Smoke inhalation risk detection

Abstract

A system for classifying smoke inhalation lung injury includes a memory that stores instructions; and a processor that executes the instructions. When executed by the processor, the instructions cause the system to: obtain a set of lung abnormalities for a classification model; apply the classification model to the set of lung abnormalities for generating a risk score for severity of smoke inhalation lung injury; generate, by the classification model, the risk score for severity of smoke inhalation lung injury.

Description

Docket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT -SMOKE INHALATION RISK DETECTIONGOVERNMENT INTEREST

[0001] This invention was made with United States government support awarded by the United States Department of Health and Human Services under the grant number HHS / ASPR / BARDA 75A50120C00097. The United States has certain rights in this invention.BACKGROUND

[0002] Lung ultrasound (LUS) is an imaging technique deployed at the point-of-care to aid in lung evaluation in cases of infectious disease, trauma, cardiac disease, and other medical conditions. Important clinical features - such as B-lines, merged B-lines, pleural line changes, consolidations, and pleural effusions can be visualized using lung ultrasound. However, the ability to accurately identify these clinical features can be challenging to learn, and the ability itself involves reviewing of multiple cineloops (also referred to as lung ultrasound “videos”), each comprising about 60 to 200 image frames, acquired throughout the lung.

[0003] Smoke inhalation lung injury (SILI) can occur after exposure to smoke in a variety of situations such as building fires, forest fires and mass casualty events. The standard of care for assessing subjects with suspected smoke inhalation lung injury in the emergency room or burn unit are chest Xray and fiber-optic bronchoscopy (FOB), both of which have poor accuracy in identifying smoke inhalation lung injury. The chest Xray and fiber-optic bronchoscopy modalities with poor accuracy are used to guide treatment decisions such as intubation, and to predict clinical course and patient outcomes. Thus, the current standard-of-care modalities for assessing subjects suspected of smoke inhalation lung injury lack sensitivity and specificity, leading to high rates of complications and fatalities among victims of smoke inhalation. Abnormalities visible in lung ultrasound may be a better predictor of patient status and clinical course, and could help with treatment decisions, but healthcare providers for smoke inhalation victims typically lack training in lung ultrasound interpretation. In addition, the presentation of lung ultrasound abnormalities can be complex and may require interpretation across up to 14 - 16 lung ultrasound cineloops that are acquired as part of a comprehensive lung ultrasound exam. Treatment decisions for victims of smoke inhalation are often time-critical and need to be madeDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - at the point of care rather than awaiting results from consultative ultrasound.

[0004] Solutions to these types of challenged such as those described herein are applicable to smoke inhalation risk detection, but are also applicable to outputs from other types of imaging modalities and to assessments of other types of injuries.SUMMARY

[0005] According to an aspect of the present disclosure, a system for classifying lung injury or disease includes a memory that stores instructions; and a processor that executes the instructions. When executed by the processor, the instructions cause the system to: obtain a set of lung abnormalities for a classification model; apply the classification model to the set of lung abnormalities for generating a risk score for severity of smoke inhalation lung injury; generate, by the classification model, the risk score for severity of smoke inhalation lung injury.

[0006] According to another aspect of the present disclosure, a method for classifying lung injury or disease includes obtaining, by a system comprising a memory that stores instructions and a processor that executes the instructions, a set of abnormalities from a plurality of ultrasound cineloops for a classification model. Each of the plurality of ultrasound cineloops includes a plurality of ultrasound frames. The method also includes applying the classification model to the set of abnormalities for generating a risk score for severity of an injury indicated by the set of abnormalities; and generating, by the classification model, the risk score for the severity of injury indicated by the set of abnormalities.

[0007] According to another aspect of the present disclosure, a system for classifying smoke inhalation injury includes a memory that stores instructions; and a processor that executes the instructions. When executed by the processor, the instructions cause the system to: obtain a set of abnormalities for a classification model; apply the classification model to the set of abnormalities for generating a risk score for severity of at least one of heart disease, or respiratory disease from smoke inhalation; and generate, by the classification model, the risk score for severity of the at least one of heart disease, or respiratory disease from smoke inhalation.BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The example embodiments are best understood from the following detailed descriptionDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - when read with the accompanying drawing figures. It is emphasized that the various features are not necessarily drawn to scale. In fact, the dimensions may be arbitrarily increased or decreased for clarity of discussion. Wherever applicable and practical, like reference numerals refer to like elements.

[0009] FIG. 1 illustrates a system for smoke inhalation risk detection, in accordance with a representative embodiment.

[0010] FIG. 2 illustrates another system for smoke inhalation risk detection, in accordance with a representative embodiment.

[0011] FIG. 3 illustrates a method for smoke inhalation risk detection, in accordance with a representative embodiment.

[0012] FIG. 4A illustrates a display of risk scores for abnormality findings in smoke inhalation risk detection, in accordance with a representative embodiment.

[0013] FIG. 4B illustrates another display of risk scores for abnormality findings in smoke inhalation risk detection, in accordance with a representative embodiment.

[0014] FIG. 5 illustrates risk scores for each of a pair of lungs in smoke inhalation risk detection, in accordance with a representative embodiment.

[0015] FIG. 6 illustrates another method for smoke inhalation risk detection, in accordance with a representative embodiment.

[0016] FIG. 7 illustrates a computer system, on which a method for smoke inhalation risk detection is implemented, in accordance with another representative embodiment.DETAILED DESCRIPTION

[0017] In the following detailed description, for the purposes of explanation and not limitation, representative embodiments disclosing specific details are set forth in order to provide a thorough understanding of embodiments according to the present teachings. However, other embodiments consistent with the present disclosure that depart from specific details disclosed herein remain within the scope of the appended claims. Descriptions of known systems, devices, materials, methods of operation and methods of manufacture may be omitted so as to avoid obscuring the description of the representative embodiments. Nonetheless, systems, devices, materials and methods that are within the purview of one of ordinary skill in the art are withinDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - the scope of the present teachings and may be used in accordance with the representative embodiments. It is to be understood that the terminology used herein is for purposes of describing particular embodiments only and is not intended to be limiting. Definitions and explanations for terms herein are in addition to the technical and scientific meanings of the terms as commonly understood and accepted in the technical field of the present teachings.

[0018] It will be understood that, although the terms first, second, third etc. may be used herein to describe various elements or components, these elements or components should not be limited by these terms. These terms are only used to distinguish one element or component from another element or component. Thus, a first element or component discussed below could be termed a second element or component without departing from the teachings of the inventive concept.

[0019] As used in the specification and appended claims, the singular forms of terms ‘a,’ ‘an’ and ‘the’ are intended to include both singular and plural forms, unless the context clearly dictates otherwise. Additionally, the terms "comprises", and / or "comprising," and / or similar terms when used in this specification, specify the presence of stated features, elements, and / or components, but do not preclude the presence or addition of one or more other features, elements, components, and / or groups thereof. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0020] Unless otherwise noted, when an element or component is said to be “connected to,” “coupled to,” or “adjacent to” another element or component, it will be understood that the element or component can be directly connected or coupled to the other element or component, or intervening elements or components may be present. That is, these and similar terms encompass cases where one or more intermediate elements or components may be employed to connect two elements or components. However, when an element or component is said to be “directly connected” to another element or component, this encompasses only cases where the two elements or components are connected to each other without any intermediate or intervening elements or components.

[0021] The present disclosure, through one or more of its various aspects, embodiments and / or specific features or sub-components, is thus intended to bring out one or more of the advantages as specifically noted below.

[0022] As set forth herein, point-of-care lung ultrasound can be applied to detect smokeDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - inhalation lung injury and other types of injuries. Smole inhalation lung injury can lead to several types of lung abnormalities visible in lung ultrasound. With the teachings herein, lung ultrasound abnormalities which are detectable in lung ultrasound obtained after smoke inhalation may correlate with the presence or absence of smoke inhalation lung injury. A method for risk stratification in patients suspected of smoke inhalation lung injury may be implemented by extracting and quantifying information about lung ultrasound abnormalities after smoke inhalation. Artificial intelligence (Al) may be applied to detect the ultrasound abnormalities and may be used to generate a risk score that can help make treatment decisions for patients with smoke inhalation lung injury and other pulmonary conditions. Output information such as risk scores can be summarized and used by healthcare providers to make monitoring and treatment decisions, including intubation, to provide optimal care for smoke inhalation lung injury victims. Notwithstanding the focus of teachings herein on ultrasound and smoke inhalation, the underlying technical solutions described herein are not particularly limited to ultrasound as an imaging modality or to smoke inhalation lung injury as an injury type.

[0023] FIG. 1 illustrates a system 100 for smoke inhalation risk detection, in accordance with a representative embodiment.

[0024] The system 100 in FIG. 1 includes an ultrasound probe 110, an ultrasound base 120, and a display 180.

[0025] The ultrasound probe 110 includes a transducer array 113 and a processing circuit 115. The transducer array 113 includes at least a first transducer element 1131, a second transducer element 1132, and an Xth transducer element 113X. The transducer array 113 converts electrical energy into sound waves which reflect off of body tissue and receives echoes of the sound waves and converts the echoes into electrical energy. The transducer array 113 may include dozens, hundreds, or thousands of individual transducer elements. The ultrasound probe 110 may transmit a beam to produce images and may detect echoes. The processing circuit 115 may comprise a memory / processor combination as in FIG. 2 (explained below), and / or an application-specific integrated circuit (ASIC) comprising hardware elements. The processing circuit 115 may process ultrasound images captured by the transducer array 113 of the ultrasound probe 110.

[0026] The ultrasound base 120 includes a first interface 121, a second interface 122, a thirdDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - interface 123, and a controller 150. A computer that can be used to implement the ultrasound base 120 is depicted in FIG. 7, though an ultrasound base 120 may include more elements than depicted in FIG. 1 and more or fewer elements than depicted in FIG. 7. One or more of the interfaces may include ports, disk drives, wireless antennas, or other types of receiver circuitry that connect the controller 150 to other electronic elements. The first interface 121 connects the ultrasound base 120 to the ultrasound probe 110, and may comprise a port, an antenna, and / or another type of physical component for wired or wireless communications. The second interface 122 connects the ultrasound base 120 to the display 180, and may also comprise a port, an antenna, and / or another type of physical component for wired or wireless communications. The third interface 123 is a user interface, and may comprise buttons, keys, a mouse, a microphone, a speaker, switches, a touchscreen, or other type of display separate from the display 180, and / or other types of physical components that allow medical personnel to interact with the ultrasound base 120 such as to enter instructions and receive output.

[0027] Display 180 may be local to the ultrasound base 120 or may be remotely connected to the ultrasound base 120, such as wirelessly. Display 180 may be a monitor such as a computer monitor, a display on a mobile device, an augmented reality display, a television, an electronic whiteboard, or another screen configured to display electronic imagery. Display 180 includes a graphical user interface 181 (GUI) that displays ultrasound images and guidance to users.Display 180 may be interfaced with other user input devices by which medical personnel can input instructions, including mouses, keyboards, thumbwheels and so on. Display 180 may also include one or more input interface(s) such as those noted above that may connect to other elements or components, as well as an interactive touch screen configured to display prompts to medical personnel and collect touch input from medical personnel.

[0028] Controller 150 includes at least a memory 151 that stores instructions and a processor 152 that executes the instructions. The memory 151 may be representative of multiple memories such as random-access memory (RAM), registers, flash memory, and other types of memory. Memory 151 may store one or more software program(s). Controller 150 may perform some of the operations described herein directly and may implement other operations described herein indirectly. For example, controller 150 may indirectly control operations such as by generating and transmitting content to be displayed on the display 180. The controller 150 may directlyDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - control other operations such as logical operations performed by the processor 152 executing instructions from the memory 151 based on input received from electronic elements and / or medical personnel via the third interface 123. Accordingly, the processes implemented by the controller 150 when the processor 152 executes instructions from the memory 151 may include steps not directly performed by the controller 150.

[0029] Using the system 100, a user may initiate ultrasound imaging using the ultrasound probe 110. The ultrasound probe 110 may comprise a one-dimensional array probe or a two- dimensional matrix array probe. The ultrasound probe 110 is used to generate ultrasound images. In FIG. 1, the controller 150 may be configured to process a set of ultrasound images. The controller 150 may obtain a set of lung abnormalities for a classification model from the set of ultrasound images. The controller 150 may apply the classification model to the set of lung abnormalities for generating a risk score for severity of smoke inhalation lung injury. The risk score may be displayed on the display 180. Although not shown in FIG. 1, a separate controller, such as in a separate mobile device or controller, may be configured to process the set of ultrasound images, obtain the set of lung abnormalities for a classification model from the set of ultrasound images, apply the classification model to the set of lung abnormalities for generating a risk score for severity of smoke inhalation lung injury. In some embodiments, the controller 150 or a separate controller, may be configured to apply a classification model to the set of abnormalities for generating a risk score for severity of at least one of heart disease, or respiratory disease from smoke inhalation, and the classification model by generating the risk score for severity of the at least one of heart disease, or respiratory disease from smoke inhalation. In other words, the classification model may be configured to assess and generate risk scores for heart disease, respiratory disease, or the lung injury caused by smoke inhalation as initially described. Moreover, in some embodiments, the imaging modality may be other than the ultrasound modality in FIG. 1.

[0030] In some embodiments, functions attributed directly to the controller 150 may be performed at a data center in a cloud environment, such as by the controller 150 providing selected content from the controller 150 over the internet to a data center in the cloud environment. In other embodiments, functions attributed directly to the controller 150 may be performed by and at an external computer, including an external system remote from theDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - controller 150 and connected over a local or wide area network to the controller 150.

[0031] FIG. 2 illustrates another system for smoke inhalation risk detection, in accordance with a representative embodiment.

[0032] The system 200 in FIG. 2 is a system for smoke inhalation risk detection and includes components that may be provided together or that may be distributed. System 200 includes an ultrasound probe 210, a network 201, and smartphone A and smartphone B. A computer that can be used to implement smartphone A and / or smartphone B is depicted in FIG. 7, though smartphone A and / or smartphone B may include more elements than depicted in FIG. 2 and more or fewer elements than depicted in FIG. 7.

[0033] Network 201 may comprise a local wireless network such as a WiFi network, though the network 201 may also or alternatively include wired elements such as USB cables or other wires connected to smartphone A and smartphone B. Smartphone A and smartphone B are representative of mobile smart devices such as smartphones and tablets or other networked and / or networkable devices with logical processing capabilities. Additionally, smartphone A and smartphone B are used as examples to show that system 200 may include multiple different smart devices with applications or other functional capabilities that can be functionally integrated with ultrasound probes in overall systems for ultrasound imaging even though they are not necessarily dedicated only to the ultrasound imaging.

[0034] The ultrasound probe 210 may comprise a portable transducer. The ultrasound probe 210 includes a transducer array 213, a lens 214, a user interface 223, a controller 250, and a wireless communication circuit 290. The controller 250 includes a memory 251 and a processor 252. The ultrasound probe 210 may comprise, for example, a TEE ultrasound probe or a TTE ultrasound probe. The memory 251 may be representative of multiple memories such as random-access memory (RAM), registers, flash memory, and / or other types of memory. The memory 251 stores data and instructions. The processor 252 processes the data and instructions. The transducer array 213 includes an array of transducer elements including at least a first transducer element 2131, a second transducer element 2132, and an Xth transducer element 213X. The transducer array 213 converts electrical energy into sound waves which bounce off of body tissue and receives echoes of the sound waves and converts the echoes into electrical energy. The transducer array 213 may include dozens, hundreds, or thousands of individual transducerDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - elements. The ultrasound probe 210 may transmit a beam to produce images and may detect echoes. The processor 252 may process ultrasound images captured by the transducer array 213 of the ultrasound probe 210. The wireless communication circuit 290 may be used to communicate with smartphone A and smartphone B via the network 201. The ultrasound probe 210 may be configured to link to an external device such as the smartphone A and smartphone B via applications installed on the external device(s). The lens 214 may be an acoustic focusing lens for transmitting the ultrasound beams and receiving echoes of the ultrasound beams. User interface 223 may be used by medical personnel to interact with the ultrasound probe 210.

[0035] Smartphone A stores and executes an ultrasound application 299A. Smartphone B stores and executes an ultrasound application 299B. The ultrasound application 299A and the ultrasound application 299B may be configured to enable smartphone A and smartphone B to interact with the ultrasound probe 210 via the network 201. For example, ultrasound application 299A and ultrasound application 299B may be configured to enable displays of ultrasound images from the ultrasound probe 210 and to generate and display guidance.

[0036] As set forth above, system 200 may comprise an ultrasound system with a controller 250 in an ultrasound probe 210, as well as smartphone A and smartphone B. When executed by processor 252, instructions stored in the memory 251 and / or in memories of smartphone A and / or smartphone B may cause the system 200 to implement some or all features of the methods in FIG. 3 and / or FIG. 6. In FIG. 2, a user interface may be generated by the instructions stored in memory 251 and may be displayed on a display of smartphone A and / or smartphone B. Alternatively, the user interface may be generated by the ultrasound application 299A for smartphone A and / or the ultrasound application 299B for smartphone B. The user interface on smartphone A and / or smartphone B may provide guidance for a user consistent with the explanations herein.

[0037] Using the system 200, a user may initiate ultrasound imaging using the ultrasound probe 210. The ultrasound probe 210 may comprise a one-dimensional array probe or a two- dimensional matrix array probe. The ultrasound probe 210 is used to generate ultrasound images. In FIG. 2, the controller 250 may be configured to process a set of ultrasound images. The controller 250 may obtain a set of lung abnormalities for a classification model from the set of ultrasound images. The controller 250 may apply the classification model to the set of lungDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - abnormalities for generating a risk score for severity of smoke inhalation lung injury. The risk score may be displayed on the display 280. Although not shown in FIG. 1, a separate controller, such as in a separate mobile device or controller, may be configured to process the set of ultrasound images, obtain the set of lung abnormalities for a classification model from the set of ultrasound images, apply the classification model to the set of lung abnormalities for generating a risk score for severity of smoke inhalation lung injury. In some embodiments, the controller 250 or a separate controller, may be configured to apply a classification model to the set of abnormalities for generating a risk score for severity of at least one of heart disease, or respiratory disease from smoke inhalation, and the classification model by generate the risk score for severity of the at least one of heart disease, or respiratory disease from smoke inhalation. In other words, the classification model may be configured to assess and generate risk scores for heart disease, respiratory disease, or the lung injury as initially described. Moreover, in some embodiments, the imaging modality may be other than the ultrasound modality in FIG. 2.

[0038] In some embodiments, functions attributed directly to the controller 250 may be performed instead at a device connected to the ultrasound probe 210, so smartphone A or smartphone B in the embodiment of FIG. 2. Alternatively, functions attributed directly to the controller 250 may be performed at a data center in a cloud environment, such as by the controller 250 or smartphone A or smartphone B providing selected content over the internet to a data center in the cloud environment. In other embodiments, functions attributed directly to the controller 250 may be performed by and at an external computer, including an external system remote from the controller 250 and connected over a local or wide area network to the controller 250 or smartphone A or smartphone B.

[0039] FIG. 3 illustrates a method for smoke inhalation risk detection, in accordance with a representative embodiment.

[0040] The method of FIG. 3 is a method for automated quantification of lung ultrasound abnormalities that correlate with smoke inhalation lung injury, and results in a display of a risk score derived from the detected abnormalities.

[0041] At S310, ultrasound cineloops are acquired after starting. The ultrasound cineloops may be acquired by the system 100 in FIG. 1 or the system 200 in FIG. 2. At least one ultrasound cineloop may be acquired in each of the left lung and the right lung. Each ultrasound cineloop isDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - acquired and provided to a cineloop classification processor. A cineloop includes multiple image frames, typically 10 to 300, acquired continuously over a period of a few seconds, typically 1 to 10 seconds. The cineloop classification processor may be implemented by the controller 150 in FIG. 1 or controller 250 in FIG. 2, or may be implemented by a controller of a computer integrated with but separate from the elements of the system 100 in FIG. 1 or the system 200 in FIG. 2.

[0042] At S320, each cineloop is analyzed by the cineloop classification processor to detect and classify abnormalities in multiple abnormality categories. Lung abnormalities are identified from ultrasound images, such as from ultrasound images of each of a left lung and a right lung. The acquired cineloops from S310 are provided to a cineloop processor for analysis in which the presence and severity of multiple abnormality features are quantified for each cineloop. The abnormality feature categories to be analyzed may include B-lines, pleural line irregularities, lung consolidation, pleural effusion, and lack of lung sliding. The cineloop processor may apply a trained artificial intelligence (Al) network to detect abnormality features for each abnormality category. The trained artificial intelligence network may comprise a classification model which is trained on inputs including at least one of B-lines, merged B-lines, abnormal pleural lines, lung consolidations, pleural effusions, or absence of lung sliding.

[0043] In a first aspect of S320, the cineloop processor may apply a trained artificial intelligence (Al) network to detect abnormality features for each abnormality category. Different trained artificial network networks may be employed for each of the categories of abnormalities, or a single (larger) network may be used to detect features of several or all categories with a single network.

[0044] In a second aspect of S320, for each abnormality category, the cineloop processor may calculate metrics based on the sizes and confidences of individual detections to determine the cineloop-level presence or absence of abnormality features, as well as the severity of features. The cineloop processor may apply a trained artificial intelligence network to the set of lung abnormalities for generating a risk score for severity of smoke inhalation lung injury. In some embodiments, the detection of abnormality features and the calculation of metrics to generate a risk score may be performed by different trained artificial intelligence models. In other embodiments, the same model may perform the detecting of abnormalities and the generating ofDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - the risk scores.

[0045] For some features such as consolidation and / or pleural effusion, severity of a feature may be identified based on a size threshold. If the detected sizes are smaller than a size threshold, such as smaller than 1 centimeter, the feature may be classified “low severity”, otherwise the feature may be classified “high severity”. For other features such as B-lines and / or pleural line abnormalities, the severity categories may be identified directly as part of the detection process or based on counting the occurrence of features. Identification as part of the detection process may be based, for example, on detecting videos with “thickened pleural line” versus videos with “interrupted pleural line” in the pleural line category. Identification as part of the detection process may be alternatively or also based on counting the occurrence of features, such as for B-lines, counting fewer than 3 B-lines per video, versus counting 3 or more B-lines per video.

[0046] At S330, the method of FIG. 3 includes summarizing abnormality findings from all cineloops for each of one or a plurality of abnormality categories. The abnormality findings may be summarized based on the fraction or percentage of videos that exhibit low-severity and / or high-severity abnormalities in the different abnormality categories. Examples of summaries of abnormality findings are shown in and described below with respect to FIG. 4A and FIG. 4B.

[0047] At S340, risk scores are determined. At least one risk score for the patient may be calculated and displayed based on the abnormality findings. Optionally, a risk score may be calculated separately for each lung, so based separately on the lung ultrasound cineloops obtained in each lung. The risk score may be based on a classifier that is developed based on training data from subjects with known smoke inhalation lung injury status in which lung ultrasound exams were obtained and mapped to the presence or absence of low-severity and / or high-severity abnormalities in the different abnormality categories. In one embodiment, Quadratic Discriminant Analysis may be used to model the distributions of abnormality percentages for subjects with and without smoke inhalation lung injury in the training data set as chi-squared distributions with n degrees of freedom, where n is the number of abnormality categories employed. For example, n may be equal to 5 if B-lines, abnormal pleural line, lung consolidation, pleural effusion, and lack of lung sliding are used. For the calculated percentages of videos in the abnormality categories in a subject to be assessed (a “test sample”), theDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT -Mahalanobis distance to the centroids of the smoke inhalation lung injury -positive and smoke inhalation lung injury-negative training distributions may be calculated, which allows calculation of the probability p+ and p- that the test sample belongs to either the smoke inhalation lung injury -positive or the smoke inhalation lung injury-negative class. The probability of being smoke inhalation lung injury -positive (p+), can be displayed as the total risk score in FIG. 5. Optionally, the probability of being smoke inhalation lung injury-negative (p-) may also be displayed, which may indicate the reliability of the risk score result. Alternative display options may include scaled (0-100%), binned (e.g., 0, 1, 2, ... 10), or otherwise modified versions of the probability (e.g., logarithmically scaled). Optionally, the user may identify the subset of lung ultrasound abnormalities that the risk score calculation should be based on. Optionally, the user may provide or modify the weighting of the contributions of different lung ultrasound abnormalities to the overall risk score. Selectable weighting parameters may be used to calculate the risk scores.

[0048] At S350, abnormality summaries and risk scores are displayed, after which the method of FIG. 3 stops. The findings from all cineloops acquired in one lung ultrasound exam are summarized at S330 and displayed at S350. The display at S350 may include the fraction or percentage of videos that include any abnormalities in the analyzed abnormality categories, the fraction or percentage of videos that include severe abnormalities in the analyzed abnormality categories, and / or an overall risk score for each lung, and / or for the patient, based on the cineloops acquired in each lung. The display at S350 may be of abnormality summaries for each abnormality category, and may be provided with risk scores for each lung and an overall risk score.

[0049] FIG. 4A illustrates a display of risk scores for abnormality findings in smoke inhalation risk detection, in accordance with a representative embodiment.

[0050] FIG. 4A illustrates a display of a summary of abnormality findings in a subject with low rates / severities of abnormalities throughout the lung exam, indicative of low risk of smoke inhalation lung injury or other serious lung conditions.

[0051] FIG. 4B illustrates another display of risk scores for abnormality findings in smoke inhalation risk detection, in accordance with a representative embodiment.

[0052] FIG. 4B illustrates a display of a summary of abnormality findings in a subject with highDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - rates / severities of some abnormalities throughout the lung exam, indicative of high risk of smoke inhalation lung injury or other serious lung conditions.

[0053] FIG. 5 illustrates risk scores for each of a pair of lungs in smoke inhalation risk detection, in accordance with a representative embodiment.

[0054] In FIG. 5, smoke inhalation lung injury scores are displayed for each lung based on the lung ultrasound cineloops obtained in each lung. FIG. 5 also shows a total smoke inhalation lung injury risk score based on all cineloops obtained in the LUS exam. The smoke inhalation lung injury risk score for the right lung in FIG. 5 is .75. The smoke inhalation lung injury risk score for the left lung in FIG. 5 is .32. The total smoke inhalation lung injury risk score in FIG. 5 is .54.

[0055] FIG. 6 illustrates another method for smoke inhalation risk detection, in accordance with a representative embodiment.

[0056] At S605, a classification model is trained. The classification model may be trained on inputs including at least one of B-lines, merged B-lines, abnormal pleural lines, lung consolidations, pleural effusions, or absence of lung sliding. The classification model may be implemented by the controller 150 in FIG. 1 or the controller 250 in FIG. 2, or by separate computers integrated with but separate from the system 100 or the system 200.

[0057] At S610, a set of lung abnormalities are detected. The set of lung abnormalities are detected for a classification model. The set of lung abnormalities may be detected by a model that includes the classification model, or the detection and the classification may be performed by same models.

[0058] At S620, the set of lung abnormalities are obtained. The lung abnormalities may be identified from ultrasound images of each of a left lung and a right lung, and provided to the classification model. Abnormalities may be summarized for each of a plurality of categories of abnormalities.

[0059] At S630, the classification model is applied to the set of obtained lung abnormalities. The classification model is applied for generating a risk score for severity of smoke inhalation lung injury. The classification model may be configured to generate risk scores for smoke inhalation lung injury, or in other embodiments for at least one of heart disease or respiratory disease. The applying may comprise applying summaries of the abnormalities for each of a plurality ofDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - categories to the classification model. A different risk score may be generated for each of the plurality of categories.

[0060] For example, at S640, one or more risk score is generated. A display may be generated for each of the summaries of the abnormalities and the different risk score for each of the plurality of categories. Examples of risk scores for pluralities of categories are shown in FIG. 4A and FIG. 4B.

[0061] Teachings herein are not limited to smoke inhalation causing lung injury or respiratory disease. Rather, the teachings herein may be applied to classifying risks of heart disease. Heart disease detection / classifi cation may be performed completely independent of “smoke inhalation” - but using the same ultrasound-based abnormality detections described for “smoke inhalation lung injury detection,” such as the presence / severity of B-lines and pleural effusions, as visible in lung ultrasound. The teachings herein relating to using artificial intelligence models on lung ultrasound images to classify presence and severity of abnormalities such as B-lines and pleural effusion, display percentages of images classified as having low severity / high severity abnormalities as in FIG. 4A and FIG. 4B, and showing a total risk score for an injury / disease as in FIG. 5 are applicable for heart disease as well. The specific calculation of the total risk score, however, for heart disease differs in that the total risk score is calculated from the individual percentages / severities of abnormalities. That calculation described herein is based on modeling the distribution of lung ultrasound abnormalities in subjects with and without a certain injury / disease, such as using quadratic discriminant analysis and Mahalanobis distance. This modeling is carried out in populations with and without heart disease in order to determine the specific parameters that allow the calculation and display of the heart disease total risk score instead of smoke inhalation lung injury total risk score.

[0062] FIG. 7 illustrates a computer system, on which a method for smoke inhalation risk detection is implemented, in accordance with another representative embodiment.

[0063] Referring to FIG. 7, the computer system 700 includes a set of software instructions that can be executed to cause the computer system 700 to perform any of the methods or computer- based functions disclosed herein. The computer system 700 may operate as a standalone device or may be connected, for example, using a network 701, to other computer systems or peripheral devices. In embodiments, a computer system 700 performs logical processing based on digitalDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - signals received via an analog-to-digital converter.

[0064] In a networked deployment, the computer system 700 operates in the capacity of a server or as a client user computer in a server-client user network environment, or as a peer computer system in a peer-to-peer (or distributed) network environment. The computer system 700 can also be implemented as or incorporated into various devices, such as a workstation that includes a controller, a stationary computer, a mobile computer, a personal computer (PC), a laptop computer, a tablet computer, or any other machine capable of executing a set of software instructions (sequential or otherwise) that specify actions to be taken by that machine. The computer system 700 can be incorporated as or in a device that in turn is in an integrated system that includes additional devices. In an embodiment, the computer system 700 can be implemented using electronic devices that provide voice, video or data communication. Further, while the computer system 700 is illustrated in the singular, the term “system” shall also be taken to include any collection of systems or sub-systems that individually or jointly execute a set, or multiple sets, of software instructions to perform one or more computer functions.

[0065] As illustrated in FIG. 7, the computer system 700 includes a processor 710. The processor 710 may be considered a representative example of a processor of a controller and executes instructions to implement some or all aspects of methods and processes described herein. The processor 710 is tangible and non-transitory. As used herein, the term “non- transitory” is to be interpreted not as an eternal characteristic of a state, but as a characteristic of a state that will last for a period. The term “non-transitory” specifically disavows fleeting characteristics such as characteristics of a carrier wave or signal or other forms that exist only transitorily in any place at any time. The processor 710 is an article of manufacture and / or a machine component. The processor 710 is configured to execute software instructions to perform functions as described in the various embodiments herein. The processor 710 may be a general- purpose processor or may be part of an application specific integrated circuit (ASIC). The processor 710 may also be a microprocessor, a microcomputer, a processor chip, a controller, a microcontroller, a digital signal processor (DSP), a state machine, or a programmable logic device. The processor 710 may also be a logical circuit, including a programmable gate array (PGA), such as a field programmable gate array (FPGA), or another type of circuit that includes discrete gate and / or transistor logic. The processor 710 may be a central processing unit (CPU), aDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - graphics processing unit (GPU), or both. Additionally, any processor described herein may include multiple processors, parallel processors, or both. Multiple processors may be included in, or coupled to, a single device or multiple devices.

[0066] The term “processor” as used herein encompasses an electronic component able to execute a program or machine executable instruction. References to a computing device comprising “a processor” should be interpreted to include more than one processor or processing core, as in a multi-core processor. A processor may also refer to a collection of processors within a single computer system or distributed among multiple computer systems. The term computing device should also be interpreted to include a collection or network of computing devices each including a processor or processors. Programs have software instructions performed by one or multiple processors that may be within the same computing device or which may be distributed across multiple computing devices.

[0067] The computer system 700 further includes a main memory 720 and a static memory 730, where memories in the computer system 700 communicate with each other and the processor 710 via a bus 708. Either or both of the main memory 720 and the static memory 730 may be considered representative examples of a memory of a controller, and store instructions used to implement some, or all aspects of methods and processes described herein. Memories described herein are tangible storage mediums for storing data and executable software instructions and are non-transitory during the time software instructions are stored therein. As used herein, the term “non-transitory” is to be interpreted not as an eternal characteristic of a state, but as a characteristic of a state that will last for a period. The term “non-transitory” specifically disavows fleeting characteristics such as characteristics of a carrier wave or signal or other forms that exist only transitorily in any place at any time. The main memory 720 and the static memory 730 are articles of manufacture and / or machine components. The main memory 720 and the static memory 730 are computer-readable mediums from which data and executable software instructions can be read by a computer (e.g., the processor 710). Each of the main memory 720 and the static memory 730 may be implemented as one or more of random access memory (RAM), read only memory (ROM), flash memory, electrically programmable read only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, a hard disk, a removable disk, tape, compact disk read only memory (CD-ROM), digital versatile diskDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT -(DVD), floppy disk, blu-ray disk, or any other form of storage medium known in the art. The memories may be volatile or non-volatile, secure and / or encrypted, unsecure and / or unencrypted.

[0068] “Memory” is an example of a computer-readable storage medium. Computer memory is any memory which is directly accessible to a processor. Examples of computer memory include, but are not limited to RAM memory, registers, and register files. References to “computer memory” or “memory” should be interpreted as possibly being multiple memories. The memory may for instance be multiple memories within the same computer system. The memory may also be multiple memories distributed amongst multiple computer systems or computing devices.

[0069] As shown, the computer system 700 further includes a video display unit 750, such as a liquid crystal display (LCD), an organic light emitting diode (OLED), a flat panel display, a solid-state display, or a cathode ray tube (CRT), for example. Additionally, the computer system 700 includes an input device 760, such as a keyboard / virtual keyboard or touch-sensitive input screen or speech input with speech recognition, and a cursor control device 770, such as a mouse or touch-sensitive input screen or pad. The computer system 700 also optionally includes a disk drive unit 780, a signal generation device 790, such as a speaker or remote control, and / or a network interface device 740.

[0070] In an embodiment, as depicted in FIG. 7, the disk drive unit 780 includes a computer- readable medium 782 in which one or more sets of software instructions 784 (software) are embedded. The sets of software instructions 784 are read from the computer-readable medium 782 to be executed by the processor 710. Further, the software instructions 784, when executed by the processor 710, perform one or more steps of the methods and processes as described herein. In an embodiment, the software instructions 784 reside all or in part within the main memory 720, the static memory 730 and / or the processor 710 during execution by the computer system 700. Further, the computer-readable medium 782 may include software instructions 784 or receive and execute software instructions 784 responsive to a propagated signal, so that a device connected to a network 701 communicates voice, video or data over the network 701. The software instructions 784 may be transmitted or received over the network 701 via the network interface device 740.

[0071] In an embodiment, dedicated hardware implementations, such as application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), programmable logic arraysDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - and other hardware components, are constructed to implement one or more of the methods described herein. One or more embodiments described herein may implement functions using two or more specific interconnected hardware modules or devices with related control and data signals that can be communicated between and through the modules. Accordingly, the present disclosure encompasses software, firmware, and hardware implementations. Nothing in the present application should be interpreted as being implemented or implementable solely with software and not hardware such as a tangible non-transitory processor and / or memory.

[0072] In accordance with various embodiments of the present disclosure, the methods described herein may be implemented using a hardware computer system that executes software programs. Further, in an exemplary, non-limited embodiment, implementations can include distributed processing, component / object distributed processing, and parallel processing. Virtual computer system processing may implement one or more of the methods or functionalities as described herein, and a processor described herein may be used to support a virtual processing environment.

[0073] Accordingly, smoke inhalation risk detection enables application of point-of-care lung ultrasound to detect smoke inhalation lung injury and other types of injuries. Smole inhalation lung injury can lead to several types of lung abnormalities visible in lung ultrasound. With the teachings herein, lung ultrasound abnormalities which are detectable in lung ultrasound obtained after smoke inhalation may correlate with the presence or absence of smoke inhalation lung injury. A method for risk stratification in patients suspected of smoke inhalation lung injury may be implemented by extracting and quantifying information about lung ultrasound abnormalities after smoke inhalation. Artificial intelligence (Al) may be applied to detect the ultrasound abnormalities and may be used to generate a risk score that can help make treatments decisions for patients with smoke inhalation lung injury and other pulmonary conditions. Output information such as risk scores can be summarized and used by healthcare providers to make monitoring and treatment decisions, including intubation, to provide optimal care for smoke inhalation lung injury victims. As noted previously, notwithstanding the focus of teachings herein on ultrasound and smoke inhalation, the underlying technical solutions described herein are not particularly limited to ultrasound as an imaging modality or to smoke inhalation lung injury as an injury type.Docket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT -

[0074] Although smoke inhalation risk detection has been described with reference to several exemplary embodiments, it is understood that the words that have been used are words of description and illustration, rather than words of limitation. Changes may be made within the purview of the appended claims, as presently stated and as amended, without departing from the scope and spirit of smoke inhalation risk detection in its aspects. Although smoke inhalation risk detection has been described with reference to particular means, materials and embodiments, smoke inhalation risk detection is not intended to be limited to the particulars disclosed; rather smoke inhalation risk detection extends to all functionally equivalent structures, methods, and uses such as are within the scope of the appended claims.

[0075] The illustrations of the embodiments described herein are intended to provide a general understanding of the structure of the various embodiments. The illustrations are not intended to serve as a complete description of all of the elements and features of the disclosure described herein. Many other embodiments may be apparent to those of skill in the art upon reviewing the disclosure. Other embodiments may be utilized and derived from the disclosure, such that structural and logical substitutions and changes may be made without departing from the scope of the disclosure. Additionally, the illustrations are merely representational and may not be drawn to scale. Certain proportions within the illustrations may be exaggerated, while other proportions may be minimized. Accordingly, the disclosure and the figures are to be regarded as illustrative rather than restrictive.

[0076] One or more embodiments of the disclosure may be referred to herein, individually and / or collectively, by the term “invention” merely for convenience and without intending to voluntarily limit the scope of this application to any particular invention or inventive concept. Moreover, although specific embodiments have been illustrated and described herein, it should be appreciated that any subsequent arrangement designed to achieve the same or similar purpose may be substituted for the specific embodiments shown. This disclosure is intended to cover any and all subsequent adaptations or variations of various embodiments. Combinations of the above embodiments, and other embodiments not specifically described herein, will be apparent to those of skill in the art upon reviewing the description.

[0077] The Abstract of the Disclosure is provided to comply with 37 C.F.R. §1.72(b) and is submitted with the understanding that it will not be used to interpret or limit the scope orDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - meaning of the claims. In addition, in the foregoing Detailed Description, various features may be grouped together or described in a single embodiment for the purpose of streamlining the disclosure. This disclosure is not to be interpreted as reflecting an intention that the claimed embodiments require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive subject matter may be directed to less than all of the features of any of the disclosed embodiments. Thus, the following claims are incorporated into the Detailed Description, with each claim standing on its own as defining separately claimed subject matter.

[0078] The preceding description of the disclosed embodiments is provided to enable any person skilled in the art to practice the concepts described in the present disclosure. As such, the above disclosed subject matter is to be considered illustrative, and not restrictive, and the appended claims are intended to cover all such modifications, enhancements, and other embodiments which fall within the true spirit and scope of the present disclosure. Thus, to the maximum extent allowed by law, the scope of the present disclosure is to be determined by the broadest permissible interpretation of the following claims and their equivalents and shall not be restricted or limited by the foregoing detailed description.

Claims

Docket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT -CLAIMS:

1. A system for classifying lung injury or disease, comprising: a memory that stores instructions; and a processor that executes the instructions, wherein, when executed by the processor, the instructions cause the system to: obtain a set of lung abnormalities for a classification model; apply the classification model to the set of lung abnormalities for generating a risk score for severity of lung injury or disease; and generate, by the classification model, the risk score for severity of lung injury or disease.

2. The system of claim 1, wherein the lung abnormalities are identified from ultrasound images of each of a left lung and a right lung, and wherein the lung injury or disease is caused by smoke inhalation.

3. The system of claim 1, wherein the classification model is also configured to generate risk scores for at least one of heart disease or respiratory disease4. The system of claim 1, wherein, when executed by the processor, the instructions further cause the processor to detect the set of lung abnormalities for the classification model.

5. The system of claim 4, wherein the applying and the detecting are performed by the same model.

6. The system of claim 1, wherein the classification model is trained on inputs including at least one of B-lines, merged B-lines, abnormal pleural lines, lung consolidations, pleural effusions, or absence of lung sliding.Docket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT -7. A method for classifying injury or disease, comprising: obtaining, by a system comprising a memory that stores instructions and a processor that executes the instructions, a set of abnormalities from a plurality of ultrasound cineloops for a classification model, each of the plurality of ultrasound cineloops comprising a plurality of ultrasound frames; applying a classification model to the set of abnormalities for generating a risk score for severity of an injury indicated by the set of abnormalities; and generating, by the classification model, the risk score for the severity of injury indicated by the set of abnormalities.

8. The method of claim 7, wherein the injury comprises a smoke inhalation lung injury.

9. The method of claim 7, wherein the classification model is configured to also generate risk scores for at least one of heart disease or respiratory disease.

10. The method of claim 7, further comprising: detecting the set of abnormalities for the classification model from the plurality of ultrasound cineloops.

11. The method of claim 10, wherein the applying and the detecting are performed by the same model.

12. The method of claim 7, wherein the classification model is trained on inputs including at least one of B-lines, merged B-lines, abnormal pleural lines, lung consolidations, pleural effusions, or absence of lung sliding13. The method of claim 7, further comprising: summarizing abnormalities for each of a plurality of categories, wherein the applying comprises applying summaries of the abnormalities for each of the plurality of categories to theDocket No. (PCIP.2583)Docket No. (2024PF00241)- PATENT - classification model, and wherein the generating comprises generating a different risk score for each of the plurality of categories; and generating a display of each of the summaries of the abnormalities and the different risk score for each of the plurality of categories.

14. The method of claim 7, further comprising: recommending a treatment for the set of abnormalities based on the risk score.

15. The method of claim 10, further comprising: providing acquired cineloops to a cineloop processor; and determining the set of abnormalities based on the acquired cineloops.

16. The method of claim 13, wherein the display comprises at least one of: a fraction of cineloops that include any abnormalities; a fraction of cineloops that include severe abnormalities; or an overall risk score for each of a left lung and right lung, based on the cineloops acquired for each lung.

17. A system for classifying smoke inhalation injury, comprising: a memory that stores instructions; and a processor that executes the instructions, wherein, when executed by the processor, the instructions cause the system to: obtain a set of abnormalities for a classification model; apply the classification model to the set of abnormalities for generating a risk score for severity of at least one of heart disease or respiratory disease from smoke inhalation; and generate, by the classification model, the risk score for severity of the at least one of heart disease or respiratory disease from smoke inhalation.

18. The system of claim 17, wherein the classification model is trained on inputs including at least one of B-lines, merged B-lines, abnormal pleural lines, lung consolidations, pleural effusions, or absence of lung slidingDocket No. (PCIP.2583)Docket No. (2024PF00241) - PATENT -19. The system of claim 17, wherein, when executed by the processor, the instructions further cause the system to: summarize abnormalities for each of a plurality of categories, wherein applying the classification model comprises applying summaries of the abnormalities for each of the plurality of categories to the classification model, and wherein generating the risk score comprises generating a different risk score for each of the plurality of categories; and generating a display of each of the summaries of the abnormalities and the different risk score for each of the plurality of categories.

20. The system of claim 17, wherein, when executed by the processor, the instructions further cause the system to: recommend a treatment for the set of abnormalities based on the risk score.25

Images