Compression guidance based on echocardiography

Abstract

An ultrasound system includes a memory that stores instructions and a processor that executes the instructions. The instructions, when executed by the processor, cause the ultrasound system to receive a plurality of echocardiogram images at a time of performing a compression, interpret each of the plurality of echocardiogram images, output each of the plurality of echocardiogram images, and generate and output real-time feedback for a quality of compression based on interpreting each of the plurality of echocardiogram images.

Description

[0001] Government Rights Statement This invention was made with the support of grant number HHS / ASPR / BARDA 75A50120C00097 from the Biomedical Advanced Research and Development Authority (BARDA) of the U.S. Department of Health and Human Services. The U.S. government holds certain rights to this invention. Background Technology

[0002] Cardiac arrest is one of the leading causes of death worldwide, with a survival rate of less than 10% for out-of-hospital cardiac arrest. For in-hospital cardiac arrest, echocardiography is sometimes used as a bedside examination to determine the cause of cardiac arrest and guide cardiac arrest resuscitation (CAR). Cardiac arrest resuscitation may involve emergency cardiopulmonary resuscitation (CPR). CPR consists of chest compressions and artificial ventilation. Ventilation is performed mouth-to-mouth or using a mechanical ventilation system.

[0003] Transesophageal echocardiography (TEE) is a form of echocardiography that involves inserting a specialized probe with an ultrasound transducer through the patient's mouth into the esophagus to monitor the heart. Recent advancements in TEE technology also include nasal insertion into the esophagus for cardiac monitoring. Nasal TEE involves inserting a single-plane TEE through the patient's nose into the esophagus. Transthoracic echocardiography (TTE) is another form of echocardiography that places a specialized probe with an ultrasound transducer in the patient's chest or abdomen to monitor the heart. Compared to TTE, the use of TEE in cardiac arrest resuscitation (CAR) has been shown to have a clinically significant impact on critically ill and hemodynamically unstable patients. TEE continuously monitors cardiopulmonary activity, provides undisturbed continuous images of myocardial activity, and offers excellent image quality for diagnosing disease states and identifying potential reversible causes of cardiac arrest. The American College of Emergency Physicians (ACEP) has recognized the use of echocardiography with TEE as a standard tool for ultrasound-guided resuscitation.

[0004] In emergency medicine and intensive care settings, medical professionals can use focused cardiac ultrasound (FoCUS) as a goal-oriented framework to quickly determine the cause of cardiac arrest and guide resuscitation. FoCUS transthoracic echocardiography is the most commonly used diagnostic method by emergency physicians today.

[0005] Current guidelines for CPR recommend that, to promote ventricular blood flow, the correct hand or machine placement for chest compressions should be "center of the patient's chest." However, radiological studies using computed tomography (CT) and cardiac magnetic resonance imaging (MRI) have found that the left ventricle (LV) is not always located in the center of the sternum; instead, in 50–80% of patients, this location is actually the left ventricular outflow tract (LVOT), aortic valve, or aortic root. Blindly compressing at the aortic root / LVOT can lead to outflow tract obstruction and is associated with poor prognosis and survival due to inadequate forward cardiac flow. Emerging research suggests that direct compressions above the LV can improve survival and prognosis, but rapid and reliable LV localization remains a significant obstacle for physicians responding to cardiac arrest for the first time.

[0006] Studies guided by TEE have confirmed these findings, indicating that healthcare professionals often perform inappropriate compressions at locations associated with outflow tract obstruction. A prospective study of cardiac arrest patients found that when compressions were performed at the center of the chest, the area of ​​maximum compression (AMC) was inappropriately located above the LVOT in 41% of patients, compared to 59% at the aorta (including the aortic valve). Another more recent study of out-of-hospital cardiac arrest patients found that inappropriate AMC occurred at the aortic root / LVOT in 53% of cases. Both studies highlight the fact that a high proportion of patients may not receive appropriate compressions in the correct location to achieve the goals of coronary and cerebral perfusion. Comparing compressions at the aortic root / LVOT with compressions at the LV, studies have found that direct LV compression improves hemodynamic efficiency and increases survival rates.

[0007] While the optimal location for hand or machine placement to generate the most effective compressions may vary from patient to patient, using TEE to guide hand placement from the aortic root or LVOT to areas with a stronger hemodynamic response has been shown to improve outcomes by increasing spontaneous circulation rate. By developing a guidance solution that assesses patient hemodynamic variables to guide the user to the appropriate area of ​​maximum compression, individualized CPR can be achieved, and outcomes for patients undergoing cardiac arrest resuscitation can be significantly improved.

[0008] Both TEE and TTE require optimization of chest compression quality to improve resuscitation outcomes. Currently, there is a lack of comprehensive, real-time guidance to improve the quality of resuscitation, and even medical professionals experience chest compression errors as high as 70% due to improper hand placement. This is partly because the position of the lower ventricle (LV) is not the same for every patient. Another reason is that existing solutions do not monitor chest compression placement. A more individualized approach is needed to optimize chest compressions during CPR.

[0009] Automated software solutions can improve the efficiency and clinical performance of cardiac ultrasound workflows. However, there is a need to develop echocardiography applications for bedside users in emergency medicine and intensive care settings, where examination time is limited, decisions must be made immediately, and user experience levels vary considerably. A particular requirement is to provide users with real-time feedback and assessment to optimize the quality of chest compressions. Summary of the Invention

[0010] According to one aspect of this disclosure, an ultrasound system includes a memory storing instructions and a processor executing the instructions. When executed by the processor, the instructions cause the ultrasound system to: receive a plurality of echocardiographic images during compression; interpret each of the plurality of echocardiographic images; output each of the plurality of echocardiographic images; and generate and output real-time feedback on the quality of compression based on the interpretation of each of the plurality of echocardiographic images.

[0011] According to another aspect of this disclosure, a method of operating an ultrasound system, the ultrasound system including a memory storing instructions and a processor executing the instructions, the method comprising: receiving a plurality of echocardiographic images during the execution of chest compressions. The method further comprises: interpreting each of the plurality of echocardiographic images; outputting each of the plurality of echocardiographic images; and generating and outputting real-time feedback on the quality of chest compressions based on the interpretation of each of the plurality of echocardiographic images.

[0012] According to another aspect of this disclosure, a tangible, non-transitory computer-readable medium stores instructions. When executed by a processor, the instructions cause the processor to: receive a plurality of echocardiographic images during compression; interpret each of the plurality of echocardiographic images; output each of the plurality of echocardiographic images; and generate and output real-time feedback on the quality of compression based on the interpretation of each of the plurality of echocardiographic images. Attached Figure Description

[0013] A better understanding of the exemplary embodiments can be achieved by reading the following detailed description in conjunction with the accompanying drawings. It should be emphasized that the various features are not necessarily drawn to scale. In fact, dimensions may be arbitrarily increased or decreased for clarity. Where applicable and practicable, the same reference numerals denote the same elements.

[0014] Figure 1 A system for echocardiography-based resuscitation guidance, according to a representative embodiment, is shown.

[0015] Figure 2 Another system for echocardiography-based resuscitation guidance, according to a representative embodiment, is shown.

[0016] Figure 3 A method for echocardiography-based resuscitation guidance is illustrated according to a representative embodiment.

[0017] Figure 4 Another method for echocardiography-based resuscitation guidance, according to a representative embodiment, is shown.

[0018] Figure 5 Another method for echocardiography-based resuscitation guidance, according to a representative embodiment, is shown.

[0019] Figure 6 A user interface for echocardiography-based resuscitation guidance is shown according to a representative embodiment.

[0020] Figure 7 The development and deployment of a model for echocardiography-based resuscitation guidance, according to a representative embodiment, are illustrated.

[0021] Figure 8 A computer system according to another representative embodiment is shown, on which a method for echocardiography-based resuscitation guidance is implemented. Detailed Implementation

[0022] In the following detailed description, representative embodiments with specific details disclosed are set forth for interpretative and not limiting purposes in order to provide a full understanding of embodiments according to this teaching. However, other embodiments consistent with this disclosure but departing from the specific details disclosed herein remain within the scope of the claims. To avoid obscuring the description of representative embodiments, descriptions of known systems, devices, materials, methods of operation, and methods of manufacture may be omitted. Nevertheless, systems, devices, materials, and methods within the knowledge of those skilled in the art remain within the scope of this teaching and can be used according to representative embodiments. It should be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. The definitions and interpretations of terms herein supplement the technical and scientific meanings of terms commonly understood and accepted in the art of this teaching.

[0023] It should 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 to these terms. These terms are only used 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.

[0024] As used herein and in the claims, unless the context clearly indicates otherwise, the singular forms of the terms “a,” “an,” and “the” are intended to include both the singular and plural forms. Additionally, when used herein, the terms “comprising,” “including,” and / or similar terms indicate the presence of the stated features, elements, and / or components, but do not exclude the presence or addition of one or more other features, elements, components, and / or combinations thereof. As used herein, the term “and / or” includes any and all combinations of one or more of the associated listed items.

[0025] Unless otherwise stated, when a component or assembly is referred to as "connected to," "coupled to," or "proximity to" another component or assembly, it should be understood that the component or assembly can be directly connected to or coupled to the other component or assembly, or that there may be intermediate components or assemblies. That is, these and similar terms cover situations where one or more intermediate components or assemblies may be used to connect two components or assemblies. However, when a component or assembly is referred to as "directly connected" to another component or assembly, this only covers situations where the two components or assemblies are directly connected without any intermediate or intermediary components or assemblies between them.

[0026] Therefore, this disclosure is intended to provide one or more advantages, as specifically pointed out below, through various aspects, embodiments and / or specific features or sub-components.

[0027] As described herein, real-time assessment and feedback of cardiopulmonary compression quality can be provided based on anatomical location, depth, frequency, hemodynamic measurements, and / or anatomical imaging. This assessment and feedback can be used to optimize coronary perfusion and individualize chest compressions for patients during CPR. Echocardiographic-based assessment and compression guidance can be used to provide an effective user experience to support resuscitation efforts by optimizing the quality of chest compressions, thereby improving patient survival.

[0028] Figure 1 A system 100 for compression guidance based on echocardiography, according to a representative embodiment, is shown.

[0029] Figure 1 System 100 is a system for echocardiographic-based compression guidance, comprising components that can be provided together or distributed. System 100 includes an ultrasound probe 110, an ultrasound host 120, and a display 180.

[0030] An ultrasound probe 110 includes processing circuitry 115 and a transducer array 113. The ultrasound probe 110 may include a TEE ultrasound probe or a TTE ultrasound probe. The processing circuitry 115 may include a memory for storing data and instructions, an application-specific integrated circuit (ASIC), and / or a processor for processing data and instructions. The transducer array 113 includes an array of transducer elements, including 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 that bounce off body tissue, and then receives the echoes of the sound waves and converts the echoes back into electrical energy. The transducer array 113 may include tens, hundreds, or thousands of individual transducer elements. The ultrasound probe 110 may emit a beam to generate an image and may detect echoes. The processing circuitry 115 may process the ultrasound images captured by the transducer array 113 of the ultrasound probe 110.

[0031] The ultrasound main unit 120 may include an ultrasound cart. The ultrasound main unit 120 includes a first interface 121, a second interface 122, a third interface 123, and a controller 150. Figure 8 The image depicts a computer that can be used to implement the ultrasound host 120, but the ultrasound host 120 may include more than Figure 1 More elements as shown may also be included than Figure 8 The diagram shows more or fewer elements. One or more interfaces may include ports, disk drives, wireless antennas, or other types of receiver circuitry that connect controller 150 to other electronic components. A first interface 121 connects ultrasound host 120 to ultrasound probe 110 and may include ports, antennas, and / or other types of physical components for wired or wireless communication. A second interface 122 connects ultrasound host 120 to display 180 and may also include ports, antennas, and / or other types of physical components for wired or wireless communication. A third interface 123 is a user interface and may include buttons, keys, a mouse, a microphone, a speaker, a switch, a touchscreen, or other types of displays separate from display 180, and / or other types of physical components that allow medical personnel to interact with ultrasound host 120, such as inputting commands and receiving outputs.

[0032] The controller 150 includes at least a memory 151 for storing instructions and a processor 152 for executing the instructions. The instructions stored in the memory 151 may include one or more software programs for generating and outputting feedback (e.g., via a display 180) on the quality of chest compressions based on the interpretation of each of multiple echocardiographic images acquired by the ultrasound probe 110. The software programs in the memory 151 may include digital signal processing algorithms that can process 2D or 3D volumetric data to classify ultrasound images, detect features within ultrasound images, segment ultrasound images to extract regions of interest, calculate cardiac measurements, render a 3D reconstruction of the heart, and model the spatiotemporal characteristics of the compressions. These features may include, for example, the location of the left ventricle, and the compression location, depth, and frequency detected relative to the compression location. In the system 100, a user interface may be generated from the instructions stored in the memory 151 and may be displayed on the display 180.

[0033] The display 180 can be local to the ultrasound host 120 or can be remotely connected to the ultrasound host 120, for example, wirelessly. The display 180 includes a graphical user interface (GUI) 181 that displays ultrasound images and guidance information to the user.

[0034] The display 180 can be connected to the ultrasound host 120 via a local wired interface (e.g., Ethernet cable) or a local wireless interface (e.g., Wi-Fi connection). The display 180 can interface with other user input devices (including a mouse, keyboard, scroll wheel, etc.) for medical personnel to input commands. The display 180 can be a monitor (e.g., a computer monitor), a display on a mobile device, an augmented reality display, a television, an electronic whiteboard, or other screen configured to display electronic images. The display 180 may also include one or more input interfaces that can connect to the other elements or components described above, and an interactive touchscreen configured to display prompts to medical personnel and collect touch input from them.

[0035] Controller 150 may directly perform some of the operations described herein, or indirectly implement other operations described herein. For example, controller 150 may indirectly control operations, such as generating and transmitting content to be displayed on display 180. Controller 150 may directly control other operations, such as logical operations performed by processor 152, which executes instructions from memory 151, based on input received via an interface from electronic components and / or medical personnel. Therefore, when processor 152 executes instructions from memory 151, the process implemented by controller 150 may include steps not directly executed by controller 150.

[0036] As described above, system 100 may include an ultrasound system having a memory 151 for storing instructions and a processor 152 for executing the instructions. The instructions, when executed by the processor 152, cause the ultrasound system to perform a reference... Figure 3 The method described by some or all of the features of the method. The method performed by system 100 may include: a controller 150 of ultrasound host 120 receiving multiple echocardiographic images from ultrasound probe 110 during compression. The echocardiographic images may be, for example, TEE ultrasound images or TTE ultrasound images. The echocardiographic images may be received in real time or near real time and interpreted by controller 150. Instructions stored in memory 151 may be executed by processor 152 to cause controller 150 to interpret each of the multiple echocardiographic images received from ultrasound probe 110. The method performed by system 100 may also include: outputting each of the multiple echocardiographic images on a graphical user interface 181 of display 180. The method performed by system 100 may also include: controller 150 generating and outputting real-time feedback on compression quality on display 180 based on the interpretation of each of the multiple echocardiographic images.

[0037] Figure 2 Another system for echocardiography-based resuscitation guidance, according to a representative embodiment, is shown.

[0038] Network 201 may include a local wireless network such as a Wi-Fi network, but network 201 may also or alternatively include wired components, such as cables connecting smartphone A and smartphone B via USB cables. Smartphone A and smartphone B represent mobile devices, such as smartphones and tablets, or other networked / network-enabled devices with logical processing capabilities. Additionally, smartphone A and smartphone B are used as examples to illustrate that system 200 may include multiple different smart devices with applications or other functional capabilities, which, even if not necessarily dedicated to ultrasound imaging, can be functionally integrated with ultrasound probes in the overall system for ultrasound imaging.

[0039] The ultrasound probe 210 may include a portable transducer. The ultrasound probe 210 includes a transducer array 213, a lens 214, a user interface 223, a controller 250, and wireless communication circuitry 290. The transducer array 213 includes 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 that are reflected back from body tissue, and then receives the echoes of the sound waves and converts the echoes back into electrical energy. The transducer array 213 may include tens, hundreds, or thousands of individual transducer elements. The ultrasound probe 210 can emit a beam to generate an image and can detect echoes. The processor 252 can process the ultrasound images captured by the transducer array 213 of the ultrasound probe 210. The lens 214 can be used to emit an ultrasound beam and receive the echoes of the ultrasound beam. Medical personnel can interact with the ultrasound probe 210 using the user interface 223. Wireless communication circuit 290 can be used to communicate with smartphone A and smartphone B via network 201. Ultrasonic probe 210 can be configured to link to external devices, such as smartphone A and smartphone B, via an application installed on one or more external devices.

[0040] Smartphone A stores and executes ultrasound application 299A. Smartphone B stores and executes ultrasound application 299B. Ultrasound applications 299A and 299B can be configured to enable smartphones A and B to interact with ultrasound probe 210 via network 201. For example, ultrasound applications 299A and 299B can be configured to display ultrasound images from ultrasound probe 210. Ultrasound applications 299A and 299B can also be configured to generate and output real-time feedback on compression quality based on the interpretation of each of multiple echocardiographic images captured by ultrasound probe 210. In some embodiments, controller 250 can generate output and send it to ultrasound applications 299A and / or 299B for display on one or more screens of smartphones A and / or B.

[0041] The controller 250 includes at least a memory 251 for storing instructions and a processor 252 for executing instructions. The memory 251 may store one or more software programs. The software programs(s) may include signal processing techniques to assist in detecting cardiac events and assessing compression quality. Such software programs may include image classification models for classifying ultrasound images / pre-imaging data, feature detection models for detecting features within ultrasound images, segmentation models for segmenting cardiac structures, and / or quantification algorithms for calculating measurements based on ultrasound images / pre-imaging data and / or post-processed images to inform the user of cardiac function and compression quality. Figure 2In this system, the user interface can be generated by instructions stored in memory 251 and can be displayed on the screen of smartphone A or smartphone B.

[0042] Controller 250 may directly perform some of the operations described herein, or indirectly implement other operations described herein. For example, controller 250 may indirectly control operations, such as generating and transmitting content to be displayed on the screen of smartphone A or smartphone B. Controller 250 may directly control other operations, such as logical operations performed by processor 252, which executes instructions from memory 251, based on input received via an interface from electronic components and / or medical personnel. Therefore, when processor 252 executes instructions from memory 251, the process implemented by controller 250 may include steps not directly executed by controller 250.

[0043] Figure 1 System 100 and Figure 2 System 200 is used to optimize chest compressions, for example, after ruling out potential causes of cardiac arrest. It can provide the user with real-time feedback on compression quality based on compression location and associated hemodynamic response. There are multiple ways to measure compression success. The system design described in this paper provides a solution where compression success can be defined by multiple variables, and each variable can serve as an independent measure of compression success. The processing results described in this paper can provide a user interface to support the detection of compression success.

[0044] Figure 1 The ultrasonic probe 110 and Figure 2 The ultrasound probe 210 can operate as a CPR monitoring device to provide the user with real-time feedback and assessment of compression quality. System 100 and System 100 are operable to provide a feedback loop during compression implementation, assessment, and correction. Real-time feedback can be provided to the user, allowing for adjustments to compressions as needed, whether by adjusting hand / mechanical device placement, compression depth, and / or compression frequency. As an example, the feedback loop can be provided to external devices providing hemodynamic parameter information, automated compression machines performing compressions, and / or automated robotic systems controlling imaging systems performing echocardiography (such as TEE acquisition).

[0045] Figure 1 System 100 and / or Figure 2System 200 may each further include one or more controllable robots and one or more sensors. For example, controller 150 and / or controller 250 may provide feedback to the automated chest compression device, to wearable or invasive sensors monitoring hemodynamic pressure, and / or to the robotic TEE imaging system to adjust the echocardiographic images, thereby improving the feedback. For example, system 100 and / or system 200 may include a robot, such as an automated chest compression device configured to perform compressions. Instructions executed by controller 150 or controller 250 may further cause system 100 or system 200 to provide real-time feedback to the automated chest compression device for compressions to adjust the positioning of the automated chest compression device for compressions. Another example of a robot that may be included in system 100 or system 200 is a robotic system configured to control the source positioning of echocardiographic images. Real-time feedback from system 100 and / or system 200 may include feedback provided to the robotic system for adjusting the view in the echocardiographic images. The robotic system can be configured to manipulate an imaging system, such as a TEE ultrasound probe or a TTE ultrasound probe, in one or more dimensions, such as anterior / posterior, right / left, clockwise or counterclockwise, or advance / retract. The robotic system can also be used to control image characteristics, such as one or more imaging angles of the ultrasound imaging system. Controller 150 and / or controller 250 can provide feedback to control the manipulation of the robotic system and the corresponding ultrasound imaging system. As another example, system 100 and / or system 200 may include one or more sensors that interact with system 100 and / or system 200. The sensors can be sensors configured to measure hemodynamic variables of a patient and provide feedback to the controller (including memory and processor). Examples of such sensors may include interventional sensors configured to receive instructions from the controller to stimulate hemodynamic variables. The sensors may also be, or alternatively, wearable sensors that receive feedback from controller 150 or controller 250 and provide measurement results to controller 150 or controller 250. Sensors can also be invasive sensors, such as the Swaz-Gantz catheter used to measure / monitor hemodynamic variables (e.g., cardiac output or stroke volume). Sensors can also be monitors, such as wearable devices used to measure and monitor vital signs.

[0046] Figure 3 A method for echocardiography-based resuscitation guidance is illustrated according to a representative embodiment.

[0047] Figure 3 The method can be performed by system 100 including controller 150 or by system 200 including controller 250.

[0048] At S310, the ultrasonic procedure begins. Figure 1 System 100 or Figure 2 The system 200 in the middle can start the ultrasound procedure. At S310, it can be started... Figure 1 The ultrasound probe 110 and / or ultrasound host 120, or by starting Figure 2 The ultrasound procedure is initiated using ultrasound probe 210 and / or ultrasound application 299A and / or ultrasound application 299B. After initiating the ultrasound procedure at S310, cannulation assistance can be provided when the echocardiogram shows a TEE.

[0049] At S320, one or more echocardiographic images are received. Multiple echocardiographic images can be obtained from... Figure 1 The ultrasound host 120 receives signals from the ultrasound probe 110, or is powered by... Figure 2 The controller 250 and / or either or both of smartphone A and / or smartphone B receive the data. Figure 2 In this context, S320 can be executed by controller 250 or any device in one or more smartphones that performs the logic processing described herein. The echocardiographic image can be a TEE-based or TTE-based echocardiographic image.

[0050] At S330, the echocardiographic images (one or more) and / or pre-imaging data are interpreted. Interpretation at S330 helps ensure that the ultrasound probe 110 or ultrasound probe 210 is in the correct position, as such feedback can be used to allow the user to confirm correct positioning or prompt the user to correct the position. Interpretation at S330 can be performed before the echocardiographic images are formed on display 180 or on smartphone A or smartphone B, but interpretation at S330 may also include identifying and classifying anatomical features in each of the multiple views, for example, by applying an object recognition model to each echocardiographic image. Interpretation at S330 may include checking whether each ultrasound image is well focused, whether each ultrasound image has an appropriate resolution, and whether each ultrasound image reflects the patient's cardiac anatomy. Interpretation at S330 may also include checking the characteristics of compressions, including the location, depth, and frequency of compressions. Interpretation at S330 may also include checking the characteristics of different cardiac disease states and / or cardiac arrest. Interpretation at S330 may also include quantifying hemodynamic variables to assess the quality of the compressions performed. Each echocardiographic image in a set of multiple echocardiographic images can be interpreted individually or as a group. For example, a set of ultrasound images can be compared to detect differences, such as ensuring that they are not completely or substantially duplicated images.

[0051] Figure 3The method may also include controller 150 or controller 250 identifying potential causes of cardiac arrest by interpreting each of multiple echocardiographic images. At least one potential cause of cardiac arrest may also be included in the real-time feedback provided to the user performing chest compressions. Therefore, the real-time feedback may include: suggestions to move the anatomical location of the compressions to align with the optimal anatomical location of the compressions individualized for the hemodynamic response of the patient receiving the compressions, and notification of one or more potential causes of cardiac arrest. After interpreting the echocardiographic images at S330, the view quality may be detected; if the view quality is poor, guidance may be generated, and the process may return to S320. If the image quality is detected at S330 and determined to be acceptable, the process may continue.

[0052] At S340, the anatomical location of the compression is identified. Controller 150 or controller 250 can identify the optimal anatomical location of the compression from multiple echocardiographic images / pre-imaging data by interpreting data acquired from each of multiple echocardiographic images / pre-imaging data. The anatomical location of the compression can also be identified using pre-acquisition data (e.g., pre-acquisition data from volumetric imaging). The anatomical location of the maximum compression can be identified as the area of ​​the heart most significantly compressed during compression. Quantization mechanisms used to identify the compression location include, but are not limited to: volumetric changes, spectral vector quantization (quantifying the spatial location and applied force during compression), pressure gradient quantization, or mechanisms for detecting tissue deformation within 2D or 3D volumetric imaging datasets. This can include speckle tissue tracking, tissue Doppler imaging, shear wave elastography, strain elastography, compression elastography, or tissue displacement modeling. These mechanisms can also be based on deep learning, such as using left ventricular segmentation. Furthermore, the compression location can be identified when the ultrasound device is paired with an external device that monitors compression implementation or hemodynamic parameters. Such devices may include automated compression devices (e.g., mechanical or robotic compression devices) or wearable or invasive devices (e.g., sensors) located on the person performing the compression, which can provide hemodynamic feedback. Mechanisms from which the location of the applied compression force can be derived may include gyroscopes, accelerometers, IMUs, RF signaling, or other integrated sensors (force sensors, proximity sensors, light sensors, pressure sensors).

[0053] Although compression location identification is primarily described as image-based, it can also be identified using other inputs, or alternatively. For example, the magnitude of the force applied during compression can be quantified and matched with the resulting deformation to identify the location, potentially without image feedback. Pre-imaging data, such as detected blood flow / strain measurements, can be derived from radio frequency (RF) data, which is part of pre-beamforming data acquisition.

[0054] At S350, one or more compression characteristics are detected. Compression characteristics may include compression frequency and compression depth. A single echocardiographic image and a set of multiple echocardiographic images can be evaluated at S350 to assess compression depth and compression frequency. For example, frequency can be detected from a set of multiple echocardiographic images to detect whether the compression is too fast or too slow. Compression depth can be detected to determine whether the user should compress the patient deeper or less.

[0055] At S360, select potential corrections. Potential corrections may include adjusting the placement of the hand / mechanical device so that compressions can be performed in the left ventricular position when the hand is not currently in the left ventricular position. Optimal compression placement may also be based on the patient's hemodynamic response, or other factors besides the hand's position relative to the left ventricle. Potential corrections may also include adjustments to compression depth, such as suggesting adjusting the compression depth to increase or decrease it. Potential corrections may also include adjustments to compression frequency, such as suggesting a faster or slower pace of compressions.

[0056] At S370, feedback is generated. The feedback can be visual and / or auditory. The feedback can indicate the current pressing position and suggest that the user move their hand in one or more directions and by how much, such as moving 3cm to the left or 3cm down. The feedback can suggest that the pressing should be sped up or slowed down, for example, once per second. Feedback regarding pressing frequency can be auditory and / or visual, and can also indicate the current pressing speed. Therefore, feedback regarding pressing frequency could indicate: Increase the pressing frequency from once every two seconds to once per second.

[0057] At S380, one or more echocardiographic images and feedback are output. The echocardiographic images are output visually, for example, via a display 180 or on the user interface of smartphone A and / or smartphone B. The feedback can be output visually and / or audibly. For example, instructions regarding adjusting hand position, increasing or decreasing compression speed, and / or increasing or decreasing compression depth can be both auditory and visual, or only auditory or visual.

[0058] although Figure 3 The steps in the process are shown as a method flow, but Figure 3 Some or all of the steps can be performed in a different order or simultaneously. For example, one or more echocardiographic images can be received and processed individually at a time, such that while the first echocardiographic image is being interpreted at S330, a second echocardiographic image is being received at S320. The output at S380 can be executed multiple times, for example, once every two seconds or every five seconds, to indicate adjustments to the compressions being performed.

[0059] Figure 4 Another method for echocardiography-based resuscitation guidance, according to a representative embodiment, is shown.

[0060] exist Figure 4 During the pressing, evaluation, and correction process, a feedback loop is provided. Feedback is provided to the user in real time. As a press is applied, data is collected and fed into an algorithm executed by controller 150, controller 250, or smartphone A or smartphone B (representing a networked / connected mobile device with logic processing capabilities). This algorithm categorizes the press quality. Poor-quality press data can be fed into a press correction algorithm to determine how the press should be corrected and to provide any one or more suggestions as recommendations to the user.

[0061] More specifically, the user applies pressure at S410. Data is input at S420. The pressure is categorized at S430, for example, by classifying pressure quality using a first algorithm. When a pressure is categorized as poor, a pressure correction algorithm is applied at S440 to determine how the pressure should be corrected. When a pressure is categorized as good, system 100 or system 200 can output a guarantee to the user that the pressure was performed well. When a pressure is categorized as poor, feedback to the user at S450 forms part of a loop so that the next pressure applied at S410 can reflect the recommendations of system 100 or system 200 regarding how to correct the pressure. Figure 4 The method is a real-time press feedback method that can be performed during focused TEE or during TTE.

[0062] A classification model can be used to detect the location and quality of chest compressions. The classification model may include an input layer, hidden layers, and an output layer with activation functions. The output layer can output an indication of whether the compression is good or bad, and the output of a bad compression can be used as input to a separate optimization algorithm that provides compression correction measures and user feedback. This feedback can be used cyclically to evaluate the quality of chest compressions and improve the quality of chest compressions when potential areas for improvement are identified.

[0063] Next, combine Figure 5 Demonstrate and explain the detailed mechanisms used for pressure detection, classification, and optimized feedback. Figure 5 In this implementation, deep learning mechanisms can be used to perform pressure-optimized feedback. In other embodiments, alternative signal processing techniques can be used to perform this. Figure 5 Optimized press feedback.

[0064] Figure 5 A method for echocardiography-based resuscitation guidance is illustrated according to a representative embodiment.

[0065] Figure 4The purpose is to illustrate the overall concept of a real-time pressure feedback system. Figure 5 One implementation method is explained in detail. Figure 5 express Figure 4 A detailed example. The diagram below illustrates an example of how it works. When the user applies pressure / relief, image data is acquired from the system, and the features of the pressure are calculated as the input vector for the first system. Each independent variable xn contains information about the pressure quality. These parameters can then be fed into a classification algorithm, such as a deep learning algorithm, which outputs a binary classification result for pressure quality (good / poor). The output of this step is a binary classification result Ŷ for pressure quality (good / poor). Poor pressure (Ŷ=1) is fed into a secondary model, which determines how to correct the pressure and communicates the result to the user in real time.

[0066] exist Figure 5 During the process, the user applies chest compressions and decompressions at S510. Data on compressions and decompressions is received, and a set of input features is extracted from the dataset to evaluate the quality of the compressions and decompressions.

[0067] At S520, the input to the classification algorithm is calculated based on the pressing and depressing actions. Multiple different inputs can be calculated, including those at different time points during the pressing and depressing process and those of different types.

[0068] At S530, a classification algorithm is applied to detect the location and quality of the pressure applied. The classification algorithm may include a trained machine learning model having an input layer, one or more hidden layers, and an output layer. The trained machine learning model may include a deep neural network for the hidden layers. The classification algorithm outputs a good or bad classification at S535 as the result of the activation function.

[0069] At S540, either system 100 or system 200 calculates a correction measure. The correction measure can be calculated at S540 using an optimized algorithm that provides pressure correction and user feedback.

[0070] At S550, the optimization algorithm executed by system 100 or system 200 provides feedback to the user for correcting the press. This feedback is provided to correct the press when the algorithm outputs an unsuitable classification. Figure 5 In this method, a first subsystem and a second subsystem can be used to perform the operation. The first subsystem can use an artificial neural network to evaluate the pressure quality. For example, a multilayer perceptron model with backpropagation or a convolutional neural network (CNN) can be used as the first subsystem.

[0071] Figure 5The method can be executed by controller 150 or controller 250, and can involve a single integrated system implemented by a single algorithm or two separate subsystems implemented by different algorithms. The first subsystem can use a deep learning classification algorithm to detect the location and quality of the press, such as... Figure 5 As shown in S530, this deep learning classification model illustrates a feedback loop used to monitor press quality. Several inputs can be used as independent variables for monitoring press quality. Figure 5 In this model, the inputs as independent variables include: x1 for identifying the area of ​​maximum compression; x2 for calculating hemodynamic variables; x3 for classifying image features in each of multiple views (e.g., the ResusTEE view); x4 for evaluating compression depth and frequency; and xn for other potential inputs. The model's output makes a judgment on compression quality based on these inputs. For example, the output of a classification model could be a binary classifier for compressions, categorizing them as either good or poor compressions.

[0072] A second algorithm can be applied to identify optimal compression positions and characteristics and to correct the user's compression. Compressions classified as poor can be fed into a secondary model that evaluates how to correct them. This secondary model can use deep learning algorithms and can transmit this information to the user in real time during compression. The secondary optimization model can be based on machine learning, such as a backpropagation algorithm that uses stochastic gradient descent to iteratively update the learning rate and vector weights to identify optimal compression positions / quality. Real-time feedback can be provided to the user on whether their compression position is correct, and feedback on compression quality can also be provided based on compression depth, frequency, and position customized to the patient's hemodynamic response.

[0073] Figure 6 A user interface for echocardiography-based resuscitation guidance is shown according to a representative embodiment.

[0074] Figure 6 Three user interfaces are shown, including a first user interface 681A, a second user interface 681B, and a third user interface 681C. Figure 6 The three user interfaces demonstrate how to detect and provide pressure feedback in a scenario involving aortic root compression.

[0075] Figure 6The user interface illustrates the concept of how to provide compression feedback in cases of improper compression placement. These three user interfaces follow a user interface workflow, illustrating an exemplary scenario of compression at the aortic root in a midesophageal long-axis view. The first user interface 681A is from an image taken in a relaxed state and includes automatic heart chamber markings. The second user interface 681B is from an image taken during compression, with the CPR detector enabled. As the compression phase progresses and the right ventricle deforms, the ascending aorta in the image on the second user interface 681B is narrowing. Changes in cardiac morphology associated with good / poor compression are detected. The second user interface 681B warns the user that inappropriate compression is being applied and redirects the user to the correct compression area, such as the left ventricle. The third user interface 681C informs the user that the area of ​​maximum compression has been detected at the fully collapsed ascending aorta and redirects the user to the correct compression position in real time.

[0076] Before proceeding, the following will detail how input is determined. The process of determining input generally includes: identifying the area of ​​maximum compression; calculating hemodynamic variables to assess the quality of the compression / decompression performed; classifying cardiac features associated with compression / decompression performance in images; and detecting compression depth and frequency.

[0077] Identifying the maximum compression region may involve defining and quantifying the maximum compression region among candidate regions. The maximum compression region can be defined as the area of ​​the heart where compression is most significant at the end of the compression phase. Quantization mechanisms may be based on, for example, acoustic compression vector quantization, pressure gradient mechanisms or similar tissue deformation detection mechanisms, calculation of the distance to the aortic root, or left ventricular segmentation. Acoustic compression vector quantization quantifies the spatial location and applied force during compression. Tissue deformation detection mechanisms other than pressure gradient mechanisms may include speckle tissue tracking, shear wave elastography, compression elastography, or tissue Doppler imaging. The distance to the aortic root is calculated using an m-mode. Left ventricular segmentation can be performed using a neural network.

[0078] Calculations of hemodynamic variables can be broadly categorized to include calculations during the compression phase and quantitative measurements of quality decompression. Examples of relevant TEE views that can be used to measure compression quality include the TEE ME 4C view, the TEE ME RV inflow-outflow view, or the TEE biatrial view. General measurements can be performed in all views. Deviations of these measurements from the mean inform the model about the quality of the hemodynamic response.

[0079] Here's an example of how the model can be applied to feedback loops and how measurement results can guide compression improvements: The first step might involve checking if the area of ​​maximum compression is above the patient's atrium to examine compression quality. Compression depth can also be checked to optimize it. Slow or incomplete chest recoil can be checked to optimize chest recoil. The second step might involve additional checks, such as left ventricular obstruction, as an indication of ineffective CPR.

[0080] The following measurements can be performed in the TEE ME 4C view: i. Two-dimensional (2D) fractional change in right ventricular (RV) area (FAC), calculated as: [end-diastolic area - end-systolic area) / end-diastolic area × 100]. This is used to estimate right ventricular (RV) function. The normal value for right ventricular (RV) systolic function is 35%, and the average fractional change in area (FAC) under chest compressions is 59.6%. ii. Measure right ventricular (RV) lateral wall offset (LWE) in M ​​mode.

[0081] iii. Left ventricular (LV) ejection fraction was measured to estimate LV function, calculated using the formula [(end-diastolic volume - end-systolic volume) / end-diastolic volume] × 100]. The average left ventricular ejection fraction under chest compressions was 53.5%. iv. Fractional change in left ventricular (LV) area (FAC).

[0082] Assess right ventricular outflow tract compression in the TEE ME RV inflow-outflow view. Measure right ventricular outflow tract short-axis shortening (FS) in M ​​mode. The mean FS during chest compressions was 45.3% (range 22.8–64.2%). These measurements can be used in feedback loops to improve compression quality.

[0083] Assess atrial compressions in the TEE biatrial view. The assessment in the TEE biatrial view may include: identifying the area of ​​maximum compression at x1, and the right atrial wall displacement under compression as observed in m-mode.

[0084] General measurements in all views may include cardiac output and stroke volume.

[0085] Quantitative measurements of decompression quality may include chest recoil quality. Good chest recoil generates negative intrathoracic pressure and reduces right atrial (RA) pressure. The mechanism for detecting pressure changes may be Doppler-based or quantitative in nature. Alternatively, quantitative measurements may include, for example, observing right ventricular (RV) lateral wall displacement, compression velocity (DS), and / or relaxation time during decompression in m-mode in a 4-chamber view of the esophagus in a TEE.

[0086] Next, combine Figure 7 Interpret image classification of cardiac features associated with compression / decompression responses.

[0087] Figure 7 The development and deployment of a model for echocardiography-based resuscitation guidance, according to a representative embodiment, are illustrated.

[0088] As described in this article, defining compression quality can involve multivariate decisions based on a series of inputs x = x1, x2, x3…xn. One of the inputs, x3, can be used to define compression quality based on morphological changes in intracardiac signs indicating good / poor compression. Figure 7 The methods used to determine the model development and deployment are illustrated. Figure 7 The algorithm in can be represented Figure 5 This is one of the many steps involved in the "Calculate Input" step. (In order to develop...) Figure 7 The algorithm in this paper uses an annotated dataset of pre-labeled images associated with good / poor compression quality to train an image classification algorithm to detect signs of good / poor compression quality. Once the model is deployed, an image associated with aortic root obstruction can be fed into it. Figure 7 In this model, the input image is detected as a sign of poor pressing quality, and the output variable x3 = poor. This variable can then be fed into... Figure 5 In the press classification algorithm, multiple inputs related to press quality are used to perform binary classification judgment on press quality.

[0089] Figure 7 The deployment model shown at S730 on the right is an example of a model used to classify good / poor compression quality. The image processing algorithm used in this deployment model can identify features indicating good / poor quality compression and decompression of cardiac chambers across all relevant imaging planes associated with echocardiographic resuscitation protocols such as those used for TEE. Classification can be performed using tissue tracking algorithms, deep learning image classification algorithms, or any quantification mechanism capable of detecting changes in cardiac chamber morphology and training the model to obtain signs of good / poor compression quality. Typically, signs of good compression include ventricular collapse and open left ventricular outflow tract, while signs of poor compression include collapse of the left ventricular outflow tract / aorta during compression.

[0090] Figure 7The set of six ultrasound images on the user interface 781 represents the training dataset for the compression quality classification algorithm used in the model to be deployed. These images are obtained from the dataset, preprocessed, and annotated with relevant features related to compression quality, indicated by asterisks. The six images in the training set on the user interface 781 are labeled C, D, E, F, G, and H in sequence. The images in the training dataset on the user interface 781 are all from the midesophageal long axis plane of patients undergoing cardiac arrest CPR. The first three images (C, D, E) show signs of good compression / decompression, and the last three images (F, G, H) show signs of poor compression / decompression. The first image, labeled C, is used for image classification features of good decompression. The first image is an image of the left ventricle during the decompression phase of mitral valve opening (asterisk). The second image, labeled D, is used for appropriate early left ventricular (LV) compression (asterisk) without left ventricular outflow tract or aortic root obstruction. The third image, marked E, is used to show aortic valve opening (asterisk) during the remaining time of the compression phase, thus allowing for the detection of stroke volume.

[0091] The fourth image marked F, the fifth image marked G, and the sixth image marked H on user interface 781 are signs of inadequate decompression. These images depict an example of complete aortic root obstruction during the compression phase of CPR, which effectively impedes forward blood flow. The asterisks indicate the location of the progressive aortic root obstruction during the compression phase. Figure 7 As shown, these features related to good / poor compression / decompression can be fed into a deep learning algorithm on an annotated dataset, which utilizes supervised learning to detect these features. Once successfully developed, the algorithm can take in new images during CPR and classify compression / decompression quality based on changes in heart morphology. The output of this classifier can be fed into an input node x3, where x3 represents the classification feature of good / poor compression / decompression. For example, x3 can be a subset of the input parameters [x1, x2, x3…...xn] to provide an overall classifier for compression quality.

[0092] At S702, the training dataset, represented by six images from the user interface 781, is classified and used for model development. At S730, the model is deployed. The input to the model deployment can be real-world image data deployed during patient resuscitation activities. The output of the deployment can be a binary classification, such as... Figure 7The input parameters can be categorized as either "poor" or "good." A poor classification will result in feedback, which can be based on a subset of x3 of the input parameters. A good classification may output a guarantee. The deployed model can be used in a feedback loop, allowing feedback from poor classifications to adjust the press / depressor being performed, and the adjusted press / depressor will be captured in real image data and fed back into the model deployed at S730. Feedback can be provided audibly or visually, such as via voice prompts, tone of voice, or other mechanisms used to convey adjustment suggestions.

[0093] Compression depth and frequency can also be measured. Current resuscitation guidelines recommend a compression depth of at least 5 cm but no more than 6 cm, and a compression rate of 100 to 120 compressions per minute, allowing chest recoil between compressions. Compression depth and frequency can be measured using deformation measurement methods such as tissue tracking, volume change measurement, Doppler imaging, compression elastography, or simplified accelerometer methods.

[0094] Figure 8 A computer system according to another representative embodiment is shown, on which a method for echocardiography-based resuscitation guidance is implemented.

[0095] See Figure 8 The computer system 800 includes a set of software instructions that can be executed to cause the computer system 800 to perform any of the methods or computer-based functions disclosed herein. The computer system 800 can operate as a standalone device or can be connected to other computer systems or peripheral devices, for example, using a network 801. In embodiments, the computer system 800 performs logical processing based on digital signals received via an analog-to-digital converter.

[0096] In a networked deployment, computer system 800 operates as a server or client user computer in a server-client user network environment, or as a peer-to-peer (or distributed) computer system in a peer-to-peer (or distributed) network environment. Computer system 800 can also be implemented as or integrated into various devices, such as workstations including controllers, fixed computers, mobile computers, personal computers (PCs), laptops, tablets, or any other machine capable of executing a set of software instructions (sequentially or otherwise) specifying actions to be taken by that machine. Computer system 800 can be implemented as a device or integrated into a device that is itself an integrated system including additional devices. In embodiments, computer system 800 can be implemented using electronic devices that provide voice, video, or data communications. Furthermore, although computer system 800 is shown in the singular, the term "system" should also include any system or collection of subsystems that individually or jointly operate one or more sets of software instructions to perform one or more computer functions.

[0097] like Figure 8 As shown, computer system 800 includes processor 810. Processor 810 can be considered a representative example of a processor of a controller that executes instructions to implement some or all aspects of the methods and processes described herein. Processor 810 is tangible and non-transitory. As used herein, the term "non-transitory" should not be interpreted as a permanent characteristic of a state, but rather as a characteristic that a state will persist for a period of time. The term "non-transitory" explicitly excludes transient characteristics, such as the characteristics of a carrier wave or signal, or other forms that exist only briefly at any time and place. Processor 810 is an article of manufacture and / or a machine component. Processor 810 is configured to execute software instructions to perform the functions described in the various embodiments herein. Processor 810 may be a general-purpose processor or may be part of an application-specific integrated circuit (ASIC). Processor 810 may also be a microprocessor, microcomputer, processor chip, controller, microcontroller, digital signal processor (DSP), state machine, or programmable logic device. Processor 810 may also be logic circuitry, including programmable gate arrays (PGAs) (such as field-programmable gate arrays (FPGAs)), or another type of circuitry including discrete gate and / or transistor logic. The processor 810 may be a central processing unit (CPU), a 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.

[0098] As used herein, the term "processor" encompasses an electronic component capable of executing programs or machine-executable instructions. References to computing devices that include "processor" should be interpreted as including multiple processors or processing cores, such as in a multi-core processor. A processor can also refer to a collection of processors within a single computer system or distributed across multiple computer systems. The term computing device should also be interpreted as including a collection or network of computing devices, each comprising one or more processors. A program has software instructions that are executed by one or more processors, which may be located within the same computing device or distributed across multiple computing devices.

[0099] Computer system 800 also includes main memory 820 and static memory 830, wherein the memories in computer system 800 communicate with each other and with processor 810 via bus 808. Either or both of main memory 820 and static memory 830 may be considered representative instances of the memory of a controller and store instructions for implementing some or all aspects of the methods and processes described herein. The memory described herein is a tangible storage medium for storing data and executable software instructions, and is non-transitory for the period during which the software instructions are stored. As used herein, the term "non-transitory" should not be interpreted as a permanent characteristic of a state, but rather as a characteristic that the state will persist for a period of time. The term "non-transitory" expressly excludes transient characteristics, such as the characteristics of a carrier wave or signal, or other forms that exist only briefly at any time and place. Main memory 820 and static memory 830 are articles of art and / or machine components. Main memory 820 and static memory 830 are computer-readable media from which a computer (e.g., processor 810) can read data and executable software instructions. Each of the main memory 820 and the static memory 830 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, hard disks, removable disks, magnetic tapes, optical disc read-only memory (CD-ROM), digital versatile disks (DVDs), floppy disks, Blu-ray discs, or any other form of storage medium known in the art. The memory may be volatile or non-volatile, secure and / or encrypted, insecure and / or unencrypted.

[0100] “Memory” is an example of a computer-readable storage medium. Computer memory is any memory that can be directly accessed by a processor. Examples of computer memory include, but are not limited to, RAM, registers, and register files. The reference to “computer memory” or “memory” should be interpreted as potentially referring to multiple memories. For example, memory can be multiple memories within the same computer system. Memory can also be multiple memories distributed across multiple computer systems or computing devices.

[0101] As shown in the figure, the computer system 800 also includes a video display unit 850, 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). Additionally, the computer system 800 includes an input device 860 (e.g., a keyboard / virtual keyboard or a touch-sensitive input screen or voice input with voice recognition), and a cursor control device 870 (e.g., a mouse or a touch-sensitive input screen or pad). The computer system 800 may also optionally include a disk drive unit 880, a signal generation device 890 (e.g., a speaker or a remote control), and / or a network interface device 840.

[0102] In an embodiment, such as Figure 8 As shown, the disk drive unit 880 includes a computer-readable medium 882 in which one or more sets of software instructions 884 (software) are embedded. The software instructions 884 are read from the computer-readable medium 882 for execution by the processor 810. Furthermore, the software instructions 884, when executed by the processor 810, perform one or more steps of the methods and processes described herein. In embodiments, the software instructions 884 reside wholly or partially within main memory 820, static memory 830, and / or processor 810 during execution by the computer system 800. Additionally, the computer-readable medium 882 may include the software instructions 884 or software instructions 884 received and executed in response to a propagation signal, causing a device connected to the network 801 to transmit voice, video, or data through the network 801. The software instructions 884 may be transmitted or received through the network 801 via a network interface device 840.

[0103] In the embodiments, dedicated hardware implementations, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), programmable logic arrays, and other hardware components, are constructed to implement one or more methods described herein. One or more embodiments described herein may implement functionality using two or more specific interconnected hardware modules or devices having associated control and data signals that can communicate between and through the modules. Therefore, this disclosure covers software, firmware, and hardware implementations. Nothing in this application should be construed as being implemented or achievable solely in software and not in hardware (e.g., tangible, non-transitory processors and / or memory).

[0104] According to various embodiments of this disclosure, the methods described herein can be implemented using a hardware computer system that executes software programs. Furthermore, in exemplary non-limiting embodiments, implementations may include distributed processing, component / object distributed processing, and parallel processing. Virtual computer system processing can implement one or more of the methods or functions described herein, and the processor described herein can be used to support virtual processing environments.

[0105] Therefore, echocardiogram-based compression guidance enables real-time assessment and feedback of cardiopulmonary compression quality based on anatomical location, depth, frequency, hemodynamic measurements, and / or anatomical images. This assessment and feedback can be used to optimize coronary perfusion and individualize chest compressions for patients during CPR. Echocardiogram-based assessment and compression guidance can provide an effective user experience, supporting resuscitation efforts by optimizing the quality of chest compressions, thereby improving patient survival. For example, echocardiogram-based compression guidance may be particularly useful for patients undergoing cardiac or high-risk surgery and postoperatively in cardiac or surgical ICUs, as they are already intubated.

[0106] Although echocardiography-based compression guidance has been described with reference to several exemplary embodiments, it should be understood that the language used is descriptive and illustrative, not restrictive. Changes may be made in accordance with the scope of the presently stated and amended claims without departing from the scope and spirit of echocardiography-based compression guidance. While echocardiography-based compression guidance has been described with reference to specific devices, materials, and embodiments, it is not intended to be limited to the specific details disclosed; rather, it extends to all functionally equivalent structures, methods, and uses within the scope of the claims.

[0107] The illustrations of the embodiments described herein are intended to provide a general understanding of the structures of various embodiments. These illustrations are not intended to be a complete description of all elements and features of the disclosure herein. Many other embodiments may arise in the mind of those skilled in the art upon reading this disclosure. Other embodiments may be utilized and derived from this disclosure, allowing for structural and logical substitutions and changes without departing from the scope of this disclosure. Furthermore, these illustrations are merely schematic and may not be drawn to scale. Some scales in the illustrations may be exaggerated, while others may be minimized. Therefore, this disclosure and the accompanying drawings are to be considered illustrative rather than restrictive.

[0108] In this disclosure, one or more embodiments may be referred to individually and / or collectively as "the present invention," merely for convenience, and are not intended to voluntarily limit the scope of this application to any particular invention or inventive concept. Furthermore, although specific embodiments have been shown and described herein, it should be understood that any subsequent configurations intended to achieve the same or similar purpose may replace the specific embodiments shown. This disclosure is intended to cover any and all subsequent modifications or variations of the various embodiments. Combinations of the above embodiments and other embodiments not specifically described herein will be apparent to those skilled in the art upon reading the specification.

[0109] This abstract of disclosure is provided to conform to 37 C. FR § 1.72(b), and it should be understood at the time of filing that it is not intended to interpret or limit the scope or meaning of the claims. Additionally, in the foregoing detailed embodiments, various features may be combined together or described in a single embodiment for the purpose of simplifying the invention. This disclosure should not be construed as reflecting an intention that the claimed embodiments require more features than expressly recited in each claim. Rather, as reflected in the claims, the inventive subject matter may address fewer features than all of any disclosed embodiment. Therefore, the claims are incorporated into the detailed description, with each claim independently defining a separate claimed subject matter.

[0110] The foregoing description of the disclosed embodiments is provided to enable any person skilled in the art to practice the concepts described in this disclosure. Therefore, the subject matter of the above disclosure should be considered illustrative rather than restrictive, and the claims are intended to cover all such modifications, alterations, and other embodiments falling within the true spirit and scope of this disclosure. Accordingly, the scope of this disclosure should be determined by the broadest permissible interpretation of the claims and their equivalents to the fullest extent permitted by law, and should not be bound or limited by the foregoing detailed description.

Claims

1. An ultrasound system (100 / Figure 2), comprising: The memory (151 / 251) stores instructions; as well as Processor (152 / 252), which executes the instructions, wherein the instructions, when executed by the processor, cause the ultrasound system to: Receive (S320) multiple echocardiographic images during compression; Interpret each of the multiple echocardiographic images (S330); Output (S380) each of the multiple echocardiographic images; and Based on the interpretation of each of the multiple echocardiogram images, real-time feedback on the quality of chest compressions is generated (S370) and output (S380).

2. The ultrasound system according to claim 1, wherein, The instructions, when executed by the processor, further enable the ultrasound system: By interpreting each of the multiple echocardiographic images, the anatomical location of the compression is identified (S340) and the quality of the compression is evaluated, wherein the quality of the compression is based on at least one of the anatomical location of the compression, the depth of the compression, the frequency of the compression, hemodynamic measurements, or anatomical images.

3. The ultrasound system according to claim 2, wherein, The real-time feedback is provided to optimize coronary perfusion and personalize cardiopulmonary resuscitation for patients receiving the compressions.

4. The ultrasound system according to claim 1, wherein, The instructions, when executed by the processor, further enable the ultrasound system: By interpreting each of the multiple echocardiographic images, the optimal anatomical location for compression in the echocardiographic images is identified (S340).

5. The ultrasound system according to claim 4, wherein, The instructions, when executed by the processor, further enable the ultrasound system: By interpreting each of the multiple echocardiographic images, the compression position in the echocardiographic image is identified (S340), wherein the real-time feedback is provided based on a comparison of the anatomical position of the compression with the anatomical position of the optimal compression.

6. The ultrasound system according to claim 5, wherein, The real-time feedback includes a suggestion to move the anatomical location of the compression to align with the anatomical location of the optimal compression, which is individualized for the hemodynamic response of the patient receiving the compression.

7. The ultrasound system according to claim 1, wherein, The instructions, when executed by the processor, further enable the ultrasound system: By interpreting each of the multiple echocardiographic images, potential causes of cardiac arrest are identified (S330); and The real-time feedback includes at least one potential cause of cardiac arrest.

8. The ultrasound system according to claim 1, wherein, The instructions, when executed by the processor, further enable the ultrasound system: Based on the interpretation of each of the multiple echocardiographic images, (S360) which of the multiple potential correction measures is used as the feedback output for compression quality.

9. The ultrasound system according to claim 1, wherein, Interpreting each of the multiple echocardiographic images includes identifying the area of ​​maximum compression.

10. The ultrasound system according to claim 1, wherein, Interpreting each of the multiple echocardiographic images involves calculating at least one hemodynamic variable.

11. The ultrasound system according to claim 1, wherein, Interpreting each of the multiple echocardiographic images involves classifying the anatomical features in each of the multiple views.

12. The ultrasound system according to claim 1, wherein, The instructions, when executed by the processor, further enable the ultrasound system: Compression characteristics were detected from the multiple echocardiographic images by evaluating compression depth and compression frequency (S350).

13. The ultrasound system according to claim 1, further comprising: A display (180 / smartphone A and / or smartphone B) is used to display the multiple echocardiographic images and the real-time feedback on the quality of the compressions.

14. A method of operating an ultrasound system, the ultrasound system comprising a memory storing instructions and a processor executing the instructions, the method comprising: Receive (S320) multiple echocardiographic images during compression; Interpret each of the multiple echocardiographic images (S330); Output (S380) each of the multiple echocardiographic images; and Based on the interpretation of each of the multiple echocardiogram images, real-time feedback on the quality of chest compressions is generated (S370) and output (S380).

15. The method of claim 14, further comprising: By interpreting each of the multiple echocardiographic images, the anatomical location of the compression is identified (S340) and the quality of the compression is evaluated, wherein the quality of the compression is based on at least one of the anatomical location of the compression, the depth of the compression, the frequency of the compression, hemodynamic measurements, or anatomical images, and wherein real-time feedback is provided to optimize coronary perfusion and personalize cardiopulmonary resuscitation for the patient receiving the compression.

16. The method of claim 14, further comprising: By interpreting each of the multiple echocardiographic images, the optimal anatomical location for compression is identified (S360) in the echocardiographic images; and By interpreting each of the multiple echocardiographic images, the anatomical location of the compression in the echocardiographic images is identified (S340), wherein the real-time feedback is provided based on the optimal anatomical location of the compression and the anatomical location of the compression, wherein the real-time feedback includes a suggestion to move the anatomical location of the compression to align with the optimal anatomical location of the compression based on the hemodynamic response of the patient receiving the compression.

17. The method of claim 14, further comprising: By interpreting each of the multiple echocardiographic images, potential causes of cardiac arrest can be identified.

18. The method of claim 14, further comprising: Based on the interpretation of each of the multiple echocardiographic images, (S360) which of the multiple potential correction measures is used as the feedback output for compression quality.

19. The method of claim 14, wherein, Interpreting each of the multiple echocardiographic images includes identifying the area of ​​maximum compression, calculating at least one hemodynamic variable, classifying or evaluating at least one of compression depth and compression frequency in each of the multiple views.

20. A tangible, non-transitory computer-readable medium storing instructions, which, when executed by a processor, cause the processor to: Receive (S320) multiple echocardiographic images during compression; Interpret each of the multiple echocardiographic images (S330); Output (S380) each of the multiple echocardiographic images; and Based on the interpretation of each of the multiple echocardiogram images, real-time feedback on the quality of chest compressions is generated (S370) and output (S380).

21. The ultrasound system according to claim 1, further comprising: An automatic chest compression device configured to perform the compression, wherein the instruction further causes the ultrasound system to provide real-time feedback to the automatic chest compression device regarding the compression, so as to adjust the positioning of the compression performed by the automatic chest compression device.

22. The ultrasound system according to claim 1, further comprising: A robotic system configured to control the localization of the source of the echocardiographic image, wherein the real-time feedback further includes feedback to the robotic system to adjust the view in the echocardiographic image.

23. The ultrasound system according to claim 1, further comprising: A sensor configured to measure a patient’s hemodynamic variables and provide feedback to a controller including the memory and the processor.

24. The ultrasound system according to claim 23, wherein, The sensor includes an interventional sensor configured to receive instructions from the controller to stimulate the hemodynamic variables.

Images