104. In embodiments, the B-line data generated for the ultrasound video loop 104 may include, for example, discrete B-line counts for each imaging frame, a maximum B-line count for all imaging frames of the video loop 104, and / or a fusion B-line indicator indicating the presence of fused B-lines in the video loop 104.

[0047] More specifically, see reference Figure 3 One or more processors (e.g., Figure 4 The processor 402 shown can be stored in a computer-readable storage medium (e.g., Figure 4 The computer-readable instructions on the memory (404) shown are configured to perform the following operations: in step 310, a lung ultrasound video loop 104 is obtained; in step 320, the lung ultrasound video loop 104 is analyzed to generate B-line data for the video loop 104; in step 330, it is determined whether the video loop 104 contains any B-lines; optionally, in step 340, if the video loop 104 does not contain any B-lines, a B-line count of zero is output (e.g., via a user interface); in step 350, it is determined whether at least N frames of the video loop 104 contain fused B-lines; optionally, in step 360, if at least N frames of the video loop 104 do not contain fused B-lines, a maximum B-line count of the video loop 104 is output (e.g., via a user interface); and in step 370, if at least N frames of the video loop 104 contain fused B-lines, a fused B-line indicator is output (e.g., via a user interface).

[0048] As described herein, N can be considered a frame threshold for the video loop 104. In embodiments, N can be an integer greater than zero. In some embodiments, N can be a predetermined number independent of the size / number of frames in the video loop 104. In other embodiments, N can be a function of the size / number of frames in the video loop 104. For example, in some embodiments, if more than 25 frames in the 250-frame lung ultrasound video loop 104 contain fused B-lines, a fused B-line indicator can be output (i.e., in step 370). In other embodiments, for example, if more than 50% of the frames in the lung ultrasound video loop 104 contain fused B-lines, a fused B-line indicator can be output (i.e., in step 370).

[0049] like Figure 4 As illustrated in the example, electronic device 108 may include one or more processors 402 and computer-readable storage 404 interconnected and / or communicating via a system bus 406, which contains conductive circuit paths through which instructions (e.g., machine-readable signals) travel to enable communication, tasks, storage, etc. Electronic device 108 may be connected to a power source (not shown), which may include an internal power source and / or an external power source. In embodiments, electronic device 108 may also include one or more additional components, such as a display 110, an input device 112, an input / output (I / O) interface 412, a networking unit 414, etc., including combinations thereof. As shown, for example, each of these components may be interconnected and / or communicating via the system bus 406.

[0050] In embodiments, one or more processors 402 may include one or more high-speed data processors sufficient to perform one or more operations of the program components described herein and / or the methods described herein. One or more processors 402 may include microprocessors, multi-core processors, multi-threaded processors, ultra-low voltage processors, embedded processors, etc., including combinations thereof. One or more processors 402 may include multiple processor cores on a single die and / or may be part of a system-on-a-chip (SoC), wherein processor 402 and other components are formed as a single integrated circuit or a single package. That is, one or more processors 402 may be a single processor, multiple independent processors, or multiple processor cores on a single die.

[0051] In embodiments, display device 110 may be configured to display information, including text, graphics, etc. In a particular embodiment, display device 110 may be configured to include a graphical user interface for B-line data generated for one or more lung ultrasound video loops 104. Display device 110 may include, but is not limited to, liquid crystal display (LCD), light-emitting diode (LED) display, touch screen or other touch-enabled display, foldable display, projection display, etc., or combinations thereof.

[0052] In an embodiment, input device 112 may be configured to receive various forms of input from a user associated with electronic device 108. Input device 112 may include, but is not limited to, one or more of the following: keyboard, keypad, touchpad, trackball(s), capacitive keyboard, controller (e.g., game controller), computer mouse, computer stylus / pen, voice input device, etc., including combinations thereof.

[0053] In an embodiment, the input / output (I / O) interface 412 may be configured to connect to and / or enable communication with one or more peripheral devices (not shown), including but not limited to additional machine-readable storage devices, diagnostic equipment, and other attachable devices. The I / O interface 412 may include one or more I / O ports providing physical connections to the one or more peripheral devices. In some embodiments, the I / O interface 412 may include one or more serial ports.

[0054] In embodiments, networking unit 414 may include one or more types of networking interfaces that facilitate wired and / or wireless communication between electronic device 108 and one or more external devices. That is, networking unit 414 can operatively connect electronic device 108 to one or more types of communication networks 216, which may include directional interconnects, the Internet, local area networks (“LANs”), metropolitan area networks (“MANs”), wide area networks (“WANs”), wired or Ethernet connections, wireless connections, cellular networks, and similar types of communication networks, including combinations thereof. In some embodiments, electronic device 108 may communicate via communication network 416 with one or more remote / cloud-based servers and / or cloud-based services (such as remote server 418).

[0055] In embodiments, memory 404 may be embodied in one or more forms of machine-accessible and machine-readable memory. In some embodiments, memory 404 includes a storage device (not shown) that may include, but is not limited to, non-transient storage media, disk storage devices, optical disk storage devices, storage device arrays, solid-state memory devices, and combinations thereof. Memory 404 may also include one or more other types of memory, such as dynamic random access memory (DRAM), static random access memory (SRAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, and combinations thereof. In embodiments, memory 404 may include one or more types of transient and / or non-transient memory.

[0056] Electronic device 108 may be configured by software components stored in memory 404 to perform one or more processes of the methods described herein. More specifically, memory 404 may be configured to store data / information 420 and computer-readable instructions 422, which, when executed by one or more processors 402, cause electronic device 108 to: (i) obtain a lung ultrasound video loop 104 of object 106, the lung ultrasound video loop 104 including a plurality of lung ultrasound imaging frames; (ii) analyze the lung ultrasound video loop 104 using a B-line classifier and a fusion B-line classifier to generate B-line data for the lung ultrasound video loop 104; and (iii) output a graphical user interface via display device 110, the graphical user interface including the B-line data generated for the lung ultrasound video loop 104. Such data 420 and computer-readable instructions 422 stored in memory 404 can form a B-line analysis package 424, which can be incorporated into, loaded from, loaded onto, or otherwise operatively made available to and obtained from electronic device 108. Therefore, in some embodiments, the B-line analysis package 424 and / or one or more individual software packages can be stored in local storage of memory 404. However, in other embodiments, the B-line analysis package 424 and / or one or more individual software packages can be loaded onto and / or updated from a remote server or service (such as server 418) via communication network 416.

[0057] In a particular embodiment, the B-line analysis package 424 includes at least one B-line classifier (e.g., Figure 6 , 12 And the B-line classifier 610 shown in 13) and at least one fusion B-line classifier (e.g., Figure 6 , 12 (and the fusion B-line classifier 612 shown in Figure 13). Each of classifiers 610, 612 can be one or more trained models, such as one or more trained machine learning models. As described in more detail below, B-line classifier 610 and fusion B-line classifier 612 can be trained machine learning models configured to receive multiple B-line features (i.e., feature scores of multiple image features) associated with a single B-line candidate and generate the probability that the single B-line candidate is a possible B-line or a possible fusion B-line. In a particular embodiment, each classifier 610, 612 is a separately trained logistic regression model. That is, B-line classifier 610 can be a first trained model, and fusion B-line classifier 612 can be a second trained model different from the first trained model.

[0058] The electronic device 108 may also include an operating system component 426, which may be stored in the memory 404. The operating system component 426 may be an executable program that facilitates the operation of the electronic device 108 and / or the ultrasonic device 102. Typically, the operating system component 426 may facilitate access to the I / O interface 412, the network interface 414, the input device 112, and the display 110, and may communicate with or control other components of the electronic device 108.

[0059] Therefore, this document provides a computer program product 424 including a non-transient computer-readable storage medium 404 having computer-readable instructions 422 stored thereon, the computer-readable instructions 422 causing one or more processors (such as processor 402) to perform one or more operations of the methods described below when executed. For example, in a particular embodiment, the computer-readable storage medium 404 may include computer-readable instructions 422 that, when executed by one or more processors (such as processor 402), cause the one or more processors to perform the following operations: (i) obtain a lung ultrasound video loop 104 of object 106, the lung ultrasound video loop 104 including a plurality of lung ultrasound imaging frames; (ii) analyze the lung ultrasound video loop 104 using a B-line classifier and a fusion B-line classifier to generate B-line data for the lung ultrasound video loop 104; and (iii) output a graphical user interface via a display device 110, the graphical user interface including the B-line data generated for the lung ultrasound video loop 104.

[0060] In another embodiment, the computer-readable storage medium 404 may include computer-readable instructions 422, which, when executed by one or more processors (such as processor 402), cause the one or more processors to perform the following operations: (i) obtaining a lung ultrasound video loop of an object, the lung ultrasound video loop comprising a plurality of lung ultrasound imaging frames; (ii) preprocessing each imaging frame of the lung ultrasound video loop to obtain a preprocessed lung ultrasound video loop; (iii) determining a region of interest for B-line analysis for each imaging frame of the preprocessed lung ultrasound video loop; (iv) analyzing the region of interest for B-line analysis of each imaging frame to identify one or more B-line candidates; (v) for each B-line candidate, extracting a set of B-line features from the imaging frames of the preprocessed lung ultrasound video loop; (vi) classifying the B-line candidates based on the corresponding set of B-line features using a B-line classifier and a fused B-line classifier. Each B-line candidate is classified to predict the probability that the B-line candidate is a possible B-line and / or a possible fused B-line; (vii) one or more possible B-lines and / or possible fused B-lines in each imaging frame of the preprocessed ultrasound video loop are identified based on the classification of each B-line candidate; (viii) a maximum B-line count is determined for the lung ultrasound video loop, wherein the maximum B-line count is the maximum number of possible B-lines appearing in any single imaging frame of the preprocessed lung ultrasound video loop; (ix) whether the lung ultrasound video loop is positive for fused B-lines is determined based on the classification of each B-line candidate; and (x) a graphical user interface is output via a display device, the graphical user interface including the maximum B-line count and / or the determination of whether the lung ultrasound video loop is positive for fused B-lines.

[0061] That is, the computer-readable storage medium 404 may include computer-readable instructions 422, which, when executed by one or more processors (such as processor 402), cause the one or more processors to perform an improved method for detecting and distinguishing B-lines and fused B-lines in lung ultrasound images according to various aspects described herein.

[0062] For example, refer to Figure 5The present disclosure illustrates a method 500 for detecting and distinguishing B-lines and fused B-lines in lung ultrasound images. As shown, method 500 may include: in step 510, preprocessing each imaging frame of a lung ultrasound video loop to obtain a preprocessed lung ultrasound video loop; in step 520, determining a region of interest (ROI) for B-line analysis for each imaging frame of the preprocessed lung ultrasound video loop; in step 530, analyzing the B-line analysis ROI of each imaging frame to identify one or more B-line candidates; in step 540, extracting a set of B-line features for each B-line candidate from / based on the imaging frames of the preprocessed lung ultrasound video loop; and in step 550, using B-line classification. The classifier and the fusion B-line classifier classify each B-line candidate based on a corresponding set of B-line features to predict the probability that the B-line candidate is a possible B-line and / or a possible fusion B-line; in step 560, based on the classification of the B-line candidates, it is determined whether each B-line candidate is eligible as a possible B-line and / or a possible fusion B-line; in step 570, the B-lines detected at the video loop level are evaluated; in step 580, the fusion B-lines detected at the video loop level are evaluated; and in step 590, the frame-level and video-level results are output.

[0063] According to various aspects of this disclosure, each step in the methods 300, 500 described herein can be implemented in several ways, including as one or more sub-steps. For example, as Figure 6 As illustrated, one embodiment of method 500 is depicted according to certain aspects of this disclosure. As shown, an ultrasound video loop 602 comprising multiple lung ultrasound imaging frames is presented / obtained. The ultrasound video loop 602 may be obtained, for example, from an ultrasound imaging device 102 comprising one or more ultrasound imaging transducers. In embodiments, the ultrasound video loop 602 may include native (i.e., pre-scan converted) ultrasound line scan data to maintain greater visual consistency across different transducer types (e.g., sector, linear, and curvilinear) and allow for the use of a consistent set of image feature extraction steps across all transducers.

[0064] In this embodiment, each frame of the ultrasound video loop 602 may be preprocessed to improve the robustness of subsequent B-line detection steps, including across different transducer types. Preprocessing steps may include, but are not limited to, image normalization, noise normalization, TGC normalization, frame mixing, and combinations thereof.

[0065] More specifically, for example, the input frames of ultrasound video loop 602 can be normalized (i.e., rescaled) to a fixed intensity distribution before B-line analysis. In an embodiment, this can be performed on the image frames by first calculating the mean and variance of the image frames after excluding outlier pixels (i.e., pixels with intensity values ​​at low and high extremes) from the distribution estimate. The image frames can then be normalized to zero mean and unit variance. In some embodiments, truncation is performed at a certain number of... Pixels outside the standard deviation are then rescaled to a 0-255 intensity scale. Therefore, the effects of low and high transducer gain are significantly reduced, and images acquired using different transducer types can have similar intensity distributions. For example, as... Figure 7A and 7B As shown, images acquired with low and high transducer gains before and after image normalization are presented.

[0066] In another embodiment, the input frames of the ultrasound video loop 602 can also be preprocessed for noise normalization, as different transducer types will generate images with different noise and speckle characteristics, which may affect B-line detection and analysis. To normalize the noise level across transducer types, the noise distribution can be calculated from images acquired using one transducer type and used to denoise images collected using another transducer, resulting in a similar noise distribution. For example, denoising can be accomplished through Gaussian smoothing, median filtering, etc.

[0067] In yet another embodiment, the time gain compensation (TGC) profile of the input frames of the ultrasound video loop 602 can vary significantly across different transducer types, thus affecting B-line detection and analysis; the TGC profile can also be normalized. The TGC profile can be calculated based on images acquired from one transducer type and used to adjust the TGC on images from different transducer types.

[0068] In yet another embodiment, frame blending can be applied as an additional preprocessing step to reduce frame-to-frame "jitter" and improve the consistency of B-line analysis results. For example, frame blending can be achieved by: (1) averaging the pixel intensity of n consecutive frames by calculating the average pixel intensity over n consecutive frames, and then (2) projecting the maximum intensity across n consecutive frames by calculating the maximum intensity over n consecutive frames. In an embodiment, the value n can be set to three frames, which would capture the current frame plus two previous frames. However, other values ​​are expected.

[0069] After obtaining and optionally preprocessing the ultrasound video loop 602 as described herein, one or more regions of interest 604 for B-line analysis are determined for the ultrasound video loop 602. That is, a region of interest (ROI) 604 for B-line analysis is determined for each imaging frame of the (preprocessed) ultrasound video loop 602. In a particular embodiment, the B-line analysis ROI 604 may be determined based on the pleural line of the object 108 seen in the imaging frame of the ultrasound video loop 602.

[0070] As used herein, the term "pleural line" refers to the interface between the soft tissue (rich in fluid) of the lung wall and the gas-rich lung tissue. In lung ultrasound, the pleural line appears as a hyperechoic line and represents the junction of the visceral and parietal pleura. Because B-lines appear in lung ultrasound images as originating from the pleural line, the B-line analysis ROI 604 can be truncated after the pleural line is detected and tracked in frames through ultrasound video loop 602.

[0071] refer to Figure 8 According to certain aspects of this disclosure, a method for detecting and tracking a pleural line of an object 108 using imaging frames of an ultrasound video loop 602 is illustrated. As shown, the process begins in step 810 by identifying the presence and location of the pleural line from the first few frames (e.g., n frames) of the ultrasound video loop 602. In embodiments, pleural line detection is based on the analysis of at least two features, including but not limited to: (1) the intensity profile along the (vertical) depth direction within a region of the image frame that may contain the pleural line, and (2) the movement of the lung above and below possible pleural line candidates within that region of the frame. In particular, the region of the image frame most likely to contain the pleural line is typically the upper half of the image, at a depth of approximately 1 to 5 cm.

[0072] If the pleural line is successfully detected, the process then includes tracking the position of the pleural line across all remaining frames of the ultrasound video loop 602 in step 820. In an embodiment, the pleural line tracking algorithm may be designed to be similar to the pleural line detection algorithm, but operates on fewer frames, uses a smaller search range (e.g., region 815B instead of region 815A), and applies a reduced image sampling density to accelerate the processing time required per frame.

[0073] Then, in step 830, the B-line analysis ROI 604 can be defined and updated in each frame of the ultrasound video loop 602 based on the location of the tracked pleural line. In a particular embodiment, the B-line analysis ROI 604 can be set to start at a predefined distance below the detected pleural line and extend toward the bottom of the image frame to a fixed image depth. For example, in some embodiments, the predefined starting position of the B-line analysis ROI 604 may be 0.25 cm below the detected pleural line. In other embodiments, the fixed image depth may be transducer-dependent.

[0074] However, it should be understood that other methods can be implemented to define the B-line analysis ROI 604 for each image frame of the ultrasound video loop 602. For example, such as... Figure 8 As shown, if no pleural line is detected, a default B-line analysis ROI 604 can be defined based on the default ROI. In some embodiments, the default ROI can be based on the specifications of the ultrasound imaging probe 102 used to obtain the ultrasound video loop 602.

[0075] Once the B-line analysis ROI 604 has been determined for each frame of the ultrasound video loop 602, B-line and fused B-line analyses can proceed. For example, as Figure 6 As shown, the B-line analysis ROI of each image frame in the ultrasound video loop 602 can then be analyzed to identify one or more B-line candidates 606. In a particular embodiment, the B-line candidate 606 may be a discrete B-line candidate and / or a fused B-line candidate.

[0076] In an embodiment, one or more B-line candidates 606 for each frame of the ultrasound video loop 602 can be identified based on local “peaks” in the intensity profile calculated within the B-line analysis ROI 604 defined for the corresponding frame. For example, refer to Figure 9 An annotated example of image frame 900A of ultrasound video loop 602 is illustrated based on certain aspects of this operation. (See also...) Figure 9 As shown in the example, image frame 900A is annotated to illustrate the pleural line and B-line analysis ROI 604 defined for image frame 900A. Below image frame 900A, a representative intensity profile 910 of ROI 604 is illustrated, which can be calculated as the median of each column of ROI 604. Therefore, the intensity profile 910 of the B-line analysis ROI 604 can be generated for each image frame (e.g., frame 900A) of ultrasound video loop 602, and one or more B-line candidates 606 can be detected by identifying one or more local peaks in the intensity profile 910.

[0077] In an embodiment, a set of B-line features can be extracted from the imaging frames of the ultrasound video loop 602 for each of the identified B-line candidates 606, as discussed in more detail below. However, in a particular embodiment, each B-line candidate 606 may be associated with a region of interest (i.e., a B-line candidate ROI). Figure 9 As illustrated, for example, an annotated image frame 900B shows seven B-line candidate ROIs 912 associated with the seven B-line candidates 606 seen in image frame 900A. Thus, in this embodiment, B-line candidate ROIs can be computed as smaller local regions of interest centered on each B-line candidate 606 before extracting the features of each set. In other words, a B-line candidate ROI 912 can be defined for each of one or more B-line candidates 606 (i.e., each of the one or more identified local peaks), where each B-line candidate ROI 912 corresponds to a B-line candidate 606. B-line features can then be extracted from each B-line candidate ROI 912 for subsequent B-line and fused B-line classification.

[0078] In a particular embodiment, the height of each B-line candidate ROI 912 can be the length of the B-line analysis ROI 604, while the width of each B-line candidate ROI 912 can be a predefined number of columns (i.e., pixels) to the left and right of the identified local peak (e.g., three pixels to the left and three pixels to the right of the local maximum of the peak, etc.). However, it should be understood that the size of the B-line candidate ROI 912 can vary, as described in more detail below.

[0079] According to other aspects of this disclosure, the detection of B-line candidates 606 (including B-line candidates and fused B-line candidates) can be computed at different spatial scales, including but not limited to two or more different spatial scales. For example, in some embodiments, the step of detecting one or more B-line candidates 606 within the B-line intensity profile of an image frame of an ultrasound video loop 602 may include: (i) smoothing the intensity profile of the B-line analysis ROI 604 of the image frame at two or more spatial scales; (ii) identifying one or more local peaks along the smoothed intensity profile(s); and (iii) defining a B-line candidate ROI 912 for each identified local peak, wherein each B-line candidate ROI 912 corresponds to a B-line candidate 606. In embodiments, two or more smoothing kernels of different sizes may be used to apply smoothing at two or more spatial scales, such that a smaller smoothing kernel results in a smaller smoothing, while a larger smoothing kernel results in a larger smoothing.

[0080] For example, refer to Figure 10An example of an image frame 1000A having intensity profiles 1010A and 1010B processed at two different spatial scales is illustrated according to various aspects of this disclosure. As shown, by applying a first smoothing kernel (i.e., a small smoothing kernel) to intensity profile 1010A, four discrete B-line candidates 606 can be identified from the local peaks of the median intensity profile 1010A. However, by applying a second smoothing kernel (i.e., a larger smoothing kernel), a single / wide-blended B-line candidate is identified from the local peaks of the median intensity profile 1010B.

[0081] In embodiments, the second smoothing kernel can be at least about 50% larger than the first smoothing kernel, including at least about 75%, at least about 100%, at least about 150%, at least about 200%, and / or at least about 300%. In other words, the size ratio of the first smoothing kernel to the second smoothing kernel can be about 1:1.5, about 1:1.75, about 1:2, about 1:2.5, about 1:3, and / or about 1:4. Therefore, in certain embodiments, the size of the second smoothing kernel can be at least two to three times larger than the size of the first smoothing kernel.

[0082] As described herein, smoothing the intensity profile of the B-line candidate ROI 604 can involve an average-type filter (e.g., a moving average filter). However, it should be understood that other types of smoothing filters with different smoothing kernels can be used, including but not limited to Gaussian smoothing filters.

[0083] In a particular embodiment, the B-line candidate ROI 912 can also be defined at two or more different spatial scales. That is, in an embodiment, one or more B-line candidate ROIs 912 can be defined at a first scale and a second scale different from the first scale. For example, refer to Figure 11 Image frame 1100 showing two B-line candidates 606 is shown and annotated using B-line candidate ROIs 1110 and 1120. As shown, each B-line candidate 606 has a narrower B-line candidate ROI 1110 (indicated by a solid line) and an extended / wider B-line candidate ROI 1120 (indicated by a dashed line).

[0084] In a particular embodiment, the width of the extended B-line candidate ROI 1120 may be at least about 50% larger than the width of the narrower B-line candidate ROI 1100, including but not limited to at least about 100%, at least about 150%, and / or at least about 200% larger than the width of the narrower B-line candidate ROI 1100. In other words, in a particular embodiment, the width of the extended B-line candidate ROI 1120 may be two or three times the width of the narrower B-line candidate ROI 1110.

[0085] return Figure 6After detecting and calculating B-line candidates 606 and B-line candidate ROIs (e.g., ROIs 912, 1110, 1120), a set of B-line features 608 can be extracted from the imaging frames of the ultrasound video loop 602 for each of the identified B-line candidates 606. That is, for each B-line candidate identified in the ultrasound video loop 602, a set of B-line features 608 is extracted from the corresponding B-line candidate ROI. Therefore, in the embodiment, multiple sets of B-line features 608 corresponding to multiple B-line candidates 606 identified in the ultrasound video loop 602 can be generated.

[0086] In an embodiment, each set of B-line features may include one or more image features extracted from a local image region (i.e., the B-line candidate ROI) surrounding the B-line candidate 606. The image features may describe the visual appearance of the B-line candidate. In some embodiments, each image feature may be represented as one or more consecutive floating-point values ​​or feature scores. One or more of the image features may be applied at multiple spatial resolution scales (e.g., smaller and larger spatial scales). Therefore, in the case where image features are computed at two or more spatial scales, two or more feature scores are determined instead of a single feature score.

[0087] In certain embodiments, a set of B-line features may include a feature score of one or a combination of statistical image features, which may be statistical features of the ultrasound image. For example, in some embodiments, feature scores may be determined for multiple B-line features, which may include, but are not limited to, B-line extent, B-line intensity variance, temporal profile, peak local significance, peak amplitude, minimum / maximum B-line intensity, B-line homogeneity, and / or B-line width. In certain embodiments, a set of B-line features includes a feature score for at least the B-line extent for each of the B-line candidates. As described herein, "B-line extent" refers to the extent to which a B-line candidate extends vertically to the bottom of the ultrasound image. In certain embodiments, "B-line extent" may be calculated as a percentage of pixel values ​​along the center of a B-line candidate having an intensity exceeding the background intensity surrounding the candidate. However, it should be understood that other methods of determining the B-line extent may be implemented.

[0088] In embodiments, one or more of these features can be computed at different spatial resolution scales, and thus produce two or more feature scores for the same image features that can be passed to B-line classifier 610 and fused B-line classifier 612. For example, in a particular embodiment, at least peak local saliency image features and / or B-line width image features can be computed at two or more different spatial resolution scales. In another embodiment, one or more image features can produce two or more feature scores at the same spatial resolution scale. For example, at least B-line minimum / maximum intensity image features and B-line uniformity image features can produce two feature scores that are included in a set of B-line features 608 and passed to B-line and / or fused B-line classifiers 610, 612.

[0089] Although specific image features are described herein, it should be understood that in some embodiments, each group of B-line features 608 may include fewer image features, while in other embodiments, each group of B-line features 608 may include additional image features. Furthermore, it should be understood that the B-line features 608 extracted for some B-line candidates may differ from the B-line features 608 extracted for other B-line candidates. For example, one or more B-line features 608 may be specific to B-line candidates that may be fused B-lines, and / or vice versa.

[0090] After extracting a set of B-line features 608 from the ultrasound video loop 602 for each of the B-line candidates 606, these sets of B-line features 608, including one or more feature scores corresponding to one or more image features, are provided to classifiers 610, 612 for B-line and fused B-line classification. In embodiments, each of classifiers 610, 612 may be one or more trained models, such as one or more trained machine learning models. For example, in some embodiments, B-line classifier 610 may be a trained machine learning model configured to receive multiple B-line features (i.e., feature scores of multiple image features) associated with a single B-line candidate and generate the probability that the single B-line candidate is a possible B-line. In other embodiments, fused B-line classifier 612 may be a different trained machine learning model configured to receive multiple B-line features (i.e., feature scores of multiple image features) associated with a single B-line candidate and generate the probability that the single B-line candidate is a possible fused B-line.

[0091] Therefore, for example, for each identified B-line candidate 606, a set of B-line features 608 associated with that candidate can be presented as a floating-point vector ( This is provided to the first machine learning classifier 610 (i.e., the B-line classifier). The B-line classifier 610 can be configured to output a single floating-point predicted score between 0 and 1. (where a higher score indicates that the candidate is a higher predictor of the B line.)

[0092] In a particular embodiment, the B-line classifier 610 is a logistic regression model with multiple optimizable parameters that are tuned during training the model on a training dataset. Optimizable parameters may include, for example, but not limited to, feature weights that determine the relative contribution of each feature to the prediction. ) and final classification threshold ( For example, in some embodiments, the input vector ( A logistic regression model can have 14 parameters, and it can have 15 optimizable parameters, including 14 feature weights that determine the relative contribution of each feature to the prediction. ) and final classification threshold ( The predicted score meets or exceeds the classification threshold. 606 B-line candidates can be classified as possible B-line 614, while 606 B-line candidates whose scores do not meet the threshold can be classified as false candidates and ignored.

[0093] Optionally, the B-line classifier 610 may use more than one classification threshold. That is, in some embodiments, two classification thresholds may be used ( and ),in, A higher threshold ( This can be applied to the B-line candidate with the highest prediction score among all candidates in a given image frame, i.e., According to these embodiments, only when the initial conditions Only when true will all remaining B-lines within a given image frame be compared with the default threshold. The comparison is then performed. Therefore, negative videos (i.e., those with a maximum B-line count of zero) can be filtered out better (i.e., more specifically) from positive videos (i.e., those with a maximum count of at least 1).

[0094] Furthermore, for each identified B-line candidate 606, a set of B-line features 608 associated with that candidate can also be presented as a floating-point vector. The B-line features 608 provided to the second machine learning classifier 612 (i.e., the fused B-line classifier) ​​should be understood to include feature scores of the same image features provided to the B-line classifier 610. However, it is also anticipated that the B-line features 608 provided to the fused B-line classifier 612 may differ from the B-line features provided to the B-line classifier 610 (i.e., may include one or more different feature scores and / or feature scores of different image features). The fused B-line classifier 612 outputs a single floating-point predicted score between 0 and 1. (where a higher score indicates that the candidate is more likely to be predicted by the fusion of the B line.)

[0095] In a particular embodiment, the fused B-line classifier 612 is a logistic regression model with multiple optimizable parameters that are tuned during training on the training dataset. Optimizable parameters may include, for example, but not limited to, feature weights that determine the relative contribution of each feature to the prediction. ) and final classification threshold ( For example, in some embodiments, the input vector ( A logistic regression model can have 14 parameters, and it can have 15 optimizable parameters, including 14 feature weights that determine the relative contribution of each feature to the prediction. ) and final classification threshold ( In the embodiment, the predicted score meets or exceeds the classification threshold (). 606 B-line candidates can be classified as possible fused B-lines 616, while 606 B-line candidates whose scores do not meet the threshold can be classified as discrete B-lines (or false candidates) and ignored.

[0096] As described above, as part of the process of detecting and classifying one or more B-line candidates 606, one or more smoothing kernels can be applied to the intensity profile of the image frames in the ultrasound video loop 602. This is used to determine the criteria for counting as a true B-line or a true fused B-line.

[0097] For example, refer to Figure 12 The figure illustrates a process 1200 for detecting and classifying B-line candidates 1208 of a single image frame using a smoothing kernel. As shown, process 1200 includes: defining a B-line analysis ROI 1202 for the image frame; calculating an intensity profile 1204 within the B-line analysis ROI 1202; applying a smoothing kernel to the intensity profile 1204 to generate a smooth intensity profile 1206; detecting one or more B-line candidates 1208 based on the smooth intensity profile 1206; defining a B-line candidate ROI 1210 for each of the detected B-line candidates 1208; extracting a set of B-line features 1212 for each B-line candidate 1208 based on the corresponding B-line candidate ROI 1210; and independently passing these sets of B-line features 1212 to a B-line classifier 610 and a fused B-line classifier 612. Figure 12In the example, B-line classifier 610 will determine whether each of the B-line candidates 1208 is a possible B-line 614, while fusion B-line classifier 612 will determine whether each of the B-line candidates 1208 is a possible fusion B-line 616. After obtaining the possible B-lines 614 and the possible fusion B-lines 616, process 1200 includes checking each possible B-line 614 to determine whether the lower-level B-line candidates 1208 are also definitively classified as possible fusion B-lines 616. If so, the fusion B-line classification takes precedence, and the B-line candidate 1208 will be recorded as a fusion B-line instead of a B-line. Otherwise, the B-line classification can be recorded.

[0098] In an embodiment, the process can be repeated for each image frame of one or more ultrasound video loops 602. Figure 12 The procedure 1200 is shown. Once procedure 1200 is repeated for a specific ultrasound video loop 602, a video level assessment (such as...) can be performed. Figure 5 As shown and discussed in more detail below.

[0099] In other embodiments, as part of the process of detecting one or more B-line candidates 606, two or more smoothing kernels may be applied to the intensity profile of image frames in the ultrasound video loop 602. For example, refer to Figure 13 The diagram illustrates the process 1300 for detecting and classifying B-line candidates 1308 in a single image frame using two smoothing kernels. As shown in the figure, process 1300 includes: defining a B-line analysis ROI 1302 for an image frame; calculating an intensity profile 1304 within the B-line analysis ROI 1302; applying a first smoothing kernel to the intensity profile 1304 to generate a first smoothed intensity profile 1306; detecting one or more B-line candidates 1308 based on the first smoothed intensity profile 1306; defining a B-line candidate ROI 1310 for each of the detected one or more B-line candidates 1308; extracting a set of B-line features 1312 for each B-line candidate 1308 based on the corresponding B-line candidate ROI 1310; passing these sets of B-line features 1312 to a B-line classifier 610; applying a second smoothing kernel to the intensity profile 1304 to generate a second smoothed intensity profile 1316; detecting one or more B-line candidates 1318 based on the second smoothed intensity profile 1316; and defining a B-line candidate ROI for each of the detected one or more B-line candidates 1318. 1320; Based on the corresponding B-line candidate ROI, 1320 extracts a set of B-line features 1322 for each B-line candidate 1318; and passes these sets of B-line features 1322 to the fused B-line classifier 610. As mentioned above, the first smoothing kernel can be smaller than the second smoothing kernel.

[0100] exist Figure 13In the example, B-line classifier 610 will determine whether each of the B-line candidates 1308 is a possible B-line 614, while fusion B-line classifier 612 will determine whether each of the B-line candidates 1318 is a possible fused B-line 616. After obtaining the possible B-lines 614 and the possible fused B-lines 616, process 1200 includes, in step 1324, searching image frames to determine whether any of the detected B-lines 614 corresponds to (i.e., overlaps with) the detected fused B-line 616. If so, the fused B-line classification takes precedence, and the B-line candidate 1308 will be recorded as a fused B-line instead of a B-line. Otherwise, the B-line classification can be recorded.

[0101] In an embodiment, the process can be repeated for each image frame of one or more ultrasound video loops 602. Figure 13 The procedure 1300 is shown. Once procedure 1300 is repeated for a specific ultrasound video loop 602, a video level assessment (such as...) can be performed. Figure 5 As shown and discussed in more detail below.

[0102] In an embodiment, possible video level assessments of the B-line and fused B-line video level assessments can be performed for a given ultrasound video loop 602. For example, also refer to... Figure 5 The process 500 may include: in step 570, processing the ultrasound video loop 602 to determine one or more video level B-line parameters; and in step 580, processing the ultrasound video loop 602 to determine one or more video level fusion B-line parameters.

[0103] In a particular embodiment, step 570 may include calculating at least a first video level parameter, such as the “maximum B-line count” of video loop 602. The maximum B-line count may be calculated as the maximum number of discrete B-lines 614 appearing in any single frame of the video loop. In some embodiments, the maximum B-line count may be reported in the form of categories (i.e., output via electronics 108), such as: “0 B-lines”, “1-2 B-lines”, “3+ B-lines”, etc. Alternatively, a raw integer count may be reported (i.e., output).

[0104] In another embodiment, step 580 may include calculating at least a second video level parameter, such as determining whether video loop 602 is positive for a fused B-line. According to various aspects of this disclosure, whether video loop 602 is positive for a fused B-line may depend on one or more of the following: (i) the number of frames containing at least one fused B-line; (ii) the average or total number of fused B-lines detected throughout the video; (iii) the average or total width of all fused B-lines throughout the video; (iv) the average or total prediction confidence score of all fused B-lines throughout the video; and / or (v) any combination of the foregoing.

[0105] For example, in a particular embodiment, if the number of imaging frames containing at least one possible fused B-line meets or exceeds a predefined minimum number of frames (e.g., ... ,in, (Optimized for each transducer type), then for the fused B line, video loop 602 can be determined to be positive.

[0106] According to certain aspects of this disclosure, system 100 can be configured to report the maximum B-line count (or associated category) only if ultrasound video loop 602 is not affirmative for fusion B-line. If video loop 602 is affirmative, a separate category (“fusion(⊥)”) can be reported. Alternatively, in all cases, the maximum B-line count (or associated category) can be reported together with the “fusion” category.

[0107] A specific feature of this disclosure is that the user 114 of system 100 may be able to visualize lung ultrasound imaging data 104 for patient 106 more quickly and accurately, including real-time visualization. While the evaluation of B-lines and fusion B-lines is important in the screening, diagnosis, and management of disease progression and treatment, it should be understood that even for experienced users, it may be difficult or impossible to quickly and accurately detect certain pathological features seen in lung ultrasound. Therefore, the system and method disclosed herein not only improve the quantification of B-lines and fusion B-lines in lung ultrasound examinations but also provide a more consistent interpretation of lung ultrasound and facilitate utilization by more experienced users.

[0108] Therefore, according to various aspects of this disclosure, the systems and methods described herein may include generating a graphical user interface including B-line data generated for one or more lung ultrasound video cycles 104, 602, and displaying the graphical user interface on a display device 110. For example, as Figure 3 As shown in the example, process 300 may include: in step 340, if no image frame in the ultrasound video loops 104, 602 contains any B-line, then outputting a zero B-line count; in step 360, if the ultrasound video loops 104, 602 are not certain about the fused B-line, then outputting a maximum B-line count for video loops 104, 602; and in step 370, if the ultrasound video loops 104, 602 are certain about the fused B-line, then outputting a fused B-line indicator. These groups are as follows... Figure 5 As illustrated in the example, one or more of these steps 340, 360, 370 and / or one or more other steps may be outlined in step 590 of process 500, which includes outputting frame-level and video-level results. In embodiments, the output results (e.g., Figure 6The results shown in 618 may include the above-described frame-level and video-level assessments, as well as one or more representative image frames (without or without annotations).

[0109] In a particular embodiment, the graphical user interface may include one or more of the following: one or more image frames from ultrasound video loops 104, 602; video level output categories; lung area indicators; overlays of B-line and / or fused B-line indicators; overlays indicating B-line analysis ROIs; and so on, including combinations thereof. For example, refer to Figure 14 The first exemplary graphical user interface 1400 includes one or more image frames from ultrasound video loops 104, 602, video level output categories, lung area indicators, overlays of B-line and / or fused B-line indicators, and overlays indicating the ROI for B-line analysis.

[0110] In another embodiment, the graphical user interface may include a lung region overview of one or more ultrasound video loops 104, 602 acquired across all scanned lung regions. For example, refer to... Figure 15 A second exemplary graphical user interface 1500 is illustrated according to various aspects of this disclosure. As shown, the graphical user interface 1500 includes eight forward-facing lung regions (R1, R2, R3, R4, L1, L2, L3, and L4) and a visual overlay of the video level output category of each of these lung regions, and four backward-facing lung regions (R5, R6, L5, and L6) and a visual overlay of the video level output category of each of these lung regions.

[0111] In yet another embodiment, the graphical user interface may include multiple image frames from lung ultrasound video loops 104, 602, corresponding to ultrasound line scan data obtained using different transducers of the ultrasound imaging probe 102. For example... Figure 16 As shown in the example, the graphical user interface 1600 includes image frames from three individual transducers (labeled as transducers S4-1, L12-4, and C5-2) of the ultrasound probe 102, which are annotated to indicate B-line and fused B-line, B-line analysis ROI, and video level output categories.

[0112] It should be understood that all combinations of the foregoing concepts and the additional concepts discussed in more detail below (assuming such concepts do not contradict each other) are considered part of the inventive subject matter disclosed herein. In particular, all combinations of the claimed subject matter appearing at the end of this disclosure are considered part of the inventive subject matter disclosed herein. It should also be understood that terms expressly adopted herein that may also appear in any disclosure incorporated by reference should conform to the meaning most consistent with the specific concepts disclosed herein.

[0113] All definitions defined and used herein should be understood as control dictionary definitions, definitions in incorporated documents by reference, and / or the general meaning of the defined terms.

[0114] Unless explicitly indicated to the contrary, the words “a” and “an” as used herein in the specification and claims shall be understood to mean “at least one”.

[0115] The phrase “and / or” as used herein in the specification and claims should be understood to mean “any one or both” of the elements so combined, that is, elements that are combined in some cases and separate in others. Multiple elements listed using “and / or” should be interpreted in the same way, that is, “one or more” of the elements so combined. In addition to the elements specifically identified by the “and / or” clause, other elements may optionally be present, whether related to or unrelated to those specifically identified.

[0116] As used herein in the specification and claims, “or” should be understood to have the same meaning as “and / or” as defined above. For example, when separating items in a list, “or” or “and / or” should be interpreted as inclusive, i.e., including multiple elements or at least one element in the list of elements, but also including more than one, and optionally additional unlisted items. Only terms that explicitly indicate the opposite (such as “only one of…” or “exact one of…”, or “consisting of…” as used in the claims) will refer to multiple elements or exactly one element in the list of elements. Generally, when preceded by an exclusive term (such as “any,” “one of…,” “only one of…,” or “exact one of…”), the term “or” as used herein should be interpreted only as indicating an exclusive alternative (i.e., “one or the other but not both”).

[0117] As used herein in the specification and claims, the phrase "at least one" in relation to a list of one or more elements should be understood to mean at least one element selected from any one or more elements in the list of elements, but does not necessarily include at least one of each and every element specifically listed in the list of elements, and does not exclude any combination of elements in the list of elements. This definition also allows for the optional presence of elements other than those specifically identified in the list of elements referred to by the phrase "at least one," whether related to or unrelated to those specifically identified elements.

[0118] As used herein, although the terms first, second, third, etc., may be used to describe various elements or components, these elements or components should not be limited by these terms. These terms are used only to distinguish one element or component from another. Therefore, without departing from the teachings of the inventive concept, the first element or component discussed below may be referred to as the second element or component.

[0119] Unless otherwise stated, when an element or component is referred to as "connected to," "coupled to," or "proximity to" another element or component, it should be understood that the element or component may be directly connected to or coupled to the other element or component, or that there may be intermediate elements or components. That is, these and similar terms cover the situation where one or more intermediate elements or components may be used to connect two elements or components. However, when an element or component is referred to as "directly connected" to another element or component, this only covers the situation where two elements or components are connected to each other without any intermediate or intermediary elements or components.

[0120] In the claims and the foregoing description, all transitional phrases (such as "comprising," "including," "carrying," "having," "containing," "involving," "holding," "consisting of," etc.) shall be understood as open-ended, meaning including but not limited to. Only the transitional phrases "consisting of" and "substantially consisting of" shall be closed or semi-closed transitional phrases, respectively.

[0121] It should also be understood that, unless expressly indicated to the contrary, in any method claimed herein that includes more than one step or action, the order of the steps or actions of the method is not necessarily limited to the order in which the steps or actions of the method are described.

[0122] The examples of the described topics described above can be implemented in any of a variety of ways. For example, some aspects can be implemented using hardware, software, or a combination thereof. When any aspect is implemented at least partially in software, the software code can execute on any suitable processor or collection of processors, whether provided in a single device or computer or distributed across multiple devices / computers.

[0123] This disclosure can be implemented as a system, method, and / or computer program product at any possible level of technical detail integration. The computer program product may include a computer-readable storage medium (or media) having computer-readable program instructions thereon for causing a processor to execute aspects of this disclosure.

[0124] Computer-readable storage media can be tangible devices capable of retaining and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer-readable storage media includes the following: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable optical disc read-only memory (CD-ROM), digital versatile disc (DVD), memory sticks, floppy disks, mechanical encoding devices (such as punched cards or raised structures in grooves on which instructions are recorded), and any suitable combination of the foregoing. As used herein, computer-readable storage media should not be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0125] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a suitable computing / processing device, or downloaded via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network) to an external computer or external storage device. The network may include copper cables, optical fibers, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to a computer-readable storage medium within the suitable computing / processing device.

[0126] Computer-readable program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and procedural programming languages ​​(such as the "C" programming language or similar programming languages). The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer via any type of network (including a local area network (LAN) or wide area network (WAN)) or may make a connection to an external computer (e.g., via the Internet using an Internet service provider). In some examples, electronic circuits including, for example, programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs) may execute computer-readable program instructions to personalize the electronic circuits in order to perform aspects of this disclosure by utilizing the status information of the computer-readable program instructions.

[0127] This document describes aspects of the disclosure with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to examples of the disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0128] Computer-readable program instructions may be provided to a processor of a special-purpose computer or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create modules for implementing one or more functions / actions specified in flowcharts and / or block diagrams. These computer-readable program instructions may also be stored in a computer-readable storage medium that can instruct a computer, programmable data processing apparatus, and / or other device to operate in a particular manner, such that the computer-readable storage medium in which the instructions are stored includes an article of writing comprising instructions for implementing aspects of the functions / actions specified in the flowcharts and / or block diagrams or blocks.

[0129] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer-implemented process, and to cause the instructions to be executed on the computer, other programmable apparatus or other device to perform one or more functions / actions specified in a flowchart and / or block diagram box.

[0130] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various examples of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of instructions, including one or more executable instructions for implementing one or more specified logical functions. In some alternative implementations, the functions marked in the blocks may occur in a non-consecutive order. For example, two blocks shown consecutively may actually be executed substantially simultaneously, or these blocks may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified function or action or executes a combination of dedicated hardware and computer instructions.

[0131] Other embodiments are within the scope of the following claims and other claims that the applicant may enjoy.

[0132] Although several embodiments of the invention have been described and illustrated herein, those skilled in the art will readily conceive of various other modules and / or structures for performing functions and / or obtaining results and / or one or more of the advantages described herein, and each such variation and / or modification is considered to be within the scope of the embodiments of the invention described herein. More generally, those skilled in the art will readily understand that all parameters, dimensions, materials, and configurations described herein are intended to be exemplary, and actual parameters, dimensions, materials, and / or configurations will depend on one or more specific applications using the teachings of the invention. Those skilled in the art will recognize or be able to determine many equivalents of the specific embodiments of the invention described herein using experimental means no more than conventional. Therefore, it should be understood that the foregoing embodiments are presented by way of example only, and that embodiments of the invention may be practiced in ways different from those specifically described and claimed within the scope of the claims and their equivalents. The embodiments of the invention disclosed herein relate to each individual feature, system, article, material, kit, and / or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and / or methods is included within the scope of the invention disclosed herein, provided that such features, systems, articles, materials, kits, and / or methods are not contradictory.

Claims

1. A system (100) for viewing and analyzing lung ultrasound images, the system comprising: An ultrasound imaging device (102) comprising one or more ultrasound imaging transducers, wherein the ultrasound imaging device (102) is configured to generate a lung ultrasound video loop (104) of an object (106), the lung ultrasound video loop (104, 602) comprising a plurality of lung ultrasound imaging frames; and An electronic device (108) communicating with the ultrasound imaging device (102), wherein the electronic device (108) includes: Display device (110), configured to display a graphical user interface; A computer-readable storage medium (404) having stored thereon computer-readable instructions (422) to be executed by one or more processors; and One or more processors (402) configured by computer-readable instructions (422) stored on a computer-readable storage medium (404) to perform the following operations: (i) obtaining a lung ultrasound video loop of an object, the lung ultrasound video loop comprising a plurality of lung ultrasound imaging frames; (ii) analyzing the lung ultrasound video loop using a B-line classifier (610) and a fusion B-line classifier (612) to generate B-line data for the lung ultrasound video loop; and (iii) outputting a graphical user interface via the display device, the graphical user interface including the B-line data generated for the lung ultrasound video loop.

2. The system (100) according to claim 1, wherein, The B-line data for the lung ultrasound video loops (104, 602) were generated through the following operations: Each imaging frame of the lung ultrasound video loop (104) is preprocessed to obtain a preprocessed lung ultrasound video loop (602); For each imaging frame of the preprocessed lung ultrasound video loop (602), the region of interest for B-line analysis is determined; The region of interest for each B-line analysis in each imaging frame is analyzed to identify one or more B-line candidates; For each B-line candidate, a set of B-line features are extracted from the imaging frames of the preprocessed lung ultrasound video loop (602); The B-line classifier (610) and the fusion B-line classifier (612) are used to classify each B-line candidate based on a corresponding set of B-line features to predict the probability that the B-line candidate is a possible B-line and / or a possible fusion B-line. Based on the classification of each of the B-line candidates, one or more possible B-lines and / or possible fused B-lines are identified in each imaging frame of the preprocessed ultrasound video loop (602); Determine the maximum B-line count for the lung ultrasound video loop (104), wherein the maximum B-line count is the maximum number of possible B-lines appearing in any single imaging frame of the preprocessed lung ultrasound video loop (602); and The lung ultrasound video loop (104) is determined to be positive for fusion B-line based on the classification of each of the B-line candidates.

3. The system (100) according to claim 2, wherein, The B-line classifier (610) is a first trained machine learning model configured to receive multiple B-line features as input and output the probability that a B-line candidate is a possible B-line, and wherein the fusion B-line classifier (612) is a second trained machine learning model configured to receive multiple B-line features as input and output the probability that a B-line candidate is a possible fusion B-line.

4. The system (100) according to claim 2, wherein, If the number of imaging frames containing possible fused B-lines in the preprocessed lung ultrasound video loop (602) meets or exceeds a predefined minimum number of imaging frames, then the lung ultrasound video loop (104) is positive for fused B-lines.

5. The system (100) according to claim 2, wherein, Identifying one or more B-line candidates within the region of interest for the B-line analysis includes performing the following operations for each imaging frame of the preprocessed lung ultrasound imaging video loop (602): The intensity profile of the region of interest for the B-line analysis of the corresponding imaging frame is smoothed. Identify one or more local peaks along the smooth intensity profile of the region of interest analyzed along the B-line for the corresponding imaging frame; as well as A region of interest for a B-line candidate is defined for each of the one or more identified local peaks, wherein each region of interest for a B-line candidate corresponds to a B-line candidate.

6. The system (100) according to claim 5, wherein, The set of B-line features for each B-line candidate is extracted from the region of interest of the B-line candidate defined in the imaging frames of the preprocessed lung ultrasound video loop.

7. The system (100) according to claim 2, wherein, Identifying one or more B-line candidates within the region of interest for the B-line analysis includes performing the following operations for each imaging frame of the preprocessed lung ultrasound imaging video loop (602): The intensity profile of the region of interest for the B-line analysis of the corresponding imaging frame is smoothed using a first smoothing kernel. Identify one or more local peaks along the intensity profile smoothed using the first smoothing kernel; A region of interest for a B-line candidate is defined for each of the one or more local peaks identified in the intensity profile smoothed by the first smoothing kernel, wherein each region of interest for a B-line candidate corresponds to a B-line candidate. The intensity profile of the region of interest for the B-line analysis of the corresponding imaging frame is smoothed using a second smoothing kernel, wherein the size of the second smoothing kernel is different from that of the first smoothing kernel; Identify one or more local peaks along the intensity profile smoothed using the second smoothing kernel; and A region of interest for a B-line candidate is defined for each of the one or more local peaks identified in the intensity profile smoothed using the second smoothing kernel, wherein each region of interest for a B-line candidate corresponds to an additional B-line candidate.

8. The system (100) according to claim 7, wherein, Extract a first set of B-line features for each B-line candidate from the region of interest of the B-line candidates defined based on the intensity profile smoothed using the first smoothing kernel, and extract a second set of B-line features for each B-line candidate from the region of interest of the B-line candidates defined based on the intensity profile smoothed using the second smoothing kernel.

9. The system (100) according to claim 2, wherein, The set of B-line features extracted from the imaging frames of the preprocessed lung ultrasound video loop (602) includes at least one B-line feature measured at two or more different spatial scales.

10. An image processing method (500), comprising: Each imaging frame of the lung ultrasound video loop of the subject is preprocessed (510) to obtain a preprocessed lung ultrasound video loop, wherein the lung ultrasound video loop includes multiple imaging frames; (520) Determine the region of interest for B-line analysis of each imaging frame of the preprocessed lung ultrasound video loop; (530) Analyze the region of interest of the B-line analysis in each imaging frame to identify one or more B-line candidates; For each B-line candidate, a set of B-line features is extracted (540) from the imaging frames of the preprocessed lung ultrasound video loop; The B-line classifier and the fusion B-line classifier are used to classify each of the B-line candidates based on a corresponding set of B-line features (550) to predict the probability that the B-line candidate is a possible B-line and / or a possible fusion B-line. Based on the classification of each of the B-line candidates, identify (560) one or more possible B-lines and / or possible fused B-lines in each imaging frame of the preprocessed ultrasound video loop; Determine (570) the maximum B-line count for the lung ultrasound video loop, wherein the maximum B-line count is the maximum number of possible B-lines appearing in any single imaging frame of the preprocessed lung ultrasound video loop; and The lung ultrasound video cycle is determined (580) as positive for fusion B-line based on the classification of each of the B-line candidates.

11. The image processing method (500) according to claim 10, wherein, The B-line classifier (610) is a first trained machine learning model configured to receive multiple B-line features as input and output the probability that a B-line candidate is a possible B-line, and wherein the fusion B-line classifier (612) is a second trained machine learning model configured to receive multiple B-line features as input and output the probability that a B-line candidate is a possible fusion B-line.

12. The image processing method (500) according to claim 10, wherein, The lung ultrasound video loop is considered positive for a fused B-line if the number of imaging frames containing possible fused B-lines in the preprocessed lung ultrasound video loop meets or exceeds a predefined minimum number of imaging frames.

13. The image processing method (500) according to claim 10, wherein, Identifying (530) one or more B-line candidates within the region of interest for the B-line analysis includes performing the following operations for each imaging frame of the preprocessed lung ultrasound imaging video loop: The intensity profile of the region of interest for the B-line analysis of the corresponding imaging frame is smoothed. Identify one or more local peaks along the smooth intensity profile of the region of interest analyzed along the B-line for the corresponding imaging frame; as well as A region of interest for a B-line candidate is defined for each of the one or more identified local peaks, wherein each region of interest for a B-line candidate corresponds to a B-line candidate.

14. The image processing method (500) according to claim 10, wherein, Identifying (530) one or more B-line candidates within the region of interest for the B-line analysis includes performing the following operations for each imaging frame of the preprocessed lung ultrasound imaging video loop: The intensity profile of the region of interest for the B-line analysis of the corresponding imaging frame is smoothed using a first smoothing kernel. Identify one or more local peaks along the intensity profile smoothed using the first smoothing kernel; A region of interest for a B-line candidate is defined for each of the one or more local peaks identified in the intensity profile smoothed by the first smoothing kernel, wherein each region of interest for a B-line candidate corresponds to a B-line candidate. The intensity profile of the region of interest for the B-line analysis of the corresponding imaging frame is smoothed using a second smoothing kernel, wherein the size of the second smoothing kernel is different from that of the first smoothing kernel; Identify one or more local peaks along the intensity profile smoothed using the second smoothing kernel; and A region of interest for a B-line candidate is defined for each of the one or more local peaks identified in the intensity profile smoothed using the second smoothing kernel, wherein each region of interest for a B-line candidate corresponds to an additional B-line candidate.

15. A computer program product (424), comprising: A non-transient computer-readable storage medium (404) storing computer-readable instructions (422) that, when executed by one or more processors, cause the one or more processors to perform the following operations: (i) obtaining a lung ultrasound video loop of an object, the lung ultrasound video loop comprising a plurality of lung ultrasound imaging frames; (ii) preprocessing each imaging frame of the lung ultrasound video loop to obtain a preprocessed lung ultrasound video loop; (iii) determining a region of interest for B-line analysis for each imaging frame of the preprocessed lung ultrasound video loop; and (iv) analyzing the region of interest for B-line analysis of each imaging frame to identify one or more B-line candidates. (v) For each B-line candidate, a set of B-line features is extracted from the imaging frames of the preprocessed lung ultrasound video loop; (vi) Each B-line candidate is classified using a B-line classifier and a fusion B-line classifier based on the corresponding set of B-line features to predict the probability that the B-line candidate is a possible B-line and / or a possible fusion B-line. (vii) Identifying one or more possible B-lines and / or possible fused B-lines in each imaging frame of the preprocessed ultrasound video loop based on the classification of each of the B-line candidates; (viii) Determining the maximum B-line count for the lung ultrasound video loop, wherein the maximum B-line count is the maximum number of possible B-lines appearing in any single imaging frame of the preprocessed lung ultrasound video loop; (ix) Determining whether the lung ultrasound video loop is positive for fused B-lines based on the classification of each of the B-line candidates; and (x) Outputting a graphical user interface via a display device, the graphical user interface including the maximum B-line count and / or the determination of whether the lung ultrasound video loop is positive for fused B-lines.

Images